gmane.comp.version-control.mercurial.devel

[issue1405] "unknown exception encountered" when qfolding multiplepatches together,where later patch removes file edited by earlier patch

Daniel Holbert — 2008-12-02T02:31:15

New submission from Daniel Holbert mozilla.com>: STEPS TO REPRODUCE: ------------------- Just run the commands in the attached "qfold_broken.sh" script The commands basically just do this: a. Clone http://hg.mozilla.org/mozilla-central b. Clone my patch queue http://hg.mozilla.org/users/dholbert_mozilla.com/smil-patches c. "hg qnew" a new patch, and try to fold the first 4 patches (up to "smil_secondFixes") into it. ACTUAL RESULTS: --------------- The "qfold" command fails with an "unknown exception encountered" error message, and it leaves my source tree in an unclean state, with 'hg qpop' and 'hg status' both failing. (details to follow) NOTE: If, instead of qfolding the 4 patches all at once, I qfold the first 3 and *then* separately qfold the 4th one, then everything works fine. (I'll attach a separate script demonstrating that.) I think this might happen because the 4th patch deletes files which are created in the 1st patch (and which are modified in the 2nd and 3rd patches). So, I'm guessing the problem is that Mercurial has a hard time when it needs to create & delete a file in a single qfold command. Mercurial version: 1.0.1 OS Version: Ubuntu 8.10 (up to date) ---------- files: qfold_broken.sh messages: 8038 nosy: dholbert priority: urgent status: unread title: "unknown exception encountered" when qfolding multiple patches together, where later patch removes file edited by earlier patch ____________________________________________________ Mercurial issue tracker selenic.com> ____________________________________________________

Re: [PATCH 3 of 3] cmdutil: speed up walkchangerevs() when bothfollowing and pruning

Matt Mackall — 2008-12-01T20:13:33

Just a note that contrib/perf.py is very useful for this sort of benchmarking.

mercurial< at >7450: 24 new changesets

Mercurial Commits — 2008-12-01T18:15:52

24 new changesets in mercurial: http://www.selenic.com/hg/rev/f21e3d0e335b changeset: 7427:f21e3d0e335b parent: 7421:4c4324476be6 user: Patrick Mezard gmail.com> date: Wed Nov 26 22:27:37 2008 +0100 summary: util_win32: fix Windows version checking (issue1358) http://www.selenic.com/hg/rev/fdcde929ce4f changeset: 7428:fdcde929ce4f user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:30 2008 +0100 summary: tests: use killdaemons in hgweb tests http://www.selenic.com/hg/rev/dbc40381620e changeset: 7429:dbc40381620e user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:31 2008 +0100 summary: tests: Skip tests if they will fail because of outer repo http://www.selenic.com/hg/rev/f0a3e87c810d changeset: 7430:f0a3e87c810d user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:32 2008 +0100 summary: util: use existing never() instead of custom lambda http://www.selenic.com/hg/rev/3d827cc616b6 changeset: 7431:3d827cc616b6 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 01:09:11 2008 +0100 summary: buildrpm: complain when hg command isn't available http://www.selenic.com/hg/rev/642754e776e2 changeset: 7432:642754e776e2 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 01:10:27 2008 +0100 summary: rpm spec: include python egg in rpm http://www.selenic.com/hg/rev/c4ce828e8074 changeset: 7433:c4ce828e8074 parent: 7426:df0962f6c54e parent: 7432:642754e776e2 user: Benoit Boissinot ens-lyon.org> date: Thu Nov 27 01:35:12 2008 +0100 summary: merge with mpm http://www.selenic.com/hg/rev/cf7741aa1e96 changeset: 7434:cf7741aa1e96 user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 16:07:13 2008 +0100 summary: kill some trailing spaces http://www.selenic.com/hg/rev/5e13df32fb74 changeset: 7435:5e13df32fb74 user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 16:07:17 2008 +0100 summary: bundlerepo doesn't really have a dirstate, throw AttributeError if requested http://www.selenic.com/hg/rev/07faba78cf5a changeset: 7436:07faba78cf5a user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 17:00:54 2008 +0100 summary: diff: fix obscure off-by-one error in diff -p http://www.selenic.com/hg/rev/3cdaac732b2b changeset: 7437:3cdaac732b2b user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 18:46:45 2008 +0100 summary: tests: fix test results for 07faba78cf5a http://www.selenic.com/hg/rev/b80b944836ec changeset: 7438:b80b944836ec user: Brendan Cully kublai.com> date: Thu Nov 27 11:04:29 2008 -0800 summary: Add hgcia to hgext http://www.selenic.com/hg/rev/7dd44ad43914 changeset: 7439:7dd44ad43914 user: Brendan Cully kublai.com> date: Thu Nov 27 11:05:50 2008 -0800 summary: Ignore eclipse droppings http://www.selenic.com/hg/rev/d2a917b27152 changeset: 7440:d2a917b27152 user: Brendan Cully kublai.com> date: Thu Nov 27 11:14:43 2008 -0800 summary: Fix test breakage introduced in 07faba78cf5a http://www.selenic.com/hg/rev/4fecd17f2de9 changeset: 7441:4fecd17f2de9 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 22:04:51 2008 +0100 summary: convert cvs: Fix branch name parsing http://www.selenic.com/hg/rev/a14ce129cfcd changeset: 7442:a14ce129cfcd user: Edouard Gomez free.fr> date: Thu Nov 27 10:13:20 2008 +0100 summary: convert: check existence of ~/.cvspass before reading it http://www.selenic.com/hg/rev/1e0677756f60 changeset: 7443:1e0677756f60 user: Dirkjan Ochtman ochtman.nl> date: Fri Nov 28 09:50:30 2008 +0100 summary: gitweb: fix graph display in IE http://www.selenic.com/hg/rev/f792c7bb2fb3 changeset: 7444:f792c7bb2fb3 user: Thomas Arendsen Hein intevation.de> date: Fri Nov 28 11:38:41 2008 +0100 summary: Improvement to 14ce129cfcd: Use try/except and pass filename on errors http://www.selenic.com/hg/rev/29c6e71b1c73 changeset: 7445:29c6e71b1c73 user: Dirkjan Ochtman ochtman.nl> date: Sun Nov 30 18:38:42 2008 +0100 summary: gitweb: fix graph again after previous faulty commit (1e0677756f60) http://www.selenic.com/hg/rev/2d2e0be762be changeset: 7446:2d2e0be762be user: Patrick Mezard gmail.com> date: Sun Nov 30 22:01:26 2008 +0100 summary: convert: work around svn.ra.get_files() not releasing input buffer http://www.selenic.com/hg/rev/4fd92687f331 changeset: 7447:4fd92687f331 user: Patrick Mezard gmail.com> date: Sun Nov 30 22:51:42 2008 +0100 summary: convert: improve reporting of invalid svn bindings http://www.selenic.com/hg/rev/7900d240c3d8 changeset: 7448:7900d240c3d8 user: Thomas Arendsen Hein intevation.de> date: Mon Dec 01 12:11:28 2008 +0100 summary: Fix non-empty $CDPATH causing failed tests. http://www.selenic.com/hg/rev/f848d7f96195 changeset: 7449:f848d7f96195 user: Georg Brandl python.org> date: Mon Dec 01 13:44:34 2008 +0100 summary: help: try topics before commands to avoid "diffs" showing help for diffstat http://www.selenic.com/hg/rev/79d1bb737c16 changeset: 7450:79d1bb737c16 tag: tip user: Patrick Mezard gmail.com> date: Mon Dec 01 14:20:20 2008 +0100 summary: hgweb: extend [paths] syntax to match repositories recursively (issue852)

[PATCH] notify: no charset conversion when testing

Christian Ebert — 2008-12-01T14:33:56

# HG changeset patch # User Christian Ebert gmx.net> # Date 1228141979 -3600 # Node ID a65f638e219cec49b0a1bc1e5516caf3ecbc670b # Parent 79d1bb737c16b343e33be7cd7e22dcc4e7bfab0b notify: no charset conversion when testing Same as for patchbomb. diff --git a/hgext/notify.py b/hgext/notify.py --- a/hgext/notify.py +++ b/hgext/notify.py < at >< at > -157,7 +157,9 < at >< at > for user in users.split(','): subs[self.fixmail(user)] = 1 subs = util.sort(subs) - return [mail.addressencode(self.ui, s, self.charsets) for s in subs] + show = self.ui.configbool('notify', 'test', True) + return [mail.addressencode(self.ui, s, self.charsets, show) + for s in subs] def url(self, path=None): return self.ui.config('web', 'baseurl') + (path or self.root) < at >< at > -186,7 +188,8 < at >< at > # create fresh mime message from msg body text = msg.get_payload() # for notification prefer readability over data precision - msg = mail.mimeencode(self.ui, text, self.charsets) + show = self.ui.configbool('notify', 'test', True) + msg = mail.mimeencode(self.ui, text, self.charsets, show) def fix_subject(subject): '''try to make subject line exist and be useful.''' < at >< at > -201,7 +204,8 < at >< at > maxsubject = int(self.ui.config('notify', 'maxsubject', 67)) if maxsubject and len(subject) > maxsubject: subject = subject[:maxsubject-3] + '...' - msg['Subject'] = mail.headencode(self.ui, subject, self.charsets) + msg['Subject'] = mail.headencode(self.ui, subject, + self.charsets, show) def fix_sender(sender): '''try to make message have proper sender.''' < at >< at > -210,7 +214,8 < at >< at > sender = self.ui.config('email', 'from') or self.ui.username() if '< at >' not in sender or '< at >localhost' in sender: sender = self.fixmail(sender) - msg['From'] = mail.addressencode(self.ui, sender, self.charsets) + msg['From'] = mail.addressencode(self.ui, sender, + self.charsets, show) msg['Date'] = util.datestr(format="%a, %d %b %Y %H:%M:%S %1%2") fix_subject(subject) < at >< at > -224,7 +229,7 < at >< at > msg['To'] = ', '.join(self.subs) msgtext = msg.as_string(0) - if self.ui.configbool('notify', 'test', True): + if show: self.ui.write(msgtext) if not msgtext.endswith('\n'): self.ui.write('\n')

[issue1404] Commit fails on a VMWare shared drive

Jukka Laurila — 2008-12-01T13:56:38

New submission from Jukka Laurila nokia.com>: Commit on a repository on a VMWare shared drive fails in certain circumstances. To replicate: X:\shared>mkdir repo X:\shared>cd repo X:\shared\repo>mkdir foo X:\shared\repo>echo foo >foo\foo.txt X:\shared\repo>hg init X:\shared\repo>hg add . adding foo\foo.txt X:\shared\repo>hg commit trouble committing foo/foo.txt! abort: The filename, directory name, or volume label syntax is incorrect: X:\shared\repo\.hg\store\data/foo/foo.txt.i X:\shared is a directory shared with an OS X Leopard host. The guest OS is WinXP SP2. Mercurial is 1.0.2, installed via the TortoiseHG distribution. To trigger the bug the top level directory must be empty. The bug does not appear if there is a file in the top level directory: X:\shared>mkdir repo X:\shared>cd repo X:\shared\repo>mkdir foo X:\shared\repo>echo foo>foo\foo.txt X:\shared\repo>echo foo>foo.txt X:\shared\repo>hg init X:\shared\repo>hg add . adding foo.txt adding foo\foo.txt X:\shared\repo>hg commit X:\shared\repo> The bug does not appear on a regular drive. ---------- messages: 8035 nosy: jlaurila priority: bug status: unread title: Commit fails on a VMWare shared drive ____________________________________________________ Mercurial issue tracker selenic.com> ____________________________________________________

Re: [PATCH 1 of 1] help: try topics before commands to avoid "diffs"showing help for diffstat

Dirkjan Ochtman — 2008-12-01T13:10:17

Nice one. Pushed to crew as f848d7f96195. Thanks, Dirkjan

mercurial/crew< at >7448: 22 outgoing changesets

Mercurial Commits — 2008-12-01T12:00:08

22 outgoing changesets in mercurial/crew: http://hg.intevation.org/mercurial/crew/rev/7900d240c3d8 changeset: 7448:7900d240c3d8 tag: tip user: Thomas Arendsen Hein intevation.de> date: Mon Dec 01 12:11:28 2008 +0100 summary: Fix non-empty $CDPATH causing failed tests. http://hg.intevation.org/mercurial/crew/rev/4fd92687f331 changeset: 7447:4fd92687f331 user: Patrick Mezard gmail.com> date: Sun Nov 30 22:51:42 2008 +0100 summary: convert: improve reporting of invalid svn bindings http://hg.intevation.org/mercurial/crew/rev/2d2e0be762be changeset: 7446:2d2e0be762be user: Patrick Mezard gmail.com> date: Sun Nov 30 22:01:26 2008 +0100 summary: convert: work around svn.ra.get_files() not releasing input buffer http://hg.intevation.org/mercurial/crew/rev/29c6e71b1c73 changeset: 7445:29c6e71b1c73 user: Dirkjan Ochtman ochtman.nl> date: Sun Nov 30 18:38:42 2008 +0100 summary: gitweb: fix graph again after previous faulty commit (1e0677756f60) http://hg.intevation.org/mercurial/crew/rev/f792c7bb2fb3 changeset: 7444:f792c7bb2fb3 user: Thomas Arendsen Hein intevation.de> date: Fri Nov 28 11:38:41 2008 +0100 summary: Improvement to 14ce129cfcd: Use try/except and pass filename on errors http://hg.intevation.org/mercurial/crew/rev/1e0677756f60 changeset: 7443:1e0677756f60 user: Dirkjan Ochtman ochtman.nl> date: Fri Nov 28 09:50:30 2008 +0100 summary: gitweb: fix graph display in IE http://hg.intevation.org/mercurial/crew/rev/a14ce129cfcd changeset: 7442:a14ce129cfcd user: Edouard Gomez free.fr> date: Thu Nov 27 10:13:20 2008 +0100 summary: convert: check existence of ~/.cvspass before reading it http://hg.intevation.org/mercurial/crew/rev/4fecd17f2de9 changeset: 7441:4fecd17f2de9 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 22:04:51 2008 +0100 summary: convert cvs: Fix branch name parsing http://hg.intevation.org/mercurial/crew/rev/d2a917b27152 changeset: 7440:d2a917b27152 user: Brendan Cully kublai.com> date: Thu Nov 27 11:14:43 2008 -0800 summary: Fix test breakage introduced in 07faba78cf5a http://hg.intevation.org/mercurial/crew/rev/7dd44ad43914 changeset: 7439:7dd44ad43914 user: Brendan Cully kublai.com> date: Thu Nov 27 11:05:50 2008 -0800 summary: Ignore eclipse droppings http://hg.intevation.org/mercurial/crew/rev/b80b944836ec changeset: 7438:b80b944836ec user: Brendan Cully kublai.com> date: Thu Nov 27 11:04:29 2008 -0800 summary: Add hgcia to hgext http://hg.intevation.org/mercurial/crew/rev/3cdaac732b2b changeset: 7437:3cdaac732b2b user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 18:46:45 2008 +0100 summary: tests: fix test results for 07faba78cf5a http://hg.intevation.org/mercurial/crew/rev/07faba78cf5a changeset: 7436:07faba78cf5a user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 17:00:54 2008 +0100 summary: diff: fix obscure off-by-one error in diff -p http://hg.intevation.org/mercurial/crew/rev/5e13df32fb74 changeset: 7435:5e13df32fb74 user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 16:07:17 2008 +0100 summary: bundlerepo doesn't really have a dirstate, throw AttributeError if requested http://hg.intevation.org/mercurial/crew/rev/cf7741aa1e96 changeset: 7434:cf7741aa1e96 user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 16:07:13 2008 +0100 summary: kill some trailing spaces http://hg.intevation.org/mercurial/crew/rev/c4ce828e8074 changeset: 7433:c4ce828e8074 parent: 7432:df0962f6c54e parent: 7426:642754e776e2 user: Benoit Boissinot ens-lyon.org> date: Thu Nov 27 01:35:12 2008 +0100 summary: merge with mpm http://hg.intevation.org/mercurial/crew/rev/642754e776e2 changeset: 7426:642754e776e2 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 01:10:27 2008 +0100 summary: rpm spec: include python egg in rpm http://hg.intevation.org/mercurial/crew/rev/3d827cc616b6 changeset: 7425:3d827cc616b6 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 01:09:11 2008 +0100 summary: buildrpm: complain when hg command isn't available http://hg.intevation.org/mercurial/crew/rev/f0a3e87c810d changeset: 7424:f0a3e87c810d user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:32 2008 +0100 summary: util: use existing never() instead of custom lambda http://hg.intevation.org/mercurial/crew/rev/dbc40381620e changeset: 7423:dbc40381620e user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:31 2008 +0100 summary: tests: Skip tests if they will fail because of outer repo http://hg.intevation.org/mercurial/crew/rev/fdcde929ce4f changeset: 7422:fdcde929ce4f user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:30 2008 +0100 summary: tests: use killdaemons in hgweb tests http://hg.intevation.org/mercurial/crew/rev/f21e3d0e335b changeset: 7421:f21e3d0e335b user: Patrick Mezard gmail.com> date: Wed Nov 26 22:27:37 2008 +0100 summary: util_win32: fix Windows version checking (issue1358)

[PATCH] hgrc 5: fix email.charsets documentation

Christian Ebert — 2008-12-01T11:53:17

# HG changeset patch # User Christian Ebert gmx.net> # Date 1228132190 -3600 # Node ID f34ab298b2e248d0a3377c68bb982873d18df4ae # Parent 7900d240c3d83d5485d998b156ec27ed50c249b5 hgrc 5: fix email.charsets documentation diff --git a/doc/hgrc.5.txt b/doc/hgrc.5.txt --- a/doc/hgrc.5.txt +++ b/doc/hgrc.5.txt < at >< at > -219,7 +219,7 < at >< at > convenient for recipients. Addresses, headers, and parts not containing patches of outgoing messages will be encoded in the first charset to which conversion from local encoding - (ui.encoding, ui.fallbackencoding) succeeds. If correct + ($HGENCODING, ui.fallbackencoding) succeeds. If correct conversion fails, the text in question is sent as is. Defaults to empty (explicit) list. < at >< at > -228,7 +228,7 < at >< at > us-ascii always first, regardless of settings email.charsets in order given by user ui.fallbackencoding if not in email.charsets - ui.encoding if not in email.charsets + $HGENCODING if not in email.charsets utf-8 always last, regardless of settings Email example:

[PATCH 1 of 1] Add a colored legend to the graph page showing howeach color in the graph corrolates to a branch

Lindsey Smith — 2008-12-01T01:31:01

# HG changeset patch # User Lindsey Smith betoptions.com> # Date 1228094835 -39600 # Node ID 1f08be66d14d377f4c4c9324f661cb0e148114fc # Parent 4fd92687f331a38ddb38d3f71753ab92cd9cb174 Add a colored legend to the graph page showing how each color in the graph corrolates to a branch. diff --git a/templates/paper/graph.tmpl b/templates/paper/graph.tmpl --- a/templates/paper/graph.tmpl +++ b/templates/paper/graph.tmpl < at >< at > -23,6 +23,8 < at >< at >

changeset

browse

diff --git a/templates/static/graph.js b/templates/static/graph.js --- a/templates/static/graph.js +++ b/templates/static/graph.js < at >< at > -50,10 +50,7 < at >< at > } } -this.setColor = function(color, bg, fg) { - -// Set the colour. -// +this.getColor = function(color, bg, fg) { // Picks a distinct colour based on an internal wheel; the bg // parameter provides the value that should be assigned to the 'zero' // colours and the fg parameter provides the multiplier that should be < at >< at > -66,7 +63,15 < at >< at > red = Math.round(red * 255); green = Math.round(green * 255); blue = Math.round(blue * 255); -var s = 'rgb(' + red + ', ' + green + ', ' + blue + ')'; +return 'rgb(' + red + ', ' + green + ', ' + blue + ')'; +} + +this.setColor = function(color, bg, fg) { + +// Set the colour. + +var s = this.getColor(color, bg, fg); + this.ctx.strokeStyle = s; this.ctx.fillStyle = s; return s; < at >< at > -78,6 +83,8 < at >< at > var backgrounds = ''; var nodedata = ''; +var branch_colors = {}; + for (var i in data) { var parity = i % 2; < at >< at > -127,11 +134,35 < at >< at > if (fold) this.columns -= 1; +// Multiple "defaults" can exist, but still be separate branches, +// so index the array based on name+color +var branch_color = this.getColor(color, 0.25, 0.75); +branch_colors[data[i][6][0].toString() + branch_color.toString()] = { +'color': branch_color, +'name': data[i][6][0] +} + } document.getElementById('nodebgs').innerHTML += backgrounds; document.getElementById('graphnodes').innerHTML += nodedata; +this.drawBranchColors(branch_colors); + +} + +this.drawBranchColors = function(colors) { + +var html = ''; + +for (var branch_name in colors) { +html = html + '

'; +// Trim the name to 10 characters so as to not overlap with +// the log display +html = html + colors[branch_name]['name'].substr(0, 10) + '

'; +} + +document.getElementById('branch_colors').innerHTML = html; } } diff --git a/templates/static/style-paper.css b/templates/static/style-paper.css --- a/templates/static/style-paper.css +++ b/templates/static/style-paper.css < at >< at > -245,3 +245,11 < at >< at > position: relative; top: -3px; } + +#branch_colors { +position: fixed; +} + +li.branch_color { +font-size: 85%; +}

[PATCH 0 of 1] Addition of a colored legend for graph page

Lindsey Smith — 2008-12-01T01:31:00

(Apologies for the first patch email, first time posting and it obviously needed some more configuring) When scrolling down large graph pages, it is easy to forget which branch is represented by each line in the graph, especially with long periods between changesets. This patch adds a simple color coded legend under the left navigation which can be referenced in such cases. The colors for each branch are simply the colors that are picked by the actual graph (the colors can repeat with a large number of branches, just as in the graph). Unnamed branches' colors are attributed to (I believe) their parents.

[PATCH] Assign unique priorities in mergetools.hgrc

Mads Kiilerich — 2008-12-01T00:45:28

# HG changeset patch # User Mads Kiilerich kiilerich.com> # Date 1228092187 -3600 # Node ID cbd6e0539c7aa15e0f837f233870eb2c5fd828b6 # Parent 4fd92687f331a38ddb38d3f71753ab92cd9cb174 Assign unique priorities in mergetools.hgrc Merge tools are chosen by priority, but in case of tools with same priority there is no well-defined order. The default is 0. This patch gives all merge tools in mergetools.hgrc a unique priority less than 0. New tools with default priority will thus get higher priority. The assigned priorities could be different. They are based on existing values and ordering and how common and usable I think they are. diff --git a/contrib/mergetools.hgrc b/contrib/mergetools.hgrc --- a/contrib/mergetools.hgrc +++ b/contrib/mergetools.hgrc < at >< at > -6,6 +6,7 < at >< at > kdiff3.regappend=\kdiff3.exe kdiff3.fixeol=True kdiff3.gui=True +kdiff3.priority=-1 gvimdiff.args=--nofork -d -g -O $local $other $base gvimdiff.regkey=Software\Vim\GVim < at >< at > -16,8 +17,10 < at >< at > merge.priority=-10 gpyfm.gui=True +gpyfm.priority=-3 meld.gui=True +meld.priority=-2 tkdiff.args=$local $other -a $base -o $output tkdiff.gui=True < at >< at > -25,27 +28,31 < at >< at > xxdiff.args=--show-merged-pane --exit-with-merge-status --title1 local --title2 base --title3 other --merged-filename $output --merge $local $base $other xxdiff.gui=True -xxdiff.priority=-8 +xxdiff.priority=-7 diffmerge.args=--nosplash --merge --title1=base --title2=local --title3=other $base $local $other diffmerge.checkchanged=True diffmerge.gui=True +diffmerge.priority=-11 p4merge.args=$base $local $other $output p4merge.regkey=Software\Perforce\Environment p4merge.regname=P4INSTROOT p4merge.regappend=\p4merge.exe p4merge.gui=True -p4merge.priority=-8 +p4merge.priority=-6 tortoisemerge.args=/base: $output /mine:$local /theirs:$other /merged:$output tortoisemerge.regkey=Software\TortoiseSVN tortoisemerge.gui=True +tortoisemerge.priority=-5 ecmerge.args=$base $local $other --mode=merge3 --title0=base --title1=local --title2=other --to=$output ecmerge.regkey=Software\Elli\xc3\xa9 Computing\Merge ecmerge.gui=True +ecmerge.priority=-12 filemerge.executable=/Developer/Applications/Utilities/FileMerge.app/Contents/MacOS/FileMerge filemerge.args=-left $other -right $local -ancestor $base -merge $output filemerge.gui=True +filemerge.priority=-4

Re: Call for testing: mercurial 1.1

TK Soh — 2008-12-01T00:30:17

No problem. Thanks for the fix :-)

Re: excanvas.js obfuscated

TK Soh — 2008-12-01T00:29:36

Ah, Explorer Canvas. Thanks for the clarification.

Re: excanvas.js obfuscated

Dirkjan Ochtman — 2008-11-30T17:43:13

It's an external package, which we don't intend to modify (it's just an IE compatibility layer for canvas), packed for small size. Kind of like coverage.py, in fact, except that size matters more. Cheers, Dirkjan

Re: [PATCH 3 of 3] cmdutil: speed up walkchangerevs() when bothfollowing and pruning

Patrick Mézard — 2008-11-30T17:25:42

Peter Arrenbrecht a écrit : I can see the improvements on your crew sample. Here are mines (real time, best of 3 runs, without and with your patches): * Firebug (http://mezard.eu/share/firebug.hg 1.5M) time hg log -fr firebug1.4:null -P performance > /dev/null 450ms / 475ms * Django (http://mezard.eu/share/django.hg 22M) time hg log -fr 615f9285682a:null -P gis > /dev/null 1.100ms / 1.130ms time hg log -fr gis:null -P 615f9285682a > /dev/null 555ms / 580ms time hg log -fr unicode:null -P 598d81dc81b8 > /dev/null 525ms / 513ms I tried to match revisions so the "outgoing" request would make sense, at least in Django tests. Perhaps, I am doing something wrong here. My problem is I saw much worse result with large repo, but I have to assess the log query made sense first. -- Patrick Mézard

Re: Information about cloning operations

Dirkjan Ochtman — 2008-11-30T14:50:00

First, I don't think you should add direct ui knowledge to those places. Instead, make them call a callback provided by a higher-up spot, which can then write to the UI as it sees fit. Second, I think one callback for changesets, one before manifests, then one for each revlog would still be useful, as it is often the combined size of the other revlogs which make clone slow, IIRC. Cheers, Dirkjan

Information about cloning operations

Stefano — 2008-11-30T14:13:04

Hello, I'm looking for a way to provide more information about what's going on during cloning operations (read as: progress bar)... As far as I can see the needed information is at revlog.py level (addgroup) but there we don't have access to ui, while at an higher level we have addchangegroup (localrepo.py) but here we can just provide one progress.update before the changesets step, another one before the manifests step etc... and the result is not so useful. Any ideas? Cheers, Stefano

mercurial/crew< at >7444: 18 outgoing changesets

Mercurial Commits — 2008-11-30T12:00:07

18 outgoing changesets in mercurial/crew: http://hg.intevation.org/mercurial/crew/rev/f792c7bb2fb3 changeset: 7444:f792c7bb2fb3 tag: tip user: Thomas Arendsen Hein intevation.de> date: Fri Nov 28 11:38:41 2008 +0100 summary: Improvement to 14ce129cfcd: Use try/except and pass filename on errors http://hg.intevation.org/mercurial/crew/rev/1e0677756f60 changeset: 7443:1e0677756f60 user: Dirkjan Ochtman ochtman.nl> date: Fri Nov 28 09:50:30 2008 +0100 summary: gitweb: fix graph display in IE http://hg.intevation.org/mercurial/crew/rev/a14ce129cfcd changeset: 7442:a14ce129cfcd user: Edouard Gomez free.fr> date: Thu Nov 27 10:13:20 2008 +0100 summary: convert: check existence of ~/.cvspass before reading it http://hg.intevation.org/mercurial/crew/rev/4fecd17f2de9 changeset: 7441:4fecd17f2de9 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 22:04:51 2008 +0100 summary: convert cvs: Fix branch name parsing http://hg.intevation.org/mercurial/crew/rev/d2a917b27152 changeset: 7440:d2a917b27152 user: Brendan Cully kublai.com> date: Thu Nov 27 11:14:43 2008 -0800 summary: Fix test breakage introduced in 07faba78cf5a http://hg.intevation.org/mercurial/crew/rev/7dd44ad43914 changeset: 7439:7dd44ad43914 user: Brendan Cully kublai.com> date: Thu Nov 27 11:05:50 2008 -0800 summary: Ignore eclipse droppings http://hg.intevation.org/mercurial/crew/rev/b80b944836ec changeset: 7438:b80b944836ec user: Brendan Cully kublai.com> date: Thu Nov 27 11:04:29 2008 -0800 summary: Add hgcia to hgext http://hg.intevation.org/mercurial/crew/rev/3cdaac732b2b changeset: 7437:3cdaac732b2b user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 18:46:45 2008 +0100 summary: tests: fix test results for 07faba78cf5a http://hg.intevation.org/mercurial/crew/rev/07faba78cf5a changeset: 7436:07faba78cf5a user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 17:00:54 2008 +0100 summary: diff: fix obscure off-by-one error in diff -p http://hg.intevation.org/mercurial/crew/rev/5e13df32fb74 changeset: 7435:5e13df32fb74 user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 16:07:17 2008 +0100 summary: bundlerepo doesn't really have a dirstate, throw AttributeError if requested http://hg.intevation.org/mercurial/crew/rev/cf7741aa1e96 changeset: 7434:cf7741aa1e96 user: Dirkjan Ochtman ochtman.nl> date: Thu Nov 27 16:07:13 2008 +0100 summary: kill some trailing spaces http://hg.intevation.org/mercurial/crew/rev/c4ce828e8074 changeset: 7433:c4ce828e8074 parent: 7432:df0962f6c54e parent: 7426:642754e776e2 user: Benoit Boissinot ens-lyon.org> date: Thu Nov 27 01:35:12 2008 +0100 summary: merge with mpm http://hg.intevation.org/mercurial/crew/rev/642754e776e2 changeset: 7426:642754e776e2 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 01:10:27 2008 +0100 summary: rpm spec: include python egg in rpm http://hg.intevation.org/mercurial/crew/rev/3d827cc616b6 changeset: 7425:3d827cc616b6 user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 01:09:11 2008 +0100 summary: buildrpm: complain when hg command isn't available http://hg.intevation.org/mercurial/crew/rev/f0a3e87c810d changeset: 7424:f0a3e87c810d user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:32 2008 +0100 summary: util: use existing never() instead of custom lambda http://hg.intevation.org/mercurial/crew/rev/dbc40381620e changeset: 7423:dbc40381620e user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:31 2008 +0100 summary: tests: Skip tests if they will fail because of outer repo http://hg.intevation.org/mercurial/crew/rev/fdcde929ce4f changeset: 7422:fdcde929ce4f user: Mads Kiilerich kiilerich.com> date: Thu Nov 27 00:57:30 2008 +0100 summary: tests: use killdaemons in hgweb tests http://hg.intevation.org/mercurial/crew/rev/f21e3d0e335b changeset: 7421:f21e3d0e335b user: Patrick Mezard gmail.com> date: Wed Nov 26 22:27:37 2008 +0100 summary: util_win32: fix Windows version checking (issue1358)

excanvas.js obfuscated

TK Soh — 2008-11-30T03:38:56

I try to take a look at excanvas.js. Unlike other java scripts in mercurial's repo, this one is completely obfuscated. Any reasons for this?

Re: Call for testing: mercurial 1.1

TK Soh — 2008-11-30T03:32:40

I noticed something I don't quite understand: the path to excanvas.js has been hardcoded to "/mercurial/crew/static/excanvas.js". In any case, I try to change it withe attached patch, based on what I saw in other template files, and I begin to see the changesets in the graph page again on IE 6. diff -r 1e0677756f60 templates/gitweb/graph.tmpl --- a/templates/gitweb/graph.tmplFri Nov 28 09:50:30 2008 +0100 +++ b/templates/gitweb/graph.tmplSun Nov 30 11:23:22 2008 +0800 < at >< at > -4,7 +4,7 < at >< at > href="{url}atom-log" title="Atom feed for #repo|escape#"/> - +

[issue1403] add support to new rebase extension for handlingconcatenating changesets usecase

Faheem Mitha — 2008-11-29T19:18:09

New submission from Faheem Mitha email.unc.edu>: Hi, Someone on IRC (I think pmezard) said the functionality in http://www.selenic.com/mercurial/wiki/index.cgi/ConcatenatingChangesets could be folded into the new rebase extension. Since this is a regularly used feature, I'm submitting a feature request for this. If it is not possible to do this, feel free to close this issue. Regards, Faheem. ---------- messages: 8021 nosy: astratto, brendan, faheem, pmezard, tonfa priority: feature status: unread title: add support to new rebase extension for handling concatenating changesets usecase topic: rebase ____________________________________________________ Mercurial issue tracker selenic.com> ____________________________________________________