| Comments: |
| Posted By: ****** Date: 2:28 AM On 06-23-2005 |
| Stuff like: <3 is captured by the HTML filter, but it shouldn't catch it since it doesn't even qualify as poorly formed HTML. |
| Posted By: ****** Date: 2:47 AM 06-23-2005 |
| This bug strongly correlates with #1 and will probably be fixed as a result of fixing #1. The use of strip_tags() when trying to determine how much of a string is HTML is clearly not a good idea, 'cause strip_tags() seems to think that everything < or > (even <<< - but not the last two of >>>, only if preceded by multiple <'s) is HTML. Examples: strip_tags("<3"); // Returns: '' strip_tags("<3<>"); // Returns: '' strip_tags("<3<<<<>"); // Returns: '' strip_tags("<3>>>>"); // Returns: '>>>' strip_tags("<<<<3>>>>"); // Returns: '' strip_tags("<<<<3>"); // Returns: '' Thus '<3' is identified as 100% HTML, which is obviously not the case, so the identification will have to become more fine-grained, also to eventually for users to allow certain HTML tags in their content. (See: #1) |
