On 2017-09-06 23:45, Richard Hipp wrote:
On 9/6/17, Stephan Beal <[email protected]> wrote:
i'm speculating, based on the fact that you're pulling "shun" info, that
you once shunned one of those files. ALL empty files have the same hash
code, so if you shunned one of them, you've shunned them all.
I was stumped. Then I read Stephan's theory and smiled. I think he
may be on to something.
I had the exact same idea in the beginning but the fact that the other
files with a length of 0 are not touched made me discard it. However,
turned out it's the case after you provided the SHA3 of an empty file
and I checked it just now.
A zero-length file has a SHA3 hash of
a7ffc6f8bf1ed76651c14756a061d662f580ff4de43b49fa82d80a4b80f8434a and a
SHA1 hash of da39a3ee5e6b4b0d3255bfef95601890afd80709. Do either of
those two hashes appear in your shun list?
Yes. Someone obviously shunned
a7ffc6f8bf1ed76651c14756a061d662f580ff4de43b49fa82d80a4b80f8434a, maybe
even me a long time ago to protect the other - private - empty files.
That makes it even more complicated for me now.
If I unshun
a7ffc6f8bf1ed76651c14756a061d662f580ff4de43b49fa82d80a4b80f8434a now the
next one to check in (run the check-in script) would cause all the other
empty files to be distributed to everyone else, wouldn't they?
Why are those not removed, by the way? They got the same SHA3 key.
I wonder if we should add some magic to Fossil that prevents the
shunning of empty files? Or, perhaps add a warning of some kind when
files less than (say) 8 bytes in length are shunned?
No idea, as in my case I don't want the files to be distributed but keep
them private for everyone. This also means for me that they shouldn't be
deleted by Fossil.
Thanks, and regards
Thomas
_______________________________________________
fossil-users mailing list
[email protected]
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
