largefiles: check existence of the file with case awareness of the filesystem Before this patch, largefiles extension always unlinks largefiles untracked on the target context in merging/updating after updating working directory. For example, it is assumed that the revision X consists of ".hglf/A" (and "A" implicitly) and revision Y consists of "a" (not ".hglf/A"). In the case of updating from X to Y, largefiles extension tries to unlink "A" after updating "a" in working directory. This causes unexpected unlinking "a" on the case insensitive filesystem. This patch checks existence of the file in the working context with case awareness of the filesystem to prevent from such unexpected unlinking. "lfcommands._updatelfile()" also unlinks target file in the case "largefile is tracked in the target context, but fails to be fetched". This patch doesn't apply "repo.dirstate.normalize()" in this case, because it should be already ensured in the manifest merging that there is no normal file colliding against any largefiles.

windows: check target type before actual unlinking to follow POSIX semantics Creation and writing into target file via vfs (a.k.a opener) is done after "unlink()" target file, if it exists. For example, it is assumed that the revision X consists of file 'A', and the revision Y consists of file 'A/B'. Merging revision X into Y tries to "unlink()" on directory 'A' of 'A/B', before creation of file 'A'. On POSIX environment, directories should be removed by "rmdir(2)", and "unlink(2)" on directories fails. "unlink()" of Mercurial (and Python) uses "unlink(2)" directly, so unlinking in the merge case above would fail. In the other hand, on Windows environment, "unlink()" of Mercurial tries to rename before actual unlinking, to follow POSIX semantics: already opened file can be unlinked safely. This causes unexpected success in unlinking in the merge case above, even though directory 'A' is renamed to another. This confuses users. This patch checks whether target is directory or not before renaming, and raises IOError(errno.EPERM) if so, to follow POSIX semantics.

Merge with stable.

manifestmerge: local unknown, remote created: don't traverse symlinks To figure out what to do with locally unknown files, Mercurial attempts to read them if they exist. When an attempt is made to read a file that exists but traverses a symlink, Mercurial aborts. With this patch, we first ensure that the file doesn't traverse a symlink before opening it. This is fine because a file being "remote created" means the symlink doesn't exist remotely, which means it will be deleted in the apply phase.

subrepo: open files in 'rb' mode to read exact data in (issue3926) Before this patch, "subrepo._calcfilehash()" opens files by "open()" without any mode specification. This implies "text mode" on Windows. When target file contains '\x00' byte, "read()" in "text mode" reads file contents in without data after '\x00'. This causes invalid SHA1 hash calculation in "subrepo._calcfilehash()". This patch opens files in 'rb' mode to read exact data in.

patch: use scmutil.marktouched instead of scmutil.addremove addremove required paths relative to the cwd, which meant a lot of extra code that transformed paths into relative ones. That code is now gone as well.

scmutil: add a function to mark that files have been operated on Several places use scmutil.addremove as a means to declare that certain files have been operated on. This is ugly because: - addremove takes patterns relative to the cwd, not paths relative to the root, which means extra contortions for callers. - addremove doesn't make clear what happens to files whose status hasn't changed. This new method accepts filenames relative to the repo root, and has a much clearer contract. It also allows future modifications that do more with files whose status hasn't changed.

scmutil.addremove: factor out code to mark added/removed/renames An upcoming patch will reuse this code in another function.

scmutil.addremove: factor out code to find renames This code will be used in a different context in upcoming patches.

scmutil.addremove: rename local 'copies' to 'renames' An upcoming patch will refactor some code out into a method called _findrenames. Having a line saying "copies = _findrenames..." is confusing. Besides, 'renames' is a more precise name for this local anyway.

scmutil.addremove: factor out dirstate walk into another function Upcoming patches will reuse and expand on this function for other purposes.

filecontext: use 'is not None' to check for filelog existence Previously we used 'if filelog:' to check if the filelog existed. If the instance did exist, this pattern then calls len() on the filelog to see if it is empty. I'm developing a filelog replacement that doesn't have len() implemented, so it's better to do an explicit 'is not None' check here instead. Also change _changeid() to return the _changeid attribute if it has it. Previously it would try to obtain it from the _changectx(), and if that did not exist it would construct the _changectx() using the linkrev. In the extension I'm working on, filectx's don't have easy access to linkrevs so avoiding this when possible is better.

filelog: use super() for calling base functions filelog had some hardcoded revlog.revlog.foo() calls. This changes it to use super() instead so that extensions can replace the filelog base class.

bookmarks: allow bookmark command to take multiple arguments This change allows setting or deleting multiple bookmarks at once. If more than one is being set and --inactive is not given, the first one is made active.

merge with stable

convert: fix bug of wrong CVS path parsing without port number (issue3678) The cvsps.py:getrepopath suffers from a string parsing bug (it returns "user@server/path/to/repository" if the CVSROOT is given like this: ":pserver:user@server/path/to/repository" ), which gives returnes the wrong value becouse cvsps.py fails to strip the prefix from filenames. With this patch for the same input we get the correct repo path that is: "/path/to/repository"

match: add comments to explain explicitdir and traversedir

match: make explicitdir and traversedir None by default With this, extensions can easily tell when traversedir and/or explicitdir don't need to be called.

dirstate.walk: cache match.explicitdir and traversedir locally

match: drop dir callback dir has been subsumed by explicitdir and traversedir.

inotify: call match.explicitdir inotify only returns explicit directories matched.

purge: hook into match.explicitdir and traversedir

localrepo.commit: hook into match.explicitdir

dirstate.walk: call match.explicitdir or traversedir as appropriate

match: introduce explicitdir and traversedir match.dir is currently called in two different places: (1) noting when a directory specified explicitly is visited. (2) noting when a directory is visited during a recursive walk. purge cares about both, but commit only cares about the first. Upcoming patches will split the two cases into two different callbacks. Why bother? Consider a hypothetical extension that can provide more efficient walk results, via e.g. watching the filesystem. That extension will need to fall back to a full recursive walk if a callback is set for (2), but not if a callback is only set for (1).

merge with stable

run-tests: when running in parallel, delete tmpdirs immediately This is especially important if extensions that use inotify are enabled, because it's very easy to hit the inotify max_user_instances limit without this.

tests: check path separator in moves