Snowfakery Recipe for Git Data Model

gistfile1.txt

# Snowfakery recipe for Git's core data model
# Generates: Objects (commits, trees, blobs, tags), References, Index entries, and Reflogs

# ============================================
# OBJECTS
# ============================================

# Blob - stores file contents
- object: Blob
  count: 20
  fields:
    object_id:
      fake: sha1
    type: blob
    filename:
      fake: file_name
    content:
      fake: text
    size:
      random_number:
        min: 100
        max: 50000

# Tree - represents a directory
- object: Tree
  count: 10
  fields:
    object_id:
      fake: sha1
    type: tree

# TreeEntry - entries within a tree (separate table for proper relationships)
- object: TreeEntry
  count: 30
  fields:
    tree_id:
      reference: Tree
    file_mode:
      random_choice:
        - "100644"  # regular file
        - "100755"  # executable file
        - "120000"  # symbolic link
        - "040000"  # directory
    object_type:
      random_choice:
        - blob
        - tree
    filename:
      fake: file_name
    referenced_object_id:
      fake: sha1

# Commit - a snapshot of the repository
- object: Commit
  count: 15
  fields:
    object_id:
      fake: sha1
    type: commit
    tree_id:
      reference: Tree
    parent_commit_id:
      random_choice:
        - choice:
            probability: 10
            pick: null  # Root commits have no parent
        - choice:
            probability: 90
            pick:
              fake: sha1
    author_name:
      fake: name
    author_email:
      fake: email
    author_timestamp:
      fake: unix_time
    author_timezone: "-0400"
    committer_name:
      fake: name
    committer_email:
      fake: email
    committer_timestamp:
      fake: unix_time
    committer_timezone: "-0400"
    message:
      fake: sentence

# Tag Object - annotated tag metadata
- object: TagObject
  count: 5
  fields:
    object_id:
      fake: sha1
    type: tag
    target_id:
      reference: Commit
    target_type: commit
    tag_name:
      fake.text:
        max_nb_chars: 10
    tagger_name:
      fake: name
    tagger_email:
      fake: email
    tagger_timestamp:
      fake: unix_time
    tagger_timezone: "-0400"
    message:
      fake: sentence

# ============================================
# REFERENCES
# ============================================

# Branch reference - points to a commit
- object: Branch
  count: 5
  fields:
    name:
      random_choice:
        - main
        - develop
        - feature/login
        - feature/signup
        - bugfix/issue-123
    ref_path: refs/heads/${{name}}
    commit_id:
      reference: Commit

# Tag reference - points to a commit or tag object
- object: Tag
  count: 5
  fields:
    name: v${{fake.random_int(min=1, max=5)}}.${{fake.random_int(min=0, max=9)}}.${{fake.random_int(min=0, max=9)}}
    ref_path: refs/tags/${{name}}
    is_annotated:
      random_choice:
        - true
        - false
    target_id:
      reference: Commit

# HEAD reference - current branch or detached commit
- object: HEAD
  count: 1
  fields:
    is_detached:
      random_choice:
        - choice:
            probability: 20
            pick: true
        - choice:
            probability: 80
            pick: false
    symbolic_ref:
      reference: Branch
    commit_id:
      reference: Commit

# Remote-tracking branch
- object: RemoteTrackingBranch
  count: 5
  fields:
    remote: origin
    branch_name:
      random_choice:
        - main
        - develop
        - feature/api
        - feature/ui
        - release/v2
    ref_path: refs/remotes/${{remote}}/${{branch_name}}
    commit_id:
      reference: Commit

# ============================================
# INDEX (Staging Area)
# ============================================

- object: IndexEntry
  count: 10
  fields:
    file_mode:
      random_choice:
        - "100644"  # regular file
        - "100755"  # executable file
        - "120000"  # symbolic link
    blob_id:
      reference: Blob
    stage_number:
      random_choice:
        - choice:
            probability: 90
            pick: 0  # Normal state
        - choice:
            probability: 10
            pick:
              random_number:
                min: 1
                max: 3  # Merge conflict states
    directory:
      random_choice:
        - src
        - lib
        - tests
        - docs
    file_name:
      fake: file_name
    file_path: ${{directory}}/${{file_name}}

# ============================================
# REFLOGS
# ============================================

- object: ReflogEntry
  count: 20
  fields:
    ref_name:
      random_choice:
        - HEAD
        - refs/heads/main
        - refs/heads/develop
    commit_id:
      reference: Commit
    previous_commit_id:
      reference: Commit
    timestamp:
      fake: unix_time
    action:
      random_choice:
        - "commit: Initial commit"
        - "commit: Add new feature"
        - "commit (amend): Fix typo"
        - "pull: Fast-forward"
        - "checkout: moving from main to develop"
        - "merge develop: Fast-forward"
        - "reset: moving to HEAD~1"
        - "rebase (finish): refs/heads/feature onto main"