Textpattern CMS

#1 2011-02-16 23:53:43

morgant

Plugin Author

From: Vermont, United States

Registered: 2004-06-22

Posts: 63

Website

Please help determine cause of HTML entities encoding in comments

I’m troubleshooting a Textpattern 4.0.7 install that’s getting certain characters in comments encoded into HTML entities, but are then further encoded so they appear as raw HTML entities in the comments on the live site. This blog definitely needs to be upgraded, and I’m working on it, but if I can fix the comment encoding issues first, that’d be ideal.

I’m using the following text string for testing:

Test & — ' " ‘ “

The POST data is:

Test+%26+%E2%80%94+%27+%22+%E2%80%98+%E2%80%9C

And the raw resulting HTML (even on the preview, so before it’s even put into MySQL) is as follows:

Test & — ' " ‘ “

Got that? :D

It’s Textpattern 4.0.7 w/a minor tweak to publish.php (only to category matching, so unrelated to comments) running under Apache 2.2.17 w/PHP 5.2.14 on Mac OS X Server. The two plug-ins that would seem to have most likely effect on parsing comments are gbp_permanent_links 0.14 patch 6 (incl. gbp_admin_library) & ngd_recaptcha. I’ve tried completely disabling both, with no change in effect.

I have also upgraded Textile to 2.2 (from the Textpattern 4.3.0 release, it was already running a newer version than was included with Textpattern 4.0.7) thinking that the extra level of HTML encoding might be occurring there, but also no change in functionality.

I’ve enabled debugging in Textpattern and gbp_permanent_links, but am not seeing any output related to comments.

Any suggestions would be greatly appreciated.

Last edited by morgant (2011-02-16 23:55:22)

---

Morgan Aldridge
http://www.makkintosshu.com/