Filter adds < before closing 'a' HTML tags

I'm not seeing this after I made a few changes. Can you please try the -dev version this evening and see if you still see it? If you do, please tell me what kind of term indicator you are using.

Okay. It's hard for me to know if it's fixed when I can't reproduce it. And I do test things.

Hmm, I still don't see it. Is this only happening on short phrases where the term is at the end? Are you using substring substitution? Have you tried it without the <strong>?

Aaarrrggghhh! WYSIWYG editors are the bane of module developers' lives. Please turn it off and see if you can recreate the problem.

I took a short vacation so I'm still trying to catch up.

Still catching up. Real Life™ seems to be very hectic lately.

I have run into this problem, and I don't use a WYSIWYG editor on my site. I have the module temporarily turned off, but I'd be happy to turn it on and give you a url if that will help.

In my case, I used G2 before updating glossary, and I'm wondering if that may actually be part of the problem.

I am seeing this problem as well. I did not use G2 before (as codeknitter had), but am using FCKeditor. I tried disabling FCKeditor, but the extra > is still showing when Glossary is enabled. It was running without problems, so I'm unsure why it suddenly started adding the extra >. I did do a recent upgrade to a newer version of FCKeditor, and it's possible this problem started after that, but I can't be sure.

Would appreciate your help on this problem. Let me know if there's anything I can do in testing.

Nate/quicksketch looked into this and found it only happens when the mbstring extension is not available. That probably explains why it's reproducable in some environments and not others.

To reproduce this on a server *not* having this problem, change the following line in glossary.module:

  $multibyte_enabled = extension_loaded('mbstring');

to:

  $multibyte_enabled = !extension_loaded('mbstring');

And in includes/unicode.inc change:

if (!function_exists('mb_strlen')) {

to:

if (function_exists('mb_strlen')) {

(basically trick it into thinking your server isn't up to par)

Then voila; arrows everywhere! :)

I can confirm that this patch fixes the issue, although I'm not positive it's the correct fix. Worthy of "needs review" at any rate.

And fwiw, our glossary settings are:

Match type: Word
Case sensitivity: Case insensitive
Replace matches: All matches
Blocked elements: abbr acronym
Link type: normal
Term indicator: use abbr element

I have added the patch fix above and it seems to work fine for me as well. thank you...

That patch does not work. Glossary module still adds < to external links in a text. Waiting for solution...

I've applied the patch, as well. It did not appear to be working at first. I cleared the cache and it seems to work fine.

bump

I'm still having this issue, the patch did not work for me.

The issue only seems to be occurring under special circumstances for me, that is, if the next character after the term is a "<", whether a closing tag wrapping the term, or an open tag unrelated to the term, then another "<" will be added to the output.

EDIT: Ok totally didn't read properly the first time. Installing php-mbstring fixed this issue for me.

I have had the same issue.
To solve it i reimplemented the _glossary_insertlink() function.

Open Issue:
I dont understand what for the rtl and ltr stuff should be? Why i have to watch for this here?
My version dosent have this. Do right to left writing people do: [code]some crazy letters[/code]. Or do they have to use normal html (codes) like all other?

bump!