gmane.comp.version-control.mercurial.devel

[PATCH] import: add similarity option (issue295)

Brendan Cully — 2008-11-22T08:22:48

# HG changeset patch # User Brendan Cully kublai.com> # Date 1227342117 28800 # Node ID c227bbf29d91a07d2572769747320cd7f7b8d384 # Parent 03b60f2f90bf14594ffd733cace1102c5c10499b import: add similarity option (issue295) diff --git a/mercurial/commands.py b/mercurial/commands.py --- a/mercurial/commands.py +++ b/mercurial/commands.py < at >< at > -1581,6 +1581,9 < at >< at > recorded in the patch. This may happen due to character set problems or other deficiencies in the text patch format. + With --similarity, hg will attempt to discover renames and copies + in the patch in the same way as 'addremove'. + To read a patch from standard input, use patch name "-". See 'hg help dates' for a list of formats valid for -d/--date. """ < at >< at > -1590,6 +1593,13 < at >< at > if date: opts['date'] = util.parsedate(date) + try: + sim = float(opts.get('similarity') or 0) + except ValueError: + raise util.Abort(_('similarity must be a number')) + if sim < 0 or sim > 100: +

[PATCH 0 of 1] --graph option for in/out/log (was: gin/gout ...)

Alpar Juttner — 2008-11-21T22:28:59

Hi, This patch implements a graphical version of incoming anf outgoing commands, namely, it adds a --graph option to the in, out and log commands. The gincoming and goutgoing commands are not defined now. Best regards, Alpar

mercurial/crew< at >7391: 5 outgoing changesets

Mercurial Commits — 2008-11-21T12:00:05

5 outgoing changesets in mercurial/crew: http://hg.intevation.org/mercurial/crew/rev/92c952c4470c changeset: 7391:92c952c4470c tag: tip user: Brendan Cully kublai.com> date: Wed Nov 19 13:00:48 2008 -0800 summary: inotify: fix status . in repo.root http://hg.intevation.org/mercurial/crew/rev/564326a6ef9c changeset: 7390:564326a6ef9c user: Patrick Mezard gmail.com> date: Wed Nov 19 13:27:57 2008 +0100 summary: patch: isolate patchfile filesystem calls into methods http://hg.intevation.org/mercurial/crew/rev/27d304c8cc03 changeset: 7389:27d304c8cc03 user: Patrick Mezard gmail.com> date: Wed Nov 19 13:27:57 2008 +0100 summary: patch: pass an opener to patchfile http://hg.intevation.org/mercurial/crew/rev/0d1c770c6be1 changeset: 7388:0d1c770c6be1 user: Patrick Mezard gmail.com> date: Tue Nov 18 13:20:55 2008 +0100 summary: util_win32: implement posixfile_nt.readlines() http://hg.inteva

[PATCH 0 of 3] gin/out commands for graphlog extension

Alpar Juttner — 2008-11-21T09:57:24

Hi, The first two patches implement the graphlog counterpart of the standard 'incoming' and 'outgoing' commands. They understand the same options except for --newest-first. The third patch wraps the original log, incoming and outgoing commands and extends them with a --graph option. It also checks whether the other given options are compatible with the --graph. I also kept the glog/gincoming/goutgoing commands because - In this way 'help graphlog' will display them - Users may want to use different default options (e.g. --style) for the graphical and the standard version. It is easier to configure if the command name is different.

[PATCH] hgweb (coal/paper): make file, diff and annotate render faster

Brendan Cully — 2008-11-21T02:22:34

Here's a patch that reworks the HTML/CSS for paper and coal to make file, diff and annotate faster. For file and diff the tables are removed entirely, but for annotate I had to keep them to get things to align properly against the user< at >rev column. Consequently annotate's speed increase is not nearly as dramatic. If we were willing to trim/pad that info to a fixed size (maybe preserving full info in a mouseover) we could really improve annotate speed. Here's the difference in speed I get on a 100,000 line file, measured in hg from start to finish page serving (actual render time is several seconds longer for the large numbers): unpatched: file: 120 seconds annotate = 308 seconds rev = 120 seconds patched: file = 28 seconds annotate = 220 seconds rev = 23 seconds # HG changeset patch # User Brendan Cully kublai.com> # Date 1227233750 28800 # Node ID d7f367b771be3f85a23bbf257aed1672e97ff3af # Parent b425a0b391dd596659d9627b6f80e20d2a79cbb5 hgweb (coal/paper): make file, diff and annotate easier to r

[PATCH 0 of 2] gout/gin commands for graphlog extension

Alpar Juttner — 2008-11-20T21:54:07

Hi, These two patches implement the graphlog counterpart of the standard 'incoming' and 'outgoing' commands. They understand the same options except for --newest-first. The first patch is almost the same as I sent before, I just removed the unused --newest-first option. Best regards, Alpar

mercurial/crew< at >7391: 5 outgoing changesets

Mercurial Commits — 2008-11-20T12:00:05

[PATCH 0 of 2] templater support for filter arguments

Brendan Cully — 2008-11-19T23:28:09

Hi all, The following patches add templater support for filter arguments. For example, instead of writing {node|short} you could write {node|shorten(12)}. I've included new filters "fill", "datefmt", "head", and "shorten" as a demonstration. "datefmt" lets you write {date|datefmt("%Y-%m-%d %H:%M")} instead of {date|isodate}, and head lets you select the number of lines of text you want to keep, so firstline is equivalent to head(1) as in {desc|head(1)}. You can also put filters in front of % expandos for some interesting new options. I've included a new hgpatch style demonstrating this (perhaps in a silly way) -- it filters out revisions from the parents list if they are less than 0. This is post-1.1 stuff, but might be worth reviewing. It passes the test suite.

[issue1391] hg pbchk throws exception

Susan Scheufele — 2008-11-19T23:07:10

New submission from Susan Scheufele sun.com>: zday% hg pbchk Enter passphrase for key '/home/susans/.ssh/id_dsa': CDDL block check: Copyright check: C style check: Header format check: Java style check: File permission check: Keywords check: ** unknown exception encountered, details follow ** report bug details to http://www.selenic.com/mercurial/bts ** or mercurial< at >selenic.com ** Mercurial Distributed SCM (version 1.0.2) Traceback (most recent call last): File "/usr/bin/hg", line 20, in ? mercurial.dispatch.run() File "build/proto/lib/python/mercurial/dispatch.py", line 20, in run File "build/proto/lib/python/mercurial/dispatch.py", line 29, in dispatch File "build/proto/lib/python/mercurial/dispatch.py", line 45, in _runcatch File "build/proto/lib/python/mercurial/dispatch.py", line 364, in _dispatch File "build/proto/lib/python/mercurial/dispatch.py", line 417, in _runcommand File "build/proto/lib/python/mercurial/dispatch.py", line 373, in checkargs File "bu

gout command for graphlog extension

Alpár Jüttner — 2008-11-19T17:57:28

Hi, Please find small extension to graphlog.py in the attachment. It provides a gout/goutgoing command, which is just the graphical counterpart of the standard out/outgoing. They eat exactly the same options. I think it is useful enough to get into the hg-crew. I used a "trial and error" approach when I wrote it, so I would really appreciate any reviews and comments. Best regards, Alpar

gout command for graphlog extension

Alpár Jüttner — 2008-11-19T18:10:00

[PATCH] Add --keepbranches option to rebase

Augie Fackler — 2008-11-19T04:03:10

hg-stable< at >6558: new changeset

Mercurial Commits — 2008-11-18T22:16:28

New changeset in hg-stable: http://www.selenic.com/repo/hg-stable/rev/b965605dfb2e changeset: 6558:b965605dfb2e tag: tip user: Patrick Mezard gmail.com> date: Sat Nov 15 15:51:26 2008 +0100 summary: convert: be even more tolerant when detecting svn tags

mercurial< at >7388: 6 new changesets

Mercurial Commits — 2008-11-18T22:15:47

6 new changesets in mercurial: http://www.selenic.com/hg/rev/b501c7f3c2ad changeset: 7383:b501c7f3c2ad user: Peter Arrenbrecht gmail.com> date: Sun Nov 16 20:55:30 2008 +0100 summary: graphlog: fix regression with filelogs introduced by 7bc62ebe7693 http://www.selenic.com/hg/rev/62b75d73c33d changeset: 7384:62b75d73c33d user: Brendan Cully kublai.com> date: Sun Nov 16 19:49:36 2008 -0800 summary: zsh: support for resolve http://www.selenic.com/hg/rev/8119c1a607c4 changeset: 7385:8119c1a607c4 user: Brendan Cully kublai.com> date: Sun Nov 16 20:56:55 2008 -0800 summary: mergetools: add pointer to the usual filemerge path http://www.selenic.com/hg/rev/86bfd65532ac changeset: 7386:86bfd65532ac user: Brendan Cully kublai.com> date: Mon Nov 17 11:45:38 2008 -0800 summary: zsh: make resolve with no options only complete unresolved files http://www.selenic.com/hg/rev/7e9a15fa6c8f changes

mercurial/crew-stable< at >6558: outgoing changeset

Mercurial Commits — 2008-11-18T12:00:04

Outgoing changeset in mercurial/crew-stable: http://hg.intevation.org/mercurial/crew-stable/rev/b965605dfb2e changeset: 6558:b965605dfb2e tag: tip user: Patrick Mezard gmail.com> date: Sat Nov 15 15:51:26 2008 +0100 summary: convert: be even more tolerant when detecting svn tags

mercurial/crew< at >7386: 4 outgoing changesets

Mercurial Commits — 2008-11-18T12:00:05

4 outgoing changesets in mercurial/crew: http://hg.intevation.org/mercurial/crew/rev/86bfd65532ac changeset: 7386:86bfd65532ac tag: tip user: Brendan Cully kublai.com> date: Mon Nov 17 11:45:38 2008 -0800 summary: zsh: make resolve with no options only complete unresolved files http://hg.intevation.org/mercurial/crew/rev/8119c1a607c4 changeset: 7385:8119c1a607c4 user: Brendan Cully kublai.com> date: Sun Nov 16 20:56:55 2008 -0800 summary: mergetools: add pointer to the usual filemerge path http://hg.intevation.org/mercurial/crew/rev/62b75d73c33d changeset: 7384:62b75d73c33d user: Brendan Cully kublai.com> date: Sun Nov 16 19:49:36 2008 -0800 summary: zsh: support for resolve http://hg.intevation.org/mercurial/crew/rev/b501c7f3c2ad changeset: 7383:b501c7f3c2ad user: Peter Arrenbrecht gmail.com> date: Sun Nov 16 20:55:30 2008 +0100 summary: graphlog: fix regression with file

[issue1390] clone throws an exception (Mac OS X 10.5 / G4 / versionb801d6e5dc83+20081020)

dkoerner — 2008-11-17T21:35:22

New submission from dkoerner googlemail.com>: After installing mercurial from here (http://mercurial.berkwood.com/ --> Version 1.0.2 + fixes through 2008-10-19) on a G4 iBook running OS X 10.5.5 I get an reproducable exception calling clone(see below). The new folder gets created but when I call hg status in it I get an "abort: incompatible revision flag 1!" error message. Same is for changing files and the commiting them. updating working directory ** unknown exception encountered, details follow ** report bug details to http://www.selenic.com/mercurial/bts ** or mercurial< at >selenic.com ** Mercurial Distributed SCM (version b801d6e5dc83+20081020) ** Extensions loaded: Traceback (most recent call last): File "/usr/local/bin/hg", line 20, in mercurial.dispatch.run() File "/Library/Python/2.5/site-packages/mercurial/dispatch.py", line 20, in run sys.exit(dispatch(sys.argv[1:])) File "/Library/Python/2.5/site-packages/mercurial/dispatch.py", line 29, in dispatch r

[issue1389] Inefficient push to repository with multiple branches

Brendan Cully — 2008-11-17T21:31:49

New submission from Brendan Cully kublai.com>: Using recent crew I've experienced a significant performance regression pushing to my mutt repository over ssh (remote hg version 1.0.1). Here's hg push --debug --lsprof of a one-line change being pushed: running ssh mutt< at >dev.mutt.org "hg -R mutt serve --stdio" sending hello command sending between command remote: 48 remote: capabilities: unbundle lookup changegroupsubset remote: 1 pushing to ssh://mutt< at >dev.mutt.org/mutt sending heads command searching for changes common changesets up to 9c9fbf98fbda 90944f375844 1da8b126c870 982532ae8410 5b142858393a 2946ce6c56c9 6934de2f9f48 a60586461eb8 034a6daeebcf 659 changesets found List of changesets: 5eaeb42f4f6988d974cd9bcb299ce1fd3bb891f3 5b142858393abe6588c2205c66951a173e274ef6 a7c3fbadd7d83c6af580c9994837143c40d1f426 2d79dcdd9812b7350b9892e48899df01a2a07392 66c23252011510035b7d58e04d47bbcb9e3014c3 a7079190b7f2efdd003abc19729309909d5336d1 ed686fe85e06c16b38e504452201c26452cff144 1405e31d20fc683454f9608f64

[PATCH] graphlog: add --only-branch and --except-branch options toglog

Peter Arrenbrecht — 2008-11-17T16:27:35

# HG changeset patch # User Peter Arrenbrecht gmail.com> # Date 1226938976 -3600 # Node ID ab6a3d686136261beb4623583ddce5edb9b0edd0 # Parent de304148dfd96f03226ff201819a2a5bbe072121 graphlog: add --only-branch and --except-branch options to glog diff --git a/hgext/graphlog.py b/hgext/graphlog.py --- a/hgext/graphlog.py +++ b/hgext/graphlog.py < at >< at > -14,7 +14,10 < at >< at > from mercurial.node import nullrev from mercurial.util import Abort, canonpath -def revisions(repo, start, stop): +def nofilter(ctx): + return True + +def revisions(repo, start, stop, filterctxfn=nofilter): """cset DAG generator yielding (rev, node, [parents]) tuples This generator function walks through the revision history from revision < at >< at > -24,12 +27,14 < at >< at > cur = start while cur >= stop: ctx = repo[cur] - parents = [p.rev() for p in ctx.parents() if p.rev() != nullrev] - parents.sort() - yield (ctx, parents) + if filterctxfn(ctx): + parents = [p.rev() fo

[Fwd: Re: HTTP 500 when using hooks with hgwebdir.cgi in 1.0.2]

Jesse Long — 2008-11-17T14:39:34

Dear Mercurial Developers, Please include the patch below in the next release. Basically, the hook method fails to restore io correctly if either _pythonhook() or _exthook() throw and exception. finally fixes it. Thanks, Jesse -------- Original Message -------- Subject: Re: HTTP 500 when using hooks with hgwebdir.cgi in 1.0.2 Date: Mon, 17 Nov 2008 16:13:40 +0200 From: Jesse Long unknown.za.net> To: mercurial< at >selenic.com References: <49216F64.5090302< at >unknown.za.net> Jesse Long wrote: Well, it looks like I'll be that kind person. See patch below. Please note, this is the first python code I've written in my life. Thanks, Jesse diff -Naur mercurial-1.0.2.orig/mercurial/hook.py mercurial-1.0.2/mercurial/hook.py --- mercurial-1.0.2.orig/mercurial/hook.py2008-08-14 00:11:47.000000000 +0200 +++ mercurial-1.0.2/mercurial/hook.py2008-11-17 16:11:09.000000000 +0200 < at >< at > -96,20 +96,21 < at >< at > oldstdout = os.dup(sys.__stdout__.fileno()) os.dup2(sys.__stderr__.fileno(), sys.__stdout__.fi

[PATCH 0 of 3] 'incoming' for heads

Peter Arrenbrecht — 2008-11-17T14:34:34

Fixed distribution of changes to patches.