Mercurial > hg > mercurial-crew-with-dirclash
changeset 5385:caadfbc439c6
Merge with crew.
| author | Bryan O'Sullivan <bos@serpentine.com> |
|---|---|
| date | Fri, 05 Oct 2007 12:17:29 -0700 |
| parents | d7d395fb7cd5 (diff) e3a0c092b4e2 (current diff) |
| children | 8051549afb32 |
| files | |
| diffstat | 30 files changed, 1005 insertions(+), 549 deletions(-) [+] |
line wrap: on
line diff
--- a/hgext/convert/__init__.py +++ b/hgext/convert/__init__.py @@ -5,14 +5,15 @@ # This software may be used and distributed according to the terms # of the GNU General Public License, incorporated herein by reference. -from common import NoRepo, converter_source, converter_sink +from common import NoRepo, SKIPREV, converter_source, converter_sink from cvs import convert_cvs from darcs import darcs_source from git import convert_git from hg import mercurial_source, mercurial_sink from subversion import convert_svn, debugsvnlog +import filemap -import os, shlex, shutil +import os, shutil from mercurial import hg, ui, util, commands from mercurial.i18n import _ @@ -40,7 +41,7 @@ def convertsink(ui, path): raise util.Abort('%s: unknown repository type' % path) class converter(object): - def __init__(self, ui, source, dest, revmapfile, filemapper, opts): + def __init__(self, ui, source, dest, revmapfile, opts): self.source = source self.dest = dest @@ -51,13 +52,15 @@ class converter(object): self.revmapfilefd = None self.authors = {} self.authorfile = None - self.mapfile = filemapper + self.maporder = [] self.map = {} try: origrevmapfile = open(self.revmapfile, 'r') for l in origrevmapfile: sv, dv = l[:-1].split() + if sv not in self.map: + self.maporder.append(sv) self.map[sv] = dv origrevmapfile.close() except IOError: @@ -199,7 +202,15 @@ class converter(object): do_copies = hasattr(self.dest, 'copyfile') filenames = [] - files, copies = self.source.getchanges(rev) + changes = self.source.getchanges(rev) + if isinstance(changes, basestring): + if changes == SKIPREV: + dest = SKIPREV + else: + dest = self.map[changes] + self.mapentry(rev, dest) + return + files, copies = changes parents = [self.map[r] for r in commit.parents] if commit.parents: prev = commit.parents[0] @@ -210,35 +221,28 @@ class converter(object): pbranch = None self.dest.setbranch(commit.branch, pbranch, parents) for f, v in files: - newf = self.mapfile(f) - if not newf: - continue - filenames.append(newf) + filenames.append(f) try: data = self.source.getfile(f, v) except IOError, inst: - self.dest.delfile(newf) + self.dest.delfile(f) else: e = self.source.getmode(f, v) - self.dest.putfile(newf, e, data) + self.dest.putfile(f, e, data) if do_copies: if f in copies: - copyf = self.mapfile(copies[f]) - if copyf: - # Merely marks that a copy happened. - self.dest.copyfile(copyf, newf) + copyf = copies[f] + # Merely marks that a copy happened. + self.dest.copyfile(copyf, f) - if not filenames and self.mapfile.active(): - newnode = parents[0] - else: - newnode = self.dest.putcommit(filenames, parents, commit) + newnode = self.dest.putcommit(filenames, parents, commit) self.mapentry(rev, newnode) def convert(self): try: self.source.before() self.dest.before() - self.source.setrevmap(self.map) + self.source.setrevmap(self.map, self.maporder) self.ui.status("scanning source...\n") heads = self.source.getheads() parents = self.walktree(heads) @@ -260,7 +264,7 @@ class converter(object): ctags = {} for k in tags: v = tags[k] - if v in self.map: + if self.map.get(v, SKIPREV) != SKIPREV: ctags[k] = self.map[v] if c and ctags: @@ -282,92 +286,6 @@ class converter(object): if self.revmapfilefd: self.revmapfilefd.close() -def rpairs(name): - e = len(name) - while e != -1: - yield name[:e], name[e+1:] - e = name.rfind('/', 0, e) - -class filemapper(object): - '''Map and filter filenames when importing. - A name can be mapped to itself, a new name, or None (omit from new - repository).''' - - def __init__(self, ui, path=None): - self.ui = ui - self.include = {} - self.exclude = {} - self.rename = {} - if path: - if self.parse(path): - raise util.Abort(_('errors in filemap')) - - def parse(self, path): - errs = 0 - def check(name, mapping, listname): - if name in mapping: - self.ui.warn(_('%s:%d: %r already in %s list\n') % - (lex.infile, lex.lineno, name, listname)) - return 1 - return 0 - lex = shlex.shlex(open(path), path, True) - lex.wordchars += '!@#$%^&*()-=+[]{}|;:,./<>?' - cmd = lex.get_token() - while cmd: - if cmd == 'include': - name = lex.get_token() - errs += check(name, self.exclude, 'exclude') - self.include[name] = name - elif cmd == 'exclude': - name = lex.get_token() - errs += check(name, self.include, 'include') - errs += check(name, self.rename, 'rename') - self.exclude[name] = name - elif cmd == 'rename': - src = lex.get_token() - dest = lex.get_token() - errs += check(src, self.exclude, 'exclude') - self.rename[src] = dest - elif cmd == 'source': - errs += self.parse(lex.get_token()) - else: - self.ui.warn(_('%s:%d: unknown directive %r\n') % - (lex.infile, lex.lineno, cmd)) - errs += 1 - cmd = lex.get_token() - return errs - - def lookup(self, name, mapping): - for pre, suf in rpairs(name): - try: - return mapping[pre], pre, suf - except KeyError, err: - pass - return '', name, '' - - def __call__(self, name): - if self.include: - inc = self.lookup(name, self.include)[0] - else: - inc = name - if self.exclude: - exc = self.lookup(name, self.exclude)[0] - else: - exc = '' - if not inc or exc: - return None - newpre, pre, suf = self.lookup(name, self.rename) - if newpre: - if newpre == '.': - return suf - if suf: - return newpre + '/' + suf - return newpre - return name - - def active(self): - return bool(self.include or self.exclude or self.rename) - def convert(ui, src, dest=None, revmapfile=None, **opts): """Convert a foreign SCM repository to a Mercurial one. @@ -461,15 +379,18 @@ def convert(ui, src, dest=None, revmapfi shutil.rmtree(dest, True) raise + fmap = opts.get('filemap') + if fmap: + srcc = filemap.filemap_source(ui, srcc, fmap) + destc.setfilemapmode(True) + if not revmapfile: try: revmapfile = destc.revmapfile() except: revmapfile = os.path.join(destc, "map") - - c = converter(ui, srcc, destc, revmapfile, filemapper(ui, opts['filemap']), - opts) + c = converter(ui, srcc, destc, revmapfile, opts) c.convert()
--- a/hgext/convert/common.py +++ b/hgext/convert/common.py @@ -17,6 +17,8 @@ def decodeargs(s): class NoRepo(Exception): pass +SKIPREV = 'hg-convert-skipped-revision' + class commit(object): def __init__(self, author, date, desc, parents, branch=None, rev=None): self.author = author @@ -44,8 +46,11 @@ class converter_source(object): def after(self): pass - def setrevmap(self, revmap): - """set the map of already-converted revisions""" + def setrevmap(self, revmap, order): + """set the map of already-converted revisions + + order is a list with the keys from revmap in the order they + appear in the revision map file.""" pass def getheads(self): @@ -91,6 +96,19 @@ class converter_source(object): except: return s.decode(encoding, "replace").encode("utf-8") + def getchangedfiles(self, rev, i): + """Return the files changed by rev compared to parent[i]. + + i is an index selecting one of the parents of rev. The return + value should be the list of files that are different in rev and + this parent. + + If rev has no parents, i is None. + + This function is only needed to support --filemap + """ + raise NotImplementedError() + class converter_sink(object): """Conversion sink (target) interface""" @@ -149,3 +167,13 @@ class converter_sink(object): pbranch: branch name of parent commit parents: destination revisions of parent""" pass + + def setfilemapmode(self, active): + """Tell the destination that we're using a filemap + + Some converter_sources (svn in particular) can claim that a file + was changed in a revision, even if there was no change. This method + tells the destination that we're using a filemap and that it should + filter empty revisions. + """ + pass
--- a/hgext/convert/cvs.py +++ b/hgext/convert/cvs.py @@ -266,3 +266,9 @@ class convert_cvs(converter_source): def gettags(self): return self.tags + + def getchangedfiles(self, rev, i): + files = self.files[rev].keys() + files.sort() + return files +
copy from hgext/convert/__init__.py copy to hgext/convert/filemap.py --- a/hgext/convert/__init__.py +++ b/hgext/convert/filemap.py @@ -1,286 +1,13 @@ -# convert.py Foreign SCM converter -# -# Copyright 2005-2007 Matt Mackall <mpm@selenic.com> +# Copyright 2007 Bryan O'Sullivan <bos@serpentine.com> +# Copyright 2007 Alexis S. L. Carvalho <alexis@cecm.usp.br> # -# This software may be used and distributed according to the terms -# of the GNU General Public License, incorporated herein by reference. - -from common import NoRepo, converter_source, converter_sink -from cvs import convert_cvs -from darcs import darcs_source -from git import convert_git -from hg import mercurial_source, mercurial_sink -from subversion import convert_svn, debugsvnlog - -import os, shlex, shutil -from mercurial import hg, ui, util, commands -from mercurial.i18n import _ - -commands.norepo += " convert debugsvnlog" - -converters = [convert_cvs, convert_git, convert_svn, mercurial_source, - mercurial_sink, darcs_source] - -def convertsource(ui, path, **opts): - for c in converters: - try: - return c.getcommit and c(ui, path, **opts) - except (AttributeError, NoRepo): - pass - raise util.Abort('%s: unknown repository type' % path) - -def convertsink(ui, path): - if not os.path.isdir(path): - raise util.Abort("%s: not a directory" % path) - for c in converters: - try: - return c.putcommit and c(ui, path) - except (AttributeError, NoRepo): - pass - raise util.Abort('%s: unknown repository type' % path) - -class converter(object): - def __init__(self, ui, source, dest, revmapfile, filemapper, opts): - - self.source = source - self.dest = dest - self.ui = ui - self.opts = opts - self.commitcache = {} - self.revmapfile = revmapfile - self.revmapfilefd = None - self.authors = {} - self.authorfile = None - self.mapfile = filemapper - - self.map = {} - try: - origrevmapfile = open(self.revmapfile, 'r') - for l in origrevmapfile: - sv, dv = l[:-1].split() - self.map[sv] = dv - origrevmapfile.close() - except IOError: - pass - - # Read first the dst author map if any - authorfile = self.dest.authorfile() - if authorfile and os.path.exists(authorfile): - self.readauthormap(authorfile) - # Extend/Override with new author map if necessary - if opts.get('authors'): - self.readauthormap(opts.get('authors')) - self.authorfile = self.dest.authorfile() - - def walktree(self, heads): - '''Return a mapping that identifies the uncommitted parents of every - uncommitted changeset.''' - visit = heads - known = {} - parents = {} - while visit: - n = visit.pop(0) - if n in known or n in self.map: continue - known[n] = 1 - commit = self.cachecommit(n) - parents[n] = [] - for p in commit.parents: - parents[n].append(p) - visit.append(p) - - return parents - - def toposort(self, parents): - '''Return an ordering such that every uncommitted changeset is - preceeded by all its uncommitted ancestors.''' - visit = parents.keys() - seen = {} - children = {} - - while visit: - n = visit.pop(0) - if n in seen: continue - seen[n] = 1 - # Ensure that nodes without parents are present in the 'children' - # mapping. - children.setdefault(n, []) - for p in parents[n]: - if not p in self.map: - visit.append(p) - children.setdefault(p, []).append(n) - - s = [] - removed = {} - visit = children.keys() - while visit: - n = visit.pop(0) - if n in removed: continue - dep = 0 - if n in parents: - for p in parents[n]: - if p in self.map: continue - if p not in removed: - # we're still dependent - visit.append(n) - dep = 1 - break - - if not dep: - # all n's parents are in the list - removed[n] = 1 - if n not in self.map: - s.append(n) - if n in children: - for c in children[n]: - visit.insert(0, c) +# This software may be used and distributed according to the terms of +# the GNU General Public License, incorporated herein by reference. - if self.opts.get('datesort'): - depth = {} - for n in s: - depth[n] = 0 - pl = [p for p in self.commitcache[n].parents - if p not in self.map] - if pl: - depth[n] = max([depth[p] for p in pl]) + 1 - - s = [(depth[n], self.commitcache[n].date, n) for n in s] - s.sort() - s = [e[2] for e in s] - - return s - - def mapentry(self, src, dst): - if self.revmapfilefd is None: - try: - self.revmapfilefd = open(self.revmapfile, "a") - except IOError, (errno, strerror): - raise util.Abort("Could not open map file %s: %s, %s\n" % (self.revmapfile, errno, strerror)) - self.map[src] = dst - self.revmapfilefd.write("%s %s\n" % (src, dst)) - self.revmapfilefd.flush() - - def writeauthormap(self): - authorfile = self.authorfile - if authorfile: - self.ui.status('Writing author map file %s\n' % authorfile) - ofile = open(authorfile, 'w+') - for author in self.authors: - ofile.write("%s=%s\n" % (author, self.authors[author])) - ofile.close() - - def readauthormap(self, authorfile): - afile = open(authorfile, 'r') - for line in afile: - try: - srcauthor = line.split('=')[0].strip() - dstauthor = line.split('=')[1].strip() - if srcauthor in self.authors and dstauthor != self.authors[srcauthor]: - self.ui.status( - 'Overriding mapping for author %s, was %s, will be %s\n' - % (srcauthor, self.authors[srcauthor], dstauthor)) - else: - self.ui.debug('Mapping author %s to %s\n' - % (srcauthor, dstauthor)) - self.authors[srcauthor] = dstauthor - except IndexError: - self.ui.warn( - 'Ignoring bad line in author file map %s: %s\n' - % (authorfile, line)) - afile.close() - - def cachecommit(self, rev): - commit = self.source.getcommit(rev) - commit.author = self.authors.get(commit.author, commit.author) - self.commitcache[rev] = commit - return commit - - def copy(self, rev): - commit = self.commitcache[rev] - do_copies = hasattr(self.dest, 'copyfile') - filenames = [] - - files, copies = self.source.getchanges(rev) - parents = [self.map[r] for r in commit.parents] - if commit.parents: - prev = commit.parents[0] - if prev not in self.commitcache: - self.cachecommit(prev) - pbranch = self.commitcache[prev].branch - else: - pbranch = None - self.dest.setbranch(commit.branch, pbranch, parents) - for f, v in files: - newf = self.mapfile(f) - if not newf: - continue - filenames.append(newf) - try: - data = self.source.getfile(f, v) - except IOError, inst: - self.dest.delfile(newf) - else: - e = self.source.getmode(f, v) - self.dest.putfile(newf, e, data) - if do_copies: - if f in copies: - copyf = self.mapfile(copies[f]) - if copyf: - # Merely marks that a copy happened. - self.dest.copyfile(copyf, newf) - - if not filenames and self.mapfile.active(): - newnode = parents[0] - else: - newnode = self.dest.putcommit(filenames, parents, commit) - self.mapentry(rev, newnode) - - def convert(self): - try: - self.source.before() - self.dest.before() - self.source.setrevmap(self.map) - self.ui.status("scanning source...\n") - heads = self.source.getheads() - parents = self.walktree(heads) - self.ui.status("sorting...\n") - t = self.toposort(parents) - num = len(t) - c = None - - self.ui.status("converting...\n") - for c in t: - num -= 1 - desc = self.commitcache[c].desc - if "\n" in desc: - desc = desc.splitlines()[0] - self.ui.status("%d %s\n" % (num, desc)) - self.copy(c) - - tags = self.source.gettags() - ctags = {} - for k in tags: - v = tags[k] - if v in self.map: - ctags[k] = self.map[v] - - if c and ctags: - nrev = self.dest.puttags(ctags) - # write another hash correspondence to override the previous - # one so we don't end up with extra tag heads - if nrev: - self.mapentry(c, nrev) - - self.writeauthormap() - finally: - self.cleanup() - - def cleanup(self): - try: - self.dest.after() - finally: - self.source.after() - if self.revmapfilefd: - self.revmapfilefd.close() +import shlex +from mercurial.i18n import _ +from mercurial import util +from common import SKIPREV def rpairs(name): e = len(name) @@ -368,122 +95,236 @@ class filemapper(object): def active(self): return bool(self.include or self.exclude or self.rename) -def convert(ui, src, dest=None, revmapfile=None, **opts): - """Convert a foreign SCM repository to a Mercurial one. - - Accepted source formats: - - CVS - - Darcs - - git - - Subversion - - Accepted destination formats: - - Mercurial +# This class does two additional things compared to a regular source: +# +# - Filter and rename files. This is mostly wrapped by the filemapper +# class above. We hide the original filename in the revision that is +# returned by getchanges to be able to find things later in getfile +# and getmode. +# +# - Return only revisions that matter for the files we're interested in. +# This involves rewriting the parents of the original revision to +# create a graph that is restricted to those revisions. +# +# This set of revisions includes not only revisions that directly +# touch files we're interested in, but also merges that merge two +# or more interesting revisions. - If no revision is given, all revisions will be converted. Otherwise, - convert will only import up to the named revision (given in a format - understood by the source). - - If no destination directory name is specified, it defaults to the - basename of the source with '-hg' appended. If the destination - repository doesn't exist, it will be created. - - If <revmapfile> isn't given, it will be put in a default location - (<dest>/.hg/shamap by default). The <revmapfile> is a simple text - file that maps each source commit ID to the destination ID for - that revision, like so: - <source ID> <destination ID> - - If the file doesn't exist, it's automatically created. It's updated - on each commit copied, so convert-repo can be interrupted and can - be run repeatedly to copy new commits. +class filemap_source(object): + def __init__(self, ui, baseconverter, filemap): + self.ui = ui + self.base = baseconverter + self.filemapper = filemapper(ui, filemap) + self.commits = {} + # if a revision rev has parent p in the original revision graph, then + # rev will have parent self.parentmap[p] in the restricted graph. + self.parentmap = {} + # self.wantedancestors[rev] is the set of all ancestors of rev that + # are in the restricted graph. + self.wantedancestors = {} + self.convertedorder = None + self._rebuilt = False + self.origparents = {} - The [username mapping] file is a simple text file that maps each source - commit author to a destination commit author. It is handy for source SCMs - that use unix logins to identify authors (eg: CVS). One line per author - mapping and the line format is: - srcauthor=whatever string you want - - The filemap is a file that allows filtering and remapping of files - and directories. Comment lines start with '#'. Each line can - contain one of the following directives: - - include path/to/file - - exclude path/to/file + def setrevmap(self, revmap, order): + # rebuild our state to make things restartable + # + # To avoid calling getcommit for every revision that has already + # been converted, we rebuild only the parentmap, delaying the + # rebuild of wantedancestors until we need it (i.e. until a + # merge). + # + # We assume the order argument lists the revisions in + # topological order, so that we can infer which revisions were + # wanted by previous runs. + self._rebuilt = not revmap + seen = {SKIPREV: SKIPREV} + dummyset = util.set() + converted = [] + for rev in order: + mapped = revmap[rev] + wanted = mapped not in seen + if wanted: + seen[mapped] = rev + self.parentmap[rev] = rev + else: + self.parentmap[rev] = seen[mapped] + self.wantedancestors[rev] = dummyset + arg = seen[mapped] + if arg == SKIPREV: + arg = None + converted.append((rev, wanted, arg)) + self.convertedorder = converted + return self.base.setrevmap(revmap, order) - rename from/file to/file - - The 'include' directive causes a file, or all files under a - directory, to be included in the destination repository. The - 'exclude' directive causes files or directories to be omitted. - The 'rename' directive renames a file or directory. To rename - from a subdirectory into the root of the repository, use '.' as - the path to rename to. - """ + def rebuild(self): + if self._rebuilt: + return True + self._rebuilt = True + pmap = self.parentmap.copy() + self.parentmap.clear() + self.wantedancestors.clear() + for rev, wanted, arg in self.convertedorder: + parents = self.origparents.get(rev) + if parents is None: + parents = self.base.getcommit(rev).parents + if wanted: + self.mark_wanted(rev, parents) + else: + self.mark_not_wanted(rev, arg) + + assert pmap == self.parentmap + return True + + def getheads(self): + return self.base.getheads() - util._encoding = 'UTF-8' + def getcommit(self, rev): + # We want to save a reference to the commit objects to be able + # to rewrite their parents later on. + self.commits[rev] = self.base.getcommit(rev) + return self.commits[rev] - if not dest: - dest = hg.defaultdest(src) + "-hg" - ui.status("assuming destination %s\n" % dest) + def wanted(self, rev, i): + # Return True if we're directly interested in rev. + # + # i is an index selecting one of the parents of rev (if rev + # has no parents, i is None). getchangedfiles will give us + # the list of files that are different in rev and in the parent + # indicated by i. If we're interested in any of these files, + # we're interested in rev. + try: + files = self.base.getchangedfiles(rev, i) + except NotImplementedError: + raise util.Abort(_("source repository doesn't support --filemap")) + for f in files: + if self.filemapper(f): + return True + return False + + def mark_not_wanted(self, rev, p): + # Mark rev as not interesting and update data structures. - # Try to be smart and initalize things when required - created = False - if os.path.isdir(dest): - if len(os.listdir(dest)) > 0: - try: - hg.repository(ui, dest) - ui.status("destination %s is a Mercurial repository\n" % dest) - except hg.RepoError: - raise util.Abort( - "destination directory %s is not empty.\n" - "Please specify an empty directory to be initialized\n" - "or an already initialized mercurial repository" - % dest) - else: - ui.status("initializing destination %s repository\n" % dest) - hg.repository(ui, dest, create=True) - created = True - elif os.path.exists(dest): - raise util.Abort("destination %s exists and is not a directory" % dest) - else: - ui.status("initializing destination %s repository\n" % dest) - hg.repository(ui, dest, create=True) - created = True + if p is None: + # A root revision. Use SKIPREV to indicate that it doesn't + # map to any revision in the restricted graph. Put SKIPREV + # in the set of wanted ancestors to simplify code elsewhere + self.parentmap[rev] = SKIPREV + self.wantedancestors[rev] = util.set((SKIPREV,)) + return + + # Reuse the data from our parent. + self.parentmap[rev] = self.parentmap[p] + self.wantedancestors[rev] = self.wantedancestors[p] + + def mark_wanted(self, rev, parents): + # Mark rev ss wanted and update data structures. + + # rev will be in the restricted graph, so children of rev in + # the original graph should still have rev as a parent in the + # restricted graph. + self.parentmap[rev] = rev + + # The set of wanted ancestors of rev is the union of the sets + # of wanted ancestors of its parents. Plus rev itself. + wrev = util.set() + for p in parents: + wrev.update(self.wantedancestors[p]) + wrev.add(rev) + self.wantedancestors[rev] = wrev - destc = convertsink(ui, dest) + def getchanges(self, rev): + parents = self.commits[rev].parents + if len(parents) > 1: + self.rebuild() + + # To decide whether we're interested in rev we: + # + # - calculate what parents rev will have if it turns out we're + # interested in it. If it's going to have more than 1 parent, + # we're interested in it. + # + # - otherwise, we'll compare it with the single parent we found. + # If any of the files we're interested in is different in the + # the two revisions, we're interested in rev. + + # A parent p is interesting if its mapped version (self.parentmap[p]): + # - is not SKIPREV + # - is still not in the list of parents (we don't want duplicates) + # - is not an ancestor of the mapped versions of the other parents + mparents = [] + wp = None + for i, p1 in enumerate(parents): + mp1 = self.parentmap[p1] + if mp1 == SKIPREV or mp1 in mparents: + continue + for p2 in parents: + if p1 == p2 or mp1 == self.parentmap[p2]: + continue + if mp1 in self.wantedancestors[p2]: + break + else: + mparents.append(mp1) + wp = i - try: - srcc = convertsource(ui, src, rev=opts.get('rev')) - except Exception: - if created: - shutil.rmtree(dest, True) - raise + if wp is None and parents: + wp = 0 + + self.origparents[rev] = parents - if not revmapfile: - try: - revmapfile = destc.revmapfile() - except: - revmapfile = os.path.join(destc, "map") + if len(mparents) < 2 and not self.wanted(rev, wp): + # We don't want this revision. + # Update our state and tell the convert process to map this + # revision to the same revision its parent as mapped to. + p = None + if parents: + p = parents[wp] + self.mark_not_wanted(rev, p) + self.convertedorder.append((rev, False, p)) + return self.parentmap[rev] + # We want this revision. + # Rewrite the parents of the commit object + self.commits[rev].parents = mparents + self.mark_wanted(rev, parents) + self.convertedorder.append((rev, True, None)) - c = converter(ui, srcc, destc, revmapfile, filemapper(ui, opts['filemap']), - opts) - c.convert() - + # Get the real changes and do the filtering/mapping. + # To be able to get the files later on in getfile and getmode, + # we hide the original filename in the rev part of the return + # value. + changes, copies = self.base.getchanges(rev) + newnames = {} + files = [] + for f, r in changes: + newf = self.filemapper(f) + if newf: + files.append((newf, (f, r))) + newnames[f] = newf -cmdtable = { - "convert": - (convert, - [('A', 'authors', '', 'username mapping filename'), - ('', 'filemap', '', 'remap file names using contents of file'), - ('r', 'rev', '', 'import up to target revision REV'), - ('', 'datesort', None, 'try to sort changesets by date')], - 'hg convert [OPTION]... SOURCE [DEST [MAPFILE]]'), - "debugsvnlog": - (debugsvnlog, - [], - 'hg debugsvnlog'), -} + ncopies = {} + for c in copies: + newc = self.filemapper(c) + if newc: + newsource = self.filemapper(copies[c]) + if newsource: + ncopies[newc] = newsource + + return files, ncopies + def getfile(self, name, rev): + realname, realrev = rev + return self.base.getfile(realname, realrev) + + def getmode(self, name, rev): + realname, realrev = rev + return self.base.getmode(realname, realrev) + + def gettags(self): + return self.base.gettags() + + def before(self): + pass + + def after(self): + pass
--- a/hgext/convert/git.py +++ b/hgext/convert/git.py @@ -122,3 +122,21 @@ class convert_git(converter_source): tags[tag] = node return tags + + def getchangedfiles(self, version, i): + changes = [] + if i is None: + fh = self.gitcmd("git-diff-tree --root -m -r %s" % version) + for l in fh: + if "\t" not in l: + continue + m, f = l[:-1].split("\t") + changes.append(f) + fh.close() + else: + fh = self.gitcmd("git-diff-tree --name-only --root -r %s %s^%s --" + % (version, version, i+1)) + changes = [f.rstrip('\n') for f in fh] + fh.close() + + return changes
--- a/hgext/convert/hg.py +++ b/hgext/convert/hg.py @@ -28,6 +28,7 @@ class mercurial_sink(converter_sink): raise NoRepo("could not open hg repo %s as sink" % path) self.lock = None self.wlock = None + self.filemapmode = False def before(self): self.wlock = self.repo.wlock() @@ -96,6 +97,10 @@ class mercurial_sink(converter_sink): pl.append(p) seen[p] = 1 parents = pl + nparents = len(parents) + if self.filemapmode and nparents == 1: + m1node = self.repo.changelog.read(bin(parents[0]))[0] + parent = parents[0] if len(parents) < 2: parents.append("0" * 40) if len(parents) < 2: parents.append("0" * 40) @@ -117,6 +122,13 @@ class mercurial_sink(converter_sink): text = "(octopus merge fixup)\n" p2 = hg.hex(self.repo.changelog.tip()) + if self.filemapmode and nparents == 1: + man = self.repo.manifest + mnode = self.repo.changelog.read(bin(p2))[0] + if not man.cmp(m1node, man.revision(mnode)): + self.repo.rollback() + self.repo.dirstate.clear() + return parent return p2 def puttags(self, tags): @@ -153,6 +165,9 @@ class mercurial_sink(converter_sink): date, tagparent, nullid) return hex(self.repo.changelog.tip()) + def setfilemapmode(self, active): + self.filemapmode = active + class mercurial_source(converter_source): def __init__(self, ui, path, rev=None): converter_source.__init__(self, ui, path, rev) @@ -162,6 +177,7 @@ class mercurial_source(converter_source) raise NoRepo("could not open hg repo %s as source" % path) self.lastrev = None self.lastctx = None + self._changescache = None def changectx(self, rev): if self.lastrev != rev: @@ -187,7 +203,10 @@ class mercurial_source(converter_source) def getchanges(self, rev): ctx = self.changectx(rev) - m, a, r = self.repo.status(ctx.parents()[0].node(), ctx.node())[:3] + if self._changescache and self._changescache[0] == rev: + m, a, r = self._changescache[1] + else: + m, a, r = self.repo.status(ctx.parents()[0].node(), ctx.node())[:3] changes = [(name, rev) for name in m + a + r] changes.sort() return (changes, self.getcopies(ctx, m + a)) @@ -211,3 +230,14 @@ class mercurial_source(converter_source) def gettags(self): tags = [t for t in self.repo.tagslist() if t[0] != 'tip'] return dict([(name, hex(node)) for name, node in tags]) + + def getchangedfiles(self, rev, i): + ctx = self.changectx(rev) + i = i or 0 + changes = self.repo.status(ctx.parents()[i].node(), ctx.node())[:3] + + if i == 0: + self._changescache = (rev, changes) + + return changes[0] + changes[1] + changes[2] +
--- a/hgext/convert/subversion.py +++ b/hgext/convert/subversion.py @@ -146,8 +146,9 @@ class convert_svn(converter_source): self.last_changed = self.latest(self.module, latest) self.head = self.revid(self.last_changed) + self._changescache = None - def setrevmap(self, revmap): + def setrevmap(self, revmap, order): lastrevs = {} for revid in revmap.keys(): uuid, module, revnum = self.revsplit(revid) @@ -206,6 +207,9 @@ class convert_svn(converter_source): return self.modecache[(file, rev)] def getchanges(self, rev): + if self._changescache and self._changescache[0] == rev: + return self._changescache[1] + self._changescache = None self.modecache = {} (paths, parents) = self.paths[rev] files, copies = self.expandpaths(rev, paths, parents) @@ -216,6 +220,11 @@ class convert_svn(converter_source): del self.paths[rev] return (files, copies) + def getchangedfiles(self, rev, i): + changes = self.getchanges(rev) + self._changescache = (rev, changes) + return [f[0] for f in changes[0]] + def getcommit(self, rev): if rev not in self.commits: uuid, module, revnum = self.revsplit(rev)
--- a/mercurial/bdiff.c +++ b/mercurial/bdiff.c @@ -12,6 +12,7 @@ #include <Python.h> #include <stdlib.h> #include <string.h> +#include <limits.h> #if defined __hpux || defined __SUNPRO_C || defined _AIX # define inline @@ -58,21 +59,17 @@ struct hunklist { struct hunk *base, *head; }; -static inline uint32_t rol32(uint32_t word, unsigned int shift) -{ - return (word << shift) | (word >> (32 - shift)); -} - int splitlines(const char *a, int len, struct line **lr) { - int g, h, i; + int h, i; const char *p, *b = a; + const char * const plast = a + len - 1; struct line *l; /* count the lines */ i = 1; /* extra line for sentinel */ for (p = a; p < a + len; p++) - if (*p == '\n' || p == a + len - 1) + if (*p == '\n' || p == plast) i++; *lr = l = (struct line *)malloc(sizeof(struct line) * i); @@ -82,24 +79,17 @@ int splitlines(const char *a, int len, s /* build the line array and calculate hashes */ h = 0; for (p = a; p < a + len; p++) { - /* - * a simple hash from GNU diff, with better collision - * resistance from hashpjw. this slows down common - * case by 10%, but speeds up worst case by 100x. - */ - h = *p + rol32(h, 7); - if ((g = h & 0xf0000000)) { - h ^= g >> 24; - h ^= g; - } - if (*p == '\n' || p == a + len - 1) { + /* Leonid Yuriev's hash */ + h = (h * 1664525) + *p + 1013904223; + + if (*p == '\n' || p == plast) { + l->h = h; + h = 0; l->len = p - b + 1; - l->h = h * l->len; l->l = b; - l->n = -1; + l->n = INT_MAX; l++; b = p + 1; - h = 0; } } @@ -117,27 +107,34 @@ int inline cmp(struct line *a, struct li static int equatelines(struct line *a, int an, struct line *b, int bn) { int i, j, buckets = 1, t; + int scale = 32; struct pos *h; /* build a hash table of the next highest power of 2 */ while (buckets < bn + 1) buckets *= 2; - h = (struct pos *)malloc(buckets * sizeof(struct pos)); - buckets = buckets - 1; + /* try to allocate a large hash table to avoid collisions */ + do { + scale /= 2; + h = (struct pos *)malloc(scale * buckets * sizeof(struct pos)); + } while (!h && scale != 1); + if (!h) return 0; + buckets = buckets * scale - 1; + /* clear the hash table */ for (i = 0; i <= buckets; i++) { - h[i].pos = -1; + h[i].pos = INT_MAX; h[i].len = 0; } /* add lines to the hash table chains */ for (i = bn - 1; i >= 0; i--) { /* find the equivalence class */ - for (j = b[i].h & buckets; h[j].pos != -1; + for (j = b[i].h & buckets; h[j].pos != INT_MAX; j = (j + 1) & buckets) if (!cmp(b + i, b + h[j].pos)) break; @@ -155,7 +152,7 @@ static int equatelines(struct line *a, i /* match items in a to their equivalence class in b */ for (i = 0; i < an; i++) { /* find the equivalence class */ - for (j = a[i].h & buckets; h[j].pos != -1; + for (j = a[i].h & buckets; h[j].pos != INT_MAX; j = (j + 1) & buckets) if (!cmp(a + i, b + h[j].pos)) break; @@ -164,7 +161,7 @@ static int equatelines(struct line *a, i if (h[j].len <= t) a[i].n = h[j].pos; /* point to head of match list */ else - a[i].n = -1; /* too popular */ + a[i].n = INT_MAX; /* too popular */ } /* discard hash tables */ @@ -179,11 +176,11 @@ static int longest_match(struct line *a, for (i = a1; i < a2; i++) { /* skip things before the current block */ - for (j = a[i].n; j != -1 && j < b1; j = b[j].n) + for (j = a[i].n; j < b1; j = b[j].n) ; /* loop through all lines match a[i] in b */ - for (; j != -1 && j < b2; j = b[j].n) { + for (; j < b2; j = b[j].n) { /* does this extend an earlier match? */ if (i > a1 && j > b1 && pos[j - 1].pos == i - 1) k = pos[j - 1].len + 1; @@ -216,6 +213,7 @@ static int longest_match(struct line *a, *omi = mi - mb; *omj = mj - mb; + return mk + mb; }
--- a/mercurial/changegroup.py +++ b/mercurial/changegroup.py @@ -33,10 +33,9 @@ def chunkiter(source): break yield c -def genchunk(data): - """build a changegroup chunk""" - header = struct.pack(">l", len(data)+ 4) - return "%s%s" % (header, data) +def chunkheader(length): + """build a changegroup chunk header""" + return struct.pack(">l", length + 4) def closechunk(): return struct.pack(">l", 0) @@ -86,7 +85,12 @@ def writebundle(cg, filename, bundletype empty = True for chunk in chunkiter(cg): empty = False - fh.write(z.compress(genchunk(chunk))) + fh.write(z.compress(chunkheader(len(chunk)))) + pos = 0 + while pos < len(chunk): + next = pos + 2**20 + fh.write(z.compress(chunk[pos:next])) + pos = next fh.write(z.compress(closechunk())) fh.write(z.flush()) cleanup = None
--- a/mercurial/localrepo.py +++ b/mercurial/localrepo.py @@ -1720,7 +1720,8 @@ class localrepository(repo.repository): # If any filenodes are left, generate the group for them, # otherwise don't bother. if len(msng_filenode_lst) > 0: - yield changegroup.genchunk(fname) + yield changegroup.chunkheader(len(fname)) + yield fname # Sort the filenodes by their revision # msng_filenode_lst.sort(cmp_by_rev_func(filerevlog)) # Create a group generator and only pass in a changenode @@ -1796,7 +1797,8 @@ class localrepository(repo.repository): nodeiter = gennodelst(filerevlog) nodeiter = list(nodeiter) if nodeiter: - yield changegroup.genchunk(fname) + yield changegroup.chunkheader(len(fname)) + yield fname lookup = lookuprevlink_func(filerevlog) for chnk in filerevlog.group(nodeiter, lookup): yield chnk
--- a/mercurial/mdiff.py +++ b/mercurial/mdiff.py @@ -245,6 +245,9 @@ def patch(a, bin): def get_matching_blocks(a, b): return [(d[0], d[2], d[1] - d[0]) for d in bdiff.blocks(a, b)] +def trivialdiffheader(length): + return struct.pack(">lll", 0, 0, length) + patches = mpatch.patches patchedsize = mpatch.patchedsize textdiff = bdiff.bdiff
--- a/mercurial/merge.py +++ b/mercurial/merge.py @@ -188,9 +188,18 @@ def findcopies(repo, m1, m2, ma, limit): if not m1 or not m2 or not ma: return {}, {} + repo.ui.debug(_(" searching for copies back to rev %d\n") % limit) + u1 = nonoverlap(m1, m2, ma) u2 = nonoverlap(m2, m1, ma) + if u1: + repo.ui.debug(_(" unmatched files in local:\n %s\n") + % "\n ".join(u1)) + if u2: + repo.ui.debug(_(" unmatched files in other:\n %s\n") + % "\n ".join(u2)) + for f in u1: checkcopies(ctx(f, m1[f]), m2, ma) @@ -204,9 +213,19 @@ def findcopies(repo, m1, m2, ma, limit): fo.remove(f) d2[f] = (of, fo) + if fullcopy: + repo.ui.debug(_(" all copies found (* = to merge, ! = divergent):\n")) + for f in fullcopy: + note = "" + if f in copy: note += "*" + if f in diverge: note += "!" + repo.ui.debug(_(" %s -> %s %s\n") % (f, fullcopy[f], note)) + if not fullcopy or not repo.ui.configbool("merge", "followdirs", True): return copy, diverge + repo.ui.debug(_(" checking for directory renames\n")) + # generate a directory move map d1, d2 = dirs(m1), dirs(m2) invalid = {} @@ -241,6 +260,9 @@ def findcopies(repo, m1, m2, ma, limit): if not dirmove: return copy, diverge + for d in dirmove: + repo.ui.debug(_(" dir %s -> %s\n") % (d, dirmove[d])) + # check unaccounted nonoverlapping files against directory moves for f in u1 + u2: if f not in fullcopy: @@ -248,6 +270,7 @@ def findcopies(repo, m1, m2, ma, limit): if f.startswith(d): # new file added in a directory that was moved, move it copy[f] = dirmove[d] + f[len(d):] + repo.ui.debug(_(" file %s -> %s\n") % (f, copy[f])) break return copy, diverge
--- a/mercurial/revlog.py +++ b/mercurial/revlog.py @@ -1087,10 +1087,16 @@ class revlog(object): if infocollect is not None: infocollect(nb) - d = self.revdiff(a, b) p = self.parents(nb) meta = nb + p[0] + p[1] + lookup(nb) - yield changegroup.genchunk("%s%s" % (meta, d)) + if a == -1: + d = self.revision(nb) + meta += mdiff.trivialdiffheader(len(d)) + else: + d = self.revdiff(a, b) + yield changegroup.chunkheader(len(meta) + len(d)) + yield meta + yield d yield changegroup.closechunk()
--- a/tests/test-convert-cvs +++ b/tests/test-convert-cvs @@ -29,16 +29,30 @@ cd .. echo % checkout source directory cvs -q checkout src +echo % commit a new revision changing b/c +cd src +echo c >> b/c +cvs -q commit -mci0 . | grep '<--' |\ + sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g' +cd .. + echo % convert fresh repo hg convert src src-hg | sed -e 's/connecting to.*cvsrepo/connecting to cvsrepo/g' cat src-hg/a cat src-hg/b/c +echo % convert fresh repo with --filemap +echo include b/c > filemap +hg convert --filemap filemap src src-filemap | sed -e 's/connecting to.*cvsrepo/connecting to cvsrepo/g' +cat src-hg/b/c +hg -R src-filemap log --template '#rev# #desc# files: #files#\n' + echo % commit new file revisions cd src echo a >> a echo c >> b/c -cvs -q commit -mci1 . | sed -e 's:.*src/\(.*\),v:src/\1,v:g' +cvs -q commit -mci1 . | grep '<--' |\ + sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g' cd .. echo % convert again @@ -46,5 +60,8 @@ hg convert src src-hg | sed -e 's/connec cat src-hg/a cat src-hg/b/c +echo % convert again with --filemap +hg convert --filemap filemap src src-filemap | sed -e 's/connecting to.*cvsrepo/connecting to cvsrepo/g' +cat src-hg/b/c +hg -R src-filemap log --template '#rev# #desc# files: #files#\n' -
--- a/tests/test-convert-cvs.out +++ b/tests/test-convert-cvs.out @@ -9,22 +9,40 @@ No conflicts created by this import % checkout source directory U src/a U src/b/c +% commit a new revision changing b/c +checking in src/b/c,v % convert fresh repo initializing destination src-hg repository connecting to cvsrepo scanning source... sorting... converting... -1 Initial revision -0 import +2 Initial revision +1 import +0 ci0 updating tags a c +c +% convert fresh repo with --filemap +initializing destination src-filemap repository +connecting to cvsrepo +scanning source... +sorting... +converting... +2 Initial revision +1 import +rolling back last transaction +0 ci0 +updating tags +c +c +2 update tags files: .hgtags +1 ci0 files: b/c +0 Initial revision files: b/c % commit new file revisions -src/a,v <-- a -new revision: 1.2; previous revision: 1.1 -src/b/c,v <-- b/c -new revision: 1.2; previous revision: 1.1 +checking in src/a,v +checking in src/b/c,v % convert again destination src-hg is a Mercurial repository connecting to cvsrepo @@ -32,8 +50,22 @@ scanning source... sorting... converting... 0 ci1 -updating tags a a c c +c +% convert again with --filemap +destination src-filemap is a Mercurial repository +connecting to cvsrepo +scanning source... +sorting... +converting... +0 ci1 +c +c +c +3 ci1 files: b/c +2 update tags files: .hgtags +1 ci0 files: b/c +0 Initial revision files: b/c
new file mode 100755 --- /dev/null +++ b/tests/test-convert-filemap @@ -0,0 +1,124 @@ +#!/bin/sh + +HGMERGE=true; export HGMERGE + +echo '[extensions]' >> $HGRCPATH +echo 'hgext.graphlog =' >> $HGRCPATH +echo 'hgext.convert =' >> $HGRCPATH + +glog() +{ + hg glog --template '#rev# "#desc#" files: #files#\n' "$@" +} + +hg init source +cd source + +echo foo > foo +echo baz > baz +mkdir dir +echo dir/file >> dir/file +echo dir/file2 >> dir/file2 +hg ci -d '0 0' -qAm '0: add foo baz dir/' + +echo bar > bar +echo quux > quux +hg copy foo copied +hg ci -d '1 0' -qAm '1: add bar quux; copy foo to copied' + +echo >> foo +hg ci -d '2 0' -m '2: change foo' + +hg up -qC 1 +echo >> bar +echo >> quux +hg ci -d '3 0' -m '3: change bar quux' + +hg up -qC 2 +hg merge -qr 3 +echo >> bar +echo >> baz +hg ci -d '4 0' -m '4: first merge; change bar baz' + +echo >> bar +echo 1 >> baz +echo >> quux +hg ci -d '5 0' -m '5: change bar baz quux' + +hg up -qC 4 +echo >> foo +echo 2 >> baz +hg ci -d '6 0' -m '6: change foo baz' + +hg up -qC 5 +hg merge -qr 6 +echo >> bar +hg ci -d '7 0' -m '7: second merge; change bar' + +echo >> foo +hg ci -m '8: change foo' + +glog + +echo '% final file versions in this repo:' +hg manifest --debug +hg debugrename copied +echo + +cd .. + +splitrepo() +{ + msg="$1" + files="$2" + opts=$3 + echo "% $files: $msg" + prefix=`echo "$files" | sed -e 's/ /-/g'` + fmap="$prefix.fmap" + repo="$prefix.repo" + for i in $files; do + echo "include $i" >> "$fmap" + done + hg -q convert $opts --filemap "$fmap" --datesort source "$repo" + glog -R "$repo" + hg -R "$repo" manifest --debug +} + +splitrepo 'skip unwanted merges; use 1st parent in 1st merge, 2nd in 2nd' foo + +splitrepo 'merges are not merges anymore' bar + +splitrepo '1st merge is not a merge anymore; 2nd still is' baz + +splitrepo 'we add additional merges when they are interesting' 'foo quux' + +splitrepo 'partial conversion' 'bar quux' '-r 3' +splitrepo 'complete the partial conversion' 'bar quux' + +rm -r foo.repo +splitrepo 'partial conversion' 'foo' '-r 3' +splitrepo 'complete the partial conversion' 'foo' + +splitrepo 'copied file; source not included in new repo' copied +hg --cwd copied.repo debugrename copied + +splitrepo 'copied file; source included in new repo' 'foo copied' +hg --cwd foo-copied.repo debugrename copied + +cat > renames.fmap <<EOF +include dir +exclude dir/file2 +rename dir dir2 +include foo +include copied +rename foo foo2 +rename copied copied2 +EOF +hg -q convert --filemap renames.fmap --datesort source renames.repo +glog -R renames.repo +hg -R renames.repo manifest --debug +hg --cwd renames.repo debugrename copied2 +echo 'copied:' +hg --cwd source cat copied +echo 'copied2:' +hg --cwd renames.repo cat copied2
new file mode 100644 --- /dev/null +++ b/tests/test-convert-filemap.out @@ -0,0 +1,154 @@ +@ 8 "8: change foo" files: foo +| +o 7 "7: second merge; change bar" files: bar baz +|\ +| o 6 "6: change foo baz" files: baz foo +| | +o | 5 "5: change bar baz quux" files: bar baz quux +|/ +o 4 "4: first merge; change bar baz" files: bar baz +|\ +| o 3 "3: change bar quux" files: bar quux +| | +o | 2 "2: change foo" files: foo +|/ +o 1 "1: add bar quux; copy foo to copied" files: bar copied quux +| +o 0 "0: add foo baz dir/" files: baz dir/file dir/file2 foo + +% final file versions in this repo: +9463f52fe115e377cf2878d4fc548117211063f2 644 bar +94c1be4dfde2ee8d78db8bbfcf81210813307c3d 644 baz +6ca237634e1f6bee1b6db94292fb44f092a25842 644 copied +3e20847584beff41d7cd16136b7331ab3d754be0 644 dir/file +75e6d3f8328f5f6ace6bf10b98df793416a09dca 644 dir/file2 +9a7b52012991e4873687192c3e17e61ba3e837a3 644 foo +bc3eca3f47023a3e70ca0d8cc95a22a6827db19d 644 quux +copied renamed from foo:2ed2a3912a0b24502043eae84ee4b279c18b90dd + +% foo: skip unwanted merges; use 1st parent in 1st merge, 2nd in 2nd +o 3 "8: change foo" files: foo +| +o 2 "6: change foo baz" files: foo +| +o 1 "2: change foo" files: foo +| +o 0 "0: add foo baz dir/" files: foo + +9a7b52012991e4873687192c3e17e61ba3e837a3 644 foo +% bar: merges are not merges anymore +o 4 "7: second merge; change bar" files: bar +| +o 3 "5: change bar baz quux" files: bar +| +o 2 "4: first merge; change bar baz" files: bar +| +o 1 "3: change bar quux" files: bar +| +o 0 "1: add bar quux; copy foo to copied" files: bar + +9463f52fe115e377cf2878d4fc548117211063f2 644 bar +% baz: 1st merge is not a merge anymore; 2nd still is +o 4 "7: second merge; change bar" files: baz +|\ +| o 3 "6: change foo baz" files: baz +| | +o | 2 "5: change bar baz quux" files: baz +|/ +o 1 "4: first merge; change bar baz" files: baz +| +o 0 "0: add foo baz dir/" files: baz + +94c1be4dfde2ee8d78db8bbfcf81210813307c3d 644 baz +% foo quux: we add additional merges when they are interesting +o 8 "8: change foo" files: foo +| +o 7 "7: second merge; change bar" files: +|\ +| o 6 "6: change foo baz" files: foo +| | +o | 5 "5: change bar baz quux" files: quux +|/ +o 4 "4: first merge; change bar baz" files: +|\ +| o 3 "3: change bar quux" files: quux +| | +o | 2 "2: change foo" files: foo +|/ +o 1 "1: add bar quux; copy foo to copied" files: quux +| +o 0 "0: add foo baz dir/" files: foo + +9a7b52012991e4873687192c3e17e61ba3e837a3 644 foo +bc3eca3f47023a3e70ca0d8cc95a22a6827db19d 644 quux +% bar quux: partial conversion +o 1 "3: change bar quux" files: bar quux +| +o 0 "1: add bar quux; copy foo to copied" files: bar quux + +b79105bedc55102f394e90a789c9c380117c1b4a 644 bar +db0421cc6b685a458c8d86c7d5c004f94429ea23 644 quux +% bar quux: complete the partial conversion +o 4 "7: second merge; change bar" files: bar +| +o 3 "5: change bar baz quux" files: bar quux +| +o 2 "4: first merge; change bar baz" files: bar +| +o 1 "3: change bar quux" files: bar quux +| +o 0 "1: add bar quux; copy foo to copied" files: bar quux + +9463f52fe115e377cf2878d4fc548117211063f2 644 bar +bc3eca3f47023a3e70ca0d8cc95a22a6827db19d 644 quux +% foo: partial conversion +o 0 "0: add foo baz dir/" files: foo + +2ed2a3912a0b24502043eae84ee4b279c18b90dd 644 foo +% foo: complete the partial conversion +o 3 "8: change foo" files: foo +| +o 2 "6: change foo baz" files: foo +| +o 1 "2: change foo" files: foo +| +o 0 "0: add foo baz dir/" files: foo + +9a7b52012991e4873687192c3e17e61ba3e837a3 644 foo +% copied: copied file; source not included in new repo +o 0 "1: add bar quux; copy foo to copied" files: copied + +2ed2a3912a0b24502043eae84ee4b279c18b90dd 644 copied +copied not renamed +% foo copied: copied file; source included in new repo +o 4 "8: change foo" files: foo +| +o 3 "6: change foo baz" files: foo +| +o 2 "2: change foo" files: foo +| +o 1 "1: add bar quux; copy foo to copied" files: copied +| +o 0 "0: add foo baz dir/" files: foo + +6ca237634e1f6bee1b6db94292fb44f092a25842 644 copied +9a7b52012991e4873687192c3e17e61ba3e837a3 644 foo +copied renamed from foo:2ed2a3912a0b24502043eae84ee4b279c18b90dd +o 4 "8: change foo" files: foo2 +| +o 3 "6: change foo baz" files: foo2 +| +o 2 "2: change foo" files: foo2 +| +o 1 "1: add bar quux; copy foo to copied" files: copied2 +| +o 0 "0: add foo baz dir/" files: dir2/file foo2 + +e5e3d520be9be45937d0b06b004fadcd6c221fa2 644 copied2 +3e20847584beff41d7cd16136b7331ab3d754be0 644 dir2/file +9a7b52012991e4873687192c3e17e61ba3e837a3 644 foo2 +copied2 renamed from foo2:2ed2a3912a0b24502043eae84ee4b279c18b90dd +copied: +foo +copied2: +foo
--- a/tests/test-convert-git +++ b/tests/test-convert-git @@ -4,6 +4,7 @@ echo "[extensions]" >> $HGRCPATH echo "convert=" >> $HGRCPATH +echo 'hgext.graphlog =' >> $HGRCPATH GIT_AUTHOR_NAME='test'; export GIT_AUTHOR_NAME GIT_AUTHOR_EMAIL='test@example.org'; export GIT_AUTHOR_EMAIL @@ -32,7 +33,7 @@ commit -a -m t1 # Remove the directory, then try to replace it with a file # (issue 754) -git rm -r d +git rm -f d/b commit -m t2 echo d > d git add d @@ -54,3 +55,78 @@ cd .. hg convert --datesort git-repo hg -R git-repo-hg tip -v + +count=10 +mkdir git-repo2 +cd git-repo2 +git init-db >/dev/null 2>/dev/null + +echo foo > foo +git add foo +commit -a -m 'add foo' + +echo >> foo +commit -a -m 'change foo' + +git checkout -b Bar HEAD^ >/dev/null 2>/dev/null +echo quux >> quux +git add quux +commit -a -m 'add quux' + +echo bar > bar +git add bar +commit -a -m 'add bar' + +git checkout -b Baz HEAD^ >/dev/null 2>/dev/null +echo baz > baz +git add baz +commit -a -m 'add baz' + +git checkout master >/dev/null 2>/dev/null +git pull --no-commit . Bar Baz > /dev/null 2>/dev/null +commit -m 'Octopus merge' + +echo bar >> bar +commit -a -m 'change bar' + +git checkout -b Foo HEAD^ >/dev/null 2>/dev/null +echo >> foo +commit -a -m 'change foo' + +git checkout master >/dev/null 2>/dev/null +git pull --no-commit -s ours . Foo > /dev/null 2>/dev/null +commit -m 'Discard change to foo' + +cd .. + +glog() +{ + hg glog --template '#rev# "#desc|firstline#" files: #files#\n' "$@" +} + +splitrepo() +{ + msg="$1" + files="$2" + opts=$3 + echo "% $files: $msg" + prefix=`echo "$files" | sed -e 's/ /-/g'` + fmap="$prefix.fmap" + repo="$prefix.repo" + for i in $files; do + echo "include $i" >> "$fmap" + done + hg -q convert $opts --filemap "$fmap" --datesort git-repo2 "$repo" + glog -R "$repo" + hg -R "$repo" manifest --debug +} + +echo '% full conversion' +hg -q convert --datesort git-repo2 fullrepo +glog -R fullrepo +hg -R fullrepo manifest --debug + +splitrepo 'octopus merge' 'foo bar baz' + +splitrepo 'only some parents of an octopus merge; "discard" a head' 'foo baz quux' +
--- a/tests/test-convert-git.out +++ b/tests/test-convert-git.out @@ -23,3 +23,68 @@ Merge branch other committer: test <test@example.org> +% full conversion +o 9 "Discard change to foo" files: foo +|\ +| o 8 "change foo" files: foo +| | +o | 7 "change bar" files: bar +|/ +o 6 "(octopus merge fixup)" files: +|\ +| o 5 "Octopus merge" files: baz +| |\ +o | | 4 "add baz" files: baz +| | | ++---o 3 "add bar" files: bar +| | +o | 2 "add quux" files: quux +| | +| o 1 "change foo" files: foo +|/ +o 0 "add foo" files: foo + +245a3b8bc653999c2b22cdabd517ccb47aecafdf 644 bar +354ae8da6e890359ef49ade27b68bbc361f3ca88 644 baz +9277c9cc8dd4576fc01a17939b4351e5ada93466 644 foo +88dfeab657e8cf2cef3dec67b914f49791ae76b1 644 quux +% foo bar baz: octopus merge +o 8 "Discard change to foo" files: foo +|\ +| o 7 "change foo" files: foo +| | +o | 6 "change bar" files: bar +|/ +o 5 "(octopus merge fixup)" files: +|\ +| o 4 "Octopus merge" files: baz +| |\ +o | | 3 "add baz" files: baz +| | | ++---o 2 "add bar" files: bar +| | +| o 1 "change foo" files: foo +|/ +o 0 "add foo" files: foo + +245a3b8bc653999c2b22cdabd517ccb47aecafdf 644 bar +354ae8da6e890359ef49ade27b68bbc361f3ca88 644 baz +9277c9cc8dd4576fc01a17939b4351e5ada93466 644 foo +% foo baz quux: only some parents of an octopus merge; "discard" a head +o 6 "Discard change to foo" files: foo +| +o 5 "change foo" files: foo +| +o 4 "Octopus merge" files: +|\ +| o 3 "add baz" files: baz +| | +| o 2 "add quux" files: quux +| | +o | 1 "change foo" files: foo +|/ +o 0 "add foo" files: foo + +354ae8da6e890359ef49ade27b68bbc361f3ca88 644 baz +9277c9cc8dd4576fc01a17939b4351e5ada93466 644 foo +88dfeab657e8cf2cef3dec67b914f49791ae76b1 644 quux
--- a/tests/test-convert-svn +++ b/tests/test-convert-svn @@ -51,3 +51,10 @@ cd .. echo % test incremental conversion hg convert $svnurl +echo % test filemap +echo 'include b' > filemap +hg convert --filemap filemap $svnurl fmap +echo '[extensions]' >> $HGRCPATH +echo 'hgext.graphlog =' >> $HGRCPATH +hg glog -R fmap --template '#rev# #desc|firstline# files: #files#\n' +
--- a/tests/test-convert-svn.out +++ b/tests/test-convert-svn.out @@ -30,3 +30,15 @@ scanning source... sorting... converting... 0 changeb +% test filemap +initializing destination fmap repository +scanning source... +sorting... +converting... +2 init +1 changea +0 changeb +o 1 changeb files: b +| +o 0 changea files: b +
--- a/tests/test-copy-move-merge.out +++ b/tests/test-copy-move-merge.out @@ -2,6 +2,14 @@ 1 files updated, 0 files merged, 2 files resolving manifests overwrite None partial False ancestor 583c7b748052 local fb3948d97f07+ remote 40da226db0f0 + searching for copies back to rev 1 + unmatched files in other: + b + c + all copies found (* = to merge, ! = divergent): + c -> a * + b -> a * + checking for directory renames a: remote moved to c -> m a: remote moved to b -> m copying a to b
--- a/tests/test-double-merge.out +++ b/tests/test-double-merge.out @@ -1,6 +1,12 @@ resolving manifests overwrite None partial False ancestor 310fd17130da local 2092631ce82b+ remote 7731dad1c2b9 + searching for copies back to rev 1 + unmatched files in other: + bar + all copies found (* = to merge, ! = divergent): + bar -> foo * + checking for directory renames foo: versions differ -> m foo: remote copied to bar -> m copying foo to bar
--- a/tests/test-issue522.out +++ b/tests/test-issue522.out @@ -4,6 +4,9 @@ 1 files updated, 0 files merged, 0 files resolving manifests overwrite None partial False ancestor bbd179dfa0a7 local 71766447bdbb+ remote 4d9e78aaceee + searching for copies back to rev 1 + unmatched files in local: + bar foo: remote is newer -> g getting foo 1 files updated, 0 files merged, 0 files removed, 0 files unresolved
--- a/tests/test-issue672.out +++ b/tests/test-issue672.out @@ -4,6 +4,12 @@ 1 files updated, 0 files merged, 1 files resolving manifests overwrite None partial False ancestor 81f4b099af3d local c64f439569a9+ remote 2f8037f47a5c + searching for copies back to rev 1 + unmatched files in other: + 1a + all copies found (* = to merge, ! = divergent): + 1a -> 1 + checking for directory renames 1: other deleted -> r 1a: remote created -> g removing 1 @@ -15,6 +21,12 @@ 1 files updated, 0 files merged, 1 files resolving manifests overwrite None partial False ancestor c64f439569a9 local ac7575e3c052+ remote 746e9549ea96 + searching for copies back to rev 1 + unmatched files in local: + 1a + all copies found (* = to merge, ! = divergent): + 1a -> 1 * + checking for directory renames 1a: local moved to 1 -> m merging 1a and 1 my 1a@ac7575e3c052+ other 1@746e9549ea96 ancestor 1@81f4b099af3d @@ -24,6 +36,12 @@ 1 files updated, 0 files merged, 1 files resolving manifests overwrite None partial False ancestor c64f439569a9 local 746e9549ea96+ remote ac7575e3c052 + searching for copies back to rev 1 + unmatched files in other: + 1a + all copies found (* = to merge, ! = divergent): + 1a -> 1 * + checking for directory renames 1: remote moved to 1a -> m copying 1 to 1a merging 1 and 1a
--- a/tests/test-merge-commit.out +++ b/tests/test-merge-commit.out @@ -23,6 +23,7 @@ 0:2665aaee66e9 resolving manifests overwrite None partial False ancestor 0a3ab4856510 local 2d2f9a22c82b+ remote 7d3b554bfdf1 + searching for copies back to rev 1 bar: versions differ -> m merging bar my bar@2d2f9a22c82b+ other bar@7d3b554bfdf1 ancestor bar@0a3ab4856510 @@ -68,6 +69,7 @@ 0:2665aaee66e9 resolving manifests overwrite None partial False ancestor 0a3ab4856510 local 2d2f9a22c82b+ remote 96ab80c60897 + searching for copies back to rev 1 bar: versions differ -> m merging bar my bar@2d2f9a22c82b+ other bar@96ab80c60897 ancestor bar@0a3ab4856510
--- a/tests/test-merge7.out +++ b/tests/test-merge7.out @@ -24,6 +24,7 @@ warning: conflicts during merge. resolving manifests overwrite None partial False ancestor faaea63e63a9 local 451c744aabcc+ remote a070d41e8360 + searching for copies back to rev 1 test.txt: versions differ -> m merging test.txt my test.txt@451c744aabcc+ other test.txt@a070d41e8360 ancestor test.txt@faaea63e63a9
--- a/tests/test-rename-dir-merge.out +++ b/tests/test-rename-dir-merge.out @@ -9,6 +9,18 @@ 2 files updated, 0 files merged, 2 files resolving manifests overwrite None partial False ancestor f9b20c0d4c51 local ce36d17b18fb+ remote 55119e611c80 + searching for copies back to rev 1 + unmatched files in local: + a/c + unmatched files in other: + b/a + b/b + all copies found (* = to merge, ! = divergent): + b/a -> a/a + b/b -> a/b + checking for directory renames + dir a/ -> b/ + file a/c -> b/c a/c: remote renamed directory to b/c -> d a/b: other deleted -> r a/a: other deleted -> r @@ -34,6 +46,18 @@ 0 files updated, 0 files merged, 1 files resolving manifests overwrite None partial False ancestor f9b20c0d4c51 local 55119e611c80+ remote ce36d17b18fb + searching for copies back to rev 1 + unmatched files in local: + b/a + b/b + unmatched files in other: + a/c + all copies found (* = to merge, ! = divergent): + b/a -> a/a + b/b -> a/b + checking for directory renames + dir a/ -> b/ + file a/c -> b/c None: local renamed directory to b/c -> d getting a/c to b/c 1 files updated, 0 files merged, 0 files removed, 0 files unresolved
--- a/tests/test-rename-merge1.out +++ b/tests/test-rename-merge1.out @@ -4,6 +4,17 @@ merge resolving manifests overwrite None partial False ancestor af1939970a1c local f26ec4fc3fa3+ remote 8e765a822af2 + searching for copies back to rev 1 + unmatched files in local: + c2 + unmatched files in other: + b + b2 + all copies found (* = to merge, ! = divergent): + c2 -> a2 + b -> a * + b2 -> a2 + checking for directory renames a2: divergent renames -> dr a: remote moved to b -> m b2: remote created -> g
--- a/tests/test-up-local-change.out +++ b/tests/test-up-local-change.out @@ -17,6 +17,9 @@ summary: 1 resolving manifests overwrite False partial False ancestor 33aaa84a386b local 33aaa84a386b+ remote 802f095af299 + searching for copies back to rev 1 + unmatched files in other: + b a: versions differ -> m b: remote created -> g merging a @@ -50,6 +53,9 @@ summary: 1 resolving manifests overwrite False partial False ancestor 33aaa84a386b local 33aaa84a386b+ remote 802f095af299 + searching for copies back to rev 1 + unmatched files in other: + b a: versions differ -> m b: remote created -> g merging a @@ -100,6 +106,7 @@ failed resolving manifests overwrite False partial False ancestor 33aaa84a386b local 802f095af299+ remote 030602aee63d + searching for copies back to rev 1 a: versions differ -> m b: versions differ -> m merging a