version 2.4

[Mars]

mars was just guessing, but he was not right

it's understandable, javascript gives me Urticaria on neurons 

[LeoNeeson]

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).

[LeoNeeson]

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).

[rejetto]

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?

[Mars]

 may be wrong but it must be about the compatibility of old default template with the new version of jquery

[LeoNeeson]

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...

[rejetto]

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.

[rejetto]

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

Now we have sorting, and much more translatable strings.

[LeoNeeson]

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' or the old legacy template that I'm sharing HERE. 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! I'll check it out and do some tests on the next few days...

[MarkV]

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.

[bmartino1]

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).

[Chris Harris]

Just installed hfs24b6.exe and the interface is clean and very easy to use.

Thanks
Chris

[Fysack]

" latest tests I did the other day, were only using" Strighten up Leo. We can absolutely not use that in season 5 Le Bureau

[LeoNeeson]

" 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!!

[LeoNeeson]

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). 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:

@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 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.