view tests/test-merge5.out @ 460:6409d9a0df43

add dirstate debugging commands -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 add dirstate debugging commands As I've played with various different merges and more recently rawcommit, I've found the following patch to be very very helpful in figuring out whether the dirstate is being left in a consistent or inconsistent state with respect to the current manifest. I attempted to deduce the invariants that were assumed by the current code, and then check it in this code. I may or may not have captured the design intent in this check; if not, I'd be very happy to hear more clearly what was intended, so that I can write tests to that expectation. Anyway, here's the patch. Not sure if it's a good idea to commit it to the mainline, or just leave it as a debugging aid. I attempted to package it so that it doesn't interfere with normal usage. Michael Fetterman (tweaked by mpm: remove -d magic) manifest hash: 869f5b5f954dc0f46ba27322359e811d5e21d71c -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCvP77ywK+sNU5EO8RArmtAKCCVuI2slANzWZ26P5edtH/ixdwNwCfZLWl 5P+V+C92II3usO4YW2MULKY= =/Pv4 -----END PGP SIGNATURE-----
author mpm@selenic.com
date Fri, 24 Jun 2005 22:51:39 -0800
parents c2b9502a4e96
children d45d1c90032e
line wrap: on
line source

+ mkdir t
+ cd t
+ hg init
+ echo This is file a1
+ echo This is file b1
+ hg add a b
+ hg commit -t 'commit #0' -d '0 0' -u user
+ echo This is file b22
+ hg commit '-tcomment #1' -d '0 0' -u user
+ hg update 0
+ rm b
+ hg commit -A '-tcomment #2' -d '0 0' -u user
+ yes k
+ hg update 1
this update spans a branch affecting the following files:
aborting update spanning branches!
(use update -m to perform a branch merge)
+ exit 0