A working example of a tool

README.md

@@ -17,4 +17,7 @@ All contributions are accepted under the license in LICENSE.
 After your first contribution please add your name to the `pyproject.toml` file.
 
 
+## Notes
 
+- https://git-scm.com/docs/git-commit I can see that there is a `--reuse-message=<commit>` option
+    in git which might be interesting.

pdm.lock

+# This file is @generated by PDM.
+# It is not intended for manual editing.
+
 [[package]]
 name = "click"
 version = "8.1.3"
@@ -13,9 +16,39 @@ version = "0.4.6"
 requires_python = "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,!=3.5.*,!=3.6.*,>=2.7"
 summary = "Cross-platform colored terminal text."
 
+[[package]]
+name = "git-filter-repo"
+version = "2.38.0"
+requires_python = ">=3.5"
+summary = "Quickly rewrite git repository history"
+
+[[package]]
+name = "gitdb"
+version = "4.0.10"
+requires_python = ">=3.7"
+summary = "Git Object Database"
+dependencies = [
+    "smmap<6,>=3.0.1",
+]
+
+[[package]]
+name = "gitpython"
+version = "3.1.31"
+requires_python = ">=3.7"
+summary = "GitPython is a Python library used to interact with Git repositories"
+dependencies = [
+    "gitdb<5,>=4.0.1",
+]
+
+[[package]]
+name = "smmap"
+version = "5.0.0"
+requires_python = ">=3.6"
+summary = "A pure Python implementation of a sliding window memory map manager"
+
 [metadata]
-lock_version = "4.0"
-content_hash = "sha256:2bdce55641fef073c86dab537aa23267eeee7c12b6d9a223d41cce0fb30b4af5"
+lock_version = "4.1"
+content_hash = "sha256:e4668d20f55756901111f3e73519dca0c61f32a1eccd679b031ad9aafd0c6225"
 
 [metadata.files]
 "click 8.1.3" = [
@@ -26,3 +59,19 @@ content_hash = "sha256:2bdce55641fef073c86dab537aa23267eeee7c12b6d9a223d41cce0fb
     {url = "https://files.pythonhosted.org/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6"},
     {url = "https://files.pythonhosted.org/packages/d8/53/6f443c9a4a8358a93a6792e2acffb9d9d5cb0a5cfd8802644b7b1c9a02e4/colorama-0.4.6.tar.gz", hash = "sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44"},
 ]
+"git-filter-repo 2.38.0" = [
+    {url = "https://files.pythonhosted.org/packages/70/99/37da3374977fb5e0915064718e58715666a395a614c42532a89d6164d958/git-filter-repo-2.38.0.tar.gz", hash = "sha256:fe1753e18f0bd6d3c9b57868c8011ae7d056a6ebab706706396a15ef5312725e"},
+    {url = "https://files.pythonhosted.org/packages/8d/0b/49d4d620327b717fdc072b2efdfcb3588eb3cf780e60ed5ba98ebedb5637/git_filter_repo-2.38.0-py2.py3-none-any.whl", hash = "sha256:c2646206dfdbf06ce27c06c5b77ea33209d8e32d1e33f6bd4c25b1ccc59f9864"},
+]
+"gitdb 4.0.10" = [
+    {url = "https://files.pythonhosted.org/packages/21/a6/35f83efec687615c711fe0a09b67e58f6d1254db27b1013119de46f450bd/gitdb-4.0.10-py3-none-any.whl", hash = "sha256:c286cf298426064079ed96a9e4a9d39e7f3e9bf15ba60701e95f5492f28415c7"},
+    {url = "https://files.pythonhosted.org/packages/4b/47/dc98f3d5d48aa815770e31490893b92c5f1cd6c6cf28dd3a8ae0efffac14/gitdb-4.0.10.tar.gz", hash = "sha256:6eb990b69df4e15bad899ea868dc46572c3f75339735663b81de79b06f17eb9a"},
+]
+"gitpython 3.1.31" = [
+    {url = "https://files.pythonhosted.org/packages/5f/11/2b0f60686dbda49028cec8c66bd18a5e82c96d92eef4bc34961e35bb3762/GitPython-3.1.31.tar.gz", hash = "sha256:8ce3bcf69adfdf7c7d503e78fd3b1c492af782d58893b650adb2ac8912ddd573"},
+    {url = "https://files.pythonhosted.org/packages/9e/8a/d1e02cc111d65b0346f70abb83c51f8593e7134bf694a4a56d1a470caaf7/GitPython-3.1.31-py3-none-any.whl", hash = "sha256:f04893614f6aa713a60cbbe1e6a97403ef633103cdd0ef5eb6efe0deb98dbe8d"},
+]
+"smmap 5.0.0" = [
+    {url = "https://files.pythonhosted.org/packages/21/2d/39c6c57032f786f1965022563eec60623bb3e1409ade6ad834ff703724f3/smmap-5.0.0.tar.gz", hash = "sha256:c840e62059cd3be204b0c9c9f74be2c09d5648eddd4580d9314c3ecde0b30936"},
+    {url = "https://files.pythonhosted.org/packages/6d/01/7caa71608bc29952ae09b0be63a539e50d2484bc37747797a66a60679856/smmap-5.0.0-py3-none-any.whl", hash = "sha256:2aba19d6a040e78d8b09de5c57e96207b09ed71d8e55ce0959eeee6c8e190d94"},
+]

pyproject.toml

@@ -7,6 +7,8 @@ authors = [
 ]
 dependencies = [
     "click>=8.1.3",
+    "git-filter-repo>=2.38.0",
+    "gitpython>=3.1.31",
 ]
 requires-python = ">=3.10"
 license = {text = "MIT"}

regit/__main__.py

+from contextlib import contextmanager
+from pathlib import Path
+import shutil
+from typing import Callable
 import click
+import csv
+import git
+import os
+import git_filter_repo as fr
+import tempfile
+import subprocess
+from dataclasses import dataclass, field
+
+# Some code borrowed from https://github.com/newren/git-filter-repo/blob/main/contrib/filter-repo-demos/lint-history
+
+
+@contextmanager
+def blob_reader():
+    process = subprocess.Popen(
+        ["git", "cat-file", "--batch"], stdin=subprocess.PIPE, stdout=subprocess.PIPE
+    )
+
+    def reader(blob_id):
+        assert process.stdin is not None
+        assert process.stdout is not None
+        process.stdin.write(blob_id + b"\n")
+        process.stdin.flush()
+        _, _, objsize = process.stdout.readline().split()
+        return process.stdout.read(int(objsize) + 1)[:-1]
+
+    yield reader
+
+    if process.stdin is not None:
+        process.stdin.close()
+
+    process.wait()
+
+
+@dataclass
+class BlobHandler:
+    reader: Callable[[str], bytes]
+    filter: fr.RepoFilter | None = None
+    blobs_handled: dict[int, int] = field(default_factory=dict)
+    is_relevant: Callable[[Path], bool] = lambda _: True
+    transform: Callable[[Path, bytes], bytes] = lambda _, b: b
+
+    def __call__(self, commit, metadata):
+        assert self.filter is not None
+        for change in commit.file_changes:
+            filename = Path(change.filename.decode("utf-8"))
+            if change.type == b"D" or not self.is_relevant(filename):
+                continue
+
+            if change.blob_id not in self.blobs_handled:
+                content = self.reader(change.blob_id)
+                blob = fr.Blob(self.transform(filename, content))
+                self.filter.insert(blob)
+                # Record our handling of the blob and use it for this change
+                self.blobs_handled[change.blob_id] = blob.id
+            change.blob_id = self.blobs_handled[change.blob_id]
+
+
+def transform_program(program: Path, args: tuple[str], folder: Path):
+    def callback(file: Path, content: bytes):
+        ix = args.index("{}")
+        if ix != -1:
+            pargs = list(args)
+            tmp_file = folder / file.name
+            with open(tmp_file, "wb") as f:
+                f.write(content)
+            pargs[ix] = str(tmp_file)
+            print(program, pargs)
+            subprocess.run([program] + pargs, check=True)
+            with open(tmp_file, "rb") as f:
+                new_content = f.read()
+            os.remove(tmp_file)
+            return new_content
+        else:
+            return subprocess.run(
+                [program] + list(args), input=content, capture_output=True, check=True
+            ).stdout
+
+    return callback
 
 
 @click.command()
-@click.option("--count", default=1, help="Number of greetings.")
-@click.option("--name", prompt="Your name?", help="The person to greet.")
-def regit(count, name):
-    """Simple program that greets NAME for a total of COUNT times."""
-    for x in range(count):
-        click.echo("Hello %s!" % name)
+@click.option("--repo", help="the repo to clone and change", default=None)
+@click.option(
+    "-o",
+    "--output",
+    help="The name of the resulting repo",
+    type=click.Path(exists=False, path_type=Path),
+)
+@click.option(
+    "-p",
+    "--pattern",
+    help="the glob-pattern to match files.",
+    type=str,
+)
+@click.option(
+    "-m",
+    "--mapping",
+    help="The file to output the csv mapping to",
+    type=click.File("w", lazy=False),
+)
+@click.argument(
+    "program",
+    # help="The program to execute on each file",
+    type=click.Path(executable=True, path_type=Path),
+)
+@click.argument(
+    "args",
+    # help="The argurments, choose '{}' to mean the file in question, omit a '{}' to use stdin",
+    nargs=-1,
+    type=str,
+)
+def regit(
+    repo,
+    pattern: str | None,
+    output: Path,
+    mapping,
+    program: Path,
+    args: tuple[str],
+):
+    """A simple program that runs a command on every commit on a repo."""
+
+    if output is None:
+        output = Path(tempfile.mkdtemp())
+
+    if repo is None:
+        repo = git.Repo().clone(path=output)
+    else:
+        repo = git.Repo.clone_from(url=repo, to_path=output)
+
+    if program.exists():
+        program = program.absolute()
+    else:
+        pp = shutil.which(str(program))
+        assert pp is not None
+        program = Path(pp)
+
+    os.chdir(output)
+    options = fr.FilteringOptions.parse_args([], error_on_empty=False)
+
+    with blob_reader() as br, tempfile.TemporaryDirectory() as folder:
+        handler = BlobHandler(
+            br,
+            is_relevant=lambda a: pattern is None or a.match(pattern),
+            transform=transform_program(program, args, Path(folder)),
+        )
+        filter = fr.RepoFilter(options, commit_callback=handler)
+        handler.filter = filter
+        filter.run()
+
+        if mapping:
+            writer = csv.writer(mapping)
+            writer.writerow(["from", "to"])
+            for fm, to in filter._commit_renames.items():
+                writer.writerow([m.decode() for m in [fm, to]])
+
+    print(repo)
 
 
 if __name__ == "__main__":