Show Likes

This section allows you to view all posts where this member received or gave a like to.

Features / Re: New revs
Nao « on October 24th, 2017, 03:33 PM »
[Commit revision 2ebdde8]
Author: Nao
Date: Tue, 24 Oct 2017 15:26:42 +0200
Stats: 3 files changed; +5 (insertions), -19 (deletions)

  • I'm officially giving up on @viewport support. This great little bit of CSS never took off and browser support is as obscure as ever. No need to waste time on it then. (sections.css, Class-CSS.php, index.template.php)
  • Updated @keyframes and flexbox support to remove prefixes on some browsers that were updated since the last time I looked into it. (Class-CSS.php)

[Commit revision ad4eada]
Author: Nao
Date: Tue, 24 Oct 2017 15:26:58 +0200
Stats: 1 file changed; +6 (insertions), -6 (deletions)

Archived fixes / Re: it keeps giving me this error in Error Log
Nao « on February 11th, 2014, 12:28 AM »
Well, at least we get to see people who silently install Wedge... :P
Features / Re: New revs
Nao « on July 22nd, 2017, 11:41 AM »
[Commit revision 389ed36]
Author: Nao
Date: Sat, 22 Jul 2017 10:28:59 +0200
Stats: 4 files changed; +0 (insertion), -0 (deletion)

  • Smaller size, dark-friendly rank icons. (assets/rank*.gif)

[Commit revision d060d06]
Author: Nao
Date: Sat, 22 Jul 2017 11:08:01 +0200
Stats: 3 files changed; +96 (insertions), -99 (deletions)

  • array() to [] conversions in the one file that screams "I'm there to optimize sizes, so optimize me!!" (Subs-Cache.php)
  • Same for Ajax.php because I can, also fixed a thought conversion call that was taking more space than needed in the database. (Ajax.php)
  • Also shaved a few bytes off serialized strings with integers in them. (Subs.php)

[Commit revision 0c3cd07]
Author: Nao
Date: Sat, 22 Jul 2017 11:10:20 +0200
Stats: 1 file changed; +14 (insertions), -7 (deletions)

  • Added proper hyphens support to Wess, now that, years later, support has grown a bit. (Class-CSS.php)
  • Dynamic functions shouldn't be cached across all CSS files... That can lead to some unexpected results. Not that it matters, really. (Class-CSS.php)

[Commit revision bcb61a6]
Author: Nao
Date: Sat, 22 Jul 2017 11:12:20 +0200
Stats: 2 files changed; +2 (insertions), -2 (deletions)

  • Simplified a query. (Memberlist.php)
  • PHP caching code was probably a bit too greedy. Wait time after uploading a file was reduced to 2 seconds max, with perfect results on my sometimes busy server. (Subs-CachePHP.php)

[Commit revision 8d4d155]
Author: Nao
Date: Sat, 22 Jul 2017 11:14:36 +0200
Stats: 2 files changed; +5 (insertions), -2 (deletions)

  • No need to log an error when someone logs off and has another tab around which still attempts to fetch notifications... It's okay, calm down. (Notifications.php)
  • Account for potential inputs inside select boxes. (sbox.js)

[Commit revision cee6d78]
Author: Nao
Date: Sat, 22 Jul 2017 11:17:20 +0200
Stats: 1 file changed; +4 (insertions), -4 (deletions)

  • While I no longer use a proxy, I've established that the most useful IP data to be recorded is the one discovered through the proxy's special headers. If nothing was found (e.g. no proxy), that data will be the same as usual anyway... Way, way more useful for most tasks. (Class-System.php)

[Commit revision ec916e4]
Author: Nao
Date: Sat, 22 Jul 2017 11:18:25 +0200
Stats: 2 files changed; +2 (insertions), -2 (deletions)

  • Why should guests want to collapse a category? Excuse me for a minute-- why would members want to, either? Anyway, guests, one less useless button for you... (Subs-Boards.php, Subs-BoardIndex.php)

[Commit revision 1937057]
Author: Nao
Date: Sat, 22 Jul 2017 11:20:23 +0200
Stats: 3 files changed; +6 (insertions), -13 (deletions)

  • Moved some of the app_error_count logic around. I haven't seen significant improvements, but it makes more sense to me. (index.php, Subs-Template.php)
  • Spacinazi. (index.template.php)

[Commit revision 3349b81]
Author: Nao
Date: Sat, 22 Jul 2017 11:25:34 +0200
Stats: 1 file changed; +1 (insertion), -96 (deletions)

  • Removed my very own sticky positioning polyfill. Why? Chrome recently added proper support for 'position: fixed', and I figured it would mean 90% of users would no longer need a polyfill. Older browsers will simply do without it, it's not a deal breaker anyway. Rest in peace polyfill, I was proud of you. (topic.js)

[Commit revision fa1b720]
Author: Nao
Date: Sat, 22 Jul 2017 11:28:22 +0200
Stats: 1 file changed; +30 (insertions), -30 (deletions)

  • Make sure search engines can't see context menus, even if some items remain for guests. This makes for uncomfortable pseudo-URLs that get parsed for no reason... Also, more array() to [] conversions. (Display.php)

[Commit revision 878793c]
Author: Nao
Date: Sat, 22 Jul 2017 11:28:43 +0200
Stats: 2 files changed; +5 (insertions), -5 (deletions)

  • Updated a few files with the https version of the Wedge URL. (ScheduledTasks.php, Credits.php)
  • I didn't need some hack in PM code... (pm.js)

[Commit revision c43f32d]
Author: Nao
Date: Sat, 22 Jul 2017 11:30:27 +0200
Stats: 4 files changed; +12 (insertions), -12 (deletions)

  • More protocol conversions. (install.php, Help.english.php, index.english.php, Admin.english.php)

[Commit revision aabdd5f]
Author: Nao
Date: Sat, 22 Jul 2017 11:33:32 +0200
Stats: 1 file changed; +8 (insertions), -8 (deletions)

  • Fixed error handling to use proper PHP logic for error constants, rather than the weird situation it was in in SMF... It was very old code, I reckon. Not finished with it though, I've got a better version live at but I need to test it further. (Errors.php)

[Commit revision 7f9ec53]
Author: Nao
Date: Sat, 22 Jul 2017 11:36:23 +0200
Stats: 1 file changed; +18 (insertions), -16 (deletions)

  • Added support for MySQL strict mode -- more specifically, the ability to disable it through a global variable. I wrote that for a previous server, and never bothered to commit it because I didn't need it anymore by the time I was committing everything. But still, can be useful to some... I know I wasted a few hours on that one..! (Class-DB.php)
  • Fixed {literal:...} code not allowing for flexible strings. (Class-DB.php)

[Commit revision 14d6a15]
Author: Nao
Date: Sat, 22 Jul 2017 11:41:01 +0200
Stats: 1 file changed; +32 (insertions), -20 (deletions)

  • Wrote a better, more self-explained version of invertAll() (kept the silly name though, for compatibility reasons), where it's made more explicit what the function does, *and* it now also ensures that events linked to those checkboxes are called as needed. (script.js)
  • Tweaked notification popup code to accomodate for any custom changes to the notification button HTML. I know it helped me. Also various additions and fixes that I won't bother listing. (script.js)
  • Reduced length of some select box strings. The idea is, you only need to escape the "<" sign... Everything else is fair game. As a result, the new stuff is more readable, and shorter of course. I always go for shorter. (script.js)
The Pub / Re: PHP 5.4?
Nao « on June 9th, 2017, 11:00 PM »
Quote from Kian on May 29th, 2017, 10:56 PM
Je ne sais plus :^^;:  peut-être un problème de PHP corrigé en éditant 2 lignes, ou alors causé finalement par un plugin je ne sais plus exactement.
Je passe mes journées à améliorer des designs et à corriger des conneries dans ce genre, hein, si besoin :P

( pour voir ce sur quoi je bosse depuis quelques mois... Ca tourne sous Wedge, bien sûr !)
Installed UC on Android KitKat, works perfectly. Dunno why you're having this...

And uc is very lazy... it took all my ram just for one tab lol... and slow to load.
The Pub / Re: PHP 5.4?
Nao « on June 12th, 2017, 12:53 PM »
It's not *much* faster in the case of Wedge. It is definitely faster, but not noticeably so, I'd say. I compared both PHP versions with it.
I absolutely recommend that everyone upgrades to PHP 7.x, of course.
But Wedge is based off SMF, which itself, at the time of forking, was compatible with PHP 4... So I started by making it PHP 5-only, then PHP 5.3, then PHP 5.4... I'm basing my assumptions off, and as you can see by going 5.4+ I cut myself off from potentially ~25% of all PHP users. It was already a bold move....
Going 7.x+ only will cut me off from 95% of all PHP users. That's not something I can do, obviously. I still have some hope that Wedge will eventually make itself known as one of the better free platforms.
The Pub / Re: PHP 5.4?
Nao « on May 28th, 2017, 01:33 AM »
Quote from developer on May 26th, 2017, 02:11 PM
U crazy?
The Pub / Re: PHP 5.4?
Nao « on May 28th, 2017, 01:33 AM »
Quote from developer on May 26th, 2017, 02:11 PM
U crazy?
Features / Re: New revs
Nao « on April 18th, 2017, 01:27 AM »
[Commit revision e80acb5]
Author: Nao
Date: Tue, 18 Apr 2017 01:27:11 +0200
Stats: 1 file changed; +12 (insertions), -20 (deletions)

  • Two new techniques for the key-value file cache, to combat possible race conditions. First of all, Wedge will no longer attempt to delete an expired file when it sees it expired. It's the script's responsibility to regenerate the cache (and thus the file) when it gets an invalid result (also, this saves more bytes, and I like saving bytes). Secondly, files are now created under a random name (like in Subs-CachePHP.php), and then renamed if everything works. This should help in a LOT of situations. And I'm serious. Just like I fixed race condition issues with the PHP minifier, I'm hoping this will also be the case here, although it happens much less often-- in fact the only reason I fought this is because Opera Chromium is bloody running two queries on my site at the exact same millisecond, even if I didn't ask for that. For once, it's Opera's fault. (Subs-Cache.php)
  • Seriously, someone should report that to them... Opera is actually loading Atom feeds even though I never added them to my feed list.
The Pub / Re: PHP 5.4?
Nao « on April 3rd, 2017, 10:51 PM »
Tiens, je me souvenais même pas que PNR était passé sous Wedge... :^^;:
Faudra faire quelque chose pour la bannière invisible, par contre... Ça fait tache ! :(
Tu postais pour quoi ? Un problème avec ta version de PHP..?