Initialize GovOPlaN meta repository

This commit is contained in:
2026-07-11 17:16:58 +02:00
commit 70109afc78
70 changed files with 11641 additions and 0 deletions

View File

@@ -0,0 +1,580 @@
#!/usr/bin/env python3
"""Import markdown backlog/planning files into Gitea issues."""
from __future__ import annotations
import argparse
import dataclasses
import hashlib
import os
import pathlib
import re
import sys
from typing import Any, Iterable
from gitea_common import GiteaClient, GiteaError, infer_target, label_ids_by_name, load_dotenv, repo_path, repo_root, require_token
DEFAULT_PRODUCT_BACKLOG = pathlib.Path("/mnt/DATA/Nextcloud/ADD ideas UG/Products/govoplan/backlog.md")
DEFAULT_WEB_TODO = pathlib.Path("/mnt/DATA/git/addideas-govoplan-website/TODO.md")
REPO_ROOTS = {
"core": pathlib.Path("/mnt/DATA/git/govoplan-core"),
"access": pathlib.Path("/mnt/DATA/git/govoplan-access"),
"mail": pathlib.Path("/mnt/DATA/git/govoplan-mail"),
"files": pathlib.Path("/mnt/DATA/git/govoplan-files"),
"campaign": pathlib.Path("/mnt/DATA/git/govoplan-campaign"),
"web": pathlib.Path("/mnt/DATA/git/addideas-govoplan-website"),
}
MODULE_LABELS = {
"core": "module/core",
"access": "module/access",
"mail": "module/mail",
"files": "module/files",
"campaign": "module/campaign",
"web": "area/marketing",
}
@dataclasses.dataclass(frozen=True)
class Candidate:
repo_key: str
title: str
body: str
labels: tuple[str, ...]
fingerprint: str
source: str
line: int
@dataclasses.dataclass
class RepoState:
target: Any
client: GiteaClient
label_ids: dict[str, int]
fingerprints: set[str]
titles: set[str]
def main() -> int:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--env-file", type=pathlib.Path, help="dotenv file to read before using GITEA_* values")
parser.add_argument("--product-backlog", type=pathlib.Path, default=DEFAULT_PRODUCT_BACKLOG)
parser.add_argument("--access-plan", type=pathlib.Path, help="optional legacy access extraction plan to import")
parser.add_argument("--web-todo", type=pathlib.Path, default=DEFAULT_WEB_TODO)
parser.add_argument("--apply", action="store_true", help="create missing Gitea issues")
args = parser.parse_args()
try:
load_dotenv(args.env_file)
candidates = list(build_candidates(args))
candidates.sort(key=lambda item: (item.repo_key, item.source, item.line, item.title))
print(f"Prepared {len(candidates)} backlog issue candidate(s).")
for repo_key, grouped in group_candidates(candidates).items():
print(f"{repo_key}: {len(grouped)} candidate(s)")
token = require_token() if args.apply else os.environ.get("GITEA_TOKEN")
if not token:
print("Dry run without GITEA_TOKEN: duplicate comparison skipped.")
print_preview(candidates)
return 0
states = load_repo_states(token, sorted({candidate.repo_key for candidate in candidates}))
missing = [candidate for candidate in candidates if not already_imported(candidate, states[candidate.repo_key])]
skipped = len(candidates) - len(missing)
print(f"Skipped {skipped} existing issue candidate(s).")
print(f"Missing {len(missing)} issue candidate(s).")
print_preview(missing)
if not args.apply:
print("Dry run only. Re-run with --apply to create missing issues.")
return 0
created_by_repo: dict[str, list[str]] = {}
for candidate in missing:
state = states[candidate.repo_key]
validate_labels(candidate, state)
issue = state.client.request_json(
"POST",
repo_path(state.target.owner, state.target.repo, "/issues"),
body={
"title": candidate.title,
"body": candidate.body,
"labels": [state.label_ids[label] for label in candidate.labels],
},
)
number = issue.get("number") or issue.get("index")
created_by_repo.setdefault(candidate.repo_key, []).append(f"#{number} {candidate.title}")
state.fingerprints.add(candidate.fingerprint)
state.titles.add(normalize_title(candidate.title))
print("Created issues:")
for repo_key, titles in created_by_repo.items():
print(f"{repo_key}:")
for title in titles:
print(f" {title}")
if not created_by_repo:
print(" none")
return 0
except GiteaError as exc:
print(f"error: {exc}", file=sys.stderr)
return 1
def build_candidates(args: argparse.Namespace) -> Iterable[Candidate]:
if args.product_backlog.exists():
yield from parse_product_backlog(args.product_backlog)
if args.access_plan and args.access_plan.exists():
yield from parse_access_plan(args.access_plan)
if args.web_todo.exists():
yield from parse_simple_todo(args.web_todo)
def parse_product_backlog(path: pathlib.Path) -> Iterable[Candidate]:
headings: list[tuple[int, str]] = []
status = ""
lines = path.read_text(encoding="utf-8").splitlines()
for index, line in enumerate(lines):
line_number = index + 1
heading_match = re.match(r"^(#{2,4})\s+(.+?)\s*$", line)
if heading_match:
level = len(heading_match.group(1))
title = heading_match.group(2).strip()
if level == 2 and title == "Completed Work Archive":
break
headings = [(h_level, h_title) for h_level, h_title in headings if h_level < level]
headings.append((level, title))
status = ""
continue
status_match = re.match(r"^\*\*Status:\*\*\s*(.+?)\s*$", line)
if status_match:
status = status_match.group(1).strip()
continue
item_match = re.match(r"^\s*[-*]\s+\[\s\]\s+(.+?)\s*$", line)
if not item_match:
continue
item = collect_continued_item(lines, index, item_match.group(1))
section = section_title(headings)
repo_key, labels = classify_product_item(item, section)
prefix = type_prefix(labels)
title = format_title(prefix, item)
source = str(path)
body = body_for_item(
fingerprint=make_fingerprint(source, line_number, repo_key, item),
source=source,
line=line_number,
section=section,
status=status,
item=item,
note="Imported from the consolidated GovOPlaN product backlog.",
)
yield Candidate(
repo_key=repo_key,
title=title,
body=body,
labels=tuple(sorted(set(labels))),
fingerprint=make_fingerprint(source, line_number, repo_key, item),
source=source,
line=line_number,
)
def parse_access_plan(path: pathlib.Path) -> Iterable[Candidate]:
text = path.read_text(encoding="utf-8")
lines = text.splitlines()
yield from access_stage_candidates(path, lines)
yield from access_immediate_candidates(path, lines)
yield from access_decision_candidates(path, lines)
def access_stage_candidates(path: pathlib.Path, lines: list[str]) -> Iterable[Candidate]:
index = 0
while index < len(lines):
match = re.match(r"^### Stage ([1-7]):\s*(.+?)\s*$", lines[index])
if not match:
index += 1
continue
stage = match.group(1)
name = match.group(2).strip()
start_line = index + 1
index += 1
block: list[str] = []
while index < len(lines) and not re.match(r"^### Stage \d+:", lines[index]) and not re.match(r"^## ", lines[index]):
block.append(lines[index])
index += 1
item = f"Access extraction Stage {stage}: {name}"
fingerprint = make_fingerprint(str(path), start_line, "core", item)
body = "\n".join(
[
f"<!-- codex-backlog-fingerprint:{fingerprint} -->",
"",
"Imported from the GovOPlaN access extraction plan.",
"",
f"- Source: `{path}`",
f"- Line: `{start_line}`",
f"- Section: `Stage {stage}: {name}`",
"",
"Plan excerpt:",
"",
"```markdown",
f"### Stage {stage}: {name}",
*block,
"```",
]
)
yield Candidate(
repo_key="core",
title=format_title("[Task]", item),
body=body,
labels=labels("core", "type/task", "area/module-system", "module/access"),
fingerprint=fingerprint,
source=str(path),
line=start_line,
)
def access_immediate_candidates(path: pathlib.Path, lines: list[str]) -> Iterable[Candidate]:
in_section = False
for index, line in enumerate(lines):
line_number = index + 1
if line == "## Immediate Backlog":
in_section = True
continue
if in_section and line.startswith("## "):
break
if not in_section:
continue
item_match = re.match(r"^\s*[-*]\s+\[\s\]\s+(.+?)\s*$", line)
if not item_match:
continue
item = collect_continued_item(lines, index, item_match.group(1))
title_item = access_title_item(item)
fingerprint = make_fingerprint(str(path), line_number, "core", item)
body = body_for_item(
fingerprint=fingerprint,
source=str(path),
line=line_number,
section="Immediate Backlog",
status="open",
item=item,
note="Imported from the GovOPlaN access extraction plan immediate backlog.",
)
yield Candidate(
repo_key="core",
title=format_title("[Task]", title_item),
body=body,
labels=labels("core", "type/task", "area/auth", "area/module-system", "module/access"),
fingerprint=fingerprint,
source=str(path),
line=line_number,
)
def access_decision_candidates(path: pathlib.Path, lines: list[str]) -> Iterable[Candidate]:
in_section = False
for index, line in enumerate(lines):
line_number = index + 1
if line == "## Open Decisions":
in_section = True
continue
if in_section and line.startswith("## "):
break
if not in_section:
continue
item_match = re.match(r"^\s*[-*]\s+(.+?)\s*$", line)
if not item_match:
continue
item = collect_continued_item(lines, index, item_match.group(1))
fingerprint = make_fingerprint(str(path), line_number, "core", item)
body = body_for_item(
fingerprint=fingerprint,
source=str(path),
line=line_number,
section="Open Decisions",
status="decision needed",
item=item,
note="Imported from the GovOPlaN access extraction plan open decisions.",
)
yield Candidate(
repo_key="core",
title=format_title("[Task]", f"Decide: {item}"),
body=body,
labels=labels("core", "type/task", "status/needs-info", "codex/needs-human", "area/auth", "module/access"),
fingerprint=fingerprint,
source=str(path),
line=line_number,
)
def parse_simple_todo(path: pathlib.Path) -> Iterable[Candidate]:
lines = path.read_text(encoding="utf-8").splitlines()
for index, line in enumerate(lines):
line_number = index + 1
item_match = re.match(r"^\s*[-*]\s+(.+?)\s*$", line)
if not item_match:
continue
item = collect_continued_item(lines, index, item_match.group(1))
fingerprint = make_fingerprint(str(path), line_number, "web", item)
yield Candidate(
repo_key="web",
title=format_title("[Task]", item),
body=body_for_item(
fingerprint=fingerprint,
source=str(path),
line=line_number,
section="TODO",
status="open",
item=item,
note="Imported from the GovOPlaN website TODO file.",
),
labels=labels("web", "type/task", "area/marketing"),
fingerprint=fingerprint,
source=str(path),
line=line_number,
)
def classify_product_item(item: str, section: str) -> tuple[str, tuple[str, ...]]:
text = f"{section} {item}".lower()
repo_key = "core"
area = "area/devex"
issue_type = "type/task"
if "dsar search/export/delete" in text:
return "core", labels("core", "type/feature", "area/governance")
if "fully stream zip uploads" in text:
return "files", labels("files", "type/task", "area/api")
if "installer packaging approach" in text:
return "core", labels("core", "type/task", "area/devex")
if "profile reselection/revalidation" in text:
return "campaign", labels("campaign", "type/task", "area/api", "module/mail")
if "module boundaries" in text:
return "core", labels("core", issue_type, "area/module-system")
if "external storage connectors" in text:
repo_key = "files"
area = "area/api"
elif any(phrase in text for phrase in ("report, evidence", "recipient import", "campaign and attachment", "collaboration and advanced governance", "address book, templates")):
repo_key = "campaign"
area = "area/webui" if any(word in text for word in ("page", "table", "display", "ui", "wizard", "editor", "flow", "polish")) else "area/api"
elif "controlled sending" in text:
repo_key = "campaign"
area = "area/api"
elif "mail profiles" in text:
repo_key = "mail"
area = "area/api"
if any(word in text for word in ("smtp", "imap", "mailbox", "mail profile", "deliverability", "bounce", "openpgp", "s/mime", "pop3", "jmap")):
repo_key = "mail"
area = "area/api"
if any(word in text for word in ("seafile", "nextcloud", "webdav", "smb", "connector", "file rename", "managed storage", "live remote file")):
repo_key = "files"
area = "area/api"
if "external storage connectors" not in text and any(
word in text
for word in (
"campaign",
"recipient",
"attachment",
"zip",
"evidence",
"wizard",
"send",
"archive",
"message preview",
"address book",
"carddav",
"mailing lists",
)
):
repo_key = "campaign"
area = "area/webui" if any(word in text for word in ("page", "table", "display", "ui", "wizard", "editor", "flow", "polish")) else "area/api"
if repo_key == "core" and any(word in text for word in ("session", "device", "auth", "rbac", "role", "tenant", "governance", "policy", "retention", "dsar", "audit", "encryption", "ldap", "oidc", "saml")):
repo_key = "core"
area = "area/auth" if any(word in text for word in ("session", "auth", "ldap", "oidc", "saml")) else "area/governance"
if repo_key == "core" and any(word in text for word in ("module", "manifest", "module boundaries", "installable", "activatable", "deactivatable", "uninstallable")):
repo_key = "core"
area = "area/module-system"
if repo_key == "core" and any(word in text for word in ("release", "version metadata", "lockfile", "package", "tagging")):
repo_key = "core"
area = "area/release"
if repo_key == "core" and not any(
phrase in text
for phrase in (
"external storage connectors",
"controlled sending",
"report, evidence",
"recipient import",
"campaign and attachment",
)
) and any(word in text for word in ("postgres", "redis", "celery", "worker", "deployment", "installer", "backup", "monitoring", "configuration", "dev profile", ".env")):
repo_key = "core"
area = "area/devex"
if any(word in text for word in ("documentation", "handbook", "docs", "runbook")):
area = "area/docs"
if "ui lists often start at index 2" in text:
issue_type = "type/bug"
area = "area/webui"
repo_key = "core"
if any(word in text for word in ("add ", "implement ", "build ", "provide ", "allow ", "make ", "extend ", "replace ")):
issue_type = "type/feature"
if any(word in text for word in ("cleanup", "remove legacy", "fragile", "deferred cleanup")):
issue_type = "type/debt"
return repo_key, labels(repo_key, issue_type, area)
def collect_continued_item(lines: list[str], index: int, initial: str) -> str:
parts = [initial.strip()]
next_index = index + 1
while next_index < len(lines):
line = lines[next_index]
if not line.startswith((" ", "\t")):
break
stripped = line.strip()
if not stripped:
break
if re.match(r"^[-*]\s+", stripped) or stripped.startswith("#"):
break
parts.append(stripped)
next_index += 1
return " ".join(parts)
def labels(repo_key: str, *extra: str) -> tuple[str, ...]:
base = {"status/triage", "source/backlog-import", "codex/ready", MODULE_LABELS[repo_key]}
base.update(extra)
if "status/needs-info" in base:
base.discard("status/triage")
return tuple(sorted(base))
def type_prefix(label_set: Iterable[str]) -> str:
labels_set = set(label_set)
if "type/bug" in labels_set:
return "[Bug]"
if "type/feature" in labels_set:
return "[Feature]"
if "type/debt" in labels_set:
return "[Debt]"
if "type/docs" in labels_set:
return "[Docs]"
return "[Task]"
def format_title(prefix: str, item: str) -> str:
clean = re.sub(r"\s+", " ", item).strip().rstrip(".")
title = f"{prefix} {clean}"
if len(title) <= 180:
return title
return f"{title[:177].rstrip()}..."
def access_title_item(item: str) -> str:
lowered = item.lower()
if "compatibility wrappers" in lowered:
return "Wire core auth compatibility through access capabilities"
return item
def body_for_item(
*,
fingerprint: str,
source: str,
line: int,
section: str,
status: str,
item: str,
note: str,
) -> str:
lines = [
f"<!-- codex-backlog-fingerprint:{fingerprint} -->",
"",
note,
"",
f"- Source: `{source}`",
f"- Line: `{line}`",
f"- Section: `{section}`",
]
if status:
lines.append(f"- Source status: `{status}`")
lines.extend(
[
"",
"Imported backlog item:",
"",
"```markdown",
f"- [ ] {item}",
"```",
]
)
return "\n".join(lines)
def section_title(headings: list[tuple[int, str]]) -> str:
return " > ".join(title for _, title in headings)
def make_fingerprint(source: str, line: int, repo_key: str, item: str) -> str:
stable = "\n".join([source, str(line), repo_key, item])
return hashlib.sha256(stable.encode("utf-8")).hexdigest()[:24]
def group_candidates(candidates: list[Candidate]) -> dict[str, list[Candidate]]:
grouped: dict[str, list[Candidate]] = {}
for candidate in candidates:
grouped.setdefault(candidate.repo_key, []).append(candidate)
return grouped
def load_repo_states(token: str, repo_keys: list[str]) -> dict[str, RepoState]:
states: dict[str, RepoState] = {}
for repo_key in repo_keys:
root = repo_root(REPO_ROOTS[repo_key])
target = infer_target(root)
client = GiteaClient(target, token)
label_ids = label_ids_by_name(client, target.owner, target.repo)
issues = client.paginate(repo_path(target.owner, target.repo, "/issues"), query={"state": "all"}, limit=100)
fingerprints: set[str] = set()
titles: set[str] = set()
for issue in issues:
body = str(issue.get("body") or "")
if issue.get("state") == "closed" and "Moved to `" in body:
continue
titles.add(normalize_title(str(issue.get("title") or "")))
fingerprints.update(re.findall(r"codex-backlog-fingerprint:([0-9a-f]{24})", body))
states[repo_key] = RepoState(target=target, client=client, label_ids=label_ids, fingerprints=fingerprints, titles=titles)
return states
def already_imported(candidate: Candidate, state: RepoState) -> bool:
return candidate.fingerprint in state.fingerprints or normalize_title(candidate.title) in state.titles
def normalize_title(title: str) -> str:
return re.sub(r"[^a-z0-9]+", " ", title.lower()).strip()
def validate_labels(candidate: Candidate, state: RepoState) -> None:
missing = sorted(label for label in candidate.labels if label not in state.label_ids)
if missing:
joined = ", ".join(missing)
raise GiteaError(f"{candidate.repo_key} is missing labels for {candidate.title!r}: {joined}")
def print_preview(candidates: list[Candidate], limit: int = 160) -> None:
for repo_key, grouped in group_candidates(candidates).items():
print(f"{repo_key}:")
for candidate in grouped[:limit]:
print(f" {candidate.title}")
if len(grouped) > limit:
print(f" ... {len(grouped) - limit} more")
if __name__ == "__main__":
raise SystemExit(main())

View File

@@ -0,0 +1,97 @@
#!/usr/bin/env python3
"""Post a standardized Codex state update comment to a Gitea issue."""
from __future__ import annotations
import argparse
import os
import pathlib
import sys
from gitea_common import GiteaClient, GiteaError, infer_target, load_dotenv, repo_path, repo_root, require_token
STATUS_LABELS = {
"started": "status/in-progress",
"progress": "status/in-progress",
"blocked": "status/blocked",
"needs-info": "status/needs-info",
"ready": "status/ready",
"done": "",
"note": "",
}
def main() -> int:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--root", type=pathlib.Path, default=pathlib.Path.cwd(), help="repository root or child path")
parser.add_argument("--remote", default="origin", help="git remote to use for target inference")
parser.add_argument("--env-file", type=pathlib.Path, help="dotenv file to read before using GITEA_* values")
parser.add_argument("--issue", type=int, required=True, help="Gitea issue number")
parser.add_argument("--status", choices=sorted(STATUS_LABELS), default="note")
parser.add_argument("--summary", action="append", default=[], help="summary bullet; may be repeated")
parser.add_argument("--changed", action="append", default=[], help="changed file path; may be repeated")
parser.add_argument("--test", action="append", default=[], help="verification command/result; may be repeated")
parser.add_argument("--next", action="append", default=[], help="next step or blocker; may be repeated")
parser.add_argument("--body-file", type=pathlib.Path, help="additional Markdown body to append")
parser.add_argument("--close", action="store_true", help="close the issue after posting the note")
parser.add_argument("--apply", action="store_true", help="post the comment and optional state update")
args = parser.parse_args()
try:
root = repo_root(args.root)
load_dotenv(args.env_file or root / ".env")
target = infer_target(root, args.remote)
token = require_token() if args.apply else os.environ.get("GITEA_TOKEN")
body = build_body(args)
print(f"Target: {target.display}")
print(f"Issue: #{args.issue}")
print(body)
if not args.apply:
print("Dry run only. Re-run with --apply and GITEA_TOKEN to post.")
return 0
client = GiteaClient(target, token)
client.request_json(
"POST",
repo_path(target.owner, target.repo, f"/issues/{args.issue}/comments"),
body={"body": body},
)
if args.close:
client.request_json(
"PATCH",
repo_path(target.owner, target.repo, f"/issues/{args.issue}"),
body={"state": "closed"},
)
print("Posted Codex note.")
return 0
except GiteaError as exc:
print(f"error: {exc}", file=sys.stderr)
return 1
def build_body(args: argparse.Namespace) -> str:
lines = [f"## Codex State: {args.status}", ""]
append_section(lines, "Summary", args.summary)
append_section(lines, "Changed Files", [f"`{item}`" for item in args.changed])
append_section(lines, "Verification", [f"`{item}`" for item in args.test])
append_section(lines, "Next / Blocked", args.next)
if STATUS_LABELS[args.status]:
lines.extend(["", f"Suggested status label: `{STATUS_LABELS[args.status]}`"])
if args.body_file:
lines.extend(["", args.body_file.read_text(encoding="utf-8").strip()])
return "\n".join(lines).rstrip() + "\n"
def append_section(lines: list[str], title: str, items: list[str]) -> None:
if not items:
return
lines.extend([f"### {title}", ""])
lines.extend(f"- {item}" for item in items)
lines.append("")
if __name__ == "__main__":
raise SystemExit(main())

View File

@@ -0,0 +1,600 @@
#!/usr/bin/env python3
"""Import backlog-like files from git.add-ideas.de repositories into Gitea."""
from __future__ import annotations
import argparse
import csv
import dataclasses
import hashlib
import os
import pathlib
import re
import subprocess
import sys
from typing import Any, Iterable
from gitea_common import GiteaClient, GiteaError, infer_target, label_ids_by_name, load_dotenv, repo_path, require_token
GIT_ROOT = pathlib.Path("/mnt/DATA/git")
PRODUCTS_ROOT = pathlib.Path("/mnt/DATA/Nextcloud/ADD ideas UG/Products")
EXCLUDED_FILE_PATTERNS = (
"/.git/",
"/.venv/",
"/venv/",
"/site-packages/",
"/node_modules/",
"/__pycache__/",
"/dist/",
"/build/",
"/.cache/",
"/.module-test-build/",
)
BACKLOG_NAME_RE = re.compile(r"(backlog|todo|roadmap|plan|issue|milestone|action)", re.IGNORECASE)
ACTION_RE = re.compile(
r"^(add|avoid|build|convert|create|define|delay|design|document|enable|ensure|expand|extract|fetch|fix|finalize|generate|"
r"implement|import|improve|integrate|keep|make|move|optimize|persist|precompute|promote|provide|"
r"publish|query|replace|review|run|seed|serve|split|start|store|support|test|track|use|wire)\b",
re.IGNORECASE,
)
SKIP_SECTION_RE = re.compile(
r"(completed|implemented|current state|recent fixes|recently completed|working assumptions|stable decisions|"
r"available now|done|archive)",
re.IGNORECASE,
)
ACTIVE_SECTION_RE = re.compile(
r"(p0|p1|p2|p3|mvp|milestone|phase|todo|backlog|roadmap|tasks|testing|frontend|backend|routing|"
r"data outputs|future|critical path|next sprint|optimization|hardening|enhancements?)",
re.IGNORECASE,
)
GENERIC_LABELS = [
("type/task", "1d76db", "Implementation, maintenance, migration, or operational work."),
("type/feature", "0e8a16", "New user-visible behavior or product capability."),
("type/debt", "fbca04", "Cleanup, refactoring, risk reduction, or deferred engineering work."),
("type/docs", "5319e7", "Documentation, process, or developer workflow work."),
("status/triage", "d4c5f9", "Needs review, ownership, priority, or acceptance criteria."),
("source/backlog-import", "bfdadc", "Imported from markdown, text, CSV, roadmap, backlog, plan, or TODO files."),
("codex/ready", "0e8a16", "Suitable for Codex to pick up with the existing issue context."),
("priority/p0", "b60205", "Immediate stop-the-line priority."),
("priority/p1", "d93f0b", "High priority for the next focused work window."),
("priority/p2", "fbca04", "Normal planned priority."),
("priority/p3", "c2e0c6", "Low priority or opportunistic cleanup."),
]
@dataclasses.dataclass(frozen=True)
class RepoInfo:
root: pathlib.Path
remote: str
owner: str
repo: str
@property
def key(self) -> str:
return self.root.name
@dataclasses.dataclass(frozen=True)
class SourceFile:
repo_key: str
path: pathlib.Path
source_kind: str
@dataclasses.dataclass(frozen=True)
class Candidate:
repo_key: str
title: str
body: str
labels: tuple[str, ...]
priority: str
milestone: str
fingerprint: str
source: str
line: int
def main() -> int:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--env-file", type=pathlib.Path, default=pathlib.Path("/home/zemion/.config/gitea/gitea.env"))
parser.add_argument("--git-root", type=pathlib.Path, default=GIT_ROOT)
parser.add_argument("--products-root", type=pathlib.Path, default=PRODUCTS_ROOT)
parser.add_argument("--apply", action="store_true")
args = parser.parse_args()
try:
load_dotenv(args.env_file)
repos = discover_hosted_repos(args.git_root)
sources = discover_sources(repos, args.products_root)
candidates = list(build_candidates(sources))
candidates = dedupe_candidates(candidates)
candidates.sort(key=lambda item: (item.repo_key, item.source, item.line, item.title))
print(f"Hosted repositories: {len(repos)}")
print(f"Backlog-like source files: {len(sources)}")
print(f"Issue candidates: {len(candidates)}")
for repo_key, count in grouped_counts(candidates).items():
print(f" {repo_key}: {count}")
token = require_token() if args.apply else os.environ.get("GITEA_TOKEN")
if not token:
print_preview(candidates)
print("Dry run without GITEA_TOKEN: duplicate comparison skipped.")
return 0
missing_by_repo: dict[str, list[Candidate]] = {}
skipped = 0
states: dict[str, RepoState] = {}
for repo_key in sorted({candidate.repo_key for candidate in candidates}):
states[repo_key] = load_repo_state(repos[repo_key], token, apply=args.apply)
for candidate in candidates:
state = states[candidate.repo_key]
if candidate.fingerprint in state.fingerprints or normalize_title(candidate.title) in state.titles:
skipped += 1
continue
missing_by_repo.setdefault(candidate.repo_key, []).append(candidate)
missing_total = sum(len(items) for items in missing_by_repo.values())
print(f"Skipped existing: {skipped}")
print(f"Missing candidates: {missing_total}")
print_preview([candidate for items in missing_by_repo.values() for candidate in items])
if not args.apply:
print("Dry run only. Re-run with --apply to create missing issues.")
return 0
for repo_key, repo_candidates in missing_by_repo.items():
state = states[repo_key]
for candidate in repo_candidates:
label_ids = [state.label_ids[label] for label in candidate.labels]
if candidate.priority not in candidate.labels:
label_ids.append(state.label_ids[candidate.priority])
milestone_id = ensure_milestone(state, candidate.milestone)
created = state.client.request_json(
"POST",
repo_path(state.target.owner, state.target.repo, "/issues"),
body={
"title": candidate.title,
"body": candidate.body,
"labels": label_ids,
"milestone": milestone_id,
},
)
number = created.get("number") or created.get("index")
print(f"created {state.target.owner}/{state.target.repo}#{number}: {candidate.title}")
state.fingerprints.add(candidate.fingerprint)
state.titles.add(normalize_title(candidate.title))
return 0
except GiteaError as exc:
print(f"error: {exc}", file=sys.stderr)
return 1
@dataclasses.dataclass
class RepoState:
target: Any
client: GiteaClient
label_ids: dict[str, int]
milestone_ids: dict[str, int]
fingerprints: set[str]
titles: set[str]
def discover_hosted_repos(git_root: pathlib.Path) -> dict[str, RepoInfo]:
repos: dict[str, RepoInfo] = {}
for gitdir in sorted(git_root.glob("*/.git")):
root = gitdir.parent
result = subprocess.run(
["git", "-C", str(root), "remote", "get-url", "origin"],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.DEVNULL,
text=True,
)
remote = result.stdout.strip()
if "git.add-ideas.de" not in remote:
continue
target = infer_target(root)
repos[root.name] = RepoInfo(root=root, remote=remote, owner=target.owner, repo=target.repo)
return repos
def discover_sources(repos: dict[str, RepoInfo], products_root: pathlib.Path) -> list[SourceFile]:
sources: list[SourceFile] = []
for repo in repos.values():
for path in repo.root.rglob("*"):
if not path.is_file() or not BACKLOG_NAME_RE.search(path.name):
continue
path_text = path.as_posix()
if any(pattern in path_text for pattern in EXCLUDED_FILE_PATTERNS):
continue
if is_excluded_repo_file(path):
continue
if is_text_backlog(path):
sources.append(SourceFile(repo_key=repo.key, path=path, source_kind="repo"))
product_map = {
"co2api": "emission-api-lib",
"govoplan": "govoplan-core",
"meubility": "meubility-workbench",
"mousehold": "mousehold",
"prvnncr": "prvnncr-server",
}
if products_root.exists():
for product_dir in sorted(path for path in products_root.iterdir() if path.is_dir()):
repo_key = product_map.get(product_dir.name)
if not repo_key or repo_key not in repos:
continue
for path in product_dir.rglob("*"):
if path.is_file() and BACKLOG_NAME_RE.search(path.name) and is_text_backlog(path):
sources.append(SourceFile(repo_key=repo_key, path=path, source_kind="product"))
unique: dict[tuple[str, pathlib.Path], SourceFile] = {}
for source in sources:
unique[(source.repo_key, source.path.resolve())] = source
return sorted(unique.values(), key=lambda item: (item.repo_key, str(item.path)))
def is_excluded_repo_file(path: pathlib.Path) -> bool:
text = path.as_posix()
name = path.name.lower()
if (
"/tools/gitea/gitea-" in text
or text.endswith("/tools/gitea/gitea_common.py")
):
return True
if "testing_plan" in name or "test_plan" in name:
return True
if text.endswith("/docs/GITEA_ISSUES.md"):
return True
if text.endswith("/docs/GOVOPLAN_MASTER_ROADMAP.md"):
return True
if text.endswith("/addideas-govoplan-website/TODO.md"):
return True
return False
def is_text_backlog(path: pathlib.Path) -> bool:
if path.suffix.lower() not in {".md", ".txt", ".csv"}:
return False
try:
chunk = path.read_bytes()[:4096]
except OSError:
return False
if b"\x00" in chunk:
return False
return True
def build_candidates(sources: list[SourceFile]) -> Iterable[Candidate]:
for source in sources:
suffix = source.path.suffix.lower()
if suffix == ".csv":
yield from parse_csv_source(source)
else:
yield from parse_text_source(source)
def parse_csv_source(source: SourceFile) -> Iterable[Candidate]:
with source.path.open("r", encoding="utf-8-sig", newline="") as handle:
reader = csv.DictReader(handle)
for index, row in enumerate(reader, start=2):
story = (row.get("story") or row.get("title") or row.get("name") or "").strip()
if not story:
continue
identifier = (row.get("id") or "").strip()
priority = normalize_priority(row.get("priority") or "")
milestone = (row.get("epic") or "Backlog").strip()
title = format_title("[Feature]", f"{identifier}: {story}" if identifier else story)
fingerprint = make_fingerprint(source.path, index, source.repo_key, story)
body = "\n".join(
[
f"<!-- codex-generic-backlog-fingerprint:{fingerprint} -->",
"",
"Imported from a CSV backlog file.",
"",
f"- Source: `{source.path}`",
f"- Line: `{index}`",
f"- Source kind: `{source.source_kind}`",
f"- Milestone: `{milestone}`",
"",
"CSV row:",
"",
"```text",
", ".join(f"{key}={value}" for key, value in row.items()),
"```",
]
)
yield Candidate(
repo_key=source.repo_key,
title=title,
body=body,
labels=("type/feature", "status/triage", "source/backlog-import", "codex/ready"),
priority=priority,
milestone=milestone,
fingerprint=fingerprint,
source=str(source.path),
line=index,
)
def parse_text_source(source: SourceFile) -> Iterable[Candidate]:
try:
lines = source.path.read_text(encoding="utf-8").splitlines()
except UnicodeDecodeError:
lines = source.path.read_text(encoding="latin-1").splitlines()
headings: list[tuple[int, str]] = []
in_tasks = False
for index, line in enumerate(lines):
line_number = index + 1
heading = parse_heading(line)
if heading:
level, title = heading
headings = [(h_level, h_title) for h_level, h_title in headings if h_level < level]
headings.append((level, title))
in_tasks = False
continue
if re.match(r"^\s*(tasks|action items|immediate todos?|recommended next sprint)\s*:?\s*$", line, re.IGNORECASE):
in_tasks = True
continue
item = parse_open_item(line)
if item is None and is_action_context(source.path, headings, in_tasks):
item = parse_plain_action_item(line)
if item is None:
continue
if not item or should_skip_item(item, headings):
continue
section = section_title(headings)
priority = priority_from_context(section, item)
milestone = milestone_from_context(source.path, section)
prefix = "[Feature]" if ACTION_RE.match(item) else "[Task]"
title = format_title(prefix, item)
fingerprint = make_fingerprint(source.path, line_number, source.repo_key, item)
body = "\n".join(
[
f"<!-- codex-generic-backlog-fingerprint:{fingerprint} -->",
"",
"Imported from a backlog-like text file.",
"",
f"- Source: `{source.path}`",
f"- Line: `{line_number}`",
f"- Source kind: `{source.source_kind}`",
f"- Section: `{section or 'none'}`",
"",
"Imported item:",
"",
"```text",
item,
"```",
]
)
yield Candidate(
repo_key=source.repo_key,
title=title,
body=body,
labels=("type/feature" if prefix == "[Feature]" else "type/task", "status/triage", "source/backlog-import", "codex/ready"),
priority=priority,
milestone=milestone,
fingerprint=fingerprint,
source=str(source.path),
line=line_number,
)
def parse_heading(line: str) -> tuple[int, str] | None:
markdown = re.match(r"^(#{1,6})\s+(.+?)\s*$", line)
if markdown:
return len(markdown.group(1)), clean_text(markdown.group(2))
underlined = re.match(r"^(.+?)\s*$", line)
if underlined and line.strip() and len(line.strip()) < 120:
return None
return None
def parse_open_item(line: str) -> str | None:
patterns = [
r"^(?P<indent>\s*)[-*]\s+\[\s\]\s+(?P<text>.+?)\s*$",
r"^(?P<indent>\s*)[-*]\s+☐\s+(?P<text>.+?)\s*$",
r"^(?P<indent>\s*)\d+[.)]\s+☐\s+(?P<text>.+?)\s*$",
r"^(?P<indent>\s*)[-*]\s+\[(?!x\]|X\])(?:[^\]]+)\]\s+(?P<text>.+?)\s*$",
]
for pattern in patterns:
match = re.match(pattern, line)
if match and leading_width(match.group("indent")) == 0:
return clean_text(match.group("text"))
return None
def parse_plain_action_item(line: str) -> str | None:
match = re.match(r"^(?P<indent>\s*)[-*]\s+(?P<text>.+?)\s*$", line)
if match and leading_width(match.group("indent")) == 0:
text = clean_text(match.group("text"))
if "" in text or text.startswith(("", "[x]", "[X]")):
return None
if ACTION_RE.match(text) or is_short_noun_task(text):
return text
return None
def is_action_context(path: pathlib.Path, headings: list[tuple[int, str]], in_tasks: bool) -> bool:
if in_tasks:
return True
context = section_title(headings)
if not context:
return False
if SKIP_SECTION_RE.search(context):
return False
if ACTIVE_SECTION_RE.search(context):
return True
return path.name.lower() in {"todo.txt", "todo.md"}
def should_skip_item(item: str, headings: list[tuple[int, str]]) -> bool:
context = section_title(headings)
if "" in item or item.startswith(("", "[x]", "[X]")):
return True
if item.endswith((",", ";")):
return True
if SKIP_SECTION_RE.search(context):
return True
if len(item) < 4:
return True
if item.endswith(":") and len(item.split()) <= 6:
return True
return False
def is_short_noun_task(text: str) -> bool:
return len(text.split()) <= 8 and not text.endswith(".") and not re.search(r"://", text)
def clean_text(text: str) -> str:
text = text.strip()
text = re.sub(r"^☐\s*", "", text)
text = re.sub(r"^✅\s*", "", text)
text = re.sub(r"\s+", " ", text)
return text.strip(" -")
def leading_width(indent: str) -> int:
return len(indent.replace("\t", " "))
def section_title(headings: list[tuple[int, str]]) -> str:
return " > ".join(title for _, title in headings)
def priority_from_context(section: str, item: str) -> str:
text = f"{section} {item}".lower()
if "p0" in text or "critical path" in text:
return "priority/p0"
if "p1" in text or "immediate" in text or "next sprint" in text:
return "priority/p1"
if "p2" in text:
return "priority/p2"
if "p3" in text or "future" in text or "later" in text:
return "priority/p3"
return "priority/p2"
def normalize_priority(value: str) -> str:
value = value.strip().lower()
if value == "p0":
return "priority/p0"
if value == "p1":
return "priority/p1"
if value == "p2":
return "priority/p2"
if value == "p3":
return "priority/p3"
return "priority/p2"
def milestone_from_context(path: pathlib.Path, section: str) -> str:
if section:
parts = [part for part in section.split(" > ") if ACTIVE_SECTION_RE.search(part)]
if parts:
return parts[-1][:120]
return path.stem.replace("_", " ").replace("-", " ").title()
def format_title(prefix: str, item: str) -> str:
title = f"{prefix} {clean_text(item).rstrip('.')}"
if len(title) <= 180:
return title
return f"{title[:177].rstrip()}..."
def make_fingerprint(path: pathlib.Path, line: int, repo_key: str, item: str) -> str:
stable = "\n".join([str(path.resolve()), str(line), repo_key, item])
return hashlib.sha256(stable.encode("utf-8")).hexdigest()[:24]
def dedupe_candidates(candidates: list[Candidate]) -> list[Candidate]:
seen: set[tuple[str, str]] = set()
unique: list[Candidate] = []
for candidate in candidates:
key = (candidate.repo_key, normalize_title(candidate.title))
if key in seen:
continue
seen.add(key)
unique.append(candidate)
return unique
def grouped_counts(candidates: list[Candidate]) -> dict[str, int]:
counts: dict[str, int] = {}
for candidate in candidates:
counts[candidate.repo_key] = counts.get(candidate.repo_key, 0) + 1
return dict(sorted(counts.items()))
def load_repo_state(repo: RepoInfo, token: str, *, apply: bool) -> RepoState:
target = infer_target(repo.root)
client = GiteaClient(target, token)
label_ids = label_ids_by_name(client, target.owner, target.repo)
if apply:
for name, color, description in GENERIC_LABELS:
if name in label_ids:
continue
created = client.request_json(
"POST",
repo_path(target.owner, target.repo, "/labels"),
body={"name": name, "color": color, "description": description, "exclusive": name.startswith(("type/", "status/", "priority/"))},
)
label_ids[str(created["name"])] = int(created["id"])
print(f"created label {target.owner}/{target.repo}:{name}")
missing = [name for name, _, _ in GENERIC_LABELS if name not in label_ids]
if missing and apply:
raise GiteaError(f"{repo.key} missing labels: {', '.join(missing)}. Re-run with --apply to create them.")
milestones = client.paginate(repo_path(target.owner, target.repo, "/milestones"), query={"state": "all"}, limit=50)
milestone_ids = {str(item["title"]): int(item["id"]) for item in milestones}
issues = client.paginate(repo_path(target.owner, target.repo, "/issues"), query={"state": "all"}, limit=50)
fingerprints: set[str] = set()
titles: set[str] = set()
for issue in issues:
body = str(issue.get("body") or "")
titles.add(normalize_title(str(issue.get("title") or "")))
fingerprints.update(re.findall(r"codex-(?:generic-)?backlog-fingerprint:([0-9a-f]{24})", body))
return RepoState(target=target, client=client, label_ids=label_ids, milestone_ids=milestone_ids, fingerprints=fingerprints, titles=titles)
def ensure_milestone(state: RepoState, title: str) -> int:
if title in state.milestone_ids:
return state.milestone_ids[title]
created = state.client.request_json(
"POST",
repo_path(state.target.owner, state.target.repo, "/milestones"),
body={"title": title, "state": "open", "description": "Created from imported backlog-like files."},
)
state.milestone_ids[title] = int(created["id"])
print(f"created milestone {state.target.owner}/{state.target.repo}:{title}")
return state.milestone_ids[title]
def normalize_title(title: str) -> str:
return re.sub(r"[^a-z0-9]+", " ", title.lower()).strip()
def print_preview(candidates: list[Candidate], limit_per_repo: int = 80) -> None:
by_repo: dict[str, list[Candidate]] = {}
for candidate in candidates:
by_repo.setdefault(candidate.repo_key, []).append(candidate)
for repo_key, items in sorted(by_repo.items()):
print(f"{repo_key}:")
for item in items[:limit_per_repo]:
print(f" {item.title} [{item.priority}; {item.milestone}]")
if len(items) > limit_per_repo:
print(f" ... {len(items) - limit_per_repo} more")
if __name__ == "__main__":
raise SystemExit(main())

View File

@@ -0,0 +1,113 @@
#!/usr/bin/env python3
"""Install Gitea issue workflow files into one or more repositories."""
from __future__ import annotations
import argparse
import pathlib
import sys
from gitea_common import GiteaError, repo_root
SOURCE_ROOT = pathlib.Path(__file__).resolve().parents[2]
WORKFLOW_FILES = (
".gitea/PULL_REQUEST_TEMPLATE.md",
".gitea/ISSUE_TEMPLATE/bug_report.md",
".gitea/ISSUE_TEMPLATE/config.yaml",
".gitea/ISSUE_TEMPLATE/docs_workflow.md",
".gitea/ISSUE_TEMPLATE/feature_request.md",
".gitea/ISSUE_TEMPLATE/task.md",
".gitea/ISSUE_TEMPLATE/tech_debt.md",
)
MODULE_LABEL_BY_REPO = {
"govoplan-core": "module/core",
"govoplan-access": "module/access",
"govoplan-admin": "module/admin",
"govoplan-tenancy": "module/tenancy",
"govoplan-policy": "module/policy",
"govoplan-audit": "module/audit",
"govoplan-mail": "module/mail",
"govoplan-files": "module/files",
"govoplan-campaign": "module/campaign",
"addideas-govoplan-website": "area/marketing",
}
def main() -> int:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("targets", nargs="*", type=pathlib.Path, help="target repository roots or child paths")
parser.add_argument("--module-label", help="module label to write into issue templates; only valid for one target")
parser.add_argument("--include-labels-file", action="store_true", help="also copy docs/gitea-labels.json")
parser.add_argument("--apply", action="store_true", help="write files instead of previewing changes")
args = parser.parse_args()
try:
target_roots = [repo_root(path) for path in (args.targets or [pathlib.Path.cwd()])]
unique_roots = list(dict.fromkeys(target_roots))
if args.module_label and len(unique_roots) != 1:
raise GiteaError("--module-label can only be used with a single target")
rel_paths = list(WORKFLOW_FILES)
if args.include_labels_file:
rel_paths.append("docs/gitea-labels.json")
for target_root in unique_roots:
module_label = args.module_label or infer_module_label(target_root)
install_files(target_root, rel_paths, module_label, apply=args.apply)
if not args.apply:
print("Dry run only. Re-run with --apply to write files.")
return 0
except GiteaError as exc:
print(f"error: {exc}", file=sys.stderr)
return 1
def infer_module_label(target_root: pathlib.Path) -> str:
repo_name = target_root.name
try:
repo_name = target_root.resolve().name
except OSError:
pass
label = MODULE_LABEL_BY_REPO.get(repo_name)
if label:
return label
if repo_name.startswith("govoplan-"):
suffix = repo_name.removeprefix("govoplan-")
if suffix:
return f"module/{suffix}"
raise GiteaError(f"cannot infer repository label for {target_root}. Use --module-label module/<name> or area/<name>.")
def install_files(target_root: pathlib.Path, rel_paths: list[str], module_label: str, *, apply: bool) -> None:
print(f"Target: {target_root} ({module_label})")
for rel_path in rel_paths:
source = SOURCE_ROOT / rel_path
target = target_root / rel_path
if not source.exists():
raise GiteaError(f"missing workflow source file: {source}")
content = source.read_text(encoding="utf-8")
content = transform_content(rel_path, content, module_label)
existing = target.read_text(encoding="utf-8") if target.exists() else None
if existing == content:
print(f"unchanged {rel_path}")
continue
action = "create" if existing is None else "update"
print(f"{action if apply else 'would ' + action} {rel_path}")
if apply:
target.parent.mkdir(parents=True, exist_ok=True)
target.write_text(content, encoding="utf-8")
def transform_content(rel_path: str, content: str, module_label: str) -> str:
if rel_path.startswith(".gitea/ISSUE_TEMPLATE/") and rel_path.endswith(".md"):
return content.replace(" - module/core", f" - {module_label}")
return content
if __name__ == "__main__":
raise SystemExit(main())

View File

@@ -0,0 +1,250 @@
#!/usr/bin/env python3
"""Move issue labels from repository-local labels to organization labels."""
from __future__ import annotations
import argparse
import os
import pathlib
import re
import sys
from typing import Any
from gitea_common import (
GiteaClient,
GiteaError,
infer_target,
load_dotenv,
org_path,
repo_path,
repo_root,
require_token,
)
def main() -> int:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--root", type=pathlib.Path, default=pathlib.Path.cwd(), help="repository root used for Gitea URL inference")
parser.add_argument("--remote", default="origin", help="git remote to use for target inference")
parser.add_argument("--env-file", type=pathlib.Path, default=pathlib.Path("/home/zemion/.config/gitea/gitea.env"))
parser.add_argument("--org", help="organization owner; defaults to inferred owner")
parser.add_argument("--repo", action="append", default=[], help="repository name to process; may be repeated")
parser.add_argument("--repo-regex", default=".*", help="only process repositories whose name matches this regex")
parser.add_argument(
"--issue-mode",
choices=("replace", "delete-add"),
default="replace",
help="replace labels in one request, or delete local ids before adding org ids",
)
parser.add_argument("--delete-repo-labels", action="store_true", help="delete matching repository-local labels after issue and PR migration")
parser.add_argument("--apply", action="store_true", help="apply changes; omit for dry-run")
args = parser.parse_args()
try:
root = repo_root(args.root)
load_dotenv(args.env_file or root / ".env")
target = infer_target(root, args.remote)
owner = args.org or target.owner
client = GiteaClient(target, require_token() if args.apply else os.environ.get("GITEA_TOKEN"))
if client.token is None:
raise GiteaError("GITEA_TOKEN is required for dry-run and apply because migration inspects live issue labels.")
repo_filter = re.compile(args.repo_regex)
org_labels = _labels_by_name(client.paginate(org_path(owner, "/labels"), limit=100))
if not org_labels:
raise GiteaError(f"{owner} has no organization labels")
repos = _selected_repositories(client, owner, args.repo, repo_filter)
totals = Totals()
print(f"Organization: {owner}")
print(f"Repositories: {len(repos)}")
print(f"Mode: {'apply' if args.apply else 'dry-run'}")
print(f"Delete repository labels: {args.delete_repo_labels}")
for index, repo in enumerate(repos, start=1):
repo_result = migrate_repository(
client,
owner=owner,
repo=repo,
org_labels=org_labels,
issue_mode=args.issue_mode,
apply=args.apply,
delete_repo_labels=args.delete_repo_labels,
)
totals.add(repo_result)
if repo_result.has_work:
print(
f"[{index}/{len(repos)}] {repo}: "
f"issue-labels={repo_result.issue_label_migrations}, "
f"delete-labels={repo_result.repo_label_deletions}, "
f"errors={len(repo_result.errors)}"
)
for error in repo_result.errors:
print(f" error: {error}")
else:
print(f"[{index}/{len(repos)}] {repo}: no matching local labels")
print("Summary:")
print(f" repositories processed: {totals.repositories}")
print(f" repositories with matching local labels: {totals.repositories_with_work}")
print(f" issue/PR local labels migrated: {totals.issue_label_migrations}")
print(f" repository labels deleted: {totals.repo_label_deletions}")
print(f" errors: {totals.errors}")
if totals.errors:
return 1
return 0
except GiteaError as exc:
print(f"error: {exc}", file=sys.stderr)
return 1
class RepoResult:
def __init__(self) -> None:
self.issue_label_migrations = 0
self.repo_label_deletions = 0
self.errors: list[str] = []
self.matched_local_labels = 0
@property
def has_work(self) -> bool:
return bool(self.matched_local_labels or self.issue_label_migrations or self.repo_label_deletions or self.errors)
class Totals:
def __init__(self) -> None:
self.repositories = 0
self.repositories_with_work = 0
self.issue_label_migrations = 0
self.repo_label_deletions = 0
self.errors = 0
def add(self, result: RepoResult) -> None:
self.repositories += 1
if result.has_work:
self.repositories_with_work += 1
self.issue_label_migrations += result.issue_label_migrations
self.repo_label_deletions += result.repo_label_deletions
self.errors += len(result.errors)
def migrate_repository(
client: GiteaClient,
*,
owner: str,
repo: str,
org_labels: dict[str, dict[str, Any]],
issue_mode: str,
apply: bool,
delete_repo_labels: bool,
) -> RepoResult:
result = RepoResult()
repo_labels = client.paginate(repo_path(owner, repo, "/labels"), limit=100)
local_labels = {
str(label.get("name") or ""): label
for label in repo_labels
if str(label.get("name") or "") in org_labels
}
result.matched_local_labels = len(local_labels)
if not local_labels:
return result
local_by_id = {_label_id(label): label for label in local_labels.values() if _label_id(label) is not None}
org_by_name = {name: _label_id(label) for name, label in org_labels.items()}
for issue_type in ("issues", "pulls"):
issues = client.paginate(
repo_path(owner, repo, "/issues"),
query={"state": "all", "type": issue_type},
limit=100,
)
for issue in issues:
issue_number = int(issue.get("number") or issue.get("index"))
issue_label_ids = [_label_id(label) for label in issue.get("labels") or []]
issue_label_ids = [label_id for label_id in issue_label_ids if label_id is not None]
final_label_ids = list(issue_label_ids)
changed = False
migrated_count = 0
local_ids_to_remove: list[int] = []
org_ids_to_add: list[int] = []
for label in issue.get("labels") or []:
local_id = _label_id(label)
if local_id is None or local_id not in local_by_id:
continue
name = str(label.get("name") or "")
org_id = org_by_name.get(name)
if org_id is None:
continue
local_ids_to_remove.append(local_id)
if org_id not in issue_label_ids and org_id not in org_ids_to_add:
org_ids_to_add.append(org_id)
final_label_ids = [label_id for label_id in final_label_ids if label_id != local_id]
if org_id not in final_label_ids:
final_label_ids.append(org_id)
changed = True
migrated_count += 1
if changed:
if apply:
if issue_mode == "delete-add":
for local_id in local_ids_to_remove:
_remove_issue_label(client, owner, repo, issue_number, local_id)
if org_ids_to_add:
client.request_json(
"POST",
repo_path(owner, repo, f"/issues/{issue_number}/labels"),
body={"labels": org_ids_to_add},
)
else:
client.request_json(
"PUT",
repo_path(owner, repo, f"/issues/{issue_number}/labels"),
body={"labels": final_label_ids},
)
result.issue_label_migrations += migrated_count
if delete_repo_labels:
for name, label in sorted(local_labels.items()):
label_id = _label_id(label)
if label_id is None:
result.errors.append(f"{name} has no label id")
continue
if apply:
try:
client.request_json("DELETE", repo_path(owner, repo, f"/labels/{label_id}"))
except GiteaError as exc:
result.errors.append(f"delete repository label {name}: {exc}")
continue
result.repo_label_deletions += 1
return result
def _selected_repositories(client: GiteaClient, owner: str, explicit: list[str], repo_filter: re.Pattern[str]) -> list[str]:
if explicit:
return sorted(set(explicit))
repos = client.paginate(org_path(owner, "/repos"), query={"type": "all"}, limit=100)
names = sorted(str(repo.get("name") or "") for repo in repos if repo.get("name"))
return [name for name in names if repo_filter.search(name)]
def _labels_by_name(labels: list[dict[str, Any]]) -> dict[str, dict[str, Any]]:
return {str(label.get("name")): label for label in labels if label.get("name") and _label_id(label) is not None}
def _label_id(label: dict[str, Any]) -> int | None:
value = label.get("id") or label.get("index")
if value is None:
return None
return int(value)
def _remove_issue_label(client: GiteaClient, owner: str, repo: str, issue_number: int, label_id: int) -> None:
try:
client.request_json("DELETE", repo_path(owner, repo, f"/issues/{issue_number}/labels/{label_id}"))
except GiteaError as exc:
if "HTTP 404" in str(exc):
return
raise
if __name__ == "__main__":
raise SystemExit(main())

View File

@@ -0,0 +1,180 @@
#!/usr/bin/env python3
"""Synchronize issue labels to a Gitea repository."""
from __future__ import annotations
import argparse
import os
import pathlib
import sys
from typing import Any
from gitea_common import (
GiteaClient,
GiteaError,
infer_target,
load_dotenv,
load_json,
org_path,
repo_path,
repo_root,
require_token,
)
DEFAULT_LABELS_FILE = pathlib.Path(__file__).resolve().parents[2] / "docs" / "gitea-labels.json"
def main() -> int:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--root", type=pathlib.Path, default=pathlib.Path.cwd(), help="repository root or child path")
parser.add_argument("--remote", default="origin", help="git remote to use for target inference")
parser.add_argument("--labels-file", type=pathlib.Path, default=DEFAULT_LABELS_FILE)
parser.add_argument("--env-file", type=pathlib.Path, help="dotenv file to read before using GITEA_* values")
parser.add_argument(
"--scope",
choices=("repository", "organization"),
default="repository",
help="sync repository labels or organization labels",
)
parser.add_argument("--org", help="organization name for --scope organization; defaults to inferred owner")
parser.add_argument("--apply", action="store_true", help="create or update labels in Gitea")
args = parser.parse_args()
try:
root = repo_root(args.root)
load_dotenv(args.env_file or root / ".env")
target = infer_target(root, args.remote)
labels = _load_labels(args.labels_file)
token = require_token() if args.apply else os.environ.get("GITEA_TOKEN")
org_name = args.org or target.owner
if args.scope == "organization":
print(f"Target: {target.base_url.rstrip('/')}/{org_name} organization labels")
else:
print(f"Target: {target.display}")
print(f"Labels file: {args.labels_file}")
if not args.apply and not token:
print("Dry run without GITEA_TOKEN: API comparison skipped.")
for label in labels:
print(f"would ensure {label['name']} ({label['color']})")
return 0
client = GiteaClient(target, token)
if args.scope == "organization":
existing = _org_labels_by_name(client, org_name)
else:
existing = _repo_labels_by_name(client, target.owner, target.repo)
creates: list[dict[str, Any]] = []
updates: list[tuple[dict[str, Any], dict[str, Any]]] = []
for label in labels:
current = existing.get(label["name"])
if current is None:
creates.append(label)
continue
patch = _diff_label(current, label)
if patch:
updates.append((current, patch))
if not creates and not updates:
print("No label changes needed.")
return 0
for label in creates:
print(f"{'create' if args.apply else 'would create'} {label['name']}")
if args.apply:
client.request_json("POST", _create_path(args.scope, org_name, target.owner, target.repo), body=label)
for current, patch in updates:
print(f"{'update' if args.apply else 'would update'} {current['name']}: {', '.join(sorted(patch))}")
if args.apply:
label_id = current.get("id") or current.get("index")
if label_id is None:
raise GiteaError(f"{current['name']} has no label id in API response")
client.request_json(
"PATCH",
_update_path(args.scope, org_name, target.owner, target.repo, int(label_id)),
body=patch,
)
return 0
except GiteaError as exc:
print(f"error: {exc}", file=sys.stderr)
return 1
def _load_labels(path: pathlib.Path) -> list[dict[str, Any]]:
payload = load_json(path)
if not isinstance(payload, list):
raise GiteaError(f"{path} must contain a JSON list")
labels: list[dict[str, Any]] = []
seen: set[str] = set()
for index, item in enumerate(payload, start=1):
if not isinstance(item, dict):
raise GiteaError(f"label #{index} must be an object")
label = {
"name": str(item.get("name", "")).strip(),
"color": str(item.get("color", "")).strip().lstrip("#").lower(),
"description": str(item.get("description", "")).strip(),
"exclusive": bool(item.get("exclusive", False)),
}
if not label["name"]:
raise GiteaError(f"label #{index} is missing name")
if label["name"] in seen:
raise GiteaError(f"duplicate label name: {label['name']}")
if not _is_hex_color(label["color"]):
raise GiteaError(f"{label['name']} has invalid color {label['color']!r}")
seen.add(label["name"])
labels.append(label)
return labels
def _repo_labels_by_name(client: GiteaClient, owner: str, repo: str) -> dict[str, dict[str, Any]]:
labels = client.paginate(repo_path(owner, repo, "/labels"))
return {str(label.get("name")): label for label in labels}
def _org_labels_by_name(client: GiteaClient, owner: str) -> dict[str, dict[str, Any]]:
labels = client.paginate(org_path(owner, "/labels"))
return {str(label.get("name")): label for label in labels}
def _create_path(scope: str, org: str, owner: str, repo: str) -> str:
if scope == "organization":
return org_path(org, "/labels")
return repo_path(owner, repo, "/labels")
def _update_path(scope: str, org: str, owner: str, repo: str, label_id: int) -> str:
if scope == "organization":
return org_path(org, f"/labels/{label_id}")
return repo_path(owner, repo, f"/labels/{label_id}")
def _diff_label(current: dict[str, Any], desired: dict[str, Any]) -> dict[str, Any]:
patch: dict[str, Any] = {}
if _normalize_color(current.get("color")) != desired["color"]:
patch["color"] = desired["color"]
if str(current.get("description") or "").strip() != desired["description"]:
patch["description"] = desired["description"]
if bool(current.get("exclusive", False)) != desired["exclusive"]:
patch["exclusive"] = desired["exclusive"]
return patch
def _normalize_color(value: Any) -> str:
return str(value or "").strip().lstrip("#").lower()
def _is_hex_color(value: str) -> bool:
if len(value) != 6:
return False
return all(char in "0123456789abcdef" for char in value)
if __name__ == "__main__":
raise SystemExit(main())

View File

@@ -0,0 +1,637 @@
#!/usr/bin/env python3
"""Mirror project documentation files into Gitea repository wikis."""
from __future__ import annotations
import argparse
import base64
import dataclasses
import hashlib
import os
import pathlib
import re
import shutil
import subprocess
import sys
import urllib.parse
from typing import Any
from gitea_common import GiteaClient, GiteaError, infer_target, load_dotenv, repo_path, require_token
GIT_ROOT = pathlib.Path("/mnt/DATA/git")
PRODUCTS_ROOT = pathlib.Path("/mnt/DATA/Nextcloud/ADD ideas UG/Products")
MANAGED_MARKER = "<!-- codex-wiki-sync:"
TEXT_SUFFIXES = {".md", ".txt", ".csv", ".toml", ".json", ".yaml", ".yml", ".rst"}
SENSITIVE_NAME_RE = re.compile(r"(password|passwd|secret|credential|api[_-]?keys?|token|private[_-]?key)", re.IGNORECASE)
EXCLUDED_PARTS = {
".git",
".gitea",
".venv",
"node_modules",
"__pycache__",
"dist",
"build",
".cache",
".module-test-build",
}
EXCLUDED_NAMES = {
"package-lock.json",
"pnpm-lock.yaml",
"yarn.lock",
"uv.lock",
}
@dataclasses.dataclass(frozen=True)
class RepoInfo:
root: pathlib.Path
owner: str
repo: str
@property
def key(self) -> str:
return self.root.name
@dataclasses.dataclass(frozen=True)
class WikiSource:
repo_key: str
path: pathlib.Path
origin: str
page_title: str
def main() -> int:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--env-file", type=pathlib.Path, default=pathlib.Path("/home/zemion/.config/gitea/gitea.env"))
parser.add_argument("--git-root", type=pathlib.Path, default=GIT_ROOT)
parser.add_argument("--products-root", type=pathlib.Path, default=PRODUCTS_ROOT)
parser.add_argument("--repo", action="append", help="limit to one or more local repository directory names")
parser.add_argument("--page", action="append", help="limit to one or more generated wiki page names")
parser.add_argument("--overwrite-unmanaged", action="store_true", help="update existing wiki pages not previously managed by this script")
parser.add_argument("--transport", choices=("git", "api"), default="git", help="sync through the wiki git repository or the Gitea REST API")
parser.add_argument("--wiki-cache-dir", type=pathlib.Path, default=pathlib.Path("/tmp/codex-gitea-wiki-sync"), help="local cache for wiki git checkouts")
parser.add_argument("--commit-message", default="Sync wiki from project files", help="commit message used by git transport")
parser.add_argument("--prune-managed", action="store_true", help="delete managed wiki pages whose source files are no longer discovered")
parser.add_argument("--apply", action="store_true")
args = parser.parse_args()
try:
load_dotenv(args.env_file)
repos = discover_hosted_repos(args.git_root)
if args.repo:
requested = set(args.repo)
repos = {key: repo for key, repo in repos.items() if key in requested}
missing = requested - set(repos)
if missing:
raise GiteaError(f"requested repos are not hosted on git.add-ideas.de locally: {', '.join(sorted(missing))}")
sources = discover_sources(repos, args.products_root)
if args.page:
requested_pages = set(args.page)
sources = [source for source in sources if source.page_title in requested_pages]
missing_pages = requested_pages - {source.page_title for source in sources}
if missing_pages:
raise GiteaError(f"requested pages were not discovered: {', '.join(sorted(missing_pages))}")
print(f"Hosted repositories: {len(repos)}")
print(f"Wiki source files: {len(sources)}")
for repo_key, count in grouped_counts(sources).items():
print(f" {repo_key}: {count}")
if not args.apply:
preview_sources(sources)
print("Dry run only. Re-run with --apply to write wiki pages.")
return 0
token = require_token() if args.transport == "api" else ""
failures = 0
for repo_key, repo_sources in group_sources(sources).items():
repo = repos[repo_key]
try:
if args.transport == "git":
sync_repo_wiki_git(
repo,
repo_sources,
cache_dir=args.wiki_cache_dir,
overwrite_unmanaged=args.overwrite_unmanaged,
commit_message=args.commit_message,
write_index=not bool(args.page),
prune_managed=args.prune_managed and not bool(args.page),
)
else:
sync_repo_wiki(
repo,
repo_sources,
token,
overwrite_unmanaged=args.overwrite_unmanaged,
write_index=not bool(args.page),
prune_managed=args.prune_managed and not bool(args.page),
)
except GiteaError as exc:
failures += 1
print(f"error syncing wiki for {repo_key}: {exc}", file=sys.stderr)
return 1 if failures else 0
except GiteaError as exc:
print(f"error: {exc}", file=sys.stderr)
return 1
def discover_hosted_repos(git_root: pathlib.Path) -> dict[str, RepoInfo]:
repos: dict[str, RepoInfo] = {}
for gitdir in sorted(git_root.glob("*/.git")):
root = gitdir.parent
result = subprocess.run(
["git", "-C", str(root), "remote", "get-url", "origin"],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.DEVNULL,
text=True,
)
if "git.add-ideas.de" not in result.stdout:
continue
target = infer_target(root)
repos[root.name] = RepoInfo(root=root, owner=target.owner, repo=target.repo)
return repos
def discover_sources(repos: dict[str, RepoInfo], products_root: pathlib.Path) -> list[WikiSource]:
sources: list[WikiSource] = []
for repo in repos.values():
for path in repo.root.rglob("*"):
if is_repo_doc(repo.root, path):
rel = path.relative_to(repo.root)
sources.append(
WikiSource(
repo_key=repo.key,
path=path,
origin="repository",
page_title=title_for_path("Repo", rel),
)
)
product_map = {
"co2api": "emission-api-lib",
"govoplan": "govoplan-core",
"meubility": "meubility-workbench",
"mousehold": "mousehold",
"prvnncr": "prvnncr-server",
}
if products_root.exists():
for product_dir in sorted(path for path in products_root.iterdir() if path.is_dir()):
repo_key = product_map.get(product_dir.name)
if not repo_key or repo_key not in repos:
continue
for path in product_dir.rglob("*"):
if is_product_doc(path):
rel = path.relative_to(product_dir)
sources.append(
WikiSource(
repo_key=repo_key,
path=path,
origin=f"product:{product_dir.name}",
page_title=title_for_path(f"Product {product_dir.name}", rel),
)
)
unique: dict[tuple[str, str], WikiSource] = {}
for source in sources:
unique[(source.repo_key, source.page_title)] = source
return sorted(unique.values(), key=lambda item: (item.repo_key, item.page_title))
def is_repo_doc(repo_root_path: pathlib.Path, path: pathlib.Path) -> bool:
if not path.is_file() or not is_text_file(path):
return False
parts = set(path.relative_to(repo_root_path).parts)
if parts & EXCLUDED_PARTS:
return False
if path.name in EXCLUDED_NAMES:
return False
if SENSITIVE_NAME_RE.search(path.name):
return False
rel = path.relative_to(repo_root_path)
if len(rel.parts) == 1 and path.name.lower().startswith(("readme", "license", "changelog", "contributing", "security")):
return True
if "generated" in rel.parts:
return False
if rel.parts[0] in {"docs", "doc", "codex"} and path.suffix.lower() in {".md", ".txt", ".csv"}:
return True
if re.search(r"(backlog|todo|roadmap|plan|architecture|workflow|manifest)", path.name, re.IGNORECASE):
return True
return False
def is_product_doc(path: pathlib.Path) -> bool:
if not path.is_file() or not is_text_file(path):
return False
if path.name in EXCLUDED_NAMES:
return False
if SENSITIVE_NAME_RE.search(path.name):
return False
if any(part in {"python_backup", "bruno", "fluege"} for part in path.parts):
return False
if re.search(
r"(readme|todo|roadmap|plan|concept|continuation|copyright|notes|whitepaper|pitch|request_response)",
path.name,
re.IGNORECASE,
) and path.suffix.lower() in {".md", ".txt"}:
return True
return False
def is_text_file(path: pathlib.Path) -> bool:
if path.suffix.lower() not in TEXT_SUFFIXES:
return False
try:
chunk = path.read_bytes()[:4096]
except OSError:
return False
return b"\x00" not in chunk
def title_for_path(prefix: str, rel: pathlib.Path) -> str:
stem = rel.with_suffix("").as_posix()
stem = re.sub(r"[^A-Za-z0-9]+", "-", stem).strip("-")
return f"{prefix}-{stem}"[:180]
def group_sources(sources: list[WikiSource]) -> dict[str, list[WikiSource]]:
grouped: dict[str, list[WikiSource]] = {}
for source in sources:
grouped.setdefault(source.repo_key, []).append(source)
return dict(sorted(grouped.items()))
def grouped_counts(sources: list[WikiSource]) -> dict[str, int]:
return {repo_key: len(items) for repo_key, items in group_sources(sources).items()}
def preview_sources(sources: list[WikiSource]) -> None:
for repo_key, repo_sources in group_sources(sources).items():
print(f"{repo_key}:")
for source in repo_sources[:80]:
print(f" {source.page_title} <- {source.path}")
if len(repo_sources) > 80:
print(f" ... {len(repo_sources) - 80} more")
def sync_repo_wiki(
repo: RepoInfo,
sources: list[WikiSource],
token: str,
*,
overwrite_unmanaged: bool,
write_index: bool = True,
prune_managed: bool = False,
) -> None:
target = infer_target(repo.root)
client = GiteaClient(target, token)
existing_pages = list_existing_wiki_pages(client, target.owner, target.repo)
existing_page_names = {
str(page.get("title")): str(page.get("sub_url") or page.get("title"))
for page in existing_pages
if page.get("title")
}
index_lines = [
f"# {target.repo} Project Wiki Index",
"",
f"{MANAGED_MARKER}index -->",
"",
"This page is generated from repository and product-directory project files.",
"",
]
for source in sources:
content = render_wiki_content(source)
changed = put_wiki_page(
client,
target.owner,
target.repo,
source.page_title,
content,
existing_page_names,
overwrite_unmanaged=overwrite_unmanaged,
)
print(f"{'updated' if changed else 'unchanged'} {target.owner}/{target.repo} wiki:{source.page_title}")
index_lines.append(f"- [{source.page_title}]({source.page_title}) - `{source.path}`")
if write_index:
put_wiki_page(
client,
target.owner,
target.repo,
"Codex-Project-Index",
"\n".join(index_lines) + "\n",
existing_page_names,
overwrite_unmanaged=True,
)
else:
print(f"skipped {target.owner}/{target.repo} wiki:Codex-Project-Index during page-limited sync")
if prune_managed and write_index:
prune_managed_wiki_pages_api(client, target.owner, target.repo, existing_page_names, sources)
def sync_repo_wiki_git(
repo: RepoInfo,
sources: list[WikiSource],
*,
cache_dir: pathlib.Path,
overwrite_unmanaged: bool,
commit_message: str,
write_index: bool = True,
prune_managed: bool = False,
) -> None:
target = infer_target(repo.root)
wiki_root = prepare_wiki_checkout(repo, cache_dir=cache_dir)
index_lines = [
f"# {target.repo} Project Wiki Index",
"",
f"{MANAGED_MARKER}index -->",
"",
"This page is generated from repository and product-directory project files.",
"",
]
for source in sources:
content = render_wiki_content(source)
status = write_wiki_page_file(
wiki_root,
source.page_title,
content,
overwrite_unmanaged=overwrite_unmanaged,
)
print(f"{status} {target.owner}/{target.repo} wiki:{source.page_title}")
index_lines.append(f"- [{source.page_title}]({source.page_title}) - `{source.path}`")
if write_index:
write_wiki_page_file(
wiki_root,
"Codex-Project-Index",
"\n".join(index_lines) + "\n",
overwrite_unmanaged=True,
)
else:
print(f"skipped {target.owner}/{target.repo} wiki:Codex-Project-Index during page-limited sync")
if prune_managed and write_index:
prune_managed_wiki_pages_git(wiki_root, sources)
if not git_has_changes(wiki_root):
print(f"unchanged {target.owner}/{target.repo} wiki repository")
return
run_git(wiki_root, "add", "-A")
if not git_has_staged_changes(wiki_root):
print(f"unchanged {target.owner}/{target.repo} wiki repository")
return
ensure_git_identity(wiki_root)
run_git(wiki_root, "commit", "-m", commit_message)
run_git(wiki_root, "push")
print(f"pushed {target.owner}/{target.repo} wiki repository")
def prepare_wiki_checkout(repo: RepoInfo, *, cache_dir: pathlib.Path) -> pathlib.Path:
remote = wiki_remote_for_repo(repo.root)
cache_dir.mkdir(parents=True, exist_ok=True)
checkout = cache_dir / f"{repo.owner}-{repo.repo}.wiki"
if (checkout / ".git").exists():
run_git(checkout, "remote", "set-url", "origin", remote)
run_git(checkout, "fetch", "origin")
branch = current_branch(checkout)
if branch:
run_git(checkout, "reset", "--hard", f"origin/{branch}")
else:
run_git(checkout, "reset", "--hard")
run_git(checkout, "clean", "-fd")
return checkout
if checkout.exists():
shutil.rmtree(checkout)
result = subprocess.run(
["git", "clone", remote, str(checkout)],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode == 0:
return checkout
checkout.mkdir(parents=True, exist_ok=True)
run_git(checkout, "init")
run_git(checkout, "remote", "add", "origin", remote)
run_git(checkout, "checkout", "-b", "master")
return checkout
def wiki_remote_for_repo(repo_root_path: pathlib.Path) -> str:
result = subprocess.run(
["git", "-C", str(repo_root_path), "remote", "get-url", "origin"],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode != 0 or not result.stdout.strip():
raise GiteaError(f"Could not read origin remote for {repo_root_path}")
remote = result.stdout.strip()
if remote.endswith(".git"):
return f"{remote[:-4]}.wiki.git"
return f"{remote}.wiki.git"
def write_wiki_page_file(wiki_root: pathlib.Path, title: str, content: str, *, overwrite_unmanaged: bool) -> str:
path = wiki_root / f"{wiki_file_stem(title)}.md"
if path.exists():
current = read_text(path)
if current == content:
return "unchanged"
if MANAGED_MARKER not in current and not overwrite_unmanaged:
return "skipped unmanaged"
path.write_text(content, encoding="utf-8")
return "updated"
def prune_managed_wiki_pages_git(wiki_root: pathlib.Path, sources: list[WikiSource]) -> None:
desired = {f"{wiki_file_stem(source.page_title)}.md" for source in sources}
desired.add(f"{wiki_file_stem('Codex-Project-Index')}.md")
for path in sorted(wiki_root.glob("*.md")):
if path.name in desired:
continue
current = read_text(path)
if MANAGED_MARKER not in current:
continue
path.unlink()
print(f"deleted stale managed wiki page {path.name}")
def prune_managed_wiki_pages_api(
client: GiteaClient,
owner: str,
repo: str,
existing_page_names: dict[str, str],
sources: list[WikiSource],
) -> None:
desired = {source.page_title for source in sources}
desired.add("Codex-Project-Index")
for title, page_name in sorted(existing_page_names.items()):
if title in desired:
continue
current = client.request_json("GET", repo_path(owner, repo, f"/wiki/page/{quote_wiki_page_name(page_name)}"))
if MANAGED_MARKER not in decode_wiki_content(current):
continue
client.request_json("DELETE", repo_path(owner, repo, f"/wiki/page/{quote_wiki_page_name(page_name)}"))
print(f"deleted stale managed wiki page {owner}/{repo}:{title}")
def wiki_file_stem(title: str) -> str:
return re.sub(r"[/\\]+", "-", title).strip() or "Home"
def git_has_changes(root: pathlib.Path) -> bool:
result = subprocess.run(
["git", "-C", str(root), "status", "--porcelain"],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode != 0:
raise GiteaError(f"git status failed in {root}: {result.stderr.strip()}")
return bool(result.stdout.strip())
def git_has_staged_changes(root: pathlib.Path) -> bool:
result = subprocess.run(
["git", "-C", str(root), "diff", "--cached", "--quiet"],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode == 0:
return False
if result.returncode == 1:
return True
raise GiteaError(f"git diff --cached failed in {root}: {result.stderr.strip()}")
def current_branch(root: pathlib.Path) -> str:
result = subprocess.run(
["git", "-C", str(root), "rev-parse", "--abbrev-ref", "HEAD"],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
branch = result.stdout.strip()
if result.returncode != 0 or branch == "HEAD":
return ""
return branch
def ensure_git_identity(root: pathlib.Path) -> None:
if not git_config_value(root, "user.email"):
run_git(root, "config", "user.email", "codex@govoplan.local")
if not git_config_value(root, "user.name"):
run_git(root, "config", "user.name", "Codex")
def git_config_value(root: pathlib.Path, key: str) -> str:
result = subprocess.run(
["git", "-C", str(root), "config", "--get", key],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
return result.stdout.strip() if result.returncode == 0 else ""
def run_git(root: pathlib.Path, *args: str) -> None:
result = subprocess.run(
["git", "-C", str(root), *args],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode != 0:
command = "git -C " + str(root) + " " + " ".join(args)
raise GiteaError(f"{command} failed: {result.stderr.strip() or result.stdout.strip()}")
def list_existing_wiki_pages(client: GiteaClient, owner: str, repo: str) -> list[dict[str, Any]]:
try:
return client.paginate(repo_path(owner, repo, "/wiki/pages"), limit=50)
except GiteaError as exc:
if "HTTP 404" in str(exc) and "/wiki/pages" in str(exc):
return []
raise
def render_wiki_content(source: WikiSource) -> str:
raw = read_text(source.path)
fingerprint = hashlib.sha256(raw.encode("utf-8")).hexdigest()[:24]
header = [
f"{MANAGED_MARKER}{fingerprint} -->",
"",
f"> Mirrored from `{source.path}`.",
f"> Origin: `{source.origin}`.",
"> Active tasks and changing state belong in Gitea issues; this wiki page is durable project context.",
"",
"---",
"",
]
if source.path.suffix.lower() == ".csv":
return "\n".join(header + ["```csv", raw.rstrip(), "```", ""])
return "\n".join(header) + raw.rstrip() + "\n"
def put_wiki_page(
client: GiteaClient,
owner: str,
repo: str,
title: str,
content: str,
existing_page_names: dict[str, str],
*,
overwrite_unmanaged: bool,
) -> bool:
body = {
"title": title,
"content_base64": base64.b64encode(content.encode("utf-8")).decode("ascii"),
"message": f"Sync {title} from project files",
}
if title in existing_page_names:
page_name = existing_page_names[title]
current = client.request_json("GET", repo_path(owner, repo, f"/wiki/page/{quote_wiki_page_name(page_name)}"))
current_content = decode_wiki_content(current)
if current_content == content:
return False
if MANAGED_MARKER not in current_content and not overwrite_unmanaged:
print(f"skip unmanaged existing wiki page {owner}/{repo}:{title}")
return False
client.request_json("PATCH", repo_path(owner, repo, f"/wiki/page/{quote_wiki_page_name(page_name)}"), body=body)
return True
client.request_json("POST", repo_path(owner, repo, "/wiki/new"), body=body)
existing_page_names[title] = title
return True
def quote_wiki_page_name(value: str) -> str:
return urllib.parse.quote(value, safe="+")
def decode_wiki_content(page: dict[str, Any]) -> str:
encoded = str(page.get("content_base64") or "")
if not encoded:
return ""
return base64.b64decode(encoded).decode("utf-8")
def read_text(path: pathlib.Path) -> str:
try:
return path.read_text(encoding="utf-8")
except UnicodeDecodeError:
return path.read_text(encoding="latin-1")
if __name__ == "__main__":
raise SystemExit(main())

View File

@@ -0,0 +1,377 @@
#!/usr/bin/env python3
"""Preview or import inline TODO-style markers as Gitea issues."""
from __future__ import annotations
import argparse
import hashlib
import json
import os
import pathlib
import re
import subprocess
import sys
from typing import Any
from gitea_common import GiteaClient, GiteaError, infer_target, label_ids_by_name, load_dotenv, repo_path, repo_root, require_token
MARKER_RE = re.compile(
r"(?P<prefix>#|//|/\*|\*|--|<!--)?\s*"
r"\b(?P<marker>TODO|FIXME|XXX|HACK)\b"
r"\s*(?:\((?P<context>[^)]{1,120})\))?"
r"\s*(?P<colon>:)?\s*(?P<text>.*)",
re.IGNORECASE,
)
DEFAULT_EXCLUDES = (
"!**/.git/**",
"!**/.venv/**",
"!**/node_modules/**",
"!**/__pycache__/**",
"!**/.module-test-build/**",
"!**/dist/**",
"!**/build/**",
"!**/.cache/**",
"!.gitea/**",
"!docs/GITEA_ISSUES.md",
"!tools/gitea/gitea-todo-import.py",
"!tools/gitea/gitea-sync-labels.py",
"!tools/gitea/gitea-codex-note.py",
"!tools/gitea/gitea_common.py",
)
def main() -> int:
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--root", type=pathlib.Path, default=pathlib.Path.cwd(), help="repository root or child path")
parser.add_argument("--remote", default="origin", help="git remote to use for target inference")
parser.add_argument("--env-file", type=pathlib.Path, help="dotenv file to read before using GITEA_* values")
parser.add_argument("--apply", action="store_true", help="create missing Gitea issues")
parser.add_argument("--include-linked", action="store_true", help="include markers that already reference an issue")
parser.add_argument(
"--module-label",
help="project/module label to apply; defaults to a known GovOPlaN mapping or module/core",
)
parser.add_argument("--no-area-labels", action="store_true", help="do not infer area/* labels from paths")
parser.add_argument("--extra-label", action="append", default=[], help="additional label name to apply")
parser.add_argument("--limit", type=int, default=0, help="maximum number of markers to process")
parser.add_argument("--json", action="store_true", help="print issue previews as JSON")
args = parser.parse_args()
try:
root = repo_root(args.root)
load_dotenv(args.env_file or root / ".env")
target = infer_target(root, args.remote)
token = require_token() if args.apply else os.environ.get("GITEA_TOKEN")
markers = scan_markers(root, include_linked=args.include_linked)
if args.limit > 0:
markers = markers[: args.limit]
module_label = args.module_label or infer_module_label(root)
previews = [
build_issue_preview(
target.owner,
target.repo,
marker,
module_label=module_label,
infer_areas=not args.no_area_labels,
extra_labels=args.extra_label,
)
for marker in markers
]
if args.json:
print(json.dumps([preview.as_dict() for preview in previews], indent=2, sort_keys=True))
else:
print(f"Target: {target.display}")
print(f"Scanned root: {root}")
print(f"Found {len(previews)} importable marker(s).")
for preview in previews:
print(f"- {preview.title}")
print(f" {preview.location}")
print(f" labels: {', '.join(preview.labels)}")
if not args.apply:
if not previews:
return 0
print("Dry run only. Re-run with --apply and GITEA_TOKEN to create issues.")
return 0
if not previews:
print("No issues to create.")
return 0
client = GiteaClient(target, token)
label_ids = load_label_ids(client, target.owner, target.repo)
missing_labels = sorted({label for preview in previews for label in preview.labels if label not in label_ids})
if missing_labels:
joined = ", ".join(missing_labels)
raise GiteaError(f"missing labels: {joined}. Run tools/gitea/gitea-sync-labels.py --apply first.")
existing_fingerprints = load_existing_fingerprints(client, target.owner, target.repo)
created = 0
skipped = 0
for preview in previews:
if preview.fingerprint in existing_fingerprints:
print(f"skip existing {preview.location}")
skipped += 1
continue
issue = client.request_json(
"POST",
repo_path(target.owner, target.repo, "/issues"),
body={
"title": preview.title,
"body": preview.body,
"labels": [label_ids[label] for label in preview.labels],
},
)
print(f"created #{issue.get('number') or issue.get('index')}: {preview.title}")
created += 1
print(f"Created {created} issue(s), skipped {skipped} existing issue(s).")
return 0
except GiteaError as exc:
print(f"error: {exc}", file=sys.stderr)
return 1
class Marker:
def __init__(self, path: pathlib.Path, line: int, column: int, marker: str, context: str, text: str, raw: str) -> None:
self.path = path
self.line = line
self.column = column
self.marker = marker.upper()
self.context = context
self.text = text
self.raw = raw.rstrip("\n")
@property
def location(self) -> str:
return f"{self.path}:{self.line}"
class IssuePreview:
def __init__(self, title: str, body: str, labels: list[str], fingerprint: str, location: str) -> None:
self.title = title
self.body = body
self.labels = labels
self.fingerprint = fingerprint
self.location = location
def as_dict(self) -> dict[str, Any]:
return {
"title": self.title,
"body": self.body,
"labels": self.labels,
"fingerprint": self.fingerprint,
"location": self.location,
}
def scan_markers(root: pathlib.Path, *, include_linked: bool) -> list[Marker]:
command = [
"rg",
"--json",
"--hidden",
"--line-number",
"--column",
"-e",
r"\b(TODO|FIXME|XXX|HACK)\b",
]
for pattern in DEFAULT_EXCLUDES:
command.extend(["--glob", pattern])
command.append(str(root))
result = subprocess.run(
command,
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode not in {0, 1}:
raise GiteaError(f"rg failed: {result.stderr.strip()}")
markers: list[Marker] = []
for line in result.stdout.splitlines():
event = json.loads(line)
if event.get("type") != "match":
continue
data = event["data"]
raw_line = data["lines"]["text"]
match = MARKER_RE.search(raw_line)
if not match:
continue
if not _looks_like_marker(match):
continue
context = (match.group("context") or "").strip()
if not include_linked and _already_linked(context, raw_line):
continue
markers.append(
Marker(
path=pathlib.Path(data["path"]["text"]).resolve().relative_to(root),
line=int(data["line_number"]),
column=int(data.get("submatches", [{}])[0].get("start", 0)) + 1,
marker=match.group("marker"),
context=context,
text=clean_marker_text(match.group("text") or raw_line),
raw=raw_line,
)
)
return markers
def build_issue_preview(
owner: str,
repo: str,
marker: Marker,
*,
module_label: str,
infer_areas: bool,
extra_labels: list[str],
) -> IssuePreview:
fingerprint = marker_fingerprint(owner, repo, marker)
summary = marker.text or f"review {marker.location}"
title = truncate_title(f"{marker.marker}: {summary} ({marker.location})")
labels = sorted(set(default_labels(marker, module_label=module_label, infer_areas=infer_areas) + list(extra_labels)))
body = "\n".join(
[
f"<!-- codex-todo-fingerprint:{fingerprint} -->",
"",
"Imported from an inline marker.",
"",
f"- Repository: `{owner}/{repo}`",
f"- Location: `{marker.location}`",
f"- Marker: `{marker.marker}`",
f"- Context: `{marker.context or 'none'}`",
"",
"Current line:",
"",
"```text",
marker.raw,
"```",
"",
"Migration rule:",
"",
"- Keep this Gitea issue as the canonical backlog item.",
"- When touching this code, remove the inline marker or replace it with a short issue reference.",
]
)
return IssuePreview(title=title, body=body, labels=labels, fingerprint=fingerprint, location=marker.location)
def clean_marker_text(text: str) -> str:
cleaned = text.strip()
cleaned = re.sub(r"\s*(?:#|//|/\*|\*|--)\s*$", "", cleaned).strip()
cleaned = re.sub(r"\s*\*/\s*$", "", cleaned).strip()
return cleaned
def default_labels(marker: Marker, *, module_label: str, infer_areas: bool) -> list[str]:
labels = ["source/todo-scan", "status/triage", module_label, marker_type_label(marker.marker)]
area = infer_area_label(marker.path) if infer_areas else ""
if area:
labels.append(area)
return labels
def marker_type_label(marker: str) -> str:
if marker == "FIXME":
return "type/bug"
if marker in {"HACK", "XXX"}:
return "type/debt"
return "type/task"
def infer_area_label(path: pathlib.Path) -> str:
text = path.as_posix()
if text.startswith("webui/"):
return "area/webui"
if text.startswith("alembic/"):
return "area/migrations"
if text.startswith("docs/"):
return "area/docs"
if text.startswith("scripts/") or text.startswith("tools/"):
return "area/devex"
if "/rbac" in text or "permission" in text:
return "area/rbac"
if "/access" in text or "auth" in text or "session" in text:
return "area/auth"
if "tenant" in text:
return "area/tenancy"
if "governance" in text or "privacy" in text or "audit" in text or "retention" in text:
return "area/governance"
if "module" in text:
return "area/module-system"
if "migration" in text:
return "area/migrations"
if "router" in text or "api" in text:
return "area/api"
if "db" in text or "model" in text or "persistence" in text:
return "area/db"
return ""
def infer_module_label(root: pathlib.Path) -> str:
known = {
"govoplan-core": "module/core",
"govoplan-access": "module/access",
"govoplan-admin": "module/admin",
"govoplan-tenancy": "module/tenancy",
"govoplan-policy": "module/policy",
"govoplan-audit": "module/audit",
"govoplan-mail": "module/mail",
"govoplan-files": "module/files",
"govoplan-campaign": "module/campaign",
"addideas-govoplan-website": "area/marketing",
}
if root.name in known:
return known[root.name]
if root.name.startswith("govoplan-"):
suffix = root.name.removeprefix("govoplan-")
if suffix:
return f"module/{suffix}"
return "module/core"
def marker_fingerprint(owner: str, repo: str, marker: Marker) -> str:
stable = "\n".join([owner, repo, marker.path.as_posix(), marker.marker, marker.context, marker.text])
return hashlib.sha256(stable.encode("utf-8")).hexdigest()[:24]
def truncate_title(title: str, limit: int = 180) -> str:
if len(title) <= limit:
return title
return f"{title[: limit - 1].rstrip()}..."
def load_label_ids(client: GiteaClient, owner: str, repo: str) -> dict[str, int]:
return label_ids_by_name(client, owner, repo)
def load_existing_fingerprints(client: GiteaClient, owner: str, repo: str) -> set[str]:
issues = client.paginate(
repo_path(owner, repo, "/issues"),
query={"state": "all", "labels": "source/todo-scan"},
)
fingerprints: set[str] = set()
for issue in issues:
body = str(issue.get("body") or "")
for match in re.finditer(r"(?:codex|govoplan)-todo-fingerprint:([0-9a-f]{24})", body):
fingerprints.add(match.group(1))
return fingerprints
def _already_linked(context: str, raw_line: str) -> bool:
linked_text = f"{context} {raw_line}".lower()
return "gitea#" in linked_text or "issue#" in linked_text or re.search(r"#\d+", linked_text) is not None
def _looks_like_marker(match: re.Match[str]) -> bool:
text = (match.group("text") or "").strip()
return bool(match.group("context") or match.group("colon") or (match.group("prefix") and text))
if __name__ == "__main__":
raise SystemExit(main())

290
tools/gitea/gitea_common.py Normal file
View File

@@ -0,0 +1,290 @@
#!/usr/bin/env python3
"""Shared helpers for Gitea maintenance scripts."""
from __future__ import annotations
import dataclasses
import json
import os
import pathlib
import re
import subprocess
import urllib.error
import urllib.parse
import urllib.request
from typing import Any
class GiteaError(RuntimeError):
"""Raised when a Gitea API request fails."""
@dataclasses.dataclass(frozen=True)
class RepoTarget:
base_url: str
owner: str
repo: str
@property
def display(self) -> str:
return f"{self.base_url.rstrip('/')}/{self.owner}/{self.repo}"
@property
def api_base(self) -> str:
return f"{self.base_url.rstrip('/')}/api/v1"
def repo_root(start: pathlib.Path) -> pathlib.Path:
result = subprocess.run(
["git", "-C", str(start), "rev-parse", "--show-toplevel"],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode != 0:
return start.resolve()
return pathlib.Path(result.stdout.strip()).resolve()
def infer_target(root: pathlib.Path, remote_name: str = "origin") -> RepoTarget:
env_url = os.environ.get("GITEA_URL")
env_owner = os.environ.get("GITEA_OWNER")
env_repo = os.environ.get("GITEA_REPO")
remote_url = _git_remote(root, remote_name)
inferred_url, inferred_owner, inferred_repo = _parse_remote(remote_url)
base_url = (env_url or inferred_url).rstrip("/")
owner = env_owner or inferred_owner
repo = env_repo or inferred_repo
missing = [
name
for name, value in (
("GITEA_URL", base_url),
("GITEA_OWNER", owner),
("GITEA_REPO", repo),
)
if not value
]
if missing:
joined = ", ".join(missing)
raise GiteaError(
f"Could not infer Gitea target. Set {joined}, or configure git remote {remote_name!r}."
)
return RepoTarget(base_url=base_url, owner=owner, repo=_strip_git_suffix(repo))
def quote_path(value: str) -> str:
return urllib.parse.quote(value, safe="")
class GiteaClient:
def __init__(self, target: RepoTarget, token: str | None) -> None:
self.target = target
self.token = token
def request_json(
self,
method: str,
path: str,
*,
body: dict[str, Any] | None = None,
query: dict[str, Any] | None = None,
) -> Any:
url = f"{self.target.api_base}{path}"
if query:
url = f"{url}?{urllib.parse.urlencode(query, doseq=True)}"
data = None
headers = {
"Accept": "application/json",
"User-Agent": "codex-gitea-maintenance",
}
if self.token:
headers["Authorization"] = f"token {self.token}"
if body is not None:
data = json.dumps(body).encode("utf-8")
headers["Content-Type"] = "application/json"
request = urllib.request.Request(url, data=data, headers=headers, method=method)
try:
with urllib.request.urlopen(request, timeout=30) as response:
payload = response.read().decode("utf-8")
if not payload:
return None
return json.loads(payload)
except urllib.error.HTTPError as exc:
message = exc.read().decode("utf-8", errors="replace")
raise GiteaError(f"{method} {url} failed with HTTP {exc.code}: {message}") from exc
except urllib.error.URLError as exc:
raise GiteaError(f"{method} {url} failed: {exc.reason}") from exc
def paginate(
self,
path: str,
*,
query: dict[str, Any] | None = None,
limit: int = 100,
) -> list[dict[str, Any]]:
items: list[dict[str, Any]] = []
page = 1
effective_limit = min(limit, 50)
seen_pages: set[tuple[Any, ...]] = set()
while True:
page_query = dict(query or {})
page_query.update({"page": page, "limit": effective_limit})
payload = self.request_json("GET", path, query=page_query)
if not isinstance(payload, list):
raise GiteaError(f"Expected a list from {path}, got {type(payload).__name__}")
if not payload:
return items
signature = tuple(
item.get("id") or item.get("number") or item.get("index") or item.get("name")
for item in payload
if isinstance(item, dict)
)
if signature in seen_pages:
return items
seen_pages.add(signature)
items.extend(payload)
if len(payload) < effective_limit:
return items
page += 1
def load_json(path: pathlib.Path) -> Any:
with path.open("r", encoding="utf-8") as handle:
return json.load(handle)
def load_dotenv(path: pathlib.Path | None) -> list[str]:
if path is None or not path.exists():
return []
loaded: list[str] = []
for line_number, line in enumerate(path.read_text(encoding="utf-8").splitlines(), start=1):
stripped = line.strip()
if not stripped or stripped.startswith("#"):
continue
if stripped.startswith("export "):
stripped = stripped[len("export ") :].strip()
if "=" not in stripped:
raise GiteaError(f"{path}:{line_number}: expected KEY=value")
key, value = stripped.split("=", 1)
key = key.strip()
value = _strip_env_value(value.strip())
if not re.match(r"^[A-Za-z_][A-Za-z0-9_]*$", key):
raise GiteaError(f"{path}:{line_number}: invalid environment key {key!r}")
if key not in os.environ:
os.environ[key] = value
loaded.append(key)
return loaded
def require_token() -> str:
token = os.environ.get("GITEA_TOKEN")
if not token:
raise GiteaError(
"GITEA_TOKEN is required when using --apply. "
"Set it in the environment, the target .env, or --env-file."
)
return token
def repo_path(owner: str, repo: str, suffix: str) -> str:
return f"/repos/{quote_path(owner)}/{quote_path(repo)}{suffix}"
def org_path(owner: str, suffix: str) -> str:
return f"/orgs/{quote_path(owner)}{suffix}"
def labels_by_name(client: GiteaClient, owner: str, repo: str | None = None) -> dict[str, dict[str, Any]]:
"""Return org labels plus optional repository labels, keyed by label name.
Gitea stores organization labels separately from repository labels. Issue
APIs can use label ids from both scopes on supported instances, so issue
creation helpers should resolve both. Repository labels intentionally win
when a repo carries a local label with the same name.
"""
labels: dict[str, dict[str, Any]] = {}
try:
for label in client.paginate(org_path(owner, "/labels")):
name = str(label.get("name") or "")
if name:
labels[name] = label
except GiteaError:
# User-owned repositories or older instances may not expose org labels.
pass
if repo:
for label in client.paginate(repo_path(owner, repo, "/labels")):
name = str(label.get("name") or "")
if name:
labels[name] = label
return labels
def label_ids_by_name(client: GiteaClient, owner: str, repo: str | None = None) -> dict[str, int]:
ids: dict[str, int] = {}
for name, label in labels_by_name(client, owner, repo).items():
label_id = label.get("id") or label.get("index")
if label_id is None:
continue
ids[name] = int(label_id)
return ids
def _git_remote(root: pathlib.Path, remote_name: str) -> str:
result = subprocess.run(
["git", "-C", str(root), "remote", "get-url", remote_name],
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode != 0:
return ""
return result.stdout.strip()
def _parse_remote(remote_url: str) -> tuple[str, str, str]:
if not remote_url:
return "", "", ""
parsed = urllib.parse.urlparse(remote_url)
if parsed.scheme in {"http", "https", "ssh"} and parsed.netloc:
path_parts = [part for part in parsed.path.strip("/").split("/") if part]
owner, repo = _owner_repo_from_parts(path_parts)
if parsed.scheme in {"http", "https"}:
prefix = "/".join(path_parts[:-2])
base_path = f"/{prefix}" if prefix else ""
return f"{parsed.scheme}://{parsed.netloc}{base_path}", owner, repo
return f"https://{parsed.hostname or parsed.netloc}", owner, repo
scp_like = re.match(r"^(?:[^@]+@)?(?P<host>[^:]+):(?P<path>.+)$", remote_url)
if scp_like:
path_parts = [part for part in scp_like.group("path").strip("/").split("/") if part]
owner, repo = _owner_repo_from_parts(path_parts)
return f"https://{scp_like.group('host')}", owner, repo
return "", "", ""
def _owner_repo_from_parts(path_parts: list[str]) -> tuple[str, str]:
if len(path_parts) < 2:
return "", ""
return path_parts[-2], _strip_git_suffix(path_parts[-1])
def _strip_git_suffix(value: str) -> str:
return value[:-4] if value.endswith(".git") else value
def _strip_env_value(value: str) -> str:
if len(value) >= 2 and value[0] == value[-1] and value[0] in {"'", '"'}:
return value[1:-1]
return value