In a git repo, lists all files larger than a given size, displays in which commit they appear, and which branches these commits are part of.

gitSearchLargeFiles.py

#! /usr/bin/env python3

sizeLimit = 500*1024 # bytes

import subprocess
import sys

def execute(shellCmd):
    return subprocess.check_output(shellCmd, shell=True, encoding='utf8')

def getLargeFiles():
    largeFiles = []
    for line in execute('git rev-list --objects --all | git cat-file --batch-check="%(objecttype) %(objectname) %(objectsize) %(rest)" | sed -n "s/^blob //p" | sort --numeric-sort --key=2').strip().split('\n'):
        tokens = line.split(' ')
        if int(tokens[1]) > sizeLimit:
            largeFiles.append({
                'blobId': tokens[0],
                'size': int(tokens[1]),
                'path': tokens[2],
            })
    return largeFiles

def getCommits(blobId):
    commits = []
    for line in execute(f'git log --all --find-object={blobId} --date=format:"%Y-%m-%d %H:%M:%S" --pretty=format:"%h -- %ad -- %cn -- %s"').strip().split('\n'):
        tokens = line.split(' -- ')
        commits.append({
                'sha': tokens[0],
                'date': tokens[1],
                'author': tokens[2],
                'msg': tokens[3],
                })
    return commits

def getBranches(sha):
    branches = []
    for line in execute(f'git branch --remote --contains={sha}').strip().split('\n'):
        branches.append(line.strip().removeprefix('origin/'))
    return branches

def humanSize(num, suffix="B"):
    for unit in ["", "Ki", "Mi", "Gi", "Ti", "Pi", "Ei", "Zi"]:
        if abs(num) < 1024.0:
            return f"{num:3.1f}{unit}{suffix}"
        num /= 1024.0
    return f"{num:.1f}Yi{suffix}"

def showLargeFile(largeFile):
    print(f'{largeFile["path"]} ({humanSize(largeFile["size"])})')
    for commit in getCommits(largeFile['blobId']):
        print(f'    in commit {commit["sha"]} by {commit["author"]} on {commit["date"]}: {commit["msg"]}')
        for branch in getBranches(commit["sha"]):
            print(f'        in branch {branch}')
    print()

print(f'Files larger than {humanSize(sizeLimit)}:\n')
for largeFile in getLargeFiles():
    showLargeFile(largeFile)