Files
ProjectStructure_HOME/zip_sequences.py

576 lines
20 KiB
Python
Raw Normal View History

2025-11-06 10:10:29 -07:00
#!/usr/bin/env python3
"""Maintain zipped render sequences for Git hooks.
Default mode scans `Renders/`, produces ZIP archives under `Renders/_zipped/`,
and stages any updated archives so commits only track compact files. Switch to
`--mode expand` to inflate the tracked archives back into the ignored working
directories after checkouts or pulls.
"""
from __future__ import annotations
import argparse
import json
import os
import shutil
2025-11-10 10:36:28 -07:00
import subprocess
2025-11-06 10:10:29 -07:00
import sys
2025-11-10 10:36:28 -07:00
import tempfile
import time
2025-11-06 10:10:29 -07:00
from concurrent.futures import ThreadPoolExecutor, as_completed
from pathlib import Path
from typing import Iterator, Sequence
RENDER_ROOT = Path("Renders")
ARCHIVE_ROOT = RENDER_ROOT / "_zipped"
SEQUENCE_EXTENSIONS = {
".png",
".jpg",
".jpeg",
".tif",
".tiff",
".exr",
}
STATE_SUFFIX = ".meta.json"
2025-11-08 02:36:20 -07:00
DEFAULT_CONFIG = {
2025-11-10 10:24:55 -07:00
"zipper": "7z",
2025-11-08 02:36:20 -07:00
"compression": 9,
2025-11-10 10:24:55 -07:00
"dailyFormat": "daily_YYMMDD",
2025-11-08 02:36:20 -07:00
}
def load_config() -> dict:
2025-11-10 11:27:44 -07:00
# First try to load from project's .config folder (current working directory)
# Then fall back to ProjectStructure repo config (next to zip_sequences.py)
cwd = Path.cwd()
project_config = cwd / ".config" / "config.json"
repo_config = Path(__file__).resolve().with_name("config.json")
config_paths = [
("project", project_config),
("repo", repo_config),
]
for source, config_path in config_paths:
try:
if config_path.exists():
text = config_path.read_text(encoding="utf-8")
try:
data = json.loads(text)
if isinstance(data, dict):
merged = DEFAULT_CONFIG.copy()
merged.update(data)
return merged
except json.JSONDecodeError:
continue
except OSError:
continue
# If no config found, return defaults
return DEFAULT_CONFIG.copy()
2025-11-08 02:36:20 -07:00
CONFIG = load_config()
2025-11-10 10:24:55 -07:00
zipper_val = CONFIG.get("zipper", "7z")
# Handle both old boolean format and new string format
if isinstance(zipper_val, bool):
ZIPPER_TYPE = "7z" if zipper_val else "zip"
else:
ZIPPER_TYPE = str(zipper_val).lower()
2025-11-10 11:27:44 -07:00
2025-11-08 02:36:20 -07:00
COMPRESSION_LEVEL = CONFIG.get("compression", 9)
if isinstance(COMPRESSION_LEVEL, str):
try:
COMPRESSION_LEVEL = int(COMPRESSION_LEVEL)
except ValueError:
COMPRESSION_LEVEL = 9
if not isinstance(COMPRESSION_LEVEL, int):
COMPRESSION_LEVEL = 9
COMPRESSION_LEVEL = max(0, min(9, COMPRESSION_LEVEL))
SEVEN_Z_EXE: str | None = None
2025-11-10 10:36:28 -07:00
if ZIPPER_TYPE == "7z":
2025-11-08 02:36:20 -07:00
SEVEN_Z_EXE = shutil.which("7z") or shutil.which("7za")
2025-11-06 10:10:29 -07:00
def parse_args() -> argparse.Namespace:
parser = argparse.ArgumentParser(description="Sync render sequences with zipped archives.")
parser.add_argument(
"--mode",
choices=("zip", "expand"),
default="zip",
help="zip sequences for commit (default) or expand tracked archives",
)
parser.add_argument("--jobs", type=int, help="max parallel workers")
parser.add_argument("--verbose", action="store_true", help="print extra progress details")
return parser.parse_args()
def max_workers(requested: int | None) -> int:
cpu = os.cpu_count() or 1
limit = max(1, min(8, cpu))
if requested and requested > 0:
return min(requested, max(1, cpu))
return limit
def log(mode: str, message: str, *, verbose_only: bool = False, verbose: bool = False) -> None:
if verbose_only and not verbose:
return
print(f"[{mode}] {message}")
2025-11-06 12:35:07 -07:00
def is_archive_path(path: Path) -> bool:
2025-11-10 19:32:31 -07:00
return any(part in ("_archive", "_CURRENT") for part in path.parts)
2025-11-06 12:35:07 -07:00
2025-11-06 10:10:29 -07:00
def find_sequence_dirs(root: Path) -> Iterator[Path]:
for dirpath, dirnames, filenames in os.walk(root):
path = Path(dirpath)
2025-11-10 19:32:31 -07:00
dirnames[:] = [d for d in dirnames if d not in ("_archive", "_CURRENT")]
2025-11-06 12:35:07 -07:00
if is_archive_path(path):
2025-11-06 10:10:29 -07:00
continue
has_frames = any(Path(dirpath, f).suffix.lower() in SEQUENCE_EXTENSIONS for f in filenames)
if has_frames:
yield path
2025-11-06 12:35:07 -07:00
def iter_sequence_files(seq_dir: Path) -> Iterator[Path]:
for dirpath, dirnames, filenames in os.walk(seq_dir):
path = Path(dirpath)
2025-11-10 19:32:31 -07:00
dirnames[:] = [d for d in dirnames if d not in ("_archive", "_CURRENT")]
2025-11-06 12:35:07 -07:00
if is_archive_path(path):
continue
for filename in filenames:
yield path / filename
2025-11-06 10:10:29 -07:00
def compute_state(seq_dir: Path) -> dict:
entries = []
2025-11-06 12:35:07 -07:00
files = sorted(
iter_sequence_files(seq_dir),
key=lambda p: p.relative_to(seq_dir).as_posix(),
)
for file_path in files:
2025-11-06 10:10:29 -07:00
stat = file_path.stat()
entries.append(
{
2025-11-06 12:35:07 -07:00
"path": file_path.relative_to(seq_dir).as_posix(),
2025-11-06 10:10:29 -07:00
"size": stat.st_size,
"mtime_ns": stat.st_mtime_ns,
}
)
return {"files": entries}
def current_state(seq_dir: Path) -> dict:
if not seq_dir.exists() or not seq_dir.is_dir():
return {"files": []}
return compute_state(seq_dir)
def load_state(state_path: Path) -> dict | None:
if not state_path.exists():
return None
try:
return json.loads(state_path.read_text())
except json.JSONDecodeError:
return None
def state_changed(seq_state: dict, stored_state: dict | None) -> bool:
if stored_state is None:
return True
return seq_state != stored_state
def archive_path_for(seq_dir: Path) -> Path:
rel = seq_dir.relative_to(RENDER_ROOT)
2025-11-10 11:27:44 -07:00
suffix = ".7z" if ZIPPER_TYPE == "7z" else ".zip"
return (ARCHIVE_ROOT / rel).with_suffix(suffix)
2025-11-06 10:10:29 -07:00
def sequence_dir_for(zip_path: Path) -> Path:
rel = zip_path.relative_to(ARCHIVE_ROOT)
return (RENDER_ROOT / rel).with_suffix("")
def state_path_for(zip_path: Path) -> Path:
return zip_path.with_suffix(zip_path.suffix + STATE_SUFFIX)
def zip_sequence(seq_dir: Path, zip_path: Path) -> None:
2025-11-10 10:36:28 -07:00
if ZIPPER_TYPE == "7z":
if SEVEN_Z_EXE is None:
2025-11-10 11:27:44 -07:00
raise RuntimeError(
"7z compression requested but 7z executable not found in PATH. "
"Please install 7z (e.g., via Chocolatey: choco install 7zip) "
"or set zipper to 'zip' in config.json"
)
2025-11-08 02:36:20 -07:00
zip_path.parent.mkdir(parents=True, exist_ok=True)
2025-11-10 11:27:44 -07:00
# If creating a .7z file, remove any existing .zip file for the same sequence
if zip_path.suffix == ".7z":
old_zip_path = zip_path.with_suffix(".zip")
if old_zip_path.exists():
old_zip_path.unlink(missing_ok=True)
old_state_path = state_path_for(old_zip_path)
if old_state_path.exists():
old_state_path.unlink(missing_ok=True)
2025-11-10 10:36:28 -07:00
# Build list of files to archive with relative paths
file_list = []
for file_path in iter_sequence_files(seq_dir):
rel_path = file_path.relative_to(seq_dir).as_posix()
file_list.append(rel_path)
if not file_list:
raise RuntimeError(f"No files found to archive in {seq_dir}")
2025-11-10 10:36:28 -07:00
2025-11-10 11:27:44 -07:00
# Create zip in temporary location first to avoid issues with corrupted existing files
temp_zip = None
list_file_path = None
2025-11-10 10:36:28 -07:00
try:
2025-11-10 11:27:44 -07:00
# Create temporary archive file path (but don't create the file - let 7z create it)
temp_zip_path = tempfile.mktemp(suffix=".7z", dir=zip_path.parent)
temp_zip = Path(temp_zip_path)
# Create list file with absolute path
fd, temp_path = tempfile.mkstemp(suffix=".lst", text=True)
list_file_path = Path(temp_path)
with os.fdopen(fd, "w", encoding="utf-8") as list_file:
for rel_path in file_list:
list_file.write(rel_path + "\n")
list_file.flush()
os.fsync(list_file.fileno()) # Ensure data is written to disk
# File is closed here by context manager, small delay to ensure OS releases handle
time.sleep(0.1)
2025-11-10 11:27:44 -07:00
# Use absolute paths for both list file and temp zip
list_file_abs = list_file_path.resolve()
2025-11-10 11:27:44 -07:00
temp_zip_abs = temp_zip.resolve()
# Create archive in temp location first (7z will create it fresh)
2025-11-10 10:36:28 -07:00
cmd = [
SEVEN_Z_EXE,
"a",
"-y",
"-bb0", # Suppress progress output
2025-11-10 10:36:28 -07:00
f"-mx={COMPRESSION_LEVEL}",
2025-11-10 11:27:44 -07:00
"-t7z", # Use 7z format, not zip
str(temp_zip_abs),
f"@{list_file_abs}",
2025-11-10 10:36:28 -07:00
]
result = subprocess.run(
cmd,
cwd=seq_dir,
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode != 0:
error_msg = result.stderr.strip() if result.stderr else "Unknown error"
if result.stdout:
error_msg += f"\nstdout: {result.stdout.strip()}"
raise RuntimeError(f"7z compression failed: {error_msg}")
2025-11-10 11:27:44 -07:00
# Move temp zip to final location, replacing any existing file
if zip_path.exists():
zip_path.unlink()
temp_zip.replace(zip_path)
temp_zip = None # Mark as moved so we don't delete it
2025-11-10 10:36:28 -07:00
finally:
2025-11-10 11:27:44 -07:00
# Clean up temp zip if it wasn't moved
if temp_zip and temp_zip.exists():
try:
temp_zip.unlink(missing_ok=True)
except OSError:
pass
# Clean up list file, with retry in case 7z still has it open
if list_file_path and list_file_path.exists():
for attempt in range(3):
try:
list_file_path.unlink(missing_ok=True)
break
except PermissionError:
if attempt < 2:
time.sleep(0.1) # Wait 100ms before retry
else:
# Last attempt failed, just log and continue
# The temp file will be cleaned up by the OS eventually
pass
2025-11-08 02:36:20 -07:00
return
2025-11-10 11:27:44 -07:00
# Use zipfile (only if ZIPPER_TYPE == "zip")
if ZIPPER_TYPE == "zip":
from zipfile import ZIP_DEFLATED, ZIP_STORED, ZipFile
2025-11-06 10:10:29 -07:00
2025-11-10 11:27:44 -07:00
zip_path.parent.mkdir(parents=True, exist_ok=True)
if COMPRESSION_LEVEL <= 0:
compression = ZIP_STORED
zip_kwargs = {}
else:
compression = ZIP_DEFLATED
zip_kwargs = {"compresslevel": COMPRESSION_LEVEL}
2025-11-08 02:36:20 -07:00
2025-11-10 11:27:44 -07:00
with ZipFile(zip_path, "w", compression=compression, **zip_kwargs) as archive:
for file_path in iter_sequence_files(seq_dir):
archive.write(file_path, arcname=file_path.relative_to(seq_dir).as_posix())
return
# Unknown ZIPPER_TYPE - fail with clear error
raise RuntimeError(
f"Unsupported ZIPPER_TYPE: {ZIPPER_TYPE!r}. "
f"Expected '7z' or 'zip'. "
f"Config zipper value: {CONFIG.get('zipper', 'not set')!r}"
)
2025-11-06 10:10:29 -07:00
def expand_sequence(zip_path: Path, seq_state: dict) -> None:
target_dir = sequence_dir_for(zip_path)
if target_dir.exists():
shutil.rmtree(target_dir)
target_dir.mkdir(parents=True, exist_ok=True)
2025-11-10 10:36:28 -07:00
if ZIPPER_TYPE == "7z":
if SEVEN_Z_EXE is None:
2025-11-10 11:27:44 -07:00
raise RuntimeError(
"7z extraction requested but 7z executable not found in PATH. "
"Please install 7z or set zipper to 'zip' in config.json"
)
2025-11-08 02:36:20 -07:00
cmd = [
SEVEN_Z_EXE,
"x",
"-y",
str(zip_path),
f"-o{target_dir}",
]
2025-11-10 11:27:44 -07:00
result = subprocess.run(
cmd,
check=False,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
)
if result.returncode != 0:
error_msg = result.stderr.strip() if result.stderr else "Unknown error"
if result.stdout:
error_msg += f"\nstdout: {result.stdout.strip()}"
raise RuntimeError(f"7z extraction failed: {error_msg}")
elif ZIPPER_TYPE == "zip":
2025-11-08 02:36:20 -07:00
from zipfile import ZipFile
with ZipFile(zip_path, "r") as archive:
archive.extractall(target_dir)
2025-11-10 11:27:44 -07:00
else:
raise RuntimeError(
f"Unsupported ZIPPER_TYPE: {ZIPPER_TYPE!r}. "
f"Expected '7z' or 'zip'. "
f"Config zipper value: {CONFIG.get('zipper', 'not set')!r}"
)
2025-11-06 10:10:29 -07:00
for entry in seq_state.get("files", []):
file_path = target_dir / entry["path"]
if file_path.exists():
os.utime(file_path, ns=(entry["mtime_ns"], entry["mtime_ns"]))
def process_zip(seq_dir: Path, zip_path: Path, state_path: Path, seq_state: dict, *, verbose: bool) -> Sequence[Path]:
log("zip", f"{seq_dir} -> {zip_path}", verbose_only=True, verbose=verbose)
zip_sequence(seq_dir, zip_path)
state_path.write_text(json.dumps(seq_state, indent=2))
return (zip_path, state_path)
def process_expand(zip_path: Path, state: dict, *, verbose: bool) -> None:
log("expand", f"{zip_path} -> {sequence_dir_for(zip_path)}", verbose_only=True, verbose=verbose)
expand_sequence(zip_path, state)
def run_zip(worker_count: int, *, verbose: bool) -> int:
work_items: list[tuple[Path, Path, Path, dict]] = []
if RENDER_ROOT.exists():
for seq_dir in find_sequence_dirs(RENDER_ROOT):
seq_state = compute_state(seq_dir)
if not seq_state["files"]:
continue
2025-11-10 11:27:44 -07:00
# Get the target archive path (will be .7z if ZIPPER_TYPE is "7z")
2025-11-06 10:10:29 -07:00
zip_path = archive_path_for(seq_dir)
state_path = state_path_for(zip_path)
2025-11-10 11:27:44 -07:00
# Check if we need to upgrade from .zip to .7z
old_zip_path = None
if ZIPPER_TYPE == "7z":
# Check if an old .zip file exists
old_zip_path = zip_path.with_suffix(".zip")
if old_zip_path.exists():
# Check if the old .zip's metadata matches current state
old_state_path = state_path_for(old_zip_path)
old_stored_state = load_state(old_state_path)
if not state_changed(seq_state, old_stored_state):
# Old .zip is up to date, skip conversion
continue
# Old .zip is out of date, will be replaced with .7z
# Check if the target archive (e.g., .7z) already exists and is up to date
2025-11-06 10:10:29 -07:00
stored_state = load_state(state_path)
if not state_changed(seq_state, stored_state):
2025-11-10 11:27:44 -07:00
# Target archive is up to date, but we might still need to clean up old .zip
if old_zip_path and old_zip_path.exists():
# Old .zip exists but we have a newer .7z, remove the old one
old_zip_path.unlink(missing_ok=True)
old_state_path = state_path_for(old_zip_path)
if old_state_path.exists():
old_state_path.unlink(missing_ok=True)
2025-11-06 10:10:29 -07:00
continue
work_items.append((seq_dir, zip_path, state_path, seq_state))
if not work_items:
2025-11-06 11:02:59 -07:00
if not RENDER_ROOT.exists():
log("zip", "Render root 'Renders' not found; nothing to zip.")
else:
log("zip", "Archives already up to date; no sequences needed zipping.")
2025-11-06 10:10:29 -07:00
return 0
updated_paths: list[Path] = []
total = len(work_items)
completed = 0
with ThreadPoolExecutor(max_workers=worker_count) as executor:
future_map = {
executor.submit(process_zip, seq_dir, zip_path, state_path, seq_state, verbose=verbose): seq_dir
for seq_dir, zip_path, state_path, seq_state in work_items
}
for future in as_completed(future_map):
updated_paths.extend(future.result())
completed += 1
seq_dir = future_map[future]
rel = seq_dir.relative_to(RENDER_ROOT)
log("zip", f"{completed}/{total} {rel}")
updated_count = len(updated_paths) // 2
log("zip", f"Updated {updated_count} sequence archive(s).", verbose=verbose)
if updated_paths:
log(
"zip",
"Archives updated. Stage manually with `git add Renders/_zipped`, if desired.",
verbose_only=True,
verbose=verbose,
)
removed = cleanup_orphan_archives(verbose=verbose)
if removed:
log("zip", f"Removed {removed} orphan archive(s).", verbose=verbose)
return updated_count
def run_expand(worker_count: int, *, verbose: bool) -> int:
if not ARCHIVE_ROOT.exists():
2025-11-06 11:02:59 -07:00
log("expand", "No archives to expand (missing 'Renders/_zipped').")
2025-11-06 10:10:29 -07:00
return 0
work_items: list[tuple[Path, dict]] = []
2025-11-10 11:27:44 -07:00
# Look for both .zip and .7z archives
archive_patterns = ["*.zip", "*.7z"]
for pattern in archive_patterns:
for zip_path in ARCHIVE_ROOT.rglob(pattern):
state_path = state_path_for(zip_path)
seq_state = load_state(state_path)
if seq_state is None:
log("expand", f"Skipping {zip_path} (missing metadata)")
continue
2025-11-06 10:10:29 -07:00
2025-11-10 11:27:44 -07:00
target_dir = sequence_dir_for(zip_path)
if current_state(target_dir) == seq_state:
continue
2025-11-06 10:10:29 -07:00
2025-11-10 11:27:44 -07:00
work_items.append((zip_path, seq_state))
2025-11-06 10:10:29 -07:00
if not work_items:
2025-11-06 11:02:59 -07:00
log("expand", "Working folders already match archives; nothing to expand.")
2025-11-06 10:10:29 -07:00
return 0
total = len(work_items)
completed = 0
with ThreadPoolExecutor(max_workers=worker_count) as executor:
future_map = {
executor.submit(process_expand, zip_path, seq_state, verbose=verbose): zip_path
for zip_path, seq_state in work_items
}
for future in as_completed(future_map):
future.result()
completed += 1
zip_path = future_map[future]
rel = zip_path.relative_to(ARCHIVE_ROOT)
log("expand", f"{completed}/{total} {rel}")
log("expand", f"Refreshed {len(work_items)} sequence folder(s).", verbose=verbose)
return len(work_items)
def cleanup_orphan_archives(*, verbose: bool) -> int:
if not ARCHIVE_ROOT.exists():
return 0
removed: list[Path] = []
2025-11-10 11:27:44 -07:00
# Look for both .zip and .7z archives
archive_patterns = ["*.zip", "*.7z"]
for pattern in archive_patterns:
for zip_path in ARCHIVE_ROOT.rglob(pattern):
seq_dir = sequence_dir_for(zip_path)
if seq_dir.exists():
continue
2025-11-06 10:10:29 -07:00
2025-11-10 11:27:44 -07:00
rel = zip_path.relative_to(ARCHIVE_ROOT)
log("zip", f"Removing orphan archive {rel}", verbose_only=True, verbose=verbose)
2025-11-06 10:10:29 -07:00
2025-11-10 11:27:44 -07:00
zip_path.unlink(missing_ok=True)
state_path = state_path_for(zip_path)
if state_path.exists():
state_path.unlink()
removed.append(zip_path)
2025-11-06 10:10:29 -07:00
if not removed:
return 0
for parent in sorted({p.parent for p in removed}, key=lambda p: len(p.parts), reverse=True):
if not parent.exists():
continue
while parent != ARCHIVE_ROOT and not any(parent.iterdir()):
parent.rmdir()
parent = parent.parent
return len(removed)
def main() -> int:
args = parse_args()
workers = max_workers(args.jobs)
if args.mode == "expand":
run_expand(workers, verbose=args.verbose)
return 0
updated = run_zip(workers, verbose=args.verbose)
return 0 if updated >= 0 else 1
if __name__ == "__main__":
try:
raise SystemExit(main())
except Exception as exc: # broad to surface unexpected errors cleanly
print(f"Sequence sync failed: {exc}", file=sys.stderr)
raise