revset: rename getkwargs() to getargsdict() This function was added recently at 48919d246a47, but its name was misleading because it processes both positional and keyword arguments.

revset: work around x:y range where x or y is wdir() All revisions must be contiguous in spanset, so we need the special case for the wdir revision.

revset: use integer representation of wdir() in revset This is the simplest way to handle wdir() revision in revset. None didn't work well because revset heavily depends on integer operations such as min(), max(), sorted(), x:y, etc. One downside is that we cannot do "wctx.rev() in set" because wctx.rev() is still None. We could wrap the result set by wdirproxyset that translates None to wdirrev, but it seems overengineered at this point. result = getset(repo, subset, tree) if 'wdir' in funcsused(tree): result = wdirproxyset(result) Test cases need the '(all() + wdir()) &' hack because we have yet to fix the bootstrapping issue of null and wdir.

localrepo: provide workingctx by integer revision This allows us to use the integer representation in revset. None doesn't work well while computing revset because revset heavily depends on and optimized for integer revisions. Still repo[wdirrev].rev() is None, which means the canonical form of the working-directory revision is None. This patch doesn't add the case for the wdirid because we can't handle short and ambiguous identifiers here. Perhaps, the wdirid will have to be handled in the changelog layer.

changeset_printer: change flush() to accept ctx instead of rev Because flush() is the function to write data buffered by show(ctx), flush(ctx) is more consistent than flush(rev). This makes sure that buffered header and hunk are always keyed by ctx.rev(). This patch will allow us to give an integer to the wdir while keeping wctx.rev() -> None.

changeset_printer: display wdirrev/wdirnode values for workingctx Because we want to eliminate "if"s in the default template, it makes sense to display wdirrev/wdirnode values for now. wdir() is still experimental, so the output of "log -r'wdir()'" may change in future.

hg: support for auto sharing stores when cloning Many 3rd party consumers of Mercurial have created wrappers to essentially perform clone+share as a single operation. This is especially popular in automated processes like continuous integration systems. The Jenkins CI software and Mozilla's Firefox release automation infrastructure have both implemented custom code that effectively perform clone+share. The common use case here is that clients want to obtain N>1 checkouts while minimizing disk space and network requirements. Furthermore, they often don't care that a clone is an exact mirror of a remote: they are simply looking to obtain checkouts of specific revisions. When multiple third parties implement a similar feature, it's a good sign that the feature is worth adding to the core product. This patch adds support for an easy-to-use clone+share feature. The internal "clone" function now accepts options to control auto sharing during clone. When the auto share mode is active, a store will be created/updated under the base directory specified and a new repository pointing to the shared store will be created at the path specified by the user. The share extension has grown the ability to pass these options into the clone command/function. No command line options for this feature are added because we don't feel the feature will be popular enough to warrant their existence. There are two modes for auto share mode. In the default mode, the shared repo is derived from the first changeset (rev 0) in the remote repository. This enables related repositories existing at different URLs to automatically use the same storage. In environments that operate several repositories (separate repo for branch/head/bookmark or separate repo per user), this has the potential to drastically reduce storage and network requirements. In the other mode, the name is derived from the remote's path/URL.

merge with stable

cmdutil: apply dirstate.normallookup on (maybe partially) committed files To detect change of a file without redundant comparison of file content, dirstate recognizes a file as certainly clean, if: (1) it is already known as "normal", (2) dirstate entry for it has valid (= not "-1") timestamp, and (3) mode, size and timestamp of it on the filesystem are as same as ones expected in dirstate This works as expected in many cases, but doesn't in the corner case that changing a file keeps mode, size and timestamp of it on the filesystem. The timetable below shows steps in one of typical such situations: ---- ----------------------------------- ---------------- timestamp of "f" ---------------- dirstate file- time action mem file system ---- ----------------------------------- ---- ----- ----- N *** *** - change "f" N - execute 'hg commit -i' - backup "f" with timestamp N - revert "f" by 'merge.update()' N with 'partially' - apply selected hunks N by 'patch.patch()' - 'repo.commit()' - 'dirstate.normal("f")' N N+1 - 'dirstate.write()' N N - restore "f" N+1 - restore timestamp of "f" N - 'hg status' shows "f" as "clean" N N N ---- ----------------------------------- ---- ----- ----- The most important point is that 'dirstate.write()' is executed at N+1 or later. This causes writing dirstate timestamp N of "f" out successfully. If it is executed at N, 'parsers.pack_dirstate()' replaces timestamp N with "-1" before actual writing dirstate out. This issue can occur when 'hg commit -i' satisfies conditions below: - the file is committed partially, and - mode and size of the file aren't changed before and after committing The root cause of this issue is that (maybe partially changed) files are restored with original timestamp but dirstate isn't updated for them. To detect changes of files correctly, this patch applies 'dirstate.normallookup()' on restored files. Status check is needed before 'dirstate.normallookup()', because status other than "n(ormal)" should be kept at failure of committing. This patch doesn't examine whether each files are committed fully or partially, because interactive hunk selection makes it difficult. After this change, timetable is changed as below: ---- ----------------------------------- ---------------- timestamp of "f" ---------------- dirstate file- time action mem file system ---- ----------------------------------- ---- ----- ----- N *** *** - change "f" N - execute 'hg commit -i' - backup "f" with timestamp N - revert "f" by 'merge.update()' N with 'partially' - apply selected hunks N by 'patch.internalpatch()' - 'repo.commit()' - 'dirstate.normal("f")' N N+1 - 'dirstate.write()' N N - restore "f" N+1 - restore timestamp of "f" N ----------------------------------- ---- ----- ----- - normallookup("f") -1 - release wlock - 'dirstate.write()' -1 -1 N ----------------------------------- ---- ----- ----- - 'hg status' shows "f" as "clean" -1 -1 N ---- ----------------------------------- ---- ----- ----- To reproduce this issue in tests certainly, this patch emulates some timing critical actions as below: - change "f" at N 'touch -t 200001010000' before command invocation changes mtime of "f" to "2000-01-01 00:00" (= N). - apply selected hunks at N 'patch.internalpatch()' with 'fakepatchtime.py' explicitly changes mtime of patched files to "2000-01-01 00:00" (= N). - 'dirstate.write()' at N+1 (or "not at N") 'pack_dirstate()' uses actual timestamp at runtime as "now", and it should be different from the "2000-01-01 00:00" of "f". BTW, in 'test-commit-interactive.t', files are sometimes treated as modified , even though they are just committed fully via 'hg commit -i' and 'hg diff' shows nothing for them. Enabling win32text causes EOL style mismatching below: - files are changed in LF style EOL => files restored after committing uses LF style EOL (1) - 'merge.update()' reverts files in CRLF style EOL - 'patch.internalpatch()' changes files in CRLF style EOL => 'dirstate.normal()' via 'repo.commit()' uses the size of files in CRLF style EOL (2) Therefore, fully committed files are treated as "modified", because 'lstat()' returns size of (1) restored files in LF style EOL, but dirstate expects size of (2) committed files in CRLF style EOL. After this patch, 'dirstate.normallookup()' on committed files forces subsequent 'hg status' to examine changes exactly, and fully committed files are treated as clean as expected. This is reason why this patch also does: - add some 'hg status' checking status of fully committed files - clear win32text configuration before size/timestamp sensitive examination

cmdutil: put recordfunc invocation into wlock scope for consistency Before this patch, 'recordfunc()' for interactive hunk selection does below outside wlock scope at 'hg commit -i' and so on: - backup files, which may be partially changed - apply selected hunks on files - restore files from backup-ed ones These should be executed inside wlock scope for consistency. To put them into wlock scope without largely changing indents in 'recordfunc()', this patch adds another wrapper function. This patch is also a preparation for subsequent patch fixing the issue to correctly recognize partially committed files as "modified".

context: write dirstate out explicitly at the end of markcommitted To detect change of a file without redundant comparison of file content, dirstate recognizes a file as certainly clean, if: (1) it is already known as "normal", (2) dirstate entry for it has valid (= not "-1") timestamp, and (3) mode, size and timestamp of it on the filesystem are as same as ones expected in dirstate This works as expected in many cases, but doesn't in the corner case that changing a file keeps mode, size and timestamp of it on the filesystem. The timetable below shows steps in one of typical such situations: ---- ----------------------------------- ---------------- timestamp of "f" ---------------- dirstate file- time action mem file system ---- ----------------------------------- ---- ----- ----- * *** *** - 'hg transplant REV1 REV2 ...' - transplanting REV1 .... N - change "f", but keep size N (via 'patch.patch()') - 'dirstate.normal("f")' N *** (via 'repo.commit()') - transplanting REV2 - change "f", but keep size N (via 'patch.patch()') - aborted while patching N+1 - release wlock - 'dirstate.write()' N N N - 'hg status' shows "r1" as "clean" N N N ---- ----------------------------------- ---- ----- ----- The most important point is that 'dirstate.write()' is executed at N+1 or later. This causes writing dirstate timestamp N of "f" out successfully. If it is executed at N, 'parsers.pack_dirstate()' replaces timestamp N with "-1" before actual writing dirstate out. This issue can occur when 'hg transplant' satisfies conditions below: - multiple revisions to be transplanted change the same file - those revisions don't change mode and size of the file, and - the 2nd or later revision of them fails after changing the file The root cause of this issue is that files are changed without flushing in-memory dirstate changes via 'repo.commit()' (even though omitting 'dirstate.normallookup()' on files changed by 'patch.patch()' for efficiency also causes this issue). To detect changes of files correctly, this patch writes in-memory dirstate changes out explicitly after marking files as clean in 'committablectx.markcommitted()', which is invoked via 'repo.commit()'. After this change, timetable is changed as below: ---- ----------------------------------- ---------------- timestamp of "f" ---------------- dirstate file- time action mem file system ---- ----------------------------------- ---- ----- ----- * *** *** - 'hg transplant REV1 REV2 ...' - transplanting REV1 .... N - change "f", but keep size N (via 'patch.patch()') - 'dirstate.normal("f")' N *** (via 'repo.commit()') ----------------------------------- ---- ----- ----- - 'dirsttate.write()' -1 -1 ----------------------------------- ---- ----- ----- - transplanting REV2 - change "f", but keep size N (via 'patch.patch()') - aborted while patching N+1 - release wlock - 'dirstate.write()' -1 -1 N - 'hg status' shows "r1" as "clean" -1 -1 N ---- ----------------------------------- ---- ----- ----- To reproduce this issue in tests certainly, this patch emulates some timing critical actions as below: - change "f" at N 'patch.patch()' with 'fakepatchtime.py' explicitly changes mtime of patched files to "2000-01-01 00:00" (= N). - 'dirstate.write()' via 'repo.commit()' at N 'fakedirstatewritetime.py' forces 'pack_dirstate()' to use "2000-01-01 00:00" as "now", only if 'pack_dirstate()' is invoked via 'committablectx.markcommitted()'. - 'dirstate.write()' via releasing wlock at N+1 (or "not at N") 'pack_dirstate()' via releasing wlock uses actual timestamp at runtime as "now", and it should be different from the "2000-01-01 00:00" of "f". BTW, this patch doesn't test cases below, even though 'patch.patch()' is used similarly in these cases: 1. failure of 'hg import' or 'hg qpush' 2. success of 'hg import', 'hg qpush' or 'hg transplant' Case (1) above doesn't cause this kind of issue, because: - if patching is aborted by conflicts, changed files are committed changed files are marked as CLEAN, even though they are partially patched. - otherwise, dirstate are fully restored by 'dirstateguard' For example in timetable above, timestamp of "f" in .hg/dirstate is restored to -1 (or less than N), and subsequent 'hg status' can detect changes correctly. Case (2) always causes 'repo.status()' invocation via 'repo.commit()' just after changing files inside same wlock scope. ---- ----------------------------------- ---------------- timestamp of "f" ---------------- dirstate file- time action mem file system ---- ----------------------------------- ---- ----- ----- N *** *** - make file "f" clean N - execute 'hg foobar' .... - 'dirstate.normal("f")' N *** (e.g. via dirty check or previous 'repo.commit()') - change "f", but keep size N - 'repo.status()' (*1) (via 'repo.commit()') ---- ----------------------------------- ---- ----- ----- At a glance, 'repo.status()' at (*1) seems to cause similar issue (= "changed files are treated as clean"), but actually doesn't. 'dirstate._lastnormaltime' should be N at (*1) above, because 'dirstate.normal()' via dirty check is finished at N. Therefore, "f" changed at N (= 'dirstate._lastnormaltime') is forcibly treated as "unsure" at (*1), and changes are detected as expected (see 'dirstate.status()' for detail). If 'hg import' is executed with '--no-commit', 'repo.status()' isn't invoked just after changing files inside same wlock scope. But preceding 'dirstate.normal()' is invoked inside another wlock scope via 'cmdutil.bailifchanged()', and in-memory changes should be flushed at the end of that scope. Therefore, timestamp N of clean "f" should be replaced by -1, if 'dirstate.write()' is invoked at N. It means that condition of this issue isn't satisfied.

tests: add extension to emulate invoking internalpatch at the specific time This extension fakes "mtime" of patched files to emulate invoking 'patch.internalpatch()' at the specific time. This is useful to reproduce timing critical issues fixed in subsequent patches.

cmdutil: remove useless dirstate.normallookup() invocation in revert() Explicit 'dirstate.normallookup()' invocation in 'revert()' is useless now, because previous patch fixed the relevant issue by writing in-memory dirstate changes out at the end of dirty check. 'dirstate.normallookup()' invocation was introduced by 21b33f0460e0 to avoid occasional test failure (see issue4583 for detail). This is partial backout of it (added tests are still left).

merge: remove useless dirstate.normallookup() invocation in applyupdates() Explicit 'dirstate.normallookup()' invocation via 'dirtysubstate()' in 'applyupdates()' is useless now, because previous patch fixed the relevant issue by writing in-memory dirstate changes out at the end of dirty check. 'dirstate.normallookup()' invocation was introduced by 6becb9dbca25 to avoid occasional test failure. This is partial backout of it (added tests are still left).

context: write dirstate out explicitly after marking files as clean To detect change of a file without redundant comparison of file content, dirstate recognizes a file as certainly clean, if: (1) it is already known as "normal", (2) dirstate entry for it has valid (= not "-1") timestamp, and (3) mode, size and timestamp of it on the filesystem are as same as ones expected in dirstate This works as expected in many cases, but doesn't in the corner case that changing a file keeps mode, size and timestamp of it on the filesystem. The timetable below shows steps in one of typical such situations: ---- ----------------------------------- ---------------- timestamp of "f" ---------------- dirstate file- time action mem file system ---- ----------------------------------- ---- ----- ----- N -1 *** - make file "f" clean N - execute 'hg foobar' - instantiate 'dirstate' -1 -1 - 'dirstate.normal("f")' N -1 (e.g. via dirty check) - change "f", but keep size N N+1 - release wlock - 'dirstate.write()' N N - 'hg status' shows "f" as "clean" N N N ---- ----------------------------------- ---- ----- ----- The most important point is that 'dirstate.write()' is executed at N+1 or later. This causes writing dirstate timestamp N of "f" out successfully. If it is executed at N, 'parsers.pack_dirstate()' replaces timestamp N with "-1" before actual writing dirstate out. Occasional test failure for unexpected file status is typical example of this corner case. Batch execution with small working directory is finished in no time, and rarely satisfies condition (2) above. This issue can occur in cases below; - 'hg revert --rev REV' for revisions other than the parent - failure of 'merge.update()' before 'merge.recordupdates()' The root cause of this issue is that files are changed without flushing in-memory dirstate changes via 'repo.commit()' (even though omitting 'dirstate.normallookup()' on changed files also causes this issue). To detect changes of files correctly, this patch writes in-memory dirstate changes out explicitly after marking files as clean in 'workingctx._checklookup()', which is invoked via 'repo.status()'. After this change, timetable is changed as below: ---- ----------------------------------- ---------------- timestamp of "f" ---------------- dirstate file- time action mem file system ---- ----------------------------------- ---- ----- ----- N -1 *** - make file "f" clean N - execute 'hg foobar' - instantiate 'dirstate' -1 -1 - 'dirstate.normal("f")' N -1 (e.g. via dirty check) ----------------------------------- ---- ----- ----- - 'dirsttate.write()' -1 -1 ----------------------------------- ---- ----- ----- - change "f", but keep size N N+1 - release wlock - 'dirstate.write()' -1 -1 - 'hg status' -1 -1 N ---- ----------------------------------- ---- ----- ----- To reproduce this issue in tests certainly, this patch emulates some timing critical actions as below: - timestamp of "f" in '.hg/dirstate' is -1 at the beginning 'hg debugrebuildstate' before command invocation ensures it. - make file "f" clean at N - change "f" at N 'touch -t 200001010000' before and after command invocation changes mtime of "f" to "2000-01-01 00:00" (= N). - invoke 'dirstate.write()' via 'repo.status()' at N 'fakedirstatewritetime.py' forces 'pack_dirstate()' to use "2000-01-01 00:00" as "now", only if 'pack_dirstate()' is invoked via 'workingctx._checklookup()'. - invoke 'dirstate.write()' via releasing wlock at N+1 (or "not at N") 'pack_dirstate()' via releasing wlock uses actual timestamp at runtime as "now", and it should be different from the "2000-01-01 00:00" of "f". BTW, this patch also changes 'test-largefiles-misc.t', because adding 'dirstate.write()' makes recent dirstate changes visible to external process.

tests: add extension to emulate invoking dirstate.write at the specific time This extension fakes 'now' for 'parsers.pack_dirstate()' to emulate invoking 'dirstate.write()' at the specific time, only when 'dirstate.write()' is invoked via functions below: - 'workingctx._checklookup()' (= 'repo.status()') - 'committablectx.markcommitted()' This is useful to reproduce timing critical issues fixed in subsequent patches.

convert: handle copies when converting from Perforce (issue4744)

convert: add config for recording the source name This creates the convert.hg.sourcename config option which will embed a user defined name into each commit created by the convert. This is useful when using the convert extension to merge several repositories together and we want to record where each commit came from.

convert: support multiple specifed revs in git source This allows specifying multiple revs/branches to convert from a git repo.

convert: add support for specifying multiple revs Previously convert could only take one '--rev'. This change allows the user to specify multiple --rev entries. For instance, this could allow converting multiple branches (but not all branches) at once from git. In this first patch, we disable support for this for all sources. Future patches will enable it for select sources (like git).

monoblue: use padding instead of position for text in footer Some installations alter monoblue style and remove margins from body element (these margins have that dark gray background) to adapt hgweb instance to an already existing site design. However, the margins hid a quirk in page footer: a block of text needlessly popped out of the footer, and when margins were gone, the whole page got a vertical scroll bar because of that. Live example: https://hg.prosody.im/prosody-modules/ To remove the potential scroll bar, this block of text now uses left padding, which doesn't make it overflow the footer, but makes it achieve the otherwise same result visually.

monoblue: don't try to show repo on hgwebdir index page Index page shows a list of accessible repositories, it doesn't have a single-repo context.

convert: handle deleted files when converting from Perforce (issue4743)

bookmarks: change bookmark within a transaction For some time, bookmark can and should be moved in the transaction. This changeset migrates the 'hg bookmarks' commands to use a transaction. Tests regarding rollback and transaction hooks are impacted for obvious reasons. Some have to be slightly updated to keep testing the same things. Some can just be dropped because they do not make sense anymore.

bookmark: remove the "touch changelog" hack Any changes to bookmarks used to touch the changelog to ensure hgweb was reloaded. This was fairly hacky and stops working when bookmarks are moved as part of the transaction. As hgweb is now explicitly tracking bookmark changes, we can remove this hack (and no tests break).

convert: add config option for disabling ancestor parent checks When converting merge commits, convert checks if any of the parents are ancestors of any of the other parents. To do this, it builds an ancestor list for every commit in the repository. On large repos this can take a long time (30min+). Let's add an option for disabling this check to preserve performance. The downside of this is that it may create unnecessary parent connections when enabled (which is unfortunate, but not incorrect). To verify, I ran the convert tests with the flag enabled, and verified the graph changes were all just to add new parents that were ancestors of existing parents.

convert: add config to not convert tags In some cases we do not want to convert tags from the source repo to be tags in the target repo (for instance, in a large repository, hgtags cause scaling issues so we want to avoid them). This adds a config option to disable converting tags.

templatekw: make {rev} return wdirrev instead of None wdirrev/wdirnode identifiers are still experimental, but {node} is mapped to wdirnode. So {rev} should do the same for consistency. I'm not sure if templatekw can import scmutil. If not, we should move intrev() to node module.

changeset_printer: use node.wdirrev to calculate meaningful parentrevs Because we defined the working-directory revision is INT_MAX, it makes sense that "hg log -r 'wdir()'" displays the "parent:" field. This is the same for two revisions that are semantically contiguous but the intermediate revisions are hidden.

workingctx: use node.wdirid constant

node: define experimental identifiers for working directory The "ff..." node was introduced at 183965a00c76, and we also need an integer that can be processed in revset. We could use len(repo), but it would be likely to hide possible bugs. Instead, using INT_MAX, we can notice such bugs by IndexError, at the cost of handling non-contiguous revisions.

templatekw: apply manifest template only if ctx.manifestnode() exists This will prevent crash by "hg log -r 'wdir()' -Tdefault". We could use the pseudo ff... hash introduced by 183965a00c76, but it isn't proven idea yet. For now, I want to make "hg log" just works in order to test 'wdir()' revset. Note that unlike its name, "{manifest}" is not a list of files in that revision, but a pair of (manifestrev, manifestnode).

merge with stable

import-checker: exclude mercurial packages installed into the system path If mercurial was installed into a directory other than the site-packages, test-module-imports.t failed as 'mercurial.node' was listed in stdlib_modules: testpackage/latesymbolimport.py relative import of stdlib module Instead, we should exclude our packages explicitly.

import-checker: recurse into subtree of sys.path only if __init__.py exists We can't assume that the site-packages is the only directory that has Python files but is not handled as a package. For example, we have dist-packages directory on Debian.

hghave: allow adding customized features at runtime Before this patch, there is no way to add customized features to `hghave` without changing `hghave` and `hghave.py` themselves. This decreases reusability of `run-tests.py` framework for third party tools, because they may want to examine custom features at runtime (e.g. existence of some external tools). To allow adding customized features at runtime, this patch makes `hghave` import `hghaveaddon` module, only when `hghaveaddon.py` file can be found in directories below: - `TESTDIR` for invocation via `run-tests.py` - `.` for invocation via command line The path to the directory where `hghaveaddon.py` should be placed is added to `sys.path` only while importing `hghaveaddon`, because: - `.` may not be added to `PYTHONPATH` - adding additional path to `sys.path` may change behavior of subsequent `import` for other features `hghave` is terminated with exit code '2' at failure of `import hghaveaddon`, because exit code '2' terminates `run-tests.py` immediately. This is a one of preparations for issue4677.

import-checker.py: exit with code 0 if no error is detected Before this patch, `import-checker.py` exits with non-0 code, if no error is detected. This is unusual as Unix command. This change may be a one of preparations for issue4677, because this can avoid extra explanation about unusual exit code of `import-checker.py` for third party tool developers.

run-tests.py: add TESTDIR to PATH if it differs from RUNTESTDIR Before this patch, `RUNTESTDIR` is added to `PATH`, but `TESTDIR` isn't. This doesn't cause any problems, if `run-tests.py` runs in `tests` directory of Mercurial source tree. In this case, `RUNTESTDIR` should be equal to `TESTDIR`. On the other hand, if `run-tests.py` runs in `tests` of third party tools, commands in that directory should be executed with explicit `$TESTDIR/` prefix in `*.t` test scripts. This isn't suitable for the policy "drop explicit $TESTDIR from executables" of Mercurial itself (see 4d2b9b304ad0). BTW, 4d2b9b304ad0 describes that "$TESTDIR is added to the path" even though `TESTDIR` isn't added to `PATH` exactly speaking, because `TESTDIR` and `RUNTESTDIR` weren't yet distinguished from each other at that time. This is a one of preparations for issue4677.

run-tests.py: add RUNTESTDIR to refer `tests` of Mercurial Before this patch, there is no way to refer files under `tests` or so of Mercurial source tree, when `run-tests.py` runs in `tests` of third party tools. In this case, `TESTDIR` refers the latter `tests`. This prevents third party tools from using useful tools in Mercurial source tree (e.g. `contrib/check-code.py`). This patch adds `RUNTESTDIR` environment variable to refer `tests` of Mercurial source tree, in which `run-tests.py` now running is placed. For example, tests of third party tools can refer `contrib/check-code.py` in Mercurial source tree as `$RUNTESTDIR/../contrib/check-code.py`. BTW, for similarity with `TESTDIR` referring `test*s*` directory, newly added environment variable isn't named as `RUNTEST*S*DIR`. In addition to it, the corresponded local variable is also named as `runtestdir`. This is a one of preparations for issue4677.

run-tests.py: execute hghave by the path relative to run-tests.py Before this patch, `run-tests.py` executes `hghave` by the path relative to `TESTDIR` (= cwd of `run-tests.py` running). This prevents third party tools for Mercurial from running `run-tests.py`, which is placed in `tests` of Mercurial source tree, in `tests` of own source tree. In such cases, `TESTDIR` refers the latter `tests`, and `hghave` doesn't exist in it. This is a one of preparations for issue4677.

templatekw: make {latesttag} a hybrid list This maintains the previous behavior of expanding {latesttag} to a string containing all of the tags, joined by ':'. But now it also allows list type operations. I'm unsure if the plural handling is correct (i.e. it seems like it is usually "{foos % '{foo}'}"), but I guess we are stuck with this because the singular form previously existed.

templatekw: allow the caller of showlist() to specify the join() separator The keyword {latesttag} currently manually joins the list of tags using ':', which prevents a transparent switch over to a hybrid list.

archive: use {changessincelatesttag} to build the metadata file This isolates the current magic when dealing with wdir() and only().

templatekw: introduce the changessincelatesttag keyword Archive is putting a value with the same name in the metadata file, to count all of the changes not covered by the latest tag, instead of just along the longest path. It seems that this would be useful to have on the command line as well. It might be nice for the name to start with 'latesttag' so that it is grouped with the other tag keywords, but I can't think of a better name. The initial version of this counted a clean wdir() and '.' as the same value, and a dirty wdir() as the same value after it is committed. Yuya objected on the grounds of consistency [1]. Since revsets can be used to conditionally select a dirty wdir() or '.' when clean, I can build the version string I need and will defer to him on this. [1] https://www.selenic.com/pipermail/mercurial-devel/2015-June/071588.html

help: support 'hg help template.somekeyword' Previously the output was simply 'abort: help section not found'.

test-convert-git: use a relative gitmodule url The absolute URL was causing this error with 1.9.5 on Windows, which had a cascading effect: @@ -466,22 +466,24 @@ > url = $TESTTMP/git-repo5 > EOF $ git commit -a -m "weird white space submodule" - [master *] weird white space submodule (glob) - Author: nottest <test@example.org> - 1 file changed, 3 insertions(+) + fatal: bad config file line 6 in $TESTTMP/git-repo6/.gitmodules + [128] $ cd .. $ hg convert git-repo6 hg-repo6 initializing destination hg-repo6 repository scanning source... For reasons unknown, there is still this delta on Windows: @@ -490,7 +490,6 @@ $ git commit -q -m "missing .gitmodules" $ cd .. $ hg convert git-repo6 hg-repo6 --traceback - fatal: Path '.gitmodules' does not exist in '*' (glob) initializing destination hg-repo6 repository scanning source... sorting...

test-convert-git: stablize for git 1.7.7.6 The output has apparently changed slightly since this version. Since they are just commits without any obvious importance to the test, and I can't figure out how to glob them away, silence them. Sample diffs were like this: @@ -468,7 +468,7 @@ $ git commit -a -m "weird white space submodule" [master *] weird white space submodule (glob) Author: nottest <test@example.org> - 1 file changed, 3 insertions(+) + 1 files changed, 3 insertions(+), 0 deletions(-) $ cd .. $ hg convert git-repo6 hg-repo6 initializing destination hg-repo6 repository

hgweb: also monitor change to bookmarks This makes changes to bookmarks visible to hgweb through the official way. There is no change to tests because there is currently another hack in place to ensure the same behavior.

hgweb: also refresh the repo on changes to the obsstore Before this change, hgweb could miss update to the obsolescence markers store if that was the only change between two commands.

hgweb: use an extensible list of files to check for refresh The refresh feature was explicitly testing if '00changelog.i' and 'phaseroots' changed. This is overlooking other important information like bookmarks and obsstore (bookmark have their own hack to work around it). We move to a more extensible system with a list of files of interest that will be used to build the repo state. The system should probably move into a more central place so that the command server and other systems are able to use it. Extension writers will also be able to add entries to ensure that changes to extension data are properly detected. Also the current key (mtime, size) is notably weak for bookmarks and phases whose files can easily change content without effect on their size. Still, this patch seems like a valuable minimal first step.

hgweb: drop the default argument for get_stat This default argument is used twice and is making things confusing. Making it explicit helps to clarify coming changesets

revset: prefetch method in "parents" As already demonstrated, saving attribute lookup gains us some minor but noticeable performance improvements. revset #0: parents(all()) before) 0.024169 after ) 0.022756 94%

convert: handle deleted files when converting from Perforce (issue4743)

forget: add a note to the command help about remove

shelve: only keep the latest N shelve backups This will keep the backup directory from growing indefinitely. The number of backups to keep can be set using the shelve.maxbackups config option (defaults to 10 backups).

shelve: always backup shelves instead of deleting them Instead of being deleted, shelve files are now moved into the .hg/shelve-backup directory. This is designed similarly to how strip saves backups into .ht/strip-backup. The goal is to prevent data loss especially when using unshelve. There are cases in which a user can complete an unshelve but lose some of the data that was shelved by, for example, resolving merge conflicts incorrectly. Storing backups will allow the user to recover the data that was shelved, at the expense of using more disk space over time.

wireproto: remove a debug print This looks like someone forgot something here.

amend: move obsmarkers creation in the "new changeset" conditional We already check if we created a new changesets right above this piece of code, so we can just drop the condition and indent the markers creation.

amend: move createmarkers evaluation earlier The value is used at multiple points in the function. Retrieving the value in the middle of the transaction scope gives the false impression that it has a single user. We move it at the start of the function to clarify this.

wireproto: correctly escape batched args and responses (issue4739) This issue appears to be as old as wireproto batching itself: I can reproduce the failure as far back as 08ef6b5f3715 trivially by rebasing the test changes in this patch, which was back in the 1.9 era. I didn't test before that change, because prior to that the testfile has a different name and I'm lazy. Note that the test thought it was checking this case, but it actually wasn't: it put a literal ; in the arg and response for its greet command, but the mangle/unmangle step defined in the test meant that instead of "Fo, =;o" going over the wire, "Gp-!><p" went instead, which doesn't contain any special characters (those being [.=;]) and thus not exercising the escaping. The test has been updated to use pre-unmangled special characters, so the request is now "Fo+<:o", which mangles to "Gp,=;p". I have confirmed that the test fails without the adjustment to the escaping rules in wireproto.py. No existing clients of RPC batching were depending on the old behavior in any way. The only *actual* users of batchable RPCs in core were: 1) largefiles, wherein it batches up many statlfile calls. It sends hexlified hashes over the wire and gets a 0, 1, or 2 back as a response. No risk of special characters. 2) setdiscovery, which was using heads() and known(), both of which communicate via hexlified nodes. Again, no risk of special characters. Since the escaping functionality has been completely broken since it was introduced, we know that it has no users. As such, we can change the escaping mechanism without having to worry about backwards compatibility issues. For the curious, this was detected by chance: it happens that the lz4-compressed text of a test file for remotefilelog compressed to something containing a ;, which then caused the failure when I moved remotefilelog to using batching for file content fetching.

merge with stable

revset: port extra() to support keyword arguments This is an example to show how keyword arguments are processed.

revset: add function to build dict of positional and keyword arguments Keyword arguments will be convenient for functions that will take more than one optional or boolean flags. For example, file(pattern[, subrepos=false]) subrepo([[pattern], status]) Because I don't think all functions should accept key=value syntax, getkwargs() does not support variadic functions such as 'ancestor(*changeset)'. The core logic is placed in the parser module because keyword arguments will be more useful in the templater, where functions take more options. Test cases will be added by the next patch.

revset: add parsing rule for key=value pair It will be used as an keyword argument. Note that our "=" operator is left-associative. In general, the assignment operator is right-associative, but we don't care because it isn't allowed to chain "=" operations.

import-checker: establish modern import convention We introduce a new convention for declaring imports and enforce it via the import checker script. The new convention is only active when absolute imports are used, which is currently nowhere. Keying off "from __future__ import absolute_import" to engage the new import convention seems like the easiest solution. It is also beneficial for Mercurial to use this mode because it means less work and ambiguity for the importer and potentially better performance due to fewer stat() system calls because the importer won't look for modules in relative paths unless explicitly asked. Once all files are converted to use absolute import, we can refactor this code to again only have a single import convention and we can require use of absolute import in the style checker. The rules for the new convention are documented in the docstring of the added function. Tests have been added to test-module-imports.t. Some tests are sensitive to newlines and source column position, which makes docstring testing difficult and/or impossible.

import-checker: establish new function for verifying import conventions A future patch will formalize the modern import convention. In preparation for that, introduce a new wrapper function that will invoke the proper function.

import-checker: resolve relative imports "from . import X" will produce an ImportFrom ast node with .module = None. This resulted in a run-time error from attempting to concatenate None with a str. Another problem with relative imports is that the prefix may be dynamic based on the "level" attribute of the import. e.g. "from ." has level 1 and "from .." has level 2. We teach the "fromlocal" function how to cope with relative imports. Where appropriate, the consumer passes in the level so relative module names may be resolved properly.

templatekw: use a list of tags in getlatesttags() instead of joining them This will be used in the next patch. It also points out that the documentation for '{latesttag}' is not quite accurate, since it says "most recent global tag" (singular). I assume it is too radical of a change to convert it to a list of strings. At least ':' is currently a reserved character in tag names.

convert: improve support for unusual .gitmodules Previously convert would throw an exception if it encountered a git commit with a .gitmodules file that was malformed (i.e. was missing, but had submodule files, or was malformed). Instead of breaking the convert entirely, let's print error messages and move on.

convert: handle .gitmodules with non-tab whitespaces The old implementation assumed .gitmodules file lines always began with tabs. It can be any whitespace, so lets trim the lines appropriately.

convert: fix bug with converting the same commit twice Convert had a bug where it relied on repo.tip() to be the newly committed commit. This was not the case if the commit already existed in the repository (since repo.commitctx() did nothing, the tip() referenced some random other commit and the revmap got corrupted). This fixes it by using the node returned by repo.commitctx().

templater: remove workaround for escaped quoted string in quoted template This patch backs out 554d6fcc3c84 which should no longer be needed. The test for '{\"invalid\"}' is removed because the parser is permissive for \"...\" literal.

merge with stable

amend: stop updating the bookmarks twice There was code to move the bookmarks around both in the 'cmdutil' help and in the main 'commit' function. We kill the 'commit' version as it is performed outside the transaction. The debug note is moved into cmdutil.

amend: collaborate with the transaction when moving bookmarks We have code moving bookmarks from the old changeset to the new one within the transaction scope. Yet this code was still writing to disk instead of handing the change to the transaction. This changeset fixes this.

sshserver: drop ancient do_{lock,unlock,addchangegroup} methods These were marked as deprecated and dangerous way back in e8c4f3d3df8c, which was first included in Mercurial 0.9.1. While it's possible that clients from that long ago are still around somewhere, they're risky for servers in that they want to lock the repo, and then might leave it locked if they died before finishing their transaction. Given that it's been 9 years, let's go ahead and cut this last lingering tie with a basically-untested protocol.

wireproto: add config knob for http header length limit Well-behaved Mercurial clients will respect the httpheader capability by not sending http headers longer than the given limit in bytes. The limit is currently hard-coded at 1024 bytes, a safe value for any web server. Since parsing headers is a notable factor in web server performance, tuning header size can nontrivially improve performance for request-heavy operations (eg. obsolete marker negotiation). Exposing the maximum header length limit as a configuration setting is a simple way to enable such tuning.

archive: fix changesincelatesttag with wdir()

revset: fix a crash in parents() when 'wdir()' is in the set The crash was "TypeError: expected string or Unicode object, NoneType found" down in revlog.parentrevs(). This fixes heads() too (which is where I found it.)

workingctx: don't report the tags for its parents This fixes the bad distance calculation for '{latesttagdistance}' mentioned in the previous patch.

hgweb: don't try to show repo on repo-not-found page in monoblue

hgweb: fix help pages title in gitweb and monoblue Help pages in gitweb and monoblue were setting a completely inappropriate title: "Branches". Let's have a proper title (copy-pasted from paper style).

hgweb: don't show summary link as active on error pages in monoblue These two error pages in monoblue think they are the summary page, when they clearly aren't.

identify: build the tag list directly instead of using wctx.tags() The current implementation of workingctx.tags() returns the tags of the parents. This causes the calculation of {lastesttagdistance} from wdir() to be wrong. The value when updated to a tag is 0, but updated to the tag's child is 2, the child of that 3, and so on. This prepares for workingctx.tags() to not report the parent tags.

identify: avoid a crash when given '-r wdir()' The crash was 'NoneType is not subscriptable' in hexfunc(ctx.node()), because the node for wdir() is None. This can be avoided simply by detecting 'wdir()' and taking the existing path for no given revision.

tests: clean up duplicated output in test-subrepo-recursion progress We have fixed a bug where two progress instance were created and competed with each other (in 95f490136e75). But we did not updated the non-hardlink section of 'test-subrepo-recursion.t'. This patch fixes it.

Added signature for changeset 540cd0ddac49

Added tag 3.4.2 for changeset 540cd0ddac49

transplant: only pull the transplanted revision (issue4692) For some reason, transplant was pulling all remote revisions when transplanting from a remote repository (unless --branch was specified).

transplant: update test to use hash for remote transplant Transplant is apparently allowing using revision numbers when transplanting through http. I call this nonsense and update the test to use an explicit hash. This "feature" will break in the next change fixing another bug.

changegroup: properly compute common base in changeggroupsubset (issue4736) The computation of roots was buggy, any ancestor of a bundled merge which was also a descendant of the parents of a bundled revision were included as part of the bundle. We fix it and add a test for strip (which revealed the problem). Check the test for a practical usecase.

templater: parse \"...\" as string for 2.9.2-3.4 compatibility (issue4733) As of Mercurial 3.4, there were several syntax rules to process nested template strings. Unfortunately, they were inconsistent and conflicted each other. a. buildmap() rule - template string is _parsed_ as string, and parsed as template - <\"> is not allowed in nested template: {xs % "{f(\"{x}\")}"} -> parse error - template escaping <\{> is handled consistently: {xs % "\{x}"} -> escaped b. _evalifliteral() rule - template string is _interpreted_ as string, and parsed as template in crafted environment to avoid double processing of escape sequences - <\"> is allowed in nested template: {if(x, "{f(\"{x}\")}")} - <\{> and escape sequences in string literal in nested template are not handled well c. pad() rule - template string is first interpreted as string, and parsed as template, which means escape sequences are processed twice - <\"> is allowed in nested template: {pad("{xs % \"{x}\"}', 10)} Because of the issue of template escaping, issue4714, 7298da81f5a9 (in stable) unified the rule (b) to (a). Then, 576d6c74784b (in default) unified the rule (c) to (b) = (a). But they disabled the following syntax that was somewhat considered valid. {if(rev, "{if(rev, \"{rev}\")}")} {pad("{files % \"{file}\"}", 10)} So, this patch introduces \"...\" literal to work around the escaped-quoted nested template strings. Because this parsing rule exists only for the backward compatibility, it is designed to copy the behavior of old _evalifliteral() as possible. Future patches will introduce a better parsing rule similar to a command substitution of POSIX shells or a string interpolation of Ruby, where extra escapes won't be necessary at all. {pad("{files % "{file}"}", 10)} ~~~~~~~~~~~~~~~~~~ parsed as a template, not as a string Because <\> character wasn't allowed in a template fragment, this patch won't introduce more breakages. But the syntax of nested templates are interpreted differently by people, there might be unknown issues. So if we want, we could instead remove db7463aa080f, 890845af1ac2 and 7298da81f5a9 from the stable branch as the bug fixed by these patches existed for longer periods. 554d6fcc3c8, "strip single backslash before quotation mark in quoted template", should be superseded by this patch. I'll remove it later.

merge with stable

demandimport: support importing builtins for Python 3 __builtin__ was renamed to builtins in Python 3. If importing __builtin__ fails, fall back to importing builtins in order to support Python 3.

demandimport: alias __builtin__ as builtins Python 3 renamed the __builtin__ module to builtins. In preparation for supporting Python 3, alias the imported module as "builtins."

bufferedinputpipe: remove N^2 computation of buffer length (issue4735) The assumption that dynamically computing the length of the buffer was N^2, but negligible because fast was False. So we drop the dynamic computation and manually keep track of the buffer length.

bufferedinputpipe: remove an outdate comment This comment is the remains of a intermediate implementation using self._buffer += data This implementation never made it to the repository and we can safely drop the comment.

statichttprepo: kill off sopener

obsolete: replace references to 'sopener' with 'svfs'

exchange: replace references to 'sopener' with 'svfs'

localrepo: kill off sopener (API) sopener is deprecated since 7034365089bf (Mercurial 2.3). It's annoying for extension authors to have to deal with both. Let's just kill it off.

share: replace reference to 'sopener' with 'svfs' sopener is deprecated. It's annoying for extension authors to have to deal with both. Let's just kill it off.

archive: don't assume '.' is being archived for changessincelatesttag Hardcoding '.' is wrong, and yielded strange results when archiving old revisions. For example, when archiving the cset that adds the signature to 3.4 (c48850339988), the resulting value was previously 51 (the number of commits on stable between 3.4 and today), even though it was a direct descendant of a tag, with a {latesttagdistance} of 2. This still includes all other _ancestor_ paths not included in {latesttag}. Note that archiving wdir() currently blows up several lines above this when building the 'base' variable. Since wdir() isn't documented, ignore that it needs work to handle wdir() here for now.

hgweb: use css for stripey background in coal Since "b8ecc3830c89 or 25dae11bb044::c229a5e7511e" paper style used css for stripes in background for browsing files, for listing branches/tags/bookmarks, and so on. Since coal borrows many paper templates (e.g. shortlogentry.tmpl), it actually tried to do the same, but it didn't have the needed css classes. You can compare https://selenic.com/hg?style=coal with https://selenic.com/hg?style=paper and see how log view in coal style has plain white background, unlike the one in paper style. This wasn't intended. Let's copy css classes directly from style-paper.css and remove parity classes from elements that don't need them anymore. This makes plain white background have stripes again and makes coal/map even more similar to paper/map (which can ease porting changes or %including paper/map in future).

templatekw: correct typo in activebookmark documentation

merge with stable

check-code: detect legacy exception syntax We just rewrote all files to use modern exception syntax. Ban the old form. This will detect the "except type, instance" and "except (type1, type2), instance" forms.

global: mass rewrite to use modern exception syntax Python 2.6 introduced the "except type as instance" syntax, replacing the "except type, instance" syntax that came before. Python 3 dropped support for the latter syntax. Since we no longer support Python 2.4 or 2.5, we have no need to continue supporting the "except type, instance". This patch mass rewrites the exception syntax to be Python 2.6+ and Python 3 compatible. This patch was produced by running `2to3 -f except -w -n .`.

check-code: detect legacy octal syntax Now that we have mass rewriting all files to use the modern octal syntax, detect and ban the legacy syntax, which is no longer supported in Python 3.

global: mass rewrite to use modern octal syntax Python 2.6 introduced a new octal syntax: "0oXXX", replacing "0XXX". The old syntax is not recognized in Python 3 and will result in a parse error. Mass rewrite all instances of the old octal syntax to the new syntax. This patch was generated by `2to3 -f numliterals -w -n .` and the diff was selectively recorded to exclude changes to "<N>l" syntax conversion, which will be handled separately.

revert: change the direction of revert -i After the discussion on the list about hg revert -i, it seems like we are satisfied with what we called proposition 2. It shows the changes to revert in the same direction as hg diff. This patch makes it the default option. It changes all the + in - and vice versa in the tests for revert -i.

revert: change a test to make the change of direction of revert -i easier Currently we are handling editing of newly-added files with the interactive interface. We are not handling editing of deleted files. In the test for revert, we were editing a newly-added file. Since we want to change the direction of revert -i, this editing of a newly-added file will become editing of a deleted file. Since we don't support that, this patch changes the test to make the rest of the series cleaner.