Jürgen Spitzmüller wrote:
Jürgen Spitzmüller wrote:
There's one remaining critical bug
(#4534), but I have already uploaded a patch to bugzilla, which just needs
to be tested by one of the Windows people who can reproduce the problem.
Unfortunately, Uwe isn't able to reproduce the problem. So here is my
proposed, untested fix: substitute some filesystem::exists calls that might
throw exceptions by a new helper function doesFileExist that calls fs::exists
itself, but catches exceptions. AFAICS, the fix cannot harm. Note, however,
that there are still plain fs::exists calls in the source.
So my questions are: (a) any objections to this patch?
No.
and (b) should I
substitute the remaining fs::exists calls as well?
Yes IMHO.
Abdel.
