On 2016-04-19 06:03:52 +0000, Dave Yeo said:

The problem is that symlinks on OS/2 (unless using TVFS.IFS) are
inherently fragile. It just takes one utility that is not aware of them
to break things and not everything is linked against kLIBC. In FFmpegs
case we use lxlite, which is a Pascal program and knows nothing of symlinks.

lxlite works with executables and we can't use symlinks for executables anyway (because the kernel doesn't support them and won't be able to load such binaries). If lxlite encounters a symlink to a dll or an exe, it will simply skip it (becuase it will only see a text file) - and this is asbolutely fine and safe. So lxlite is not a problem. You also speak theoretically, like KOMH. In practice, I personally don't know a single problem that would come out of symlinks and require something non-trivial to solve it. And I do A LOT of stuff.

I personally prefer to only use symlinks when needed, such as
_virtualenv, and currently using my older environment can often pass all
the FATE tests, whereas the new environment has more failures.
As leaving it in doesn't hurt your use case anyways, my vote is to leave
it in.

I always use symlinks and have absolutely no problems with that. You use the RPM/YUM UNIXROOT environment too (IIRC) so you indirectly use them as well. I can assure you, you wouldn't see all the current OS/2 progress if we didn't switch to the Unix-based toolchain. And symlinks is part of it. So it's not about the number of votes, it's about progress and future.

http://fate.ffmpeg.org/history.cgi?slot=x86.os2.446
http://fate.ffmpeg.org/history.cgi?slot=x86.os2.492

Regarding tests, that's a different story. I believe symlinks have a tiny contribution in test failures (if at all). And that's in general, again. In case of FFmpeg, symlinks don't make any harm (at least you didn't present any proof of that so far).

--
Kind regards,
Dmitriy Kuminov
CPO of bww bitwise works GmbH


_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Reply via email to