Reason Developer Zone

• Improve UTF-8 handling and work around php5 htmlspecialchars bug
  • Description: There are a number of related issues regarding utf-8 handling, some of which are specific to php5:
    1. in php5, htmlspecialchars() returns an empty string if passed a string containing an invalid utf-8 character (Bug report.) The workaround is apparently to not pass the function the encoding parameter.
    2. There may still be some places where Reason is mangling UTF-8, because there are still odd invalid character glyphs that pop up from time to time. unhtmlentities() is a suspect, but it appears to only be used when prepping strings for use in headering. We should follow a string from userland to the database on the Reason backend and make sure there are no iffy steps
    3. We should be enforcing UTF-8 validity in more locations. The most important ones seem to me to be in the database (which is does not recognize the data as being UTF-8 encoded at the moment) and at the application perimiter, as part of ensuring that the data is in good shape in the first place (this is something of a security issue, as IE6 has a bug in handling invalid UTF-8)
  • Current State: In Planning
  • Deadline: Not currently defined
  • Project Lead: mryan, nwhite
  • Scale: medium
  • Initiation Date: Feb 27th, 2008
• Rework String Escape Method(s)
  • Description: Reason uses a hodgepoge of methods to escape strings for use in SQL queries. For security and compatibility reasons, it would be a good idea to standardize the way we are escaping these values.
  • Current State: In Planning
  • Deadline: Not currently defined
  • Project Lead: mryan
  • Scale: medium
  • Initiation Date: Apr 16th, 2007

Show all projects