Merge branch 'ps/object-read-stream' into seen

The "git_istream" abstraction has been revamped to make it easier
to interface with pluggable object database design.

Comments?

* ps/object-read-stream:
  streaming: drop redundant type and size pointers
  streaming: move into object database subsystem
  streaming: refactor interface to be object-database-centric
  streaming: move logic to read packed objects streams into backend
  streaming: move logic to read loose objects streams into backend
  streaming: make the `odb_read_stream` definition public
  streaming: get rid of `the_repository`
  streaming: rely on object sources to create object stream
  packfile: introduce function to read object info from a store
  streaming: move zlib stream into backends
  streaming: create structure for filtered object streams
  streaming: create structure for packed object streams
  streaming: create structure for loose object streams
  streaming: create structure for in-core object streams
  streaming: allocate stream inside the backend-specific logic
  streaming: explicitly pass packfile info when streaming a packed object
  streaming: propagate final object type via the stream
  streaming: drop the `open()` callback function
  streaming: rename `git_istream` into `odb_read_stream`

Author: gitster

Makefile

@@ -1201,6 +1201,7 @@ LIB_OBJS += object-file.o
 LIB_OBJS += object-name.o
 LIB_OBJS += object.o
 LIB_OBJS += odb.o
+LIB_OBJS += odb/streaming.o
 LIB_OBJS += oid-array.o
 LIB_OBJS += oidmap.o
 LIB_OBJS += oidset.o
@@ -1295,7 +1296,6 @@ LIB_OBJS += split-index.o
 LIB_OBJS += stable-qsort.o
 LIB_OBJS += statinfo.o
 LIB_OBJS += strbuf.o
-LIB_OBJS += streaming.o
 LIB_OBJS += string-list.o
 LIB_OBJS += strmap.o
 LIB_OBJS += strvec.o

archive-tar.c

@@ -12,8 +12,8 @@
 #include "tar.h"
 #include "archive.h"
 #include "odb.h"
+#include "odb/streaming.h"
 #include "strbuf.h"
-#include "streaming.h"
 #include "run-command.h"
 #include "write-or-die.h"
 
@@ -129,22 +129,20 @@ static void write_trailer(void)
  */
 static int stream_blocked(struct repository *r, const struct object_id *oid)
 {
-	struct git_istream *st;
-	enum object_type type;
-	unsigned long sz;
+	struct odb_read_stream *st;
 	char buf[BLOCKSIZE];
 	ssize_t readlen;
 
-	st = open_istream(r, oid, &type, &sz, NULL);
+	st = odb_read_object_stream(r->objects, oid, NULL);
 	if (!st)
 		return error(_("cannot stream blob %s"), oid_to_hex(oid));
 	for (;;) {
-		readlen = read_istream(st, buf, sizeof(buf));
+		readlen = odb_read_stream_read(st, buf, sizeof(buf));
 		if (readlen <= 0)
 			break;
 		do_write_blocked(buf, readlen);
 	}
-	close_istream(st);
+	odb_read_stream_close(st);
 	if (!readlen)
 		finish_record();
 	return readlen;

archive-zip.c

@@ -10,9 +10,9 @@
 #include "gettext.h"
 #include "git-zlib.h"
 #include "hex.h"
-#include "streaming.h"
 #include "utf8.h"
 #include "odb.h"
+#include "odb/streaming.h"
 #include "strbuf.h"
 #include "userdiff.h"
 #include "write-or-die.h"
@@ -309,7 +309,7 @@ static int write_zip_entry(struct archiver_args *args,
 	enum zip_method method;
 	unsigned char *out;
 	void *deflated = NULL;
-	struct git_istream *stream = NULL;
+	struct odb_read_stream *stream = NULL;
 	unsigned long flags = 0;
 	int is_binary = -1;
 	const char *path_without_prefix = path + args->baselen;
@@ -347,12 +347,11 @@ static int write_zip_entry(struct archiver_args *args,
 			method = ZIP_METHOD_DEFLATE;
 
 		if (!buffer) {
-			enum object_type type;
-			stream = open_istream(args->repo, oid, &type, &size,
-					      NULL);
+			stream = odb_read_object_stream(args->repo->objects, oid, NULL);
 			if (!stream)
 				return error(_("cannot stream blob %s"),
 					     oid_to_hex(oid));
+			size = stream->size;
 			flags |= ZIP_STREAM;
 			out = NULL;
 		} else {
@@ -429,7 +428,7 @@ static int write_zip_entry(struct archiver_args *args,
 		ssize_t readlen;
 
 		for (;;) {
-			readlen = read_istream(stream, buf, sizeof(buf));
+			readlen = odb_read_stream_read(stream, buf, sizeof(buf));
 			if (readlen <= 0)
 				break;
 			crc = crc32(crc, buf, readlen);
@@ -439,7 +438,7 @@ static int write_zip_entry(struct archiver_args *args,
 							    buf, readlen);
 			write_or_die(1, buf, readlen);
 		}
-		close_istream(stream);
+		odb_read_stream_close(stream);
 		if (readlen)
 			return readlen;
 
@@ -462,7 +461,7 @@ static int write_zip_entry(struct archiver_args *args,
 		zstream.avail_out = sizeof(compressed);
 
 		for (;;) {
-			readlen = read_istream(stream, buf, sizeof(buf));
+			readlen = odb_read_stream_read(stream, buf, sizeof(buf));
 			if (readlen <= 0)
 				break;
 			crc = crc32(crc, buf, readlen);
@@ -486,7 +485,7 @@ static int write_zip_entry(struct archiver_args *args,
 			}
 
 		}
-		close_istream(stream);
+		odb_read_stream_close(stream);
 		if (readlen)
 			return readlen;
 

builtin/cat-file.c

@@ -18,13 +18,13 @@
 #include "list-objects-filter-options.h"
 #include "parse-options.h"
 #include "userdiff.h"
-#include "streaming.h"
 #include "oid-array.h"
 #include "packfile.h"
 #include "pack-bitmap.h"
 #include "object-file.h"
 #include "object-name.h"
 #include "odb.h"
+#include "odb/streaming.h"
 #include "replace-object.h"
 #include "promisor-remote.h"
 #include "mailmap.h"
@@ -95,7 +95,7 @@ static int filter_object(const char *path, unsigned mode,
 
 static int stream_blob(const struct object_id *oid)
 {
-	if (stream_blob_to_fd(1, oid, NULL, 0))
+	if (odb_stream_blob_to_fd(the_repository->objects, 1, oid, NULL, 0))
 		die("unable to stream %s to stdout", oid_to_hex(oid));
 	return 0;
 }

builtin/fsck.c

@@ -13,11 +13,11 @@
 #include "fsck.h"
 #include "parse-options.h"
 #include "progress.h"
-#include "streaming.h"
 #include "packfile.h"
 #include "object-file.h"
 #include "object-name.h"
 #include "odb.h"
+#include "odb/streaming.h"
 #include "path.h"
 #include "read-cache-ll.h"
 #include "replace-object.h"
@@ -340,7 +340,8 @@ static void check_unreachable_object(struct object *obj)
 			}
 			f = xfopen(filename, "w");
 			if (obj->type == OBJ_BLOB) {
-				if (stream_blob_to_fd(fileno(f), &obj->oid, NULL, 1))
+				if (odb_stream_blob_to_fd(the_repository->objects, fileno(f),
+							  &obj->oid, NULL, 1))
 					die_errno(_("could not write '%s'"), filename);
 			} else
 				fprintf(f, "%s\n", describe_object(&obj->oid));

builtin/index-pack.c

@@ -16,12 +16,12 @@
 #include "progress.h"
 #include "fsck.h"
 #include "strbuf.h"
-#include "streaming.h"
 #include "thread-utils.h"
 #include "packfile.h"
 #include "pack-revindex.h"
 #include "object-file.h"
 #include "odb.h"
+#include "odb/streaming.h"
 #include "oid-array.h"
 #include "oidset.h"
 #include "path.h"
@@ -762,7 +762,7 @@ static void find_ref_delta_children(const struct object_id *oid,
 
 struct compare_data {
 	struct object_entry *entry;
-	struct git_istream *st;
+	struct odb_read_stream *st;
 	unsigned char *buf;
 	unsigned long buf_size;
 };
@@ -779,7 +779,7 @@ static int compare_objects(const unsigned char *buf, unsigned long size,
 	}
 
 	while (size) {
-		ssize_t len = read_istream(data->st, data->buf, size);
+		ssize_t len = odb_read_stream_read(data->st, data->buf, size);
 		if (len == 0)
 			die(_("SHA1 COLLISION FOUND WITH %s !"),
 			    oid_to_hex(&data->entry->idx.oid));
@@ -798,24 +798,21 @@ static int compare_objects(const unsigned char *buf, unsigned long size,
 static int check_collison(struct object_entry *entry)
 {
 	struct compare_data data;
-	enum object_type type;
-	unsigned long size;
 
 	if (entry->size <= repo_settings_get_big_file_threshold(the_repository) ||
 	    entry->type != OBJ_BLOB)
 		return -1;
 
 	memset(&data, 0, sizeof(data));
 	data.entry = entry;
-	data.st = open_istream(the_repository, &entry->idx.oid, &type, &size,
-			       NULL);
+	data.st = odb_read_object_stream(the_repository->objects, &entry->idx.oid, NULL);
 	if (!data.st)
 		return -1;
-	if (size != entry->size || type != entry->type)
+	if (data.st->size != entry->size || data.st->type != entry->type)
 		die(_("SHA1 COLLISION FOUND WITH %s !"),
 		    oid_to_hex(&entry->idx.oid));
 	unpack_data(entry, compare_objects, &data);
-	close_istream(data.st);
+	odb_read_stream_close(data.st);
 	free(data.buf);
 	return 0;
 }

builtin/log.c

@@ -16,6 +16,7 @@
 #include "refs.h"
 #include "object-name.h"
 #include "odb.h"
+#include "odb/streaming.h"
 #include "pager.h"
 #include "color.h"
 #include "commit.h"
@@ -35,7 +36,6 @@
 #include "parse-options.h"
 #include "line-log.h"
 #include "branch.h"
-#include "streaming.h"
 #include "version.h"
 #include "mailmap.h"
 #include "progress.h"
@@ -584,7 +584,7 @@ static int show_blob_object(const struct object_id *oid, struct rev_info *rev, c
 	fflush(rev->diffopt.file);
 	if (!rev->diffopt.flags.textconv_set_via_cmdline ||
 	    !rev->diffopt.flags.allow_textconv)
-		return stream_blob_to_fd(1, oid, NULL, 0);
+		return odb_stream_blob_to_fd(the_repository->objects, 1, oid, NULL, 0);
 
 	if (get_oid_with_context(the_repository, obj_name,
 				 GET_OID_RECORD_PATH,
@@ -594,7 +594,7 @@ static int show_blob_object(const struct object_id *oid, struct rev_info *rev, c
 	    !textconv_object(the_repository, obj_context.path,
 			     obj_context.mode, &oidc, 1, &buf, &size)) {
 		object_context_release(&obj_context);
-		return stream_blob_to_fd(1, oid, NULL, 0);
+		return odb_stream_blob_to_fd(the_repository->objects, 1, oid, NULL, 0);
 	}
 
 	if (!buf)

builtin/pack-objects.c

@@ -22,7 +22,6 @@
 #include "pack-objects.h"
 #include "progress.h"
 #include "refs.h"
-#include "streaming.h"
 #include "thread-utils.h"
 #include "pack-bitmap.h"
 #include "delta-islands.h"
@@ -33,6 +32,7 @@
 #include "packfile.h"
 #include "object-file.h"
 #include "odb.h"
+#include "odb/streaming.h"
 #include "replace-object.h"
 #include "dir.h"
 #include "midx.h"
@@ -404,7 +404,7 @@ static unsigned long do_compress(void **pptr, unsigned long size)
 	return stream.total_out;
 }
 
-static unsigned long write_large_blob_data(struct git_istream *st, struct hashfile *f,
+static unsigned long write_large_blob_data(struct odb_read_stream *st, struct hashfile *f,
 					   const struct object_id *oid)
 {
 	git_zstream stream;
@@ -417,7 +417,7 @@ static unsigned long write_large_blob_data(struct git_istream *st, struct hashfi
 	for (;;) {
 		ssize_t readlen;
 		int zret = Z_OK;
-		readlen = read_istream(st, ibuf, sizeof(ibuf));
+		readlen = odb_read_stream_read(st, ibuf, sizeof(ibuf));
 		if (readlen == -1)
 			die(_("unable to read %s"), oid_to_hex(oid));
 
@@ -513,17 +513,19 @@ static unsigned long write_no_reuse_object(struct hashfile *f, struct object_ent
 	unsigned hdrlen;
 	enum object_type type;
 	void *buf;
-	struct git_istream *st = NULL;
+	struct odb_read_stream *st = NULL;
 	const unsigned hashsz = the_hash_algo->rawsz;
 
 	if (!usable_delta) {
 		if (oe_type(entry) == OBJ_BLOB &&
 		    oe_size_greater_than(&to_pack, entry,
 					 repo_settings_get_big_file_threshold(the_repository)) &&
-		    (st = open_istream(the_repository, &entry->idx.oid, &type,
-				       &size, NULL)) != NULL)
+		    (st = odb_read_object_stream(the_repository->objects, &entry->idx.oid,
+						 NULL)) != NULL) {
 			buf = NULL;
-		else {
+			type = st->type;
+			size = st->size;
+		} else {
 			buf = odb_read_object(the_repository->objects,
 					      &entry->idx.oid, &type,
 					      &size);
@@ -577,7 +579,7 @@ static unsigned long write_no_reuse_object(struct hashfile *f, struct object_ent
 			dheader[--pos] = 128 | (--ofs & 127);
 		if (limit && hdrlen + sizeof(dheader) - pos + datalen + hashsz >= limit) {
 			if (st)
-				close_istream(st);
+				odb_read_stream_close(st);
 			free(buf);
 			return 0;
 		}
@@ -591,7 +593,7 @@ static unsigned long write_no_reuse_object(struct hashfile *f, struct object_ent
 		 */
 		if (limit && hdrlen + hashsz + datalen + hashsz >= limit) {
 			if (st)
-				close_istream(st);
+				odb_read_stream_close(st);
 			free(buf);
 			return 0;
 		}
@@ -601,15 +603,15 @@ static unsigned long write_no_reuse_object(struct hashfile *f, struct object_ent
 	} else {
 		if (limit && hdrlen + datalen + hashsz >= limit) {
 			if (st)
-				close_istream(st);
+				odb_read_stream_close(st);
 			free(buf);
 			return 0;
 		}
 		hashwrite(f, header, hdrlen);
 	}
 	if (st) {
 		datalen = write_large_blob_data(st, f, &entry->idx.oid);
-		close_istream(st);
+		odb_read_stream_close(st);
 	} else {
 		hashwrite(f, buf, datalen);
 		free(buf);

entry.c

@@ -2,13 +2,13 @@
 
 #include "git-compat-util.h"
 #include "odb.h"
+#include "odb/streaming.h"
 #include "dir.h"
 #include "environment.h"
 #include "gettext.h"
 #include "hex.h"
 #include "name-hash.h"
 #include "sparse-index.h"
-#include "streaming.h"
 #include "submodule.h"
 #include "symlinks.h"
 #include "progress.h"
@@ -139,7 +139,7 @@ static int streaming_write_entry(const struct cache_entry *ce, char *path,
 	if (fd < 0)
 		return -1;
 
-	result |= stream_blob_to_fd(fd, &ce->oid, filter, 1);
+	result |= odb_stream_blob_to_fd(the_repository->objects, fd, &ce->oid, filter, 1);
 	*fstat_done = fstat_checkout_output(fd, state, statbuf);
 	result |= close(fd);
 

meson.build

@@ -397,6 +397,7 @@ libgit_sources = [
   'object-name.c',
   'object.c',
   'odb.c',
+  'odb/streaming.c',
   'oid-array.c',
   'oidmap.c',
   'oidset.c',
@@ -491,7 +492,6 @@ libgit_sources = [
   'stable-qsort.c',
   'statinfo.c',
   'strbuf.c',
-  'streaming.c',
   'string-list.c',
   'strmap.c',
   'strvec.c',