Mon, 13 Jun 2005 17:29:04 -0800 hg pull: more query fixes
mpm@selenic.com [Mon, 13 Jun 2005 17:29:04 -0800] rev 328
hg pull: more query fixes -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hg pull: more query fixes Pull was getting confused by single-element branches because it had seen the top end already. Also, the previous change to do seenbranch on the base of the branch was bogus. Minor cleanup and a bit more diagnostics in the branch-narrowing code. manifest hash: 71d4dced9f74bef43d3a54d086dccd0a96d68e26 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrjLgywK+sNU5EO8RAsxSAJ92U31Pz7K14Ctje4SEec2857Ut7QCgoVwY ho65mloslCSidLsEnxYfBDI= =Wts9 -----END PGP SIGNATURE-----
Mon, 13 Jun 2005 15:01:38 -0800 Tweak README for branch, push, and .hgrc
mpm@selenic.com [Mon, 13 Jun 2005 15:01:38 -0800] rev 327
Tweak README for branch, push, and .hgrc -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Tweak README for branch, push, and .hgrc manifest hash: f1937ad49013e7e40e6555f2acf1210563e6a96d -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrhBSywK+sNU5EO8RAr0OAJ9T+DcMcIvXp9ajI46kXM40GuHosgCfRxfY 8uhwZqdaR0JWlWff0nC6t1Y= =3I60 -----END PGP SIGNATURE-----
Mon, 13 Jun 2005 14:53:33 -0800 mdiff: fix the fix
mpm@selenic.com [Mon, 13 Jun 2005 14:53:33 -0800] rev 326
mdiff: fix the fix -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 mdiff: fix the fix I saw you check in the thing back in but with another typo. You check in "la < lenb", should be "la < lena". So we are even :-) Chris manifest hash: 57157502250b0120bcc12a75a530d4c133673b39 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrg5tywK+sNU5EO8RAvLPAKCOs76LKPniKHFa15YwzWWsNZ5/xACfesl3 WCiUjdeYUtU3jVKCEwmjAT0= =+GFy -----END PGP SIGNATURE-----
Mon, 13 Jun 2005 13:43:48 -0800 mdiff: reinstate new algorithm
mpm@selenic.com [Mon, 13 Jun 2005 13:43:48 -0800] rev 325
mdiff: reinstate new algorithm -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 mdiff: reinstate new algorithm This unreverts the new algorithm with a fix from Chris (s/len/lena) and adds some comments on what it's doing. manifest hash: 75fc1acee1926e57d495f67a44cd88d9555f2356 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrf4UywK+sNU5EO8RAoRzAKCA2vpUAGNqTkDeba3YHo6XXht7VgCfXQK0 /j5yv5cucnsYezCdclpftOA= =FNMD -----END PGP SIGNATURE-----
Mon, 13 Jun 2005 12:04:04 -0800 hg merge: some getchangegroup fixups
mpm@selenic.com [Mon, 13 Jun 2005 12:04:04 -0800] rev 324
hg merge: some getchangegroup fixups -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hg merge: some getchangegroup fixups - - filter more duplicates - - request batching - - lookup ordering - - request counting After these fixes, grabbing 1800 changesets on top of the 28k changeset kernel history takes 23 round trips rather than a couple hundred. manifest hash: 78e893b68000267a8c7ea49fd38758f2fc96269f -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrea0ywK+sNU5EO8RAv3cAKCQveivdApJIlslTCMLMp2VogRXfACeNRnx Ihsf5yLz60fFpqlzAmZBACk= =Euku -----END PGP SIGNATURE-----
Mon, 13 Jun 2005 12:01:12 -0800 lazyparser speed ups
mpm@selenic.com [Mon, 13 Jun 2005 12:01:12 -0800] rev 323
lazyparser speed ups -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 lazyparser speed ups When we do __contains__ on a map, we might as well load the whole index. Not doing this was slowing down finding new changesets quite by a factor of 20. When we do a full load, we also attempt to replace the revlog's index and nodemap with normal Python objects to avoid the lazymap overhead. manifest hash: 9b2b20aacc508f9027d115426c63a381d28e5485 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCreYIywK+sNU5EO8RAoNHAJ9+LmXqsTQb9Bh3mZHq0A0VfQOleQCffHmn jC/O0vnfx5FCRsX2bUFG794= =BDTz -----END PGP SIGNATURE-----
Mon, 13 Jun 2005 11:58:33 -0800 hgweb: Fix reloading of index
mpm@selenic.com [Mon, 13 Jun 2005 11:58:33 -0800] rev 322
hgweb: Fix reloading of index -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hgweb: Fix reloading of index We were reloading every request because we weren't saving the old mtime. manifest hash: 507330727957ff53c48b38f427f0b70469a49271 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCreVpywK+sNU5EO8RAp/YAJ9cXh2II4DGAQtRc58yj4hi49C8jgCfT7bV 22w73T2FiYBjzMk8N8STUVI= =eghX -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 22:05:08 -0800 Transparent proxy support
mpm@selenic.com [Sun, 12 Jun 2005 22:05:08 -0800] rev 321
Transparent proxy support -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Transparent proxy support Originally from "Michael S. Tsirkin" <mst@mellanox.co.il> manifest hash: 74cf7456ef35ff8d4c007544f0d1a57c69d3c929 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrSIUywK+sNU5EO8RAje1AJ41ALW8soF78Mo3UTraV1QQvJoFSQCgrqvc I9ohlI4hzdjOD+wSwRGlERQ= =Ugfi -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 20:39:08 -0800 hg push: propagate return code
mpm@selenic.com [Sun, 12 Jun 2005 20:39:08 -0800] rev 320
hg push: propagate return code -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hg push: propagate return code manifest hash: 3c4c5a0eba4c2eb431e4830fafe3138c12f9bddf -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrQ3sywK+sNU5EO8RAkGpAJ40HsYprOWEZzmqw47ZYvZukMzKvgCeIGfD H6gB6idJV4K2SpEaFUByRYE= =LV1Z -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 19:35:41 -0800 stopgap hg push support
mpm@selenic.com [Sun, 12 Jun 2005 19:35:41 -0800] rev 319
stopgap hg push support -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 stopgap hg push support This does hg push by running hg serve and tunnelling it over ssh to a client at the destination doing hg pull. manifest hash: 552607f1e8bea238b4c585209af93f0594121dd1 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrP8NywK+sNU5EO8RAmGJAKCJ9kQSz6Bs/4cYCvKU7TpDXPHSbgCdEE1X FfZe2y2xtgSeHQa/uZLqEBQ= =KlPp -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 10:20:07 -0800 mdiff: revert grouping optimization for the time being
mpm@selenic.com [Sun, 12 Jun 2005 10:20:07 -0800] rev 318
mdiff: revert grouping optimization for the time being -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 mdiff: revert grouping optimization for the time being This had trouble with Ted T'so import test while the original didn't. manifest hash: e2fc49b5277096bd4c5081558af5efe9964d5310 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrHzXywK+sNU5EO8RArocAJwKlxrnyVpdYaKzgJG/b4gSVOYBTwCgkl2t zD807fsMULRDdDe1k9jVPcU= =Iivz -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 10:18:32 -0800 hg commit: user and date options
mpm@selenic.com [Sun, 12 Jun 2005 10:18:32 -0800] rev 317
hg commit: user and date options -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hg commit: user and date options - From Ted Ts'o manifest hash: f6ce29eed387664c6f0631b666158eb930876cb3 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrHx4ywK+sNU5EO8RAp/uAKCEGhvm1tpBJ9Bub8BlalWaIpMeqgCfUejo IoevMbk4cjxQ5P3Zo2pSFs8= =Ibp3 -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 08:51:11 -0800 Add new convert-repo script
mpm@selenic.com [Sun, 12 Jun 2005 08:51:11 -0800] rev 316
Add new convert-repo script -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Add new convert-repo script This is the beginnings of a generalized framework for converting repositories. Currently hardwired to convert from git to hg. manifest hash: dc3b72de2c45bfdaffcc1cf71da530228793facd -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrGf/ywK+sNU5EO8RAi9EAJ0eQ++cwSgn5j2PHiTvF7r3JNiv4gCePY+X do12pUvCczyBKVCoBN7y/uI= =YtzI -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 08:49:49 -0800 Remove the old convert-repo script for .4-era hg repositories
mpm@selenic.com [Sun, 12 Jun 2005 08:49:49 -0800] rev 315
Remove the old convert-repo script for .4-era hg repositories -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Remove the old convert-repo script for .4-era hg repositories manifest hash: f203e562f8e66750b7abe5dd5a3040bae7b0951b -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrGetywK+sNU5EO8RAslSAJ9Eav+ufW7jjtP4k79waEXxUQVUMgCgivZ/ 053y3CQhBPFCe/TaqViVEfE= =jw59 -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 08:48:01 -0800 More tweaking to rawcommit for repo conversion
mpm@selenic.com [Sun, 12 Jun 2005 08:48:01 -0800] rev 314
More tweaking to rawcommit for repo conversion -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 More tweaking to rawcommit for repo conversion manifest hash: 912e7d61fc15845fd24baada9f553c2d38b26f06 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrGdBywK+sNU5EO8RApbrAKCPVvxnSgkzAWpwc6uRDCCYAsqO7QCeMonn tZS4URbMI2aDQYw15wDEZHg= =C2Vn -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 08:23:29 -0800 Minor caching improvement for manifest
mpm@selenic.com [Sun, 12 Jun 2005 08:23:29 -0800] rev 313
Minor caching improvement for manifest -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Minor caching improvement for manifest manifest hash: 91d827f4041f1151875a52199618cdd1a219ac79 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrGGBywK+sNU5EO8RAotVAJ93JTk6XoHnZrbGzcZQ1N+P7kNKiACfYj+F RlcmcLHc3rEMkcM127BszDg= =5zAN -----END PGP SIGNATURE-----
Sun, 12 Jun 2005 08:11:43 -0800 Optimize diff and status in subdirectories
mpm@selenic.com [Sun, 12 Jun 2005 08:11:43 -0800] rev 312
Optimize diff and status in subdirectories -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Optimize diff and status in subdirectories This walks the tree from the subdirectory down. Based on a patch by Arun Sharma manifest hash: 3a8545369b4eb6911b9d0fcdd8e4dc77784fa3d8 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCrF6/ywK+sNU5EO8RAosUAKC0KWcMfsLErEInDk1X1vpM54HFEgCfUwcN gXhaAviQZJ8h0ssl3BMmMfI= =KnRD -----END PGP SIGNATURE-----
Sat, 11 Jun 2005 10:10:06 -0800 fix KeyErrors from reading empty dirstate
mpm@selenic.com [Sat, 11 Jun 2005 10:10:06 -0800] rev 311
fix KeyErrors from reading empty dirstate -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 fix KeyErrors from reading empty dirstate manifest hash: 04a797f1b53e34153c8ea18950a5dd0eefb4694d -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD4DBQFCqyj+ywK+sNU5EO8RAoepAJYp2TjjaMq3l+ZrRfogyvUGCebgAJ9nKPdr nP/Y3F1GvOle58NTIQ6UBA== =0k3O -----END PGP SIGNATURE-----
Fri, 10 Jun 2005 14:10:07 -0800 hg patch: add default commit text
mpm@selenic.com [Fri, 10 Jun 2005 14:10:07 -0800] rev 310
hg patch: add default commit text -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hg patch: add default commit text I was doing some benchmarks via hg import and found a few patches where the commit string ended up empty. hg would then try to prompt you for the commit string. The patch below forces a default string based on the patch name. Signed-off-by: Chris Mason <mason@suse.com> Modified to only be added if there's no commit text - mpm manifest hash: 35da46e3832a98e2d288604e6be8c82e85405ef8 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCqg+/ywK+sNU5EO8RAqCIAJ4/Nj6c3MusY3HyD8HeHjqc+GlFMQCbBWA5 HhCmOoJuL5Uoz7WlGPlpumw= =M3/j -----END PGP SIGNATURE-----
Fri, 10 Jun 2005 13:48:08 -0800 add easy profiling support
mpm@selenic.com [Fri, 10 Jun 2005 13:48:08 -0800] rev 309
add easy profiling support -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 add easy profiling support Now you can do hg -p [command] and it will run it under the hotshot profiler manifest hash: fcad346798243ad2434bc5458ed8d3456707c599 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCqgqYywK+sNU5EO8RAp9BAJsFI7iUWYEMvXWYaVt4VmrlOF3enwCdEFRJ l/EZmMLecbfQcbbh5oo0yHs= =voa0 -----END PGP SIGNATURE-----
Fri, 10 Jun 2005 12:49:25 -0800 Add tip to tags
mpm@selenic.com [Fri, 10 Jun 2005 12:49:25 -0800] rev 308
Add tip to tags -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Add tip to tags This makes a tag for the current tip that always overrides what's in the tags files. manifest hash: b1a17d0238aa84ee30773be79e4a32dca321a0c3 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCqfzVywK+sNU5EO8RAgdEAJ41VvLHIIzNZbIC/nHmoJOpZL0+4gCfRxpC eSq341nfmMZNcrqmGYYALe4= =miH1 -----END PGP SIGNATURE-----
Fri, 10 Jun 2005 12:48:12 -0800 add tip to repo.lookup
mpm@selenic.com [Fri, 10 Jun 2005 12:48:12 -0800] rev 307
add tip to repo.lookup
Fri, 10 Jun 2005 09:36:38 -0800 Merge with TAH
mpm@selenic.com [Fri, 10 Jun 2005 09:36:38 -0800] rev 306
Merge with TAH -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Merge with TAH manifest hash: 9ce3631dbbe2070e6e0b8a72e197790c911896e2 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCqc+mywK+sNU5EO8RAshrAJ4rqQXCJqB1rqgDQ5ujwOhFqqRuuQCdEUsK wpBCgWEm3xr9mxQ6rLLsihU= =Ozg9 -----END PGP SIGNATURE-----
Fri, 10 Jun 2005 09:36:02 -0800 more merge fixes
mpm@selenic.com [Fri, 10 Jun 2005 09:36:02 -0800] rev 305
more merge fixes -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 more merge fixes Logic for getting a remote change should work going backwards and forwards Mark all the changes in the merge that we don't get or merge so we have the right list of files in the changeset manifest hash: 88be742f510a2c58b276172ae538f9533f5e7491 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCqc+CywK+sNU5EO8RAjOKAJwNkC7phUO77IGW20R7MKyWBGP01gCgjxP6 K2CqUESHr2+0J96q3N74Siw= =zb2f -----END PGP SIGNATURE-----
Fri, 10 Jun 2005 07:52:39 +0100 Use vi if $EDITOR is unset.
Thomas Arendsen Hein <thomas@intevation.de> [Fri, 10 Jun 2005 07:52:39 +0100] rev 304
Use vi if $EDITOR is unset.
Fri, 10 Jun 2005 07:50:20 +0100 Check if $DISPLAY is set before using tkdiff or kdiff3.
Thomas Arendsen Hein <thomas@intevation.de> [Fri, 10 Jun 2005 07:50:20 +0100] rev 303
Check if $DISPLAY is set before using tkdiff or kdiff3.
Fri, 10 Jun 2005 00:29:36 -0800 various fixups for git import
mpm@selenic.com [Fri, 10 Jun 2005 00:29:36 -0800] rev 302
various fixups for git import -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 various fixups for git import cache manifest flags properly after add rawcommit handle flags handle multiple parents better commit a complete manifest verify changesets and manifests are no longer necessarily 1:1 check for duplicate nodes manifest hash: 5bf65f506d2b0da94d42d9fb1f792d58f6453c58 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCqU9wywK+sNU5EO8RApUUAJ9sXVIKMcmLChdk2M95U6DW37qNNQCgid08 /1E9Y4vw25HcxwJlXopV1lU= =oohz -----END PGP SIGNATURE-----
Fri, 10 Jun 2005 00:26:29 -0800 revlog: allow duplicates
mpm@selenic.com [Fri, 10 Jun 2005 00:26:29 -0800] rev 301
revlog: allow duplicates -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 revlog: allow duplicates If two branches make the same change to the same parent, the result will be an identical hash. Git apparently does this all the time. Deal with it gracefully. manifest hash: c6217eab4b310e1ae529dd75ab90e717dbe5d55d -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCqU61ywK+sNU5EO8RAkFqAJ9KhWUQgjZbzzB/+mTkolH0GkT1awCfa+Mj ulbI4xCRZcvfQE492mcNwQA= =N6In -----END PGP SIGNATURE-----
Fri, 10 Jun 2005 00:25:04 -0800 hg init: when hardlinking, remove dirstate
mpm@selenic.com [Fri, 10 Jun 2005 00:25:04 -0800] rev 300
hg init: when hardlinking, remove dirstate -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hg init: when hardlinking, remove dirstate manifest hash: 48d32cdfe7eacf17b012226a884384e48377b0d8 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFCqU5gywK+sNU5EO8RAmzxAJ4hQUt/qrIcPzuGHf5dazJkKc9wgwCgpqV/ zSC4SHAvJQ2VptJSSVCEOAg= =QVG+ -----END PGP SIGNATURE-----
Thu, 09 Jun 2005 09:48:50 -0800 Merge with TAH
mpm@selenic.com [Thu, 09 Jun 2005 09:48:50 -0800] rev 299
Merge with TAH -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Merge with TAH manifest hash: 5a7d047acc6df6ba03db532657f54c0d38e4cffe -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD4DBQFCqIECywK+sNU5EO8RAg56AJ95hB6uw7Fc55XgwgZls4UUNRt1jACYrkrq V9VJhhv29Fd/LEZy+ydoGQ== =unEK -----END PGP SIGNATURE-----
(0) -300 -100 -50 -30 +30 +50 +100 +300 +1000 +3000 +10000 +30000 tip