Opened 3 years ago
Last modified 7 months ago
#6754 reopened defect
libavfilter/signature_lookup.c:494: suspicious compare ?
| Reported by: | dcb | Owned by: | |
|---|---|---|---|
| Priority: | normal | Component: | avfilter |
| Version: | unspecified | Keywords: | |
| Cc: | Blocked By: | ||
| Blocking: | Reproduced by developer: | no | |
| Analyzed by developer: | no |
Description (last modified by cehoyos)
ffmpeg-3.4/libavfilter/signature_lookup.c:494]: (warning) Result of operator '|' is always true if one operand is non-zero. Did you intend to use '&'?
Source code is
if (meandist < minmeandist ||
status == STATUS_END_REACHED | STATUS_BEGIN_REACHED ||
mode == MODE_FAST){
maybe better code
if (meandist < minmeandist ||
status == (STATUS_END_REACHED | STATUS_BEGIN_REACHED) ||
mode == MODE_FAST){
Change History (10)
comment:1 Changed 3 years ago by cehoyos
comment:2 Changed 3 years ago by dcb
How did you find the issues?
A static analyser called cppcheck, available from sourceforge.
I find it very useful for C and C++ development. It catches
a lot of problems compilers don't find.
comment:3 Changed 7 months ago by Balling
- Resolution set to duplicate
- Status changed from new to closed
Merged in #8156.
comment:4 Changed 7 months ago by Balling
- Resolution duplicate deleted
- Status changed from closed to reopened
What the... Okay.
comment:5 follow-up: ↓ 6 Changed 7 months ago by cehoyos
- Description modified (diff)
comment:6 in reply to: ↑ 5 ; follow-up: ↓ 7 Changed 7 months ago by Balling
Replying to cehoyos:
Suspecious? Really? Now it never become 0, so mode is never checked. "Result of operator '|' is always true if one operand is non-zero." And == have higher precedence. In this case STATUS_BEGIN_REACHED is 2. So...
comment:7 in reply to: ↑ 6 ; follow-up: ↓ 8 Changed 7 months ago by cehoyos
comment:8 in reply to: ↑ 7 Changed 7 months ago by Balling
comment:9 follow-up: ↓ 10 Changed 7 months ago by cehoyos
Now I have two questions:
Why do you believe I disagree?
Why didn't you sent a patch made with git format-patch to the development mailing list?
comment:10 in reply to: ↑ 9 Changed 7 months ago by Balling
Replying to cehoyos:
Now I have two questions:
Why do you believe I disagree?
Why didn't you sent a patch made with git format-patch to the development mailing list?
Well, because it is already midnight here?
Please send your patches made with git format-patch to the FFmpeg development mailing list, patches (and patch suggestions) are ignored here.
How did you find the issues?