convert: cvs.py - Allow user to use built-in CVS changeset code.

tests: add two testcases for CVS conversion with builtin CVS
including a testcase for issue 1148.

tests/test-convert-cvs-branch

@@ -0,0 +1,60
+#!/bin/sh
+
+# This is http://www.selenic.com/mercurial/bts/issue1148
+
+"$TESTDIR/hghave" cvs || exit 80
+
+cvscall()
+{
+    cvs -f "$@"
+}
+
+echo "[extensions]" >> $HGRCPATH
+echo "convert = " >> $HGRCPATH
+echo "graphlog = " >> $HGRCPATH
+echo "[convert]" >> $HGRCPATH
+echo "cvsps=builtin" >> $HGRCPATH
+
+echo % create cvs repository
+mkdir cvsrepo
+cd cvsrepo
+export CVSROOT=`pwd`
+export CVS_OPTIONS=-f
+cd ..
+
+cvscall -q -d "$CVSROOT" init
+
+echo % Create a new project
+
+mkdir src
+cd src
+echo "1" > a > b
+cvscall import -m "init" src v0 r0
+cd ..
+cvscall co src
+cd src
+
+echo % Branch the project
+
+cvscall tag -b BRANCH
+cvscall up -r BRANCH
+
+echo % Modify file a, then b, then a 
+
+echo "2" > a
+cvscall ci -m "mod a" | grep '<--' | sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g'
+
+echo "2" > b
+cvscall ci -m "mod b" | grep '<--' | sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g'
+
+echo "3" > a
+cvscall ci -m "mod a again" | grep '<--' | sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g'
+
+echo % Convert
+
+cd ..
+hg convert src | sed -e 's/connecting to.*cvsrepo/connecting to cvsrepo/g'
+
+echo % Check the result
+
+hg -R src-hg glog --template '#rev# (#branches#) #desc# files: #files#\n'

tests/test-convert-cvs-branch.out

@@ -0,0 +1,53
+% create cvs repository
+% Create a new project
+N src/a
+N src/b
+
+No conflicts created by this import
+
+cvs checkout: Updating src
+U src/a
+U src/b
+% Branch the project
+cvs tag: Tagging .
+T a
+T b
+cvs update: Updating .
+% Modify file a, then b, then a
+cvs commit: Examining .
+checking in src/a,v
+cvs commit: Examining .
+checking in src/b,v
+cvs commit: Examining .
+checking in src/a,v
+% Convert
+assuming destination src-hg
+initializing destination src-hg repository
+using builtin cvsps
+collecting CVS rlog
+7 log entries
+creating changesets
+5 changeset entries
+connecting to cvsrepo
+scanning source...
+sorting...
+converting...
+4 Initial revision
+3 init
+2 mod a
+1 mod b
+0 mod a again
+updating tags
+% Check the result
+o  5 () update tags files: .hgtags
+|
+| o  4 (BRANCH) mod a again files: a
+| |
+| o  3 (BRANCH) mod b files: b
+| |
+| o  2 (BRANCH) mod a files: a
+| |
+| o  1 (v0) init files:
+|/
+o  0 () Initial revision files: a b
+

tests/test-convert-cvs-builtincvsps

@@ -0,0 +1,99
+#!/bin/sh
+
+"$TESTDIR/hghave" cvs || exit 80
+
+cvscall()
+{
+    cvs -f "$@"
+}
+
+echo "[extensions]" >> $HGRCPATH
+echo "convert = " >> $HGRCPATH
+echo "graphlog = " >> $HGRCPATH
+echo "[convert]" >> $HGRCPATH
+echo "cvsps=builtin" >> $HGRCPATH
+
+echo % create cvs repository
+mkdir cvsrepo
+cd cvsrepo
+export CVSROOT=`pwd`
+export CVS_OPTIONS=-f
+cd ..
+
+cvscall -q -d "$CVSROOT" init
+
+echo % create source directory
+mkdir src-temp
+cd src-temp
+echo a > a
+mkdir b
+cd b
+echo c > c
+cd ..
+
+echo % import source directory
+cvscall -q import -m import src INITIAL start
+cd ..
+
+echo % checkout source directory
+cvscall -q checkout src
+
+echo % commit a new revision changing b/c
+cd src
+sleep 1
+echo c >> b/c
+cvscall -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
+cvscall -q commit -mci1 . | grep '<--' |\
+    sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g'
+cd ..
+
+echo % convert again
+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 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'
+
+echo % commit branch
+cd src
+cvs -q update -r1.1 b/c
+cvs -q tag -b branch
+cvs -q update -r branch
+echo d >> b/c
+cvs -q commit -mci2 . | grep '<--' |\
+    sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g'
+cd ..
+
+echo % convert again
+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 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'
+
+echo "graphlog = " >> $HGRCPATH
+hg -R src-hg glog --template '#rev# (#branches#) #desc# files: #files#\n'

tests/test-convert-cvs-builtincvsps.out

@@ -0,0 +1,139
+% create cvs repository
+% create source directory
+% import source directory
+N src/a
+N src/b/c
+
+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
+using builtin cvsps
+collecting CVS rlog
+5 log entries
+creating changesets
+3 changeset entries
+connecting to cvsrepo
+scanning source...
+sorting...
+converting...
+2 Initial revision
+1 import
+0 ci0
+updating tags
+a
+c
+c
+% convert fresh repo with --filemap
+initializing destination src-filemap repository
+using builtin cvsps
+collecting CVS rlog
+5 log entries
+creating changesets
+3 changeset entries
+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
+checking in src/a,v
+checking in src/b/c,v
+% convert again
+using builtin cvsps
+collecting CVS rlog
+7 log entries
+creating changesets
+4 changeset entries
+connecting to cvsrepo
+scanning source...
+sorting...
+converting...
+0 ci1
+a
+a
+c
+c
+c
+% convert again with --filemap
+using builtin cvsps
+collecting CVS rlog
+7 log entries
+creating changesets
+4 changeset entries
+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
+% commit branch
+U b/c
+T a
+T b/c
+checking in src/b/c,v
+% convert again
+using builtin cvsps
+collecting CVS rlog
+8 log entries
+creating changesets
+5 changeset entries
+connecting to cvsrepo
+scanning source...
+sorting...
+converting...
+0 ci2
+a
+a
+c
+d
+% convert again with --filemap
+using builtin cvsps
+collecting CVS rlog
+8 log entries
+creating changesets
+5 changeset entries
+connecting to cvsrepo
+scanning source...
+sorting...
+converting...
+0 ci2
+c
+d
+4 ci2 files: b/c
+3 ci1 files: b/c
+2 update tags files: .hgtags
+1 ci0 files: b/c
+0 Initial revision files: b/c
+o  5 (branch) ci2 files: b/c
+|
+| o  4 () ci1 files: a b/c
+| |
+| o  3 () update tags files: .hgtags
+| |
+| o  2 () ci0 files: b/c
+|/
+| o  1 (INITIAL) import files:
+|/
+o  0 () Initial revision files: a b/c
+

hgext/convert/cvs.py

@@ -1,315 +1,355
 # CVS conversion code inspired by hg-cvs-import and git-cvsimport
 
 import os, locale, re, socket
 from cStringIO import StringIO
 from mercurial import util
+from mercurial.i18n import _
 
 from common import NoRepo, commit, converter_source, checktool
+import cvsps
 
 class convert_cvs(converter_source):
     def __init__(self, ui, path, rev=None):
         super(convert_cvs, self).__init__(ui, path, rev=rev)
 
         cvs = os.path.join(path, "CVS")
         if not os.path.exists(cvs):
             raise NoRepo("%s does not look like a CVS checkout" % path)
 
+        checktool('cvs')
         self.cmd = ui.config('convert', 'cvsps', 'cvsps -A -u --cvs-direct -q')
         cvspsexe = self.cmd.split(None, 1)[0]
-        for tool in (cvspsexe, 'cvs'):
-            checktool(tool)
+        self.builtin = cvspsexe == 'builtin'
+
+        if not self.builtin:
+            checktool(cvspsexe)
 
         self.changeset = {}
         self.files = {}
         self.tags = {}
         self.lastbranch = {}
         self.parent = {}
         self.socket = None
         self.cvsroot = file(os.path.join(cvs, "Root")).read()[:-1]
         self.cvsrepo = file(os.path.join(cvs, "Repository")).read()[:-1]
         self.encoding = locale.getpreferredencoding()
-        self._parse()
+
+        self._parse(ui)
         self._connect()
 
-    def _parse(self):
+    def _parse(self, ui):
         if self.changeset:
             return
 
         maxrev = 0
         cmd = self.cmd
         if self.rev:
             # TODO: handle tags
             try:
                 # patchset number?
                 maxrev = int(self.rev)
             except ValueError:
                 try:
                     # date
                     util.parsedate(self.rev, ['%Y/%m/%d %H:%M:%S'])
                     cmd = '%s -d "1970/01/01 00:00:01" -d "%s"' % (cmd, self.rev)
                 except util.Abort:
                     raise util.Abort('revision %s is not a patchset number or date' % self.rev)
 
         d = os.getcwd()
         try:
             os.chdir(self.path)
             id = None
             state = 0
             filerevids = {}
-            for l in util.popen(cmd):
-                if state == 0: # header
-                    if l.startswith("PatchSet"):
-                        id = l[9:-2]
-                        if maxrev and int(id) > maxrev:
-                            # ignore everything
-                            state = 3
-                    elif l.startswith("Date"):
-                        date = util.parsedate(l[6:-1], ["%Y/%m/%d %H:%M:%S"])
-                        date = util.datestr(date)
-                    elif l.startswith("Branch"):
-                        branch = l[8:-1]
-                        self.parent[id] = self.lastbranch.get(branch, 'bad')
-                        self.lastbranch[branch] = id
-                    elif l.startswith("Ancestor branch"):
-                        ancestor = l[17:-1]
-                        # figure out the parent later
-                        self.parent[id] = self.lastbranch[ancestor]
-                    elif l.startswith("Author"):
-                        author = self.recode(l[8:-1])
-                    elif l.startswith("Tag:") or l.startswith("Tags:"):
-                        t = l[l.index(':')+1:]
-                        t = [ut.strip() for ut in t.split(',')]
-                        if (len(t) > 1) or (t[0] and (t[0] != "(none)")):
-                            self.tags.update(dict.fromkeys(t, id))
-                    elif l.startswith("Log:"):
-                        # switch to gathering log
-                        state = 1
-                        log = ""
-                elif state == 1: # log
-                    if l == "Members: \n":
-                        # switch to gathering members
-                        files = {}
-                        oldrevs = []
-                        log = self.recode(log[:-1])
-                        state = 2
-                    else:
-                        # gather log
-                        log += l
-                elif state == 2: # members
-                    if l == "\n": # start of next entry
-                        state = 0
-                        p = [self.parent[id]]
-                        if id == "1":
-                            p = []
-                        if branch == "HEAD":
-                            branch = ""
-                        if branch:
-                            latest = None
-                            # the last changeset that contains a base
-                            # file is our parent
-                            for r in oldrevs:
-                                latest = max(filerevids.get(r, None), latest)
-                            if latest:
-                                p = [latest]
+
+            if self.builtin:
+                # builtin cvsps code
+                ui.status(_('using builtin cvsps\n'))
+
+                db = cvsps.createlog(ui, cache='update')
+                db = cvsps.createchangeset(ui, db,
+                      fuzz=int(ui.config('convert', 'cvsps.fuzz', 60)),
+                      mergeto=ui.config('convert', 'cvsps.mergeto', None),
+                      mergefrom=ui.config('convert', 'cvsps.mergefrom', None))
+
+                for cs in db:
+                    if maxrev and cs.id>maxrev:
+                        break
+                    id = str(cs.id)
+                    cs.author = self.recode(cs.author)
+                    self.lastbranch[cs.branch] = id
+                    cs.comment = self.recode(cs.comment)
+                    date = util.datestr(cs.date)
+                    self.tags.update(dict.fromkeys(cs.tags, id))
+
+                    files = {}
+                    for f in cs.entries:
+                        files[f.file] = "%s%s" % ('.'.join([str(x) for x in f.revision]),
+                                                  ['', '(DEAD)'][f.dead])
 
-                        # add current commit to set
-                        c = commit(author=author, date=date, parents=p,
-                                   desc=log, branch=branch)
-                        self.changeset[id] = c
-                        self.files[id] = files
-                    else:
-                        colon = l.rfind(':')
-                        file = l[1:colon]
-                        rev = l[colon+1:-2]
-                        oldrev, rev = rev.split("->")
-                        files[file] = rev
+                    # add current commit to set
+                    c = commit(author=cs.author, date=date,
+                             parents=[str(p.id) for p in cs.parents],
+                             desc=cs.comment, branch=cs.branch or '')
+                    self.changeset[id] = c
+                    self.files[id] = files
+            else:
+                # external cvsps
+                for l in util.popen(cmd):
+                    if state == 0: # header
+                        if l.startswith("PatchSet"):
+                            id = l[9:-2]
+                            if maxrev and int(id) > maxrev:
+                                # ignore everything
+                                state = 3
+                        elif l.startswith("Date"):
+                            date = util.parsedate(l[6:-1], ["%Y/%m/%d %H:%M:%S"])
+                            date = util.datestr(date)
+                        elif l.startswith("Branch"):
+                            branch = l[8:-1]
+                            self.parent[id] = self.lastbranch.get(branch, 'bad')
+                            self.lastbranch[branch] = id
+                        elif l.startswith("Ancestor branch"):
+                            ancestor = l[17:-1]
+                            # figure out the parent later
+                            self.parent[id] = self.lastbranch[ancestor]
+                        elif l.startswith("Author"):
+                            author = self.recode(l[8:-1])
+                        elif l.startswith("Tag:") or l.startswith("Tags:"):
+                            t = l[l.index(':')+1:]
+                            t = [ut.strip() for ut in t.split(',')]
+                            if (len(t) > 1) or (t[0] and (t[0] != "(none)")):
+                                self.tags.update(dict.fromkeys(t, id))
+                        elif l.startswith("Log:"):
+                            # switch to gathering log
+                            state = 1
+                            log = ""
+                    elif state == 1: # log
+                        if l == "Members: \n":
+                            # switch to gathering members
+                            files = {}
+                            oldrevs = []
+                            log = self.recode(log[:-1])
+                            state = 2
+                        else:
+                            # gather log
+                            log += l
+                    elif state == 2: # members
+                        if l == "\n": # start of next entry
+                            state = 0
+                            p = [self.parent[id]]
+                            if id == "1":
+                                p = []
+                            if branch == "HEAD":
+                                branch = ""
+                            if branch:
+                                latest = None
+                                # the last changeset that contains a base
+                                # file is our parent
+                                for r in oldrevs:
+                                    latest = max(filerevids.get(r, None), latest)
+                                if latest:
+                                    p = [latest]
 
-                        # save some information for identifying branch points
-                        oldrevs.append("%s:%s" % (oldrev, file))
-                        filerevids["%s:%s" % (rev, file)] = id
-                elif state == 3:
-                    # swallow all input
-                    continue
+                            # add current commit to set
+                            c = commit(author=author, date=date, parents=p,
+                                       desc=log, branch=branch)
+                            self.changeset[id] = c
+                            self.files[id] = files
+                        else:
+                            colon = l.rfind(':')
+                            file = l[1:colon]
+                            rev = l[colon+1:-2]
+                            oldrev, rev = rev.split("->")
+                            files[file] = rev
+
+                            # save some information for identifying branch points
+                            oldrevs.append("%s:%s" % (oldrev, file))
+                            filerevids["%s:%s" % (rev, file)] = id
+                    elif state == 3:
+                        # swallow all input
+                        continue
 
             self.heads = self.lastbranch.values()
         finally:
             os.chdir(d)
 
     def _connect(self):
         root = self.cvsroot
         conntype = None
         user, host = None, None
         cmd = ['cvs', 'server']
 
         self.ui.status("connecting to %s\n" % root)
 
         if root.startswith(":pserver:"):
             root = root[9:]
             m = re.match(r'(?:(.*?)(?::(.*?))?@)?([^:\/]*)(?::(\d*))?(.*)',
                          root)
             if m:
                 conntype = "pserver"
                 user, passw, serv, port, root = m.groups()
                 if not user:
                     user = "anonymous"
                 if not port:
                     port = 2401
                 else:
                     port = int(port)
                 format0 = ":pserver:%s@%s:%s" % (user, serv, root)
                 format1 = ":pserver:%s@%s:%d%s" % (user, serv, port, root)
 
                 if not passw:
                     passw = "A"
                     pf = open(os.path.join(os.environ["HOME"], ".cvspass"))
                     for line in pf.read().splitlines():
                         part1, part2 = line.split(' ', 1)
                         if part1 == '/1':
                             # /1 :pserver:[e-mail]:2401/cvsroot/foo Ah<Z
                             part1, part2 = part2.split(' ', 1)
                             format = format1
                         else:
                             # :pserver:[e-mail]:/cvsroot/foo Ah<Z
                             format = format0
                         if part1 == format:
                             passw = part2
                             break
                     pf.close()
 
                 sck = socket.socket()
                 sck.connect((serv, port))
                 sck.send("\n".join(["BEGIN AUTH REQUEST", root, user, passw,
                                     "END AUTH REQUEST", ""]))
                 if sck.recv(128) != "I LOVE YOU\n":
                     raise util.Abort("CVS pserver authentication failed")
 
                 self.writep = self.readp = sck.makefile('r+')
 
         if not conntype and root.startswith(":local:"):
             conntype = "local"
             root = root[7:]
 
         if not conntype:
             # :ext:user@host/home/user/path/to/cvsroot
             if root.startswith(":ext:"):
                 root = root[5:]
             m = re.match(r'(?:([^@:/]+)@)?([^:/]+):?(.*)', root)
             # Do not take Windows path "c:\foo\bar" for a connection strings
             if os.path.isdir(root) or not m:
                 conntype = "local"
             else:
                 conntype = "rsh"
                 user, host, root = m.group(1), m.group(2), m.group(3)
 
         if conntype != "pserver":
             if conntype == "rsh":
                 rsh = os.environ.get("CVS_RSH") or "ssh"
                 if user:
                     cmd = [rsh, '-l', user, host] + cmd
                 else:
                     cmd = [rsh, host] + cmd
 
             # popen2 does not support argument lists under Windows
             cmd = [util.shellquote(arg) for arg in cmd]
             cmd = util.quotecommand(' '.join(cmd))
             self.writep, self.readp = os.popen2(cmd, 'b')
 
         self.realroot = root
 
         self.writep.write("Root %s\n" % root)
         self.writep.write("Valid-responses ok error Valid-requests Mode"
                           " M Mbinary E Checked-in Created Updated"
                           " Merged Removed\n")
         self.writep.write("valid-requests\n")
         self.writep.flush()
         r = self.readp.readline()
         if not r.startswith("Valid-requests"):
             raise util.Abort("server sucks")
         if "UseUnchanged" in r:
             self.writep.write("UseUnchanged\n")
             self.writep.flush()
             r = self.readp.readline()
 
     def getheads(self):
         return self.heads
 
     def _getfile(self, name, rev):
 
         def chunkedread(fp, count):
             # file-objects returned by socked.makefile() do not handle
             # large read() requests very well.
             chunksize = 65536
             output = StringIO()
             while count > 0:
                 data = fp.read(min(count, chunksize))
                 if not data:
                     raise util.Abort("%d bytes missing from remote file" % count)
                 count -= len(data)
                 output.write(data)
             return output.getvalue()
 
         if rev.endswith("(DEAD)"):
             raise IOError
 
         args = ("-N -P -kk -r %s --" % rev).split()
         args.append(self.cvsrepo + '/' + name)
         for x in args:
             self.writep.write("Argument %s\n" % x)
         self.writep.write("Directory .\n%s\nco\n" % self.realroot)
         self.writep.flush()
 
         data = ""
         while 1:
             line = self.readp.readline()
             if line.startswith("Created ") or line.startswith("Updated "):
                 self.readp.readline() # path
                 self.readp.readline() # entries
                 mode = self.readp.readline()[:-1]
                 count = int(self.readp.readline()[:-1])
                 data = chunkedread(self.readp, count)
             elif line.startswith(" "):
                 data += line[1:]
             elif line.startswith("M "):
                 pass
             elif line.startswith("Mbinary "):
                 count = int(self.readp.readline()[:-1])
                 data = chunkedread(self.readp, count)
             else:
                 if line == "ok\n":
                     return (data, "x" in mode and "x" or "")
                 elif line.startswith("E "):
                     self.ui.warn("cvs server: %s\n" % line[2:])
                 elif line.startswith("Remove"):
                     l = self.readp.readline()
                     l = self.readp.readline()
                     if l != "ok\n":
                         raise util.Abort("unknown CVS response: %s" % l)
                 else:
                     raise util.Abort("unknown CVS response: %s" % line)
 
     def getfile(self, file, rev):
         data, mode = self._getfile(file, rev)
         self.modecache[(file, rev)] = mode
         return data
 
     def getmode(self, file, rev):
         return self.modecache[(file, rev)]
 
     def getchanges(self, rev):
         self.modecache = {}
         files = self.files[rev]
         cl = files.items()
         cl.sort()
         return (cl, {})
 
     def getcommit(self, rev):
         return self.changeset[rev]
 
     def gettags(self):
         return self.tags
 
     def getchangedfiles(self, rev, i):
         files = self.files[rev].keys()
         files.sort()
         return files