From f95308c5ebe548dce2d2da20b73748b09e5efdc7 Mon Sep 17 00:00:00 2001
From: Elijah Newren <newren@gmail.com>
Date: Tue, 30 Oct 2018 17:55:54 -0700
Subject: [PATCH] filter-repo: add handling of 'original-oid' directive

This will be used later to help with commit message rewriting (so that
commits can continue to refer to other commits in their history, using
the new rewritten hashes for those commits), and perhaps also in
removing blobs by id.

Signed-off-by: Elijah Newren <newren@gmail.com>
---
 git-filter-repo | 42 ++++++++++++++++++++++++++++++++++++++----
 1 file changed, 38 insertions(+), 4 deletions(-)

diff --git a/git-filter-repo b/git-filter-repo
index 4ffa584..9947b19 100755
--- a/git-filter-repo
+++ b/git-filter-repo
@@ -272,12 +272,15 @@ class Blob(_GitElementWithId):
   way of representing file contents).
   """
 
-  def __init__(self, data):
+  def __init__(self, data, original_id = None):
     _GitElementWithId.__init__(self)
 
     # Denote that this is a blob
     self.type = 'blob'
 
+    # Record original id
+    self.original_id = original_id
+
     # Stores the blob's data
     self.data = data
 
@@ -382,6 +385,7 @@ class Commit(_GitElementWithId):
                file_changes,
                from_commit = None,
                merge_commits = [],
+               original_id = None,
                **kwargs):
     _GitElementWithId.__init__(self)
 
@@ -391,6 +395,9 @@ class Commit(_GitElementWithId):
     # Record the affected branch
     self.branch = branch
 
+    # Record original id
+    self.original_id = original_id
+
     # Record author's name
     self.author_name  = author_name
 
@@ -500,7 +507,8 @@ class Tag(_GitElement):
   """
 
   def __init__(self, ref, from_ref,
-               tagger_name, tagger_email, tagger_date, tag_msg):
+               tagger_name, tagger_email, tagger_date, tag_msg,
+               original_id = None):
     _GitElement.__init__(self)
 
     # Denote that this is a tag element
@@ -512,6 +520,9 @@ class Tag(_GitElement):
     # Store the entity being tagged (this should be a commit)
     self.from_ref = from_ref
 
+    # Record original id
+    self.original_id = original_id
+
     # Store the name of the tagger
     self.tagger_name  = tagger_name
 
@@ -740,6 +751,11 @@ class FastExportFilter(object):
       self._advance_currentline()
     return filechange
 
+  def _parse_original_id(self):
+    original_id = self._currentline[len('original-oid '):].rstrip()
+    self._advance_currentline()
+    return original_id
+
   def _parse_ref_line(self, refname):
     """
     Parses string data (often a branch name) from current-line. The name of
@@ -795,12 +811,17 @@ class FastExportFilter(object):
     # Parse the Blob
     self._advance_currentline()
     id_ = self._parse_optional_mark()
+
+    original_id = None
+    if self._currentline.startswith('original-oid'):
+      original_id = self._parse_original_id();
+
     data = self._parse_data()
     if self._currentline == '\n':
       self._advance_currentline()
 
     # Create the blob
-    blob = Blob(data)
+    blob = Blob(data, original_id)
 
     # If fast-export text had a mark for this blob, need to make sure this
     # mark translates to the blob's true id.
@@ -860,6 +881,10 @@ class FastExportFilter(object):
     self._seen_refs[branch] = None
     id_ = self._parse_optional_mark()
 
+    original_id = None
+    if self._currentline.startswith('original-oid'):
+      original_id = self._parse_original_id();
+
     author_name = None
     if self._currentline.startswith('author'):
       (author_name, author_email, author_date) = self._parse_user('author')
@@ -968,6 +993,7 @@ class FastExportFilter(object):
                     file_changes,
                     from_commit,
                     merge_commits,
+                    original_id,
                     stream_number = _CURRENT_STREAM_NUMBER)
 
     # If fast-export text had a mark for this commit, need to make sure this
@@ -1022,13 +1048,20 @@ class FastExportFilter(object):
     from_ref = self._parse_optional_parent_ref('from')
     if from_ref is None:
       raise SystemExit("Expected 'from' line while parsing tag %s" % tag)
+
+    original_id = None
+    if self._currentline.startswith('original-oid'):
+      original_id = self._parse_original_id();
+
     (tagger_name, tagger_email, tagger_date) = self._parse_user('tagger')
     tag_msg = self._parse_data()
     if self._currentline == '\n':
       self._advance_currentline()
 
     # Create the tag
-    tag = Tag(tag, from_ref, tagger_name, tagger_email, tagger_date, tag_msg)
+    tag = Tag(tag, from_ref,
+              tagger_name, tagger_email, tagger_date, tag_msg,
+              original_id)
 
     # Call any user callback to allow them to modify the tag
     if self._tag_callback:
@@ -1583,6 +1616,7 @@ def run_fast_filter():
     fe_orig = None
   else:
     fep_cmd = ['git', 'fast-export',
+               '--show-original-ids',
                '--no-data', '--use-done-feature'] + args.revisions
     fep = subprocess.Popen(fep_cmd, stdout=subprocess.PIPE)
     input = fep.stdout