Speed up initial commit processing

There is no reason to run blame on files in the initial commit,
since all the lines there are known to have the commit author as
their author. Rewrite the initial commit processing to only
read the files, which speeds up the first commit processing
substantially.

Author: Jaakko Kangasharju

githammer/hammer.py

@@ -14,6 +14,7 @@
 
 import datetime
 import os
+import io
 import re
 from operator import itemgetter
 
@@ -115,26 +116,32 @@ def _build_commit_map(self, session):
             if db_detail.test_count:
                 self._shas_to_commits[db_detail.commit_id].test_counts[db_detail.author] = db_detail.test_count
 
+    def _process_lines_into_line_counts(self, repository, commit, path, lines, line_counts, test_counts):
+        is_test_file = _is_test_file(repository.configuration, path)
+        author = self._names_to_authors[_author_line(commit)]
+        line_counts[author] = line_counts.get(author, 0) + len(lines)
+        if is_test_file:
+            for line in lines:
+                if re.search(repository.test_line_regex, line):
+                    test_counts[author] = test_counts.get(author, 0) + 1
+
     def _blame_blob_into_line_counts(self, repository, commit_to_blame, path, line_counts, test_counts):
         if not _is_source_file(repository.configuration, path):
             return
-        is_test_file = _is_test_file(repository.configuration, path)
         blame = repository.git_repository.blame(commit_to_blame, path, w=True)
         for commit, lines in blame:
-            author = self._names_to_authors[_author_line(commit)]
-            line_counts[author] = line_counts.get(author, 0) + len(lines)
-            if is_test_file:
-                for line in lines:
-                    if re.search(repository.test_line_regex, line):
-                        test_counts[author] = test_counts.get(author, 0) + 1
+            self._process_lines_into_line_counts(repository, commit, path, lines, line_counts, test_counts)
 
     def _make_full_commit_stats(self, repository, commit):
         line_counts = {}
         test_counts = {}
         for git_object in commit.tree.traverse(visit_once=True):
             if git_object.type != 'blob':
                 continue
-            self._blame_blob_into_line_counts(repository, commit, git_object.path, line_counts, test_counts)
+            if not _is_source_file(repository.configuration, git_object.path):
+                continue
+            lines = [line.decode('utf-8') for line in io.BytesIO(git_object.data_stream.read()).readlines()]
+            self._process_lines_into_line_counts(repository, commit, git_object.path, lines, line_counts, test_counts)
         return line_counts, test_counts
 
     def _make_diffed_commit_stats(self, repository, commit, previous_commit, previous_commit_line_counts,