changeset 33388:0823f0983eaa

convert: transcode CVS log messages by specified encoding (issue5597)

Converting from CVS to Mercurial assumes that CVS log messages in "cvs
rlog" output are encoded in UTF-8 (or basic Latin-1). But cvs itself
is usually unaware of encoding of log messages, in practice.

Therefore, if there are commits, of which log message is encoded in
other than UTF-8, log message of corresponded revisions in the
converted repository will be broken.

To avoid such broken log messages, this patch transcodes CVS log
messages by encoding specified via "convert.cvsps.logencoding"
configuration.

This patch accepts multiple encoding for convenience, because
"multiple encoding mixed in a repository" easily occurs. For example,
UTF-8 (recent POSIX), cp932 (Windows), and EUC-JP (legacy POSIX) are
well known encoding for Japanese.

--- a/hgext/convert/__init__.py	Mon Jul 10 23:09:52 2017 +0900
+++ b/hgext/convert/__init__.py	Tue Jul 11 02:10:04 2017 +0900
@@ -228,6 +228,12 @@
         part of a changeset then the default may not be long enough.
         The default is 60.
 
+    :convert.cvsps.logencoding: Specify encoding name to be used for
+        transcoding CVS log messages. Multiple encoding names can be
+        specified as a list (see :hg:`help config.Syntax`), but only
+        the first acceptable encoding in the list is used per CVS log
+        entries. This transcoding is executed before cvslog hook below.
+
     :convert.cvsps.mergeto: Specify a regular expression to which
         commit log messages are matched. If a match occurs, then the
         conversion process will insert a dummy revision merging the

--- a/hgext/convert/cvsps.py	Mon Jul 10 23:09:52 2017 +0900
+++ b/hgext/convert/cvsps.py	Tue Jul 11 02:10:04 2017 +0900
@@ -12,6 +12,7 @@
 from mercurial.i18n import _
 from mercurial import (
     encoding,
+    error,
     hook,
     pycompat,
     util,
@@ -491,6 +492,35 @@
 
     ui.status(_('%d log entries\n') % len(log))
 
+    encodings = ui.configlist('convert', 'cvsps.logencoding')
+    if encodings:
+        def revstr(r):
+            # this is needed, because logentry.revision is a tuple of "int"
+            # (e.g. (1, 2) for "1.2")
+            return '.'.join(pycompat.maplist(pycompat.bytestr, r))
+
+        for entry in log:
+            comment = entry.comment
+            for e in encodings:
+                try:
+                    entry.comment = comment.decode(e).encode('utf-8')
+                    if ui.debugflag:
+                        ui.debug("transcoding by %s: %s of %s\n" %
+                                 (e, revstr(entry.revision), entry.file))
+                    break
+                except UnicodeDecodeError:
+                    pass # try next encoding
+                except LookupError as inst: # unknown encoding, maybe
+                    raise error.Abort(inst,
+                                      hint=_('check convert.cvsps.logencoding'
+                                             ' configuration'))
+            else:
+                raise error.Abort(_("no encoding can transcode"
+                                    " CVS log message for %s of %s")
+                                  % (revstr(entry.revision), entry.file),
+                                  hint=_('check convert.cvsps.logencoding'
+                                         ' configuration'))
+
     hook.hook(ui, None, "cvslog", True, log=log)
 
     return log

--- a/tests/test-convert-cvs.t	Mon Jul 10 23:09:52 2017 +0900
+++ b/tests/test-convert-cvs.t	Tue Jul 11 02:10:04 2017 +0900
@@ -498,3 +498,157 @@
   
 
   $ cd ..
+
+Test transcoding CVS log messages (issue5597)
+=============================================
+
+To emulate commit messages in (non-ascii) multiple encodings portably,
+this test scenario writes CVS history file (*,v file) directly via
+python code.
+
+Commit messages of version 1.2 - 1.4 use u3042 in 3 encodings below.
+
+|encoding  |byte sequence | decodable as:      |
+|          |              | utf-8 euc-jp cp932 |
++----------+--------------+--------------------+
+|utf-8     |\xe3\x81\x82  |  o      x     x    |
+|euc-jp    |\xa4\xa2      |  x      o     o    |
+|cp932     |\x82\xa0      |  x      x     o    |
+
+  $ mkdir -p cvsrepo/transcoding
+  $ python <<EOF
+  > fp = open('cvsrepo/transcoding/file,v', 'w')
+  > fp.write(('''
+  > head	1.4;
+  > access;
+  > symbols
+  > 	start:1.1.1.1 INITIAL:1.1.1;
+  > locks; strict;
+  > comment	@# @;
+  > 
+  > 
+  > 1.4
+  > date	2017.07.10.00.00.04;	author nobody;	state Exp;
+  > branches;
+  > next	1.3;
+  > commitid	10059635D016A510FFA;
+  > 
+  > 1.3
+  > date	2017.07.10.00.00.03;	author nobody;	state Exp;
+  > branches;
+  > next	1.2;
+  > commitid	10059635CFF6A4FF34E;
+  > 
+  > 1.2
+  > date	2017.07.10.00.00.02;	author nobody;	state Exp;
+  > branches;
+  > next	1.1;
+  > commitid	10059635CFD6A4D5095;
+  > 
+  > 1.1
+  > date	2017.07.10.00.00.01;	author nobody;	state Exp;
+  > branches
+  > 	1.1.1.1;
+  > next	;
+  > commitid	10059635CFB6A4A3C33;
+  > 
+  > 1.1.1.1
+  > date	2017.07.10.00.00.01;	author nobody;	state Exp;
+  > branches;
+  > next	;
+  > commitid	10059635CFB6A4A3C33;
+  > 
+  > 
+  > desc
+  > @@
+  > 
+  > 
+  > 1.4
+  > log
+  > @''' + u'\u3042'.encode('cp932') + ''' (cp932)
+  > @
+  > text
+  > @1
+  > 2
+  > 3
+  > 4
+  > @
+  > 
+  > 
+  > 1.3
+  > log
+  > @''' + u'\u3042'.encode('euc-jp') + ''' (euc-jp)
+  > @
+  > text
+  > @d4 1
+  > @
+  > 
+  > 
+  > 1.2
+  > log
+  > @''' + u'\u3042'.encode('utf-8') +  ''' (utf-8)
+  > @
+  > text
+  > @d3 1
+  > @
+  > 
+  > 
+  > 1.1
+  > log
+  > @Initial revision
+  > @
+  > text
+  > @d2 1
+  > @
+  > 
+  > 
+  > 1.1.1.1
+  > log
+  > @import
+  > @
+  > text
+  > @@
+  > ''').lstrip())
+  > EOF
+
+  $ cvscall -q checkout transcoding
+  U transcoding/file
+
+Test converting in normal case
+------------------------------
+
+(filtering by grep in order to check only form of debug messages)
+
+  $ hg convert --config convert.cvsps.logencoding=utf-8,euc-jp,cp932 -q --debug transcoding transcoding-hg | grep 'transcoding by'
+  transcoding by utf-8: 1.1 of file
+  transcoding by utf-8: 1.1.1.1 of file
+  transcoding by utf-8: 1.2 of file
+  transcoding by euc-jp: 1.3 of file
+  transcoding by cp932: 1.4 of file
+  $ hg -R transcoding-hg --encoding utf-8 log -T "{rev}: {desc}\n"
+  5: update tags
+  4: import
+  3: \xe3\x81\x82 (cp932) (esc)
+  2: \xe3\x81\x82 (euc-jp) (esc)
+  1: \xe3\x81\x82 (utf-8) (esc)
+  0: Initial revision
+  $ rm -rf transcoding-hg
+
+Test converting in error cases
+------------------------------
+
+unknown encoding in convert.cvsps.logencoding
+
+  $ hg convert --config convert.cvsps.logencoding=foobar -q transcoding transcoding-hg
+  abort: unknown encoding: foobar
+  (check convert.cvsps.logencoding configuration)
+  [255]
+  $ rm -rf transcoding-hg
+
+no acceptable encoding in convert.cvsps.logencoding
+
+  $ hg convert --config convert.cvsps.logencoding=utf-8,euc-jp -q transcoding transcoding-hg
+  abort: no encoding can transcode CVS log message for 1.4 of file
+  (check convert.cvsps.logencoding configuration)
+  [255]
+  $ rm -rf transcoding-hg

--- a/tests/test-convert.t	Mon Jul 10 23:09:52 2017 +0900
+++ b/tests/test-convert.t	Tue Jul 11 02:10:04 2017 +0900
@@ -171,6 +171,12 @@
                     single changeset. When very large files were checked in as
                     part of a changeset then the default may not be long enough.
                     The default is 60.
+      convert.cvsps.logencoding
+                    Specify encoding name to be used for transcoding CVS log
+                    messages. Multiple encoding names can be specified as a list
+                    (see 'hg help config.Syntax'), but only the first acceptable
+                    encoding in the list is used per CVS log entries. This
+                    transcoding is executed before cvslog hook below.
       convert.cvsps.mergeto
                     Specify a regular expression to which commit log messages
                     are matched. If a match occurs, then the conversion process