Files
libretro/scripts/artifacts.py
T
Abdessamad Derraz c31ce0b693 refactor: split the pack builder into its layers
generate_pack.py held six responsibilities in 4744 lines. Five move out
in dependency order, so nothing above reaches back down: destinations,
core extras, resolution with its storage tiers, the notes shipped inside
a pack, and the pack verifier. generate_pack.py keeps the build and the
command line, and re-exports the rest.

Two things the move surfaced. The offline switch was a module global the
command line assigned, which a re-export would have copied and frozen at
False; it is set through a call now and not re-exported. And a facade
placed after the entry point binds too late: importing the module worked,
running it did not, so the manifest run died on a name the tests never
exercised because tests import.

Verified against the previous revision on frozen inputs: every generated
artefact is identical, save the catalogue that embeds hashes of files
carrying a build timestamp.
2026-08-12 12:20:17 +02:00

103 lines
3.6 KiB
Python

"""Writing generated files, and not writing them.
A generated artefact carries a timestamp but must not be rewritten
when only the clock moved: the CI freshness guard is a git diff, and
it is only meaningful if the comparison ignores the hour."""
from __future__ import annotations
import contextlib
import os
import re
_TIMESTAMP_PATTERNS = [
re.compile(r'"generated_at":\s*"[^"]*"'), # database.json
re.compile(r'"imported_at":\s*"[^"]*"'), # provenance snapshots
re.compile(r"\*Auto-generated on [^*]*\*"), # README.md
re.compile(r"\*Generated on [^*]*\*"), # docs site pages
# The decorated pages carry the same stamp again as a rendered element,
# and missing it rewrote every page on every run for the clock alone.
re.compile(r'<div class="rb-timestamp">[^<]*</div>'),
]
def write_if_changed(path: str, content: str, normalize=None) -> bool:
"""Write content to path only if the non-timestamp content differs.
Compares new and existing content after stripping timestamp lines.
Returns True if the file was written, False if skipped (unchanged).
A caller that writes a file in two passes -a body, then the same body
wrapped in front matter -passes ``normalize`` to reduce both sides to the
part it owns. Without it the second pass always sees a difference, the
file is rewritten, and the fresh timestamp defeats the comparison.
"""
if os.path.exists(path):
with open(path) as f:
existing = f.read()
before, after = (
(normalize(existing), normalize(content))
if normalize
else (existing, content)
)
if _strip_timestamps(before) == _strip_timestamps(after):
return False
with open(path, "w") as f:
f.write(content)
return True
def _strip_timestamps(text: str) -> str:
"""Remove known timestamp patterns for content comparison."""
result = text
for pattern in _TIMESTAMP_PATTERNS:
result = pattern.sub("", result)
return result
class ArtifactLockBusy(RuntimeError):
"""Raised when another process already holds the artifact directory."""
@contextlib.contextmanager
def artifact_lock(directory: str, exclusive: bool = True):
"""Serialize access to a shared artifact directory across processes.
Two pipeline runs building the same dist/ leave readers looking at
half-written ZIPs, which surfaces as BadZipFile far from its cause.
Writers take the lock exclusively, readers share it. On platforms
without flock the lock is a no-op.
"""
try:
import fcntl
except ImportError:
yield
return
os.makedirs(directory, exist_ok=True)
lock_path = os.path.join(directory, ".lock")
mode = fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH
with open(lock_path, "w") as handle:
try:
fcntl.flock(handle, mode | fcntl.LOCK_NB)
except OSError as exc:
raise ArtifactLockBusy(
f"{directory} is in use by another run "
f"(lock: {lock_path}). Wait for it to finish."
) from exc
try:
yield
finally:
fcntl.flock(handle, fcntl.LOCK_UN)
def _build_timestamp(db: dict | None = None) -> str:
"""Timestamp for generated artifacts.
Reads the database snapshot the artifact was built from, so rebuilding
the same data twice yields the same value. Falls back to the clock only
when no database is at hand.
"""
stamp = (db or {}).get("generated_at")
if isinstance(stamp, str) and stamp:
return stamp
from datetime import datetime, timezone
return datetime.now(timezone.utc).strftime("%Y-%m-%dT%H:%M:%SZ")