Files
ucsinfer/ucsinfer/util.py
2025-08-30 20:14:09 -07:00

141 lines
3.8 KiB
Python

import subprocess
import json
import os
from typing import NamedTuple, Optional
from re import match
def ffmpeg_description(path: str) -> Optional[str]:
result = subprocess.run(['ffprobe', '-show_format', '-of',
'json', path], capture_output=True)
try:
result.check_returncode()
except:
return None
stream = json.loads(result.stdout)
fmt = stream.get("format", None)
if fmt:
tags = fmt.get("tags", None)
if tags:
return tags.get("comment", None)
class UcsNameComponents(NamedTuple):
"""
Components of a UCS filename
"""
cat_id: str
user_cat: str | None
vendor_cat: str | None
fx_name: str
creator: str | None
source: str | None
user_data: str | None
def validate(self):
"""
Check if fields do not contain invalid characters
"""
if not match(r"[A-Z]+[a-z]+", self.cat_id):
return False
if self.user_cat and not match(r"[^\-_]+", self.user_cat):
return False
if self.vendor_cat and not match(r"[^\-_]+", self.vendor_cat):
return False
if not match(r"[^\-_]+", self.fx_name):
return False
if self.creator and not match(r"[^_]+", self.creator):
return False
if self.source and not match(r"[^_]+", self.source):
return False
if self.user_data and not match(r"[^.]+", self.user_data):
return False
def normalize_ucs(basename: str, catid_list: list[str]):
"""
Take any filename and normalize it into the UCS system
"""
n, ext = os.path.splitext(basename)
r = parse_ucs(n, catid_list)
if r:
pass
else:
pass
return f"aaa.{ext}"
def build_ucs(components: UcsNameComponents, extension: str) -> str:
"""
Build a UCS filename
"""
assert components.validate(), \
"UcsNameComponents contains invalid characters"
cat_segment = components.cat_id
if components.user_cat:
cat_segment += f"-{components.user_cat}"
name_segment = components.fx_name
if components.vendor_cat:
name_segment = f"{components.vendor_cat}-{components.fx_name}"
all_comps = [cat_segment, name_segment]
if components.creator:
all_comps += [components.creator]
if components.source:
all_comps += [components.source]
if components.user_data:
all_comps += [components.user_data]
root_name = "_".join(all_comps)
return root_name + '.' + extension
def parse_ucs(rootname: str,
catid_list: list[str]) -> Optional[UcsNameComponents]:
"""
Parse the UCS components from a file name root.
:param rootname: filename root, the basename of the file without extension
:param catid_list: a list of all UCS CatIDs
:returns: the components, or `None` if the filename is not in UCS format
"""
regexp1 = r"^(?P<CatID>[A-z]+)(-(?P<UserCat>[^_]+))?_"
regexp2 = r"((?P<VendorCat>[^-]+)-)?(?P<FXName>[^_]+)"
regexp3 = r"(_(?P<CreatorID>[^_]+)(_(?P<SourceID>[^_]+)"
regexp4 = r"(_(?P<UserData>[^.]+))?)?)?"
regexp = regexp1 + regexp2 + regexp3 + regexp4
matches = match(regexp, rootname)
if matches is None:
return None
if matches.group('CatID') not in catid_list:
return None
return UcsNameComponents(cat_id=matches.group('CatID'),
user_cat=matches.group('UserCat'),
vendor_cat=matches.group('VendorCat'),
fx_name=matches.group('FXName'),
creator=matches.group('CreatorID'),
source=matches.group('SourceID'),
user_data=matches.group('UserData'))