fileset: add hgignore

fileset: add resolved and unresolved predicates

fileset: prescan parse tree to optimize status usage We only call status if needed to avoid walking the working directory or comparing manifests. Similarly, we scan for whether unknown or ignored files are mentioned so we can include them.

fileset: add support for file status predicates This forcibly walks the tree looking for unknown and ignored files, which is suboptimal. A better approach would scan the tree first to find required status components and skip the status check entirely if it's unused.

fileset: add some basic predicates

match: introduce basic fileset support

match: allow passing a context object to match core

fileset: drop matchfn This is now built into contexts

scmutil: match no longer accepts repo objects

scmutil: switch match users to supplying contexts The most appropriate context is not always clearly defined. The obvious cases: For working directory commands, we use None For commands (eg annotate) with single revs, we use that revision The less obvious cases: For commands (eg status, diff) with a pair of revs, we use the second revision For commands that take a range (like log), we use None

scmutil: match now accepts a context or a repo

context: add a match builder method This is groundwork for matching against filesets in contexts

help/hgignore: refer to the builtin help instead of external URLs

sslutil: make messages for Python without certificate handling more helpful We now explain why there is no certificate to verify and do not give hints about cacerts when they can't be used anyway.

sslutil: fall back to commonName when no dNSName in subjectAltName (issue2798) Any entries in subjectAltName would prevent fallback to using commonName, but RFC 2818 says: If a subjectAltName extension of type dNSName is present, that MUST be used as the identity. Otherwise, the (most specific) Common Name field in the Subject field of the certificate MUST be used. We now only consider dNSNames in subjectAltName. (dNSName is known as 'DNS' in OpenSSL/Python.)

merge with i18n

svn subrepos: work around checkout obstructions (issue2752) We do this by ensuring the working copy is clean and then blowing away the working copy and replacing it with one from the desired path. We could probably use 'svn switch' to do this more efficiently, but there's some subtle logic required to get that right and this is more likely to work reliably.

update: check wc.dirty() before setting overwrite=True merge.update() was missing a few dirtiness checks from workingcontext, including subrepo cleanliness checks. Using wc.dirty() instead of one-off checks for various forms of dirtiness will be significantly safer.

tags: don't allow environment errors to be raised from _writetagscache See https://bitbucket.org/tortoisehg/thg/issue/719

help/config: replace docutils markup

test-i18n, i18n-pt_BR: update test message This changes both a test and a translation, so I'm sending a patch; please let me know if I should push this directly to i18n instead.

config: undo change of empty line definition We don't allow spaces in front of comment markers

patch: make filestore store data in memory and fallback to fs

i18n: merge in latest Romanian translations

i18n-ro: synchronized with 3c7907dc95ca

i18n-ro: outstanding changes

help/config: quote config section names consistently

help/config: fix small typo

help/config: do not refer to config files as hgrc files

help/config: fix 'Mercurial' casing

revset: add desc(string) to search in commit messages Like keyword(), but does not search in filenames and users. No grepdesc() or descgrep() added, because it might be bad to introduce grepfoo() versions of too many string searches.

revset: update sorting of symbols

http: fix variable name in unexpected response message

serve: add --cmdserver option to communicate with hg over a pipe

localrepo: don't attempt to open .hgtags twice if the error isn't ENOENT

log: do not display hidden changeset We add a --hidden option to display them.

hidden: Add ``hidden`` method for context

hidden: Add a hiddenrevs attributes to changelog. This attributes hold the set of all revisions that should be ommited by command and tools displaying changesets. This set is given as a hit. Command and tools are responsible to check it in order to filter they outpur. Code adding revisions to the set are responsible to the consistency of it's data.

config: handle comment lines in continuations (issue2854)

http: report unexpected unparsable push responses (issue2777)

dispatch: write shell alias output to ui out descriptor

commands: use ui descriptors when reading/writing from stdin/out

cmdutil: return a dummy, closable file object if it cannot be duped If the ui I/O descriptors aren't real descriptors, they cannot be duped. Instead, we return a wrapper object that behaves the same, and can be closed (by overriding close and doing nothing).

cmdutil: use ui descriptors in makefileobj

mq: use ui.fin when importing patch from '-'

cmdutil, logmessage: use ui.fin when reading from '-'

dagutil: fix missing import of i18n._

hgmanpage: adapt to Docutils 0.8 API change

manifest: use "\0" instead of "\000" Though both give the same result (a NUL byte), I found that I tend to read "\000" as "\0" + "00", which is something completely different. I did not change the occurance of "\000" in archival.py since there are other octal constants in that file.

commands: add pointer to bookmarks command in branch help

glossary: add entry for "Bookmark"

glossary: add entry for "Tag"

merge with i18n

i18n-ru: translated all standard commands and config-environment additional help topics

i18n-ru: translated push-rename

i18n-ru: translated parents-pull

setdiscovery: batch heads and known(ownheads) This means that we now discover both subset conditions (local<remote and remote<local) in a single roundtrip without ever constructing an actual sample (which takes a bit of client CPU).

wireproto: make a number of commands batchable Makes lookup, heads, known, branchmap, pushkey, and listkeys batchable. It could, for instance, be interesting to use this to batch calls to lookup when a pull or clone has multiple --rev arguments. The next patch is going to batch heads and known to slightly tune discovery.

wireproto: add batching support to wirerepository Adds the plumbing and wire call for batched execution, but does not batch-enable any methods yet.

wireproto: add basic command batching infrastructure Note that localbatch will not be used until we actually have a localpeer to use it with.

mq: make qrefresh/qfold keep wlock until saving patch status Because q.refresh() changes nodeid, .hg/patches/status gets invalid until q.savedirty(). This patch changes mq not to unlock repository of incomplete state.

bugzilla: fix typo in documentation

dispatch: fix for-loop variable name

notify: send changesets on 'outgoing' hook, updated doc

sslutil: Restore missing imports of socket and httplib to sslutil Two imports were omitted in the restructure of the code creating sslutil.py, socket and httplib are required when the 'ssl' module cannot be imported, restoring these imports allows mercurial to run on python2.4+2.5.

dispatch: assign I/O descriptors from the request to the ui

ui: use I/O descriptors internally and as a result: - fix webproto to redirect the ui descriptors instead of sys.stdout/err - fix sshserver to use the ui descriptors

dispatch: add I/O descriptors to the request

ui: add I/O descriptors

import: add --bypass option This feature is more a way to test patching without a working directory than something people asked about. Adding a --rev option to specify the parent patch revision would make it a little more useful. What this change introduces is patch.repobackend class which let patches be applied against repository revisions. The caller must supply a filestore object to receive patched content, which can be turned into a memctx with patch.makememctx() helper.

import: separate parents selection from working dir update This will be useful when patching without updating the dirstate

patch: extend filtestore to store an optional copy source This will help wrapping filestores in memctx.

paper/coal: use fixed width for diffstat +/- link The plus and minus characters are normally not the same width in a non-monospace font, and this made the line length change when the diffstat display was toggled. The square brackets are not rendered in a monospace font to ensure that they align with the parenthesis on the same line.

hg: rename opts argument to peeropts in clone This should make it clearer that the options are used with remoteui only and are not used otherwise in hg.clone.

hg: move peerschemes back to schemes This will avoid breaking things with extensions until peers are fully separated from repos.

hg: rearrange peer scheme lookup There is now only peer scheme lookup. Repository lookup goes through peer scheme lookup. When peer and repo types are finally separated, repo lookup will use peer.local() to get a repository object. The underbar is dropped so that extensions can patch the table.

forget, remove: don't note on nonexistent file twice before: $ hg forget foo foo: No such file or directory not removing foo: file is already untracked after: $ hg forget foo foo: No such file or directory

localrepo: local() returns self Temporary measure for introducing peers

tests: avoid instability in test-convert-cvs-synthetic

dispatch: propagate ui command options to the local ui (issue2523) so the ui object passed to pre/post python hooks has the verbose flag (and the rest) set correctly

mq: catch correct exception when calling changelog.rev()

rebase: preserve mq series order after rebasing (issue2849)

run-tests: verbose log ignored test

record: add white space diff options

mq: rename abort_if_wdir_patched to abortifwdirpatched

mq: rename partial_name to partialname

mq: rename write_list to writelist

mq: rename series_dirty to seriesdirty

mq: rename applied_dirty to applieddirty

mq: rename guards_dirty to guardsdirty

mq: rename active_guards to activeguards

mq: rename guards_path to guardspath

mq: rename status_path to statuspath

mq: rename series_path to seriespath

mq: rename series_end to seriesend

mq: rename full_series_end to fullseriesend

mq: rename check_reserved_name to checkreservedname

mq: rename check_localchanges to checklocalchanges

mq: rename check_substate to checksubstate

mq: rename check_toppatch to checktoppatch

mq: rename save_dirty to savedirty

mq: rename explain_pushable to explainpushable

mq: rename set_active to setactive

mq: rename set_guards to setguards

mq: rename check_guard to checkguard

mq: rename parse_series to parseseries

mq: rename find_series to findseries

mq: rename series_guards to seriesguards

mq: rename full_series to fullseries

paper, coal: display diffstat on the changeset page Displayed in a row of the changeset summary table, underneath the list of files. When the page is loaded, only the diff summary is displayed. The full diffstat is only displayed when the [+] link is selected.

web: provide diff summary to the changeset page This is the same message displayed at the end of the "diff --stat" command. For example, "9 files changed, 1651 insertions(+), 2 deletions(-)". The webutil.diffstatgen function allows the diffstat data to be lazily calculated only once and then re-used.

tests: fix up date ambiguity in cvs test

hg: split peer and repo lookup tables

tests: rename test-git-import.t into test-import-git.t This is consistent with other test-import* tests.

patch: generalize the use of patchmeta in applydiff() - Add patchmeta.copy() and emit copies from iterhunks. Modifying patchmeta instances in applydiff() makes things simpler. - Rename selectfile() into makepatchmeta(). It is responsible for creating patchmeta for regular patches. - Pass patchmeta objects to patchfile() directly patchmeta instances were associated with git patches, for regular patches we had to pass additional variables to tell the patch intent to patchfile(). Instead, we generate patchmeta for regular patches and pass them. This will also help with patch filtering by matcher objects.

patch: stop updating changed files set in applydiff() This information is more correctly returned by backends. The extra updated file removed from test-mq-merge.t output came from changes from git patches being counted before being really applied in some cases.

patch: turn patch() touched files dict into a set

paper, coal: define the diffstat templates The templates output the filename, the total number of changes to the file and a bar graph of the adds and removes. The filename is a link into the diff output. The diffstat is not yet displayed, but it can be manually added to the changeset page template.

web: provide the file number to the diffstat templates This allows the diffstat templates to link into the diff output. For example, the URLs of the first three files within the diff are #l1.1, #l2.1, #l3.1.

web: include all files in the diffstat The webutil.diffstat function now returns a diffstat template for each file in the diff. It previously returned a template for each file returned by ctx.files() which did not work well for merge changesets.

hgignore: ignore MANIFEST.in (generated since 2ce7dfe17bc5)

bookmarks: recognize the current bookmark when the local encoding isn't UTF-8 The current bookmark is stored in bookmark.current, supposingly in UTF-8. But the call to encoding.fromlocal() is missing, therefore Hg is not able to recognize the current bookmark in the case that bookmark uses characters of which the bit stream is different between local encoding and UTF-8. For example, the Chinese version of Windows cmd uses gbk(cp936), not UTF-8. Therefore I won't be able to make a Chinese bookmark current. By wrapping mark in a encoding.fromlocal() call, the problem is solved.

distutils: Create MANIFEST.in instead of MANIFEST in Makefile When running 'python setup.py sdist' many files were omitted from the resulting tarball that are required for a complete build, including the .h file for the inotify extension due to a python2.7+ distutils bug. This would cause 'make dist' to fail to build a correct dist on python2.7 By building a MANIFEST.in that is the same as the old MANIFEST with 'include' at the start of each line will cause a correct MANIFEST to be built and used by distutils.

revert: is reverting file states, not just file contents e.g. x bit is reverted as well Also makes this paragraph more consistent with the preceding one, which already uses the term "state" (not "content").

hg: change various repository() users to use peer() where appropriate This gets all the easy cases (peers that aren't also used as repositories).

hg: have peer() call _lookup directly Note that this means that peers no longer call the extension repository setup hooks.

hg: add peer method

hg: add opts argument to clone for internal remoteui

hg: use capable instead of capabilities

fileset: basic pattern and boolean support debugfileset can now generate file lists for things like: "* and not hg*"

tests: fix updated monotone version requirement 0969d91fad5c only updated the comment, not the actual check.

check-code: catch misspellings of descendant This word is fairly common in Mercurial, and easy to misspell.

revert: drop obvious paragraph about filenames

revert: actually add pointer to backout

revert: replace mention of 'roll back' with pointer to 'backout'

revert: simplify description of effect of -r

revert: rearrange the date help

revert: remove some redundancy in basic description

revert: drop requirement to use -r to revert with two parents This reduces documentation confusion between the need to: a) hg revert -a -r . (drop all changes from a merge) b) hg up -C . (drop the second parent entirely) Currently revert is one of two commands (the other being tag) that still complains about uncommitted merges, dating from its former use of a generic defaultrev function that aborted.

revert: simplify usage note This points people looking for other commands to the right place with a minimum of verbiage.

revert: rewrite help summary New users have a tendency to mistake 'revert' as the command to use to check out old revisions. They also occasionally mistake revert for a generalized undo (compare rollback). This version intentionally aims to avoid mentioning 'earlier' and thus intentionally no longer alludes to the (secondary) -r behavior (which in fact is not actually limited to 'earlier'). Instead, we mention checkout state, to convey that we can restore things to the way they were when checked out.

extensions: make disabled()/disabledext() load prebuilt index if available __index__ contains a dict of {name: desc} of all extensions.

setup: add command to generate index of extensions It generates prebuilt index of all extensions, which will be used by frozen exe when running 'hg help extensions'. Now py2exe invokes this command automatically.

sshrepo: fix addchangegroup's signature Make it mirror the usage found in localrepo.push. This should only affect pushes to pre-unbundle ssh servers.

subrepo: be more careful with deletions of .hgsub and .hgsubstate (issue2844) We shouldn't try to process subrepos if .hgsub isn't present and we should remove .hgsubstate if .hgsub is marked for removal.

patch: dot not ignore hunk of files marked as 'deleted' git 'deleted' flag was processed unconditionnally and the file removed even if the related hunk was not matching.

patch: fix patchmeta/hunk synchronization in iterhunks() Synchronizing on bfile does not work on file removal where bfile is /dev/null. We match items on afile or bfile instead. The incorrect code makes iterhunks() to emit patchmeta and hunks separately in some cases. This is currently hidden by applydiff() being too tolerant when processing patchmeta, and will be fixed later.

patch: remove unnecessary exists() call in selectfile()

import: deprecate --base This option was a convenient hack introduced in Mercurial early days which unfortunately survived until now.

patch: remove redundant islink() call

extensions: update doc of enabled() and disabled() according to d5b525697ddb

localrepo: add savecommitmessage() to write last-message.txt

memctx: simplify constructor

scmutil: improve path calculation for install-relative RC files (issue2841) This code was ugly and might do the wrong thing with symlinks.

prepush: show details about new remote heads with --verbose previously, this info was only shown with --debug Note that the new remote heads are only shown if the command aborts, that is, they are *not* shown if --force was specified. See also change 91cb08a9e7fb.

prepush: print short hash of first new head in abort message and don't mention the branch if it's the default branch. OLD: $ hg push ../a pushing to ../a searching for changes abort: push creates new remote heads on branch 'default'! (you should pull and merge or use push -f to force) NEW: $ hg push ../a pushing to ../a searching for changes abort: push creates new remote head 1e108cc5548c! (you should pull and merge or use push -f to force) This helps to identify which local head is causing troubles. See also change 91cb08a9e7fb.

prepush: eliminate 'if branch' branch is always set anyway No change in program behavior.

revlog: bail out earlier in group when we have no chunks

localrepo: simplify file bundling code and fix progress bug Progress for files was off by one, and the code was rather hackish.

localrepo: use list comprehension instead of generators The lists in question are iterated and kept in memory further down the code path anyway, and this simplifies things.

localrepo: add total to changeset progress in bundle/push

statichttprepo: use in-place difference for shorter line

workingctx: eliminate remove function Inlining it into it's last remaining call place in cmdutil.copy. Note that cmdutil.copy is called with the wlock already held, so no additional locking is needed to call util.unlinkpath. We do not need to wrap the util.unlinkpath call into a try block, because at that point we already know whether abssrc exists or not -- thanks to the preceding util.copyfile call. Adding a new local 'srcexists' in cmdutil.copy for that purpose.

test-copy2: add case for moving a missing file

color: check if ui is already a subclass of colorui before wrapping it since it's possible to reuse the ui object (see 80c599eee3f3) between dispatch calls, the ui might have been wrapped already. we could probably move this test up a bit but leaving it here allows updates to take effect by extstyles() and configstyles().

util: add helper function isatty(fd) to check for tty-ness

web: prevent TypeError thrown when using the diffstat templates Remove the lambda used to wrap webutil.diffstat because: - functions passed to the templater must accept keyword arguments - webutil.diffstat is a generator, so already calculates the diffstat lazily This reverts the changes made to 1d3e2349304a after it was submitted to the mailing list but before it was queued.

fileset: fix long line

COPYING: refresh with current address from fsf.org

filesets: introduce basic fileset expression parser

dispatch: add repo to the request allows callers of dispatch.dispatch to pass in an initialized repo object to be used. -R/--repository overrides the request repo.

merge with stable

Added signature for changeset 733af5d9f6b2

Added tag 1.8.4 for changeset 733af5d9f6b2

subrepo: make stdin for svn a pipe for non-interactive use (issue2759) This certainly can't hurt, so go ahead and do it, potentially along with --non-interactive if that flag is safe for the given subcommand.

subrepo: svn abort now depends on exit code (issue2833)

httprepo: send URL redirection notices to stderr (issue2828)

httprepo: proper handling of invalid responses without content-type (issue2019) This can currently be tested on http://sf.net/

strip: make it clear that --force discards changes (issue310)

tags: catch more corruption during cache parsing (issue2779)

localrepo: don't add deleted files to list of modified/added files (issue2761) If a file is deleted (rm, not 'hg rm') from the working dir an attempt to run 'hg diff -r X', with the file being present in X will cause an abort. We didn't check if the file has been deleted from the working dir and later on tried to open it to compare with the one from X, causing the abort. This fix adds that check. Consequently, no output will be returned.

localrepo: ignore tags to unknown nodes (issue2750)

bookmarks: do not forward merged bookmark (issue1877)

rebase: restore mq guards after rebasing (issue2107) Guards on rebased mq patches were lost. This patch restores them after the qimporting step.

revset: report a parse error if a revset is not parsed completely (issue2654)

color: fix --color=always when output is piped Also restructures the mode determination as discussed on IRC. I tried to simplify the code too, while keeping the original semantics except for the case of --color=always.

check-code: flag 0/1 used as constant Boolean expression

run-tests: allow whitelisting tests that should always run It's desirable to run some tests all the time, for example test-check-pyflakes.t and test-check-code-hg.py. This allows passing --whitelist as a path to a file (flag can be specified more than once) which contains a list of files to whitelist. Whitelisted tests are run even if they're blacklisted or wouldn't match a --keyword test run. For example, to do a quick test of usehttp2, one can now do $ cat > test-whitelist <<EOF > test-check-pyflakes.t > test-check-code-hg.py > EOF $ (cd tests && ./run-tests.py --extra-config-opt 'ui.usehttp2=true' > -k http -j 8 --whitelist test-whitelist) and have all http-specific tests run as well as the two code linters.

subrepo: make stdin for svn a pipe for non-interactive use (issue2759) This certainly can't hurt, so go ahead and do it, potentially along with --non-interactive if that flag is safe for the given subcommand.

tests: glob version number in test-hgrc

web: provide diffstat to the changeset page This includes all affected files, so it can be used for an extended view of the files or as a replacement for the filenodelink and filenolink templates.

tests: update for handling of missing hgrc includes

subrepo: svn abort now depends on exit code (issue2833)

fixed issue2833: aborts nows depends on svn exit code

config: ignore include errors for nonexistent files

Make pull -u behave like pull && update Previously, pull would not update if new branch heads were received, whereas pull && update would move to the tipmost branch head. Also change the "crosses branches" abort in merge.update from "crosses branches (merge branches or use --check to force update)" to "crosses branches (merge branches or update --check to force update)" since it can no longer assume the user is running hg update.

requires: note apparent corruption

readrequires: provide easier to understand error message NEW: $ hg id abort: unknown repository format: requires feature 'foo' (upgrade Mercurial)! OLD: $ hg id abort: requirement 'foo' not supported!

introduce new function scmutil.readrequires for reading and parsing the .hg/requires file

merge with stable

merge with crew

merge with i18n

merge with stable

merge with hg-i18n

merge with hg-i18n

test-gendoc: update for new Russian translation

i18n-ru: branches-import translated

i18n-ru:bisect,backout,branch translated

i18n-ru: translated add, addremove, annotate, archive

i18n-ru:translation started,convert translated

i18n-pt_BR: synchronized with ab687820c4cc

subrepo: don't crash when git .hgsubstate is empty (issue2716)

mq: no need to make a copy of seriesopts the command decorator takes care of copying the options (introduced in 5b48ad1e7f1a)

workingctx.remove: use self.forget

commands: use repo.branchheads in heads command

commands: remove unused variable from heads command The b variable has not been needed since fd511e9eeea6. Renamed ls variable to bheads for clarity.

mq: print "'foo' 'bar'", not "['foo', 'bar']" when showing guards The internal list representation of guards was leaking into the output. The guards were always printed using repr(guard) and that style was kept. When "hg qguard -l" prints several guards for a patch, it does so by joining the names with " " and that style was used for the error messages too.

hg: remove underscores in clone function

hg-ssh: fix dispatch call to use dispatch.request() 08bfec2ef031 changed dispatch.dispatch() to accept a request object instead of a list of arguments

gendoc: config help topic is in hgrc.5, do not include it in hg.1

doc: improve merge between hgrc.5 and config help topic

doc: rebuild hgrc.5 man and HTML page when help/config changes

help/config: separate terms with a blank line This makes it easier for translators since they can then translate each term individually.

help/config: fix rendering of definition list Without the blank line, the minirst parser renders Term-1 Line-1 Line-2 Term-2 Line-1 as Term-1 Line-1 Line-2 Term-2 Line-1 because the second term is seen as a paragraph.

help: move part of hgrc.5 man page config help topic

hgrc.5: make minirst find headings correctly The minirst parser is stricter than Docutils here and require a blank after a heading. Otherwise the heading is classified as a paragraph.

run-tests: fix --blacklist (broken by 95715c2f90bf)

patch: do not patch unknown files (issue752)

patch: use temporary files to handle intermediate copies git patches may require copies to be handled out-of-order. For instance, take the following sequence: * modify a * copy a into b Here, we have to generate b from a before its modification. To do so, applydiff() was scanning for copy metadata and performing the copies before processing the other changes in-order. While smart and efficient, this approach complicates things by handling file copies and file creations at different places and times. While a new file must not exist before being patched a copied file already exists before applying the first hunk. Instead of copying the files at their final destination before patching, we store them in a temporary file location and retrieve them when patching. The filestore always stores file content in real files but nothing prevents adding a cache layer. The filestore class was kept separate from fsbackend for at least two reasons: - This class is likely to be reused as a temporary result store for a future repository patching call (entries just have to be extended to contain copy sources). - Delegating this role to backends might be more efficient in a repository backend case: the source files are already available in the repository itself and do not need to be copied again. It also means that third-parties backend would have to implement two other methods. If we ever decide to merge the filestore feature into backend, a minimalistic approach would be to compose with filestore directly. Keep in mind this copy overhead only applies for copy/rename sources, and may even be reduced to copy sources which have to handled ahead of time.

patch: refactor file creation/removal detection The patcher has to know if a file is being created or removed to check if the target already exists, or to actually unlink the file when a hunk emptying it is applied. This was done by embedding the creation/removal information in the first (and only) hunk attached to the file. There are two problems with this approach: - creation/removal is really a property of the file being patched and not its hunk. - for regular patches, file creation cannot be deduced at parsing time: there are case where the *stripped* file paths must be compared. Modifying hunks after their creation is clumsy and prevent further refactorings related to copies handling. Instead, we delegate this job to selectfile() which has all the relevant information, and remove the hunk createfile() and rmfile() methods.

commands.remove: don't use workingctx.remove(list, unlink=True) workingctx.remove(list, unlink=True) is unsuited here, because it does too much: it also unlinks added files. But the command 'hg remove' is specified to *never* unlink added files. Instead, we now unlink the files at the commands.remove level (if --after was not specified) and use workingctx.forget for all files. As an added bonus, this happens to eliminate a wlock acquire/release pair, since the previous implementation caused acquire wlock release wlock acquire wlock release wlock where the first pair of acquire/release was caused by the workingctx.forget call, and the second by the workingctx.remove call.

tests: add a test to check for duplicate command options

mq: allow --mq for qrecord

run-tests: fix --blacklist with jobs > 1 filter blacklisted tests before partitioning them to the children. maintains the 'Skipped...' output but not the 's'.

run-tests: slightly simplify blacklist check

coal: use preformatted text for descriptions (issue2835) The coal style was broken here by an earlier change to paper, which shares files.

rebase: add option --tool/-t for 'pull --rebase' This makes 'pull --rebase' consistent with the merge command, which already provides that option to control the merges

subrepo: refactor writestate for clarity

cmdutil: make private copies of option lists to avoid sharing monkeypatches

record: fix options placeholder

subrepo: bare git repos considered dirty Currently, if there is a bare git subrepo, but it is at the "right" revision, calling dirty() will error because diff-index does not work on bare repos. This patch makes it so bare subrepos are always considered dirty.

dispatch: use the request to store the ui object and check if we got one before creating. note that the contents of the ui object might change after dispatch() returns (by options passed through --config for example), to ensure it doesn't, pass a copy() of it.

dispatch: wrap dispatch related information in a request class currently only stores the arguments.

patch: restore the previous output of 'diff --stat' Restore the previous diffstat behaviour of scaling by the maximum number of changes to a single file. Changeset f03f08240c32 modified the diffstat to be scaled by the total number of changes. This seems to have been unintentional.

wireproto: enable optional args for known() for future extensibility Firstly, I think we should do this for all new wire commands, just to be on the safe side. So I want to get this into the 1.9 release. Secondly, there actually is potential here that sometimes the server can know that the number of its nodes which can possibly still be undecided on the client is small. It might then just send them along directly (cutting short the end game). This, however, requires walking the graph on the server, which can be expensive, so for the moment we're not actually doing it.

context: make forget work like commands.forget Switch users of wctx.delete(..., False) to forget.

dirstate: rename forget to drop It has substantially different semantics from forget at the command layer, so change it to avoid confusion. We can't simply combine it with remove because we need to explicitly drop non-added files in some cases like commit.

minirst: read test input from stdin

tests: update monotone version requirement

test-git-import: test patching existing copy targets

httprepo: handle large lengths by bypassing the len() operator

workingctx.remove: don't stat files again after unlinking we already know at this point that they have been unlinked

httprepo: send URL redirection notices to stderr (issue2828)

record: alias qrecord to qnew -i/--interactive

record: add qrefresh -i/--interactive interactively select changes to refresh

record: add an option to backup all wc modifications Also, don't create a backup dir if we have no files to backup. This is essential for qrefresh --interactive. Since we can't select individual files to qrefresh without eliminating already present changes, we have to backup all changes in the working copy to avoid refreshing unaccepted hunks. (thanks to Patrick for the idea)

record: check patch name is valid before prompting in qrecord

mq: use checkpatchname This also fixes a bug where qrename would allow renaming a patch to a reserved name.

mq: wrap patch file name checks in a function

pure parsers: properly detect corrupt index files This new Python code should be equivalent in behavior to the if statement at line 312 of parsers.c. Without this, the pure-python parsers improperly ignore truncated revlogs as created in test-verify.t.

workingctx: unlink paths while holding the wlock

wireproto: do not hash when heads == ['force'] Changeset 88f0e41d8802 introduced the unbundlehash capability and unconditionally hashed the heads on the client side. By mistake, the heads were also cased in the heads == ['force'] case.

patch: remove EOL support from linereader class This was only used when reading patched files which is now done by backends.

subrepo: handle local added subrepo case correctly

graphlog: remove unused arg from _wrapcmd

extensions: raise when trying to find an extension that failed to load extensions that depend on other extensions (such as record) use this pattern to check if the dependant extension is available: try: mq = extensions.find('mq') except KeyError: return but since if an error occurs while loading an extension it leaves its entry in the _extensions map as None, we want to raise in that situation too. (rather than adding another check if the return value is None)

remove: clarify help text about added files never being deleted from disk

pure: provide more correct implementation of posixfile for Windows requires ctypes Why is posixfile a class? Because the implementation needs to use the Python library call os.fdopen [1], which sets the 'name' attribute on the Python file object it creates to the mostly meaningless string '<fdopen>', since file descriptors don't have a name. But users of posixfile depend on the name attribute [2] being set to a proper value, like Python's built-in 'open' function sets it on file objects. Python file's name attribute is read-only, so we can't just assign to it after the file object has alrady been created. To solve this problem, we save the name of the file on a wrapper object, and delegate the file function calls to the wrapped (private) file object using __getattr__. [1] http://docs.python.org/library/os.html#os.fdopen [2] http://docs.python.org/library/stdtypes.html#file.name

bundlerepo: make getremotechanges support filtering of incoming Extensions can hook discovery.findcommonincoming to filter out unwanted remote changesets. This patch makes getremotechanges respect the changed remote heads returned by such extensions.

tests: add tests for discovery/pull without cgsubset

tests: add tests for partial pulls with treediscovery

tests: support multiple caps in notcapable

record: use cmdutil.command decorator

record: suggest the right command when running non interactively

applyupdates: audit merged files protects changing flags on merged files (util.setflags call on line 341)

applyupdates: audit path on flag changes we're using the auditor of the repo wopener, since this is a path that belongs to the tree we're updating to

opener: add audit function

templatekw: use diffstatsum in diffstat keyword

patch: use diffstatsum in diffstat

patch: add diffstatsum helper

diffstatdata: no longer a generator This produces a smallish amount of data and all consumers needed to buffer it anyway.

add new option --all to manifest command prints a list of all files in all revisions of the repo obsoletes the cifiles extension

applyupdates: audit unlinking of renamed files and directories

patchbomb: pass --git argument to diffstat

mq: strip all leading slashes from url when importing When trying to import a url that ends with a slash os.path.basename() would return an empty string. So when seeing a leading slash, remove it and infer the patch name from the remaining url. e.g. `hg qimport http://paste.pocoo.org/raw/xxx/` deduced '.' to be the patch name. Now we'll deduce 'xxx'.

mq: check patch name is valid before reading imported file

tests: update monotone output for v1.0 changes

revlog: stop exporting node.short

patch: fast-path git case in selectfile() We avoid a lot of complicated heuristics in git cases, where these heurestics may even be broken when copies are involved.

patch: unify backend file access interface - Rename readlines() into getfile(), return data and mode - Make setfile() write a data buffer instead of lines, make mode mandatory.

patch: merge backend setmode() into writelines() Copy handling will be easier to handle in a single method.

patch: stop modifying gitpatch objects gitpatch objects emitted by iterhunks() were referencing file paths unmodified from the input patch. _applydif() made them usable by modifying the gitpatch objects in-place with specified path strip level. The same modified objects were then reused by iterhunks() generator. _applydiff() now copies and update the paths which completely decouples both routines. As a side effect, the "git" event now receives only metadata about copies/renames to perform the necessary copies ahead of time. Other actions are handled in the "file" event.

patch: stop handling hunkless git blocks out of stream Patch changes are emitted by iterhunks() in two separate events: 'file' when hunks have to be applied and 'git' to describe other modifications like copies or mode changes. Note that a file which mode is changed and which content is modified by the same patch will be emitted in both events. It is more convenient to handle all file modifications in a single event. This patch "zips" git actions with regular changes so both kinds can be emitted at the same place.

patch: reindent code

patch: unify iterhunks() afile/bfile handling git afile/bfile are extracted twice, once when reading a 'diff --git', and again when reading a unified hunk. The problem is not all git blocks have unified hunks (renames just have metadata) and they were not extracted the same way. This is what this patch unifies.

patch: git metadata was ignored if strip > 1 gitpatch objects emitted by iterhunks() are modified in place by applydiff(). Processing them earlier improves iterhunks() isolation. applydiff() modifying them should still be fixed though.

patch: construct and parse binary hunks at the same time

patch: refactor iterhunks() regular and binary files emission

patch: remove patch.patch() cwd argument

patch: merge _updatedir() into externalpatch()

merge with stable

changelog: convert user and desc from local encoding early Failing to do so makes it impossible to use the memctx API to create a changeset with a commit message or username outside of the current encoding.encoding setting.

merge crew with main

clone: make default path absolute for all local paths This fixes the default path when cloning local repositories with hg-git, since hg-git cancopy() is false even for local repositories.

httpclient: import revision fc731618702a of py-nonblocking-http

httpconnection: improved logging formatting I had to use this debugging output for the first time recently when looking for a problem, and the lack of good formatting made things difficult.

bash_completion: enable alias auto-complete When auto-completing, set HGPLAINEXCEPT=alias to list aliases in the suggestions. Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

ui: enable alias exception when reading config in plain mode When in plain mode with "alias" present in the exception list, keep the aliases. This will be used later to enable auto-completion. Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

ui: test plain mode against exceptions Let ui.plain() accept an optional parameter in the form of a feature name (as a string) to exclude from plain mode. The result of ui.plain is now: - False if HGPLAIN is not set or the requested feature is in HGPLAINEXCEPT - True otherwise Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

revlog: drop base() again deltaparent does what's needed, and more "portably".

patch: add a workingbackend dirstate layer on top of fsbackend _updatedir() is no longer used by internalpatch() The change in test-mq-missingfiles.t comes from workingbackend not considering the missing 'b' file as changed, thus not calling addremove() on it.

patch: handle binary copies as regular ones This introduces a performance regression for large files, as they will be copied just to be clobbered afterwards since binary patching does not use deltas. But it simplifies the code and the previous optimization will be reintroduced later in a better way.

patch: remove files while patching, not in updatedir() At this point, updatedir() only reads the working directory and update the dirstate.

patch: set desired mode when patching, not in updatedir() This patch and the following aim at merging _updatedir() actions into _applydiff().

patch: stop changing current directory before patching _applydiff() patcher argument was added to help hgsubversion like extension monkeypatching the patching process. While it could be removed at this point, I prefer to leave it until patch.py is completely refactored and there is a valid and tested alternative.

revlog: linearize created changegroups in generaldelta revlogs This greatly improves the speed of the bundling process, and often reduces the bundle size considerably. (Although if the repository is already ordered, this has little effect on both time and bundle size.) For non-generaldelta clients, the reduced bundle size translates to a reduced repository size, similar to shrinking the revlogs (which uses the exact same algorithm). For generaldelta clients the difference is minor. When the new bundle format comes, reordering will not be necessary since we can then store the deltaparent relationsships directly. The eventual default behavior for clients and servers is presented in the table below, where "new" implies support for GD as well as the new bundle format: old client new client old server old bundle, no reorder old bundle, no reorder new server, non-GD old bundle, no reorder[1] old bundle, no reorder[2] new server, GD old bundle, reorder[3] new bundle, no reorder[4] [1] reordering is expensive on the server in this case, skip it [2] client can choose to do its own redelta here [3] reordering is needed because otherwise the pull does a lot of extra work on the server [4] reordering isn't needed because client can get deltabase in bundle format Currently, the default is to reorder on GD-servers, and not otherwise. A new setting, bundle.reorder, has been added to override the default reordering behavior. It can be set to either 'auto' (the default), or any true or false value as a standard boolean setting, to either force the reordering on or off regardless of generaldelta. Some timing data from a relatively branch test repository follows. All bundling is done with --all --type none options. Non-generaldelta, non-shrunk repo: ----------------------------------- Size: 276M Without reorder (default): Bundle time: 14.4 seconds Bundle size: 939M With reorder: Bundle time: 1 minute, 29.3 seconds Bundle size: 381M Generaldelta, non-shrunk repo: ----------------------------------- Size: 87M Without reorder: Bundle time: 2 minutes, 1.4 seconds Bundle size: 939M With reorder (default): Bundle time: 25.5 seconds Bundle size: 381M

revlogdag: add linearize function See the docstring for a detailed explanation. The linearizer was originally written by Benoit Boissinot.

localrepo, sshrepo: use Boolean create argument in __init__

commands: replace 'x = f(); return x' with 'return f()'

merge with stable

subrepo: respect non-default path for incoming/outgoing Push and pull set repo._subtoppath before pushing or pulling subrepos. The incoming and outgoing commands needed to do the same.

run-tests: print a newline after all warnings

annotate: fix alignment of columns in front of line numbers (issue2807)

merge with stable

revset: the name is optional for the tag predicate

revset: note case-sensitive match in grep

revset: note case-insensitive matches in keyword and user

revset: expand help for contains predicate

patch: remove unused variable to make pyflakes test happy

patch: add lexists() to backends, use it in selectfile() At this point, all applydiff() filesystem calls should pass through fsbackend.

patch: move copyfile() into backends, abstract basedir

patch: merge makerejlines() into write_rej()

patch: extract fs access from patchfile into fsbackend Most filesystem calls are already isolated in patchfile but this is not enough: renames are performed before patchfile is available and some chmod calls are even done outside of the applydiff call. Once all these calls are extracted into a backend class, we can provide cleaner APIs to write to a working directory context directly into the repository.

mq: fix qpush changes detection of renamed files patch.changedfile() was not considering renamed file source as changed.

httpconnection: correctly handle redirects from http to https Previously the connection cache for keepalives didn't keep track of ssl. This meant that when we connected to an https server after that same server via http, both on the default port, we'd incorrectly reuse the non-https connection as the default port meant the connection cache key was the same.

win32.py: add argtypes and restype This is a feature of ctypes. Without these, pypy complains with RuntimeWarning: C function without declared arguments called RuntimeWarning: C function without declared return type called As a side effect of specifying restypes, the return value of e.g. CreateFileA is now implicitly converted to an instance of _HANDLE, so we also need to change the definition _INVALID_HANDLE_VALUE = -1 to _INVALID_HANDLE_VALUE = _HANDLE(-1).value Otherwise, tests for equality to _INVALID_HANDLE_VALUE in code like def _getfileinfo(name): fh = _kernel32.CreateFileA(name, 0, _FILE_SHARE_READ | _FILE_SHARE_WRITE | _FILE_SHARE_DELETE, None, _OPEN_EXISTING, 0, None) if fh == _INVALID_HANDLE_VALUE: _raiseoserror(name) would now fail to detect an invalid handle, which in turn would lead to exceptions raised with wrong errno values, like e.g. >>> nlinks('missing.txt') Traceback (most recent call last): ... OSError: [Errno 9] missing.txt: The handle is invalid. instead of the correct (as per this patch and before it) >>> nlinks('missing.txt') Traceback (most recent call last): ... OSError: [Errno 2] missing.txt: The system cannot find the file specified.

win32.py: more explicit definition of _STD_ERROR_HANDLE

revset: add follow(filename) to follow a filename's history across copies

revset: introduce filelog() to emulate log's fast path filelog() only reports revisions present in the matching filelogs.

httpclient: import revision b8c3511a8cae from py-nonblocking-http Fixes issues with SSL_ERROR_WANT_READ incorrectly breaking the response read.

run-tests: replace chdir() with Popen cwd option

run-tests: make --debug run in the temporary test directory

run-tests: use the common test path on Windows and Java The alternate one did not run in the test directory and controlling the jobs execution with threads instead of process made it harder to fix.

run-tests: ignore timeout when Popen.terminate is unavailable Popen.terminate was introduced in python 2.6

run-tests: fix hg.bat python reference hg.bat expects to live in pythonxx/scripts and the python interpreter to be in pythonxx. run-tests.py file layout is a little different and python location must be fixed.

run-tests: fix python executable detection and copy on Windows

changelog: don't use generaldelta

revlog: get rid of defversion defversion was a property (later option) on the store opener, used to propagate the changelog revlog format to the other revlogs, so they would be created with the same format. This required that the changelog instance was created before any other revlog; an invariant that wasn't directly enforced (or documented) anywhere. We now use the revlogv1 requirement instead, which is transfered to the store opener options. If this option is missing, v0 revlogs are created.

bash_completion: Use "hg paths -q" instead of piping through sed

paths: Add support for -q/--quiet Suppresses output (resolved paths or "not found!") when searching a path, similar to "grep -q". Sample usage: hg paths -q foo || echo "there is no foo" Just prints path names (instead of "name = result") when listing all path definitions, like "hg bookmarks -q". Sample usage: hg paths -q | while read i; do hg incoming "$i"; done

fix test broken in merge from stable

merge with stable

simplemerge: do not allow binary files to abort an entire merge When used as the default merge tool, or used as a --tool override, the simplemerge script should not be allowed to raise a util.Abort just because one of the files being merged is binary. Instead, return 1 and mark the file unresolved.

extdiff: grammar "allows to" -> "allows one to" The verb to allow requires a direct object. Lintian, a Debian tool to find common mistakes, reported it. I'm not a native english speaker but I think this is correct.

debugrevlog: add --dump flag to dump graphable per-revision statistics

revlog: restore the base method

Hardcoding ARCHFLAGS breaks the build against fink's python Fink's python is either i386 or amd64, but not universal. Setting ARCHFLAGS to the empty string produces a successful build against both OS X python and fink python. The modules will no longer be universal -- if that is an issue, we can change the test to extract ARCHFLAGS from distutils.sysconfig and remove ppc if necessary.

debugindex etc.: add --changelog and --manifest options These open the changelog and manifest, respectively, directly so you don't need to specify the path. The options have been added to debugindex, debugdata and debugrevlog. The patch also fixes some minor usage-related bugs.

scmutil: drop aliases in cmdutil for match functions

scmutil: drop some aliases in cmdutil

scmutil: fold in wdutil

scmutil: move revsingle/pair/range from cmdutil This allows users at levels below the command layer to avoid import loops.

help: consolidate topic hooks in help.py This removes loops like cmdutil->revset->help->extensions->cmdutil and simplifies the code.

extensions: move moduledoc to break import loop with help

extensions: drop maxlength from enabled and disabled This is a bad/silly API. Instead calculate maxlength in one place in help it's used and simplify all the callers.

Makefile: remove pure modules on clean

hgcia: Handle URL like in notify (issue2406) The URL must be composed with baseurl and webroot like in notify

util: make str(url) return file:/// for abs paths again str(url) was recently changed to return only file:/. However, the canonical way to represent absolute local paths is file:/// [1], which is also expected by at least hgsubversion. Relative paths are returned as file:the/relative/path. [1] http://en.wikipedia.org/wiki/File_URI_scheme

subrepo: use a safer revision check for hg repos This avoids a traceback if the expected revision has been stripped.

graphlog: use cmdutil.command decorator

purge: use cmdutil.command decorator

patchbomb: use cmdutil.command decorator

transplant: use cmdutil.command decorator

monoblue: improve readability of the filenolink template Whitespace changes only. No difference to the rendered page.

rebase: use cmdutil.command decorator

debugrevlog: many improvements

add debugrevlog command Displays data and statistics about revlogs, including generaldelta details.

check-code: add /= to operator list

order commands alphabetically

revlog: improve delta generation heuristics for generaldelta Without this change, pulls (and clones) into a generaldelta repository could generate very inefficient revlogs, the size of which could be at least twice the original size. This was caused by the generated delta chains covering too large distances, causing new chains to be built far too often. This change addresses the problem by forcing a delta against second parent or against the previous revision, when the first parent delta is in danger of creating a long chain.

keyword: use cmdutil.command decorator

gpg: use cmdutil.command decorator

mq: use cmdutil.command decorator

commands: use a decorator to build table incrementally this allows to define the table entries near the command functions

revlog: fix bug in chainbase cache The bug didn't cause corruption, and thus wasn't caught in hg verify or in tests. It could lead to delta chains longer than normally allowed, by affecting the code that decides when to add a full revision. This could, in turn, lead to performance regression.

setup.py: workaround for missing bz2 module in IronPython IronPython does not provide the bz2 module on its own. This patch skips importing it to allow setup to continue. (minor tweaks by mpm)

httpconnection: fix debug logging option for httpclient

httpclient: import f4c380237fd5 to fix keepalive not working

revlog: add docstring to _addrevision

cmdutil: make_file to makefileobj

cmdutil: make_filename -> makefilename

cmdutil: bail_if_changed to bailifchanged

store: change names to comply with project coding standards as stated at http://mercurial.selenic.com/wiki/CodingStyle (see also http://selenic.com/pipermail/mercurial-devel/2011-May/031139.html ) name changes done at module scope: _build_lower_encodefun -> _buildlowerencodefun _windows_reserved_filenames -> _winreservednames (see 23cd7eeff678) MAX_PATH_LEN_IN_HGSTORE -> _maxstorepathlen DIR_PREFIX_LEN -> _dirprefixlen _MAX_SHORTENED_DIRS_LEN -> _maxshortdirslen (no users of these outside the store module) changed locals: win_reserved -> winreserved space_left -> spaceleft

filelog: add file function to open other filelogs

help: add -c/--command flag to only show command help (issue2799)

help: give hint about 'hg help -e' when appropriate The hint is only given if a command shadows an extension with the same name and when that extension has a multi-line module docstring.

help: add -e/--extension switch to display extension help text

debugbuilddag: read DAG from stdin if not given on cmdline

run-tests: remove check on --shell argument No shell is needed when executing Python tests and finding a good shell might be difficult/impossible on some platforms (e.g., Windows).

subrepo: create subrepos using clone instead of pull Subrepositories used to be created empty and then filled with data using pull. This is wasteful when you do a clone from a local source since it means that no hardlinks are created for the subrepos. This patch make the hgsubrepo._get method check for an empty subrepo and in that case do a clone instead of a pull. This brings in the same data as before, but creates hardlinks when possible.

progress: add speed format This is not enabled by default, but the user can add it by setting progress.format. We might want to let the estimate format output a speed instead of an ETA if there is not total. Currently estimate just returns the empty string in that case.

debugbuilddag: output progress information

test-debugbuilddag: replace unneeded 'rm -rf' with 'rm -r'

hgrc.5: document positional arguments in aliases

merge with stable

hgrc.5: document shell aliases

merge with crew

rename util.is_exec to isexec

posix: delete unused expand_glob

rename util.find_exe to findexe

revlog: support writing generaldelta revlogs With generaldelta switched on, deltas are always computed against the first parent when adding revisions. This is done regardless of what revision the incoming bundle, if any, is deltaed against. The exact delta building strategy is subject to change, but this will not affect compatibility. Generaldelta is switched off by default.

cmdutil: fix errors reported by pyflakes test

merge with stable

check-code: complain about set.isdisjoint

fix bookmarks rollback behavior Before this patch undo.bookmarks was created on bookmarks write and not with other transaction-related files. There were two issues: first is that if you have changed bookmarks few times after a transaction happened, rollback will give you a state which can point to non-existing revision. Second is that if you have not changed bookmarks after a transaction, rollback will touch your state anyway. This change also adds `localrepo._writejournal` method, which can be used by other extensions to save their transaction-related backup in right time.

add positional arguments to non-shell aliases

run-tests: take the basepath when constructing the test temp dir so it's possible to run tests that are outside the test/ dir

run-tests: compare absolute paths in _checkhglib

util: rename _windows_reserved_filenames and _windows_reserved_chars

opener: rename _can_symlink to _cansymlink

patch: make patch()/internalpatch() always update the dirstate

patch: move updatedir() from cmdutil into patch Also, create an artificial wdutil.py to avoid import cycles between patch.py and cmdutil.py.

record: unconditionally update the working dir after patching

mq: explicitly updatedir() even if patch() fails It already works that way in practice, and we intend to merge updatedir() into patch().

mq: allow to qpop/push with a dirty working copy (issue2780) It's safe to do so if the sets of changed files in the working copy and patches are disjoint.

patch: introduce changedfiles returns the set of all changed files in a given patch

debugindex: change output for generaldelta revlogs For generaldelta revlogs, reporting the deltaparent instead of the chain base makes more sense, since that's what's actually stored in the revlog.

revlog: support reading generaldelta revlogs Generaldelta is a new revlog global flag. When it's turned on, the base field of each revision entry holds the deltaparent instead of the base revision of the current delta chain. This allows for great potential flexibility when generating deltas, as any revision can serve as deltaparent. Previously, the deltaparent for revision r was hardcoded to be r - 1. The base revision of the delta chain can still be accessed as before, since it is now computed in an iterative fashion, following the deltaparents backwards.

revlog: calculate base revisions iteratively This is in preparation for generaldelta, where the revlog entry base field is reinterpreted as the deltaparent. For that reason we also rename the base function to chainbase. Without generaldelta, performance is unaffected, but generaldelta will suffer from this in _addrevision, since delta chains will be walked repeatedly. A cache has been added to eliminate this problem completely.

revlog: remove the last bits of punched/shallow Most of it was removed in e7483ec3c374, but a few pieces were accidentally left behind.

util: make readfile() operate in binary mode writefile() and appendfile() are already working that way.

cmdutil: normalize log message eols when reading from file This will be necessary once util.readfile() operates in binary mode. While changelog.add() already normalizes the message, doing so in logmessage() is required as ui.edit() or others expect messages with LF only.

match: make 'listfile:' split on LF and CRLF We want util.readfile() to operate in binary mode, so EOLs have to be handled correctly depending on the platform. It seems both easier and more convenient to treat LF and CRLF the same way on all platforms.

progress: remove useless statement left by bab267e7fc1a

debugbuilddag: simplify tags serialization a bit

httprepo: send 100-continue on POSTs if using http2

url: use new http support if requested by the user The new http library is wired in via an extra module (httpconnection.py), as it requires similar but different plumbing to connect the library to Mercurial's internals and urllib2. Eventualy we should be able to remove all of keepalive.py and its associated tangle in url.py and replace it all with the code in httpconnection.py. To use the new library, set 'ui.usehttp2' to true. The underlying http library uses the logging module liberally, so if things break you can use 'ui.http2debuglevel' to set the log level to INFO or DEBUG to get that logging information (for example, ui.http2debuglevel=info.)

Import new http library as mercurial.httpclient. This is revision a4229f13c374 of http://py-nonblocking-http.googlecode.com/ with a no-check-code comment added to the end of each file using `for fi in $(hg manifest | grep mercurial/httpclient/) ; echo '# no-check-code' >> $fi`.

commands, i18n: add translation hint for split message

mq: don't hide the patch module

patch: remove unused ui arg to iterhunks

mq: loosen except clause when reading patch headers

merge with abuehl

rename util.set_signal_handler to setsignalhandler

rename util.executable_path to executablepath

rename util.os_link to oslink

rename explain_exit to explainexit

rename util.set_binary to setbinary

rename util.set_flags to setflags

rename util.parse_patch_output to parsepatchoutput

rename util.lookup_reg to lookupreg

rename util.set_hgexecutable to _sethgexecutable

rename util.main_is_frozen to mainfrozen

scmutil: rename local function _add_dir_if_not_there

rename scmutil.user_rcpath to userrcpath

rename scmutil.system_rcpath to systemrcpath

rename scmutil.os_rcpath to osrcpath

merge with stable

merge with crew

merge with stable The change from 7b627923739f (parent 2) is not needed after switching to the new url parser.

rename path_auditor to pathauditor The Mercurial 1.9 release is moving a lot of stuff around anyway and we are already moving path_auditor from util.py to scmutil.py for that release. So this seems like a good opportunity to do such a rename. It also strengthens the current project policy to avoid underbars in names.

revlog: use real Booleans instead of 0/1 in nodesbetween

patch: fix clash between local variable and exception instance The local err variable would be bound to PatchError thrown and it keeps this value even after the except block is executed. The whole thing worked anyway since the rejected variable is set in the except block and this makes the function return -1 when a PatchError is thrown.

patchfile: use real Booleans instead of 0/1

annotate: use real Booleans instead of 0/1

hbisect: use real Booleans instead of 0/1

summary: run discovery only once for in/out against same repo Reuses the result of findcommonincoming if default and default-push are the same.

discovery: resurrect findoutgoing as findcommonoutgoing for extension hooks discovery.findoutgoing used to be a useful hook for extensions like hgsubversion. This patch reintroduces this version of findcommonincoming which is meant to be used when computing outgoing changesets.

transplant: remove unneeded loop over repo in revset

transplant: fix revset doc

mq: add a 'mq()' revset predicate that returns applied mq csets

pyflakes: ignore files marked no-check-code

revlog: compute correct deltaparent in the deltaparent function It now returns nullrev for chain base revisions, since they are conceptually deltas against nullrev. The revdiff function was updated accordingly.

changelog: fixes leaked file handle

setdiscovery: limit lines to 80 characters

subrepo: use code from 71ea5b2b9517 only if Python needs it (issue2795) With Python >= 2.6 the original code already works correct, therefore the fix for issue2556 on Python <= 2.5 broke relative subrepositories with newer versions of Python.

sslutil: extracted ssl methods from httpsconnection in url.py This makes it easier to share ssl cert validation with other http implementations.

check-code: fix checking for sh style in .t tests Restore checks after they were disabled by 64de9ca66511.

run-tests: add --shell command line flag This makes it easy to test with, say, both /bin/bash and /bin/dash instead of changing the system-wide /bin/sh.

run-tests: use type of default to convert environment variable Before, defaults could only be integers, they can now also be strings, floats and what have you.

test-treediscovery: fix escaping for both dash and bash A '\n' on the command line is turned into a newline by dash, but kept as-is by bash, which resulted in a syntax error in the config file. Luckily, dash wont turn '\n' into a newline when it is part of a here-doc, so we can write the config file using that technique.

treediscovery: fix regression when run against older repos (issue2793) I ran the entire test suite with "known" and "getbundle" disabled in localrepository. This generated failures because the old findoutgoing had always queried remote's heads explicitly and thus always got them back in the returned heads. treediscovery.findcommonincoming now correctly returns remote's heads in all cases. Also adds a dedicated test for running treediscovery against a pre-getbundle HTTP server.

commands: use util.Abort's hint some more

commands: use double-quotes for strings with single-quotes The globalopts table contain a couple of \' still, but it was normalized like that in 83238c1db6de.

revlog: remove support for punched/shallow The feature was never finished, and there has been restructuring going on since it was added.

revlog: remove support for parentdelta We will introduce a more powerful and general delta concept instead, called generaldelta.

store: break up reference cycle introduced in 9cbff8a39a2a see also 996c1cd8f530

test-https.t: clean up superfluous trailing whitespace

run-tests.py: correctly handle list options with parallel tasks

mq: don't suggest to refresh when qpushing with no applied patches

bundlerepo: fix closing and docstring of getremotechanges I did not in fact implement what I documented about closing of the original repo. And there was a bit of leftover from an older version of the API.

bookmarks: allow deactivating current bookmark with -i

bookmarks: allow to create/move bookmark without making it current (issue2788) Adds a -i/--inactive flag to the bookmarks command to create or move a bookmark without making it the current one. This lets you use a bookmark like a shareable, movable 'tag' and less like a 'branch'. Thanks to Alexander Solovyov <alexander@solovyov.net> for most of the work of implementation, and Matt Mackall <mpm@selenic.com> for suggesting the option name.

tests: test subrepos with ssh urls

tests: share dummyssh

tests: fix test-ssh.t after 6bd9778ae749

fix errors reported by pyflakes test

merge with crew

tests: fix deprecated use of hg debugdata/debugindex For filelogs, debugindex and debugdata can be called with the file name directly instead of the path to the revlog. Since in the future filelogs will no longer be valid revlogs, calling with a path to the revlog is deprecated for debugdata. For debugindex it is expected to still work, but I changed them as well for consistency.

merge with stable

merge with crew

hgext: fixup a couple missed file().read() instances

test-casecollision.t: add test for case changing rename

convert: fix error in git solaris code

test: add pyflakes checking for unable to detect undefined names

tests: add pyflakes checking for assigned to but never used

remove unused variables

test-pyflake: improve sorting algorithm

add make check: tests rule

ui: add configint function and tests

mq: strip extra whitespace from node ids in header (issue2790) This makes `hg qpush --exact` work with patches created by `hg export`.

check-code: disallow calling opener(...).read() and opener(..).write()

prevent transient leaks of file handle by using new helper functions These leaks may occur in environments that don't employ a reference counting GC, i.e. PyPy. This implies: - changing opener(...).read() calls to opener.read(...) - changing opener(...).write() calls to opener.write(...) - changing open(...).read(...) to util.readfile(...) - changing open(...).write(...) to util.writefile(...)

util & scmutil: adapt read/write helpers as request by mpm

store: use a subclass, not a function for fncacheopener

util.termwidth: never return 0 for terminal width Catch a case where the termwidth was being reported as 0 when I was connecting with TLMTask instead of NSTask in OSX. This caused the progress extension to print no progress. The termwidth should never return 0 so in case we would return 0, simply fall back to the default termwidth below which is 80.

discovery: add new set-based discovery Adds a new discovery method based on repeatedly sampling the still undecided subset of the local node graph to determine the set of nodes common to both the client and the server. For small differences between client and server, it uses about the same or slightly fewer roundtrips than the old tree-based discovery. For larger differences, it typically reduces the number of roundtrips drastically (from 150 to 4, for instance). The old discovery code now lives in treediscovery.py, the new code is in setdiscovery.py. Still missing is a hook for extensions to contribute nodes to the initial sample. For instance, Augie's remotebranches could contribute the last known state of the server's heads. Credits for the actual sampler and computing common heads instead of bases go to Benoit Boissinot.

debugbuilddag: use memctx for speed This drops the options to run arbitrary shell commands within commits and the option to create a file that gets appended to in every revision. It now supports to not write file data at all, which is very fast for generating a pure 00changelog.i (useful for discovery tests, for instance). Timings for 1000 linear nodes: * Old `hg debugbuilddag -o '+1000'` took 4.5 secs. * New `hg debugbuilddag -o '+1000'` takes 2 secs. * New `hg debugbuilddag '+1000'` takes 0.8 secs. (The last one creates only 00changelog.i).

localrepo: reuse parent manifest in commitctx if no files have changed This speeds up the in-memory version of debugbuilddag that I'm working on considerably for the case where we want to build just a 00changelog.i (for discovery tests, for instance). There are a couple of test changes because node ids in tests have changed. The changes to the patch names in test-mq-qdelete.t were required because they could collide with nodeid abbreviations and newly actually do (patch "c" collides with id "cafe..." for patch "b").

bundlerepo: fix and improve getremotechanges Fixes the regression where incoming could show local changes introduced by rev 72c84f24b420.

hgrc.5: mention new 'case collision' handling for ui.portablefilenames

win32: remove obsolete bookmarks and parentrevspec extensions from mercurial.ini

bundlerepo: drop struct

bisect: fix indent level

share: create 'hgrc' using an opener, like clone

backout 3a3584967a93 - the "unreachable code" was apparently not unreachable All tests repeatedly passes with 3a3584967a93 on some machines, but on other machines it regularly causes failure in test-mv-cp-st-diff.t, such as: @@ -203,6 +203,7 @@ - working to root: --rev 0 M a + M x/x A b a

merge with mpm

revset: avoid over-aggresive optimizations of non-filtering functions (issue2549) When limit, last, min and max were evaluated they worked on a reduced set in the wrong way. Now they work on an unrestricted set (the whole repo) and get limited later on.

convert: handle invalid subversion source paths

convert: also catch missing revlogs when introduced in repo roots The previous behaviour was almost as if convert.hg.ignoreerrors was always set for revisions without parents, except that errors were silently ignored. Revlog errors are handled as a side effect of getcopies(), but getcopies() was only called when convert.hg.ignoreerrors was set. Now we always call self.getcopies for root revisions, not only when convert.hg.ignoreerrors is set, just like we do on all other revisions. The extra call might be a bit expensive, but the proper fix for that would be to catch these errors in another way.

status: remove unreachable code for status between revisions The manifest value of a file will never be false when "not parentworking", and the expensive content comparision would thus fortunately never be reached. (If it was reached it would be wrong for example in case of renames.) This code once handled status against working directory, but that has been done elsewhere for a long time.

httprepo: proper handling of invalid responses without content-type (issue2019) This can currently be tested on http://sf.net/

httprepo: use the original exception after falling back to static-http failed Before the fine exception handling in httprepo was never shown. The static-http exception will now only be shown when static-http is requested explicitly.

test: test that backslash is preserved by the url class

tests: use raw string for url tests of '\' handling

color: fix using multiple effects

unbundler: separate delta and header parsing Add header parsing for changelog and manifest (currently no headers might change for next-gen bundle).

changegroup: new bundler API

bundlerepo: port to new bundle API

bundler: make parsechunk return the base revision of the delta

tests: add pyflakes checking for unused imports

remove unused imports

scmutil: introduce casecollisionauditor and cleaning up portability functions

check-code: check for repo in revlog and ui in util

check-code: allow for checking multiple pattern sets per file

check-code: adding debug flag

run-tests: add flag to provide extra hgrc options for test runs

graphlog: always sort revisions topologically The grapher cannot really handled revisions if they are not emitted in topological order. The previous 'reverse()' revset was not enough to achieve that and was replaced by an explicit sort call for simplicity. The --limit option is now also handled as usual with cmdutil.loglimit() instead of a 'limit' revset.

graphlog: take the union of --rev specs instead of the intersection

graphmod: correctly emit nodes with more than 2 predecessors The grandparent() function was returning only the closest predecessor of a missing parent while it must return all of them to display a correct ancestry graph.

graphlog: display nodes with more than 2 predecessors While nodes with more than 2 parents do not exist in revision graphs, they do appear when you transform them by removing subgraphs while trying to preserve ancestry links. This code was borrowed from Peter Arrenbrecht <peter.arrenbrecht@gmail.com> pbranch extension.

context: provide an efficient iterator for workingctx This avoids needing to call status or build a synthetic manifest.

merge with stable

merge with nicdumz

tests: export patches only once in test-impexp-branch

tests: do not export several times the same patch/diff in test-import

tests: move rebase-keep-branch into rebase-named-branches It's not worth building a repository just to test keep branch when we can do it our bigger, other, already built repositories.

tests: move testcase from rebase-named-branches to rebase-parameters It saves us a clone of a repository.

tests: change test-rebase-parameters to use bundle/rebase.hg

tests: change test-rebase-detach to use bundle/rebase.hg

tests: simplify test-rebase-named-branches We were cloning a repo to perform a smoke test that could have been performed earlier.

tests: upgrade bundles/rebase.hg to support test-rebase-collapse Add two changesets to the scenario so that the bundle can be reused within three tests. Before: @ 5: 'F' | | o 4: 'E' |/| o | 3: 'D | | | o 2: 'C' |/ | o 1: 'B' |/ o 0: 'A' After: @ 7: 'H' | | o 6: 'G' |/| o | 5: 'F' | | | o 4: 'E' |/ | o 3: 'D' | | | o 2: 'C' | | | o 1: 'B' |/ o 0: 'A' Revisions 0-1 keep the same number/label. Others were translated by an offset of 2 (2.C -> 4.E)

tests: introduce a rebase bundle to use with rebase tests Allows saving a few seconds in test runs by not constructing over and over the same repository.

tests: create a bundle to bootstrap tests using a remote repository It should be faster to use a single common bundle instead of recreating 4 times the same repository manually.

tests: move test bundles in a bundles/ subdirectory

tests: move test-issue436 in test-dispatch where fancyopts is tested

tests: remove test-issue433 as it's covered by test-parents

tests: remove executable bits from unified tests Those files are not executable.

tests: remove test-issue322 as it's fully included in test-issue660

Added signature for changeset 3cb1e95676ad

Added tag 1.8.3 for changeset 3cb1e95676ad

merge with i18n

i18n-pt_BR: synchronized with 42d5165975ad

merge with i18n

i18n-pt_BR: minor spacing fix

i18n-pt_BR: synchronized with 3c753f9a2fbc

zeroconf: notify the Zeroconf threads when hg exits Zeroconf launches two threads in the background, and they wait on Condition objects to exit. We need to call Zeroconf.close() to release those conditions so that threads can gracefully exit. This means that an interrupt on the hg process will now gracefully propagate to the Zeroconf children, fixing that bug which did not allow us to kill an `hg serve` process.

color: use ui.formatted() to test TTYness, not sys.stdout.isatty() This fixes the color extension not working with pager (broken in 877390020477). The pager extension already sets ui.formatted=True to allow this use case.

pushkey: add hooks for pushkey/listkeys

test-patchbomb.t: clean up progress tests

util: really drop size from readfile

util: add readfile() & writefile() helper functions These two functions allow quickly reading or writing a file, without relying on reference counting to close the file handle afterwards.

revset aliases

opener: add read & write utility methods The two new methods are useful for quickly opening a file for reading or writing. Unlike 'opener(...).read()', they ensure they the file is immediately closed without relying on CPython reference counting.

dispatch: handle IndexErrors

color: verify stdout is a tty before using curses Without this change, curses complains when invoked in certain contexts because stdout isn't a tty (such as emacs integration) but we ask it to check for various bits of information from terminfo.

http: minor tweaks to long arg handling x-arg -> x-hgarg replace itertools.count(1)

httprepo: long arguments support (issue2126) Send the command arguments in the HTTP headers. The command is still part of the URL. If the server does not have the 'httpheader' capability, the client will send the command arguments in the URL as it did previously. Web servers typically allow more data to be placed within the headers than in the URL, so this approach will: - Avoid HTTP errors due to using a URL that is too large. - Allow Mercurial to implement a more efficient wire protocol. An alternate approach is to send the arguments as part of the request body. This approach has been rejected because it requires the use of POST requests, so it would break any existing configuration that relies on the request type for authentication or caching. Extensibility: - The header size is provided by the server, which makes it possible to introduce an hgrc setting for it. - The client ignores the capability value after the first comma, which allows more information to be included in the future.

store: rename the 'opener' argument to 'openertype' The 'opener' argument wasn't, in fact, an actual opener instance, but rather something expected to return an opener. The normal argument, from localrepository, is the scmutil.opener type; hence 'openertype'.

statichttprepo: make the opener a subclass of abstractopener

add filteropener abstraction for store openers

opener: introduce an abstact superclass of it Currently, this class doesn't do anything useful.

graphlog: use a set for inclusion test This makes a big difference in performance in the special case where all revisions are being graphed.

graphmod: restore generator nature of dagwalker 9966c95b8c4f introduced the ability to walk the DAG given arbitrary revisions, but changed the behaviour of it to return a list of all nodes (and create a changectx for each one) rather than doing it lazily. This has a pretty significant impact on performance for large repositories (tested on CPython repo, with output disabled): $ time hg glog real 0m2.642s user 0m2.560s sys 0m0.080s Before 9966c95b8c4f: $ time hg glog real 0m0.143s user 0m0.112s sys 0m0.032s And after this fix: $ time hg glog real 0m0.213s user 0m0.184s sys 0m0.028s

graphlog: log -G --follow file does not work, forbid it We do not have revsets to follow file history.

graphlog: unify log -G revset translation

graphlog: fix log -G --prune follow() revset really means '::.' while we want something based on the passed argument. Also, ancestors() revset does not include the parent revisions.

graphlog: escape log -G single value option arguments

graphlog: fix log -G --removed Current version was mispelled and processed like a valued argument.

graphlog: fix log -G option types when translating to revset For instance, --keyword is a multiple value option while it was processed as a single value option.

revset: add tests for multiple and mixed ^ and ~ operators

extensions: obsolete and remove parentrevspec extension As of 305c97670d7a, revsets implements the ^ and ~ operators that this extension provides, so it's no longer necessary.

wix: a new mercurial.js has replaced graph.js

util.url: copy urllib.unquote() into util to improve startup times The ui class uses util.hasscheme() in a couple of places, causing hg to import urllib even when it doesn't need to. This copies urllib.unquote() to avoid that import. perfstartup time before the URL refactoring (8796fb6af67e): ! wall 0.050692 comb 0.000000 user 0.000000 sys 0.000000 (best of 100) before this change: ! wall 0.064742 comb 0.000000 user 0.000000 sys 0.000000 (best of 100) after this change: ! wall 0.052126 comb 0.000000 user 0.000000 sys 0.000000 (best of 100

url: move URL parsing functions into util to improve startup time The introduction of the new URL parsing code has created a startup time regression. This is mainly due to the use of url.hasscheme() in the ui class. It ends up importing many libraries that the url module requires. This fix helps marginally, but if we can get rid of the urllib import in the URL parser all together, startup time will go back to normal. perfstartup time before the URL refactoring (8796fb6af67e): ! wall 0.050692 comb 0.000000 user 0.000000 sys 0.000000 (best of 100) current startup time (139fb11210bb): ! wall 0.070685 comb 0.000000 user 0.000000 sys 0.000000 (best of 100) after this change: ! wall 0.064667 comb 0.000000 user 0.000000 sys 0.000000 (best of 100)

revlog: introduce _chunkbase to allow filelog to override Used by revlog.revision to retrieve the base-chunk in a delta chain.

filelog: extract metadata parsing and packing _parsemeta returns the dictionary and a list of keys in the order they appear in metadata. This can be used to repack the dictionary in the same order. _packmeta creates metadata from a dictionary and an optional key-order list. In _parsemeta, we use slices and re.search indead of str.index so we can accept both buffers and strings.

discovery: drop findoutgoing and simplify findcommonincoming's api This is a long desired cleanup and paves the way for new discovery. To specify subsets for bundling changes, all code should use the heads of the desired subset ("heads") and the heads of the common subset ("common") to be excluded from the bundled set. These can be used revlog.findmissing instead of revlog.nodesbetween. This fixes an actual bug exposed by the change in test-bundle-r.t where we try to bundle a changeset while specifying that said changeset is to be assumed already present in the target. This used to still bundle the changeset. It no longer does. This is similar to the bugs fixed by the recent switch to heads/common for incoming/pull.

revset: add missing whitespace

url: remove unused/obsolete functions

revset: add ^ and ~ operators from parentrevspec extension ^ (Nth parent) and ~ (Nth first ancestor) are infix operators that match certain ancestors of the set: set^0 the set set^1 (also available as set^) the first parent of every changeset in set set^2 the second parent of every changeset in set set~0 the set set~1 the first ancestor (i.e. the first parent) of every changeset in set set~2 the second ancestor (i.e. first parent of first parent) of every changeset in set set~N the Nth ancestor (following first parents only) of every changeset in set; set~N is equivalent to set^1^1..., with ^1 repeated N times.

encoding: add an encoding-aware lower function

add: notify when adding a file that would cause a case-folding collision On a case-sensitive file system, files can be added with names that differ only in case (a "case collision"). This would cause an error on case-insensitive filesystems. A warning or error is now given for such collisions, depending on the value of ui.portablefilenames ('warn', 'abort', or 'ignore'): $ touch file File $ hg add --config ui.portablefilenames=abort File abort: possible case-folding collision for File $ hg add File warning: possible case-folding collision for File

scmutil: refactor ui.portablefilenames processing The ui.portablefilenames config handling is generally useful for notifying the user of various portability problems.

pure bdiff: don't use a generator Generators are slow, and currently defeat the PyPy JIT.

pure mpatch: avoid using list.insert(0, ...) In Python lists are implemented as arrays with overallocation. As a result, list.insert(0, ...) is O(n), whereas list.append() has an amortised running time of O(1). Reversing the internal representation of the list should cause a slight speedup for pure Python builds.

remove unused imports and variables

Zeroconf: catch both ValueError and KeyError in get() This prevents a traceback during discovery of available hosts.

run-tests: print a new line before writing the diff

revsets: add a last function last(set, n) = reverser(limit(reverse(set), n))

bundle: more comments about the different header types, remove useless if

merge with mpm

hgweb: initialize permhooks at definition time This is simpler than creating it empty and then appending the default checkauthz hook 50 lines below.

revset: fix undefined name ParseError

branchcache: improve speed relative to the amount of heads Updating the branch cache is quadratic to the amount of heads in the repository. One consequence of this was that cloning a pathological repository with 10,000 heads (and nothing else) took hours of CPU time. This patch makes one of the inner loop much faster, by removing a changectx instantiation, and removes another entirely in cases where there are no candidate branch heads which descend from other branch heads.

drop {short,hex}(ctx.node()) calls in favor of ctx methods

mq: be more explicit on invalid patch name message

fix broken tests test-debugcomplete.t broken by 58e58406ed19 test-highlight.t broken by b24e5a708fad

merge with stable

mq: add '.' and '..' to list of forbidden patch names When an empty string is being passed to normname it would return '.' causing checkfile() to always return that a patch with that name exists.

subrepo: handle svn tracked/unknown directory collisions This happens more often than expected. Say you have an svn subrepository with python code. Python would have generated unknown .pyc files. Now, you rebase this setup on a revision where a directory containing python code does not exist. Subversion is first asked to remove this directory when updating, but will not because it contains untracked items. Then it will have to bring back the directory after the merge but will fail because it now collides with an untracked directory. Using --force is not very elegant and only works with svn >= 1.5 but the only alternative I can think of is to write our own purge command for subversion.

subrepo: backout bcc6ed0f6c3b

wireproto: add test for new optional arg missing on server New argument is silently ignored by both HTTP and SSH servers. This means we can, for instance, add new flags to getbundle() to request advanced features (like lightweight-copy-aware bundles), and older servers will silently ignore this request and send back a plain bundle.

merge with stable

web: Made elapsed time calculation dynamic (javascript). This allow safe caching of the pages by the browser and still display the right amount of elapsed time upon page refresh. If javascript is disabled, absolute time is displayed, leaving it readable. All the templates have been updated.

extdiff: add repository root as a variable Some external diff tools (notably Plan 9 diff(1)) require the absolute path to the file being diffed for proper function. A root variable was added to inform an external tool of the repository root (the tool is invoked with the cwd set to tmproot).

help: move hgignore man page into built-in help (issue2769)

graphlog: make use of graphmod's revset support

graphmod: use revsets internally Thanks for the idea and most of the implementation to Klaus Koch Backs revisions() and filerevs() with DAG walker which can iterate through arbitrary list of revisions instead of strict one by one iteration from start to stop. When a gap occurs in a revisions (i.e. in file log), the next topological parent within the revset is searched and the connection to it is printed in the ascii graph. File graph can draw sometimes more connections than previous version, because graph is produced according to the revset, not according to a file's filelog. In case the graph contains several branches where the left parent is null, the graphs for each are printed sequentially, not in parallel as it was a case earlier (see for example the graph for README in hg-dev).

svn subrepo: attempt work around obstructed checkouts (issue2752) It should be possible to do better than this with 'svn switch', but the logic required woud be significantly more complex. Until someone needs the performance improvements of using switch, we'll just use the same strategy for everything.

churn: strip key earlier to avoid false negative seach in aliases

bookmarks: further restrict IOError on write This won't risk losing the undo file when the error was something other than file not found

tags: loosen IOError filtering when reading localtags

run-tests: don't count test as succeeded if it failed regressed around ec4ae5727f07

changegroup: do not count closed new heads (issue2697) If a closed head gets pulled, we currently see (example): $ hg pull pulling from $TESTTMP/repo2 searching for changes adding changesets adding manifests adding file changes added 2 changesets with 1 changes to 1 files (+1 heads) (run 'hg heads' to see heads, 'hg merge' to merge) A subsequent 'hg heads' doesn't show that head because it is closed. This patch improves the UI response texts for that same use case to: $ hg pull pulling from $TESTTMP/repo2 searching for changes adding changesets adding manifests adding file changes added 2 changesets with 1 changes to 1 files (run 'hg update' to get a working copy) That is, the part "(+1 heads)" is not shown in that case any longer.

merge with crew

shrink-revlog: defend against null first parents

test-contrib: add a test for shrink-revlog We only verify the extension runs and that verify passes at the end as constructing a test repository that would result in a shrink would be nontrivial.

test-simplemerge-cmd.t: move all tests into test-contrib.t

Move test-dumprevlog to test-contrib in anticipation of more contrib tests

shrink-revlog: use a bundler object (see d69c9510d648)

shrink-revlog: update util.opener to scmutil.opener after d13913355390

shrink-revlog: remove \ from docstring

bookmarks: be more restrictive in our Exception catching

strip: make it clear that --force discards changes (issue310)

subrepo: tell Subversion when we are non-interactive (issue2759) $ hg clone repo repo-clone -v updating to branch default resolving manifests getting .hgsub getting .hgsubstate abort: svn: OPTIONS of 'https://subversion.srv/project': Server certificate verification failed: issuer is not trusted (https://subversion.srv)

merge with stable

tests: check for (*/sec) instead of (*B/sec) to match (X bytes/sec) I just had: transferred 606 bytes in 2.9 seconds (211 bytes/sec)

test-repair-strip: turn into new style test

extdiff: fix broken symlinks handling (issue1909)

tags: catch more corruption during cache parsing (issue2779)

run-tests: do chdir for tests under a lock for thread safety

run-tests: add iolock to vlog

patch: remove redundant variable in iterhunks()

merge with stable

merge default heads

commit: improve --date help text

merge with stable

test-transplant: fix missing file addition

tests: add script to disable a selected wire protocol capability This can be used to test that the client and server both fall back to the previous wire protocol when the capability is not supported. It is currently used by test-push-http.t and I plan to use it for the HTTP long argument support tests.

mq: prevent traceback when qfinish patches not in series. When mq status entry referencing a patches that is not in series `hg qfinish -a` used to issue a traceback. This states is inconsistent but might happen regularly when people misuse hg up -mq. This changeset prevent hg from crashing. The faulty entry is finished anyway and a warning is issued.

check-code: separate warnings to avoid repetitive str.startswith

atomictempfile: rewrite docstring to clarify rename() vs. close().

atomictempfile: avoid infinite recursion in __del__(). The problem is that a programmer using atomictempfile directly can make an innocent everyday mistake -- not enough args to the constructor -- which escalates badly. You would expect a simple TypeError crash in that case, but you actually get an infinite recursion that is surprisingly difficult to kill: it happens between __del__() and __getattr__(), and Python does not handle infinite recursion from __del__() well. The fix is to not implement __getattr__(), but instead assign instance attributes for the methods we wish to delegate to the builtin file type: write() and fileno(). I've audited mercurial.* and hgext.* and found no users of atomictempfile using methods other than write() and rename(). I audited third-party extensions and found one (snap) passing an atomictempfile to util.fstat(), so I also threw in fileno(). The last time I submitted a similar patch, Matt proposed that we make atomictempfile a subclass of file instead of wrapping it. Rejected on grounds of unnecessary complexity: for one thing, it would make the Windows implementation of posixfile quite a bit more complex. It would have to become a subclass of file rather than a simple function -- but since it's written in C, this is non-obvious and non-trivial. Furthermore, there's nothing wrong with wrapping objects and delegating methods: it's a well-established pattern that works just fine in many cases. Subclassing is not the answer to all of life's problems.

run-tests: display diff before prompting with --interactive

check-code: warn about naked except clauses

eliminate various naked except clauses

run-tests: fix some missing i/o locks

run-tests: add a lock for console I/O

run-tests: switch timeout handling from alarm to helper thread This should be slightly more portable than signals and be compatible with threaded dispatch.

run-tests: add locking on results struct

run-tests: fix --interactive (after 994ad067ac6e) - Do not prompt if there is no .err file - Fix source and target paths

color: code simplification

test-status-color: fix terminfo code compatibility problem On a linux host in xterm mode, 'none' is translated to: \x1b[m\x1b(B While it is the following on osx: \x1b(B\x1b[m Take advantage of the new color.color.X option to force it to a common value.

hg-ssh: fix duplicate word in docstring

run-tests: move test loop into a helper function

run-tests: use a results dict

run-tests: move blacklist and retest filtering to runone

run-tests: keep a list of passed tests

run-tests: move keyword checking into runone

run-tests: add ignores list

run-tests: move existence/name format check into runone

run-tests: move interactive handling into runone

color: add support for terminfo-based attributes and color Using terminfo instead of hard-coding ECMA-48 control sequences provides a greater assurance that the terminal codes are correct for the current terminal type; not everything supports the ANSI escape codes. It also allows us to use a wider range of colors when a terminal emulator supports it (such as 16- or 256-color xterm), and a few more non-color attributes, such as the ever-popular blink.

move system_rcpath and user_rcpath to scmutil

move os_rcpath from util to scmutil

move rcpath from util to scmutil

merge with stable

merge with i18n

merge with i18n

i18n-pt_BR: synchronized with 1416b9118540

i18n: merge with stable

i18n-da: synchronize with 1184bb274cb3

i18n-de: improve grammar

win32: Wine doesn't know about hardlinks

move walkrepos from util to scmutil

move checkfilename from util to scmutil checkfilename is specific to Mercurial, since it contains the knowledege that Mercurial can't track files with \n or \r in the name.

scmutil: fix erroneous Abort call This fixes d13913355390 (affected Windows only).

move path_auditor from util to scmutil

move canonpath from util to scmutil

move opener from util to scmutil

merge with stable

convert: make filemap prune useless branch closing revs (issue2774) A branch closing revision only applies if one of its parents belongs to the branch being closed. Otherwise the filemap can prune it too.

dispatch: improve repository not found message This should help clarify what the problem is in various problematic cases like ssh.

hgweb: support disabling page cache By default, hgweb_mod supports caching via the ETag header. This can cause some confusion with browsers which cache aggressively. This change preserves existing behavior while giving the administrator a knob to disable the ETag header.

merge with stable

hgweb: support alternate logo url Clicking on the logo image/text in the hgweb interface brings the user to the Mercurial project page. The majority of users expect that this would bring them to the top level index. I have added a new template variable named `logourl' which allows an administrator to change this behavior. To stay compatible with existing behavior, `logourl' will default to http://mercurial.selenic.com/. This change is very useful in large installations where jumping to the index is common.

identify: further clarification of help Following on to 1416b9118540, correct a minor grammar issue, re-wrap to 72 columns, and further clarify wording.

add: introduce a warning message for non-portable filenames (issue2756) (BC) On POSIX platforms, the 'add', 'addremove', 'copy' and 'rename' commands now warn if a file has a name that can't be checked out on Windows. Example: $ hg add con.xml warning: filename contains 'con', which is reserved on Windows: 'con.xml' $ hg status A con.xml The file is added despite the warning. The warning is ON by default. It can be suppressed by setting the config option 'portablefilenames' in section 'ui' to 'ignore' or 'false': $ hg --config ui.portablefilenames=ignore add con.xml $ hg sta A con.xml If ui.portablefilenames is set to 'abort', then the command is aborted: $ hg --config ui.portablefilenames=abort add con.xml abort: filename contains 'con', which is reserved on Windows: 'con.xml' On Windows, the ui.portablefilenames config setting is irrelevant and the command is always aborted if a problematic filename is found.