convert_git: avoid returning two entries for the same file in getchanges

This could happen in merge changesets if the merged file was different
from both parents.

hgext/convert/git.py

@@ -1,123 +1,128 b''
 # git support for the convert extension
 
 import os
 from mercurial import util
 
 from common import NoRepo, commit, converter_source
 
 def recode(s):
     try:
         return s.decode("utf-8").encode("utf-8")
     except:
         try:
             return s.decode("latin-1").encode("utf-8")
         except:
             return s.decode("utf-8", "replace").encode("utf-8")
 
 class convert_git(converter_source):
     # Windows does not support GIT_DIR= construct while other systems
     # cannot remove environment variable. Just assume none have
     # both issues.
     if hasattr(os, 'unsetenv'):
         def gitcmd(self, s):
             prevgitdir = os.environ.get('GIT_DIR')
             os.environ['GIT_DIR'] = self.path
             try:
                 return os.popen(s)
             finally:
                 if prevgitdir is None:
                     del os.environ['GIT_DIR']
                 else:
                     os.environ['GIT_DIR'] = prevgitdir
     else:
         def gitcmd(self, s):
             return os.popen('GIT_DIR=%s %s' % (self.path, s))
     
     def __init__(self, ui, path):
         if os.path.isdir(path + "/.git"):
             path += "/.git"
         self.path = path
         self.ui = ui
         if not os.path.exists(path + "/objects"):
             raise NoRepo("couldn't open GIT repo %s" % path)
 
     def getheads(self):
         fh = self.gitcmd("git-rev-parse --verify HEAD")
         return [fh.read()[:-1]]
 
     def catfile(self, rev, type):
         if rev == "0" * 40: raise IOError()
         fh = self.gitcmd("git-cat-file %s %s 2>%s" % (type, rev,
                                                       util.nulldev))
         return fh.read()
 
     def getfile(self, name, rev):
         return self.catfile(rev, "blob")
 
     def getmode(self, name, rev):
         return self.modecache[(name, rev)]
 
     def getchanges(self, version):
         self.modecache = {}
         fh = self.gitcmd("git-diff-tree --root -m -r %s" % version)
         changes = []
+        seen = {}
         for l in fh:
-            if "\t" not in l: continue
+            if "\t" not in l:
+                continue
             m, f = l[:-1].split("\t")
+            if f in seen:
+                continue
+            seen[f] = 1
             m = m.split()
             h = m[3]
             p = (m[1] == "100755")
             s = (m[1] == "120000")
             self.modecache[(f, h)] = (p and "x") or (s and "l") or ""
             changes.append((f, h))
         return changes
 
     def getcommit(self, version):
         c = self.catfile(version, "commit") # read the commit hash
         end = c.find("\n\n")
         message = c[end+2:]
         message = recode(message)
         l = c[:end].splitlines()
         manifest = l[0].split()[1]
         parents = []
         for e in l[1:]:
             n, v = e.split(" ", 1)
             if n == "author":
                 p = v.split()
                 tm, tz = p[-2:]
                 author = " ".join(p[:-2])
                 if author[0] == "<": author = author[1:-1]
                 author = recode(author)
             if n == "committer":
                 p = v.split()
                 tm, tz = p[-2:]
                 committer = " ".join(p[:-2])
                 if committer[0] == "<": committer = committer[1:-1]
                 committer = recode(committer)
                 message += "\ncommitter: %s\n" % committer
             if n == "parent": parents.append(v)
 
         tzs, tzh, tzm = tz[-5:-4] + "1", tz[-4:-2], tz[-2:]
         tz = -int(tzs) * (int(tzh) * 3600 + int(tzm))
         date = tm + " " + str(tz)
         author = author or "unknown"
 
         c = commit(parents=parents, date=date, author=author, desc=message)
         return c
 
     def gettags(self):
         tags = {}
         fh = self.gitcmd('git-ls-remote --tags "%s" 2>%s' % (self.path,
                                                              util.nulldev))
         prefix = 'refs/tags/'
         for line in fh:
             line = line.strip()
             if not line.endswith("^{}"):
                 continue
             node, tag = line.split(None, 1)
             if not tag.startswith(prefix):
                 continue
             tag = tag[len(prefix):-3]
             tags[tag] = node
 
         return tags

tests/test-convert-git

@@ -1,19 +1,46 b''
 #!/bin/sh
 
 "$TESTDIR/hghave" git || exit 80
 
 echo "[extensions]" >> $HGRCPATH
 echo "convert=" >> $HGRCPATH
 
+GIT_AUTHOR_NAME='test'; export GIT_AUTHOR_NAME
+GIT_AUTHOR_EMAIL='[e-mail]'; export GIT_AUTHOR_EMAIL
+GIT_AUTHOR_DATE="2007-01-01 00:00:00 +0000"; export GIT_AUTHOR_DATE
+GIT_COMMITTER_NAME="$GIT_AUTHOR_NAME"; export GIT_COMMITTER_NAME
+GIT_COMMITTER_EMAIL="$GIT_AUTHOR_EMAIL"; export GIT_COMMITTER_EMAIL
+GIT_COMMITTER_DATE="$GIT_AUTHOR_DATE"; export GIT_COMMITTER_DATE
+
+count=10
+commit()
+{
+    GIT_AUTHOR_DATE="2007-01-01 00:00:$count +0000"
+    GIT_COMMITTER_DATE="$GIT_AUTHOR_DATE"
+    git commit "$@" >/dev/null 2>/dev/null || echo "git commit error"
+    count=`expr $count + 1`
+}
+
 mkdir git-repo
 cd git-repo
 git init-db >/dev/null 2>/dev/null
 echo a > a
 git add a
-git commit -m t1 >/dev/null 2>/dev/null || echo "git commit error"
+commit -m t1
+
 echo b >> a
-git commit -a -m t2 >/dev/null || echo "git commit error"
+commit -a -m t2.1
+
+git checkout -b other HEAD^ >/dev/null 2>/dev/null
+echo c > a
+echo a >> a
+commit -a -m t2.2
+
+git checkout master >/dev/null 2>/dev/null
+git pull --no-commit . other > /dev/null 2>/dev/null
+commit -m 'Merge branch other'
 cd ..
 
-hg convert git-repo
+hg convert --datesort git-repo
 
+hg -R git-repo-hg tip -v

tests/test-convert-git.out

@@ -1,7 +1,22 b''
 assuming destination git-repo-hg
 initializing destination git-repo-hg repository
 scanning source...
 sorting...
 converting...
-1 t1
-0 t2
+3 t1
+2 t2.1
+1 t2.2
+0 Merge branch other
+changeset:   3:f0873470732d
+tag:         tip
+parent:      1:cb991dbbb06b
+parent:      2:600bef931ca4
+user:        test <[e-mail]>
+date:        Mon Jan 01 00:00:13 2007 +0000
+files:       a
+description:
+Merge branch other
+
+committer: test <[e-mail]>
+
+