feat: update

admake/adtools/adgit.py

@@ -48,8 +48,8 @@ class AdGitHelper:
        __ -> --
        _v=None -> -v
        __version=None -> --version
-       __depth_=1 -> --depth 1
-       __depth=1 -> --depth=1
+       __depth_=1 -> --depth=1
+       __depth=1 -> --depth 1
        NOTE: All the dict arguments will be ahead of list arguments"""
 
     def __init__(self, git=None, dry_run=False):

adtools-light/adtools/adgit.py

@@ -2,7 +2,7 @@
 # -*- coding: utf-8; mode: python; tab-width: 4; indent-tabs-mode: nil -*-
 # vim: tabstop=4 expandtab shiftwidth=4 softtabstop=4 fileencoding=utf-8
 #
-# THIS FILE IS PART OF lrio PROJECT
+# THIS FILE IS PART OF adtools PROJECT
 #
 #           DO WHAT THE FUCK YOU WANT TO PUBLIC LICENSE
 #                   Version 2, December 2004
@@ -27,6 +27,7 @@ import json
 import datetime
 import argparse
 import textwrap
+import uuid
 from urllib.parse import quote_plus, urlparse
 
 import crayons
@@ -231,9 +232,78 @@ def __create_argparser():
         help=textwrap.dedent("""Only output command, never do operation."""),
     )
 
+    # lblob sub-command
+    sub_parser = subparsers.add_parser('lblob', help='List blob size of each commit', parents=parent_parser)
+    sub_parser.set_defaults(func=__list_all_blobs)
+    sub_parser.add_argument(
+        "repo", help=textwrap.dedent("Repository directory.")
+    )
+    sub_parser.add_argument(
+        "-l", "--limit", type=int,
+        default=10 * 1024 * 1024, help=textwrap.dedent("The upper limit, default: 10MB.")
+    )
+    sub_parser.add_argument(
+        "--dry-run",
+        default=False,
+        action="store_true",
+        help=textwrap.dedent("""Only output command, never do operation."""),
+    )
+
     return parser
 
 
+def __list_all_blobs(gh, args, extra_args):
+    git_repo_path = os.path.abspath(args.repo)
+    limit = args.limit
+
+    with AdUtil.chdir(git_repo_path):
+        all_commits = []
+        code, raw_all_commits = gh.rev_list(__all=None, __pretty='oneline')
+        raw_all_commits = raw_all_commits.split('\n')
+
+        for commit in raw_all_commits:
+            commit = commit.strip()
+            if not commit:
+                continue;
+            items = commit.split(maxsplit=2)
+            if len(items) < 1:
+                continue
+            commit = items[0]
+            all_commits.append(commit)
+        # Create a temporary file for following steps
+        commit_file_path = '.{}'.format(str(uuid.uuid4()))
+        open(commit_file_path, 'w+').writelines(all_commits)
+
+        commits = {}
+
+        # Parse commits one by one
+        for commit in all_commits:
+            commits[commit] = []
+
+            code, content = gh.diff_tree(commit, _r=None, _c=None, _M=None, _C=None, __no_commit_id=None)
+            content = content.strip()
+            if code != 0 or not content:
+                continue
+            objs = content.split('\n')
+            for obj in objs:
+                items = obj.split()
+                cid = items[3]
+                path = items[5]
+
+                # Deleted item
+                if cid == "0000000000000000000000000000000000000000":
+                    cid = items[2]
+
+                code, size = gh.cat_file(cid, _s=None)
+                if code != 0:
+                    continue
+                size = int(size.strip())
+                commits[commit].append({'path': path, 'id': cid, 'size': size})
+
+                if size > limit:
+                    print(f"{commit}, {path}, {size}")
+
+
 def main():
     parser = __create_argparser()