v0.0.3 Release

[OpenBagTwo]

Summary

Release staging for v0.0.3

List of Changes

• Backup Deletion by @OpenBagTwo in Backup Deletion #33
• Add py.typed to the gsb package by @OpenBagTwo in Add py.typed to the gsb package #39
• Allow gsb backup to combine with previous backups by @OpenBagTwo in Allow gsb backup to combine with previous backups #41
• Always include most recent revision in gsb history and rewind by @OpenBagTwo in Always include most recent revision in gsb history and rewind #45
• Prevent tag conflicts on gsb rewinds by @OpenBagTwo in Prevent tag conflicts on gsb rewinds #46
• Implement exporting of backups by @OpenBagTwo in Implement exporting of backups #48
• Add gsb test entrypoint by @OpenBagTwo in Add gsb test entrypoint #49
• Prevent gsb backup from overwriting existing tags by @OpenBagTwo in Prevent gsb backup from overwriting existing tags #50

Tech Debt and Other Concerns

To Do

• Add pipx installation instructions #38

Optional

• Add Python 3.12 to the Actions matrix #36 (already open as Add test runner for Python 3.12 #40)

Validation Performed

• I have been using the GSB v0.0.3 release candidate on my systems for at least a week without issue
• I have fully reviewed a recent pylint report and am comfortable with its findings

PR Type

• This PR introduces a breaking change (will
  require a bump in the minor version)
• The changes in this PR are high urgency and necessitate a hotfix or patch
  release (will require rebasing off of release)
• This is a release (staging) PR (maintainer use only)

Checklist:

• I have read the contributor's guide
• I have run mkdocs serve locally and ensured that all API docs and
  changes I have made to the static pages are rendering correctly, with all links
  working
• All tech debt concerns have been resolved, documented as issues, or otherwise
  accepted
• I agree to license my contribution to this project under
  the GNU Public License v3

[OpenBagTwo]

In proofing the HTML docs, I discovered that the mambaforge link was busted. Turns out, mamba ships with miniforge now! Consequently, I've added #53 to the scope of this release.

[OpenBagTwo]

I'm (mostly) satisfied with the pylint report

************* Module gsb.cli
gsb/cli.py:44:9: W0511: TODO: when we add log files, set this to minimum log level across all handlers (fixme)
gsb/cli.py:105:12: W0632: Possible unbalanced tuple unpacking with sequence defined at line 94 of gsb.history: left side has 2 labels, right side has 0 values (unbalanced-tuple-unpacking)
gsb/cli.py:209:0: R0913: Too many arguments (6/5) (too-many-arguments)
gsb/cli.py:469:4: C0415: Import outside toplevel (pytest) (import-outside-toplevel)
************* Module gsb.history
gsb/history.py:40:0: R0913: Too many arguments (7/5) (too-many-arguments)
gsb/history.py:40:0: R0914: Too many local variables (16/15) (too-many-locals)
************* Module gsb._git
gsb/_git.py:219:40: E1101: Module 'pygit2' has no 'Object' member (no-member)
gsb/_git.py:553:67: E1101: Module 'pygit2' has no 'Object' member (no-member)
************* Module gsb.fastforward
gsb/fastforward.py:185:11: E1120: No value for argument 'starting_point' in function call (no-value-for-parameter)
************* Module gsb.manifest
gsb/manifest.py:36:4: C0103: Method name "of" doesn't conform to snake_case naming style (invalid-name)
************* Module gsb._make_zip_archive
gsb/_make_zip_archive.py:8:0: E0611: No name 'GIT_FILEMODE_LINK' in module 'pygit2' (no-name-in-module)
gsb/_make_zip_archive.py:8:0: E0611: No name 'Commit' in module 'pygit2' (no-name-in-module)
gsb/_make_zip_archive.py:8:0: E0611: No name 'Oid' in module 'pygit2' (no-name-in-module)
gsb/_make_zip_archive.py:8:0: E0611: No name 'Tree' in module 'pygit2' (no-name-in-module)
gsb/_make_zip_archive.py:65:15: W0718: Catching too general exception Exception (broad-exception-caught)

I would love to address the issues with _make_zip_archive, but it's a balancing act of trying to keep that code relatively close in style to the source so that I can contribute it upstream.

[OpenBagTwo]

Observations after a week of daily driving:

• The interface works with my intuition (which you'd hope would be the case!), explicitly around:
  • using gsb history -gavn<N> and then copying the commits to delete into gsb delete
  • gsb backup -cc --tag "<Message>" producing the desired gsb history and git log
    • gsb backup -cc --tag "<Message>" combines beautifully with my scripted backup workflow which just does runs gsb backup (no-tag) nightly
• For a reasonably sized savegame (MBs in size), operations are near instantaneous, but for my Minecraft backups (8GB of saves, 47GB of Git history), operations are painfully slow, from simple backups, which take minutes, to delete operations, which take more like 10.
  • Option to use Git command-line tool #52 should hopefully alleviate that, and while, objectively, that should be low priority (as my usecase is niche), the pain of dailying might put it in line for the MVP release.