On 3/13/2017 5:17 PM, Richard Hipp wrote:
I think the best solution here would be to change the RE to be
"([0-9a-f]{40,64})". Does that fix the problem?
That improves things, and moves to a new problem. A number of commands
(sensibly) trim their output to a 40-character prefix of the UUID. So
now I have a different collection of tests that are doing exact matches
and failing. My planed next step is to regularize how the test suite
tests UUIDs for equality, and in most cases assume that a matched 40
character prefix is as good as equal. I'm sure that will move the
problem further down field.
To push progress forwards, I patched login.c to have an implementation
of strtok_r() which it suddenly needs but which is not provided in
MinGW's version of libc (which is really a thin veneer over MSVCRT.DLL).
The only previous usage of strtok_r() was in a cgi.c, in code that is
currently dead inside a #if 0 block.
--
Ross Berteig r...@cheshireeng.com
Cheshire Engineering Corp. http://www.CheshireEng.com/
+1 626 303 1602
_______________________________________________
fossil-dev mailing list
fossil-dev@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/fossil-dev
