mirror of
https://github.com/ilri/csv-metadata-quality.git
synced 2024-10-31 19:43:00 +01:00
csv_metadata_quality/check.py: update title in citation check
Initialize the titles and citations before the for loop so we can access them later. This makes it easier to check if the item actua- lly has a citation.
This commit is contained in:
parent
e02678cd7c
commit
4d5696c4cb
@ -419,6 +419,11 @@ def title_in_citation(row):
|
||||
|
||||
Function prints a warning if the title does not appear in the citation.
|
||||
"""
|
||||
# Initialize some variables at global scope so that we can set them in the
|
||||
# loop scope below and still be able to access them afterwards.
|
||||
title = ""
|
||||
citation = ""
|
||||
|
||||
# Iterate over the labels of the current row's values to get the names of
|
||||
# the title and citation columns. Then we check if the title is present in
|
||||
# the citation.
|
||||
@ -430,17 +435,15 @@ def title_in_citation(row):
|
||||
# Find the name of the title column
|
||||
match = re.match(r"^(dc|dcterms)\.title.*$", label)
|
||||
if match is not None:
|
||||
title_column_name = label
|
||||
title = row[label]
|
||||
|
||||
# Find the name of the citation column
|
||||
match = re.match(r"^.*?[cC]itation.*$", label)
|
||||
if match is not None:
|
||||
citation_column_name = label
|
||||
citation = row[label]
|
||||
|
||||
if row[citation_column_name] != "":
|
||||
if row[title_column_name] not in row[citation_column_name]:
|
||||
print(
|
||||
f"{Fore.YELLOW}Title is not present in citation: {Fore.RESET}{row[title_column_name]}"
|
||||
)
|
||||
if citation != "":
|
||||
if title not in citation:
|
||||
print(f"{Fore.YELLOW}Title is not present in citation: {Fore.RESET}{title}")
|
||||
|
||||
return
|
||||
|
Loading…
Reference in New Issue
Block a user