branch: operate on branch names in local string space where possible

Previously, branch names were ideally manipulated as UTF-8 strings,
because they were stored as UTF-8 in the dirstate and the changelog
and could not be safely converted to the local encoding and back.

However, only about 80% of branch name code was actually using the
right encoding conventions. This patch uses the localstr addition to
allow working on branch names as local strings, which simplifies
handling so that the previously incorrect code becomes correct.

mercurial/archival.py

@@ -8,7 +8,7 b''
 from i18n import _
 from node import hex
 import cmdutil
-import util
+import util, encoding
 import cStringIO, os, stat, tarfile, time, zipfile
 import zlib, gzip
 
@@ -243,7 +243,7 b' def archive(repo, dest, node, kind, deco'
     if repo.ui.configbool("ui", "archivemeta", True):
         def metadata():
             base = 'repo: %s\nnode: %s\nbranch: %s\n' % (
-                repo[0].hex(), hex(node), ctx.branch())
+                repo[0].hex(), hex(node), encoding.fromlocal(ctx.branch()))
 
             tags = ''.join('tag: %s\n' % t for t in ctx.tags()
                            if repo.tagtype(t) == 'global')

mercurial/cmdutil.py

@@ -801,7 +801,6 b' class changeset_printer(object):'
         branch = ctx.branch()
         # don't show the default branch name
         if branch != 'default':
-            branch = encoding.tolocal(branch)
             self.ui.write(_("branch:      %s\n") % branch,
                           label='log.branch')
         for tag in self.repo.nodetags(changenode):
@@ -1357,8 +1356,7 b' def commitforceeditor(repo, ctx, subs):'
     if ctx.p2():
         edittext.append(_("HG: branch merge"))
     if ctx.branch():
-        edittext.append(_("HG: branch '%s'")
-                        % encoding.tolocal(ctx.branch()))
+        edittext.append(_("HG: branch '%s'") % ctx.branch())
     edittext.extend([_("HG: subrepo %s") % s for s in subs])
     edittext.extend([_("HG: added %s") % f for f in added])
     edittext.extend([_("HG: changed %s") % f for f in modified])

mercurial/commands.py

@@ -488,15 +488,14 b' def branch(ui, repo, label=None, **opts)'
         repo.dirstate.setbranch(label)
         ui.status(_('reset working directory to branch %s\n') % label)
     elif label:
-        utflabel = encoding.fromlocal(label)
-        if not opts.get('force') and utflabel in repo.branchtags():
+        if not opts.get('force') and label in repo.branchtags():
             if label not in [p.branch() for p in repo.parents()]:
                 raise util.Abort(_('a branch of the same name already exists'
                                    " (use 'hg update' to switch to it)"))
-        repo.dirstate.setbranch(utflabel)
+        repo.dirstate.setbranch(label)
         ui.status(_('marked working directory as branch %s\n') % label)
     else:
-        ui.write("%s\n" % encoding.tolocal(repo.dirstate.branch()))
+        ui.write("%s\n" % repo.dirstate.branch())
 
 def branches(ui, repo, active=False, closed=False):
     """list repository named branches
@@ -525,9 +524,8 b' def branches(ui, repo, active=False, clo'
 
     for isactive, node, tag in branches:
         if (not active) or isactive:
-            encodedtag = encoding.tolocal(tag)
             if ui.quiet:
-                ui.write("%s\n" % encodedtag)
+                ui.write("%s\n" % tag)
             else:
                 hn = repo.lookup(node)
                 if isactive:
@@ -543,10 +541,10 b' def branches(ui, repo, active=False, clo'
                     notice = _(' (inactive)')
                 if tag == repo.dirstate.branch():
                     label = 'branches.current'
-                rev = str(node).rjust(31 - encoding.colwidth(encodedtag))
+                rev = str(node).rjust(31 - encoding.colwidth(tag))
                 rev = ui.label('%s:%s' % (rev, hexfunc(hn)), 'log.changeset')
-                encodedtag = ui.label(encodedtag, label)
-                ui.write("%s %s%s\n" % (encodedtag, rev, notice))
+                tag = ui.label(tag, label)
+                ui.write("%s %s%s\n" % (tag, rev, notice))
 
 def bundle(ui, repo, fname, dest=None, **opts):
     """create a changegroup file
@@ -1830,8 +1828,7 b' def heads(ui, repo, *branchrevs, **opts)'
             heads += [repo[h] for h in ls if rev(h) in descendants]
 
     if branchrevs:
-        decode, encode = encoding.fromlocal, encoding.tolocal
-        branches = set(repo[decode(br)].branch() for br in branchrevs)
+        branches = set(repo[br].branch() for br in branchrevs)
         heads = [h for h in heads if h.branch() in branches]
 
     if not opts.get('closed'):
@@ -1844,7 +1841,7 b' def heads(ui, repo, *branchrevs, **opts)'
     if branchrevs:
         haveheads = set(h.branch() for h in heads)
         if branches - haveheads:
-            headless = ', '.join(encode(b) for b in branches - haveheads)
+            headless = ', '.join(b for b in branches - haveheads)
             msg = _('no open branch heads found on branches %s')
             if opts.get('rev'):
                 msg += _(' (started at %s)' % opts['rev'])
@@ -2209,7 +2206,7 b' def identify(ui, repo, source=None,'
             output.append(str(ctx.rev()))
 
     if repo.local() and default and not ui.quiet:
-        b = encoding.tolocal(ctx.branch())
+        b = ctx.branch()
         if b != 'default':
             output.append("(%s)" % b)
 
@@ -2219,7 +2216,7 b' def identify(ui, repo, source=None,'
             output.append(t)
 
     if branch:
-        output.append(encoding.tolocal(ctx.branch()))
+        output.append(ctx.branch())
 
     if tags:
         output.extend(ctx.tags())
@@ -2623,7 +2620,7 b' def merge(ui, repo, node=None, **opts):'
         node = opts.get('rev')
 
     if not node:
-        branch = repo.changectx(None).branch()
+        branch = repo[None].branch()
         bheads = repo.branchheads(branch)
         if len(bheads) > 2:
             raise util.Abort(_(

mercurial/context.py

@@ -7,7 +7,7 b''
 
 from node import nullid, nullrev, short, hex
 from i18n import _
-import ancestor, bdiff, error, util, subrepo, patch
+import ancestor, bdiff, error, util, subrepo, patch, encoding
 import os, errno, stat
 
 propertycache = util.propertycache
@@ -109,7 +109,7 b' class changectx(object):'
     def description(self):
         return self._changeset[4]
     def branch(self):
-        return self._changeset[5].get("branch")
+        return encoding.tolocal(self._changeset[5].get("branch"))
     def extra(self):
         return self._changeset[5]
     def tags(self):
@@ -591,9 +591,8 b' class workingctx(changectx):'
         if extra:
             self._extra = extra.copy()
         if 'branch' not in self._extra:
-            branch = self._repo.dirstate.branch()
             try:
-                branch = branch.decode('UTF-8').encode('UTF-8')
+                branch = encoding.fromlocal(self._repo.dirstate.branch())
             except UnicodeDecodeError:
                 raise util.Abort(_('branch name not in UTF-8!'))
             self._extra['branch'] = branch
@@ -715,7 +714,7 b' class workingctx(changectx):'
         assert self._clean is not None  # must call status first
         return self._clean
     def branch(self):
-        return self._extra['branch']
+        return encoding.tolocal(self._extra['branch'])
     def extra(self):
         return self._extra
 
@@ -1048,7 +1047,7 b' class memctx(object):'
     def clean(self):
         return self._status[6]
     def branch(self):
-        return self._extra['branch']
+        return encoding.tolocal(self._extra['branch'])
     def extra(self):
         return self._extra
     def flags(self, f):

mercurial/dirstate.py

@@ -7,7 +7,7 b''
 
 from node import nullid
 from i18n import _
-import util, ignore, osutil, parsers
+import util, ignore, osutil, parsers, encoding
 import struct, os, stat, errno
 import cStringIO
 
@@ -201,7 +201,7 b' class dirstate(object):'
         return [self._validate(p) for p in self._pl]
 
     def branch(self):
-        return self._branch
+        return encoding.tolocal(self._branch)
 
     def setparents(self, p1, p2=nullid):
         self._dirty = self._dirtypl = True
@@ -210,8 +210,8 b' class dirstate(object):'
     def setbranch(self, branch):
         if branch in ['tip', '.', 'null']:
             raise util.Abort(_('the name \'%s\' is reserved') % branch)
-        self._branch = branch
-        self._opener("branch", "w").write(branch + '\n')
+        self._branch = encoding.fromlocal(branch)
+        self._opener("branch", "w").write(self._branch + '\n')
 
     def _read(self):
         self._map = {}

mercurial/hg.py

@@ -32,24 +32,22 b' def addbranchrevs(lrepo, repo, branches,'
         return revs, revs[0]
     branchmap = repo.branchmap()
 
-    def primary(butf8):
-        if butf8 == '.':
+    def primary(branch):
+        if branch == '.':
             if not lrepo or not lrepo.local():
                 raise util.Abort(_("dirstate branch not accessible"))
-            butf8 = lrepo.dirstate.branch()
-        if butf8 in branchmap:
-            revs.extend(node.hex(r) for r in reversed(branchmap[butf8]))
+            branch = lrepo.dirstate.branch()
+        if branch in branchmap:
+            revs.extend(node.hex(r) for r in reversed(branchmap[branch]))
             return True
         else:
             return False
 
     for branch in branches:
-        butf8 = encoding.fromlocal(branch)
-        if not primary(butf8):
+        if not primary(branch):
             raise error.RepoLookupError(_("unknown branch '%s'") % branch)
     if hashbranch:
-        butf8 = encoding.fromlocal(hashbranch)
-        if not primary(butf8):
+        if not primary(hashbranch):
             revs.append(hashbranch)
     return revs, revs[0]
 
@@ -365,8 +363,7 b' def clone(ui, source, dest=None, pull=Fa'
                     except error.RepoLookupError:
                         continue
                 bn = dest_repo[uprev].branch()
-                dest_repo.ui.status(_("updating to branch %s\n")
-                                    % encoding.tolocal(bn))
+                dest_repo.ui.status(_("updating to branch %s\n") % bn)
                 _update(dest_repo, uprev)
 
         return src_repo, dest_repo

mercurial/localrepo.py

@@ -105,7 +105,7 b' class localrepository(repo.repository):'
         self._tags = None
         self._tagtypes = None
 
-        self._branchcache = None  # in UTF-8
+        self._branchcache = None
         self._branchcachetip = None
         self.nodetagscache = None
         self.filterpats = {}
@@ -435,7 +435,6 b' class localrepository(repo.repository):'
             bt[bn] = tip
         return bt
 
-
     def _readbranchcache(self):
         partial = {}
         try:
@@ -455,7 +454,8 b' class localrepository(repo.repository):'
                 if not l:
                     continue
                 node, label = l.split(" ", 1)
-                partial.setdefault(label.strip(), []).append(bin(node))
+                label = encoding.tolocal(label.strip())
+                partial.setdefault(label, []).append(bin(node))
         except KeyboardInterrupt:
             raise
         except Exception, inst:
@@ -470,7 +470,7 b' class localrepository(repo.repository):'
             f.write("%s %s\n" % (hex(tip), tiprev))
             for label, nodes in branches.iteritems():
                 for node in nodes:
-                    f.write("%s %s\n" % (hex(node), label))
+                    f.write("%s %s\n" % (hex(node), encoding.fromlocal(label)))
             f.rename()
         except (IOError, OSError):
             pass
@@ -659,7 +659,8 b' class localrepository(repo.repository):'
         except IOError:
             ds = ""
         self.opener("journal.dirstate", "w").write(ds)
-        self.opener("journal.branch", "w").write(self.dirstate.branch())
+        self.opener("journal.branch", "w").write(
+            encoding.fromlocal(self.dirstate.branch()))
         self.opener("journal.desc", "w").write("%d\n%s\n" % (len(self), desc))
 
         renames = [(self.sjoin("journal"), self.sjoin("undo")),
@@ -717,7 +718,7 b' class localrepository(repo.repository):'
                 except IOError:
                     self.ui.warn(_("Named branch could not be reset, "
                                    "current branch still is: %s\n")
-                                 % encoding.tolocal(self.dirstate.branch()))
+                                 % self.dirstate.branch())
                 self.invalidate()
                 self.dirstate.invalidate()
                 self.destroyed()

mercurial/templatekw.py

@@ -148,7 +148,6 b' def showauthor(repo, ctx, templ, **args)'
 def showbranches(**args):
     branch = args['ctx'].branch()
     if branch != 'default':
-        branch = encoding.tolocal(branch)
         return showlist('branch', [branch], plural='branches', **args)
 
 def showchildren(**args):

mercurial/wireproto.py

@@ -44,14 +44,7 b' class wirerepository(repo.repository):'
             branchmap = {}
             for branchpart in d.splitlines():
                 branchname, branchheads = branchpart.split(' ', 1)
-                branchname = urllib.unquote(branchname)
-                # Earlier servers (1.3.x) send branch names in (their) local
-                # charset. The best we can do is assume it's identical to our
-                # own local charset, in case it's not utf-8.
-                try:
-                    branchname.decode('utf-8')
-                except UnicodeDecodeError:
-                    branchname = encoding.fromlocal(branchname)
+                branchname = encoding.tolocal(urllib.unquote(branchname))
                 branchheads = decodelist(branchheads)
                 branchmap[branchname] = branchheads
             return branchmap
@@ -162,7 +155,7 b' def branchmap(repo, proto):'
     branchmap = repo.branchmap()
     heads = []
     for branch, nodes in branchmap.iteritems():
-        branchname = urllib.quote(branch)
+        branchname = urllib.quote(encoding.fromlocal(branch))
         branchnodes = encodelist(nodes)
         heads.append('%s %s' % (branchname, branchnodes))
     return '\n'.join(heads)

tests/test-encoding.t

@@ -240,6 +240,4 b' hg log (dolphin)'
   abort: decoding near '\xe9': 'ascii' codec can't decode byte 0xe9 in position 0: ordinal not in range(128)! (esc)
   [255]
   $ cp latin-1-tag .hg/branch
-  $ HGENCODING=latin-1 hg ci -m 'should fail'
-  abort: branch name not in UTF-8!
-  [255]
+  $ HGENCODING=latin-1 hg ci -m 'auto-promote legacy name'