Skip to content

Latest commit

 

History

History
84 lines (60 loc) · 3.79 KB

README.md

File metadata and controls

84 lines (60 loc) · 3.79 KB

This is a simple script to migrate issues from Google Code to Github.

For a full history of changes, please consult the change log.

THIS SCRIPT WILL SEND A LOT OF EMAILS TO ALL WATCHERS

Github's API does not support creating issues or adding comments without notifying everyone watching the repository. As a result, running this script targetting an existing repository with watchers who do not want to recieve a very large number of emails is probably not a good idea.

I do not know of any way around this other than deleting and recreating the repository immediately before running the import.

How it works

The script iterates over the issues and comments in a Google Code repository, creating matching issues and comments in Github. This has some limitations:

  • All migrated issues and comments are authored by the user running the script, and lose their original creation date. We try to mitigate this by adding a non-obtrusive header to each issue and comment stating the original author and creation date.

  • Github doesn't support attachments for issues, so any attachments are simply listed as links to the attachment on Google Code.

  • Support for Merged-into links for duplicate issues are not implemented.

Otherwise almost everything is preserved, including labels, issue state (open/closed), and issue status (invalid, wontfix, duplicate).

The script can be run repeatedly to migrate new issues and comments, without mucking up what's already on Github.

Required Python libraries

Run pip install -r requirements.txt to install all required libraries.

Usage

migrateissues.py [options] <google project name> <github username> <github project>

  google_project_name       The project name (from the URL) from google code
  github_user_name          The Github username
  github_project            The Github project name, e.g. username/project

Options:
  -h, --help                Show this help message and exit
  -a, --assign-owner        Assign owned issues to the Github user
  -d, --dry-run             Don't modify anything on Github
  -p, --omit-priority       Don't migrate priority labels
  -s, --synchronize-ids     Ensure that migrated issues keep the same ID
  -c, --google-code-cookie  Supply cookies to use for scraping Google Code
  --skip-closed             Skip all closed bugs

    You will be prompted for your github password.

--assign-owner automatically assigns any issues that currently have an owner to your Github user (the one running the script), even if you weren't the original owner. This is used to save a little time in cases where you do in fact own most issues.

--dry-run does as much as possible without actually adding anything to Github. It's useful as a test, to turn up any errors or unexpected behaviors before you run the script, irreversibly, on your real repository.

--omit-priorities skips migration of Google Code Priority labels, since many projects don't actually use them, and would just remove them from Github anyway.

--synchronize-ids attempts to ensure that every Github issue gets the same ID as its original Google Code issue. Normally this happens anyway, but in some cases Google Code skips issue numbers; this option fills the gaps with dummy issues to ensure that the next real issue keeps the same numbering. This only works, of course, if the migration starts with a fresh Github repistory.

--google-code-cookie takes a HTTP header encoded cookie to use when fetching pages from Google Code. Google Code normally mangles names for spam prevention, and getting the raw names requires being logged in and having filled out a CAPTCHA.

--skip-closed will skip migrating issues that were closed.