Blob: vend raw or cooked content

Some blob representations require raw blob content, however, the 'blob'
view is unconditionally handed cooked (utf8-encoded) content, thus
representations which need raw content are forced to reload the blob in
raw form, which is ugly and expensive.

The ultimate goal is to eliminate the wasteful blob reloading when raw
content is needed. Toward that end, teach Blob how to vend raw or cooked
content.

Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Alberto Bertogli <albertito@blitiri.com.ar>

2 files changed, 14 insertions, 10 deletions

diff --git a/git.py b/git.py
index 78f5b1e..9f73fd1 100644
--- a/git.py
+++ b/git.py
@@ -339,12 +339,12 @@ class Repo:
             ref = self.branch
         return Tree(self, ref)
 
-    def blob(self, path, ref = None, raw = False):
+    def blob(self, path, ref = None):
         """Returns a Blob instance for the given path."""
         if not ref:
             ref = self.branch
         cmd = self.cmd('cat-file')
-        cmd.raw(raw)
+        cmd.raw(True)
         cmd.batch = None
 
         if isinstance(ref, unicode):
@@ -356,7 +356,7 @@ class Repo:
         if not head or head.strip().endswith('missing'):
             return None
 
-        return Blob(out.read(), raw)
+        return Blob(out.read())
 
     def last_commit_timestamp(self):
         """Return the timestamp of the last commit."""
@@ -559,8 +559,12 @@ class Tree:
 class Blob:
     """A git blob."""
 
-    def __init__(self, content, raw):
-        if raw:
-            self.raw_content = content
-        else:
-            self.utf8_content = content
+    def __init__(self, raw_content):
+        self.raw_content = raw_content
+        self._utf8_content = None
+
+    @property
+    def utf8_content(self):
+        if not self._utf8_content:
+            self._utf8_content = self.raw_content.decode('utf8', 'replace')
+        return self._utf8_content
diff --git a/utils.py b/utils.py
index be6ab78..7223303 100644
--- a/utils.py
+++ b/utils.py
@@ -102,9 +102,9 @@ def embed_image_blob(repo, dirname, fname):
     mimetype = mimetypes.guess_type(fname)[0]
 
     # Unfortunately, bottle seems to require utf-8 encoded data.
-    # We have to refetch the blob with raw=True, because the utf-8 encoded
+    # We have to refetch the blob as raw data, because the utf-8 encoded
     # version of the blob available in the bottle template discards binary data.
-    raw_blob = repo.blob(dirname + fname, raw = True)
+    raw_blob = repo.blob(dirname + fname)
 
     return '<img style="max-width:100%;" src="data:{0};base64,{1}" />'.format( \
                                     mimetype, base64.b64encode(raw_blob.raw_content))