• last updated 14 hours ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
If there's no whitespace before truncate_len, then just truncate at the given index, instead of leaving us with an empty string

add an apm_version_sortable public function, add a test which checks the db upgrade scripts are well ordered.

  1. … 4 more files in changeset.
merge of changes on oacs-5-1 branch between tags jcd-merge-5-1-20040517 and jcd-merge-5-1-20040628 to HEAD

  1. … 687 more files in changeset.
Added more description on how to correctly use the ad_html_text_convert proc.

Added an example.

fix from Sloan - 100 iterations was not enough in some cases

add some output to help diagnose error in ad_parse_html_attributes_upvar

merge all the changes on the oacs-5-1 and dotlrn-2-0 branches up to tag jcd-merge-20040420

  1. … 647 more files in changeset.
Fixed bug #1450: Don't dress img src URLs as links

- Fixed bug: default for ad_html_text_convert was set to truncate at 70 chars

- The algorith in util_close_html_tags for including ellipsis in determining when to break was flawed -- caused it to break too son

  1. … 1 more file in changeset.
- Improved text formatting and clipping

- Added test cases

  1. … 1 more file in changeset.
merge of changes between oacs-5-0-0b4 and oacs-5-0-0-final to HEAD

  1. … 286 more files in changeset.
Fix handling of HTML comments in util_close_html_tags. Fixes bug #1386.

Changed ad_quotehtml to use "string map", which is faster than the old

regexp approach. Thanks to Daniel Mantione...

Slight hack to make HTMLified text not have whitespace at the end

  1. … 1 more file in changeset.
merge of changes on oacs-5-0 branch to oacs-5-0-0b4 tag

  1. … 598 more files in changeset.
starting to debug bug 810 and added some debug printouts

Cleaning up the api-doc of ad_html_text_convert, after searching half an hour for something that would format the contents of a richtext widget and only finding this proc after looking into the display code of the form builder. (rant: It'd be awesome if those who add cool functionality would find the extra minute to at least mention it in the api-doc, /endofrant). Also changed it so that it returns the empty string when input is empty, so it can deal with NULL content/format pairs in the database conveniently.

patch 360 bug 1154

hopefully the last problem with -- bug 1086

add -- to all string_truncate calls

  1. … 11 more files in changeset.
add a -- to ad_html_text_convert, make util_close_html_tags not eat < not part of a tag

Fixed bug: hadn't removed all references to allowed_url_attributes_list

Let AllowedAttributes, AllowedProtocol and AllowedTags to contain a * to make it easier to run an insecure site if you really don't care. Especially having to explicitly enumerate allowed tags can be a pain.

Don't print the illegal tag

Don't put HTML in the security error message. There's no need.

Fixed bug #466, plus another bug in ad_html_to_text

  1. … 1 more file in changeset.
Fix and test-case for bug #409: ad_html_to_text bombs. Turns out the bug was fixed a while ago, but there was another bug.

  1. … 1 more file in changeset.
Applied patch #135: ad_html_to_text -no_format param should ignore img

  1. … 1 more file in changeset.
ad_text_to_html: return empty if the text is empty. Make the first <p> tag not have margin-top

Fixed two bugs in ad_html_to_text:

- Tags that start with whitespace were still considered tags, and thus < p> would get treated equivalently to <p>.

- The regexp to find malformed tags wasn't bound to start and end of the inside of the tag, thus tags that had just a subset of welformedness would get interpreted.