Until the html entities in code is resolved

[sir_mud]

Here is a small program to decode all html entities found in a file. Tested with .24 and up, on windows compile then just drop a file on the exe, everywhere else just call the executable with the path to the file to fix.

https://gist.github.com/mudhairless/8887091

[Mentat]

*cough*sed*cough*. :P

Edit: I think I found the bug. "(" translates to "(" and the old code is broken up into multiple html spans.

One for "&#" with a green color, one for "40" with pink, and a standalone ";".

code:
<span style="color: #66cc66;">&#</span>
<span style="color: #cc66cc;">40</span>
;

[v1ctor]

The () error on old posts should be fixed now.

No idea why the board used to convert those chars to &#40; and &#41; back then - the sources are stored like that in the data base.

[TJF]

There seem to be other issues with the data base as well. Ie when searching for

• Compiling fbc for BeagleBone Black (ARM)

I get a list of 9 matches. But only the first two link to posts. All other links report " Page not found", although the posts were found by the search function.

[aurelVZAB]

This problematic code board is still under errors with all this crazy signs...
Is there a way to replace this geshi thing with standard - without syntax coloring .
I think that is far better to have uncolored which work properly than this one which is broken.

[TJF]

I think that is far better to have uncolored which work properly than this one which is broken.

Agreed. And, the geshi highlighting doesn't work (or is incomplete) either. And it's still a problem that the geshi code blocks in board styles other than Maxthon are unlimited in length and we have to scroll a lot when we use a non-standard style.