1
0
mirror of https://github.com/ilri/csv-metadata-quality.git synced 2025-11-15 11:39:41 +01:00

1 Commits

Author SHA1 Message Date
renovate[bot]
58ff726203 chore(deps): update dependency uv_build to >=0.9.5,<0.10.0 2025-10-22 02:57:43 +00:00
4 changed files with 361 additions and 645 deletions

View File

@@ -7,7 +7,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
## Unreleased
### Changed
- New AGROVOC REST API URL
- Use urllib from Python stdlib instead of manual replacement for unquoting URLs
## [0.7.0] - 2025-01-31
### Added

View File

@@ -40,7 +40,7 @@ csv-metadata-quality = 'csv_metadata_quality.__main__:main'
# See: https://docs.astral.sh/uv/concepts/build-backend/
[build-system]
requires = ["uv_build>=0.8.8,<0.9.0"]
requires = ["uv_build>=0.9.5,<0.10.0"]
build-backend = "uv_build"
[dependency-groups]
@@ -49,7 +49,6 @@ dev = [
"isort~=6.0",
"csvkit~=2.0",
"ipython~=8.31",
"black~=25.9",
]
[tool.isort]

View File

@@ -3,7 +3,6 @@
import logging
import re
from unicodedata import normalize
from urllib.parse import unquote
import country_converter as coco
import pandas as pd
@@ -452,8 +451,12 @@ def normalize_dois(field):
if match:
new_value = re.sub(pattern, "doi.org", new_value)
# Replace %xx escapes with their single-character equivalent.
new_value = unquote(new_value)
# Convert erroneous %2f to /
pattern = re.compile("%2f")
match = re.findall(pattern, new_value)
if match:
new_value = re.sub(pattern, "/", new_value)
# Replace values like doi: 10.11648/j.jps.20140201.14
pattern = re.compile(r"^doi: 10\.")

993
uv.lock generated

File diff suppressed because it is too large Load Diff