| Summary: | Missing braces in code | ||
|---|---|---|---|
| Product: | fontconfig | Reporter: | Pedro Gimeno <usr352> |
| Component: | library | Assignee: | Keith Packard <keithp> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | trivial | ||
| Priority: | high | ||
| Version: | 2.2 | ||
| Hardware: | x86 (IA32) | ||
| OS: | All | ||
| Whiteboard: | |||
| i915 platform: | i915 features: | ||
Good catch. I checked HEAD and found that it already had braces around that block. This "shouldn't" cause any failures though on most machines as a null pointer is generally "less" than any non-null pointer. I fixed the bug, but I'm not planning on cutting another release for it. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.
Looking at the differences between FC 2.2.1 and 2.2.2 I've found this snippet: --- fontconfig-2.2.1/src/fcxml.c 2003-04-16 01:38:06.000000000 +0200 +++ fontconfig-2.2.2/src/fcxml.c 2003-11-10 19:10:15.000000000 +0100 @@ -963,6 +963,8 @@ FcStrtod (char *s, char **end) v = strtod (buf, &buf_end); if (buf_end) buf_end = s + (buf_end - buf); + if (buf_end > dot) + buf_end -= dlen - 1; if (end) *end = buf_end; } Note the lack of braces around the sentences within the IF. It seems to work just by luck (unless 'dot' can be negative for some reason but I don't think that's possible).