Bitbucket data source

Ingests Bitbucket Cloud metadata into bitbucket_nodes and bitbucket_edges. No source parsing yet — metadata only.

Setup — CLI

# App-password auth (most common):
export BITBUCKET_TOKEN=<app-password>
kyma create-database bitbucket
kyma datasource add bitbucket atlassian/python-bitbucket \
  --username me@example.com \
  --app-password "$BITBUCKET_TOKEN" \
  --start

# Or with a PAT (Bitbucket Server / Data Center):
kyma datasource add bitbucket my-ws/my-repo --token bbtok-... --start

Auth

Bitbucket Cloud accepts two auth shapes:

• App password + username — stored as a basic credential. Get one at Account → Personal settings → App passwords, scope repository:read (and issue:read for issues).
• PAT (Server / Data Center, or token-issuing organizations) — stored as a pat credential.

The CLI picks the right shape: if both --username and --app-password are present, basic auth; otherwise PAT.

Token sources (PAT path)

1. --token <pat>
2. --credential-id <uuid>
3. $BITBUCKET_TOKEN
4. $BB_TOKEN

Modules

• repos — repo nodes (description, language, project, is_private).
• branches — branch nodes + BELONGS_TO_REPO edges.
• pulls — pull-request nodes + OPENED_BY / MERGED_INTO.
• issues — issue nodes + OPENED_BY / CLOSED_BY.

Defaults

• Schedule: every 5 minutes (schedule_ms = 300000).
• Auth mode: PAT unless --username AND --app-password are present.