qurator-spk/modstool
1
0
Fork 0
mirror of https://github.com/qurator-spk/modstool.git synced 2025-08-14 12:09:52 +02:00
Code Issues Releases Wiki Activity

🧹 MODS: Handle mods:languageTerm with authority=iso639-2/rfc3066

Browse source
This commit is contained in:
Gerber, Mike 2025-08-07 20:29:40 +02:00
parent 5f1c1f68a7
commit 64aafbb88c
2 changed files with 13 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

12
src/mods4pandas/lib.py
View file

@ -187,6 +187,18 @@ class TagGroup:
warnings.warn("Changed scriptTerm authority to lower case")
return self
def fix_language_term(self) -> TagGroup:
for e in self.group:
if e.attrib["authority"] == "iso639-2":
e.attrib["authority"] = "iso639-2b"
warnings.warn("Changed languageTerm authority to iso639-2b")
if e.attrib["authority"] == "rfc3066":
if e.text == "de":
e.attrib["authority"] = "iso639-2b"
e.text = "deu"
warnings.warn("Changed languageTerm authority from rfc3066 to iso639-2b")
return self
def add_missing_type_text(self) -> TagGroup:
for e in self.group:
if not e.attrib.get("type") == "text":

1
src/mods4pandas/mods4pandas.py
View file

@ -255,6 +255,7 @@ def mods_to_dict(mods, raise_errors=True):
elif tag == "{http://www.loc.gov/mods/v3}languageTerm":
value["languageTerm"] = (
TagGroup(tag, group)
.fix_language_term()
.has_attributes({"authority": "iso639-2b", "type": "code"})
.text_set()
)