odb: Direct writes are back

DIRECT WRITES ARE BACK AND FASTER THAN EVER. The streaming writer to the
ODB was an overkill for the smaller objects like Commit and Tags; most
of the streaming logic was taking too long.

This commit makes Commits, Tags and Trees to be built-up in memory, and
then written to disk in 2 pushes (header + data), instead of streaming
everything.

This is *always* faster, even for big files (since the git_filebuf class
still does streaming writes when the memory cache overflows). This is
also a gazillion lines of code smaller, because we don't have to
precompute the final size of the object before starting the stream (this
was kind of defeating the point of streaming, anyway).

Blobs are still written with full streaming instead of loading them in
memory, since this is still the fastest way.

A new `git_buf` class has been added. It's missing some features, but
it'll get there.

1 files changed, 2 insertions, 0 deletions

diff --git a/src/repository.h b/src/repository.h
index bcf9b2bc8..1cf426128 100644
--- a/src/repository.h
+++ b/src/repository.h
@@ -11,6 +11,7 @@
 #include "index.h"
 #include "cache.h"
 #include "refs.h"
+#include "buffer.h"
 
 #define DOT_GIT ".git"
 #define GIT_DIR DOT_GIT "/"
@@ -44,5 +45,6 @@ void git_object__free(void *object);
 
 int git__parse_oid(git_oid *oid, const char **buffer_out, const char *buffer_end, const char *header);
 int git__write_oid(git_odb_stream *src, const char *header, const git_oid *oid);
+void git_oid__writebuf(git_buf *buf, const char *header, const git_oid *oid);
 
 #endif