lfs: prefetch lfs blobs during revert The revert command oddly prints out what it will do before requesting the files to be prefetched. But the 'need to transfer' line indicates the blobs are being grouped.

cmdutil: convert _revertprefetch() to a generic stored file hook (API) This will be used by LFS to fetch required files in a group for multiple commands, prior to being accessed. That avoids the one-at-a-time fetch when the filelog wrapper goes to access it, and it is missing locally (which costs two round trips to the server.) The core command list that needs this is probably at least: - annotate - archive (which is also used by extdiff) - cat - diff - export - grep - verify (sadly) - anything that has the '{data}' template There are no core users of the revert prefetch hook, and never have been since it was introduced in 45e02cfad4bd for remotefilelog. Thanks to Yuya for figuring out a way to reliably trigger the deprecated warning. Unfortunately, it wanted to blame the caller of revert. Passing along an adjusted stack level seemed the least bad choice (although it still blames a core function). One thing to note is that the store lock isn't being held when this is called. I'm not at all familiar with remotefilelog or its locking requirements, so this may not be a big deal. Currently, LFS doesn't hold a lock when downloading files. Even though largefiles doesn't either, I'm starting to think it should, and maybe the .hg/store/lock isn't good enough to cover the globally shared cache. .. api:: The cmdutil._revertprefetch() hook point for prefetching stored files has been replaced by the command agnostic cmdutil._prefetchfiles(). The new function takes a list of files, instead of a list of lists of files.

lfs: prefetch lfs blobs when applying merge updates In addition to merge, this method ultimately gets called by many commands: - backout - bisect - clone - fetch - graft - import (without --bypass) - pull -u - rebase - strip - share - transplant - unbundle - update Additionally, it's also called by histedit, shelve, unshelve, and split, but it seems that the related blobs should always be available locally for these. For `hg update`, it happens after the normal argument checking and pre-update hook processing, and remote corruption is detected prior to manipulating the working directory. Other commands could use this treatment (archive, cat, revert, etc), but this covers so many of the frequently used bulk commands, it seems like a good starting point. Losing the verbose message that prints the file name before a corrupt blob aborts the command is a little sad, because there's no easy way to go from oid to file name. I'd like to change that message to list the file name so it looks cleaner and less cryptic, but the pointer object is nowhere near where it needs to be to do this. So punt on that for now.

lfs: factor out a method for extracting the pointer of a single file This will be useful for filesets, among other things, instead of traversing the whole context.

py3: add b'' to literals in check-config.py # skip-blame because we are just adding b'' Differential Revision: https://phab.mercurial-scm.org/D2046

check-config: specify the mode 'rb' to open the file Differential Revision: https://phab.mercurial-scm.org/D2045

py3: use open() instead of file() file() is not present in Python 3 Differential Revision: https://phab.mercurial-scm.org/D2044

py3: use pycompat.strkwargs() to convert kwargs' key to str Differential Revision: https://phab.mercurial-scm.org/D2043

py3: add __bytes__() for mq.patchheader and make sure __str__ returns str Before this patch, __str__() function for mq.patchheader class return bytes which is not str on Python 3. So let's move that logic to __bytes__() and for __str__() convert the return value of __bytes__() to str. Differential Revision: https://phab.mercurial-scm.org/D2042

py3: use "%d" to convert integer to bytes Differential Revision: https://phab.mercurial-scm.org/D2041

py3: use .startswith() instead of bytes[0] Differential Revision: https://phab.mercurial-scm.org/D2040

py3: slice on bytes to prevent getting the ascii values Differential Revision: https://phab.mercurial-scm.org/D2039

py3: add r'' to convert keys to keyword arguments to str # skip-blame as we are just adding r'' Differential Revision: https://phab.mercurial-scm.org/D2038

py3: add b'' to tweakdefaults config string The tweakdefaults config string is enclosed inside triple quotes and the transformer does not adds b'' there. # skip-blame because we are just adding b'' Differential Revision: https://phab.mercurial-scm.org/D2037

cmdutil: introduce deprecated aliases c8e2d6ed1f9e moved some objects used by Evolve and hence broke the latest Evolve revision. Next Evolve version will use the new objects when available but introduce deprecated aliases so users using older version of Evolve won't have a broken Evolve extension. Differential Revision: https://phab.mercurial-scm.org/D2023

makefile: remove Ubuntu Yakkety and Zesty docker targets Yakkety (16.10) was supported until 2017-07-20 and Zesty (17.04) was supported until 2018-01-13.

merge: use operation-provided labels (ex: dest/source) in several merge-tools Tools that did not use labels already, used only one, or used some label other than the exact strings of "local" or "other" were unmodified. Cases that used the label "base" were modified as well, if they were otherwise changed in this CL; "merged" was *not* changed. There are other possible changes we might want to make, but I didn't: - bcompare (linux and osx) uses the labels "parent1" and "parent2" instead of "local" and "other", so it was left alone, even though beyondcompare3 (windows) *was* changed. - araxis used the labels "Other", "Base", and "Local :$local", so it was also left alone. - UltraCompare didn't provide a label for 'local', just for 'base' and 'other', so it was left alone. Differential Revision: https://phab.mercurial-scm.org/D2012

filemerge: support passing labels to external merge tools This adds $labellocal, $labelother, and $labelbase to the replacement set for merge-tools.<tool>.args config variables, and to the environment as HG_MY_LABEL, HG_OTHER_LABEL, and HG_BASE_LABEL, respectively. We also add merge-tools.<tool>.mergemarkers and merge-tools.<tool>.mergemarkertemplate config variables as counterparts of the variables available in [ui]. We are intentionally *not* respecting ui.mergemarkers when calling out to external merge programs; too often the default template will be too wide to display comfortably in most GUIs. Differential Revision: https://phab.mercurial-scm.org/D2011

bookmarks: drop deprecated methods (API) The ProgrammingError prevents accidental usage of the dict base class methods. .. api:: The following deprecated methods have been removed from bookmarks: __setitem__(), __delitem__(), update(), and recordchange(). Use bookmarks.applychanges() instead.

archive: rewrite default metadata template as a multi-line bytes literal This fixes test-directaccess.t on Python 3.

py3: drop b'' from repr() of smartset cmdutil._maybebytestr() is moved to pycompat.

py3: always drop b'' prefix from repr() of bytestr Perhaps this is what we wanted for py2-3 compatibility.

py3: format revision number as '%d' in debugrevspec Now "hg debugrevspec" works on Python 3.

py3: build repr() of smartset as bytes then convert to str This isn't pretty, but we have no way to teach Python 3 that our __repr__() would like to return a byte string.

py3: factor out helpers to apply string conversion recursively

py3: replace "if ispy3" by pycompat.sysbytes() or util.forcebytestr()

py3: replace "if ispy3" by pycompat.bytestr()

py3: replace "if ispy3" by encoding.strtolocal()

localrepo: drop the deprecated walk() method (API) .. api:: The deprecated localrepo.walk() has been removed, and replaced by repo[node].walk().

dirstate: drop deprecated methods (API) .. api:: beginparentchange() and endparentchange() have been replaced by the parentchange context manager.

templatekw: drop the deprecated '{troubles}' keyword

obsutil: drop deprecated methods (API) .. api:: The following deprecated methods have been removed from obsutil: marker.precnode() and allprecursors(). Use marker.prednode() and allpredecessors() instead.