rejetto forum

Software => HFS ~ HTTP File Server => Beta => Topic started by: rejetto on September 20, 2018, 09:34:58 PM

Title: version 2.4
Post by: rejetto on September 20, 2018, 09:34:58 PM
download latest version (https://github.com/rejetto/hfs2/releases/download/v2.4-rc07/hfs.exe)

new default template, mobile friendly and responsive.
unicode
ipv6
session login, with logout
Title: Re: version 2.4
Post by: dj on September 21, 2018, 02:43:41 AM
http://localhost/fa/font-awesome.css 404 Not Found


addons tested with version 2.4:

addoncheckedadjustment
audioplayerlight (http://rejetto.com/forum/index.php?topic=11754.msg1063922#msg1063922)(playbutton in footer)
logout (http://rejetto.com/forum/index.php?topic=11754.msg1062970#msg1062970)var ref=document.querySelector('button[onclick*=user]')
jpgimgthumbs
pictureviewercoming tomorrow
archive_as_zip (http://rejetto.com/forum/index.php?topic=11754.msg1063899#msg1063899)todo

themes tested with version 2.4: dark
Code: [Select]
[+common-head]
<style>
#files {background:#555}
</style>
Title: Re: version 2.4
Post by: bmartino1 on September 21, 2018, 04:39:19 PM
Have downloaded and will test here in a bit with Android pixel 2 phone amoung other testing device.
Title: Re: version 2.4
Post by: rejetto on September 21, 2018, 06:27:00 PM
oh shit, you are missing ALL the icons :(
i'll publish another version ASAP
Title: Re: version 2.4
Post by: rejetto on September 21, 2018, 07:29:51 PM
ok guys, i'm sorry for wasting your time, please REdownload the link above.
It should be fine now.
Title: Re: version 2.4
Post by: Mars on September 21, 2018, 10:04:05 PM
tests done in firefox on windows

There is an forgetting in the template, is missing a passage in the following line concerning links, to see image

(https://i27.servimg.com/u/f27/16/40/25/39/temp11.jpg)

by removing this line the default disappears
.item-link { float:left; }


There is a problem with the selection tool, when we check a box, it performs the same action as clicking on the name





Title: Re: version 2.4
Post by: dj on September 23, 2018, 05:24:55 PM
There is a problem with the selection tool, when we check a box, it performs the same action as clicking on the name

change '.item-selectable a' to '.item-selectable' (line 932) might help

update: and optionally remove .item-link { float:left; }
Title: Re: version 2.4
Post by: TEA-Time on September 23, 2018, 05:58:43 PM
If I try to edit a comment, I get the "span" tags with the double quote in between in the edit box like in the attached.
Title: Re: version 2.4
Post by: TEA-Time on September 24, 2018, 04:52:32 AM
I'm not sure if this exists in other templates, but I found an interesting gotcha while messing around with uploading from my phone.  Not that anyone would probably do this under normal operation, but I uploaded a file, deleted it, then reuploaded the same file again to try some other functions, and when I clicked the "Reload page" button after the upload finished, the file was deleted without warning.

If I manually reload the page using the pull-down-to-reload function of the browser instead, it tells me that the form (the delete command, but that might not be obvious to some people) will be resubmitted.  But pressing the "Reload page" button bypasses that warning and.. poof.

So I propose instead of doing an actual reload that resubmits the previous form, the "Reload page" button should just do a regular GET for the folder you're sitting at.

Edit: I just noticed that the corresponding entry in hfs.comments.txt does not get removed when any kind of delete happens.
Title: Re: version 2.4
Post by: TEA-Time on September 24, 2018, 05:32:10 AM
Out of the 8 links in the default "ip-services" list, the 5 below error out, causing HFS to fail to obtain the external IP over half the time.  The first 2 result in a page not found error, whereas the rest appear to have parsing issues due to the pages being redesigned.

Code: [Select]
http://www.melauto.it/public/rejetto/ip.php|
http://rejetto.webfactional.com/hfs/ip.php|
http://www.whatsmyrealip.com/|IP is
http://2ip.ru|d_clip_button">
http://www.mario-online.com/mio_indirizzo_ip.php|<strong>


Even if it fails, it only appears to check every "ip-services-time" (not sure what units that value is in), so I propose it tries each link until it succeeds or runs out of links when there is a failure.
Title: Re: version 2.4
Post by: LeoNeeson on September 24, 2018, 06:46:05 AM
The new template looks awesome, congratulations! :)

And now, after several tests I did on this weekend, the reports...

- New jQuery version broke old browser support
I did quite a few tests, and I've found some problems. Since you have updated jQuery from v1.4.2 to v3.3.1, it's probable that you unintentionally have broke support to a wide range of 'not so old' browsers (remember not all Android phones have the latest browser installed), or you could have used some not wide-compatible code syntax too. I've tried several browsers, and the new template only work OK with the latest browser versions, but on most (two years old) browsers, all have the same problem: those new nice buttons doesn't get displayed at all, including the folder path which doesn't get displayed at all (the webfonts are OK). On a desktop browser, the console shows one 'Uncaught SyntaxError: Unexpected identifier' and two 'Unexpected token'. The most easy way to debug this, is to download a somewhat old version of a Chromium-based browser, like this: SRWare Iron Portable v31 (http://download1.srware.net/old/iron/win/portable/IronPortable_31.zip) (and here (http://download1.srware.net/old/iron/win/portable/) you can also find other versions if you want to do some more tests). Regardless, this new jQuery change gives also some unexpected results if an user wants to use an old template that depends on jQuery v1.X.X (I'll publish those results too, if you are interested).

- Small functional details
If you click on the 'Selection' > 'Mask' button, it will say 'Please enter the file mask to select: *', but there is NO button to accept it. If you are on a desktop browser, you can press the 'Enter' key, but on a mobile device an 'OK' button is needed. This also happens on the 'Move' button, that doesn't have a 'OK'. Another detail: the 'Selection' button is also ONLY displayed, for example, if there is a permission to delete in the current folder, but this is wrong, since if you also have the 'Archive' button visible, you need the 'Selection' button to being able to archive some files (without this, you will archive the whole folder).

- Offline use of 'Font Awesome' icons
Since HFS is often used on private LANs (without external internet connections), it would be best if the 'font-awesome.min.css' (along with the required WebFont files), are included inside the HFS executable (like the jQuery already is). Without that, if you use HFS offline, this new template doesn't work correctly.

- SVG vs WebFonts (or font optimization)
If you want to optimize this (since HFS barely uses 10 icons or less), it would be great if you only use the SVG/PNG format, instead of using the complete 'Font Awesome' WebFont. There are at least two sources of ready-to-use 'Font Awesome' icons in SVG/PNG format, here (https://github.com/ladjs/font-awesome-assets) and here (https://github.com/encharm/Font-Awesome-SVG-PNG). You can also check this (https://css-tricks.com/icon-fonts-vs-svg/) very interesting article describing why SVG is better than using icon WebFonts, but that decision is up to you. Another recommendation is to build a customized (https://github.com/FortAwesome/Font-Awesome/wiki/Customize-Font-Awesome) 'Font Awesome' version, leaving only the icons used by HFS (to reduce it's size), and you could do that easily on Fontello.com (http://fontello.com/).

Well, that's all... :)

Cheers,
Leo.-
Title: Re: version 2.4
Post by: dj on September 25, 2018, 04:20:02 AM
- SVG vs WebFonts (or font optimization)
If you want to optimize this (since HFS barely uses 10 icons or less), it would be great if you only use the SVG/PNG format, instead of using the complete 'Font Awesome' WebFont.

I try to use svg. Incomplete, not IE (must change i tags for IE)
update: v2 should work with IE11 (currently untested)
v3 can resize icons

Quote
New jQuery version broke old browser support
I'm not fit in jquery, but it should support Internet Explorer: 9+ (https://jquery.com/browser-support/)
the some not wide-compatible code syntax is ECMAScript6 (https://caniuse.com/#search=es6)
I also think, IE11 should still be supported
Title: Re: version 2.4
Post by: rejetto on September 30, 2018, 07:48:20 PM
thank you all for your comments, they are very useful!

- New jQuery version broke old browser support
I did quite a few tests, and I've found some problems. Since you have updated jQuery from v1.4.2 to v3.3.1, it's probable that you unintentionally have broke support to a wide range of 'not so old' browsers

the problem is hardly jquery 3. I'm using a recent version of javascript. I'm now experimenting to see if I can do without.
I'm now targeting Chrome 48, from Jan 2016. Quite fair i think.

Quote
Regardless, this new jQuery change gives also some unexpected results if an user wants to use an old template that depends on jQuery v1.X.X (I'll publish those results too, if you are interested).

I opted for version 3 because even 2 is no longer supported/fixed. I will experiment reverting to 2 or even 1.

Quote
on a mobile device an 'OK' button is needed.

you don't have the enter-like button on your mobile keyboard?

Quote
Another detail: the 'Selection' button is also ONLY displayed, for example, if there is a permission to delete in the current folder, but this is wrong, since if you also have the 'Archive' button visible, you need the 'Selection' button to being able to archive some files (without this, you will archive the whole folder).

i'll look into this problem

Quote
- Offline use of 'Font Awesome' icons

it's tiresome, but i guess it's good to support those few who use it in such situation.

Quote
If you want to optimize this (since HFS barely uses 10 icons or less),

I made a list of the current situation, and i'm using 26 icons:
asterisk
ban =deny
bars =menu
check-circle
check-square
clock-o
coffee
download
edit
file-archive-o
folder
globe
home
key
lock
quote-left
refresh
retweet
search
star
times-circle
trash
truck
upload
user
user-circle

Also, I checked the size of the downloaded files in Chrome, and it's 100KB.
I tried 'fontello', and it took me almost an hour but now it's 10KB (woff format only).

Quote
You can also check this (https://css-tricks.com/icon-fonts-vs-svg/) very interesting article describing why SVG is better than using icon WebFonts, but that decision is up to you.

Interesting. Anyway, i tried the svg with Chrome 48 and it's not supported. For the moment the woff seems a good choice.
Title: Re: version 2.4
Post by: rejetto on September 30, 2018, 08:13:30 PM
Out of the 8 links in the default "ip-services" list, the 5 below error out, causing HFS to fail to obtain the external IP over half the time.

just updated the list
Title: Re: version 2.4
Post by: rejetto on September 30, 2018, 08:31:09 PM
If I try to edit a comment, I get the "span" tags with the double quote in between in the edit box like in the attached.

fixed in next release
Title: Re: version 2.4
Post by: rejetto on September 30, 2018, 10:24:43 PM
So I propose instead of doing an actual reload that resubmits the previous form, the "Reload page" button should just do a regular GET for the folder you're sitting at.

Edit: I just noticed that the corresponding entry in hfs.comments.txt does not get removed when any kind of delete happens.

thanks for reporting these problems. Should be fixed in next release.
Title: Re: version 2.4
Post by: rejetto on September 30, 2018, 10:29:55 PM
give a look at this new release
http://www.rejetto.com/hfsfiles/hfs24b2.exe

jquery 3 and selection problem remain to be addressed, for next release.
Title: Re: version 2.4
Post by: rejetto on October 01, 2018, 02:46:02 PM
jquery 1.12 seems to be enough for this template, so i'll try to embed that instead.

I've found that the 'move' functionality was broken in beta2.
You'll find attached the template for the next beta.


Title: Re: version 2.4
Post by: dj on October 02, 2018, 04:21:14 PM
I try to replace the script in the menu panel with css  // leave space for the fixed panel

delete the script
replace #menu-panel { position: fixed;
with      #menu-panel { position: -webkit-sticky; position: sticky; margin-bottom:5px;
Title: Re: version 2.4
Post by: LeoNeeson on October 04, 2018, 07:35:28 AM
replace #menu-panel { position: fixed;
with      #menu-panel { position: -webkit-sticky; position: sticky; margin-bottom:5px;
Thanks for this DJ! I've added that in the template I've attached to this post (with some extra (https://daker.me/2013/07/how-to-make-sticky-elements-using-position-sticky.html) code to have a wide open browser support, as it's recommended here (https://developers.google.com/web/updates/2012/08/Stick-your-landings-position-sticky-lands-in-WebKit) (I've used 0.3em instead of 5px as margin-bottom), as follows:

Code: [Select]
#menu-panel { position: -webkit-sticky; position: -moz-sticky; position: -ms-sticky; position: -o-sticky; position: sticky; margin-bottom:0.3em;
...and now: BIG NEWS! :)

After almost 5 hours of hard work (I'm not exaggerating), I did get rid of almost every incompatible JavaScript expression (ECMAScript 6), with the help of the BabelJS.io (https://babeljs.io/) tool (this tool lets you put in next-gen JavaScript, and get browser-compatible JavaScript out). It was a truly VERY painful job, because I've translated every function, one by one at hand, checking every line very carefully (just to be sure nothing goes wrong). So we now have a very compatible template that works beautifully! :D

I hope you like it (Rejetto: you can EASILY see and review all the changes HERE (https://rejetto.com/forum/index.php?action=dlattach;topic=13060.0;attach=8820) or using this online tool (https://www.diffnow.com/)). Please report if something is not working.

jquery 1.12 seems to be enough for this template, so i'll try to embed that instead.
Remember that the latest stable version of HFS (v2.3m), used jQuery version 1.4.2 and not 1.12.

Another thing: you've used .on( method of jQuery, which was implemented on jQuery v1.7, so, before switching back to jQuery v1.4.2 (http://ajax.googleapis.com/ajax/libs/jquery/1.4.2/jquery.min.js) you will either have to modify any function using it, or use this (https://gist.github.com/drublic/3457c27e8681a34b75d9) tiny polyfill (found here (https://github.com/twbs/bootstrap/issues/2018), not tested). Also the .split( method of jQuery may not be found on jQuery v1.4.2 (but I'm not sure about that).
Title: Re: version 2.4
Post by: dj on October 04, 2018, 11:44:19 AM
...and now: BIG NEWS! :)

Quickly tested. Also works with IE11 except for the More Options buttons.

Replace in css display:initial with display:inline-block (line326 in Leos tpl) and it works also with IE11
Title: Re: version 2.4
Post by: rejetto on October 04, 2018, 09:22:33 PM
Quote
After almost 5 hours of hard work (I'm not exaggerating), I did get rid of almost every incompatible JavaScript expression (ECMAScript 6),

why did you do this?  :o
i don't think it should support every browser.
what browser are you trying to support?

Quote
Remember that the latest stable version of HFS (v2.3m), used jQuery version 1.4.2 and not 1.12.

it's ok if we get compatibility with other templates, it's not necessary to use the same jquery version.

Quote
Also the .split( method of jQuery may not be found on jQuery v1.4.2 (but I'm not sure about that).

split is a String method, not jquery
Title: Re: version 2.4
Post by: rejetto on October 04, 2018, 11:17:21 PM
www.rejetto.com/hfsfiles/hfs24b3.exe

check it out!

i quickly tested the templates Live and ToG, for jquery 1.12 compatibility, and they look good.

here is a preview of the sources, mars asked
https://takeafile.com/?f=wawubukeno
Title: Re: version 2.4
Post by: dj on October 05, 2018, 02:48:23 AM
the checkboxes are broken (http://rejetto.com/forum/index.php?topic=13060.msg1064021#msg1064021) again
Title: Re: version 2.4
Post by: LeoNeeson on October 05, 2018, 07:25:29 AM
why did you do this?  :o
i don't think it should support every browser.
what browser are you trying to support?
I use SRWare Iron 31 (Portable edition) (http://download1.srware.net/old/iron/win/portable/IronPortable_31.zip), because it's the fastest browser on Windows XP (other browsers are very slow and take a lot of resources). If you don't care about me, don't do this for me, but think that the same Chromium engine is used by default on many Android's browsers (Android 4.0.3 to 4.4), and not everybody have Android 6 to 8 with the latest browsers.

www.rejetto.com/hfsfiles/hfs24b3.exe

check it out!
I see you didn't used the most important parts of my changed template :'( (after all the time it took me to do it). I really can't understand why you insist on using 'ECMAScript 6'. For as much web developers try to make (aka 'force') people upgrade your browser, there are many situations (like old hardware) where it's impossible to always have the latest browser and operating system. On my case, Windows 7 is too slow for old hardware, and the support for new browsers on Windows XP is every day more limited.
> EDIT: I guess you don't care about users like me, the same as more and more people everyday care less for the others (we have the 'global warm' just because that attitude of 'not caring about his neighbor'). OK, don't worry, I accept you don't want to get rid of 'ECMAScript 6', after all you are the boss (è solo fa male l'atteggiamento). [Sorry, but I had a bad day, and now I come to the forum and find this, completing my bad day]
> EDIT2: @Rejetto: È così che mi ringrazi?! Mi sforzo, ti aiuto a segnalare errori e tu mi ignori in questo modo? Mi rende molto triste, veramente...


Bugs found on HFS v2.4 (using the default original template)

- Possible bug found #1: add a real folder, then on root set 'access' and 'delete' to anyone, then try to rename that folder trough the browser, and you will see that the folder it's now NOT shared anymore (the folder is renamed OK on the server's hard disk, but it's not updated on the HFS VFS).

- Possible bug found #2: add a real folder (with several sub-folders inside), give to those inside sub-folders a different comment to each folder. Now rename one of those folders, and the comment for that folder is gone (what really happens?: when HFS rename that folder, it doesn't update the folder name in the 'descript.ion' file, so the comment doesn't get displayed). The same happens if you rename a file that had a comment.

- Here (http://rejetto.com/forum/index.php?topic=13060.msg1064023#msg1064023) is another different bug report about comments.

- Design issue?: if you give a comment to a folder, that comment gets displayed next to the folder path, without any design or indication that it's a comment. Here (https://i.imgur.com/MQpzJ8g.png) is a screenshot of what I'm talking about.



I'm attaching (https://rejetto.com/forum/index.php?action=dlattach;topic=13060.0;attach=8821) to this post an updated template, adding the changes from your latest template of HFS v2.4 Build 3 (hfs24b3.exe) [you can see here (https://rejetto.com/forum/index.php?action=dlattach;topic=13060.0;attach=8822) the differences from the template of Build 3], just in case you care about it.
 
Title: Re: version 2.4
Post by: Mars on October 05, 2018, 01:24:13 PM
Quote
bug report about comments.
recurring problem since very old versions
the management of the comments is delicate because it involves 3 different types of storage
- "menu > VFS > Load single comment file"
- "menu > VFS > Support DESCRIPT.ION"
- hfs.comments file

the last template still contains an old usage of adding comments instead of using the appropriate macros
{.if| {.length|%user%.} |{: {.append| %folder-resource%\hfs.comments.txt |{.filename|%item-resource%.}=uploaded by %user%
/append.} :}/if.}


the correct formula would be preferably the following because it automatically manages the storage of information:
{.if| {.length|%user%.} |{: {.set  item| %item-name% |comment={.filename|%item-resource%.}=uploaded by %user%
/set.} :}/if.}



Quote
- Possible bug found #1 & #2

that's another concern I've reported to rejetto recently,

several solutions are possible, but in the immediate future the safest solution is not to allow such actions on a VFS element

remotely modify the vfs may conflict with the option "Save options >> Autosave Options"






Title: Re: version 2.4
Post by: rejetto on October 05, 2018, 08:18:18 PM
the checkboxes are broken (http://rejetto.com/forum/index.php?topic=13060.msg1064021#msg1064021) again

i never met the problem before now. I studied it and have found the problem: it's the transparent border in the <a>.
I will remove it.
Title: Re: version 2.4
Post by: LeoNeeson on October 06, 2018, 03:56:47 AM
@Mars/Anyone: Do you know why if you use the new template on the old HFS v2.3m, it doesn't work properly? (even using a modern browser). I don't see the reason why it could fail, since all the required files are self-contained on the template.

It gives the following error in the browser's console:

Quote
TypeError: newQ(...).on is not a function.
Quote
TypeError: $('body').on is not a function. (In '$('body').on', '$('body').on' is undefined)

(http://rejetto.com/forum/Themes/default/images/post/question.gif) I'm curious to know why it fails...
Title: Re: version 2.4
Post by: rejetto on October 06, 2018, 10:20:31 AM
but think that the same Chromium engine is used by default on many Android's browsers (Android 4.0.3 to 4.4), and not everybody have Android 6 to 8 with the latest browsers.

so you suggest to support android 4 system browser.
Anyway, we don't have to have all the features working on old browsers. Browsing and downloading would be a basic set. Browsing means also login. I guess that a very old hardware won't be the "main" system, and just need to access to files remotely. What you think?
Does your template work well with that browser (Chrome 31) ?

Quote
I see you didn't used the most important parts of my changed template :'(

i'm still considering.

Quote
I really can't understand why you insist on using 'ECMAScript 6'.

because older javascript is a boring Language. Even when i was using it i had made a function myself trying to emulate the => arrow functions.

Quote
> EDIT: I guess you don't care about users like me, the same as more and more people everyday care less for the others (we have the 'global warm' just because that attitude of 'not caring about his neighbor').

don't rush, my dear  :) Opinions are mostly made by experience. Everybody's experience is limited and doesn't always produce the best of results. I don't see any XP in my life, these days, so you can understand my starting point. I'm just not doing something before understanding why, still discussing, exchanging views and information.

Quote
OK, don't worry, I accept you don't want to get rid of 'ECMAScript 6', after all you are the boss (è solo fa male l'atteggiamento). [Sorry, but I had a bad day, and now I come to the forum and find this, completing my bad day]
> EDIT2: @Rejetto: È così che mi ringrazi?! Mi sforzo, ti aiuto a segnalare errori e tu mi ignori in questo modo? Mi rende molto triste, veramente...

look, i'm sorry you are feeling this way.  :( I think it's the result of coming to conclusions too fast. I didn't decide about this thing, but that's exactly why i didn't include your work yet. I'm discussing it, to understand and try to make a decision.
I thank you for your hard work, i undertand it was very long, but the decision should not be made on the number of hours. Do you agree with this?
Title: Re: version 2.4
Post by: Mars on October 06, 2018, 10:24:31 AM
in all likelihood, this would come from an incompatibility of versions of javascript used as a resource by hfs: 1.4.2 in v2.3m, 1.12.x in v2.4
Title: Re: version 2.4
Post by: rejetto on October 06, 2018, 11:51:03 AM
@Mars/Anyone: Do you know why if you use the new template on the old HFS v2.3m, it doesn't work properly? (even using a modern browser). I don't see the reason why it could fail, since all the required files are self-contained on the template.

i think it's because 2.4 fixed some bugs related to the template loading.
Title: Re: version 2.4
Post by: Mars on October 06, 2018, 01:36:27 PM
why not consider having both versions of jquery.min.js in the exe as resources and somehow make a detection of the new template related to hfs 2.4 and utlser the most appropriate jquery to keep a compatibility


  if data.urlvars.values['mode'] = 'jquery' then
    begin
    if tplversion<'T4' then replyWithString(getRes('jquery'))
         else replyWithString(getRes('jquery2'));
    conn.reply.contentType:='text/javascript';
    exit;
    end;

it would be possible to add a specific instruction in the new templates in order to allow the differentiation of jquery to use
Title: Re: version 2.4
Post by: rejetto on October 06, 2018, 01:43:21 PM
Quote
- Possible bug found #1:
- Possible bug found #2:

thanks for reporting. These bug are of the template itself. I will work on them asap.

Quote
- Design issue?: if you give a comment to a folder,

i had actually not worked on this part yet. Now fixed.

What kind of use you make of your XP computer? Is it a media server? a secondary  computer? a primary computer?
Title: Re: version 2.4
Post by: rejetto on October 06, 2018, 01:45:43 PM
why not consider having both versions of jquery.min.js in the exe as resources and somehow make a detection of the new template related to hfs 2.4 and utlser the most appropriate jquery to keep a compatibility

to get what? maybe you didn't read my previous  post. I tested some old templates and they were compatible with newer jquery.
Let me know if you find some template that is not compatible.
Title: Re: version 2.4
Post by: Mars on October 06, 2018, 01:56:23 PM
I do not know what to answer, for now with hfs 2.4 and the template of 2.3, i can not check some files but COMMENT button does not work

maybe,  am I in special conditions?

I confirm, the good functioning depends on the version jquery in resource

 I updated data.rc  as it

Quote
1 24 "WindowsXP.manifest"
defaultTpl TEXT default.tpl
copyright TEXT copyright.txt
dmBrowserTpl TEXT dmBrowser.tpl
invertban TEXT invertban.txt
filelistTpl TEXT filelist.tpl
uploadDisabled TEXT upload_disabled.txt
uploadHowTo TEXT upload_how.txt
alias TEXT alias.txt
shell GIF shell.gif
IPservices TEXT ipservices.txt
jquery TEXT jquery.min.js
jquery2 TEXT jquery2.min.js

added in main.pas

Quote
  if data.urlvars.values['mode'] = 'jquery2' then
    begin
    replyWithString(getRes('jquery2'));
    conn.reply.contentType:='text/javascript';
    exit;
    end;

in the new template

Quote
[]
{.$common-head.}
   <title>{.!HFS.} %folder%</title>
    <script type="text/javascript" src="/?mode=jquery2"></script>
   <style class='trash-me'>
   .onlyscript, button[onclick] { display:none; }
   </style>

I put in the directory of exe two tpls named hfs23.tpl and hfs24.tpl corresponding to the versions to be tested, which allows from the menu to quickly change used template and refreh the web page

working perfectly, and the compatibility with the old templates remains whole ;D

 use this link to load a demo version  (https://takeafile.com/?f=kayeyosuni)






Title: Re: version 2.4
Post by: rejetto on October 06, 2018, 03:44:27 PM
I do not know what to answer, for now with hfs 2.4 and the template of 2.3, i can not check some files but COMMENT button does not work
maybe,  am I in special conditions?
I confirm, the good functioning depends on the version jquery in resource

good finding, thanks.
Compatibility with template "TR2" is important, as a lot of people have a "customized template" and would have problems updating to HFS 2.4 .
I think i have a better solution than keeping 2 different libraries: I'm appending to jquery file the "jquery-migrate-1.4.1.min.js" that will give compatibility with just 10KB (4 when gzipped). I just tested it for the "comment" feature, and it fixed it.
Title: Re: version 2.4
Post by: Mars on October 06, 2018, 03:50:13 PM
with just 10KB (4 when gzipped)

women often say that it is not the size that is the most important, but the quality of the "work" and the pleasure it gives  ;D

can you load the demo version that I put and test the addition of userCFG and see result in hfs.ini  ;)

user data can be defined with SET INI and GET INI for use in a template, and managed from the options dialog (F7 last tab)

these data can be used as persistent variables between two hfs launches
Title: Re: version 2.4
Post by: dj on October 06, 2018, 04:16:13 PM
you are discussing about different versions of jquery.

jQuery was born many years ago when browser implementations were very different, and much of the advanced functionality (like querySelectorAll) was not implemented. Now with the new javascript, there's very little reason for jQuery.

why not use the change and write the template without jquery?
For the old templates you can leave jquery in hfs or change one line in the old template.

If I have (5 hours) time, I will try to rewrite the template without jquery.
Title: Re: version 2.4
Post by: Mars on October 06, 2018, 04:29:50 PM
@rejetto

discovery new bug with comment  hfs 2.3 and 2.4 are same

comments in hfs.comment.txt or descript.ion  seem saved trough "htmlEconde" function

    case s of
      '&': p:='&amp;';
      '<': p:='&lt;';
      '>': p:='&gt;';
      '"': p:='&quot;';
      '''': p:='&#039;';
      else p:=s;
      end;

test.txt=there is a &quot;bug&quot;\nwith comments\ntext is saved as\nhtml code \nas with &quot;htmlencode&quot; function



when used multilines in comments, lines are serapated wit '\n'  (that is correct), but when editing
comment with web page \n are converted as <br> and appear on screen

when comments are saved in descript.ion, each line of comment is ended with code 0x04,0xC,x0D,x0A

Quote
Here, I mean that the descript.ion file contains a backslash and a letter 'n' where I typed a line break, and two special characters 04 C2 at the end of the comment. In addition, the line is ended by a Windows line break 0D 0A.

Apparently, the two extra characters at the end of the line signal the end of a multiline comment. If I remove them, the comment is rendered as a single line in the GUI, and the '\n' sequences are displayed literally.
Title: Re: version 2.4
Post by: rejetto on October 06, 2018, 05:32:25 PM
with just 10KB (4 when gzipped)
women often say that it is not the size that is the most important, but the quality of the "work" and the pleasure it gives  ;D

that you mention is not the gain, but the price to pay. The gain is that we have 1 library ;)

Quote
user data can be defined with SET INI and GET INI for use in a template, and managed from the options dialog (F7 last tab)

Sorry mars but i will consider that feature later.
At the moment my priorities are the new template and bugs.
I'm struggling to find the time to make this new version.
Title: Re: version 2.4
Post by: rejetto on October 06, 2018, 05:40:23 PM
why not use the change and write the template without jquery?

Well, we are targeting some old browser too.
By the way, i've decided to restrict to ES5. (LEO read here, this concerns you)
I've also just noticed that these resources (jquery included) are not correctly cached. I'll have to fix this too. :(
Title: Re: version 2.4
Post by: LeoNeeson on October 06, 2018, 05:53:49 PM
First of all, sorry for my misplaced comment :-[ (I felt bad yesterday, and I thought you will never notice my hidden comment). I'm glad to see that you are open minded and that you appreciate my time, and there is no hurry (take your time to make the decision). My initial worries were that you could have been using 'ECMAScript 6' on purpose, as part of the massive 'browser planned obsolescence (https://www.google.com/search?q=browser+planned+obsolescence&sa=N)' there is currently going on (but now after reading your comments I notice that I was wrong, so, sorry again for my comment). Feel free to take whatever decision you think is best. :)

What kind of use you make of your XP computer? Is it a media server? a secondary  computer? a primary computer?
One of them I used them a media server on WinXP (for local sharing files, offline), but I also have one WinXP online (with a firewall) as one of my primary computers (that PC is used as client of the offline PC), and then I have another much faster PC only for gaming and specific applications, loaded with Win7 (but it's mostly turned off, because it's too noisy to have it on 24/7, and mostly offline).

Does your template work well with that browser (Chrome 31) ?
It does! :D it works perfect not only with Chrome 31 but even with Chrome 19! (as well it works fine with Firefox 24, Firefox 31, K-Meleon Web Browser 75, and others I've tested). Of course it doesn't work with Firefox 3.x or IE6, but nobody use them nowadays.

don't rush, my dear  :) Opinions are mostly made by experience. Everybody's experience is limited and doesn't always produce the best of results. [...] I think it's the result of coming to conclusions too fast.
I agree, and it's something I work every day on being better. I'm somewhat "passional", in the meaning of sometimes I take decisions 'being carried away by feelings or thoughts, without rationally thinking about the consequences' (I took definition from the dictionary).

because older javascript is a boring Language. Even when i was using it i had made a function myself trying to emulate the => arrow functions.
That explains all! :) and it was my mistake thinking it was on purpose (part of the 'planned obsolescence (https://en.wikipedia.org/wiki/Planned_obsolescence)' that is currently happening on many things). Sorry for my mistake.

I thank you for your hard work, i undertand it was very long, but the decision should not be made on the number of hours. Do you agree with this?
Absolutely! Feel free to take whatever decision you think is best. :)

I have a possible solution to all this: perhaps this could be fixed doing a small routine, that detects if the browser doesn't support "the => arrow functions" (or other 'ECMAScript 6' expressions), and then it 'dynamically translates' those functions/expressions with older JavaScript code. Doing this, you only do that routine once, and you don't have to change every function. What do you think?... :)
Title: Re: version 2.4
Post by: Mars on October 06, 2018, 07:21:22 PM
Quote
when used multilines in comments, lines are serapated wit '\n'  (that is correct), but when editing
comment with web page \n are converted as <br> and appear on screen

same bug  in previous versions,  the offender is discovered, it is necessary to correct the setcomment function of the template

        function setComment() {
            var value = it.find('.comment-text').html() || ''    // change .html() by .text()
            ask(this.innerHTML, {type:'textarea', value}, s=>
                s===value || ajax('comment', { text:s, files:name }))
        }//setComment

Quote
test.txt=there is a &quot;bug&quot;\nwith comments\ntext is saved as\nhtml code \nas with &quot;htmlencode&quot; function
failing to do better: , this patch provides a temporary solution

procedure Tfile.setDynamicComment(cmt:string);
var
  s, path, name: string;
  i: integer;
begin
//as a possible function htmlDecode(s:string):string
cmt:=xtpl(cmt,[
      '&amp;', '&',
      '&lt;', '<',
      '&gt;', '>',
      '&quot;','"',
      '&#039;',''''
      ]);
if not isTemp() then
  begin
  comment:=cmt; // quite easy
  exit;
  end;
....

although in reality the previous conversion is not necessary, there is the problem of the codification of the languages of certain languages and their restitution.

an example is better than all explanations

how to define  comments in Russian characters, store them, and restore them correctly to the web page

this comment как возможная функция saved as %EF%B7%90%D0%BA%D0%B0%D0%BA %D0%B2%D0%BE%D0%B7%D0%BC%D0%BE%D0%B6%D0%BD%D0%B0%D1%8F %D1%84%D1%83%D0%BD%D0%BA%D1%86%D0%B8%D1%8F%EF%B7%AF in comments file

the question is how to force the creation of comment files in UTF8 and not in ANSI by default

Title: Re: version 2.4
Post by: rejetto on October 07, 2018, 10:17:55 AM
First of all, sorry for my misplaced comment :-[

don't worry at all, i perfectly understood the situation and took absolutely no offense or whatever.
Just as a side comment on the planned obsolescence, that's a very complex matter and i cannot discuss it extensively, i want just to say that the new version of the programming language has little to do with it. Each p.language, the abstract concept, have always evolved without requiring any faster computer by itself. So ES6 doesn't require a faster PC by itself. Of course this is entangled in the process of obsolescence, but it's just a pawn. We can have better software with old computers, within limits. Then other factors occur causing the obsolescence we know. While some of these factors are malicious, my opinion is that most are not.

Quote
I have a possible solution to all this: perhaps this could be fixed doing a small routine, that detects if the browser doesn't support "the => arrow functions" (or other 'ECMAScript 6' expressions), and then it 'dynamically translates' those functions/expressions with older JavaScript code. Doing this, you only do that routine once, and you don't have to change every function. What do you think?... :)

Sadly the translation is not as trivial as it seems. The software to make the conversion is big, and in general the effort is not worth the result.
Title: Re: version 2.4
Post by: rejetto on October 07, 2018, 01:05:16 PM
same bug  in previous versions,  the offender is discovered, it is necessary to correct the setcomment function of the template

        function setComment() {
            var value = it.find('.comment-text').html() || ''    // change .html() by .text()

thanks, the solution becomes complete simply by removing the {.encode html.} in the [ajax.comment]
Title: Re: version 2.4
Post by: rejetto on October 07, 2018, 01:28:36 PM
- Possible bug found #1: add a real folder, then on root set 'access' and 'delete' to anyone, then try to rename that folder trough the browser, and you will see that the - Possible bug found #2: add a real folder (with several sub-folders inside), give to those inside sub-folders a different comment to each folder. Now rename one of those
- Design issue?: if you give a comment to a folder, that comment gets displayed next to the folder path, without any design or indication that it's a comment.

all of these should be fixed in next release.

Quote
- Here (http://rejetto.com/forum/index.php?topic=13060.msg1064023#msg1064023) is another different bug report about comments.
this was already fixed, AFAIK
Title: Re: version 2.4
Post by: rejetto on October 07, 2018, 02:47:49 PM
http://www.rejetto.com/hfsfiles/hfs24b4.exe

give it a try
Title: Re: version 2.4
Post by: Mars on October 07, 2018, 05:54:09 PM
     {.set item|{.force ansi|%folder%{.^fn.}.}|comment={.force ansi|{.postvar|text.}.}.}

not working:   the tag <BR> always appears when editing the comment

however, what I proposed works despite everything

var value = it.find('.comment-text').text() || ''

Title: Re: version 2.4
Post by: Mars on October 07, 2018, 06:03:34 PM
hfs24b4.exe
give it a try

there is no menu to login or other action on web page, as if new template use the old JQUERY 1.4.2
Title: Re: version 2.4
Post by: rejetto on October 07, 2018, 08:24:28 PM
not working:   the tag <BR> always appears when editing the comment

strange, i just tested comments with multiple lines, and no problems at all.
Are you sure you are using the default template with no customization.
Do other people have problems?

when comments are saved in descript.ion, each line of comment is ended with code 0x04,0xC,x0D,x0A

yes, you can find a function explicitly adding the bytes, escapeIon()
Title: Re: version 2.4
Post by: Mars on October 07, 2018, 09:38:10 PM
Are you sure you are using the default template with no customization.
Do other people have problems?


Sorry, i was using the new template of  24b3 as external template new 24b4

yes, you can find a function explicitly adding the bytes, escapeIon()

that's what I understood but too late


EDIT:
@rejetto
do you think you can make a drop-down menu for the "more options" button instead of a central popup, as in the following image?


EDIT: hfs24b4 with internal template
I am again confronted with a reported problem: check boxes checked have the same effect as clicking the name of the associated item

EDIT:
a really strange thing, with the last exe in two different directories, on one the upload progress works but not the other, I realized that in the one that works I had a default.tpl while in the other I did not have it, I ended up comparing the default.tps of the first with the internal hfs.tpl of the second one, and I realized that réjetto had made a dumpling (fluff, stupidity)

by changing "upload-panel" by "upload panel" he had thus transformed $('#upload-panel') into $('#upload panel') leaving <div id="upload-panel"

EDIT:
@rejetto
do you think you can make a drop-down menu for the "more options" button instead of a central popup, as in the following image?
Title: Re: version 2.4
Post by: LeoNeeson on October 08, 2018, 12:02:23 AM
http://www.rejetto.com/hfsfiles/hfs24b4.exe

give it a try
I can't believe it! YOU DID IT! (https://i.imgur.com/Rr0Mejs.gif)
I had forgotten how amazing you are! :D

Now it works great from Chrome v19.0 to Chrome v31.0 (and beyond).

I am again confronted with a reported problem: check boxes checked have the same effect as clicking the name of the associated item
Changing:
Code: [Select]
.item-link { float:left; }
To:
Code: [Select]
.item-link { float:none; }
...seems to solve the problem. (https://www.w3schools.com/cssref/pr_class_float.asp)

by changing "upload-panel" by "upload panel" he had thus transformed $('#upload-panel') into $('#upload panel') leaving <div id="upload-panel"
I did notice that when I was doing a comparison between 'Build 2' vs 'Build 3' with DiffNow, but I didn't say anything because it thought it was a proper fix. By the way, with the last edition, he deleted this line (I don't know is has any effect, because it seems to work OK without that):
Code: [Select]
<link rel="stylesheet" href="/?mode=section&id=icons.css" type="text/css">
do you think you can make a drop-down menu for the "more options" button instead of a central popup?
That could be nice, but (IMHO) only as an option for desktop browsers, because I think on small screen devices, that could lead to usability problems (like clicking by mistake outside the menu and having to start again). Perhaps this can be done using only CSS, but distinguishing between mobile vs desktop by its size is not easy nowadays, since new devices have big screen resolution. As an idea, maybe a new 'mobile' icon can be added (along with the 'lightbulb'), to switch between mobile and desktop, so in the default mobile theme we can have the current modalbox, and in the desktop theme the dropdown proposed by Mars. But that's in the case Rejetto is interested on this.

how to define  comments in Russian characters, store them, and restore them correctly to the web page

this comment как возможная функция
On my PC (using the Build 4), that comment gets stored and displayed as:
Code: [Select]
??? ????????? ???????
...and my browser can display russian characters without problems. Perhaps unicode comments could be stored on Base64 (using `atob()` and `btoa()`). Just an idea...



In the dark theme, the 'foldercomment' needs his own CSS code (new line to add marked in red, and I've used a slightly different color to not to be confused with a file comment):
Quote
body.dark-theme .item .comment { background-color:#444; color:#888; }
body.dark-theme #foldercomment { background-color:#333; color:#999; }

I did a quick test, and it seems all the comment problems I've reported, are solved now (that's great!). :)
Title: Re: version 2.4
Post by: dj on October 08, 2018, 03:19:03 AM
works also with IE11

Quote from: Mars on October 07, 2018, 09:38:10 PM
I am again confronted with a reported problem: check boxes checked have the same effect as clicking the name of the associated item

Changing:
Code: [Select]
.item-link { float:left; }
To:
Code: [Select]
.item-link { float:none; }...seems to solve the problem.

This alone don't solve the problem just as a non transparent border (http://rejetto.com/forum/index.php?topic=13060.msg1064048#msg1064048) don't.
The checkboxes must be inserted before the link as already reported (http://rejetto.com/forum/index.php?topic=13060.msg1064021#msg1064021) (line 952 in the new tpl)

Quote
do you think you can make a drop-down menu for the "more options"
not a drop-down, but always vertical
Code: [Select]
[+common-head]
<style>
.item-menu-dialog {width: initial}
.item-menu-dialog .pure-button {display:block}
</style>
Title: Re: version 2.4
Post by: rejetto on October 14, 2018, 10:12:33 PM
do you think you can make a drop-down menu for the "more options" button instead of a central popup, as in the following image?

it would take time and i see no real gain

Quote
I am again confronted with a reported problem: check boxes checked have the same effect as clicking the name of the associated item

ok, i hope this time is fixed for real

Quote
by changing "upload-panel" by "upload panel" he had thus transformed $('#upload-panel') into $('#upload panel') leaving <div id="upload-panel"

fixed, thanks
Title: Re: version 2.4
Post by: rejetto on October 14, 2018, 10:23:50 PM
Perhaps unicode comments could be stored on Base64 (using `atob()` and `btoa()`).

that would break the descript.ion standard format, i guess

Quote
In the dark theme, the 'foldercomment' needs his own CSS code (new line to add marked in red, and I've used a slightly different color to not to be confused with a file comment):

thanks i'll use that
Title: Re: version 2.4
Post by: rejetto on October 14, 2018, 10:26:27 PM
i'm also going for the vertical item-menu suggestion

www.rejetto.com/hfsfiles/hfs24b5.exe

did you guys try the new forum template? it supports also dual dark/light mode, cool
Title: Re: version 2.4
Post by: Mars on October 14, 2018, 10:35:28 PM
another bug with comment ?? 

when creating or modifying a comment, the multilines are sent with LF and not with CRLF, therefore the comments of the items of the vfs, as well as those recorded in the ".comment" files are saved in the wrong format and do not appear more correctly in a new edition

the solution found is the following:

Quote
procedure Tfile.setDynamicComment(cmt:string);
var
  s, path, name: string;
  i: integer;
begin
if countSubstr(#13#10,cmt) = 0 then
  cmt:=xtpl(cmt,[#10,#13#10]);

if not isTemp() then
  begin
  comment:=cmt; // quite easy
  exit;
  end;

....
the bug is also effective in version of hfs24b5.exe

another unimportant detail
Quote
    function getAccountRedirect():string;
.....
    // if it's not a complete url, it may require some fixing
    // mod mars 10-2018
    result:=xtpl(result,['\','/']); // if result ansiStart with '\' not necessary to add '/' with  next line of code
    if not ansiStartsStr('/', result) then result:='/'+result;
    end; // getAccountRedirect
Title: Details pending for fixing on v2.4 Beta
Post by: LeoNeeson on October 23, 2018, 07:16:04 AM
Details pending for fixing on v2.4 Beta:

- jQuery: As Mars (http://rejetto.com/forum/index.php?action=profile;u=2913) also reported here (http://rejetto.com/forum/index.php?topic=13060.msg1064051#msg1064051) and also dj (http://rejetto.com/forum/index.php?action=profile;u=32265) commented here (http://rejetto.com/forum/index.php?topic=13060.msg1064059#msg1064059), and after several tests I did on this weekend, I've found that there is an important difference between jQuery v1.4.2 (released on 02/2010 (https://blog.jquery.com/2010/02/19/jquery-142-released/)) and v1.12.4 (released on 05/2016 (https://blog.jquery.com/2016/05/20/jquery-1-12-4-and-2-2-4-released/)) [and not only because there are six years of difference!]. If you do a quick test using old templates, everything 'seems' to be OK (on a superficial look), but if you do some deep and detailed tests, you will find that the functionality is NOT 100% the same (as some buttons get broken or either not display, since depend on functions not found or not compatible with jQuery v1.12.4). This is a serious problem, since this will broke many old templates. Possible solution: include v1.4.2 by default (like it was on v2.3m), and embed v1.12.4 as a section on the new template, calling it as: "/?mode=section&id=jquery.js" and then adding on the section [jquery.js|no log|cache] the jQuery v1.12.4 file. So, by default, when an old template needs jQuery (using "/?mode=jquery"), it will use v1.4.2, and everything will work. I think this is the most easy and elegant way to solve this and keep compatibility. Just my 5 cents.

- HFS.lng untranslated strings: As you already may noticed, there are only a few strings that currently can't be translated because the way the code is written (without this, we can't have a complete translation). You can easily spot and find those strings, if you use a dummy 'hfs.lng' translation file that I'm attaching to this post (it translates every string to 'xxxxxxxxxxxx', so, if you can read some text in English, then that text couldn't be translated and need a fix). Switch to expert mode to see all the untranslated menus and also open the VFS's Properties. To help you finding a solution, I've noticed that most 'dynamic' menus have this problem, except one: + Menu -> Limits -> "Connection inactivity timeout" (so the way that code is written, could help you to fix the rest of the code). Keep in touch if there is something I could do to help. :)
Title: Re: version 2.4
Post by: LeoNeeson on October 29, 2018, 10:20:43 PM
@Rejetto: Don't let my last comment stop you from continuing with v2.4 beta testings. I do understand that you want to modernize HFS, and make it (even more) awesome by delivering by default jQuery v1.12.4, so you can forget about this jQuery thing if you are not interested (after all this doesn't affect me directly, since I know how to 'patch' old templates to make them use old jQuery if I need). But please don't forget (if you have a few minutes to spare, on checking how we can translate those untranslated 'dynamic' menus (if you prefer I can open a new thread about this). And I could help you to edit the source code if you can point me out in the right direction on how to re-write those menus to make them translatable. My message doesn't mean I'm requesting a fix to those untranslated menus 'right now' (this could be fixed on future version, like v2.4b, v2.4c or v2.4d), not pressure at all! :)
Title: Re: version 2.4
Post by: rejetto on October 31, 2018, 04:48:33 PM
i just had less time lately, but i'm not stopped ;)
thanks for your tests with jquery. If it's breaking other templates i will consider to just use old version, as i doubt it is worth having both versions.
Can you please tell me what old template and what feature it breaks?
I don't really understand why you mentioned mars and dj in your argument: mars was just guessing, but he was not right, and dj was suggesting to stop using jquery, any version.

hfs.lng: i'll have a look at it. I have to check the documentation of the translator to see how dynamic strings must be handled, i don't know yet.
Title: Re: version 2.4
Post by: Mars on October 31, 2018, 10:03:49 PM
Quote from: rejetto
mars was just guessing, but he was not right
it's understandable, javascript gives me Urticaria on neurons  ;D
Title: Re: version 2.4
Post by: LeoNeeson on November 02, 2018, 08:38:23 AM
hfs.lng: i'll have a look at it. I have to check the documentation of the translator to see how dynamic strings must be handled, i don't know yet.
Oki Doki :) Keep me posted if you have any advance on this...

Can you please tell me what old template and what feature it breaks?
Oh! :-[ I haven't saved the results of my tests, so, I will have to make those tests again on this weekend, at least with the default old template (and report back the results).
Title: Re: version 2.4
Post by: LeoNeeson on November 03, 2018, 09:45:15 PM
Can you please tell me what old template and what feature it breaks?
I'm afraid you won't like to hear this: on modern browsers the functionality seems to be OK, but on old browsers it gets broken (and the problems are too many to list here, but they are mostly buttons or some sections that doesn't showing up). So, summarizing: if some user was using an old template on an old browser, it will be fucked up (will have a very limited functionality), but if he is using a modern browser, will see no difference (everything seems to work fine). So, it's up to you if you keep jQuery v1.12.4 (breaking the compatibility with old browsers), or use jQuery v1.4.2 (maintaining compatibility).
Title: Re: version 2.4
Post by: rejetto on November 04, 2018, 06:47:50 PM
i just tried the old default template with iron31, and these features work correctly:
- change comment
- rename
- archive of selection
- delete
- new folder

don't you remember if the problems were with the default template, or not?
Title: Re: version 2.4
Post by: Mars on November 04, 2018, 07:38:34 PM
 may be wrong but it must be about the compatibility of old default template with the new version of jquery
Title: Re: version 2.4
Post by: LeoNeeson on November 04, 2018, 08:42:37 PM
don't you remember if the problems were with the default template, or not?
may be wrong but it must be about the compatibility of old default template with the new version of jquery
Yes, the problem was about compatibility of old default template (of 2.3m) with the new version of jQuery v1.12.4.



The latest tests I did the other day, were only using the default template of 2.3m. Using Iron 31 worked fine (and anything more modern than that). When I said 'you won't like to hear this' it was because the compatibility problem was with 'really' old browsers, like Firefox 3.x and IE6 (and could be more old browsers).

I could report exactly what it fails (with screenshots), since I have access to those browsers (in case you care about). But those were 'action' buttons not showing up (in Firefox 3.x), and IE6 was working better (this surprised me), only having the file 'selection panel' doesn't showing up. It could be more missing, since I can't remember exactly all now.

I reported this is issue because the default template of v2.3m was working absolutely fine BEFORE with those old browsers (and because some users could be still using and depending on those old browsers), but like I've said on previous posts, this issue particularly doesn't affect me (one because I mostly use Iron 31 which works fine now, and second because I could patch the old template to use jQuery v1.4.2 in case I need). But that doesn't mean this issue still exists, and that it could affect other old custom templates too.



ANOTHER THING: I also discovered that the new template of v2.4 is not 100% compatible with the old jQuery v1.4.2. If my tests are not wrong (at least when using Iron 31), when using the new template with the old jQuery, you will find all the main buttons on the header missing, and the 'more options' menu not responding when clicking on it. So, I have my doubts now if this is worth all the hassle, just to keep jQuery v1.4.2. But if you fix the new template to be working with jQuery v1.4.2, then go ahead with jQuery v1.4.2 (this would be the best solution, if it's not too complicated for you to fix it).

That's why I leave this decision up to you. In my opinion, when you release v2.4 (if you want to keep jQuery v1.12.4), you should at least warn users that old templates could be broken when using a truly old browser. But the best solution (that's my suggestion only), would be fixing the new template of v2.4 to be working with jQuery v1.4.2 (so you could include jQuery v1.4.2 by default).

I'm attaching the new template of v2.4 with old jQuery v1.4.2 embedded on the same file, to make your tests easier (but don't rely on it, since I can't be sure if it's from Beta 4 or 5). You will find it doesn't work as expected...

I hope this helps... ;)
Title: Re: version 2.4
Post by: rejetto on November 04, 2018, 09:45:28 PM
ok, i see.
I don't see a good reason to support IE6 and FF3.
I understand your suggestion to use jquery 1.4 is also in favor of IE6 and FF3.
Title: Re: version 2.4
Post by: rejetto on November 04, 2018, 09:48:10 PM
http://www.rejetto.com/hfsfiles/hfs24b6.exe

Now we have sorting, and much more translatable strings.
Title: Re: version 2.4
Post by: LeoNeeson on November 05, 2018, 07:06:45 AM
ok, i see.
I don't see a good reason to support IE6 and FF3.
I understand your suggestion to use jquery 1.4 is also in favor of IE6 and FF3.
Yeah, it's OK. :) Who cares about IE6 and FF3 anyway? They are not relevant today (I don't use it at all, I just did some tests on it, only because the old template worked fine on it). But if some user needs support for those old browsers, could still use 'The really fast Throwback (retro) template (http://rejetto.com/forum/index.php?topic=12055.0)' or the old legacy template that I'm sharing HERE (https://rejetto.com/forum/index.php?topic=13082.0). After all, this is a brand new version, so, it's OK to drop support for those 'truly very-old browsers'.

http://www.rejetto.com/hfsfiles/hfs24b6.exe

Now we have sorting, and much more translatable strings.
Cool! 8) I'll check it out and do some tests on the next few days...
Title: Re: version 2.4
Post by: MarkV on November 08, 2018, 10:37:16 PM
Hi, just passing by, and I see you are all still busy. HFS has come a long way (I still remember the good old days when it started). No reason to stop it. :)
Title: Re: version 2.4 feature request update - revamp
Post by: bmartino1 on November 09, 2018, 11:23:15 PM
Sorry, been out the past month or so with medical issues... i was going to post some things regardeing testing and other stuff. But for now, id like to add a potential change if applicable to the new beta.

The main idea is adding "a cookie-based login system", so modern mobile browsers could use a form based login system (and not rely on that old and obsolete login popup that most new browsers are now deprecating).

it would be nice to propose new additions (at server level).

I think it would be the perfect time to request having a "cookie-based login system" instead of (using a form based user/pass login).
--This way controls to login in and out can potential be code for a full login / logout function.

Delphi google searches for codes groups and info:

http://coding.derkeiler.com/Archive/Delphi/comp.lang.pascal.delphi.misc/2004-10/0357.html

http://groups.google.com.mx/group/borland.public.delphi.internet.winsock/browse_thread/thread/929c7359d8e321d7/e250f2e8f245ef0e?lnk=st&q=&rnum=1&hl=es#e250f2e8f245ef0e

http://codeverge.com/embarcadero.delphi.tools/how-to-use-indy-10-to-post-to-a-web/1066980

https://www.example-code.com/delphiDll/http_addCookies.asp

https://softwareengineering.stackexchange.com/questions/178105/understanding-http-cookies-in-indy-10-for-delphi-xe2

*essential adding more authentication (things i have seen at the windows server level):
https://www.example-code.com/delphiDll/http_authentication.asp
--(Delphi DLL) HTTP Authentication (Basic, NTLM, Digest, Negotiate/Kerberos)

**********************************
Although a "cookie-based login system" could be done at client level (only using JavaScript and Ajax)
https://www.codementor.io/mayowa.a/how-to-build-a-simple-session-based-authentication-system-with-nodejs-from-scratch-6vn67mcy3

https://stackoverflow.com/questions/42843030/javascript-login-using-cookies

https://www.htmlgoodies.com/beyond/javascript/article.php/3888551/JavaScript-Class-How-Can-I-Set-A-Cookie-Based-On-A-Users-Selection-On-A-Form.htm

https://developer.atlassian.com/cloud/jira/platform/jira-rest-api-cookie-based-authentication/

The main idea is adding "a cookie-based login system", so modern mobile browsers could use a form based login system (and not rely on that old and obsolete login popup that most new browsers are now deprecating).

Title: Re: version 2.4
Post by: Chris Harris on November 10, 2018, 04:41:46 AM
Just installed hfs24b6.exe and the interface is clean and very easy to use.

Thanks
Chris
Title: Re: version 2.4
Post by: Fysack on November 11, 2018, 12:20:13 AM
" latest tests I did the other day, were only using" Strighten up Leo. We can absolutely not use that in season 5 Le Bureau
Title: Re: version 2.4
Post by: LeoNeeson on November 11, 2018, 03:26:26 AM
" latest tests I did the other day, were only using" Strighten up Leo. We can absolutely not use that in season 5 Le Bureau
That's a 'high' post!! (https://i.imgur.com/NSJPeHf.gif)(https://i.imgur.com/bHTRoCG.gif)
Title: Re: version 2.4
Post by: LeoNeeson on November 11, 2018, 03:28:45 AM
http://www.rejetto.com/hfsfiles/hfs24b6.exe

Now we have sorting, and much more translatable strings.
Could you please generate a new base language file (hfs.lng), because I've generated a new one (using kdlscan.exe and lngupdate.exe), but I can't find any new string to translate (and neither any new modification is found). The utility kdlscan.exe mark with "(!)" every untranslated (new & modified) string, but in this case (and after doing every step right), it does't find new or modified strings. Are you sure there are new strings to translate? (because also after doing a 'visual' inspection, I don't see anything new/changed). ???



• Another thing: When using your new template on a desktop PC with a screen resolution of 800x600px, your new 'Sorting' button makes the 'menu-bar' look ugly (like this (http://i.imgur.com/kN3lBab.png)). I guess a small change needs to be done, so if that resolution is used, then show up the short/small mobile buttons instead.

You could solve this, changing 48em to 50em (http://pxtoem.com/):

Quote
@media (min-width: 50em) {
#toggleTs { display: none }
}
@media (max-width: 50em) {
#menu-panel button { padding: .4em .6em; }
.additional-panel button span,
#menu-bar button span { display:none } /* icons only */
#menu-bar i { font-size:120%; } /* bigger icons */
#menu-bar button { width: 3em; max-width:10.7vw; padding: .4em 0; }
.hideTs .item-ts { display:none }
}

...or using another method you want, so when 7 buttons are displayed on a small screen (800x600), the small buttons are shown.



• Another thing more: "developer notes.txt" needs this line to be added:
Code: [Select]
Kryvich's Delphi Localizer https://sites.google.com/site/kryvich/localizer


Sorry, been out the past month or so with medical issues. [...] the main idea is adding "a cookie-based login system", so modern mobile browsers could use a form based login system (and not rely on that old and obsolete login popup that most new browsers are now deprecating).
I hope you are well now, it's nice to see you again. :) I agree, that would be great. After seeing your post I've search on the forum and found out that Rejetto was working some time ago on a similar solution, but then he didn't say anything about it. This (http://rejetto.com/forum/index.php?topic=9699.msg1054351#msg1054351) was that post:

1. cookie authentication is on the way. Sadly i had to stop few months ago because of the lack of time. It's harder than you may think since i'm making it backward compatible, so that current templates will continue to work. It's already quite working but i have to solve some problems. Sharing it with you may be a good idea to speed up the process.
I know that post was -very- old, but may be rejetto has somewhere stored a partial-done code for a 'cookie authentication' system, that he could share it "as is" here, so we can continue developing it. ;)
Title: Re: version 2.4
Post by: LeoNeeson on November 19, 2018, 10:12:05 AM
use this link to load a demo version  (https://takeafile.com/?f=kayeyosuni)
@Mars: This link is down. Could you please upload/share it again?... :)
Title: Re: version 2.4
Post by: Mars on November 19, 2018, 05:03:27 PM
I do not know which version it was, but we can work around the problem more simply to perform compatibility tests (as could have suggest rejetto)

add a JQUERY folder to the vfs by configuring it as HIDE

put the two files jquery.min.js jquery2.min.js inside

change the reference in every template
<script type="text/javascript" src="/?mode=jquery"></script>

towards
<script type="text/javascript" src="/jquery/jquery.min.js"></script>    -->  hfs23.tpl
or
<script type="text/javascript" src="/jquery/jquery2.min.js"></script>  -->  hfs24.tpl

so it is not necessary to make a change in the exe

if that's not enough, come back to me  ;)

Title: Re: version 2.4
Post by: LeoNeeson on November 21, 2018, 08:04:33 AM
if that's not enough, come back to me  ;)
Thank you, I appreciate your reply, but it's much better what you have proposed here (http://rejetto.com/forum/index.php?topic=13060.msg1064053#msg1064053): having automatic 'template revision' detection (without needed to modify any template). So, if TR2 is detected, then use old jQuery of v2.3m, and if TR3 is detected, then use new jQuery of v2.4. Even if I currently don't needed this change, programming is fun.

Since now I can compile HFS by myself (http://rejetto.com/forum/index.php?topic=13088.0), I would like to experiment adding whatever idea comes to my mind, no matter how crazy the idea is. That's the fun of having your own fork/modification of something: experiment. That's why I've asked if you can share again (that (https://takeafile.com/?f=kayeyosuni)) file, but if you don't have it (or whatever), don't worry, it's OK :) (I will try to implement it by myself).
Title: Re: version 2.4
Post by: LeoNeeson on November 21, 2018, 09:04:16 AM
if that's not enough, come back to me  ;)
Thank you, I appreciate your reply, but it's much better what you have proposed here (http://rejetto.com/forum/index.php?topic=13060.msg1064053#msg1064053): having automatic 'template revision' detection (without needed to modify any template). So, if TR2 is detected, then use old jQuery of v2.3m, and if TR3 is detected, then use new jQuery of v2.4. Even if I currently don't needed this change, programming is fun. 8)

Since now I can compile HFS by myself (http://rejetto.com/forum/index.php?topic=13088.0), I would like to experiment adding whatever idea comes to my mind, no matter how crazy the idea is. That's the fun of having your own fork or modification of something: experiment. That's why I've asked if you can share again (that (https://takeafile.com/?f=kayeyosuni)) file, but if you don't have it (or whatever), don't worry, it's OK :) (I will try to implement it by myself).
Title: Re: version 2.4
Post by: Fysack on December 15, 2018, 12:36:01 AM
That's a 'high' post!! (https://i.imgur.com/NSJPeHf.gif)(https://i.imgur.com/bHTRoCG.gif)
;D ;D ;D
Title: Re: version 2.4
Post by: Fysack on December 15, 2018, 01:01:47 AM
https://youtu.be/yo4pmauhugo
Title: Re: version 2.4
Post by: rejetto on January 06, 2019, 07:21:17 PM
Hi, just passing by, and I see you are all still busy. HFS has come a long way (I still remember the good old days when it started). No reason to stop it. :)

hi Mark, nice to see you around :)
Title: Re: version 2.4
Post by: rejetto on January 20, 2019, 05:04:21 PM
The main idea is adding "a cookie-based login system", so modern mobile browsers could use a form based login system (and not rely on that old and obsolete login popup that most new browsers are now deprecating).

such thing was actually ongoing years ago, then with the hiatus i've lost track of the progress.
i won't consider it for this version anyway, as it would mean further delaying in the publishing of an official update.
Title: Re: version 2.4
Post by: rejetto on January 20, 2019, 05:20:09 PM
Could you please generate a new base language file (hfs.lng), because I've generated a new one (using kdlscan.exe and lngupdate.exe), but I can't find any new string to translate (and neither any new modification is found).

it does the same to me. I don't know why it seems to ignore the fact that i've been converting many constants to "resourcestring".
I will investigate it.

Quote
You could solve this, changing 48em to 50em (http://pxtoem.com/):

ok, it seems a quite safe fix to do, so i'll introduce it without testing it myself

Quote
• Another thing more: "developer notes.txt" needs this line to be added:
Code: [Select]
Kryvich's Delphi Localizer https://sites.google.com/site/kryvich/localizer

it is already there
Title: Re: version 2.4
Post by: rejetto on January 20, 2019, 05:29:59 PM
it does the same to me. I don't know why it seems to ignore the fact that i've been converting many constants to "resourcestring".
I will investigate it.

apparently i had to use the "build", so it generated a .drc file, and then kdlscan worked correctly.
You'll find all new strings at the end.
Sorry for keeping you waiting.
Title: Re: version 2.4
Post by: LeoNeeson on January 21, 2019, 07:34:42 AM
@Rejetto: Thanks, I see your changes on [ResourceStrings], but still doesn't seem to work (at least using the latest beta hfs24b6.exe).

Do this simple test:
1) Open with a text editor the 'hfs.lng' file.
2) Do a 'search and replace' of the folowing:
    Search this: "Max simultaneous downloads"
    Change with: "Massimo di download simultanei"
3) Save the changes, and now open HFS to see if it worked...
4) Go to: Menu > Limits and you will see that the text was not translated.

If the text was translated fine (using the latest 'still unreleased' beta build you have), then the next build hfs24b7.exe will make those changes visible. If that string was not translated, then there is something else to fix on the code.

I've noticed that a similar string ("Connections inactivity timeout") is being translated OK, so, perhaps you can use that as reference. Keep in touch if there is something I could do to help you to debug this out. ;)
Title: Re: version 2.4
Post by: rejetto on January 24, 2019, 11:21:40 PM
i tried with the "Switch ON" and it worked, so it's the exe.
try this http://www.rejetto.com/hfsfiles/hfs24rc1.exe
Title: Re: version 2.4
Post by: LeoNeeson on January 26, 2019, 09:26:00 PM
i tried with the "Switch ON" and it worked, so it's the exe.
try this http://www.rejetto.com/hfsfiles/hfs24rc1.exe
Well... I appreciate you effort, but for as much as I would like to say it's solved, it isn't solved... :-\ The "Switch ON" works, but that's not the main problem.

The problem comes with menu items that has a variable string, like "%s". That's why I've proposed you to make a test with the menu option 'Max simultaneous downloads: %s ...', since it has the variable "%s" that causes the string go untranslated.

There are also another buttons, like "You are in Expert mode" that also can't be translated. And the problem with that button is similar, since it's a button that changes its caption when it's clicked.

(If this doesn't have an easy fix, don't worry, leave it as is for the time being until you find a solution)
Title: Re: version 2.4
Post by: rejetto on January 27, 2019, 06:54:04 PM
the problem exists but i didn't find a solution.
The %s is not the problem: "total out" works for example. It must be something else. :-\
The problem is not even the changing, because "switch on" and "switch off" do the same, but work.
I'm sorry, but i'm quite lost.
I'm now leaving for a 3-weeks journey in India and Nepal. I can't say about my availability next days.
Title: Re: version 2.4
Post by: dj on January 28, 2019, 06:05:23 AM
Addons tested with version 2.4rc1:

addoncheckedadjustment
check boxes (http://rejetto.com/forum/index.php?topic=13060.msg1064021#msg1064021)bugfix selection
 
audioplayerlight (http://rejetto.com/forum/index.php?topic=11754.msg1063922#msg1063922)add mime types* (* .mp3 audio/mpeg, .ogg audio/ogg)
logout (http://rejetto.com/forum/index.php?topic=11754.msg1062970#msg1062970)var ref=document.querySelector('button[onclick*=user]')
jpgimgthumbs (http://rejetto.com/forum/index.php?topic=13060.msg1064013#msg1064013)
pictureviewer (http://rejetto.com/forum/index.php?topic=13060.msg1064013#msg1064013)
archive_as_zip (http://rejetto.com/forum/index.php?topic=13060.msg1064013#msg1064013)save or change script
dnd_upload and viewthumbs (http://rejetto.com/forum/index.php?topic=13077.msg1064110#msg1064110)
comment for standard-template (http://rejetto.com/forum/index.php?topic=11754.msg1064076#msg1064076)comment in utf-8
searchcategorie (http://rejetto.com/forum/index.php?topic=13060.msg1064013#msg1064013)search for images, audios,videos

@rejetto, have a nice trip

jpgimgthumbs_v2 can fallback to full-img-thumbs, _v3 works also with progressive jpegs (update)
Title: Re: version 2.4
Post by: LeoNeeson on February 05, 2019, 01:05:09 AM
I'm now leaving for a 3-weeks journey in India and Nepal. I can't say about my availability next days.
@Rejetto: Happy holidays! :)
(https://i.imgur.com/VGhtmwj.png)
Title: Re: version 2.4
Post by: rejetto on February 11, 2019, 02:12:37 PM
thanks. I'm in Nepal right now, i'll be back sunday.
Anyway, i fear that i'll have to publish 2.4 as it is if we don't find a solution to this translation problem.
here attached the latest source changes.
i shall find the time to give a look at dj's stuff.
Title: Re: version 2.4
Post by: rejetto on February 12, 2019, 03:48:32 PM
rc2 only fixes this bug http://rejetto.com/forum/index.php?topic=13112.0

http://www.rejetto.com/hfsfiles/hfs24rc2.exe
Title: Re: version 2.4
Post by: LeoNeeson on February 13, 2019, 04:54:56 AM
Anyway, i fear that i'll have to publish 2.4 as it is if we don't find a solution to this translation problem.
Yeah, for me it's OK :) (feel free to release v2.4 as is)

I can't promise anything, but if I have some spare time, I will try to contact the original developer of 'Kryvich's Delphi Localizer' to see if he could give us some 'light' about this issue. In the meantime, feel free to release v2.4 as is, when you feel it's the right time to do it (we could leave this translation issue for a future version).
Title: Re: version 2.4
Post by: LeoNeeson on February 13, 2019, 04:58:55 AM
Anyway, i fear that i'll have to publish 2.4 as it is if we don't find a solution to this translation problem.
Yeah, for me it's OK :) (feel free to release v2.4 as is)

I can't promise anything, but if I have some spare time, I will try to contact the original developer of 'Kryvich's Delphi Localizer' to see if he could give us some 'light' about this issue. In the meantime, feel free to release it as is (when you feel it's the right time to do it). We could leave this translation issue for a future version.
Title: Re: version 2.4
Post by: rejetto on February 14, 2019, 12:28:13 PM
sometimes you get files with long name and no dash/spaces, and it looks bad especially on mobile with little horizontal space, causing horizontal scrolling.
I want to add this css rule for final 2.4
.item-link { word-break: break-word; }

Please give me some feedback
Title: Re: version 2.4
Post by: LeoNeeson on February 14, 2019, 09:06:14 PM
I want to add this css rule for final 2.4
.item-link { word-break: break-word; }
Yes, that looks OK (at least on my side). Don't forget that the original line had "float:left;"

Old line:
Code: [Select]
.item-link { float:left; }
New line:
Code: [Select]
.item-link { word-break: break-all; float:left; }
Also, I've found that "break-all" could work better that "break-word". The W3 specification (http://www.w3.org/TR/css-text-3/) seem to suggest that "word-break: break-all" is also more compatible with non-english text (like Chinese, Japanese, and Korean), so that would be more unicode compatible, if I'm not mistaken (source (https://stackoverflow.com/a/20985433)). Anyway, that decision is up to you.



Another thing: Don't forget to change 48em to 50em (like you said you would do here (http://rejetto.com/forum/index.php?topic=13060.msg1064317#msg1064317), as I originally I've reported here (http://rejetto.com/forum/index.php?topic=13060.msg1064178#msg1064178)).
Title: Re: version 2.4 - I want to add this css rule
Post by: dj on February 15, 2019, 09:34:26 AM
alternatively truncate string
.item-link { overflow: hidden; text-overflow: ellipsis }
Title: Re: version 2.4 - I want to add this css rule
Post by: LeoNeeson on February 15, 2019, 09:16:55 PM
alternatively truncate string
.item-link { overflow: hidden; text-overflow: ellipsis }
@DJ: That would be nice, but it would also needs adding "white-space: nowrap;". But sadly, your code adds a extra line, one for the file name and another below for the time and file size, even on short file names.

Do some tests, making two kind of long file names (the one without spaces is to recreate non-english languages):

Code: [Select]
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna esse.txt
Code: [Select]
Loremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaesseutenimadminimveniamq.txt
Do some tests and see the results, using "white-space: nowrap;" and not using it (and with and without "float: left;").

Code: [Select]
.item-link { white-space: nowrap; overflow: hidden; text-overflow: ellipsis; }
Code: [Select]
.item-link { white-space: nowrap; overflow: hidden; text-overflow: ellipsis; float: left; }
Adding the "float: left;" we eliminate the "extra line" on short file names, but that makes appear the unwanted horizontal scrolling.

So, what solution do you propose for this, DJ?... ???
Title: Re: version 2.4 - I want to add this css rule
Post by: dj on February 16, 2019, 05:03:23 AM
@Leo: thanks for testing. It was only a untested idea. I've seen it on my android filemanager.

Code: [Select]
.item-link { float: left; overflow: hidden; text-overflow: ellipsis; max-width: 100%}should do it. Tested on chrome and FF.
Title: Re: version 2.4
Post by: LeoNeeson on February 16, 2019, 06:32:35 AM

@dj: Perfect, now it looks good. :) Adding "white-space: nowrap;" is also recommended to support long file names (without spaces):

Code: [Select]
.item-link { float:left; overflow:hidden; text-overflow:ellipsis; white-space:nowrap; max-width:100%; }
(Anyway, Rejetto has the final word if he decides to add this or not, we are just giving out ideas)
Title: Re: version 2.4
Post by: thexfile on February 18, 2019, 02:18:30 AM
There's no place like home. ;D
Title: Re: version 2.4 - a new addon - lazy loading images
Post by: dj on March 02, 2019, 12:10:13 PM
Lazy loading images loads full-img-thumbs at the moment of need (scrolling) to save bandwidth
You can use this, if you have no thumbnails (in images and server). Tested with chrome and FF.

add Well-Controlled Scrolling additionally (chrome and safari).

lazyimages2 prevent load on cellular network (chrome android)

autotheme preselects dark theme at night
Title: Re: version 2.4
Post by: SilentPliz on March 08, 2019, 11:05:35 PM
Hi boss !  ;)

No sources for these RC ?
Title: Re: version 2.4
Post by: LeoNeeson on March 16, 2019, 06:18:15 AM
Hi boss !  ;)

No sources for these RC ?
I also join the petition, but I guess Rejetto is waiting to release the stable version to release the sources. Perhaps he plans to release the final v2.4 stable version soon?... 8)
Title: Re: version 2.4
Post by: rejetto on March 17, 2019, 07:20:55 PM
Also, I've found that "break-all" could work better that "break-word".

i tried both when making the decision. Break-all diminishes readability of file names, with bad and unnecessary breaking.
Title: Re: version 2.4
Post by: rejetto on March 17, 2019, 07:24:14 PM
alternatively truncate string
.item-link { overflow: hidden; text-overflow: ellipsis }

didn't test it, but i don't like the idea of truncating the file names. What when you have a difference at the end of them (a common case): you cannot tell the difference anymore.
Title: Re: version 2.4
Post by: rejetto on March 17, 2019, 07:24:56 PM
No sources for these RC ?

i published the only file changing, it's a bit of extra work for you, i'm sorry :)
Title: Re: version 2.4 - language
Post by: dj on March 18, 2019, 04:23:44 AM
while testing languages-addon (http://rejetto.com/forum/index.php?topic=11754.msg1064456#msg1064456), i found some strings in hfs.tpl not translated:
"Search" (line 105) and "Delete" (line 154) in hfs24rc2
Title: Changes needed in default template for v2.4 RC3
Post by: LeoNeeson on March 18, 2019, 11:46:47 PM
while testing languages-addon (http://rejetto.com/forum/index.php?topic=11754.msg1064456#msg1064456), i found some strings in hfs.tpl not translated:
"Search" (line 105) and "Delete" (line 154) in hfs24rc2
Good find! :) There were TOO many strings (more than 15 changes) that were needed to be modified in the default template. Check the attached 'hfs.tpl' along with the updated spanish translation diff, that I've updated here (http://rejetto.com/forum/index.php?topic=11880.msg1064490#msg1064490)... ;)

I'm uploading the modified file so Rejetto could use it in a future hfs24rc3.exe (obviously, he could see the changes using an online service like DiffNow (https://www.diffnow.com/v1) or viewing a report of changes, that I've also upload here). I've also changed the "48em to 50em" thing (as it was previously reported here (http://rejetto.com/forum/index.php?topic=13060.msg1064178#msg1064178), and accepted by Rejetto here (http://rejetto.com/forum/index.php?topic=13060.msg1064317#msg1064317)). Any experienced user is welcome to confirm and verify if my modifications are good or not... ;)



» Silly question: does "hfs.diff.tpl" ONLY works on sub-folders? (because currently, if "hfs.diff.tpl" is on root, it doesn't work). Is that the normal behavior? (it doesn't should work on root too?).
 
Title: Re: version 2.4
Post by: Mars on March 24, 2019, 01:53:03 AM
I am in favor of these, exepted modifications for two things
line 448 -> HFS is the fixed name of the program, do not use {.! HFS.}
line 556 -> Anonymous is an account name, do not translate it with {.! Anonymous.}.}
Title: Re: version 2.4
Post by: LeoNeeson on March 24, 2019, 04:28:16 AM
@Mars: You are right! :) I agree on both, but mostly on the first (because that introduced a bug!). I've didn't realize that 'Anonymous' (on line 560) is case sensitive, so {.!Anonymous.} needs to be replaced with {.!anonymous.} (thanks to SilentPliz (https://rejetto.com/forum/index.php?action=profile;u=3008) for point me out that 'Anonymous' was case sensitive). ;)

This line needs to be replaced:
Quote
{.add to log|{.if|%user%|{.!User.} %user%|{.!Anonymous.}.} {.!renamed.} "{.^x.}" {.!to.} "{.^yn.}".}

With this correct line (it's only one letter change):
Quote
{.add to log|{.if|%user%|{.!User.} %user%|{.!anonymous.}.} {.!renamed.} "{.^x.}" {.!to.} "{.^yn.}".}

And I've added {.!HFS.} on line 452, because that already was on another place, in line 31, and that could be useful if someone wants to hide or change the name of the server (for security purposes), but since anyone could edit the template, it's not needed after all. But no harm is done if that is left (you can see {.!HFS.} was also included on the default template of v2.3m and even on v2.3k).
Title: Re: version 2.4 - bug
Post by: dj on March 24, 2019, 08:40:53 AM
"Switch On" and "Switch Off" button is reversed

update:
either only I have the error or nobody has noticed
Find the mistake between the two pictures! Image shows server turned on
Title: Re: version 2.4
Post by: SilentPliz on March 24, 2019, 09:12:36 AM
This line needs to be replaced:
With this correct line:
(By the way, talking about macro syntax, I'm curious to know why "%user%|{.!User.}" is working fine without giving an error, but "%user%|{.!Anonymous.}." throws an error, being that both values are on the same place).

Hi !

Try anonymous in place of Anonymous
Case sensitive (?)

Edit :  post deleted by mistake.  Restored here. I'am much better.  :o

Title: Re: version 2.4 - bug
Post by: SilentPliz on March 24, 2019, 11:38:08 PM
"Switch On" and "Switch Off" button is reversed

Not verified, I remember (I hope):  :)

Switch ON .... Proposes to turn ON the server, so it's currently OFF.

Switch OFF .... Proposes to turn OFF the server, so it's currently ON.

It's like a switch: turn off when it's on; turn on when it's off.
Title: Re: version 2.4
Post by: Mars on March 25, 2019, 12:33:04 AM
could not reproduce the bug with Anonymous, 
Quote
[]
{.$common-head.}
   <title>{.!HFS.} %folder%</title>
   {.add to log|{.if|%user%|{.!User.} %user%|{.!Anonymous.}.} {.!renamed.} "{.^x.}" {.!to.} "{.^yn.}".}
    <script type="text/javascript" src="/?mode=jquery"></script>

01:30:21 127.0.0.1:1567 Got 397 bytes
01:30:22 127.0.0.1:1567 Anonymous renamed "" to ""
01:30:22 127.0.0.1:1567 Requested GET /
Title: Re: version 2.4 - several bugs
Post by: LeoNeeson on March 25, 2019, 08:38:23 AM
"Switch On" and "Switch Off" button is reversed
I can confirm this bug! It was introduced on hfs24b6.exe (http://www.rejetto.com/hfsfiles/hfs24b6.exe) (hfs24b5.exe (http://www.rejetto.com/hfsfiles/hfs24b5.exe) had this function working correctly). If you look carefully, the icon shows the correct function, meanwhile the text is wrong. I guess this happened to Rejetto when doing some automatic 'search and replace' on the source code, without looking if every replaced instance was the correct (it's something common, that it could happen to anyone).

could not reproduce the bug with Anonymous, 
This is VERY weird! I'm pulling my hairs trying to reproduce the error, but now the bug is not coming back, neither using {.!Anonymous.} or {.!anonymous.}. There must be something with some specific configuration I had, but now I can't reproduce this error anymore (is crazy!). :o

I've saved a screenshot yesterday (just in case), but anyway the bug was not important, since it only was on the template (on client side), not on the HFS program (not on the server side, because the file gets renamed successfully anyway).

This was the error shown (see screenshot):
Code: [Select]
Error: {.break|result=bad session}
ok

But on the logs everything was normal:
Code: [Select]
19:42:22 127.0.0.1:2737 Requested GET /
19:42:25 127.0.0.1:2739 Requested GET /?mode=jquery
19:42:34 127.0.0.1:2740 anonymous renamed "/folder/SomeFile1.gif" to "SomeFile2.gif"

This error happened with some folder having the 'Hidden' flag, and the 'Upload' setting to 'Anyone'. Anyway, this error can't be reproduced anymore (but yesterday was solved only if {.!anonymous.} in lower case was used).



But I've found another interesting things (two possible small bugs):

CASE 1) If you have an empty 'hfs.ini', that file is IGNORED and the settings are saved directly on the registry (the file 'hfs.ini' is deleted on exit, and the account 'can change password' is created normally). Expected behavior: save all the settings on that empty 'hfs.ini'.

CASE 2) If you have a 'hfs.ini' file with a minimal content (like a simple 'space' character, or containing a word like 'HFS' at start), the file is recognized and used to save the settings there (the registry is not used), and if you already have a 'hfs.tpl' along with the 'hfs.exe' file, HFS is NOT creating the default account 'can change password' (if there is no 'hfs.tpl' file, the account 'can change password' is created normally. Expected behavior: the account 'can change password' should be created and saved on the 'hfs.ini' file on this specific scenario (also the settings should be saved in a file, if an empty 'hfs.ini' file is found).

I hope this could be easily fixed... :-\

Keep in touch if you need more details.
Cheers,
Leo.-
Title: Re: version 2.4 - bug
Post by: Mars on March 25, 2019, 06:20:51 PM
"Switch On" and "Switch Off" button is reversed

update:
either only I have the error or nobody has noticed
Find the mistake between the two pictures! Image shows server turned on

only texts OFF & ON  are reversed, the icons and the function is correct

reversed, the icons and the function is correct

Code: [Select]
SwitchON1.imageIndex:=if_(srv.active, 11, 4);
SwitchON1.caption:='Switch '+if_(srv.active, 'OFF', 'ON');

and managed from these lines until the rb5, since this there may have been a false manipulation in the sources

I'm arguing that something has been changed to look like this

Code: [Select]
SwitchON1.caption:=if_(srv.active, 'Switch ON', 'Switch OFF');
exactly
  S_OFF = 'Switch OFF';
  S_ON = 'Switch ON';
  SwitchON1.caption:=if_(srv.active, S_ON, S_OFF);

should be
  SwitchON1.caption:=if_(srv.active, S_OFF, S_ON);

in order to facilitate the literal translations and no longer fragmented, but that there was a misleading text

I fiddled the exe to correct the inversion pending the new version

http://dl.free.fr/vAaA050dO



Title: Re: version 2.4
Post by: Mars on March 25, 2019, 07:06:33 PM
This was the error shown (see screenshot):
Error: {.break|result=bad session}
ok

It is not a real bug, it comes from a syntax error in the model at the level of the section of the alias which appears only under very particular conditions because most of the time the text is considered as a comment

we notice that the message of the error does not represent a complete macro

Quote
[special:alias]
check session=if|{.{.cookie|HFS_SID_.} != {.postvar|token.}.}|{:{.cookie|HFS_SID_|value=|expires=-1.} {.break|result=bad session}:}

lit's not new and the problem is present in the default model since the 2.3f, which use cookie to check session
Title: Re: version 2.4
Post by: AlexALSP on March 27, 2019, 05:53:43 AM
Much time has passed. But there is no support for other languages. It's sad.  :'(
Title: Re: version 2.4
Post by: AlexALSP on March 27, 2019, 05:54:33 AM
Much time has passed. But there is no support for other languages. It's sad. :'(
Title: Re: version 2.4
Post by: Mars on March 27, 2019, 10:36:51 AM
as long as a definitive solution is not adopted to manage the languages, only a few will serve as support for the study of the translation.Later a wider choice will be available for those who wish to participate in the translation. 
Title: Re: version 2.4 - files with long name
Post by: dj on March 29, 2019, 07:35:00 PM
didn't test it, but i don't like the idea of truncating the file names. What when you have a difference at the end of them (a common case): you cannot tell the difference anymore.

thats right (a additional details button is needed)
alternatively hyphenate (https://developer.mozilla.org/en-US/docs/Web/CSS/hyphens) words (tested with FF, chrome android and edge in mobil-light_V4.4beta)

update: the additional Details button are the item menu buttons

Code: [Select]
[+common-head]
<style>
.item-link {float: left; overflow: hidden; text-overflow: ellipsis; max-width: 100%}
.item-menu-dialog {word-break: break-all}
</style>
Title: Re: version 2.4
Post by: Chris Harris on May 25, 2019, 07:03:59 AM
I am wondering when you will release the final version 2.4?
Chris
Title: Re: version 2.4
Post by: thepoo on May 25, 2019, 06:59:07 PM
Just tried out HFS for the first time ....

1.   Can we have a check box beside the file names so we can download selected items?

2.   The server I'm trying to download from contains a "check mark" character in all the filenames (somefile √.mkv) alt-251 ...... the files are listed and appear to be downloadable, but when you attempt to download them it says there are no files to download?    I don't believe the owner of the files wants to remove/change all the check marks .... can HFS be changed to allow the files with special characters to be downloaded?
Title: Re: version 2.4
Post by: LeoNeeson on May 26, 2019, 01:48:24 AM
I am wondering when you will release the final version 2.4?
Chris
I also have the same question... just wondering: perhaps he wants to add some new feature?... ??? at least it would be nice to have RC3. ;)

@Rejetto: Before you release the final or the release candidate (RC3) version, please don't forget to check THIS (http://rejetto.com/forum/index.php?topic=13060.msg1064489#msg1064489) post (with some needed changes).

1.   Can we have a check box beside the file names so we can download selected items?
It's already implemented on v2.4, you have to click on the 'Selection' button, and the checkboxes will appear.

2.   The server I'm trying to download from contains a "check mark" character in all the filenames (somefile √.mkv) alt-251 ...... the files are listed and appear to be downloadable, but when you attempt to download them it says there are no files to download?
I can confirm this. I'm attaching an example file to demonstrate this (just uncompress/unrar the file on some shared folder). When you try to download the file, you will get a "Not found" message (like if the file doesn't exists).
Title: Re: version 2.4
Post by: AlexALSP on May 26, 2019, 11:57:31 AM
Where to download the latest language file?  :-\
Title: Re: version 2.4
Post by: LeoNeeson on June 04, 2019, 05:25:37 AM
Where to download the latest language file?  :-\
The latest English language file (hfs.lng) is HERE (http://rejetto.com/forum/index.php?action=dlattach;topic=13060.0;attach=8889) (this was posted by Rejetto here (http://rejetto.com/forum/index.php?topic=13060.msg1064318#msg1064318)). You can translate this file using IniTranslator which you can get it from HERE (https://sourceforge.net/projects/initranslator/files/latest/download) (thanks SilentPliz (http://rejetto.com/forum/index.php?topic=13134.msg1064631#msg1064631) for the info!), or use a portable version of that program, that you can download it from HERE (https://portableapps.com/apps/development/initranslator-portable).
Title: Re: version 2.4
Post by: AlexALSP on June 09, 2019, 12:34:18 AM
The latest English language file (hfs.lng) is HERE (http://rejetto.com/forum/index.php?action=dlattach;topic=13060.0;attach=8889) (this was posted by Rejetto here (http://rejetto.com/forum/index.php?topic=13060.msg1064318#msg1064318)). You can translate this file using IniTranslator which you can get it from HERE (https://sourceforge.net/projects/initranslator/files/latest/download) (thanks SilentPliz (http://rejetto.com/forum/index.php?topic=13134.msg1064631#msg1064631) for the info!), or use a portable version of that program, that you can download it from HERE (https://portableapps.com/apps/development/initranslator-portable).

Thank you!
Title: Re: version 2.4
Post by: eranavni on June 12, 2019, 10:32:08 AM
Is it possible to add an option to block not just by IP but also to block by user agent in case it's identidied?
For example if I want users to be able to download from Chrome but not from Firefox ?
Title: Re: version 2.4
Post by: bmartino1 on June 12, 2019, 07:53:47 PM
Is it possible to add an option to block not just by IP but also to block by user agent in case it's identidied?
For example if I want users to be able to download from Chrome but not from Firefox ?

only thing o can think of 9s a hyml web redirect.

2-3 problems

1 you can use html/jQuery to find what browser has connect to your server:

https://stackoverflow.com/questions/12089942/php-html-css-if-firefox-if-chrome-if-safari

ex
Code: [Select]
   // Safari CSS and Webkit Google Chrome
if ($.browser.webkit) {
   $("#element").css('top', '2px');
} else if ( $.browser.safari ) //not fully supported on 1.7 jQuery {
   $("#element").css('top', '2px');
// Opera CSS
} else if ( $.browser.opera ) {
   $("#element").css('top', '2px');
// Internet Explorer CSS
} else if ( $.browser.msie ) {
   $("#element").css('top', '2px');
// Mozilla FireFox CSS
} else if ( $.browser.mozilla ) {
   $("#element").css('top', '2px');
// Normal Revert, careful and note your the use of !important
} else {
   $("#element").css('top', '2px');
   // You can have normal JavaScript between these too
   document.getElementById("element").style.top="2px";

the above will edit css based on browser

2, last i checked the latest chrome usses Firefox web agent, so theres no guarantee for that browser

3  hfs built in block level is at the protcal level not html browser level

html code

ex to stop ie from html visit

https://www.quora.com/Is-there-a-way-to-block-Internet-Explorer-from-my-website-using-HTML-If-so-is-there-a-way-to-modify-that-code-for-other-browsers
Title: Re: version 2.4
Post by: AlexALSP on June 20, 2019, 07:49:14 PM
The latest English language file (hfs.lng) is HERE (http://rejetto.com/forum/index.php?action=dlattach;topic=13060.0;attach=8889) (this was posted by Rejetto here (http://rejetto.com/forum/index.php?topic=13060.msg1064318#msg1064318)). You can translate this file using IniTranslator which you can get it from HERE (https://sourceforge.net/projects/initranslator/files/latest/download) (thanks SilentPliz (http://rejetto.com/forum/index.php?topic=13134.msg1064631#msg1064631) for the info!), or use a portable version of that program, that you can download it from HERE (https://portableapps.com/apps/development/initranslator-portable).

This format

http://rejetto.com/forum/index.php?topic=13134.0

I loaded in PASSOLO program without any problems and created a translation project.

With ini, there is a problem, PASSOLO does not understand this format. It is necessary to translate all the lines on the new. It is sad.

Again, all from the beginning. (((

And I can't figure out how to make a translation in the INI TRT program and save it in its original format ..
Explain, if not difficult.

 :-[


The program functions are inverse values:

64929_main_S_OFF = Switch OFF = Switch ON
64930_main_S_ON = Switch ON = Switch OFF

64929_main_S_OFF = Switch ON
64930_main_S_ON = Switch OFF
Title: Re: version 2.4
Post by: AlexALSP on June 21, 2019, 04:33:54 AM
I found a solution, but it is .......

Question?

Is this the final file or will it be supplemented.?

A lot of text without translation.
Title: Re: version 2.4
Post by: Mars on September 24, 2019, 01:12:57 PM
impossible to understand why, but with the last template under firefox on my desktop computer it became impossible to select an item from the check boxes without this having the same effect as clicking directly on the item, as if it were just a picture that changes its appearance
Title: Re: version 2.4
Post by: surikat on October 21, 2019, 08:32:33 PM

This version hangs even more often than version 2.3 m !

I have a gigabit connection and it is impossible to use. Simultaneous connections up to a maximum of 40. The speed is limited for

each IP to 2000 KB. The program freezes every 4-5 hours. It has become unrealistic to use..

Please give a link to version 2.3 K ! But from the official website rejetto. I'll try to use it.

Title: Re: version 2.4
Post by: Mars on October 21, 2019, 09:41:42 PM
https://sourceforge.net/projects/hfs/files/HFS/
Title: Re: version 2.4
Post by: surikat on October 22, 2019, 02:05:02 PM

Version 2.3 K also freezes .. Useless!
Title: Re: version 2.4
Post by: surikat on October 26, 2019, 06:51:07 PM
Maybe the problem is in connecting IPv6?
What are these displayed in hfs?
Memory max. hfs takes about 160 MB!
Download speed about 120-180 Mbps
Large file uploads from 700 MB to 4 GB
I use real folders in the program!

Settings :
Speed Limit for Single address : 1800 KB/s
Max Simultaneous  downloads for single address : 2
Title: Re: version 2.4
Post by: AlexALSP on December 20, 2019, 11:52:54 AM
Will we ever see the final version 2.4?  :)
Title: Re: version 2.4
Post by: Mars on December 20, 2019, 02:04:47 PM
Without wanting to appear defeatist, hfs has not evolved directly to https without having to use SSL, another evolution of the web is moving towards ipV6, this will probably be the limit of use for hfs.
Title: Re: version 2.4
Post by: Fysack on January 04, 2020, 02:34:32 AM
LOVE
Title: Re: version 2.4
Post by: surikat on January 05, 2020, 03:58:50 PM
So, this is going like filling the bathtub, whereby 100 megabits = 1 faucet or gigabit = 10 faucets simultaneously and a flooding problem.  For connection speeds like that, HFS needs changed to schedule dealing with the data, pro-rata higher than accumulating it.

Instead of waiting for an update, I suggest that you buy the cheapest USB to 100 megabit Ethernet adapter from ebay, for $2.  So, just fix the ip address, aim HFS at it; and if you continue to have trouble, try (ethernet adapter settings of windows) set it to a lower speed.  Given the added adapter, you won't have to slow down your whole server; however, it will take another Ethernet cable.

Alternatively, you could try OpenMediaVault, which is a Linux-based operating system for x86 and many of the little ARM systems.  Or, for Windows, some FTP servers (https://www.lifewire.com/windows-ftp-servers-free-817577) can do heavy duty workloads.

I Disabled the speed limit for single ip. And generally turn off the speed limit.

Set only Maximum simultaneous downloads for single address: 2

The program has stopped freezing!  Hallelujah!!!  This all happens to cause a speed limit.  Now everything is clear!

I will not have to limit the speed otherwise it will freeze.
Title: Re: version 2.4
Post by: surikat on January 05, 2020, 04:07:02 PM
Without wanting to appear defeatist, hfs has not evolved directly to https without having to use SSL, another evolution of the web is moving towards ipV6, this will probably be the limit of use for hfs.

I have never displayed in the ip address bar  ipV6 ?  Are these ip addresses somehow replaced by simple ipV4 ?   :o

Title: Re: version 2.4
Post by: Fysack on January 11, 2020, 01:30:19 AM
2X love
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 04:26:30 PM
impossible to understand why, but with the last template under firefox on my desktop computer it became impossible to select an item from the check boxes without this having the same effect as clicking directly on the item, as if it were just a picture that changes its appearance

thanks for reporting. I can workaround this problem by replacing this line
.change(selectionChanged)

with this line
.click(ev=>{ setTimeout(()=>{ ev.target.checked ^= 1; selectionChanged() }); return false })
Title: Re: version 2.4
Post by: Mars on January 29, 2020, 09:37:36 PM
hello boss
have a look at this post
http://rejetto.com/forum/index.php?topic=13094.msg1065170#msg1065170
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 09:52:53 PM
thanks, i'll include this fix too
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 10:27:09 PM
someone else already answered your first question

2.   The server I'm trying to download from contains a "check mark" character in all the filenames (somefile √.mkv) alt-251 ...... the files are listed and appear to be downloadable, but when you attempt to download them it says there are no files to download?    I don't believe the owner of the files wants to remove/change all the check marks .... can HFS be changed to allow the files with special characters to be downloaded?

i'm sorry but i fear that for the moment this kind of filenames will just not be supported. :(
It will take a long work to make it
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 10:35:48 PM
Is is possible to automatically log in a certain user from a fixed ip address?

something similar.
easy version: add in hfs the same folder twice, published with different names, A and B. You should protect A with password/account, and just hide folder B. Give the address of B only to trusted people, so they don't have to use a password.

complex version: B should contain a "diff template" that will block access from unwanted IP addresses, by using scripting commands.

Is it possible to set max number of connections limit, which is Not applied to browsing (no limit on file listing)?

menu > limits > max simultaneous downloads from single address
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 10:42:33 PM
Right click the tray icon and choose Switch ON, actually switches off
Page load while downloading a big file fast, also switches off   :o (needs watchdog feature added)

I fixed the inverted "switch", thanks.
If you are a programmare you may consider making a simple watchdog using events (menu > edit events scripts)
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 11:01:26 PM
I'm uploading the modified file so Rejetto could use it in a future

ok, thanks. Sorry for the late reply :)

Quote
» Silly question: does "hfs.diff.tpl" ONLY works on sub-folders? (because currently, if "hfs.diff.tpl" is on root, it doesn't work). Is that the normal behavior? (it doesn't should work on root too?).

it's supposed to work on root too, and i remember using it in the past.
Try entering this:
Code: [Select]
[]
test
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 11:10:42 PM
line 556 -> Anonymous is an account name, do not translate it with {.! Anonymous.}.}

it's not an account name, it's only  saying there is no user specified.
And I don't see the problem with keeping it upper case.
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 11:14:35 PM
I am wondering when you will release the final version 2.4?

when i feel like it's stable. We are still making many fixes, but i hope soon
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 11:18:09 PM
CASE 2) If you have a 'hfs.ini' file with a minimal content (like a simple 'space' character, or containing a word like 'HFS' at start), the file is recognized and used to save the settings there (the registry is not used), and if you already have a 'hfs.tpl' along with the 'hfs.exe' file, HFS is NOT creating the default account 'can change password' (if there is no 'hfs.tpl' file, the account 'can change password' is created normally. Expected behavior: the account 'can change password' should be created and saved on the 'hfs.ini' file on this specific scenario (also the settings should be saved in a file, if an empty 'hfs.ini' file is found).

it's not HFS creating that account, but the template itself. See it contains this
[special:import]
{.new account|can change password|enabled=1|is group=1|notes=accounts members of this group will be allowed to change their password.}
Title: Re: version 2.4
Post by: Mars on January 29, 2020, 11:21:23 PM
ok, thanks. Sorry for the late reply :)

it's supposed to work on root too, and i remember using it in the past.
Try entering this:
Code: [Select]
[]
test

i guess it means the  root as an real folder with an hfs.diff.tpl in it
what is placed in the diff template in real or virtual mode works
Title: Re: version 2.4
Post by: rejetto on January 29, 2020, 11:25:47 PM
This was the error shown (see screenshot):
Code: [Select]
Error: {.break|result=bad session}
ok

that's a bug, caused by the missing dot after "bad session"
thanks for reporting!
Title: Re: version 2.4
Post by: Mars on January 29, 2020, 11:44:57 PM
another reported  bug needed to be fixed at
http://rejetto.com/forum/index.php?topic=13218.msg1065090#msg1065090
Title: version 2.4 rc2 template
Post by: danny on February 11, 2020, 10:45:14 PM
My apologies!
During my recent cleaning conniption, I shouldn't have removed this file (attached).

It is Rejetto's HFS 2.4 template with: checkbox patch, practical archive scope, move requires related permissions, automated error-handler, faster uni icons instead of system icons, an efficient blue-effects background plus contrast effects, and DJ's really nice music player.

If I hadn't seen HFS until today, I'd want those features and 2 more: 
Ability to download fast and browse/upload, concurrently?
Security-first focus involving available opt-outs and native https?

Those 2 seem beyond the scope of a template; so, their status is not changed by this template file.
I think that it may be desired for reference.
Title: Re: version 2.4 rc2 template
Post by: LeoNeeson on February 11, 2020, 11:02:49 PM
Edit!  Also, especially awesome for lan-only use, because the music plays reliably and the search finds everything.
As a side note, you should know that you could very EASILY (see my post here (https://rejetto.com/forum/index.php?topic=13088.0)) recompile HFS by yourself, and edit the source code removing any hard-codded URL (about updates and self-test), and have a rock-solid secure server (that you personally could trust). It's not recommended to remove or edit parts of the source code without knowledge, but it can be done (Linux users compile executable binaries files all the time, and if you done that in the past, this is much easier).
Title: Re: version 2.4
Post by: rejetto on February 16, 2020, 10:04:06 PM
i just published latest fixes in RC3 (first post)
Title: Re: version 2.4
Post by: dj on February 17, 2020, 04:34:53 AM
i just published latest fixes in RC3 (first post)
windows defender:
rc3    1 danger  :'( Trojan:Win32/Detplock
rc2    0 danger
2.3m 0 danger


Title: Re: version 2.4
Post by: LeoNeeson on February 17, 2020, 05:06:24 AM
latest version http://www.rejetto.com/hfsfiles/hfs24rc3.exe
Nice to see RC3! :D According to my tests (if I'm not mistaken, correct me if I'm wrong), RC3 (hfs24rc3.exe) comes with 'exactly' the same default template than RC2 (hfs24rc2.exe). Perhaps Rejetto forgot to include the changes on the template. I did a list of what's fixed...

» Addressed issues, possibly fixed on RC3:
- Response headers with multiple content length (https://rejetto.com/forum/index.php?topic=13125.0) (untested by me)
- Color issue in log window, adjusted to system colors (https://rejetto.com/forum/index.php?topic=13233.msg1065234#msg1065234) (untested by me)
- 'Switch On' and 'Switch Off' button was reversed (https://rejetto.com/forum/index.php?topic=13060.msg1064516#msg1064516) (tested and fixed OK!)
- 'hfs.diff.tpl' works ony on sub-folders? (https://rejetto.com/forum/index.php?topic=13060.msg1065199#msg1065199) (not a bug, it's working OK!)

» Issues possibly fixed on RC4 (template related):
- Selecting an item has the same effect as download (http://rejetto.com/forum/index.php?topic=13060.msg1065193#msg1065193)
- Make some strings translatable (https://rejetto.com/forum/index.php?topic=13060.msg1064489#msg1064489), and change '48em to 50em' (http://rejetto.com/forum/index.php?topic=13060.msg1064178#msg1064178)
- Minor fixes on the template ('bad session' was missing a dot) (https://rejetto.com/forum/index.php?topic=13060.msg1065204#msg1065204)

» 'Won't fix' list (at the moment):
- Cut/word-wrap long file names (https://rejetto.com/forum/index.php?topic=13060.msg1064478#msg1064478)
- Problematic unicode filenames (https://rejetto.com/forum/index.php?topic=13060.msg1065196#msg1065196)

@Rejetto: a silly but still somewhat important question: what is exactly the version of the compiler you are currently using to build hfs24rc3.exe (I ask this just for references purposes only, and because I'm trying to figuring out why there were some memory problems/issues reported by several users, starting from versions v2.3k or v2.3m (like for example, someone did here (http://rejetto.com/forum/index.php?topic=13060.msg1065023#msg1065023)). Are you still using Turbo Delphi (http://rejetto.com/forum/index.php?topic=3061.msg1031534#msg1031534)? (if you don't feel more comfortable answering this, just ignore the question) ...or perhaps have you changed the compiling platform, let say, from Windows 7 to 10, and that affected some users? (By the way, I currently don't have any memory problem with my Windows, but I'm curious to know if a compiler change could be the reason).

windows defender:
rc3    1 danger  :'( Trojan:Win32/Detplock
rc2    0 danger
2.3m 0 danger
I was just writing this post, when I see your message. You can rest assured is a false positive (hfs24rc3.exe was tested by myself, and I found it safe, just be sure the file you have downloaded has the same checksum, with SHA1: A98EE8B9D8D2C523466F5609EE71436BEDD69B22 (https://www.virustotal.com/gui/file/ef373572bf9463e3501a5c151b388277701bd6dbb08835b3f64f57dbc78539e8/detection)). Besides that, both RC2 and RC3 have exactly the same size 2,12 MB (2.224.640 bytes), so I doubt it could have anything bad. On each version, it's the same issue with antivirus companies, as always some of them flag HFS as not safe (and mostly because HFS is not digitally signed).

Cheers,
Leo.-
Title: Re: version 2.4
Post by: rejetto on February 17, 2020, 09:49:24 AM
windows defender:
rc3    1 danger  :'( Trojan:Win32/Detplock
rc2    0 danger

i just scanned using windows defender and no threat was detected.
i really don't know what to do about this. I stopped using the exe compression to avoid false positives, but yet this madness won't stop while these "security" softwares will consider HFS itself as a menace, "A Variant Of Win32/Server-Web.HFS"
https://www.virustotal.com/gui/file/ef373572bf9463e3501a5c151b388277701bd6dbb08835b3f64f57dbc78539e8/detection
Title: Re: version 2.4
Post by: rejetto on February 17, 2020, 09:56:08 AM
Nice to see RC3! :D According to my tests (if I'm not mistaken, correct me if I'm wrong), RC3 (hfs24rc3.exe) comes with 'exactly' the same default template than RC2 (hfs24rc2.exe).

shit, i also made a quick test to verify the new template was bundled, but apparently something went wrong :(
i'll have to release again, sorry, and thanks for reporting

Quote
the version of the compiler you are currently using to build hfs24rc3.exe

the old turbo delphi. I would like to use the new one but i couldn't solve all the issues
Title: Re: version 2.4
Post by: rejetto on February 17, 2020, 11:38:13 PM
template corrected http://www.rejetto.com/hfsfiles/hfs24rc4.exe

(hopefully)
Title: Re: version 2.4
Post by: LeoNeeson on February 20, 2020, 05:46:22 AM
template corrected http://www.rejetto.com/hfsfiles/hfs24rc4.exe
(Sorry, I had no time to check this yesterday)

Cool! 8) now everything seems to be working OK!...
At first sight, I didn't found anything wrong to report.
I will do more in-depth tests on the next days.
Title: Re: version 2.4 - addon
Post by: dj on February 23, 2020, 06:30:26 AM
Select 🎉 songs for playlist (same as in mobil-light_V5.4.1 (http://rejetto.com/forum/index.php?topic=11754.msg1065283#msg1065283))
Title: Re: version 2.4
Post by: LeoNeeson on February 25, 2020, 05:32:05 AM
Since HFS v2.4 is oriented to modern standards, I was surprised when I've found that HFS v2.4 has missing MIME support for SVG files (in default hfs.ini, it should have declared in 'mime-types=' setting "*.svg|image/svg+xml").

Without this, on some browsers .svg files are open 'as download', and NOT directly viewed on browser (as is meant). Even if the setting 'open-in-browser=' has '*.svg' on it, those files are not opened (viewed) directly on the browser (it ask for download). The same happens with .swf files (Flash files), which without having the MIME type declared, the browser open it 'as download', when it's meant to be opended directly. So, I suggest to add those two MIME types (or at least only SVG, since Flash is getting deprecated). See the text marked in red for .swf and on green for .svg:

Quote
mime-types=*.htm;*.html|text/html|*.jpg;*.jpeg;*.jpe|image/jpeg|*.gif|image/gif|*.png|image/png|*.bmp|image/bmp|*.ico|image/x-icon|*.mpeg;*.mpg;*.mpe|video/mpeg|*.avi|video/x-msvideo|*.txt|text/plain|*.css|text/css|*.js|text/javascript|*.swf|application/x-shockwave-flash|*.svg|image/svg+xml

What do you think about adding it as default?...
Title: Re: version 2.4
Post by: LeoNeeson on February 25, 2020, 05:34:45 AM
(https://i.imgur.com/8DS1zXA.gif) As promised, I did some more tests this weekend, including a review of changes in the new RC4's template (see the attachment with the results from DiffNow.com).

(https://i.imgur.com/k5aSWir.gif) Obviously, RC4 dropped support for old browsers (like Chrome 40, 31 and older), which were supported on RC2/RC3 (but don't worry, at least for me that's not a problem anymore, since I've upgraded most of my browsers). Tested on Chrome (http://download1.srware.net/old/iron/win/portable/) 48 (http://download1.srware.net/old/iron/win/portable/IronPortable_48.zip) and Chrome v75.0.3770.80 (64-bit), and on both they work well.

I encourage everyone (anyone who reads this topic), to leave some comment and opinion about the latest v2.4 RC4 (especially those who lately reported some bug, error or had some issue).

So far, everything is working as expected... (https://i.imgur.com/lgmXfyh.gif)

Cheers,
Leo.-
Title: Re: version 2.4 - addon
Post by: dj on March 08, 2020, 08:32:28 AM
sets dark mode by OS (https://docs.microsoft.com/en-us/windows-hardware/customize/desktop/set-dark-mode)

Code: [Select]
[+]
<script>
if(window.matchMedia('(prefers-color-scheme:dark)').matches) document.body.classList.add('dark-theme')  //light remove
</script>


a better dark theme (with sufficiently high contrast ratio)

Code: [Select]
[+]
<style>
body.dark-theme .item-props {color:#8A8A8A}
body.dark-theme #title-bar {color:#C9C9C9}
body.dark-theme a.pure-button {color:#303030}
body.dark-theme .item .comment {color:#B3B3B3 }
</style>
increase Lighthouse Accessibility 7%
other todo to increase Accessibility: add alt to img, add title to all buttons


Easter bunny will bring's lazyimages (thumbs) with picviewer 🐇
🆕 also works with the current Firefox
Title: Re: version 2.4
Post by: giggy on April 19, 2020, 03:53:24 PM
latest version http://www.rejetto.com/hfsfiles/hfs24rc4.exe

there's a new default template, mobile friendly and responsive.
i'm almost finished.
Would you start to give it a try?

I love the script. :-)
Question: How can I log out? I was trying to log in as another user but I can't find an option to log out of the current user.
Thank you.
Title: Re: version 2.4
Post by: Mars on April 19, 2020, 04:43:04 PM
there is no logout linked to the login of hfs, to die a connection you have to close the navigator
Title: Re: version 2.4
Post by: dj on April 19, 2020, 04:57:12 PM
I love the script. :-)
Question: How can I log out? I was trying to log in as another user but I can't find an option to log out of the current user.
Thank you.

Logout is possible with IE and Edge, but
you need to add a addon (http://rejetto.com/forum/index.php?topic=13060.msg1064345#msg1064345) (diff tpl)
Title: Re: version 2.4
Post by: giggy on April 19, 2020, 06:55:47 PM
there is no logout linked to the login of hfs, to die a connection you have to close the navigator
I either have to clear cash or use another browser. Just closing and reopening a browser did not work. Thanks.
Title: Re: version 2.4
Post by: giggy on April 19, 2020, 06:56:35 PM
Logout is possible with IE and Edge, but
you need to add a addon (http://rejetto.com/forum/index.php?topic=13060.msg1064345#msg1064345) (diff tpl)
Thank you. * I wanted to use firefox or chrome.
Title: Re: version 2.4
Post by: NateBR on April 24, 2020, 12:58:13 PM
Any suggestion when this will become final version?
I noticed this beta contacts 2 or 3 IP addresses daily, is that normal?
Title: Re: version 2.4
Post by: Mars on April 24, 2020, 01:54:22 PM
a finalized version is a version that will no longer evolve, this would mean that the hfs project stops or restarts on other bases, what is most lacking in the current version is to be able to communicate directly with the SSL protocol without having to use Stunnel.

regularly, hfs sends a request to the rejetto site to check if there is not a new version, it also connects to different addresses allowing to retrieve the external ip address for updating the DNS, that's why hfs connects to various ip daily
Title: Re: version 2.4
Post by: rejetto on April 29, 2020, 03:44:25 PM
1) Any suggestion when this will become final version?
2) I noticed this beta contacts 2 or 3 IP addresses daily, is that normal?

1) nope, i'm stuck on a problem now, can't say when i'm done
2) there are few features that contact other servers, but you can disable or not use them. One is the "auto check every day" for new versions. Another is the "find external ip address".


ah, only now i saw mars' reply :)
Title: Re: version 2.4
Post by: nightpurrer on May 03, 2020, 03:19:02 PM
I'm sorry if I'm repeating the subject, but the thread is too big to review.
Will there be a terminal version for Linux?
Title: Re: version 2.4
Post by: SilentPliz on May 03, 2020, 03:55:43 PM
Welcome nightpurrer.  :)

I'm sorry if I'm repeating the subject, but the thread is too big to review.Will there be a terminal version for Linux?

No, sorry.
Title: Re: version 2.4
Post by: nightpurrer on May 03, 2020, 07:25:14 PM
Why? This software is unique and i can't find similar fileserver for the Linux.
Title: Re: version 2.4
Post by: LeoNeeson on May 03, 2020, 07:32:23 PM
Why? This software is unique and i can't find similar fileserver for the Linux.
As far I know, you can run HFS on Linux using Wine, but not as command line (terminal version). Perhaps if you know someone who could port HFS to Lazarus, then a native Linux version could be compiled (but I'm not experienced with Linux).
Cheers,
Leo.-
Title: Re: version 2.4
Post by: bmartino1 on May 10, 2020, 01:12:38 PM
wine code breaks area and others keep changing the wine code, there are too many installation methods and a bunch of libs between release, so their is no default wine installation.

HFS3 and rejeto had sources for a per attempt with linux php , but i have lost that thread and base code. That project is still in alpha, and was never completed nor worked on. I here it was stoped.

If using linux, i higly recomend ubuntu and  "LAMP"
https://www.digitalocean.com/community/tutorials/how-to-install-linux-apache-mysql-php-lamp-stack-ubuntu-18-04
what made HFS unique is how it was a file upload server:
https://www.w3schools.com/php/php_file_upload.asp

I Personaly like to use Build Ubuntu Mate, with is Debian baase, Leo i hear like to use Retro Arch (if memroy serves, its RPM based.)
https://ubuntu-mate.org/

i made a wine installing page before, but parts of it are Right now, and the others parts no longer exist to due code changes and updates.

its not a clear search for this thread, but this was my latest instruction:
http://rejetto.com/forum/index.php?topic=11952.msg1062518#msg1062518
Title: Re: version 2.4
Post by: rejetto on May 10, 2020, 04:29:30 PM
Why? This software is unique and i can't find similar fileserver for the Linux.

i don't have time to do it. If someone else wants to do it...


the repo bmartino1 is talking about is https://github.com/rejetto/hfs
Title: Re: version 2.4
Post by: rejetto on May 17, 2020, 05:56:14 PM
sorry but we are starting our labeling again from beta 1 because of the many internal changes.
Now 2.4 includes unicode and a 'logout' button.
https://github.com/rejetto/hfs2/releases/tag/v2.4-beta01
Title: Re: version 2.4
Post by: dj on May 21, 2020, 03:51:40 PM
Addons tested with version 2.4beta1:

Lazy loading images (http://rejetto.com/forum/index.php?topic=13060.msg1065357#msg1065357) loads full-img-thumbs at the moment of need (scrolling) to save bandwidt
more to come

more on github (https://github.com/dj0001/hfs-template/tree/master/addons)
files beginning with a capital letter works also with the default(2.4) template

Title: Re: version 2.4
Post by: rejetto on May 21, 2020, 04:21:37 PM
i just saw that your diff tpl is not loaded correctly in beta 1 -_-
It will in beta 2.
I'm happy to see you are already using that feature :D
Title: Re: version 2.4
Post by: dj on May 21, 2020, 04:39:40 PM
It works if you copy the text in VFS-Prop.-Diff template,
but not if you put the file in the folder of hfs.exe.

(https://raw.githubusercontent.com/dj0001/pwa/gh-pages/test/screenshot.jpg)
Title: Re: version 2.4
Post by: rejetto on May 23, 2020, 07:58:31 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-beta02

fixes and IPv6
Title: Re: version 2.4
Post by: MarkV on May 24, 2020, 03:27:33 AM
Hi,

thanks for the IPv6 support, think I was one of the folks bringing up the subject, some time ago.  ;) Great to see that there is still progress and the project alive.
Github repo is another nice addon, no need to search the whole forums for the one download you might have missed...  :)

However, there is no light without shadow, so I have to announce that the 2.4 versions of HFS finally, and officially, stopped working on Windows 98 SE!  ::)
Absolutely nothing needs to be done about this, it was inevitable. 2.3m works fine on 98SE, and it's amazing that it outlived the OS by many years.  8)
Just FYI, 2.4beta1 still runs fine on Windows 2000 SP4. :o

Gotta stop now, running out of smileys.  :P Keep up the good work and stay safe! Bye for now.
Title: Re: version 2.4
Post by: NaitLee on May 24, 2020, 04:56:37 AM
It's a great new that 2.4 entered beta stage. But...
 Beta2 seems broken in certain conditions...

Tested with various browsers, both Windows10 and Wine/Linux (Windows7), with firewall whitelisted and anti-virus software closed:

The default template's action bar and location bar sometimes are not shown. Try enter various pages and refresh them.

Most severe ones are:
 View pages with Chrome/Edge, there is a chance that HFS transfers some data, then "die" for a while without responding data. (not sure)
 View pages with Firefox/Nightly, HFS only transfers a few data, then "died" too, and never respond any data to any browsers. (this is true)
 Switching on/off have no work to them.
 Above problems also comes with: not all required files of a page are transferred, thus my index page is broken since no full CSS.
 After temporarily resetting options, there's a chance that HFS cannot open any port to switch on.

When I switch back to 2.3, everything is well... ???
Additionally, It seems that (maybe) these things are there after adopting throwback/takeback templates' initial settings.

Did these only happened to me? :-\

The last thing: How about adding the translation system (.lng file) back ::)
Title: Re: version 2.4
Post by: rejetto on May 24, 2020, 09:37:56 AM
However, there is no light without shadow, so I have to announce that the 2.4 versions of HFS finally, and officially, stopped working on Windows 98 SE!  ::)

hi MarkV! win98? wow :D thanks for reporting.
I'm glad we are still ok with XP, since some people here still use it. I'll do my best without killing me tho. Time is short.
Title: Re: version 2.4
Post by: rejetto on May 24, 2020, 09:43:40 AM
View pages with Chrome/Edge, there is a chance that HFS transfers some data, then "die" for a while without responding data. (not sure)
...
Did these only happened to me? :-\

The "action bar" disappears if the request for "?mode=section&id=lib.js" fails.
i was able to reproduce a problem like this by having a ban rule with range syntax (address1-address2).
Do you have a ban like this?

Quote
The last thing: How about adding the translation system (.lng file) back ::)

at this point i should give a look if the feature can be restored. No promises tho.
Title: Re: version 2.4
Post by: rejetto on May 24, 2020, 06:15:46 PM
the reintroduction of the lng file seems to work.
It will be included in next beta.
Title: Re: version 2.4
Post by: rejetto on May 24, 2020, 06:24:49 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-beta03

i introduced also a change with the template system: custom templates will be loaded as they were diff tpl. This mean that if you need a resource in your template, a section, that is present in the default template, you can avoid duplicating it in yours, and just consider as it is present. You can do it for css, and for the ajax.* sections for example. I hope this is not breaking anything.
Title: Re: version 2.4
Post by: dj on May 25, 2020, 12:47:46 AM
View pages with Chrome/Edge, there is a chance that HFS transfers some data, then "die" for a while without responding data. (not sure)
...
Did these only happened to me? :-\
I can confirm that.
I have no bans.

! The script from "http://192.168.137.1/?mode=section&id=lib.js" was loaded, although its MIME type ("") is not a MIME type valid for JavaScript.
! Loading failed for the <script> with the source "http://192.168.137.1/?mode=section&id=lib.js"

Sometimes ?mode=jquery is pending.
beta1 works fine. Beta2 and 3 seems broken.

Title: Re: version 2.4
Post by: win98se on May 25, 2020, 07:46:50 AM
i introduced also a change with the template system: custom templates will be loaded as they were diff tpl. This mean that if you need a resource in your template, a section, that is present in the default template, you can avoid duplicating it in yours, and just consider as it is present. You can do it for css, and for the ajax.* sections for example. I hope this is not breaking anything.
The switch theme button at the top right corner doesn't work at all, and there are errors indicating getCookie and setCookie are not defined.
Title: Re: version 2.4
Post by: LeoNeeson on May 25, 2020, 07:48:45 AM
ok Leo, i've used flashget and i've found the bug in HFS. It will work in next release.
(Sorry for not leaving this message earlier, yesterday)

I've tested Beta 3 (https://github.com/rejetto/hfs2/releases/tag/v2.4-beta03) (I've skipped Beta 2), and the FlashGet problem was finally SOLVED!. So far, I have not encountered any other problems. :)

Stay safe people, and congrats to Rejetto for how v2.4 is going (the subtle change of colors was a nice touch). Glad to see lng support is back. :D
Cheers,
Leo.-

Quote
» Personal note: From now (until further notice), it will not be possible for me doing 'deep' testings of new builds as before, for as much as I would like to. As I've said previously, on the last past week I've started working again, and my free time is again being very limited. Besides that, I don't have a IPv6 connection to make tests, so, I hope any of you make those tests and check everything is OK. If you have anything to say to me, please leave a message on the public forum to get my attention (since I won't be logging here unless I have something to post, and I won't be checking emails as often as before). However, I'll keep an eye on the forum as always.
Title: Re: version 2.4
Post by: Rapid on May 25, 2020, 02:19:49 PM
I can confirm that.
I have no bans.

! The script from "http://192.168.137.1/?mode=section&id=lib.js" was loaded, although its MIME type ("") is not a MIME type valid for JavaScript.
! Loading failed for the <script> with the source "http://192.168.137.1/?mode=section&id=lib.js"

Sometimes ?mode=jquery is pending.
beta1 works fine. Beta2 and 3 seems broken.
Looks like TWSocketServer have some unexpected behavior. I revert to TWSocket and it work as before
Title: Re: version 2.4
Post by: rejetto on May 25, 2020, 04:26:22 PM
I can confirm that. I have no bans.

ok, i've found the problem
https://github.com/rejetto/hfs2/releases/tag/v2.4-beta04


Looks like TWSocketServer have some unexpected behavior. I revert to TWSocket and it work as before

exactly.
twsocketserver is not compatible with the way i'm currently accepting connections.
in the future, when i shall probably switch to it for ssl, will have to make deepers changes. For the moment it was easier to just add a second socket for ipv6.
Title: Re: version 2.4
Post by: TEA-Time on May 25, 2020, 08:02:24 PM

Hi rejetto,


I ran into a bug with the new "Copy URL with password" links.  After logging in, I think it's the style sheet that doesn't load.  The page is just text and basic buttons with no graphics.  If I try to upload something it fails, but only after the entire file has been sent.  I have no idea where the file was being put as it was uploading, but it wasn't in the folder after it finished.  I found out the hard way with an almost 14 GB file a friend tried to send.  Doh! lol  The problem appears to be with the fact that the part starting with "?mode=auth&u=...." is still in the address bar immediately after logging in.  If I click a folder button or link, that part is removed and the style sheet loads and uploads work.


Also, the "Leave disconnected connections" option is missing from the right click menu for the current connection pane.


Thanks!
-Tim
Title: Re: version 2.4
Post by: rejetto on May 25, 2020, 08:58:53 PM
hi Tea, what version were you using?
the "leave disconnected" option was removed. If you need to keep track of connections you can use the log.
Title: Re: version 2.4
Post by: TEA-Time on May 25, 2020, 09:07:58 PM
hi Tea, what version were you using?


Sorry!  v2.4 beta 4.


Quote
the "leave disconnected" option was removed. If you need to keep track of connections you can use the log.


Ah.. ok.  I usually ban from there, but now I see it's possible from the log.  Works for me. :)
Title: Re: version 2.4
Post by: MarkV on May 26, 2020, 05:19:16 AM
hi MarkV! win98? wow :D thanks for reporting.
I'm glad we are still ok with XP, since some people here still use it. I'll do my best without killing me tho. Time is short.

I hope you've seen the smileys. Just checked some old VMs and PCs and noticed. Personally, I'll start complaining when it stops working on Windows 7, OK? :)

Currently, Windows 2000 and up are still OK (last tested 2.4beta1). Please note that IPv6 support is an addon package for Windows 2000 and Windows XP. No idea how HFS will perform if the IPv6 package is not installed on these OS.
Might check it out. That's what we did in the old days, correct? ;)
Title: Re: version 2.4
Post by: MarkV on May 26, 2020, 09:37:50 AM
First results are in:

I couldn't get it to run on Windows 2000 SP4, so far. Have to check what the fresh system still lacks, in comparison to the established one that runs HFS fine. Maybe it is the IPv6 TP.

Important hint for Windows XP users:

On a standard Windows XP installation with SP3, HFS will start, no problem, but it will be disabled and unable to be put online. That's because a standard install is missing the IPv6 component.
Server will be permanently OFFLINE until you:
1. Open a command prompt,
2. issue a
Code: [Select]
netsh int ipv6 install command, and
3. reboot.

With activated IPv6, HFS will  be started and online.
Title: Re: version 2.4
Post by: rejetto on May 26, 2020, 11:46:11 AM
I couldn't get it to run on Windows 2000 SP4, so far. Have to check what the fresh system still lacks, in comparison to the established one that runs HFS fine. Maybe it is the IPv6 TP.

thanks. did you try older builds?

Quote
On a standard Windows XP installation with SP3, HFS will start, no problem, but it will be disabled and unable to be put online. That's because a standard install is missing the IPv6 component.

thanks for reporting. Solved in next release, for those who don't want to install ipv6

I hope you've seen the smileys. Just checked some old VMs and PCs and noticed. Personally, I'll start complaining when it stops working on Windows 7, OK? :)

i saw the smileys, but i was not worried, just surprised :D
Title: Re: version 2.4
Post by: rejetto on May 26, 2020, 01:04:26 PM
I ran into a bug with the new "Copy URL with password" links.  After logging in, I think it's the style sheet that doesn't load.  The page is just text and basic buttons with no graphics. 

i just tried but cannot reproduce this problem. Let me know if you find more on it.

Quote
If I try to upload something it fails, but only after the entire file has been sent.  I have no idea where the file was being put as it was uploading, but it wasn't in the folder after it finished. 

ok i'll see to fix it for next release

Quote
I found out the hard way with an almost 14 GB file a friend tried to send.  Doh! lol  The problem appears to be with the fact that the part starting with "?mode=auth&u=...." is still in the address bar immediately after logging in.  If I click a folder button or link, that part is removed and the style sheet loads and uploads work.

oh man, sorry but i wouldn't use HFS for such job, as HTTP doesn't support resume on upload. Unless it's in LAN, and it takes few minutes.
TBH i suspect nowadays upload-resume could be implemented via javascript. Maybe for a future release.
Title: Re: version 2.4
Post by: rejetto on May 26, 2020, 02:54:37 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-beta05

a few fixes
Title: Re: version 2.4
Post by: MarkV on May 26, 2020, 10:30:17 PM
Yes, the 2.3m build #300 works fine on that vanilla W2K SP4. So, there is at least a working fallback build.

No clue if it helps. but here's the Dr. Watson (drwtsn32.exe) report for the crash:
Code: [Select]


Application exception occurred:
        App:  (pid=1140)
        When: 5/27/2020 @ 00:10:25.788
        Exception number: c0000025
()

*----> System Information <----*
        Computer Name: VBX-W2KTEST
        User Name: Test
        Number of Processors: 1
        Processor Type: x86 Family 15 Model 2 Stepping 0
        Windows 2000 Version: 5.0
        Current Build: 2195
        Service Pack: 4
        Current Type: Uniprocessor Free
        Registered Organization: *
        Registered Owner: Test

*----> Task List <----*
   0 Idle.exe
   8 System.exe
 180 smss.exe
 208 csrss.exe
 228 WINLOGON.exe
 256 SERVICES.exe
 268 LSASS.exe
 392 VBoxService.exe
 480 svchost.exe
 508 spoolsv.exe
 536 AvastSvc.exe
 568 svchost.exe
 608 regsvc.exe
 624 mstask.exe
 540 winmgmt.exe
 704 svchost.exe
1168 explorer.exe
1280 VBoxTray.exe
1336 AvastUI.exe
1340 internat.exe
1564 explorer.exe
1140 hfs.exe
 784 drwtsn32.exe
   0 _Total.exe

(00400000 - 00979000)
(77F80000 - 77FFC000)
(77570000 - 775A0000)
(77E10000 - 77E6F000)
(7C570000 - 7C624000)
(77F40000 - 77F7D000)
(7C2D0000 - 7C335000)
(77D30000 - 77D9F000)
(77800000 - 7781E000)
(76620000 - 76631000)
(76B30000 - 76B6E000)
(70A70000 - 70AD6000)
(78000000 - 78045000)
(71710000 - 71794000)
(7CF30000 - 7D176000)
(77820000 - 77827000)
(759B0000 - 759B6000)
(779B0000 - 77A4C000)
(7CE20000 - 7CF0F000)
(7CDC0000 - 7CE10000)
(7C340000 - 7C34E000)
(77BF0000 - 77C01000)
(77980000 - 779A5000)
(75050000 - 75058000)
(75030000 - 75044000)
(75020000 - 75028000)
(77950000 - 7797B000)
(751C0000 - 751C6000)
(75150000 - 75160000)
(75E60000 - 75E7A000)
(6E420000 - 6E426000)
(7C950000 - 7C9DF000)
(64E40000 - 64E61000)
(78B20000 - 78DDE000)
(77840000 - 7787E000)
(770C0000 - 770E3000)
(772B0000 - 7731D000)
(782C0000 - 782CC000)
(77340000 - 77353000)
(77520000 - 77525000)
(77320000 - 77337000)
(773B0000 - 773DF000)
(77380000 - 773A3000)
(77830000 - 7783E000)
(77880000 - 7790E000)
(7C0F0000 - 7C154000)
(774E0000 - 77514000)
(774C0000 - 774D1000)
(77530000 - 77552000)
(77360000 - 77379000)
(777E0000 - 777E8000)
(777F0000 - 777F5000)

State Dump for Thread Id 0x5b8

eax=0012f914 ebx=0012ffb0 ecx=00000000 edx=ffffffff esi=0012f98c edi=00000000
eip=77facc94 esp=0012f908 ebp=0012f974 iopl=0         nv up ei pl nz na pe nc
cs=001b  ss=0023  ds=0023  es=0023  fs=003b  gs=0000             efl=00200202


function: RtlTraceDatabaseEnumerate
        77facc6f eb23             jmp RtlConvertUlongToLargeInteger+0x10c4 (77fbc294)
        77facc71 f6460401         test    byte ptr [esi+0x4],0x1       ds:00bb9872=??
        77facc75 742a             jz      wcstombs+0x16a9 (77fb93a1)
        77facc77 c745a0250000c0                                  ss:00bb985a=????????
                                  mov     dword ptr [ebp+0xa0],0xc0000025
        77facc7e 8d45a0           lea     eax,[ebp+0xa0]         ss:00bb985a=????????
        77facc81 c745a401000000   mov   dword ptr [ebp+0xa4],0x1 ss:00bb985a=????????
        77facc88 50               push    eax
        77facc89 8975a8           mov     [ebp+0xa8],esi         ss:00bb985a=????????
        77facc8c 894db0           mov     [ebp+0xb0],ecx         ss:00bb985a=????????
        77facc8f e80cf8ffff       call    RtlRaiseException (77fac4a0)
FAULT ->77facc94 8b1b             mov     ebx,[ebx]              ds:0012ffb0=0012ffe0
        77facc96 83fbff           cmp     ebx,0xff
        77facc99 0f8540ffffff     jne     RtlTraceDatabaseEnumerate+0xb3 (77facbdf)
        77facc9f eb08             jmp     wcstol+0x183 (77fb7ca9)
        77facca1 b001             mov     al,0x1
        77facca3 eb06             jmp     allmul+0xb (77fb4fab)
        77facca5 834e0408         or     dword ptr [esi+0x4],0x8 ds:00bb9872=????????
        77facca9 32c0             xor     al,al
        77faccab 5f               pop     edi
        77faccac 5e               pop     esi
        77faccad 5b               pop     ebx
        77faccae c9               leave

*----> Stack Back Trace <----*

FramePtr ReturnAd Param#1  Param#2  Param#3  Param#4  Function Name
0012F974 77F91BC6 0012F98C 0012F9BC 0012F98C 0012F9BC ntdll!RtlTraceDatabaseEnumerate
0012FCE0 00682EE2 0EEDFADE 00000001 00000007 0012FCF8 ntdll!KiUserExceptionDispatcher
0012FD4C 00683D83 0012FDD8 00000000 012B948C 0012FD88 !TMethodImplementationIntercept
0012FD60 006AD7DB 0012FE04 0012FD94 006AD7F1 0012FD88 !TMethodImplementationIntercept
0012FD88 006B5341 0012FE04 0012FDA0 006B5486 0012FE10 !TMethodImplementationIntercept
0012FE10 00799606 00000000 0012FE48 007A1D2B 0012FE50 !TMethodImplementationIntercept
0012FE1C 007A1D2B 0012FE50 007A1DA3 0012FE48 0089050C !TMethodImplementationIntercept
0012FE48 007A9AF7 0012FED8 007A9C8C 0012FED0 00000000 !TMethodImplementationIntercept
0012FED0 007BF53B 0012FF00 007BF5ED 0012FEF4 01396A7C !TMethodImplementationIntercept
0012FEF4 007BFA28 0012FF8C 0012FF94 007BFF1E 0012FF8C !TMethodImplementationIntercept
0012FF8C 007D653E 0012FFB0 007D6560 0012FFC0 00000000 !TMethodImplementationIntercept
0012FFC0 7C5989D5 00000000 00000000 7FFDF000 00000000 !TMethodImplementationIntercept
0012FFF0 00000000 007D6380 00000000 000000C8 00000100 kernel32!ProcessIdToSessionId

*----> Raw Stack Dump <----*
0012f908  c0 fc 12 00 14 fd 12 00 - 8c f9 12 00 25 00 00 c0  ............%...
0012f918  01 00 00 00 8c f9 12 00 - 94 cc fa 77 00 00 00 00  ...........w....
0012f928  e8 56 13 00 f8 f9 12 00 - 64 f9 12 00 60 f9 12 00  .V......d...`...
0012f938  88 f9 12 00 00 00 00 00 - cc fc 12 00 00 00 00 00  ................
0012f948  48 31 f8 77 ff ff ff ff - a8 f6 12 00 14 00 00 00  H1.w............
0012f958  8c f9 12 00 1f 00 00 00 - 14 00 16 00 38 b4 16 00  ............8...
0012f968  00 00 13 00 00 70 12 00 - 00 00 00 00 e0 fc 12 00  .....p..........
0012f978  c6 1b f9 77 8c f9 12 00 - bc f9 12 00 8c f9 12 00  ...w............
0012f988  bc f9 12 00 de fa ed 0e - 01 00 00 00 00 00 00 00  ................
0012f998  b1 bc 59 7c 07 00 00 00 - e2 2e 68 00 e0 45 2a 01  ..Y|......h..E*.
0012f9a8  a8 3d 68 00 00 00 00 00 - d8 fd 12 00 4c fd 12 00  .=h.........L...
0012f9b8  14 fd 12 00 3f 00 01 00 - 00 00 00 00 00 00 00 00  ....?...........
0012f9c8  00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00  ................
0012f9d8  72 13 ff ff 00 01 ff ff - ff ff ff ff 02 74 40 00  r............t@.
0012f9e8  1b 00 3a 07 74 ba 29 01 - 23 00 ff ff 00 00 00 00  ..:.t.).#.......
0012f9f8  00 00 00 00 00 00 00 d4 - 01 d0 01 e8 00 80 34 40  ..............4@
0012fa08  00 d2 00 da 00 ca 00 de - 35 40 00 c8 00 e6 00 40  ........5@.....@
0012fa18  00 e8 35 40 00 e6 00 e8 - 00 c2 00 dc 35 40 00 cc  ..5@........5@..
0012fa28  00 de 00 ea 00 dc 35 40 - 00 de 00 ea 00 dc 00 c8  ......5@........
0012fa38  35 40 00 50 01 a0 01 94 - 01 80 34 40 00 00 00 00  5@.P......4@....

State Dump for Thread Id 0x3fc

eax=77d37dd0 ebx=00145560 ecx=0012dce8 edx=00000000 esi=00145400 edi=00000100
eip=77f88b37 esp=015bfe28 ebp=015bff74 iopl=0         nv up ei pl nz na po nc
cs=001b  ss=0023  ds=0023  es=0023  fs=0038  gs=0000             efl=00000206


function: ZwReplyWaitReceivePortEx
        77f88b2c b8ac000000       mov     eax,0xac
        77f88b31 8d542404         lea     edx,[esp+0x4]          ss:02049d0f=????????
        77f88b35 cd2e             int     2e
        77f88b37 c21400           ret     0x14
        77f88b3a 8bff             mov     edi,edi

*----> Stack Back Trace <----*

FramePtr ReturnAd Param#1  Param#2  Param#3  Param#4  Function Name
015BFF74 77D58E4A 77D3AEED 00145400 77D50101 00130000 ntdll!ZwReplyWaitReceivePortEx
015BFFA8 77D37DE8 00144C38 015BFFEC 7C57B3BC 00145560 rpcrt4!TowerConstruct
015BFFB4 7C57B3BC 00145560 77D50101 00130000 00145560 rpcrt4!I_RpcConnectionInqSockBuffSize2
015BFFEC 00000000 00000000 00000000 00000000 00000000 kernel32!lstrcmpiW

State Dump for Thread Id 0x2f8

eax=7ce20000 ebx=00000102 ecx=00147b28 edx=00000000 esi=77f88398 edi=016bff74
eip=77f883a3 esp=016bff60 ebp=016bff7c iopl=0         nv up ei pl nz na po nc
cs=001b  ss=0023  ds=0023  es=0023  fs=0038  gs=0000             efl=00000206


function: NtDelayExecution
        77f88398 b832000000       mov     eax,0x32
        77f8839d 8d542404         lea     edx,[esp+0x4]          ss:02149e47=????????
        77f883a1 cd2e             int     2e
        77f883a3 c20800           ret     0x8
        77f883a6 8bff             mov     edi,edi

*----> Stack Back Trace <----*

FramePtr ReturnAd Param#1  Param#2  Param#3  Param#4  Function Name
016BFF7C 7C59A2CC 0000EA60 00000000 7CE89ACC 0000EA60 ntdll!NtDelayExecution
00007530 00000000 00000000 00000000 00000000 00000000 kernel32!Sleep

*----> Raw Stack Dump <----*
016bff60  01 a3 59 7c 00 00 00 00 - 74 ff 6b 01 31 a0 59 7c  ..Y|....t.k.1.Y|
016bff70  28 7b 14 00 00 ba 3c dc - ff ff ff ff 30 75 00 00  ({....<.....0u..
016bff80  cc a2 59 7c 60 ea 00 00 - 00 00 00 00 cc 9a e8 7c  ..Y|`..........|
016bff90  60 ea 00 00 a7 5b e8 7c - 00 00 00 00 00 00 e2 7c  `....[.|.......|
016bffa0  28 7b 14 00 ec ff 6b 01 - 28 7b 14 00 ff 5a e8 7c  ({....k.({...Z.|
016bffb0  85 7d e3 7c 70 7d e3 7c - bc b3 57 7c 28 7b 14 00  .}.|p}.|..W|({..
016bffc0  85 7d e3 7c 70 7d e3 7c - 28 7b 14 00 00 c0 fd 7f  .}.|p}.|({......
016bffd0  62 00 64 00 c0 ff 6b 01 - 62 00 64 00 ff ff ff ff  b.d...k.b.d.....
016bffe0  60 21 5c 7c 08 2b 57 7c - 00 00 00 00 00 00 00 00  `!\|.+W|........
016bfff0  00 00 00 00 e5 5a e8 7c - 28 7b 14 00 00 00 00 00  .....Z.|({......
016c0000  c8 00 00 00 00 01 00 00 - ff ee ff ee 02 10 00 00  ................
016c0010  00 00 00 00 00 fe 00 00 - 00 00 10 00 00 20 00 00  ............. ..
016c0020  00 02 00 00 00 20 00 00 - 9e 01 00 00 ff ef fd 7f  ..... ..........
016c0030  06 00 08 06 00 00 00 00 - 00 00 00 00 00 00 00 00  ................
016c0040  00 00 00 00 98 05 6c 01 - 0f 00 00 00 f8 ff ff ff  ......l.........
016c0050  50 00 6c 01 50 00 6c 01 - 40 06 6c 01 00 00 00 00  P.l.P.l.@.l.....
016c0060  00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00  ................
016c0070  00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00  ................
016c0080  00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00  ................
016c0090  00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00  ................

State Dump for Thread Id 0x41c

eax=778321fe ebx=00000003 ecx=0012f258 edx=00000000 esi=77f88ef8 edi=00000003
eip=77f88f03 esp=0182fd24 ebp=0182fd70 iopl=0         nv up ei pl zr na po nc
cs=001b  ss=0023  ds=0023  es=0023  fs=0038  gs=0000             efl=00000246


function: NtWaitForMultipleObjects
        77f88ef8 b8e9000000       mov     eax,0xe9
        77f88efd 8d542404         lea     edx,[esp+0x4]          ss:022b9c0b=????????
        77f88f01 cd2e             int     2e
        77f88f03 c21400           ret     0x14
        77f88f06 8bff             mov     edi,edi

*----> Stack Back Trace <----*

FramePtr ReturnAd Param#1  Param#2  Param#3  Param#4  Function Name
0182FD70 7C59A180 0182FD48 00000001 00000000 00000000 ntdll!NtWaitForMultipleObjects
0182FFB4 7C57B3BC 00000004 000B000A 7C3280D0 0016D3D0 kernel32!WaitForMultipleObjects
0182FFEC 00000000 00000000 00000000 00000000 00000000 kernel32!lstrcmpiW


Version 2.4 beta05 seems to work fine on vanilla XP SP3 (no POS hack).

Edit: 2.4beta05 crashes with the same report on my old W2K, too. Think it's end of the line and will keep using 2.3m.

Minimum OS is now Windows XP SP3!
Title: Re: version 2.4
Post by: TEA-Time on May 27, 2020, 01:04:58 AM

i just tried but cannot reproduce this problem. Let me know if you find more on it.


ok i'll see to fix it for next release


oh man, sorry but i wouldn't use HFS for such job, as HTTP doesn't support resume on upload. Unless it's in LAN, and it takes few minutes.
TBH i suspect nowadays upload-resume could be implemented via javascript. Maybe for a future release.


I'm using 2.4 beta 5 now and the same thing is happening.  It happens in Chrome, Firefox, and Vivaldi (Chrome based).  All those things appear to be related to the address bar still having the "?mode=auth&u=...." part.  After clicking a folder button or link, that part is removed and everything starts working fine.


We've transferred large files before and I'm happy to report we have had no issues.  :D


I created a rejetto folder and user and restricted it to just that folder with full access, and the log below is from just the login process.  I will message you the link to log into it to see what happens if you'd like.  Don't worry, nothing malicious in there.  8)   Feel free to upload, create a folder, delete, whatever.


Code: [Select]
5/26/2020 5:48:44 PM Server start
5/26/2020 5:49:24 PM <IP address>:56202 Connected
5/26/2020 5:49:24 PM <IP address>:56203 Connected
5/26/2020 5:49:24 PM <IP address>:56203 Got 444 bytes
5/26/2020 5:49:24 PM rejetto@<IP address>:56203 Requested GET <login link with credentials removed>
5/26/2020 5:49:24 PM rejetto@<IP address>:56203 Sent 1460 bytes
5/26/2020 5:49:24 PM rejetto@<IP address>:56203 Served 2.7 K
5/26/2020 5:49:25 PM rejetto@<IP address>:56203 Requested GET /?mode=section&id=style.css
5/26/2020 5:49:25 PM <IP address>:56204 Connected
5/26/2020 5:49:25 PM rejetto@<IP address>:56203 Redirected to /rejetto
5/26/2020 5:49:25 PM <IP address>:56205 Connected
5/26/2020 5:49:25 PM <IP address>:56204 Got 435 bytes
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Requested GET /?mode=jquery
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Sent 1460 bytes
5/26/2020 5:49:25 PM <IP address>:56206 Connected
5/26/2020 5:49:25 PM <IP address>:56205 Got 446 bytes
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Served 42.5 K
5/26/2020 5:49:25 PM <IP address>:56206 Got 445 bytes
5/26/2020 5:49:25 PM rejetto@<IP address>:56206 Requested GET /rejetto
5/26/2020 5:49:25 PM rejetto@<IP address>:56206 Sent 60 bytes
5/26/2020 5:49:25 PM rejetto@<IP address>:56206 Not served: 301 - Moved permanently
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Requested GET /~img_folder
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Served 562
5/26/2020 5:49:25 PM rejetto@<IP address>:56203 Disconnected by server - 2966 bytes sent
5/26/2020 5:49:25 PM rejetto@<IP address>:56206 Disconnected by server - 60 bytes sent
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Requested GET /rejetto/
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Served 2.7 K
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Login failed
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Requested GET /favicon.ico
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Not served: 403 - Deny
5/26/2020 5:49:25 PM rejetto@<IP address>:56204 Disconnected by server - 47381 bytes sent
5/26/2020 5:49:29 PM <IP address>:56202 Disconnected


And here is the log from clicking the "Test folder" folder at which time the "?mode=auth&u=...." part is removed, then everything starts working correctly:


Code: [Select]
5/26/2020 5:50:33 PM <IP address>:56214 Connected
5/26/2020 5:50:33 PM <IP address>:56215 Connected
5/26/2020 5:50:33 PM <IP address>:56215 Got 546 bytes
5/26/2020 5:50:33 PM rejetto@<IP address>:56215 Requested GET /rejetto/Test folder/
5/26/2020 5:50:33 PM rejetto@<IP address>:56215 Sent 1460 bytes
5/26/2020 5:50:33 PM rejetto@<IP address>:56215 Served 2.6 K
5/26/2020 5:50:33 PM <IP address>:56216 Connected
5/26/2020 5:50:33 PM <IP address>:56217 Connected
5/26/2020 5:50:33 PM <IP address>:56216 Got 400 bytes
5/26/2020 5:50:33 PM rejetto@<IP address>:56216 Requested GET /?mode=jquery
5/26/2020 5:50:33 PM rejetto@<IP address>:56216 Not modified, use cache
5/26/2020 5:50:33 PM <IP address>:56217 Got 411 bytes
5/26/2020 5:50:33 PM rejetto@<IP address>:56217 Requested GET /?mode=section&id=lib.js
5/26/2020 5:50:33 PM rejetto@<IP address>:56217 Not modified, use cache
5/26/2020 5:50:33 PM rejetto@<IP address>:56216 Disconnected by server
5/26/2020 5:50:33 PM rejetto@<IP address>:56217 Disconnected by server
5/26/2020 5:50:38 PM <IP address>:56214 Disconnected
Title: Re: version 2.4
Post by: TEA-Time on May 27, 2020, 05:09:38 AM
It's something in my configuration.  I tested a clean install on another computer and it worked fine.  So I renamed hfs.ini out of the way on the original computer and it's working there too.

So I'm trying to narrow it down.


Edit: Nothing is standing out.  Even if I temporarily reset the options in the Debug menu, it doesn't help.  Only renaming hfs.ini out of the way makes a difference.  I need to get to bed.  Let me know if I should PM you my hfs.ini file.
Title: Re: version 2.4
Post by: rejetto on May 27, 2020, 10:25:05 AM
I created a rejetto folder and user and restricted it to just that folder with full access, and the log below is from just the login process.  I will message you the link to log into it to see what happens if you'd like.  Don't worry, nothing malicious in there.  8)   Feel free to upload, create a folder, delete, whatever.

your link was precious because i've found ANOTHER bug.   ;D

If the account has a redirection, the mode=auth doesn't work well, you get the redirection on the next request (style.css).
I've decided that redirection should not take place at all with an auth link. It will work only with manual login. Ok?

I'll publish another beta later today. Already collected 6 small changes.

So I'm trying to narrow it down.
Edit: Nothing is standing out.  Even if I temporarily reset the options in the Debug menu, it doesn't help.  Only renaming hfs.ini out of the way makes a difference.  I need to get to bed.  Let me know if I should PM you my hfs.ini file.

i don't think the ini would be enough
Title: Re: version 2.4
Post by: NaitLee on May 27, 2020, 02:37:08 PM
The Unicode problem we dealt in alpha is still exists in beta5's strings(filenames) of upload, make folder, renaming etc.
Title: Re: version 2.4
Post by: rejetto on May 27, 2020, 03:04:52 PM
i just tested upload and comments with chinese characters, and had no problems.
Can you be more specific?
Title: Re: version 2.4
Post by: TEA-Time on May 27, 2020, 06:06:44 PM
your link was precious because i've found ANOTHER bug.   ;D


Excellent!  Well it's an honor having you log into my server.  ;D

Quote
If the account has a redirection, the mode=auth doesn't work well, you get the redirection on the next request (style.css).


Yup, that was it!  I thought for sure I'd suspected and tested without the redirect, but apparently not.  I've removed the redirect from your account and you can see that it works now.  Color me red from embarrassment.  :-[


Quote
I've decided that redirection should not take place at all with an auth link. It will work only with manual login. Ok?


Yes, that actually occurred to me since the auth link already has a specific folder in it, but I forgot to mention it.

Quote
I'll publish another beta later today. Already collected 6 small changes.


Cool, I will test it this evening.

Quote
i don't think the ini would be enough


Alright, I think we'll be good after this anyway.




Oh, one more thing.  If I log in with an auth link, then log out without doing anything else in between that removes the auth part, it just automatically logs back in since it's still an auth link.  Would it make sense and would it be possible to just make the auth part go away while logging in?


And a feature suggestion.  When banning an IP address, maybe a button when entering the Comment to insert the date and time?
Title: Re: version 2.4
Post by: rejetto on May 27, 2020, 06:37:16 PM
Oh, one more thing.  If I log in with an auth link, then log out without doing anything else in between that removes the auth part, it just automatically logs back in since it's still an auth link.  Would it make sense and would it be possible to just make the auth part go away while logging in?

ah, you are right, find this fix in this release

https://github.com/rejetto/hfs2/releases/tag/v2.4-beta06
Title: Re: version 2.4
Post by: TEA-Time on May 28, 2020, 03:42:46 AM
v2.4 Beta 6 looks good!  No initially missing style sheet now.  :D


Now a question as to how things are supposed to work.  I added the redirect back to the rejetto account, which sends it to \rejetto, the only folder it has access to.  But if I try to manually log in at the root, I just get prompted to log in again instead of being redirected to the \rejetto folder.  Is that working as designed?  I kind of expected the redirect to happen, but at least I probably won't actually have anyone trying to do that.


And I think I found another bug.  If I click within or highlight the IP address part of a "user@xxx.xxx.xxx.xxx" entry in the log and try to ban it, the IP address field is blank (not that I would actually want to ban the IP address of someone who has an account. heh).
Title: Re: version 2.4
Post by: NaitLee on May 28, 2020, 04:26:37 AM
(I'm rushing to leave when doing previous post, in a hurry.)
The last-utf-char-corruption still exists, but this cannot be seen on rejetto's computer.

Uploading a file with specified last utf8 char can reproduce the problem on most (other) computers.

eg. 哲学.ppt -> 哲�?ppt    (note that dot disappeared too)  Then "filename is forbidden".

This seems can be fixed with the (almost) same way we used in alpha to the uploading function.

Maybe file names on windows are in ansi but not utf-8.

Another thing: Most of the templates of version 2.3 (including mine) seems cannot work correctly in 2.4, with bad file action functions. Should this solved by template makers or rejetto? ::)

Problems found in Takeback:
 Uploads a blank file, no uploading results are shown (showing a filelist page while a POST request)
 Renaming & moving do not work
 Bad session when creating a folder
Title: Re: version 2.4
Post by: rejetto on May 28, 2020, 01:28:47 PM
Now a question as to how things are supposed to work.  I added the redirect back to the rejetto account, which sends it to \rejetto, the only folder it has access to.  But if I try to manually log in at the root, I just get prompted to log in again instead of being redirected to the \rejetto folder.  Is that working as designed?  I kind of expected the redirect to happen, but at least I probably won't actually have anyone trying to do that.

you are right. Fixed in next release (https://github.com/rejetto/hfs2/commit/ad33cbf263180d5713bce22d7fd102a5009549b4).

Quote
And I think I found another bug.  If I click within or highlight the IP address part of a "user@xxx.xxx.xxx.xxx" entry in the log and try to ban it, the IP address field is blank (not that I would actually want to ban the IP address of someone who has an account. heh).

right. Fixed too.
Title: Re: version 2.4
Post by: rejetto on May 28, 2020, 02:48:23 PM
The last-utf-char-corruption still exists, but this cannot be seen on rejetto's computer.
Uploading a file with specified last utf8 char can reproduce the problem on most (other) computers.
eg. 哲学.ppt -> 哲�?ppt    (note that dot disappeared too)  Then "filename is forbidden".

yes, i tried this specific file name and it's working here.
What Windows version are you using?
I see in the screenshot that you are using a chinese translated HFS.
Can you try a clean one, without any customization, tpl, etc?

Quote
This seems can be fixed with the (almost) same way we used in alpha to the uploading function.

what are you talking about? i don't remember.

Quote
Maybe file names on windows are in ansi but not utf-8.

modern windows uses utf-16 internally, AFAIK.
it's me then converting it to utf-8 for some uses.

Quote
Another thing: Most of the templates of version 2.3 (including mine) seems cannot work correctly in 2.4, with bad file action functions. Should this solved by template makers or rejetto? ::)

i expect 2.3 compatible templates to work, and I'll do my best.
Anyway, i tested takeback 0.13b, and it's using the experimental and undocumented _PASSWORD_MD5.
I don't think it's worth keeping it.
I thought you had it updated to the new method. Maybe I saw an old version? If yes, can you link me the new one?

Quote
Problems found in Takeback:
 Uploads a blank file, no uploading results are shown (showing a filelist page while a POST request)
 Renaming & moving do not work
 Bad session when creating a folder

I just tried old 2.3 template, created a folder, uploaded a blank file (0 bytes), no problems found.
While testing takeback I saw a 'bad session'. Looked at the SID, and saw a '+' symbol, that was interpreted as a blank space by chrome while resending it.
I didn't dig it deeply, decided to just avoid it, and from next release the SID will have only alphanumerics.
But now I just cannot log in because of the problem i said above.
Title: Re: version 2.4
Post by: rejetto on May 28, 2020, 02:55:30 PM
all the fixes afore mentioned at https://github.com/rejetto/hfs2/releases/tag/v2.4-beta07b

i also added support for get https resources, for example through {.load.}
HFS still doesnt' support serving through https.
(thanks Rapid)
Title: Re: version 2.4
Post by: rejetto on May 28, 2020, 04:55:15 PM
i just noticed that the old template with its old login system is not working anymore.
It should work in next release.

It stopped working with alpha 8. I think at the time I made my tests with the mod that applied the new login to the old template.
Title: Re: version 2.4
Post by: TEA-Time on May 29, 2020, 02:30:03 AM
you are right. Fixed in next release (https://github.com/rejetto/hfs2/commit/ad33cbf263180d5713bce22d7fd102a5009549b4).

Confirmed fixed in v2.4 beta7.



Uhoh, now the IP address field is always blank when I try to ban one from the log.  :-\
Title: Re: version 2.4
Post by: NaitLee on May 29, 2020, 04:20:49 AM
@rejetto:

The translations are just because of the hfs.lng file.

You have once solved the Unicode problem around here: http://rejetto.com/forum/index.php?topic=13288.msg1065652#msg1065652
Try to callback what the method you have used :D

For other things we mentioned I will do more actions a few days later to decide what to do next.
Title: Re: version 2.4
Post by: ivominchev on May 29, 2020, 05:47:21 AM
I want to test 2.4 but can't see any .exe file. Help?
Title: Re: version 2.4
Post by: TEA-Time on May 29, 2020, 05:56:12 AM
I want to test 2.4 but can't see any .exe file. Help?

First post in this thread. ;)

https://rejetto.com/forum/index.php?topic=13060.0 (https://rejetto.com/forum/index.php?topic=13060.0)

Edit: Oops, he corrected the more recent post, but not the first one.  Here's the working link.

https://github.com/rejetto/hfs2/releases/tag/v2.4-beta07b (https://github.com/rejetto/hfs2/releases/tag/v2.4-beta07b)
Title: Re: version 2.4
Post by: ivominchev on May 29, 2020, 07:23:49 AM
Thanks. I like the new options and view.
Title: Re: version 2.4
Post by: NaitLee on May 29, 2020, 02:30:07 PM
A truth that cannot beat: HFS is going better and better!



I found another thing: I see the ok/cancel in port specifiction now can be effect by the hfs.lng translation file, but they seems have conflicted IDs to some built-in localization type strings.

See the snapshot, or place the English (unchanged) hfs.lng to HFS folder and relaunch to see.
Title: Re: version 2.4
Post by: TEA-Time on May 29, 2020, 04:52:39 PM
Thanks. I like the new options and view.

You're welcome.  Yes, excellent improvements to an already awesome program. :)
Title: Re: version 2.4
Post by: LeoNeeson on May 30, 2020, 07:25:05 AM
I found another thing: I see the ok/cancel in port specifiction now can be effect by the hfs.lng translation file, but they seems have conflicted IDs to some built-in localization type strings.
@Rejetto: This reminds me that it could come handy if you can generate a new "hfs.lng" base (english) file, based on the latest build, to make it available for those who want to translate HFS. Maybe you can also upload it to GitHub, to make it easier to find. (I don't need this, I only comment you this as a reminder to you. ;))
Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 09:26:13 AM
Uhoh, now the IP address field is always blank when I try to ban one from the log.  :-\

i tested with several types and couldn't reproduce your problem.
Please paste here a line from the log that gives you problems. You can replace 2-3 digits of the IP for privacy.
Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 09:43:11 AM
The translations are just because of the hfs.lng file.

i know, but since you are having unicode problems the tests must as clean as possible, please.
Otherwise we don't know if the HFS.LNG itself can be the problem.

Quote
You have once solved the Unicode problem around here: http://rejetto.com/forum/index.php?topic=13288.msg1065652#msg1065652
Try to callback what the method you have used :D

ok, i see what you mean, but the fix is still there, it was never removed or changed.

I found another thing: I see the ok/cancel in port specifiction now can be effect by the hfs.lng translation file, but they seems have conflicted IDs to some built-in localization type strings

ok, i cannot find those IDs in the new hfs.lng i will now publish. I guess you were working with an old version. Please have a look at the new one.
Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 09:44:11 AM
@Rejetto: This reminds me that it could come handy if you can generate a new "hfs.lng" base (english) file, based on the latest build, to make it available for those who want to translate HFS. Maybe you can also upload it to GitHub, to make it easier to find. (I don't need this, I only comment you this as a reminder to you. ;))

sure! somebody let me know if this file is good, then we'll decide how to publish it
Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 10:42:44 AM
I've found there are some serious problems making a login system compatible with the old one.
I fear we'll have to say goodbye to old templates, unless they are adapted for the new version.
Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 01:39:51 PM
there is a huge problem in beta 7 in case the random SID has the character "D".
Title: Re: version 2.4
Post by: TEA-Time on May 30, 2020, 01:43:00 PM

i tested with several types and couldn't reproduce your problem.
Please paste here a line from the log that gives you problems. You can replace 2-3 digits of the IP for privacy.


None of the types here work.  I even renamed the ini file out of the way and created a basic account, turned on a bunch of logging, then logged in and back out.


Code: [Select]
5/30/2020 6:38:34 AM Server start
5/30/2020 6:39:07 AM 192.168.1.100:49549 Connected
5/30/2020 6:39:07 AM 192.168.1.100:49549 Got 427 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49549 Sent 477 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49549 Disconnected by server - 477 bytes sent
5/30/2020 6:39:07 AM 192.168.1.100:49550 Connected
5/30/2020 6:39:07 AM 192.168.1.100:49551 Connected
5/30/2020 6:39:07 AM 192.168.1.100:49552 Connected
5/30/2020 6:39:07 AM 192.168.1.100:49550 Got 391 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49551 Got 362 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49551 Requested GET /?mode=jquery
5/30/2020 6:39:07 AM 192.168.1.100:49552 Got 373 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49551 Sent 1460 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49551 Served 42.5 K
5/30/2020 6:39:07 AM 192.168.1.100:49552 Sent 9073 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49552 Disconnected by server - 9569 bytes sent
5/30/2020 6:39:07 AM 192.168.1.100:49551 Not modified, use cache
5/30/2020 6:39:07 AM 192.168.1.100:49550 Sent 13459 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49553 Connected
5/30/2020 6:39:07 AM 192.168.1.100:49550 Not modified, use cache
5/30/2020 6:39:07 AM 192.168.1.100:49554 Connected
5/30/2020 6:39:07 AM 192.168.1.100:49553 Got 412 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49553 Not modified, use cache
5/30/2020 6:39:07 AM 192.168.1.100:49551 Disconnected by server - 43515 bytes sent
5/30/2020 6:39:07 AM 192.168.1.100:49553 Disconnected by server
5/30/2020 6:39:07 AM 192.168.1.100:49550 Disconnected by server - 13459 bytes sent
5/30/2020 6:39:07 AM 192.168.1.100:49554 Got 397 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49554 Sent 496 bytes
5/30/2020 6:39:07 AM 192.168.1.100:49554 Disconnected by server - 496 bytes sent
5/30/2020 6:39:12 AM 192.168.1.100:49555 Connected
5/30/2020 6:39:12 AM 192.168.1.100:49555 Got 599 bytes
5/30/2020 6:39:12 AM 192.168.1.100:49555 Requested POST /?mode=login
5/30/2020 6:39:12 AM 192.168.1.100:49555 Sent 22 bytes
5/30/2020 6:39:12 AM 192.168.1.100:49555 Served 22
5/30/2020 6:39:12 AM Test@192.168.1.100:49555 Requested GET /
5/30/2020 6:39:12 AM Test@192.168.1.100:49555 Served 2.4 K
5/30/2020 6:39:12 AM Test@192.168.1.100:49555 Not modified, use cache
5/30/2020 6:39:12 AM 192.168.1.100:49556 Connected
5/30/2020 6:39:12 AM 192.168.1.100:49557 Connected
5/30/2020 6:39:12 AM 192.168.1.100:49556 Got 401 bytes
5/30/2020 6:39:12 AM Test@192.168.1.100:49556 Not modified, use cache
5/30/2020 6:39:12 AM 192.168.1.100:49557 Got 412 bytes
5/30/2020 6:39:12 AM Test@192.168.1.100:49557 Not modified, use cache
5/30/2020 6:39:12 AM Test@192.168.1.100:49555 Disconnected by server - 2527 bytes sent
5/30/2020 6:39:12 AM Test@192.168.1.100:49557 Disconnected by server
5/30/2020 6:39:12 AM Test@192.168.1.100:49556 Disconnected by server
5/30/2020 6:39:12 AM 192.168.1.100:49558 Connected
5/30/2020 6:39:12 AM 192.168.1.100:49558 Got 397 bytes
5/30/2020 6:39:12 AM Test@192.168.1.100:49558 Requested GET /favicon.ico
5/30/2020 6:39:12 AM Test@192.168.1.100:49558 Sent 576 bytes
5/30/2020 6:39:12 AM Test@192.168.1.100:49558 Served 576
5/30/2020 6:39:18 AM Test@192.168.1.100:49558 Got 446 bytes
5/30/2020 6:39:18 AM 192.168.1.100:49558 Requested POST /?mode=logout
5/30/2020 6:39:18 AM 192.168.1.100:49558 Sent 22 bytes
5/30/2020 6:39:18 AM 192.168.1.100:49558 Served 22
5/30/2020 6:39:18 AM 192.168.1.100:49558 Disconnected by server - 1074 bytes sent
5/30/2020 6:39:19 AM 192.168.1.100:49559 Connected
5/30/2020 6:39:19 AM 192.168.1.100:49560 Connected
5/30/2020 6:39:19 AM 192.168.1.100:49561 Connected
5/30/2020 6:39:19 AM 192.168.1.100:49559 Got 433 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49560 Got 401 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49561 Got 412 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49559 Not modified, use cache
5/30/2020 6:39:19 AM 192.168.1.100:49560 Not modified, use cache
5/30/2020 6:39:19 AM 192.168.1.100:49561 Not modified, use cache
5/30/2020 6:39:19 AM 192.168.1.100:49559 Disconnected by server
5/30/2020 6:39:19 AM 192.168.1.100:49561 Disconnected by server
5/30/2020 6:39:19 AM 192.168.1.100:49560 Disconnected by server
5/30/2020 6:39:19 AM 192.168.1.100:49562 Connected
5/30/2020 6:39:19 AM 192.168.1.100:49562 Got 527 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49562 Sent 497 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49562 Disconnected by server - 497 bytes sent
5/30/2020 6:39:19 AM 192.168.1.100:49563 Connected
5/30/2020 6:39:19 AM 192.168.1.100:49564 Connected
5/30/2020 6:39:19 AM 192.168.1.100:49563 Got 433 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49565 Connected
5/30/2020 6:39:19 AM 192.168.1.100:49564 Got 401 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49563 Not modified, use cache
5/30/2020 6:39:19 AM 192.168.1.100:49565 Got 412 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49565 Not modified, use cache
5/30/2020 6:39:19 AM 192.168.1.100:49564 Not modified, use cache
5/30/2020 6:39:19 AM 192.168.1.100:49563 Disconnected by server
5/30/2020 6:39:19 AM 192.168.1.100:49564 Disconnected by server
5/30/2020 6:39:19 AM 192.168.1.100:49565 Disconnected by server
5/30/2020 6:39:19 AM 192.168.1.100:49566 Connected
5/30/2020 6:39:19 AM 192.168.1.100:49566 Got 397 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49566 Sent 497 bytes
5/30/2020 6:39:19 AM 192.168.1.100:49566 Disconnected by server - 497 bytes sent
Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 02:00:32 PM
ok, i can see the problem. In my "locale" the time has no 'AM' part.
I will find a solution.
Title: Re: version 2.4
Post by: TEA-Time on May 30, 2020, 02:19:40 PM
Ah, does that mean you use 24 hour time?

As you can see, I was up way too early for a Saturday. lol
Title: Re: version 2.4
Post by: Rapid on May 30, 2020, 02:26:31 PM
sure! somebody let me know if this file is good, then we'll decide how to publish it
I think one of a problems is that all resourcestrings in functions you named as MSG:
Code: [Select]
[ResourceStrings]
64820_main_MSG{1}=The current template is using macros.\^Do you want to cancel this action?
...
64820_main_MSG{1}=Max simultaneous addresses downloading.
...
64820_main_MSG{1}=Max lines on screen
...
64820_main_MSG{1}=Max lines on screen
64820_main_MSG{1}=Here you can specify how to format the log file complying Apache standard.\^Leave blank to get bare copy of screen on file.\^\^Example:\^   %h %l %u %t "%r" %>s %b
64820_main_MSG{1}=This option creates an .md5 file for every new calculated fingerprint.\^Use with care to get not your disk invaded by these files.

Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 02:32:48 PM
I think one of a problems is that all resourcestrings in functions you named as MSG:

I didn't really dig into it, as it was presented as an automatic solution.
i expected it to handle them correctly and give different IDs.
This means that if refactor the name it will break translation. Nice -_-
I guess i'll have to rename all local ones. Worse, I should globalize them to avoid collisions, probably. Or you have a different suggestion?
Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 03:32:22 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-beta08

- new templates (with new login system) will be recognized by the section [unauth] in place of [unauthorized]
- 'selection' was renamed to 'files'
- sha256 was moved to [sha256.js] section, so you don't need to include it in your tpl
Title: Re: version 2.4
Post by: TEA-Time on May 30, 2020, 04:50:50 PM
The ban IP address field looks good in all cases now.

Can the text cursor be automatically moved to where you right click in the log instead of having to left click to move it first, since the right click menu has to do where that is?
Title: Re: version 2.4
Post by: rejetto on May 30, 2020, 06:17:40 PM
Can the text cursor be automatically moved to where you right click in the log instead of having to left click to move it first, since the right click menu has to do where that is?

agreed
Title: Re: version 2.4
Post by: NaitLee on May 30, 2020, 11:24:25 PM
I see the resources can be translated now. Good job!

I also see the locale strings(Arabic, Baltic, Chinese etc.) appeard at the end of the file were "pushed" with different IDs, compaired with old hfs.lng. If they always have a different IDs each time they are generated, it will be annoying since we should work with them again and again.

Upload unicode problem is not solved in beta8 :(
Check the template side? (Though with little expection...)

Edit: I see this problem is there from alpha6, the version which general Unicode problem had been fixed. I did not tested it out that time, sorry...
Title: Re: version 2.4
Post by: LeoNeeson on May 31, 2020, 05:26:44 AM
sure! somebody let me know if this file is good, then we'll decide how to publish it
Good, thanks for the new file (I'm sure it will come handy for those who want to translate HFS). From time to time, there are some users who doesn't know how (http://rejetto.com/forum/index.php?topic=8433.msg1065837#msg1065837) to translate HFS, and finding the 'hfs.lng' on the middle of a long thread like this (http://rejetto.com/forum/index.php?topic=13060.msg1065836#msg1065836), make it sometimes hard to find (even for me!, because after a few weeks I forget where you have uploaded the file). You can upload it wherever you think is more visible for the users (it was only a suggestion to upload it on GitHub). :)

If they always have a different IDs each time they are generated, it will be annoying since we should work with them again and again.
I could be wrong, but I think this will only happen once (I mean, the new 'hfs.lng' has different IDs because Rejetto has updated the component that makes the translation possible). HFS v2.3m was making use of "Kryvich's Delphi Localizer v3.2", and Rejetto is now using "Kryvich's Delphi Localizer v4.1" which has full unicode support (that's why it could now have different IDs). But I think this would not happen again soon, or at least not very often (until Rejetto decides to update that component again on the future). Someone please correct me if I'm mistaken... ;)

Cheers,
Leo.-
Title: Re: version 2.4
Post by: rejetto on May 31, 2020, 01:30:16 PM
ok, dedicated topic for a start
http://rejetto.com/forum/index.php?topic=13313.new#new (http://rejetto.com/forum/index.php?topic=13313.new#new)

@naitlee, i've sent you an email
Title: Re: version 2.4
Post by: dj on May 31, 2020, 04:02:25 PM
I've found there are some serious problems making a login system compatible with the old one.
I fear we'll have to say goodbye to old templates, unless they are adapted for the new version.

additional manual add [api level] to template
tested with Takeback, Throwback and Live
Title: Re: version 2.4
Post by: Mars on May 31, 2020, 04:52:16 PM
for compatibility reasons, we can add this in diff

[unauth]
{.$unauthorized.}
Title: Re: version 2.4
Post by: ivominchev on May 31, 2020, 05:28:57 PM
This error came with beta08 and the did not start. I'm back to 07.
Title: Re: version 2.4
Post by: rejetto on May 31, 2020, 05:35:03 PM
do you have a file "hfs.lng" ?
it's probably an old version and not compatible.
Title: Re: version 2.4
Post by: ivominchev on May 31, 2020, 06:12:40 PM
Yes. Problem fixed.
Title: Re: version 2.4
Post by: rejetto on June 01, 2020, 09:00:30 AM
The last-utf-char-corruption still exists, but this cannot be seen on rejetto's computer.
Uploading a file with specified last utf8 char can reproduce the problem on most (other) computers.
eg. 哲学.ppt -> 哲�?ppt    (note that dot disappeared too)  Then "filename is forbidden".


thanks NaitLee for giving a server for tests.
I just uploaded 3 different files named  哲学, one with Chrome, one with Edge, the other with Firefox.
I also tried to change the comment. I saw no problems.
Is this test good? If yes, then the problem is not in HFS but in the browser you are using.

Title: Re: version 2.4
Post by: NaitLee on June 01, 2020, 10:12:27 AM
Is this test good? If yes, then the problem is not in HFS but in the browser you are using.
(Corrected sth and moved to the topic for unicode (http://rejetto.com/forum/index.php?topic=13315.0).)
Title: Re: version 2.4
Post by: rejetto on June 02, 2020, 12:24:09 PM
new version is out and unicode problem is fixed

https://github.com/rejetto/hfs2/releases/tag/v2.4-beta09

it's 2.4.0 now because i plan to use the third number for bugfixes releases, instead of letters (a,b,c) as in 2.3 . It's more standard.
Title: Re: version 2.4
Post by: ivominchev on June 02, 2020, 04:01:11 PM
Windows Defender reported Beta 09 as potentialy unwanted app PUA:Win32/Pearfoos.A!ml
The functions of the app works better if its started "As Administrator".
Title: Re: version 2.4
Post by: NaitLee on June 02, 2020, 05:03:50 PM
Windows Defender reported Beta 09 as potentialy unwanted app PUA:Win32/Pearfoos.A!ml
Windows/Microsoft Defender/SmartScreen almost always reports exes without a developer signature unwanted/non-trivial.

Some reports by Defender I had met:
 The beta version 2.4rc4 also gets a virus detection (Trojan:Win32/Fuery.C!cl)
 SilentPliz's translation version of HFS was reported a Trojan:Win32/Wacatac.C!ml
But all of them have no problems :) and seems that the trojan types are random...

So, do not care that report, ignore it and allow HFS or add HFS to whitelist :D
Title: Re: version 2.4
Post by: NaitLee on June 02, 2020, 05:40:02 PM
new version is out and unicode problem is fixed

https://github.com/rejetto/hfs2/releases/tag/v2.4-beta09

Oh yeah, that's almost perfect! :D

I see in default template the file actions are done by adding a <form> element to page, edit and submit it with jQuery,
 and the traditional ajax method of deleting a file from default template of 2.3m in Takeback cannot work (loaded with change of [unauth] and dj's adaption).
What should be done to this :D
Title: Re: version 2.4
Post by: dj on June 02, 2020, 06:15:28 PM
Oh yeah, that's almost perfect! :D

I see in default template the file actions are done by adding a <form> element to page, edit and submit it with jQuery,
 and the traditional ajax method of deleting a file from default template of 2.3m in Takeback cannot work (loaded with change of [unauth] and dj's adaption).
What should be done to this :D
now I realized
what rejetto meant by:
- 'selection' was renamed to 'files (http://rejetto.com/forum/index.php?topic=13060.msg1065847#msg1065847)

fixed in my template (http://rejetto.com/forum/index.php?topic=11754.msg1065895#msg1065895)
Title: Re: version 2.4
Post by: NaitLee on June 03, 2020, 02:16:23 AM
I found that in Wine/Linux, after minimizing HFS beta9 to tray, I cannot call it up.
It serves files normally.
Title: Re: version 2.4
Post by: rejetto on June 03, 2020, 09:00:05 AM
by the way, minimizing to tray is no longer the default option for HFS.
Title: Re: version 2.4
Post by: MelkiyProd. on June 03, 2020, 02:45:33 PM
When you move mouse to file's icon at list of VFS
Title: Re: version 2.4
Post by: rejetto on June 03, 2020, 03:21:19 PM
When you move mouse to file's icon at list of VFS

thanks, fixed in next release
Title: Re: version 2.4
Post by: rejetto on June 04, 2020, 09:09:38 AM
i waited a bit to see if there were more bug reports, but since the last bug can be quite annoying i'm releasing this just for it
https://github.com/rejetto/hfs2/releases/tag/v2.4-beta10
Title: Re: version 2.4
Post by: MarkV on June 04, 2020, 10:33:24 PM
Small note regarding the representation of IPv6 in address lines and logs. As you know, IPv6 uses colons as separators, unless IPv4, which uses dots. In order to solve the ambiguity that arises with appended ports (which, stupidly, also use a colon as separator), literal IPv6 addresses are written inside of square brackets "[" and "]".

HFS shows the IPv6 correctly in its address line, but incorrectly in the log. The square brackets need to be added to logged IPv6 addresses, too. Otherwise, serious confusion might arise while analyzing the HFS log, especially with shorted IPv6 addresses.


Example: 2001:db8::7:2:8080 is bad to read, as you won't know if the port is part of the IPv6 address or not. Showing [2001:db8::7:2]:8080 is much better, and, by the way, the only correct syntax.


That's my 2 cents for today.  ;)


Edit: Btw, does the Boru_V1 theme have a few image errors or is it my browser?

https://rejetto.com/forum/Themes/Boru_v1/images/useron.gif

The image above will only display an error message.
Title: Re: version 2.4
Post by: Chris Harris on June 05, 2020, 02:22:33 AM
I have loaded beta 9 and I found Windows 10 Defender said it had a virus and took it off my system. I have seen this on earlier versions.

Is here anything I can do to stop hfs being quarantined?
Chris
Title: Re: version 2.4
Post by: NaitLee on June 05, 2020, 02:41:16 AM
Is here anything I can do to stop hfs being quarantined?

Click: Windows Defender in tray -> Virus & threat Protection (the shield on the left side) -> Below the "virus & threat protection settings" header, "Manage Settings" -> scroll down, see "exclusions" header, click "add or delete exclusions" -> add an exclusion, Process, input "hfs.exe" "hfs"; or add a folder you want to download/use hfs.exe every time -> Done ;)

Edit: seems the 2.4 process name is "hfs" rather than "hfs.exe". Please use that.

After this seems SmartScreen still reports something.  You can ignore them as they are easy to deal.
Title: Re: version 2.4
Post by: NaitLee on June 05, 2020, 05:15:25 PM
Files with commas in its name cannot be downloaded in Google Chrome, with the error ERR_RESPONSE_HEADERS_MULTIPLE_CONTENT_DISPOSITION.

In most other servers this is an already-solved problem. Please refer to here in Stack Overflow (https://stackoverflow.com/questions/13578428/duplicate-headers-received-from-server) for the solution.
Title: Re: version 2.4
Post by: Mars on June 05, 2020, 05:53:13 PM
under firefox the file name is truncated from the semicolon when download it,
on the other hand this phenomenon is not present with hfs 2.3m
Title: Re: version 2.4
Post by: NaitLee on June 06, 2020, 05:19:25 PM
How about having a encryption when we change our password?
(This can be put off, since HFS is open source, and the method we use should be careful.)

(@rejetto: seems that you're busy these days, or feeling hard to deal things mentioned my me. just do what you want, be patient, try not feel stressed about these things, be your best!)
Title: Re: version 2.4
Post by: Mars on June 06, 2020, 05:44:51 PM
sorry for the explanations of the link which are in french

https://benjamin-balet.info/developpement/chiffrement-rsa-partiel-en-golang-javascript/

you can use a subterfuge by encrypting the transmission of the password using a private key and a public key

encoding will be done on the client side in javascrypt

on the server side you have to see if it is feasible from macro in a specialized script or using an external program, otherwise it could study an additional macro to include in the exe
Title: Re: version 2.4
Post by: NaitLee on June 06, 2020, 06:09:15 PM
Seems simple (or complex, depending on our math skills) encryption can be done in just a template (client/page: javascript, server/template: math macros (https://rejetto.com/wiki/index.php?title=HFS:_scripting_commands#Data_manipulation) in changepwd ajax)
Title: Re: version 2.4
Post by: rejetto on June 06, 2020, 06:16:43 PM
Small note regarding the representation of IPv6 in address lines and logs. As you know, IPv6 uses colons as separators, unless IPv4, which uses dots. In order to solve the ambiguity that arises with appended ports (which, stupidly, also use a colon as separator), literal IPv6 addresses are written inside of square brackets "[" and "]".


suggestion accepted.
Apparently apache log don't require changes instead.



Quote
Edit: Btw, does the Boru_V1 theme have a few image errors or is it my browser?
https://rejetto.com/forum/Themes/Boru_v1/images/useron.gif (https://rejetto.com/forum/Themes/Boru_v1/images/useron.gif)
The image above will only display an error message.


yes, it's broken. I may fix it someday.

I have loaded beta 9 and I found Windows 10 Defender said it had a virus and took it off my system. I have seen this on earlier versions.


here it comes again -_-


i have less free time now guys, so be patient
Title: Re: version 2.4
Post by: rejetto on June 07, 2020, 08:58:41 AM
Files with commas in its name cannot be downloaded in Google Chrome, with the error ERR_RESPONSE_HEADERS_MULTIPLE_CONTENT_DISPOSITION.

thanks for reporting. Fixed in next release.
Title: Re: version 2.4
Post by: rejetto on June 07, 2020, 09:15:58 AM
How about having a encryption when we change our password?
(This can be put off, since HFS is open source, and the method we use should be careful.)

i overlooked this problem, and it's not as easy as the login.
The link posted by mars should be technically ok but requires generation of keys using another program (openssl), and it takes several seconds so it should be done before.
I will postpone this problem for a later version (2.5 or later), and for the time being i'm just adding a warning  about the low security of the action.
Title: Re: version 2.4
Post by: rejetto on June 07, 2020, 10:57:51 AM
as an alternative method, we can just use the old password to encrypt the new one (no openssl required).
Title: Re: version 2.4
Post by: Mars on June 07, 2020, 02:53:38 PM
this is what I tried to do last night but I have a problem of transferring values by using a form to process the data

for now I explore the way to use ask () by inserting three input fields instead of one, which would simplify the transfer problem that I am facing



an attempt with the attached template, but I am faced with the processing of the form to send it to the section as with ASK, it follows that the data is found in the url



Title: Re: version 2.4
Post by: Mars on June 07, 2020, 06:46:52 PM
the exchange of data between the form and the section is now under control, it only remains to implement the encryption protocols...
Title: Re: version 2.4
Post by: rejetto on June 07, 2020, 10:29:06 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-rc01

(again) first release candidate :)

mars, i appreciate your effort, but the encryption was the only problem
Title: Re: version 2.4
Post by: dj on June 08, 2020, 04:13:33 AM
a bug in rc1:
not all requests are loaded (I've no limits set, 4 requests, 1 request denied with HTTP 429 Too Many Requests)

beta08 works fine
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 07:55:05 AM
how can i reproduce your problem?
Title: Re: version 2.4
Post by: dj on June 08, 2020, 08:20:19 AM
it happens with my template (https://github.com/dj0001/hfs-template/releases/latest)

it doesn't happen with beta08. beta9 and 10 I haven't tested.
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 09:05:37 AM
templates like yours are a quite special case, with the main section not requiring the file list, because it does it later with a further request.
What happens now is that you actually cause the listing to be done twice on the server, with the first one just discarded without being used.
To support cases like this (btw, same method used by hfs3 project) I will see to introduce a section-flag to let you specify that you don't need the %list%.
I considered having this automatic without flag, but it would require the server to execute the template to know if the %list% is used or not, because you can include other sections at run-time. I want it to be done before execution.

p.s. TBH, before this post i had tested your tpl, but by accident i ran an older hfs, so i saw no problems
Title: Re: version 2.4
Post by: MarkV on June 08, 2020, 09:46:07 AM
as an alternative method, we can just use the old password to encrypt the new one (no openssl required).
You can also use the current wall clock time string from the system, as a source of randomness. Could also be used to salt the password if necessary.
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 09:58:27 AM
ok dj, RC2 should make your template work without changes in a non-very-busy server.
To have perfect support also on busy servers you should  apply these changes
1. add this text to the very top: [|no list]
2. support error 429 when you fetch, and retry the fetch.
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 09:59:33 AM
You can also use the current wall clock time string from the system, as a source of randomness. Could also be used to salt the password if necessary.

thanks mark but it's not randomness we need for this (salt excluded)
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 10:13:49 AM
i'll better explain this new anti-DoS mechanism i just introduced:

since HFS listing of folders can be heavy (especially with its stupid programming language ;D) there is a limiter for these kind of requests:
1) 1 ip can make 1 request at a time, otherwise it is postponed
2) after finishing, the same ip cannot make another request for 1 second.
3) there is a limit of 3 concurrent folder listings. Having more would slow the users' experience uselessly.

In next release (RC2) I'm improving rule#2 by applying the 1 second only if the server is very busy (that is: having the max number of concurrent listings).

At the moment there is no 'queue', the browser is just told to try later (and it will automatically in a couple of seconds). This relieves the server from some burden.

If necessary i will introduce an option to configure this behavior, we'll see.

Other requests, like for files or icons, are not subject to this limiter.
Title: Re: version 2.4
Post by: NaitLee on June 08, 2020, 03:04:26 PM
The 429 code also happens in Takeback (0.14b) with background image function enabled (by editing tpl file). Only in RC1.
This function also requires a filelist, which acts similar to the "get list" in default template of 2.3.
Thus the page bg is always blank without an image.

Seems this anti-DoS is too strict...
Title: Re: version 2.4
Post by: NaitLee on June 08, 2020, 03:24:44 PM
@Windows Defender:
  Ah, yeah, you now said that straightly.
  But I don't know why you think a web server is a threat and unwanted.
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 03:27:12 PM
The 429 code also happens in Takeback (0.14b) with background image function enabled (by editing tpl file). Only in RC1.


i tested it with RC2 and no problem.
But you should support the 429 code, because it may happen on a busy server.
In case of 429 you have to retry.
If you don't, the bg may not show on a busy server. Nobody dies.
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 03:27:57 PM
@Windows Defender:
  Ah, yeah, you now said that straightly.
  But I don't know why you think a web server is a threat and unwanted.

they assume you are clueless and you don't know what's happening, somebody else put the server on your machine :)
Title: Re: version 2.4
Post by: NaitLee on June 08, 2020, 03:36:58 PM
they assume you are clueless and you don't know what's happening, somebody else put the server on your machine :)

The "somebody" is the HFS update/rollback batch :D

The retry-on-429 is done ;)


Another thing:

I see some information about HFS is appended at end of an error page. But it also always includes the script jquery & "lib.js", even when in other templates and it's useless. This extra-loaded js will slow down some templates that needs response fast like Throwback (if it updated).

And the appended "HFS" object, how about add "var" before, to be compatible with "use strict"; ?
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 04:36:29 PM
it's because of the new system that automatically inherits the default tpl.
enter this section as empty in your tpl:
[error-page]

Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 04:40:26 PM
you should consider to actually use the error-page feature to put the common code shared by all your error pages
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 04:53:28 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-rc02
Title: Re: version 2.4
Post by: NaitLee on June 08, 2020, 05:03:50 PM
I tried an empty [error-page] but that did no work.
So I had a look to default template, and got that:
Code: [Select]
[error-page]
%content%
With THIS in tpl makes the error pages work and pure ;)
That %content% places for specific error pages. Feel free to add other things.
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 05:31:51 PM
consider i made a test with your tpl just putting [error-page] at the very end, and it fixed the problem
Title: Re: version 2.4
Post by: MarkV on June 08, 2020, 05:39:41 PM
thanks mark but it's not randomness we need for this (salt excluded)
Oh, so I didn't understand right. You want to encrypt the new password with the old one, send it and have HFS decrypt it, before replaceing and dispose of the old credentials.

That's good, however, one small thing to consider. If HFS follows good security practices, it will NEVER store the password itself. Instead, it will calculate a strong digest (like SHA512) and store that, probably salted. When a client sends the password, calculate the SHA512 digest of the sent password and then compare the two hashes.
Advantages: Passwords cannot be stolen from the server configs as the server doesn't store them; strong hashes are irreversible and have no known collisions.

In that case, you would encrypt the new password with the old password's hash, not with the password itself.


Edit: Default template also looks good in dark:
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 05:47:36 PM
correct, and i knew that, it's in the to-do since long :)
for a later release, we have even too much stuff here. Initially i meant 2.4 just for the mobile tpl ;D
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 05:50:13 PM
Edit: Default template also looks good in dark:

i see you are using some other way to have it dark :) but i trust you know that you get a similar result by clicking the bulb icon on the top-right corner
Title: Re: version 2.4
Post by: MarkV on June 08, 2020, 05:56:53 PM
correct, and i knew that, it's in the to-do since long :)
for a later release, we have even too much stuff here. Initially i meant 2.4 just for the mobile tpl ;D
Sorry, I was long absent. Lost quite some (dev) time.

i see you are using some other way to have it dark :) but i trust you know that you get a similar result by clicking the bulb icon on the top-right corner
It's an extension in my browser. :)

Will pressing the bulb be saved, possibly by account? Is it controllable, possibly even by day/night cycle?
Title: Re: version 2.4
Post by: Mars on June 08, 2020, 06:04:56 PM
@rejetto

could you add this macro in scriptlib.pas during the new release of hfs
    ....
Quote
    if name = 'sha1' then
      result:=strSHA1(p);
    if name = 'sha256' then
      result:=strSHA256(p);


it is to allow an intercation with the function sha256(s) that you placed in the template


thank you in advance
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 06:12:10 PM
Will pressing the bulb be saved, possibly by account? Is it controllable, possibly even by day/night cycle?


it's saved  as a cookie in the browser.
the 'cycle' could be a plugin, i don't think i will include such feature in the base.
i guess the best would be to detect what you want from the OS, as they are starting to be dark-aware.


Thanks to the new *.diff.tpl feature it will be much easier to make "plugins".
I have many ideas but little time.


could you add this macro in scriptlib.pas during the new release of hfs
it is to allow an intercation with the function sha256(s) that you placed in the template


sure, next release
Title: Re: version 2.4
Post by: dj on June 08, 2020, 06:25:57 PM
Will pressing the bulb be saved, possibly by account? Is it controllable, possibly even by day/night cycle?

autotheme addon (http://rejetto.com/forum/index.php?topic=13060.msg1065357#msg1065357) sets dark mode by OS
Title: Re: version 2.4
Post by: MarkV on June 08, 2020, 08:44:33 PM

it's saved  as a cookie in the browser.
the 'cycle' could be a plugin, i don't think i will include such feature in the base.
i guess the best would be to detect what you want from the OS, as they are starting to be dark-aware.


Thanks to the new *.diff.tpl feature it will be much easier to make "plugins".
I have many ideas but little time.

My browser kills cookies after a few minutes or session end, thanks for the note. I would like to have at least the option to make dark the default (theme always starts as dark instead of light).

autotheme addon (http://rejetto.com/forum/index.php?topic=13060.msg1065357#msg1065357) sets dark mode by OS
Will not work for me, I'm on Windows 7. Could be set by day time easily.
Title: Re: version 2.4
Post by: rejetto on June 08, 2020, 09:31:11 PM
My browser kills cookies after a few minutes or session end, thanks for the note. I would like to have at least the option to make dark the default (theme always starts as dark instead of light).


put this file in hfs folder.
i love this feature.


autotheme addon sets dark mode by OS


cool! i forgot about it.
I will use the method for the default tpl. I hope you don't mind :)
Title: Re: version 2.4
Post by: MarkV on June 08, 2020, 09:55:37 PM
Seems to work, although it unconditionally forces dark mode (reset to dark at every refresh or directory change as the theme gets re-read). Well, good enough for me!

Just be sure to skip the OS method on non-conforming OS (everything before Windows 10 1803 IIRC).


Edit: Small objection with the dark theme in general. Every refresh or directory change, it flashes bright before returning to dark again. This is very painful for the eyes, especially in dark environments. Maybe it's easy to improve.
Title: Re: version 2.4
Post by: dj on June 09, 2020, 02:45:37 AM
cool! i forgot about it.
I will use the method for the default tpl. I hope you don't mind :)

That's what it's there for.
(It's also build in in my tpl)
Title: Re: version 2.4
Post by: LeoNeeson on June 09, 2020, 03:50:35 AM
i'll better explain this new anti-DoS mechanism i just introduced:
[...]
If necessary i will introduce an option to configure this behavior, we'll see.
» Suggestion: Why don't you introduce an option to also configure how many connections per second (per IP) are accepted? (for example, with a new variable 'MAX_PER_SECOND'). That, along with the current variable 'MAX_CONCURRENTS (https://github.com/rejetto/hfs2/commit/782702c192e424d1ff439030fdc2d637fcb8d2ac)'. It could be a menu setting, or (even better) only a hidden config on hfs.ini. This could be useful for advanced server admins that doesn't care about having a DoS protection, for example if he only uses HFS on a private or local LAN environment. Alternatively, if any of those settings are set to "0", it could disable this DoS protection (never delivering a 429 error, but giving unlimited/unrestricted connections, like on the old 2.3m version). It's just an idea... :)

» Another important thing: Have you thought on how this new setting could affect the use of HFS along with Stunnel? (I haven't tested since I don't use Stunnel, but this new setting could give trouble if all requests have the same IP, like when it's used with Stunnel). An option to disable all this DoS protection, could come handy for that exactly reason. ;)

Cheers,
Leo.-
Title: Re: version 2.4
Post by: NaitLee on June 09, 2020, 08:49:16 AM
Leo says right. My homepage index.html needs lots of css and js requests, and RC2 also caused some of them 429.
Because they should be loaded from start, retrying cannot solve all problems.
Making this adjustable means respecting users freedom ;)
Or only apply the anti-DoS on a same request by default?
Title: Re: version 2.4
Post by: rejetto on June 09, 2020, 09:39:27 AM
» Suggestion: Why don't you introduce an option to also configure how many connections per second (per IP) are accepted?


as i just said, if we find that's necessary to have an option we'll have an option.


Quote
» Another important thing: Have you thought on how this new setting could affect the use of HFS along with Stunnel?


to be honest, no, that can be a serious problem. I may need to introduce the 'option' just for this.
As i understand, stunnel doesn't work at http level, and won't "forward" the original IP address to HFS. Correct?
An "http proxy" would do that, and you would have no problem.
Sadly it's not easy to find a free one. Surely can be done by configuring nginx or apache as reverse proxy.
https://docs.nginx.com/nginx/admin-guide/web-server/reverse-proxy/ (https://docs.nginx.com/nginx/admin-guide/web-server/reverse-proxy/)
https://www.scaleway.com/en/docs/how-to-configure-nginx-reverse-proxy/


Leo says right. My homepage index.html needs lots of css and js requests, and RC2 also caused some of them 429.


DoS protection doesn't apply to files, css or js.
Tell me how to reproduce the problem please, it may be a bug.
Title: Re: version 2.4
Post by: Mars on June 09, 2020, 10:20:13 AM
any external connection made through stunnel is seen as a local connection in 127.0.0.1 by hfs, it is impossible to go back to the user and the use of the ban therefore is rendered useless

Quote
transparent = yes | no (Unix only)
Transparent proxy mode

Rewrite the addresses so that they appear to come from the SSL client machine
rather than one that runs stunnel. This option is only available in local mode
(exec option) with the LD_PRELOADing env.so shared library shared library and
in remote mode (connect option) on Linux 2.2 kernels compiled with the option
transparent proxy and only in server mode. This option cannot be combined with the mode
proxy (connect) unless the default route from the client to the target goes through
the host running stunnel, which cannot be localhost.

transmission of external ip is only available on UNIX
Title: Re: version 2.4
Post by: illskillz24 on June 09, 2020, 10:42:04 AM
One question, can custom templates be used for hfs24?

I keep getting an error.
Title: Re: version 2.4
Post by: rejetto on June 09, 2020, 11:21:30 AM
One question, can custom templates be used for hfs24?
I keep getting an error.

if the template was made for 2.3 some changes must be done.
template makers can get help on this separate thread http://rejetto.com/forum/index.php?topic=13326.0
Title: Re: version 2.4
Post by: NaitLee on June 09, 2020, 11:25:30 AM
DoS protection doesn't apply to files, css or js.
Tell me how to reproduce the problem please, it may be a bug.

I'm wrong, there's a limit set in my configuration by a template...

One question, can custom templates be used for hfs24?
I keep getting an error.

Updates requires new templates have sth different from old ones:

[unauthorized] to [unauth] as recognition, inheritable selections like [error-page],
new file action methods ('section' to 'file'), and new login system (encryptable password, logout).
Maybe auto-retry is needed, too.
Title: Re: version 2.4
Post by: dj on June 09, 2020, 11:48:11 AM
at the moment Mobil-light (http://rejetto.com/forum/index.php?topic=11754.msg1061351#msg1061351) and Takeback (http://rejetto.com/forum/index.php?topic=13287.msg1065533#msg1065533) works
Title: Re: version 2.4
Post by: rejetto on June 09, 2020, 06:13:11 PM
i've found some time and experimenting with nginx.
I'm already connecting to HFS through https.
Now i just need to find how to enable the IP forwarding.
Title: Re: version 2.4
Post by: rejetto on June 09, 2020, 09:15:57 PM
https://drive.google.com/file/d/1n2Th-Os_CYNhL7p-JCWQCkFEliJd_VqE/view?usp=sharing

unzip and use "run.bat"

this nginx is configured to give you HTTPS without losing the original address.
It is configured to connect to HFS on port 80.
A certificate is provided, of course it's not a valid one, but the user can decide to accept it anyway.

(optional) You can generate your own certificate at https://www.selfsignedcertificate.com/

we should investigate this now
https://certbot.eff.org/lets-encrypt/windows-nginx
Title: Re: version 2.4
Post by: rejetto on June 09, 2020, 10:21:34 PM
i successfully used certbot and just had to edit nginx.conf to point to the new certificate files.
now my home HFS is perfectly working with https and i see real IPs. Happy :)
Title: Re: version 2.4
Post by: MarkV on June 09, 2020, 10:30:05 PM
It seems to work fine. I just changed the ports (my HFS runs on 8080) and used my own self-signed cert (Let's Encrypt only works for Internet-accessible servers, due to the verification process).

I'm using OpenSSL to create my certs, though, so I don't have to rely on a random server in the Internet.

Code: [Select]
openssl.exe req -new -config openssl.cfg -outform PEM -newkey rsa:2048 -x509 -sha384 -days 730 -nodes -verbose -out "cert.crt" -keyout "cert.key"
-new - create new cert
-config openssl.cfg is the name of the OpenSSL config file (you can have many)
-outform PEM - output format PEM
-newkey rsa:2048 - RSA key with 2048 bits strength
-x509 - don't do a certificate request (for CA)
-sha384 - hash used for cert
-days 730 - validity period
 -out "cert.crt" -keyout "cert.key" - names of the created cert and key.

Get OpenSSL for Win here: https://slproweb.com/products/Win32OpenSSL.html
Title: Re: version 2.4
Post by: LeoNeeson on June 10, 2020, 06:15:24 AM
as i just said, if we find that's necessary to have an option we'll have an option.
OK, I already knew you will not like my suggestion... :-[ ::) But hey, you did a great job with Nginx! :) (you can be happy about it). Rejetto: from now on, you can breathe relieved when you see a message from me, since I will not be reporting new more things about v2.4.x, because I don't want to be the guy who always 'break the balls' with small details (non voglio essere il che 'rompe le palle' sempre con piccoli dettagli :D). I officially leave the "reporting things" task for the rest of you (beside that, my free time is much more limited now).

@MarkV: thanks for the command line explanation of generating a cert with OpenSSL. It's always nice not having to rely on some external service. I will try it as soon I have some free time. (It's nice to have you on-board :))

Cheers,
Leo.-
Title: Re: version 2.4
Post by: rejetto on June 10, 2020, 08:35:38 AM
OK, I already knew you will not like my suggestion... :-[ ::)

:) actually the rule i learnt in time is that you should not put an option (on the screen) until it is necessary.
This IF you can make an automatic behavior that's always good. And sometimes even beyond that.
It's a general rule that I want to apply.

Quote
since I will not be reporting new more things about v2.4.x, because I don't want to be the guy who always 'break the balls' with small details

i don't understand why you say that, but you are very wrong. Reporting problems is very important.
Don't do it if you don't want, but don't see it as breaking balls.
Title: Re: version 2.4
Post by: MarkV on June 10, 2020, 02:56:57 PM
When I find some time, I'll also document my changes in openssl.cfg, which will greatly ease the creation process, and will allow "Subject Alternate Names (SANs)".
Title: Re: version 2.4
Post by: Mars on June 10, 2020, 04:48:54 PM
Me not happy, rejetto forget sugar in coffee  :'(

Missing  macro 'sha256' in rc2

Waiting quickly rc3
Title: Re: version 2.4
Post by: rejetto on June 10, 2020, 05:31:07 PM
Missing  macro 'sha256' in rc2

what are you going to do with it?
Title: Re: version 2.4
Post by: Mars on June 10, 2020, 07:27:39 PM

could you add this macro in scriptlib.pas during the new release of hfs
it is to allow an intercation with the function sha256(s) that you placed in the template

sure, next release

because,
first: you promised
secondly: failing  to be able to use rsa client-server encryption, this is useful for validating an exchange of encrypted passwords

I thought of inserting the data to be transmitted in a zip protected by a sha256 based on the connection pass, but it requires too many resources to add in javascript and in hfs
Title: Re: version 2.4
Post by: rejetto on June 10, 2020, 09:29:41 PM
first: you promised

calm down mars, i asked what, not why :D
if you look on github you'll see the command is ready. I'll publish asap.
Title: Re: version 2.4
Post by: Mars on June 10, 2020, 09:41:04 PM
 ;D ;D

here is what to test its use in the password change by "form"

old password
new password
confirm password

Passing over the eyes with the mouse the passwords are displayed in clear


positioning the mouse at eye level the passwords are displayed in clear

(https://i27.servimg.com/u/f27/16/40/25/39/pass10.png)

the old password is not sent in clear but its sha256 is combined with the new and sha256 of the confirmation to check the validity at the server level by macros

everything takes place in the template and only requires the existence of the macro sha256

it's not the best we can do, but it's better than a clear password, and it allows accidental password change if we don't know the old one, the rest will come later when RSA becomes available jointly using the certificate for HTTPS
Title: Re: version 2.4
Post by: rejetto on June 10, 2020, 10:24:16 PM
it's nice and it's worth making a plugin out of it.
for the default tpl i'll wait to have encryption.
i understand that the https certificate would be ok to be used, if available, but i guess that it would be nice to have a solution that works even if you are not using https and probably have no cert.
Title: Re: version 2.4
Post by: Mars on June 10, 2020, 11:38:38 PM
in the version of silentpliz using stunnel, he introduced a tab on the main page to manage the creation of such a certificate,

one solution would be to integrate it as a tab in the options or as an additional module, we could easily create his own certificate, and use it either to configure stunnel.conf or nginx.conf without using an external program.

 for the transmission of change password you should consider a solution like my encryption in the absence of a certificate, and abandon sending it by your old method

I think that with the help of silentpliz we could manage to offer you a module that would be easily integrated into hfs
Title: Re: version 2.4
Post by: LeoNeeson on June 11, 2020, 03:58:33 AM
I think that with the help of silentpliz we could manage to offer you a module that would be easily integrated into hfs
I vote to have this by default (even if we don't have HTTPS yet). It's better to have some security, than no security at all. I like the work Mars has done. :)
Title: Re: version 2.4
Post by: dj on June 11, 2020, 04:42:33 AM
you added (http://rejetto.com/forum/index.php?topic=11737.0) the mime type for mkv (works with chrome)

why not add

 .mp3 audio/mp3
 .mp4 video/mp4
(.m3u8 application/x-mpegURL)
Title: Re: version 2.4
Post by: NaitLee on June 11, 2020, 05:40:29 AM
Maybe the "sha256" macro can be used to encrypt passwords easily.
The sha256 result can be sent to client, then just slice some of its result with some rule and offset password bytes in some way?
To encrypt new password with old pass, we need a method to get it. As well as the oldpass-newpass-newagain form, may need this.
Title: Re: version 2.4
Post by: rejetto on June 11, 2020, 09:18:26 AM
in the version of silentpliz using stunnel, he introduced a tab on the main page to manage the creation of such a certificate,

hfs will handle the certificate when it support https.
making it just for the password-change it is too much, and there would even be a design problem: the change is a tpl-only feature at the moment. If you move even one bit to the server, then you move also the [ajax...] part, for consistency, for good design.
In the meantime it can be done as a plugin/tpl. You have {.exec.} and you can call openssl.
I was considering making a plugin for https, with nginx, openssl, and possibly letsencrypt. Sadly I have little time now.

The sha256 result can be sent to client, then just slice some of its result with some rule and offset password bytes in some way?

the password is to be communicated to the server, not to the client. And there's no way to solve this problem only with hashing, that i know.

you added (http://rejetto.com/forum/index.php?topic=11737.0) the mime type for mkv (works with chrome)
why not add

agreed
Title: Re: version 2.4
Post by: rejetto on June 11, 2020, 10:03:31 PM
https://github.com/rejetto/hfs2/releases/download/v2.4-rc03/hfs.exe
Title: Re: version 2.4
Post by: NaitLee on June 12, 2020, 03:49:55 PM
Should this thing (http://rejetto.com/forum/index.php?topic=13060.msg1065897#msg1065897) be fixed? It's still there.
Additionally, in Wine/Linux HFS only shows the IP 127.0.0.1.

Because of the quotes added in a fix, now the filename on download have no problem with Chrome.
But in Firefox appears to have the quote attached before and after.
Snapshot shows the Firefox version 79.0a1 (Nightly), a quote after extension made the judge of filetype not matched, with double extensions.
And in Firefox ver 69.0.2 (on GNU/Linux), the filename always comes with _underlines_ before and after.
Title: Re: version 2.4
Post by: rejetto on June 12, 2020, 10:35:37 PM
Should this thing (http://rejetto.com/forum/index.php?topic=13060.msg1065897#msg1065897) be fixed? It's still there.

well, why not. I don't use linux and I'm not going to do it. You? :) Anyone can contribute.

Because of the quotes added in a fix, now the filename on download have no problem with Chrome.
But in Firefox appears to have the quote attached before and after.

aaargh. Ok, i've changed solution. No more double quotes, i will instead encode commas (and semicolons).
I just tried with chrome and firefox a file with a space, and comma and semicolon. All well with the next release.
Thanks for reporting.
Title: Re: version 2.4
Post by: danny on June 12, 2020, 11:04:34 PM
During testing, I was really happy to see proactive (self-repair) error messages such as 'busy' which lasted only a second and then it had repaired itself. 
Awesome!

Here are some more examples of proactive/automated error handler.
Code: [Select]
[overload]
{.if not|%user%|{:{.if|{.%url% = /.}|{:{.disconnect.}:}.}:}.}{.add header|Cache-Control: no-cache, max-age=0.}<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"><html><head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<META HTTP-EQUIV="Refresh" CONTENT="3;URL=./">
<TITLE>Overload</TITLE>
<body bgcolor="black" text="white" alink="white" link="white" vlink="white">
<center><h2><br>High traffic mode engaged.</h2>Returning to previous page after overload has cleared.</center></body></html>

[max contemp downloads]
{.add header|Cache-Control: no-cache, max-age=0.}<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"><html><head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<META HTTP-EQUIV="Refresh" CONTENT="3;URL=./">
<TITLE>Downloads</TITLE>
<body bgcolor="black" text="white" alink="white" link="white" vlink="white">
<center><h2><br>There are ongoing downloads.</h2>More available <i>after</i> current downloads finish.</center></body></html>

[not found]
{.if|{.match|*.php*;*.js;*.py;*.vbs*;*.exe|%url%.}|{:{.disconnect.}:}.}{.add header|Cache-Control: no-cache, max-age=0.}<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"><html><head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<META HTTP-EQUIV="Refresh" CONTENT="1;URL=../">
<TITLE>404</TITLE>
<body bgcolor="black" text="white" alink="white" link="white" vlink="white">
<center><h2><br>You have found the 404 page.</h2>Redirecting.</center></body></html>

[deny]
{.if not|%user%|{:{.disconnect.}:}.}{.if|{.match|*.php*;*.js;*.py;*.vbs*;*.exe|%url%.}|{:{.disconnect.}:}.}{.add header|Cache-Control: no-cache, max-age=0.}<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"><html><head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<META HTTP-EQUIV="Refresh" CONTENT="1;URL=../">
<TITLE>Denied</TITLE>
<body bgcolor="black" text="white" alink="white" link="white" vlink="white"><center><h1><br><br>No</h1></center></body></html>

[ban]
{.disconnect.}
Title: Re: version 2.4
Post by: dj on June 13, 2020, 03:35:42 AM
After trying to change password with default tpl and rc3 (rc2 works fine)
Does this only happen to me?



in my template (http://rejetto.com/forum/index.php?topic=11754.msg1066068#msg1066068) change password works and will be sent base64 encoded (https://github.com/dj0001/hfs-template/blob/gh-pages/mobil-light_V5.5.tpl#L179) to the server
Title: Re: version 2.4
Post by: NaitLee on June 13, 2020, 05:10:09 AM
Does this only happen to me?

No, me too.

Seems it's jQuery failed to handle something.
A lot of lines of script was put into a VM with many <br>.
Title: Re: version 2.4
Post by: rejetto on June 13, 2020, 03:03:04 PM
sorry guys, i've broken the ajax() function with address ?~ajax instead of /~ajax  :(
Title: Re: version 2.4
Post by: rejetto on June 13, 2020, 07:32:30 PM
sorry guys, i've broken the ajax() function with address ?~ajax instead of /~ajax  :(

actually, only ?mode=section&id=ajax is good, because the folder path is important. I had forgot. Too many years have passed. Added a comment in the source to remember. ;D

Anyway, next release will revert the 'files' to 'selection' and accept the 'unauthorized' section, to help tpl to be compatible with 2.3.
To tell if the tpl is compatible with 2.4 i will introduce a dedicated method, that is, a mandatory section [api level] with a number inside. For now the number is 2.
Code: [Select]
[api level]
2
This is a method that we can use also in the future if we have breaking changes.

WOuld someone of you template makers please give a look at this preview and tell me that the news are working well for you? Thanks
https://drive.google.com/file/d/1cpNFwGlYcBZIEFObFo1_WFsyKlquU_6E/view?usp=sharing
Title: Re: version 2.4
Post by: dj on June 14, 2020, 04:26:42 AM
quickly tested (https://github.com/dj0001/hfs-template/blob/gh-pages/mobil-light_V5.5.tpl), seems ok
Title: Re: version 2.4
Post by: danny on June 14, 2020, 04:37:45 AM
had to update, if bad password, refresh page
reverted change to delete script for selection vs files.

Cool new bonus found!  HFS2.4 is now compatible with UTF-8 template so you can put icons as a regular text character in the template.  Very efficient!!  So, please default {.set ini|use-system-icons=no.} for HFS2.4

request:  please don't use "api level" for censor because, for the future, it could be bad if .exe automatic update suddenly changes server-owner's template at random times.  That conflicts .exe updates, because api-level moving target is not feasible for production server.  Your previous idea of banning the [unauthorized] section was much better. 

I chose to continue omitting the [unauthorized] section because it is never reached due to diy /~login page.  Therefore, it seems fine to let the 404 handler filter and redirect the missing page.  There was in fact, no use for unreached [unauthorized] section.

Bug:  {.if|!%user%|{:howdy stranger:}.} worked in HFS2.3, fails in HFS2.4.  Need to double-check the ! reverse logic shortcut.
For Throwback, I replaced previously working !%user% with currently working not|%user%  And, that is okay.

auto-ban not achievable, so Security Feature Request:  Need a menu/limits  option to limit bad login flooding/attack/volume to 5 per hour (or "x number per hour).   
Title: Re: version 2.4
Post by: rejetto on June 14, 2020, 07:52:57 AM
Cool new bonus found!  HFS2.4 is now compatible with UTF-8 template files (and maybe more?) so you can put icons as a regular text character in the template.  Very efficient!!  Please default {.set ini|use-system-icons=no.} for HFS2.4

I will surely change that in the future but it's too early.
def settings should match the def tpl, and it still using the system icons.

Quote
request:  please don't use "api level" for censor because, for the future, it could be dreadful if .exe automatic update suddenly changes server-owner's template at random unannounced times.  That is in conflict with .exe updates.  The api-level moving target is not feasible for production server.  Your previous idea of banning the [unauthorized] section was much better. 

You are right about automatic updates, it can be a disrupting experience. I don't really have a better solution a this moment.
Maybe it's better if tpl change is not force, and let the server unable to login people until the admin acts? What's worst? IDK.
Anway, banning [unauthorized] is exactly the same thing. You update and you get the error and are switched to default tpl. I don't understand why you think there is a difference. Probably you are not thinking of old tpl, but of tpl that were adapted for 2.4

Quote
I chose to continue omitting the [unauthorized] section because it is never reached due to diy /~login page. 

indeed, you are talking about 2.4-aware tpl.
Of course they will work.
But you will have to comply. In 2.3 the [unauthorized] section is sent when the credentials fail, i think. The documentation is bad about it, tells nothing.
I tried to have a "smart" method, but got sick of it. Sometimes explicit declaration is better than euristhic methods.

Quote
Bug:  {.if|!%user%|{:howdy stranger:}.} worked in HFS2.3, fails in HFS2.4.  Need to double-check the ! (reverse logic shortcut), because it faulted a bit.

i just tried. It never worked. It is just not supported and I don't think someone said it is. It's not in the documentation either.
I also warned you about this error in your tpl in another forum thread.
Make a better test in 2.3, that one of yours is probably fooling you {.if|!1|wrong|good.}

Quote
auto-ban not achievable, so Security Feature Request

this request is in to-do since forever and will likely be addressed soon. Anyway i don't see why you say "not achievable".
Title: Re: version 2.4
Post by: rejetto on June 14, 2020, 08:51:52 AM
https://github.com/rejetto/hfs2/releases/download/v2.4-rc04/hfs.exe

there is a slight difference with the exe from few posts ago, now the def tpl has the [api level]. It is necessary for people who decide to edit the tpl.
Title: Re: version 2.4
Post by: danny on June 14, 2020, 06:11:39 PM
...
...auto-ban not achievable, so Security Feature Request:  Need a menu/limits  option to limit bad login flooding/attack/volume to 5 per hour (or "x number per hour).
this request is in to-do since forever and will likely be addressed soon. Anyway i don't see why you say "not achievable".
New for 2.4: auto-ban not achievable in a template, because any page anywhere can post a form. 
It doesn't have to be a template--you can make a different page or script to post a form.
So, the best place to prevent/interrupt a form input, is the .exe
Title: Re: version 2.4
Post by: rejetto on June 14, 2020, 07:49:20 PM
yes, and that's why i told you that you need to use the [unauthorized] event, that's in "events", not in the tpl.
If i'll find the time i will do it for you.
Title: Re: version 2.4
Post by: danny on June 15, 2020, 12:06:30 PM
yes, and that's why i told you that you need to use the [unauthorized] event, that's in "events", not in the tpl.
If i'll find the time i will do it for you.
I found some posts about it:  https://rejetto.com/forum/index.php?topic=6757.msg1043503;topicseen#msg1043503
But, I couldn't get [+unauthorized] to work in the events file. 
Title: Re: version 2.4
Post by: rejetto on June 15, 2020, 12:47:48 PM
sorry, i've just found that the new login system is not using the 'unauthorized'.
I'll see to have this fixed for next release.
Title: Re: version 2.4
Post by: danny on June 15, 2020, 10:19:44 PM
Without functional [unauthorized], users/clients can be lost on a mostly white page that says bad password.  Sometimes (not often).

Another form-related hijink is:
menu > other options > user accounts
After login redirect to: 
doesn't work like 2.3

In both cases, redirection to usable content, is missing (doesn't deliver).

The white page problem is rare; but, there is errata, sometimes.
Title: Re: version 2.4
Post by: rejetto on June 16, 2020, 07:50:56 AM
In both cases, redirection to usable content, is missing (doesn't deliver).

i don't understand what you mean by usable content. Please tell me what you do, what's the expected behavior, and what happens instead.
I'm working to make a script to limit login failures in time. It's good because I find other bugs and fix them.
Title: Re: version 2.4
Post by: danny on June 16, 2020, 04:44:07 PM
i don't understand what you mean by usable content.
A list of files to download. 
Please tell me what you do, what's the expected behavior, and what happens instead.  I'm working to make a script to limit login failures in time. It's good because I find other bugs and fix them.
I log in, and it usually works.  What was expected was one of these:  [unauth], or [unauthorized], or successful login, or 'after login redirect' as specified in the Users menu.  There is a little more variety than just those.

Sometimes there's a white screen message or didn't end up at the right folder.  I thought to fix the landing spot by using HFS menu > Users > 'after login redirect' But that doesn't do it. 
Title: Re: version 2.4
Post by: rejetto on June 16, 2020, 05:26:45 PM
A list of files to download. 

...a folder? you are saying that you specify the path to a folder in the 'after login redirect' account field and you are not taken there?
you give too few details. You don't specify the path you entered, you talk about a "white screen message" but don't report what it says.
I'm tested this right now. I started hfs, clean, no options, no tpl. Add a folder "upload". Create an account. Set the redirect for it to "/upload". I access the home with the browser, log in, and i'm taken to the upload folder.
I try again and set a path to a file that's inside the folder, a txt. Logout, login again, i'm take to the file and see the file content.
Let me know if you have clear directions on how i should change what i just did to run into the same problem as you, otherwise it's hard for me to help.
Title: Re: version 2.4
Post by: rejetto on June 16, 2020, 05:29:44 PM
if you get a different behavior you want to show me sharing your screen on google hangout if you want
Title: Re: version 2.4
Post by: danny on June 16, 2020, 09:17:44 PM
if you get a different behavior you want to show me sharing your screen on google hangout if you want
I tested again.  The first time, I got a white screen that said ok.  But, the next time I tested, it went to /upload perfectly fine. 
And, I tested several more times.  It worked fine all of those times as well. 
Title: Re: version 2.4
Post by: rejetto on June 16, 2020, 09:24:05 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-rc05

guys, with this release we switch to the 'public' flag in place of the 'private'.
Please update your tpl.
Title: Re: version 2.4
Post by: danny on June 17, 2020, 05:30:07 AM
There was some missing parts until I cleared the browser's cache.  After that it worked nicely. 

P.S. Post#1 link (this same thread) links to RC4.
Title: Re: version 2.4
Post by: danny on June 19, 2020, 05:11:31 AM
I'm working to make a script to limit login failures in time.

However, on the phone browser, there could be a white screen with ok (too small to read--it just looks blank) or or white screen with bad password or white screen that said something about username (too small to read but I think it meant the username was typed wrong). 

The 'lost on a white screen' problem is a little more frequent for the phone than the PC.
The user may be confused or believe the server is broken/offline.

Feature request:
I'd like a 1 second (maybe 1.5s) redirect on those 3 white screens, such as:
if ok, wait 1s then redirect to ../ (if accidentally at server response screen)
if bad password, wait 1s then redirect to /~login (if password typo)
if bad username, wait 1s then redirect to /~unauth (if can't type own name)

The suggestion is a 'fallback' just to let the users escape from the white screens.  I think that problem is raw server-response screens (like white underwear), not developed web pages.  But actually redirecting to any page with links is fine--Just really anything other than the mostly blank white screens. 

P.S.  I did not test with the inbuilt template.  This is a basic usability difference between 2.3 vs 2.4:  While 2.3 would always reach a web page with clickable links or [section], 2.4 sometimes doesn't.  That needs some server-side validation redirect-to-section added.

P.P.S.  For me, this is the only impasse between rc versus production release--it is why I can't use 2.4 on my own home server.
Title: Re: version 2.4
Post by: rejetto on June 19, 2020, 07:38:39 PM
i don't know how to reproduce your problem, i get no white screen. I tested with default tpl.
This is the wrong place to discuss this, as there is a dedicated topic for that script.
Title: Re: version 2.4
Post by: MarkV on June 20, 2020, 08:16:08 AM
I did experiment a bit with nginx, as there were a few quirks (such as DualStack not working).

Found out the fix is actually simple: Change the line

listen 443 ssl;

to

listen [::]:443 ssl;


Done, works for me.

_____

One question: The reverse proxy usage of course cloaks the true identities of the connecting machines. Can HFS be configured to show the IP from the X-Forwarded-For (or X-Real-IP) headers, if they exist?
Title: Re: version 2.4
Post by: rejetto on June 20, 2020, 05:17:45 PM
that feature is enabled by default, you should see the original ip if the reverse proxy is on the same machine (localhost)
Title: Re: version 2.4
Post by: danny on June 20, 2020, 06:01:00 PM
i don't know how to reproduce your problem, i get no white screen...
The various javascript workarounds to catch server response, can't take into account all browser differences. 
When the phone gets stuck on a white screen, one can long-touch the url, edit the url, backspace over ?mode-login, touch the ok/go button to try again. 
That is a rather lengthy ordeal. 
Since the response is too small to read, it is likely that the user assumes the server is broken/offline. 
I would rather see a web page instead of these:
Title: Re: version 2.4
Post by: MarkV on June 20, 2020, 10:37:21 PM
that feature is enabled by default, you should see the original ip if the reverse proxy is on the same machine (localhost)
That works for IPv4, but not at all for IPv6. The headers contain the correct IPv6 address, though. Maybe HFS needs to be adjusted a bit?

Btw., nginx, by default, uses HTTP/1.0 for the reverse proxy. Add the following line under [server]:

proxy_http_version 1.1;
Title: Re: version 2.4
Post by: rejetto on June 21, 2020, 08:33:53 AM
When the phone gets stuck on a white screen, one can long-touch the url, edit the url, backspace over ?mode-login, touch the ok/go button to try again. 

those are not pages, those are responses intended for javascript (ajax), and that's how 2.4 works.
It's not as simple as it was 2.3 but this was already common 10 years ago.
If you are showing those pages then your tpl is doing something wrong that should be fixed. You can study what the default tpl does, or ask for help in this topic https://rejetto.com/forum/index.php?topic=13326.0
Title: Re: version 2.4
Post by: danny on June 22, 2020, 07:58:27 AM
those are not pages, those are responses intended for javascript (ajax), and that's how 2.4 works. It's not as simple as it was 2.3 but this was already common 10 years ago. If you are showing those pages then your tpl is doing something wrong that should be fixed. You can study what the default tpl does, or ask for help in this topic https://rejetto.com/forum/index.php?topic=13326.0
Is it possible to get a "fallback" action of wait 1.5s then redirect ../ so that the white screen doesn't stay overlong? 
Because, not all browsers do the same thing (a little worse on mobile).
Title: Re: version 2.4
Post by: NaitLee on June 22, 2020, 10:59:45 AM
Is it possible to get a "fallback" action of wait 1.5s then redirect ../ so that the white screen doesn't stay overlong?

After the HFS macros are executed, their results are mostly sent to client as HTML.
For redirect you can simply add a refresh meta to changepwd ajax. But this is not a recommended fix to your problem.
See this  (http://rejetto.com/forum/index.php?topic=13326.msg1066156#msg1066156)would help.
Title: Re: version 2.4
Post by: rejetto on June 22, 2020, 10:12:41 PM
Is it possible to get a "fallback" action of wait 1.5s then redirect ../ so that the white screen doesn't stay overlong? 

user should never see those messages, or the white screen, even for 0.1s, if things are handled properly.
Anything the default tpl does, other tpls can do to. One can just go by modifying the default tpl or writing by scratch if he is skilled enough.
We should definitely have an easy way to integrate the standard login form in other tpls.
Title: Re: version 2.4
Post by: danny on June 23, 2020, 12:40:10 PM
We should definitely have an easy way to integrate the standard login form in other tpls.
That would be fantastic! 
For sure, updates would work much better (because the login would update to match). 
Title: Re: version 2.4
Post by: danny on June 24, 2020, 10:32:41 AM
In preparation for 2.4 mainline release, I've been doing some testing. 

--------------------------------------------------------------------------------
Testing with the default template, I did nothing other than log in and then click Archive.
In half an hour, the server crashed out.  The archive function was not confined to doable.
At the top of the screen, there is one click for crash.
Patch is:  Previously published in https://rejetto.com/forum/index.php?topic=13060.msg1065272#msg1065272
Result:  high impact / please revise before 2.4 mainline

-------------------------------------------------------------------------------
Forced template change is still in effect.  Cost is that some server owners may have to ban/disable updates to prevent automatic/unattended template change (there are some specialty use applications that could break).  It is possible to sporadically alter the server while the owner is absent from the console.
At the minimum, we need to make sure that change can't happen until After the pop-up message is closed/acknowledged/consent.
Result:  medium impact / please revise before 2.4 mainline

==================================================
I didn't test *everything; but, it looks like 2.4 is ready for mainstream, except you need to, clamp the archive to a doable scope, and postpone automatic change until after a message box is manually closed/acknowledged. . . because the server-owner is not always present at the console (a difference between testing versus real use). 

Title: Re: version 2.4
Post by: danny on June 26, 2020, 11:47:12 AM
Just in case it was needed, I've made a favicon alternative, by manually compressing it to 178 bytes and then converting to base64.
Code: [Select]
<link rel="shortcut icon" href="data:image/gif;base64,R0lGODlhEAAQALMAABYAABMmTDZXa2ZkVRxeqzl7zkZvmkJ71m6DLWKPsah+Ko5zV5WUZ8alc/zsKtPU1CH5BAEAAAEALAAAAAAQABAAAARfMMhJq00vv2GlYVrWDBxlGOFzjAhiGkd2EOSALNNJEEVvkK0FLrArHhKGVrBU3B2VAiGzSUgIrjVBAMDter9cRxjgKIsV5jF53RV7zWKzAoB2zNlxMH7sDpfvehJgEQAAOw==">However, for [login], [signin], [overload], [unauth] we could 'turn off' the browsers request for favicon.ico file. 
Because, it is best practice to avoid external dependencies on those pages.
Code: [Select]
<link rel="icon" href="data:,"> or {.if not|%user%|{:<link rel="icon" href="data:,">:}.}
If login is required for / then there is no access to file before login. In that case, cached contents of favicon.ico file has text login, inside, not an image; that is cached, and then 'sometimes' favicon file can't show after login. 

A related error can happen for a direct-download, such as message/response/login is the insides of your file instead of the expected content. 

At login, my browser was actually fetching 2 html pages, one displayed on-screen, and an identical copy inside of the favicon.ico file.  I guess, if I had only turned off favicon request for the login page, then it would display normally after logging in. 
Edit:  Tested, and that guess was true--if any of the data links (above) are used for the login page, then the hfs ico file works normally after login.

I had a similar, except hilarious glitch on the [overload] page when it tried to fetch the favicon file by additional request, which counts against the metric.  oops!  :)

The base64 link looks like this:
Title: Re: version 2.4
Post by: rejetto on June 27, 2020, 09:39:12 AM
It is possible to sporadically alter the server while the owner is absent from the console.
At the minimum, we need to make sure that change can't happen until After the pop-up message is closed/acknowledged/consent.

it is already so, the "update automatically" option is off by default.
Unless you are talking about the message that version 2.4 displays after it is installed.
Title: Re: version 2.4
Post by: rejetto on June 27, 2020, 10:00:24 AM
danny's post moved to http://rejetto.com/forum/index.php?topic=13338.0
Title: Re: version 2.4
Post by: rejetto on June 27, 2020, 10:26:28 AM
Testing with the default template, I did nothing other than log in and then click Archive.
In half an hour, the server crashed out.  The archive function was not confined to doable.
At the top of the screen, there is one click for crash.
Patch is:  Previously published in https://rejetto.com/forum/index.php?topic=13060.msg1065272#msg1065272
Result:  high impact / please revise before 2.4 mainline

i surely have to take care of the 'archive' feature.
I studied your proposal, and it amounts to this
      {.if|{.%connections% < 50.}|{:{.if|{.%total-kbytes% <= 4000000 .}|{:{.if|{.get|can archive.}|
      <button id='archiveBtn' class='pure-button' onclick='ask("{.!Download these files as a single archive?.}", function() { submit({ selection: getSelectedItemsName() }, "{.get|url|mode=archive.}") })'>

Please guys, it can be lengthy to dig into a huge file just to search for the changes you are vaguely referring to. Report them directly instead.
You are making the sub-folders not working, for a start. That doesn't seem a great solution and it's not what people expect. You want the archive especially in cases where you want to download whole trees.
Also, the kbytes should not affect in anyway. The operation gets heavier with the number of files instead.
I think the archives should be subject to the anti-dos mechanism.
Title: Re: version 2.4
Post by: danny on June 27, 2020, 11:01:32 AM
you are not satisfied with just reading %speed-out% ?...
I haven't featured out how to read peak/max/top speed data for use in the template.
But it exists. 
HFS has the top-speed data already, in the top-right corner of the screen (in the graph).

...Also, the kbytes should not affect in anyway. The operation gets heavier with the number of files instead.
Both, actually.  There is a functional size limit.  What I listed, isn't the maximum, but it is very close. 
I guess 4194304 is closest to breakage, Or a smaller figure works (if connection rate feasible). 

As far as validating the size that might be useful on a given connection speed, you'd have to know the top speed to do that; and, then you can calculate how many hours or days required to download the archive.

...it can be lengthy to dig into a huge file...
So true--it is gigantic. 

...You are making the sub-folders not working...
I couldn't figure out how to validate the archive function without doing a recursive search to predict the outcome of a recursive archive. 
But, it was easier to predict the outcome of a non-recursive archive. 

If recursive is wanted, I think it is possible to do a search for validation purposes, After the archive button is clicked--maybe the archive function would lead to its own page and section?   If it was in its own section, then maybe it would be efficient to do .zip instead of .tar?   

P.S.  Worth another mention:  For a good part of the validation, if the user had a timeframe estimate, I think they would click 'cancel' most of the time.
Title: Ideas to enhance the anti-DoS mechanism
Post by: LeoNeeson on June 28, 2020, 07:05:27 AM
the anti-dos mechanism.
@Rejetto: After doing some research on how a typical 'Denial-of-Service (https://en.wikipedia.org/wiki/Denial-of-service_attack)' (DoS) is done, which basically consists on overloading a server, I want to contribute with my overall opinion about the Anti-DoS feature.

IMHO, the current implementation is an overkill (I mean, is nice you have implemented some anti-DoS, but for me it's way too-much over-protective). My ideas are:

A) Have different limits for upload, than for download.
B) Give downloads a more relaxed limit than the current one.
C) Count how many requests were done every 5 sec (read below)
D) Limit uploads to only one per 1/second or one per 5/secs.
E) Limit the repetition of downloading the same file, 1 file per 10/secs.
F) Limit/slowdown the request of serving pages (not the internal elements)
G) Have a 'maintenance mode' for extreme cases, limiting everyone but the admin

Some points are self-explained.

About point "C" (counting how many request were done every 5 sec, for the same SessionID), if you give the server admin an option, the admin could configure his server according how many pages are typically needed to be requested (when doing normal usage). For example, seeing Danny's example, if he has a photo gallery, and his page needs to serve 20 thumbnails at once (20 is an invented number for this example). Then, it would be normal to have 20 requests in a 5 seconds time-frame.

About point "F", if a page needs to serve only 10 elements (1 html + 1 css + 2 js + 6 images), then it's normal to have 10 requests in 1 or 2 seconds. So, an even smarter (automatic way), would be HFS automatically counting how many elements the page have (when parsing the template), and applying on-the-fly the limits (if the requests exceed the elements found for the requested page). This would be a behavior limit, since HFS knows how many elements needs the page, and it would let HFS distinguish a legitimate user from an attacker. This, along with limiting how many pages could be served per second, could let us have a more relaxed download rate for elements (like images, css, js), but a more strict limit to request new pages, for example, when exploring folders, we could only serve 1 every 2 seconds, avoiding (or delaying) when the user opens several tabs at the same time.

About point "G", finally, if HFS detects is being attacked (in an very extreme/hard way), then the server could automatically go in 'maintenance mode' for 1 hour (applying a more strict request rate for everyone during that hour), and in that time-frame, it will only allows to login the admin (so he can take care of his server and review the configuration).

This (https://www.feistyduck.com/library/apache-security/online/apachesc-CHP-5.html) is a good read I recommend about DoS. Also, please read THIS (https://blog.radware.com/security/2013/06/distinguish-between-legitimate-users-and-attackers-the-secret-sauce-of-ddos-protection/) (it explains why 'Rate limitation is not the way to go').

Well, I leave you these ideas.
Do what you think is best... :)
Cheers,
Leo.-
Title: Re: version 2.4
Post by: rejetto on June 28, 2020, 09:07:49 AM
Leo, why you are suggesting different limits for uploads and downloads when there is no limit at all for uploads and downloads? :)
guys, this mechanism only affects file listing.
If you want to serve 20 thumbnails at once, or 200, it doesn't matter, there is no limit.
HFS doesn't try to understand if you are attacked, just limits a single address on a very specific operation, because it's heavy.
Title: Re: version 2.4
Post by: danny on June 28, 2020, 02:08:01 PM
Leo, why you are suggesting different limits for uploads and downloads when there is no limit at all for uploads and downloads? :)  guys, this mechanism only affects file listing.
I think it works well.  The busy error message is proactive/automated.  I like that! 
The self-fix error message and list limiter are of especially high quality.

You know how the throwback 404 does: display 404, wait a sec, redirect ../ (loops Up the tree until valid page is found)
Yes, this is the sort of thing that I like. 

And, the thumbnails thing isn't all about limits; but, just exactly the opposite (coding the page so that limits aren't required).
For example, Throwback14 for HFS2.3 required connection limits range 5 to 9, to prevent a flood of requests. 
However, Throwback 14 for HFS2.4 has timers to reduce requests-per-second, so that limiter is not required. 
The limiting factor is not particular to HFS but rather exactly the same for every single-thread server, including Nginx for windows.

P.S.  Leo realized a good enhancement to the limiter menu = requests per second. 
Title: Re: version 2.4
Post by: rejetto on June 28, 2020, 07:20:24 PM
i think we should have a timeout on file listing operation, like: the list is built but if it's not finished after a minute, it stops and gives you what has been found.
Title: Re: version 2.4
Post by: rejetto on June 28, 2020, 10:24:02 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-rc06

- some security fixes
- limit file listing operations to 1 minute
Title: Re: version 2.4
Post by: danny on June 29, 2020, 01:00:16 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-rc06

- some security fixes
- limit file listing operations to 1 minute
Can you make an exception for recursive search? 

Sometimes I would do a search through a library, and the result was still wanted, even if it takes more than a minute. 

I checked and search broke in a minute, with only a page-full of results.  Ouch.
Also checked with rc5 and my search needs almost 7 minutes, because I have a quarter-million files.   Please repair RC6

P.S. Time limit didn't prevent archive hangs; so, Feature request: archive limit to 4gB.
Note:  List limits are unnecessary for Takeback and Throwback because those don't block reflow.
Also consider:  Remove minute limit and replace with kB limit because connection speeds/time vary. 
Title: Re: version 2.4
Post by: rejetto on June 30, 2020, 09:03:59 AM
i will remove the limit on the search and consider it just as an option.
the amount of results can be also zero if you search billions of files, the point was for how long you will keep the server busy.
of course if it's you with your computer you don't want limits.
sadly if you launch a very long request but decide to stop it, Just closing the tab won't do, because the server receives no communication about it. (tested with Chrome)
This can keep you unable to do anything else.

I'm not sure what you mean by "archive hangs", and I don't know how to reproduce your problem with archives' size: i just tested getting a 11GB archive (32 files), and the server was still working after it. If it's not saturating its single-core CPU it will also still respond and let me browse the rest of the files while downloading.

I just made another test: 9.6G archive with 7k+ files in it, all ok. You should find out what's really is causing this 4G limit on your side, because sometimes things are not what they seem. I don't know...maybe you are saving to a FAT32 drive ? Or with a very old browser?

I'm having a serious problem with archives but doesn't match the description you sent me privately
Quote
A minute was long enough to try 2TB, but only 4GB is feasible so the server quits responding to connections, even though the ui is alive and set to on.   Manually clicking off and then on, restores connectivity.

what i get instead is that too many files are in the archive (and not their size) hang HFS forever, no UI and no WEB and i can only kill it.
On my computer it was ok up to 50k files, but since this number probably depends on the specific machine I thought it was better to limit on the time it takes to collect them.
That's why a "kb" limit wouldn't help.
Title: Re: version 2.4
Post by: rejetto on June 30, 2020, 09:44:06 AM
i should always let you preview new release when i introduce strange changes :)
this is a preview with just the limit on the search removed
https://drive.google.com/file/d/1YaqQO5mQpDZPXUvcqzgEGdM60J-jEcUI/view?usp=sharing
i'll be back on it asap but now have to work
Title: Re: version 2.4
Post by: danny on June 30, 2020, 10:25:45 PM
i should always let you preview new release when i introduce strange changes :) this is a preview with just the limit on the search removed https://drive.google.com/file/d/1hqz3kO-LdnQIa5ARVwKInB_N3--L00TO/view?usp=sharing i'll be back on it asap but now have to work
THANKS!  The security of RC7 is maximized.  But, the cost was that I can't get a file list.  Maybe it is a version change cleaning labor?  brb... (cleaning cache and rebooting)... Well, no files...
P.S.  With RC6, if the minute(s) limit was in the UI limits menu and hfs.ini, then it could have been good.
P.P.S. RC5 is still best-yet. 
...I'm having a serious problem with archives but doesn't match the description you sent me privately.  what i get instead is that too many files are in the archive (and not their size) hang HFS forever, no UI and no WEB and i can only kill it...
This isn't bad, because the UI is alive.  You can do, at interval, check if responsive, if not, cycle off/on. 
Convenience:  No need for minutes-limit.  Just if not-responsive then cycle off/on.
Title: Re: version 2.4
Post by: rejetto on July 01, 2020, 11:31:21 AM
sorry to have wasted your time, i missed a basic check
https://drive.google.com/file/d/1YaqQO5mQpDZPXUvcqzgEGdM60J-jEcUI/view?usp=sharing

P.P.S. RC5 is still best-yet. This isn't bad, because the UI is alive. 

nope, it's not alive, that's what i meant with "no UI and no WEB and i can only kill it".
That IS bad :) And when the app is freezezd the app cannot recover by itself.
Title: Re: version 2.4
Post by: danny on July 01, 2020, 12:54:18 PM
Our archive testing is going differently.   When I ask it for too much, and then cancel; the ui is alive, but can't connect. 
For me, archive of any number of files is fine, so long as the sum is less than 4gb (I believe that terabytes archive is not feasible from home internet connections).  And, for users, .tar file isn't desirable; so, current trouble is not very relevant until we go for .zip

I think that if given {.top-speed.} information, then it may be possible to limit both kb and time for archive. 
Or
Could use the single thread standard of divert--every x kb or every x millisecond, divert to update the ui (keyscan mousescan screenupdate)--a sort of scheduler.  That sort of interleaving (task swap/scheduler) is standard for single thread applications, to keep-alive the ui while it is also doing other work.  I see that it has one for downloads because it can download several files simultaneously (actually with task-swap interleaved action).   So, it exists, but probably should be modified to include the ui? 
aka:  instead of time out after too long and then stop, time out far more often and then continue (handoff to ui keepalive).
...[recursive archive]...That IS bad...
It could be a little more simple, because the user didn't want.tar files.  So, I suggest triage (non-recursive + size limit) for .tar files; and, then go for .zip later.

I'm really impressed by the new HFS2.4RC7, which I tested with Throwback14.  RC7 goes much faster than expected.  That doesn't quite knacker the recursive .tar archive problem of the inbuilt template.  But, the really fast speed has me thinking--this is candy.
2.4 is ready to go!!!
Title: Re: version 2.4
Post by: rejetto on July 06, 2020, 07:59:16 AM
guys, in order to give other tpls an easy way to have the login feature (something i want for 2.4), i'm trying to make the default tpl work without jquery. I'll let you know.
Title: Re: version 2.4
Post by: dj on July 06, 2020, 09:20:40 AM
perhaps this is a good start
Code: [Select]
const $ = document.querySelector.bind(document);
Also remember (http://youmightnotneedjquery.com/)
Title: Re: version 2.4
Post by: NaitLee on July 06, 2020, 04:40:17 PM
Is the "archive selected files" missing in 2.4? I cannot achieve that...
Title: Re: version 2.4
Post by: rejetto on July 06, 2020, 04:44:01 PM
the button is the same, "archive", but first select some files
Title: Re: version 2.4
Post by: danny on July 07, 2020, 06:41:10 AM
Feature redaction request:
Speed limit
Speed limit from single address
Max connections
Max connections from single address

I'm asking that these be deleted from HFS2.4's menu, because they should be a validation data, not a limiting factor. 
Title: Re: version 2.4
Post by: rejetto on July 07, 2020, 09:44:28 AM
i don't understand what you mean by "validation data", and how people can benefit from removing the "speed limit" option.

Please consider my english is not good as yours.
Title: Re: version 2.4
Post by: rejetto on July 07, 2020, 02:06:31 PM
a preview of the next "no jquery" default tpl.
The final goal is to let other tpl easily use the new login system, but I still have to finalize that.
Title: Re: version 2.4
Post by: danny on July 07, 2020, 03:15:03 PM
i don't understand what you mean by "validation data", and how people can benefit from removing the "speed limit" option.
The 2 speed limit and 2 connections limit cause extra work for the server and sometimes cause faults.   
So, the expectation that these settings would help, is too-different from their actual functionality (more stuck/dos).

Alternative instead of removing:  Could do just one sort of limit (if speed limit then no other limit, if connections limit then no other limit). 
Title: Re: version 2.4
Post by: rejetto on July 07, 2020, 03:22:26 PM
i don't see how the speed limit can cause faults. Can you tell?
The extra work is good work for those who need the feature.
Title: Re: version 2.4
Post by: danny on July 07, 2020, 04:55:02 PM
a preview of the next "no jquery" default tpl.
Just 2 things:  archive is unlimited and the per-file menu has some undefined
For archive limit, I suggest, no archive at root, no archive if current folder is already 4gb, no archive if more than 1 download, no archive if more than 0 upload.  That isn't enough, but it is better.  Could add a 'manual' validation with the pop-up message stating that it could take a while to get a .tar file (most users will click 'cancel' at that point because they don't want .tar file anyway). 
Title: Re: version 2.4
Post by: rejetto on July 08, 2020, 01:27:03 PM
Just 2 things:  archive is unlimited and the per-file menu has some undefined

wow, what a bad quality pre-release :D i'm always on a rush because of job :( trying again with this new attachment

Quote
For archive limit, I suggest, no archive at root, no archive if current folder is already 4gb, no archive if more than 1 download, no archive if more than 0 upload.  That isn't enough, but it is better. 

i find these rules only apparently related to the problem

Quote
Could add a 'manual' validation with the pop-up message stating that it could take a while to get a .tar file (most users will click 'cancel' at that point because they don't want .tar file anyway). 

you may be right, let's try
Title: Re: version 2.4
Post by: danny on July 09, 2020, 06:14:08 PM
Archive message is good, but 'Cancel' button is missing.  So, the appearance is:  only choice is click ok? 
Also probably don't need archive at root (a curious person like me, might click it just to see what happens).
Idea!
I think that you could relocate the archive button to the "Selection" sub-menu (archive can be next to the move button); so, then we can archive only what is selected.  That would be helpful. 
Title: Re: version 2.4
Post by: Richard_F on July 11, 2020, 11:08:15 AM
Have downloaded and will test here in a bit with Android pixel 2 phone amoung other testing device.

did it work on the pixel??
Title: Re: version 2.4
Post by: danny on July 15, 2020, 11:21:39 PM
did it work on the pixel??
There are 4 templates that work for HFS2.4 (default, takeback, throwback, mobil-light), and they all work for android phone/tablet/mobile. 
The HFS2.4 new login is actually better for mobile compatibility; and, that's why the form-based login and other new work started.
Title: Re: version 2.4
Post by: LeoNeeson on July 18, 2020, 05:32:14 AM
a preview of the next "no jquery" default tpl.
This is good news! :D (congrats). I've quickly tested and seems to work fine...
Title: Re: version 2.4
Post by: danny on July 24, 2020, 08:52:16 AM
Archive-only-selected (http://rejetto.com/forum/index.php?topic=13341.0) will get the job done. 
Instead of a default to archive whole server (error), just move the button (gone from top, move to selection sub-menu),
replacing archive-all, with selection-only archive instead, is enough to mainline HFS2.4.

I did not discover any other error.  But, I did discover that RC7 runs faster.  Kudos!!
Title: 2.4 beta, freezes when there are many files
Post by: duk6046 on July 30, 2020, 09:31:49 AM
Freezes when browsing if there are many files in the Real folder (about 250 files and folders). v2.3 is fine.
in more detail, the thinking state continues almost infinitely and freezing or lagging.

And this is a problem from previous version,
if upload speed is more than about 30MB/s and there are many connections (about 10), often server is dead without error message.
but the menu interface works normally. and minimizing instead of tray icon is a little less. my ram is 16GB and internet line is 500Mbps.
in Korea, 200~500Mbps is common.
Title: Re: 2.4 beta, freezes when there are many files
Post by: danny on July 31, 2020, 10:45:09 PM
Freezes when browsing if there are many files in the Real folder (about 250 files and folders). v2.3 is fine. in more detail, the thinking state continues almost infinitely and freezing or lagging.
To fix, just make a new VFS, and also clean/default the settings.
And this is a problem from previous version, if upload speed is more than about 30MB/s and there are many connections (about 10), often server is dead without error message.
but the menu interface works normally. and minimizing instead of tray icon is a little less. my ram is 16GB and internet line is 500Mbps. in Korea, 200~500Mbps is common.
For HFS, you can use a 2nd network interface, such as PCI or USB to ethernet set to 100 megabits, fix address, use just for HFS. Or use netlimiter software.  Also, to automatically get unstuck, use the WatchCat script--it has fair documentation so you can adjust the activities and timeframes to suit. 
Title: Re: 2.4 beta, freezes when there are many files
Post by: duk6046 on July 31, 2020, 11:47:10 PM
To fix, just make a new VFS and or clean/default the settings.

I can't solved it and i missed one thing, if i add other files under the real folder, problems occur. also v2.3 is fine.
Title: Re: 2.4 beta, freezes when there are many files
Post by: danny on August 01, 2020, 11:18:29 AM
I can't solved it and i missed one thing, if i add other files under the real folder, problems occur. also v2.3 is fine.
To see if the difference was template, can you try HFS 2.4 RC7 + Throwback template?
Title: Re: 2.4 beta, freezes when there are many files
Post by: duk6046 on August 01, 2020, 01:14:17 PM
To see if the difference was template, can you try HFS 2.4 RC7 + Throwback template?

I tried it, but the same issue.
Title: Re: version 2.4
Post by: rejetto on August 02, 2020, 11:17:24 AM
Archive message is good, but 'Cancel' button is missing.  So, the appearance is:  only choice is click ok? 

ok, i'll add ok/cancel buttons

Quote
Also probably don't need archive at root (a curious person like me, might click it just to see what happens).

assuming people don't need archive at root is both debatable and not solving the problem

Quote
Idea!  I think that you could relocate the archive button to the "Selection" sub-menu (archive can be next to the move button); so, then we can archive only what is selected.  That would be helpful. 

this is not solving but for the time being it could be a way to discourage wild actions
Title: Re: version 2.4
Post by: rejetto on August 02, 2020, 03:07:36 PM
I did not discover any other error.  But, I did discover that RC7 runs faster.  Kudos!!

the only optimization i know recently was in rc6, about file listing. Is that it?
Title: Re: version 2.4
Post by: rejetto on August 02, 2020, 03:26:44 PM
Freezes when browsing if there are many files in the Real folder (about 250 files and folders). v2.3 is fine.

hi just tried a real folder with 1250 files and it took 2.5 seconds.
There must be something causing your problem but i don't know what.
See what happens if you click Menu > debug > temporarily reset options

Also, specify exactly the version you are using, and if you see "customized template" in the lower bar of the HFS window.

Quote
if upload speed is more than about 30MB/s and there are many connections (about 10), often server is dead without error message.
but the menu interface works normally.

do you confirm you are talking about "upload" to the server ?
what do you mean by "dead", what does the browser say?
does pressing F4 twice (start/stop) make it work again?
Title: Re: version 2.4
Post by: rejetto on August 02, 2020, 04:36:10 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-rc07
Title: Re: version 2.4
Post by: duk6046 on August 02, 2020, 05:56:01 PM
hi just tried a real folder with 1250 files and it took 2.5 seconds.
There must be something causing your problem but i don't know what.
See what happens if you click Menu > debug > temporarily reset options
A: Only when added files under the real folder by dragging, problems occur even in a small amount. i also tried the reset options. the same issue on two desktops.

Also, specify exactly the version you are using, and if you see "customized template" in the lower bar of the HFS window.
A: I just applied danny's method, originally using default template. and all templates are same issue

do you confirm you are talking about "upload" to the server ?
A: It is a download from point of the visitor

what do you mean by "dead", what does the browser say?
does pressing F4 twice (start/stop) make it work again?
A: Yes
Title: Re: version 2.4
Post by: danny on August 02, 2020, 11:04:10 PM
https://github.com/rejetto/hfs2/releases/tag/v2.4-rc07
For load testing, I asked of it several impossible things simultaneously.  It worked really well.  It did large file-list, large searches, photos, music, etc... That was just the initial test. 

Also checked with the default template and see that it was revised to archive selected.  This is good.  However, I did not get any archives.    And, then I tried again, selecting just 2 files for archive.  Didn't get an archive.   Also it didn't get stuck.   So, mostly good news. 
Title: Re: version 2.4
Post by: I_LOVE_HFS on August 03, 2020, 03:39:31 PM
Hi HFS people (and Admin/developer).

I'm not sure if this is related to RC6 only or all recent builds in general, but i've noticed that when someone downloads a file over LAN (and this is better noticable with big files that take longer to transfer), the HFS UI becomes completely unresponsive i.e. i can't click/nor interact with any button. The file transfer however still continues.

After the file transfer completes, the UI becomes responsive again.

When i used version 2.3m in the past there was no such problem, so i thought this could be something important to share/let people know.
Maybe it is a problem in a certain HFS build, maybe other factors are causing this. I hope that the problem can at least be reproduced.

FYI: Host OS is Windows 10 Enterprise Edition x64 (Version 2004, build 19041.388). Client OS also is Windows 10 (don't have other info yet).
Title: Re: version 2.4
Post by: LeoNeeson on August 04, 2020, 05:52:05 AM
Hi, welcome to the forum!... :)

the HFS UI becomes completely unresponsive
Have you tried (on the PC where you run HFS) to temporally disable any security suite (like an antivirus, etc), because that unresponsiveness could be caused by your antivirus that automatically scans every uploaded/downloaded file.

When i used version 2.3m in the past there was no such problem
Have you recently tested v2.3m? It would be better if you could make a new test, just to be 100% is a bug/problem introduced of later versions (but before doing any test, you must backup hfs.ini and any *.vfs that your current version of HFS uses, just to be safe).

Another important thing: are you using the default template or a modified one?.

Cheers,
Leo.-
Title: Re: version 2.4
Post by: NaitLee on August 04, 2020, 03:40:50 PM
Just found something while uploading some eBooks from my phone to computer.
Another thing: uploading many files (>20) at once (in an upload <input> with multiselect) always fails.
Title: Re: version 2.4
Post by: rejetto on August 04, 2020, 05:19:42 PM
Also checked with the default template and see that it was revised to archive selected.  This is good.  However, I did not get any archives.    And, then I tried again, selecting just 2 files for archive.  Didn't get an archive.   Also it didn't get stuck.   So, mostly good news. 

i always try to take my time in preparing the next release, but in the end i often make mistakes like this.
I'm not pissed off for the mistake itself, but because now the release is compromised and i have to make another.
irony appreciated tho.
Title: Re: version 2.4
Post by: rejetto on August 04, 2020, 09:01:29 PM
in the meantime take this fixed tpl
Title: Re: version 2.4
Post by: rejetto on August 04, 2020, 09:07:10 PM
I'm not sure if this is related to RC6 only or all recent builds in general, but i've noticed that when someone downloads a file over LAN (and this is better noticable with big files that take longer to transfer), the HFS UI becomes completely unresponsive i.e. i can't click/nor interact with any button. The file transfer however still continues.
After the file transfer completes, the UI becomes responsive again.

this is a known problem. I think it happens also with HFS 2.3, you could check. This problem is happening with faster connections. HFS is not coping with it, and the only solution i know for the moment is to set a limit for the max speed in hfs. I advice setting it slightly lower than the actual max speed you are getting. Eg: if you are getting 100, try setting 90. But I noticed even setting 100 will give you some improvement. LMK
I hope someday we'll find a real solution to this.
Title: Re: version 2.4
Post by: danny on August 04, 2020, 09:43:57 PM
i always try to take my time in preparing the next release, but in the end i often make mistakes like this.  I'm not pissed off for the mistake itself, but because now the release is compromised and i have to make another.  irony appreciated tho.
  I saw a change to much better quality menu arrangement and I really do like it.

P.S.
Since HFS2.3, including HFS2.4, the speed limits makes stuck worse if fast internet; so,
the only helpful limits are max downloads and max download from single address.
Title: Re: version 2.4
Post by: MarkV on August 04, 2020, 11:31:39 PM
Hi,

the latest changes in the template seem to have killed the "always dark" diff template you kindly offered. Is it still possible to force dark theme with the new default.tpl?


About the UI freezing, it is unfortunately not the only problem.

At least in conjunction with the nginx reverse proxy (for getting TLS support), as long as that one download runs,
1. the UI is frozen (known),
2. I can not browse anymore, be it on the same or other machines,
3. Further downloads cannot be started.

In any case, as long as the download runs, nginx will give "Gateway timeout", as there is no reply from HFS anymore.

I see only two ways to solve that. First would be to make HFS multithreaded, possibly the programming language won't support it (is it still Delphi?). Second would be separating the interface from the engine, and have both intercommunicate over TCP/IP or UDP connections.
As a hacky workaround, I have enabled nginx' caching facilities. It will locally get the requested download from HFS as fast as possible and store it in the cache. Once the caching is complete, HFS is released and the client is served from the cache until the download completes. HFS will still freeze, but not as long as with non-cached transfers. Same for uploads, although it has less benefits.
Title: Re: version 2.4
Post by: LeoNeeson on August 05, 2020, 04:18:36 AM
» Suggestion: For those having the UI freeze, just start testing every Alpha (https://github.com/rejetto/hfs2/releases) and Beta (https://github.com/rejetto/hfs2/tags) version, one by one, going backwards from latest to older versions, until you find the latest working version that doesn't freeze the UI. This will help Rejetto a lot, since it will help him to find out where the problem was introduced. Without this, it will be very hard to track and fix this issue.

Quote from: Ramblings about Win10 (Off-topic)
Off-topic: Sadly, I can't help on this issue, since I don't use Windows 10 on purpose (and I don't have Win10 installed on any of my PCs). I'm happily living in the past, like 'Doctor Brown (https://en.wikipedia.org/wiki/Emmett_Brown) in Back from the future', and I really don't like this current present of modern computers at all (were you give away your entirely personal life, since every single action on your PC is sent to a foreign server). Yes, telemetry and Cortana can be disabled, but it's not enough (for example, read this (https://arstechnica.com/information-technology/2015/08/even-when-told-not-to-windows-10-just-cant-stop-talking-to-microsoft/) and this (https://old.reddit.com/r/sysadmin/comments/44i7xk/windows_10_enterprise_still_talks_constantly_to/)). The same goes with mobile phones, so, I only use old trusty phones (old smartphones), but I don't like to depend and rely on them neither. I don't use what I can't control, and I'm totally against Win 8/10 business model, where final users are the product. Because my boring life is only mine, and not to be part of a statistic. And this is not something against some specific country (it's not that I don't trust "X" country but I trust "this" another one, no, it applies to every single country in the world, since IMHO, all data and activities on our devices are only meant to be ours). And these are not only my words, you can clearly read HERE (http://techne.alaya.net/?p=12499) this: "retain your privacy and control over your systems" (because if you don't do it, you are automatically submitting info). For those reading this, here (https://pixelprivacy.com/resources/windows-privacy-settings/) is an updated guide and also a software (https://www.filehorse.com/download-shutup10/) to tweak Win10 privacy settings (but it's too troublesome for me, and I directly prefer to avoid Win10 at all).
» Note to Mars: please keep the off-topic text small, and don't edit and enlarge my own quoted comment (if someone wants to read it, it can copy and paste to read it). It's small because it not an important comment (but it contains my thoughts about Win10). Thanks!...

Cheers,
Leo.-