qurator-spk/modstool
1
0
Fork 0
mirror of https://github.com/qurator-spk/modstool.git synced 2025-11-04 11:24:14 +01:00
Code Issues Releases Wiki Activity

⚡ Make get_sets_file faster by using a lookup table

Browse source
This commit is contained in:
Gerber, Mike 2023-12-09 12:05:20 +01:00
parent 912e5d2b4a
commit 8fc4eeeb3b
1 changed files with 10 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

12
qurator/mods4pandas/mods4pandas.py
View file

@ -279,10 +279,18 @@ def pages_to_dict(mets, raise_errors=True) -> List[Dict]:
div_physSequence = structMap_PHYSICAL[0]
assert div_physSequence.attrib.get("TYPE") == "physSequence"
# Build a look-up table to get mets:file by @ID
# This cuts retrieving the mets:file down to half the time.
mets_file_by_ID = {}
def _init_mets_file_by_ID():
for f in fileSec.iterfind('./mets:fileGrp/mets:file', ns):
mets_file_by_ID[f.attrib.get("ID")] = f
_init_mets_file_by_ID()
def get_mets_file(*, ID):
if ID:
file_ = fileSec.find(f'./mets:fileGrp/mets:file[@ID="{ID}"]', ns)
return file_
return mets_file_by_ID[ID]
def get_mets_div(*, ID):
if ID: