Version 0.6.1

Generated with poetry export:

    $ poetry export --without-hashes -f requirements.txt > requirements.txt
    $ poetry export --without-hashes --with dev -f requirements.txt > requirements-dev.txt

I am trying `--without-hashes` to work around an error on pip install
when running in CI:

    ERROR: In --require-hashes mode, all requirements must have their versions pinned with ==

.drone.yml

@@ -9,7 +9,7 @@ steps:
   commands:
   - id
   - python -V
-  - apt update && apt install -y gcc g++ libicu-dev pkg-config
+  - apt update && apt install -y gcc g++ libicu-dev pkg-config git
   - pip install -r requirements-dev.txt
   - pytest
   - python setup.py install
@@ -39,37 +39,7 @@ steps:
   commands:
   - id
   - python -V
-  - apt update && apt install -y gcc g++ libicu-dev pkg-config
-  - pip install -r requirements-dev.txt
-  - pytest
-  - python setup.py install
-  # Basic test
-  - csv-metadata-quality -i data/test.csv -o /tmp/test.csv
-  # Basic test with unsafe fixes
-  - csv-metadata-quality -i data/test.csv -o /tmp/test.csv -u
-  # Geography test
-  - csv-metadata-quality -i data/test-geography.csv -o /tmp/test.csv
-  # Geography test with unsafe fixes
-  - csv-metadata-quality -i data/test-geography.csv -o /tmp/test.csv -u
-  # Test with experimental checks
-  - csv-metadata-quality -i data/test.csv -o /tmp/test.csv -e
-  # Test with AGROVOC validation
-  - csv-metadata-quality -i data/test.csv -o /tmp/test.csv --agrovoc-fields dcterms.subject
-  # Test with AGROVOC validation (and dropping invalid)
-  - csv-metadata-quality -i data/test.csv -o /tmp/test.csv --agrovoc-fields dcterms.subject -d
-
----
-kind: pipeline
-type: docker
-name: python38
-
-steps:
-- name: test
-  image: python:3.8-slim
-  commands:
-  - id
-  - python -V
-  - apt update && apt install -y gcc g++ libicu-dev pkg-config
+  - apt update && apt install -y gcc g++ libicu-dev pkg-config git
   - pip install -r requirements-dev.txt
   - pytest
   - python setup.py install

.github/workflows/python-app.yml

@@ -12,14 +12,15 @@ on:
 jobs:
   build:
 
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.10
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: '3.10'
+        cache: 'pip'
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip

CHANGELOG.md

@@ -4,7 +4,24 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [0.6.0] = 2022-09-02
+## [0.6.1] - 2023-02-23
+### Fixed
+- Missing region check should ignore subregion field, if it exists
+
+### Changed
+- Use SPDX license data from SPDX themselves instead of spdx-license-list
+because it is deprecated and outdated
+- Require Python 3.9+
+- Don't run `fix.separators()` on title or abstract fields
+- Don't run whitespace or newline fixes on abstract fields
+- Ignore some common non-SPDX licenses
+- Ignore `__description` suffix in filenames meant for SAFBuilder when checking
+for uncommon file extensions
+
+### Updated
+- Python dependencies
+
+## [0.6.0] - 2022-09-02
 ### Changed
 - Perform fix for "unnecessary" Unicode characters after we try to fix encoding
 issues with ftfy

README.md

@@ -8,7 +8,7 @@
 
 A simple, but opinionated metadata quality checker and fixer designed to work with CSVs in the DSpace ecosystem (though it could theoretically work on any CSV that uses Dublin Core fields as columns). The implementation is essentially a pipeline of checks and fixes that begins with splitting multi-value fields on the standard DSpace "||" separator, trimming leading/trailing whitespace, and then proceeding to more specialized cases like ISSNs, ISBNs, languages, unnecessary Unicode, AGROVOC terms, etc.
 
-Requires Python 3.8 or greater. CSV support comes from the [Pandas](https://pandas.pydata.org/) library.
+Requires Python 3.9 or greater. CSV support comes from the [Pandas](https://pandas.pydata.org/) library.
 
 If you use the DSpace CSV metadata quality checker please cite:
 

csv_metadata_quality/app.py

@@ -90,12 +90,14 @@ def run(argv):
 
             continue
 
-        # Fix: whitespace
-        df[column] = df[column].apply(fix.whitespace, field_name=column)
-
-        # Fix: newlines
         if args.unsafe_fixes:
-            df[column] = df[column].apply(fix.newlines, field_name=column)
+            match = re.match(r"^.*?abstract.*$", column)
+            if match is None:
+                # Fix: whitespace
+                df[column] = df[column].apply(fix.whitespace, field_name=column)
+
+                # Fix: newlines
+                df[column] = df[column].apply(fix.newlines, field_name=column)
 
         # Fix: missing space after comma. Only run on author and citation
         # fields for now, as this problem is mostly an issue in names.
@@ -121,10 +123,14 @@ def run(argv):
         # Fix: unnecessary Unicode
         df[column] = df[column].apply(fix.unnecessary_unicode)
 
-        # Fix: invalid and unnecessary multi-value separators
-        df[column] = df[column].apply(fix.separators, field_name=column)
-        # Run whitespace fix again after fixing invalid separators
-        df[column] = df[column].apply(fix.whitespace, field_name=column)
+        # Fix: invalid and unnecessary multi-value separators. Skip the title
+        # and abstract fields because "|" is used to indicate something like
+        # a subtitle.
+        match = re.match(r"^.*?(abstract|title).*$", column)
+        if match is None:
+            df[column] = df[column].apply(fix.separators, field_name=column)
+            # Run whitespace fix again after fixing invalid separators
+            df[column] = df[column].apply(fix.whitespace, field_name=column)
 
         # Fix: duplicate metadata values
         df[column] = df[column].apply(fix.duplicates, field_name=column)

csv_metadata_quality/check.py

@@ -9,13 +9,12 @@ import country_converter as coco
 import pandas as pd
 import requests
 import requests_cache
-import spdx_license_list
 from colorama import Fore
 from pycountry import languages
 from stdnum import isbn as stdnum_isbn
 from stdnum import issn as stdnum_issn
 
-from csv_metadata_quality.util import is_mojibake
+from csv_metadata_quality.util import is_mojibake, load_spdx_licenses
 
 
 def issn(field):
@@ -34,7 +33,6 @@ def issn(field):
 
     # Try to split multi-value field on "||" separator
     for value in field.split("||"):
-
         if not stdnum_issn.is_valid(value):
             print(f"{Fore.RED}Invalid ISSN: {Fore.RESET}{value}")
 
@@ -57,7 +55,6 @@ def isbn(field):
 
     # Try to split multi-value field on "||" separator
     for value in field.split("||"):
-
         if not stdnum_isbn.is_valid(value):
             print(f"{Fore.RED}Invalid ISBN: {Fore.RESET}{value}")
 
@@ -174,7 +171,6 @@ def language(field):
 
     # Try to split multi-value field on "||" separator
     for value in field.split("||"):
-
         # After splitting, check if language value is 2 or 3 characters so we
         # can check it against ISO 639-1 or ISO 639-3 accordingly.
         if len(value) == 2:
@@ -287,6 +283,11 @@ def filename_extension(field):
 
     # Iterate over all values
     for value in values:
+        # Strip filename descriptions that are meant for SAF Bundler, for
+        # example: Annual_Report_2020.pdf__description:Report
+        if "__description" in value:
+            value = value.split("__")[0]
+
         # Assume filename extension does not match
         filename_extension_match = False
 
@@ -313,13 +314,26 @@ def spdx_license_identifier(field):
     Prints the value if it is invalid.
     """
 
+    # List of common non-SPDX licenses to ignore
+    # See: https://ilri.github.io/cgspace-submission-guidelines/dcterms-license/dcterms-license.txt
+    ignore_licenses = {
+        "All rights reserved; no re-use allowed",
+        "All rights reserved; self-archive copy only",
+        "Copyrighted; Non-commercial educational use only",
+        "Copyrighted; Non-commercial use only",
+        "Copyrighted; all rights reserved",
+        "Other",
+    }
+
     # Skip fields with missing values
-    if pd.isna(field):
+    if pd.isna(field) or field in ignore_licenses:
         return
 
+    spdx_licenses = load_spdx_licenses()
+
     # Try to split multi-value field on "||" separator
     for value in field.split("||"):
-        if value not in spdx_license_list.LICENSES:
+        if value not in spdx_licenses:
             print(f"{Fore.YELLOW}Non-SPDX license identifier: {Fore.RESET}{value}")
 
     return
@@ -512,9 +526,9 @@ def countries_match_regions(row, exclude):
         if match is not None:
             country_column_name = label
 
-        # Find the name of the region column
+        # Find the name of the region column, but make sure it's not subregion!
         match = re.match(r"^.*?region.*$", label)
-        if match is not None:
+        if match is not None and "sub" not in label:
             region_column_name = label
 
         # Find the name of the title column
@@ -550,7 +564,7 @@ def countries_match_regions(row, exclude):
 
             if un_region != "not found" and un_region not in regions:
                 print(
-                    f"{Fore.YELLOW}Missing region ({un_region}): {Fore.RESET}{row[title_column_name]}"
+                    f"{Fore.YELLOW}Missing region ({country} → {un_region}): {Fore.RESET}{row[title_column_name]}"
                 )
 
     return

csv_metadata_quality/fix.py

@@ -327,9 +327,9 @@ def countries_match_regions(row, exclude):
         if match is not None:
             country_column_name = label
 
-        # Find the name of the region column
+        # Find the name of the region column, but make sure it's not subregion!
         match = re.match(r"^.*?region.*$", label)
-        if match is not None:
+        if match is not None and "sub" not in label:
             region_column_name = label
 
         # Find the name of the title column

csv_metadata_quality/util.py

@@ -1,5 +1,9 @@
 # SPDX-License-Identifier: GPL-3.0-only
 
+
+import json
+from importlib.resources import files
+
 from ftfy.badness import is_bad
 
 
@@ -49,3 +53,13 @@ def is_mojibake(field):
     else:
         # Encodable as CP-1252, Mojibake alert level high
         return True
+
+
+def load_spdx_licenses():
+    """Returns a Python list of SPDX short license identifiers."""
+
+    with open(files("csv_metadata_quality").joinpath("data/licenses.json")) as f:
+        licenses = json.load(f)
+
+    # List comprehension to extract the license ID for each license
+    return [license["licenseId"] for license in licenses["licenses"]]

csv_metadata_quality/version.py

@@ -1,3 +1,3 @@
 # SPDX-License-Identifier: GPL-3.0-only
 
-VERSION = "0.6.0"
+VERSION = "0.6.1"

data/abstract-check.csv

@@ -0,0 +1,17 @@
+id,dc.title,dcterms.abstract
+1,Normal item,This is an abstract
+2,Leading whitespace, This is an abstract
+3,Trailing whitespace,This is an abstract 
+4,Consecutive whitespace,This is  an abstract
+5,Newline,"This
+is an abstract"
+6,Newline with leading whitespace," This
+ is an abstract"
+7,Newline with trailing whitespace,"This 
+is an abstract "
+8,Newline with consecutive whitespace,"This
+is an  abstract"
+9,Multiple newlines,"This
+is
+an
+abstract"

data/test.csv

@@ -1,38 +1,39 @@
-dc.title,dcterms.issued,dc.identifier.issn,dc.identifier.isbn,dcterms.language,dcterms.subject,cg.coverage.country,filename,dcterms.license,dcterms.type,dcterms.bibliographicCitation,cg.identifier.doi,cg.coverage.region
- Leading space,2019-07-29,,,,,,,,,,,
-Trailing space ,2019-07-29,,,,,,,,,,,
-Excessive  space,2019-07-29,,,,,,,,,,,
-Miscellaenous ||whitespace | issues ,2019-07-29,,,,,,,,,,,
-Duplicate||Duplicate,2019-07-29,,,,,,,,,,,
-Invalid ISSN,2019-07-29,2321-2302,,,,,,,,,,
-Invalid ISBN,2019-07-29,,978-0-306-40615-6,,,,,,,,,
-Multiple valid ISSNs,2019-07-29,0378-5955||0024-9319,,,,,,,,,,
-Multiple valid ISBNs,2019-07-29,,99921-58-10-7||978-0-306-40615-7,,,,,,,,,
-Invalid date,2019-07-260,,,,,,,,,,,
-Multiple dates,2019-07-26||2019-01-10,,,,,,,,,,,
-Invalid multi-value separator,2019-07-29,0378-5955|0024-9319,,,,,,,,,,
-Unnecessary Unicode​,2019-07-29,,,,,,,,,,,
-Suspicious character||foreˆt,2019-07-29,,,,,,,,,,,
-Invalid ISO 639-1 (alpha 2) language,2019-07-29,,,jp,,,,,,,,
-Invalid ISO 639-3 (alpha 3) language,2019-07-29,,,chi,,,,,,,,
-Invalid language,2019-07-29,,,Span,,,,,,,,
-Invalid AGROVOC subject,2019-07-29,,,,LIVESTOCK||FOREST,,,,,,,
+dc.title,dcterms.issued,dc.identifier.issn,dc.identifier.isbn,dcterms.language,dcterms.subject,cg.coverage.country,filename,dcterms.license,dcterms.type,dcterms.bibliographicCitation,cg.identifier.doi,cg.coverage.region,cg.coverage.subregion
+ Leading space,2019-07-29,,,,,,,,,,,,
+Trailing space ,2019-07-29,,,,,,,,,,,,
+Excessive  space,2019-07-29,,,,,,,,,,,,
+Miscellaenous ||whitespace | issues ,2019-07-29,,,,,,,,,,,,
+Duplicate||Duplicate,2019-07-29,,,,,,,,,,,,
+Invalid ISSN,2019-07-29,2321-2302,,,,,,,,,,,
+Invalid ISBN,2019-07-29,,978-0-306-40615-6,,,,,,,,,,
+Multiple valid ISSNs,2019-07-29,0378-5955||0024-9319,,,,,,,,,,,
+Multiple valid ISBNs,2019-07-29,,99921-58-10-7||978-0-306-40615-7,,,,,,,,,,
+Invalid date,2019-07-260,,,,,,,,,,,,
+Multiple dates,2019-07-26||2019-01-10,,,,,,,,,,,,
+Invalid multi-value separator,2019-07-29,0378-5955|0024-9319,,,,,,,,,,,
+Unnecessary Unicode​,2019-07-29,,,,,,,,,,,,
+Suspicious character||foreˆt,2019-07-29,,,,,,,,,,,,
+Invalid ISO 639-1 (alpha 2) language,2019-07-29,,,jp,,,,,,,,,
+Invalid ISO 639-3 (alpha 3) language,2019-07-29,,,chi,,,,,,,,,
+Invalid language,2019-07-29,,,Span,,,,,,,,,
+Invalid AGROVOC subject,2019-07-29,,,,LIVESTOCK||FOREST,,,,,,,,
 Newline (LF),2019-07-30,,,,"TANZA
-NIA",,,,,,,
-Missing date,,,,,,,,,,,,
-Invalid country,2019-08-01,,,,,KENYAA,,,,,,
-Uncommon filename extension,2019-08-10,,,,,,file.pdf.lck,,,,,
-Unneccesary unicode (U+002D + U+00AD),2019-08-10,,978-­92-­9043-­823-­6,,,,,,,,,
-"Missing space,after comma",2019-08-27,,,,,,,,,,,
-Incorrect ISO 639-1 language,2019-09-26,,,es,,,,,,,,
-Incorrect ISO 639-3 language,2019-09-26,,,spa,,,,,,,,
-Composéd Unicode,2020-01-14,,,,,,,,,,,
-Decomposéd Unicode,2020-01-14,,,,,,,,,,,
-Unnecessary multi-value separator,2021-01-03,0378-5955||,,,,,,,,,,
-Invalid SPDX license identifier,2021-03-11,,,,,,,CC-BY,,,,
-Duplicate Title,2021-03-17,,,,,,,,Report,,,
-Duplicate Title,2021-03-17,,,,,,,,Report,,,
-Mojibake,2021-03-18,,,,Publicaçao CIAT,,,,Report,,,
-"DOI in citation, but missing cg.identifier.doi",2021-10-06,,,,,,,,,"Orth, A. 2021. DOI in citation, but missing cg.identifier.doi. doi: 10.1186/1743-422X-9-218",,
-Title missing from citation,2021-12-05,,,,,,,,,"Orth, A. 2021. Title missing f rom citation.",,
-Country missing region,2021-12-08,,,,,Kenya,,,,,,
+NIA",,,,,,,,
+Missing date,,,,,,,,,,,,,
+Invalid country,2019-08-01,,,,,KENYAA,,,,,,,
+Uncommon filename extension,2019-08-10,,,,,,file.pdf.lck,,,,,,
+Unneccesary unicode (U+002D + U+00AD),2019-08-10,,978-­92-­9043-­823-­6,,,,,,,,,,
+"Missing space,after comma",2019-08-27,,,,,,,,,,,,
+Incorrect ISO 639-1 language,2019-09-26,,,es,,,,,,,,,
+Incorrect ISO 639-3 language,2019-09-26,,,spa,,,,,,,,,
+Composéd Unicode,2020-01-14,,,,,,,,,,,,
+Decomposéd Unicode,2020-01-14,,,,,,,,,,,,
+Unnecessary multi-value separator,2021-01-03,0378-5955||,,,,,,,,,,,
+Invalid SPDX license identifier,2021-03-11,,,,,,,CC-BY,,,,,
+Duplicate Title,2021-03-17,,,,,,,,Report,,,,
+Duplicate Title,2021-03-17,,,,,,,,Report,,,,
+Mojibake,2021-03-18,,,,Publicaçao CIAT,,,,Report,,,,
+"DOI in citation, but missing cg.identifier.doi",2021-10-06,,,,,,,,,"Orth, A. 2021. DOI in citation, but missing cg.identifier.doi. doi: 10.1186/1743-422X-9-218",,,
+Title missing from citation,2021-12-05,,,,,,,,,"Orth, A. 2021. Title missing f rom citation.",,,
+Country missing region,2021-12-08,,,,,Kenya,,,,,,,
+Subregion field shouldn’t trigger region checks,2022-12-07,,,,,Kenya,,,,,,Eastern Africa,Baringo

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "csv-metadata-quality"
-version = "0.6.0"
+version = "0.6.1"
 description="A simple, but opinionated CSV quality checking and fixing pipeline for CSVs in the DSpace ecosystem."
 authors = ["Alan Orth <alan.orth@gmail.com>"]
 license="GPL-3.0-only"
@@ -11,26 +11,32 @@ homepage = "https://github.com/ilri/csv-metadata-quality"
 csv-metadata-quality = 'csv_metadata_quality.__main__:main'
 
 [tool.poetry.dependencies]
-python = "^3.8"
-pandas = "^1.4.0"
-python-stdnum = "^1.13"
-requests = "^2.28.1"
-requests-cache = "^0.9.6"
-pycountry = "^22.3.5"
+python = "^3.9"
+pandas = "^1.5.2"
+python-stdnum = "^1.18"
+requests = "^2.28.2"
+requests-cache = "^0.9.8"
 langid = "^1.1.6"
-colorama = "^0.4.5"
-spdx-license-list = "^0.5.2"
+colorama = "^0.4.6"
 ftfy = "^6.1.1"
-country-converter = "^0.7.7"
+country-converter = {git = "https://github.com/alanorth/country_converter.git", rev = "myanmar-region"}
+pycountry = {git = "https://github.com/alanorth/pycountry", rev = "iso-codes-4.12.0"}
 
 [tool.poetry.dev-dependencies]
-pytest = "^7.1.3"
-flake8 = "^5.0.4"
+pytest = "^7.2.1"
+flake8 = "^6.0.0"
 pytest-clarity = "^1.0.1"
-black = "^22.8.0"
-isort = "^5.10.1"
-csvkit = "^1.0.7"
+black = "^23.1.0"
+isort = "^5.12.0"
+csvkit = "^1.1.0"
+
+[tool.poetry.group.dev.dependencies]
+ipython = "^8.10.0"
 
 [build-system]
 requires = ["poetry>=0.12"]
 build-backend = "poetry.masonry.api"
+
+[tool.isort]
+profile = "black"
+line_length=88

requirements-dev.txt

@@ -1,68 +1,80 @@
-agate-dbf==0.2.2 ; python_version >= "3.8" and python_version < "4.0"
-agate-excel==0.2.5 ; python_version >= "3.8" and python_version < "4.0"
-agate-sql==0.5.8 ; python_version >= "3.8" and python_version < "4.0"
-agate==1.6.3 ; python_version >= "3.8" and python_version < "4.0"
-appdirs==1.4.4 ; python_version >= "3.8" and python_version < "4.0"
-attrs==22.1.0 ; python_version >= "3.8" and python_version < "4.0"
-babel==2.10.3 ; python_version >= "3.8" and python_version < "4.0"
-black==22.8.0 ; python_version >= "3.8" and python_version < "4.0"
-cattrs==22.1.0 ; python_version >= "3.8" and python_version < "4.0"
-certifi==2022.6.15 ; python_version >= "3.8" and python_version < "4"
-charset-normalizer==2.1.1 ; python_version >= "3.8" and python_version < "4"
-click==8.1.3 ; python_version >= "3.8" and python_version < "4.0"
-colorama==0.4.5 ; python_version >= "3.8" and python_version < "4.0"
-commonmark==0.9.1 ; python_version >= "3.8" and python_version < "4.0"
-country-converter==0.7.7 ; python_version >= "3.8" and python_version < "4.0"
-csvkit==1.0.7 ; python_version >= "3.8" and python_version < "4.0"
-dbfread==2.0.7 ; python_version >= "3.8" and python_version < "4.0"
-et-xmlfile==1.1.0 ; python_version >= "3.8" and python_version < "4.0"
-exceptiongroup==1.0.0rc9 ; python_version >= "3.8" and python_version <= "3.10"
-flake8==5.0.4 ; python_version >= "3.8" and python_version < "4.0"
-ftfy==6.1.1 ; python_version >= "3.8" and python_version < "4"
-future==0.18.2 ; python_version >= "3.8" and python_version < "4.0"
-greenlet==1.1.3 ; python_version >= "3.8" and (platform_machine == "aarch64" or platform_machine == "ppc64le" or platform_machine == "x86_64" or platform_machine == "amd64" or platform_machine == "AMD64" or platform_machine == "win32" or platform_machine == "WIN32") and python_version < "4.0"
-idna==3.3 ; python_version >= "3.8" and python_version < "4"
-iniconfig==1.1.1 ; python_version >= "3.8" and python_version < "4.0"
-isodate==0.6.1 ; python_version >= "3.8" and python_version < "4.0"
-isort==5.10.1 ; python_version >= "3.8" and python_version < "4.0"
-langid==1.1.6 ; python_version >= "3.8" and python_version < "4.0"
-leather==0.3.4 ; python_version >= "3.8" and python_version < "4.0"
-mccabe==0.7.0 ; python_version >= "3.8" and python_version < "4.0"
-mypy-extensions==0.4.3 ; python_version >= "3.8" and python_version < "4.0"
-numpy==1.23.2 ; python_version < "4.0" and python_version >= "3.8"
-olefile==0.46 ; python_version >= "3.8" and python_version < "4.0"
-openpyxl==3.0.10 ; python_version >= "3.8" and python_version < "4.0"
-packaging==21.3 ; python_version >= "3.8" and python_version < "4.0"
-pandas==1.4.4 ; python_version >= "3.8" and python_version < "4.0"
-parsedatetime==2.4 ; python_version >= "3.8" and python_version < "4.0"
-pathspec==0.10.1 ; python_version >= "3.8" and python_version < "4.0"
-platformdirs==2.5.2 ; python_version >= "3.8" and python_version < "4.0"
-pluggy==1.0.0 ; python_version >= "3.8" and python_version < "4.0"
-pprintpp==0.4.0 ; python_version >= "3.8" and python_version < "4.0"
-py==1.11.0 ; python_version >= "3.8" and python_version < "4.0"
-pycodestyle==2.9.1 ; python_version >= "3.8" and python_version < "4.0"
-pycountry==22.3.5 ; python_version >= "3.8" and python_version < "4"
-pyflakes==2.5.0 ; python_version >= "3.8" and python_version < "4.0"
-pygments==2.13.0 ; python_version >= "3.8" and python_version < "4.0"
-pyparsing==3.0.9 ; python_version >= "3.8" and python_version < "4.0"
-pytest-clarity==1.0.1 ; python_version >= "3.8" and python_version < "4.0"
-pytest==7.1.3 ; python_version >= "3.8" and python_version < "4.0"
-python-dateutil==2.8.2 ; python_version >= "3.8" and python_version < "4.0"
-python-slugify==6.1.2 ; python_version >= "3.8" and python_version < "4.0"
-python-stdnum==1.17 ; python_version >= "3.8" and python_version < "4.0"
-pytimeparse==1.1.8 ; python_version >= "3.8" and python_version < "4.0"
-pytz==2022.2.1 ; python_version >= "3.8" and python_version < "4.0"
-requests-cache==0.9.6 ; python_version >= "3.8" and python_version < "4.0"
-requests==2.28.1 ; python_version >= "3.8" and python_version < "4"
-rich==12.5.1 ; python_version >= "3.8" and python_version < "4.0"
-setuptools==65.3.0 ; python_version >= "3.8" and python_version < "4"
-six==1.16.0 ; python_version >= "3.8" and python_version < "4.0"
-spdx-license-list==0.5.2 ; python_version >= "3.8" and python_version < "4.0"
-sqlalchemy==1.4.40 ; python_version >= "3.8" and python_version < "4.0"
-text-unidecode==1.3 ; python_version >= "3.8" and python_version < "4.0"
-tomli==2.0.1 ; python_version >= "3.8" and python_version < "4.0"
-typing-extensions==4.3.0 ; python_version >= "3.8" and python_version < "3.10"
-url-normalize==1.4.3 ; python_version >= "3.8" and python_version < "4.0"
-urllib3==1.26.12 ; python_version >= "3.8" and python_version < "4"
-wcwidth==0.2.5 ; python_version >= "3.8" and python_version < "4"
-xlrd==2.0.1 ; python_version >= "3.8" and python_version < "4.0"
+agate-dbf==0.2.2 ; python_version >= "3.9" and python_version < "4.0"
+agate-excel==0.2.5 ; python_version >= "3.9" and python_version < "4.0"
+agate-sql==0.5.9 ; python_version >= "3.9" and python_version < "4.0"
+agate==1.7.1 ; python_version >= "3.9" and python_version < "4.0"
+appdirs==1.4.4 ; python_version >= "3.9" and python_version < "4.0"
+appnope==0.1.3 ; python_version >= "3.9" and python_version < "4.0" and sys_platform == "darwin"
+asttokens==2.2.1 ; python_version >= "3.9" and python_version < "4.0"
+attrs==22.2.0 ; python_version >= "3.9" and python_version < "4.0"
+babel==2.11.0 ; python_version >= "3.9" and python_version < "4.0"
+backcall==0.2.0 ; python_version >= "3.9" and python_version < "4.0"
+black==23.1.0 ; python_version >= "3.9" and python_version < "4.0"
+cattrs==22.2.0 ; python_version >= "3.9" and python_version < "4.0"
+certifi==2022.12.7 ; python_version >= "3.9" and python_version < "4"
+charset-normalizer==3.0.1 ; python_version >= "3.9" and python_version < "4"
+click==8.1.3 ; python_version >= "3.9" and python_version < "4.0"
+colorama==0.4.6 ; python_version >= "3.9" and python_version < "4.0"
+country-converter @ git+https://github.com/alanorth/country_converter.git@myanmar-region ; python_version >= "3.9" and python_version < "4.0"
+csvkit==1.1.1 ; python_version >= "3.9" and python_version < "4.0"
+dbfread==2.0.7 ; python_version >= "3.9" and python_version < "4.0"
+decorator==5.1.1 ; python_version >= "3.9" and python_version < "4.0"
+et-xmlfile==1.1.0 ; python_version >= "3.9" and python_version < "4.0"
+exceptiongroup==1.1.0 ; python_version >= "3.9" and python_version < "3.11"
+executing==1.2.0 ; python_version >= "3.9" and python_version < "4.0"
+flake8==6.0.0 ; python_version >= "3.9" and python_version < "4.0"
+ftfy==6.1.1 ; python_version >= "3.9" and python_version < "4"
+greenlet==2.0.2 ; python_version >= "3.9" and (platform_machine == "aarch64" or platform_machine == "ppc64le" or platform_machine == "x86_64" or platform_machine == "amd64" or platform_machine == "AMD64" or platform_machine == "win32" or platform_machine == "WIN32") and python_version < "4.0"
+idna==3.4 ; python_version >= "3.9" and python_version < "4"
+iniconfig==2.0.0 ; python_version >= "3.9" and python_version < "4.0"
+ipython==8.10.0 ; python_version >= "3.9" and python_version < "4.0"
+isodate==0.6.1 ; python_version >= "3.9" and python_version < "4.0"
+isort==5.12.0 ; python_version >= "3.9" and python_version < "4.0"
+jedi==0.18.2 ; python_version >= "3.9" and python_version < "4.0"
+langid==1.1.6 ; python_version >= "3.9" and python_version < "4.0"
+leather==0.3.4 ; python_version >= "3.9" and python_version < "4.0"
+markdown-it-py==2.2.0 ; python_version >= "3.9" and python_version < "4.0"
+matplotlib-inline==0.1.6 ; python_version >= "3.9" and python_version < "4.0"
+mccabe==0.7.0 ; python_version >= "3.9" and python_version < "4.0"
+mdurl==0.1.2 ; python_version >= "3.9" and python_version < "4.0"
+mypy-extensions==1.0.0 ; python_version >= "3.9" and python_version < "4.0"
+numpy==1.24.2 ; python_version < "4.0" and python_version >= "3.9"
+olefile==0.46 ; python_version >= "3.9" and python_version < "4.0"
+openpyxl==3.1.1 ; python_version >= "3.9" and python_version < "4.0"
+packaging==23.0 ; python_version >= "3.9" and python_version < "4.0"
+pandas==1.5.3 ; python_version >= "3.9" and python_version < "4.0"
+parsedatetime==2.6 ; python_version >= "3.9" and python_version < "4.0"
+parso==0.8.3 ; python_version >= "3.9" and python_version < "4.0"
+pathspec==0.11.0 ; python_version >= "3.9" and python_version < "4.0"
+pexpect==4.8.0 ; python_version >= "3.9" and python_version < "4.0" and sys_platform != "win32"
+pickleshare==0.7.5 ; python_version >= "3.9" and python_version < "4.0"
+platformdirs==3.0.0 ; python_version >= "3.9" and python_version < "4.0"
+pluggy==1.0.0 ; python_version >= "3.9" and python_version < "4.0"
+pprintpp==0.4.0 ; python_version >= "3.9" and python_version < "4.0"
+prompt-toolkit==3.0.37 ; python_version >= "3.9" and python_version < "4.0"
+ptyprocess==0.7.0 ; python_version >= "3.9" and python_version < "4.0" and sys_platform != "win32"
+pure-eval==0.2.2 ; python_version >= "3.9" and python_version < "4.0"
+pycodestyle==2.10.0 ; python_version >= "3.9" and python_version < "4.0"
+pycountry @ git+https://github.com/alanorth/pycountry@iso-codes-4.12.0 ; python_version >= "3.9" and python_version < "4.0"
+pyflakes==3.0.1 ; python_version >= "3.9" and python_version < "4.0"
+pygments==2.14.0 ; python_version >= "3.9" and python_version < "4.0"
+pytest-clarity==1.0.1 ; python_version >= "3.9" and python_version < "4.0"
+pytest==7.2.1 ; python_version >= "3.9" and python_version < "4.0"
+python-dateutil==2.8.2 ; python_version >= "3.9" and python_version < "4.0"
+python-slugify==8.0.0 ; python_version >= "3.9" and python_version < "4.0"
+python-stdnum==1.18 ; python_version >= "3.9" and python_version < "4.0"
+pytimeparse==1.1.8 ; python_version >= "3.9" and python_version < "4.0"
+pytz==2022.7.1 ; python_version >= "3.9" and python_version < "4.0"
+requests-cache==0.9.8 ; python_version >= "3.9" and python_version < "4.0"
+requests==2.28.2 ; python_version >= "3.9" and python_version < "4"
+rich==13.3.1 ; python_version >= "3.9" and python_version < "4.0"
+six==1.16.0 ; python_version >= "3.9" and python_version < "4.0"
+sqlalchemy==1.4.46 ; python_version >= "3.9" and python_version < "4.0"
+stack-data==0.6.2 ; python_version >= "3.9" and python_version < "4.0"
+text-unidecode==1.3 ; python_version >= "3.9" and python_version < "4.0"
+tomli==2.0.1 ; python_version >= "3.9" and python_version < "3.11"
+traitlets==5.9.0 ; python_version >= "3.9" and python_version < "4.0"
+typing-extensions==4.5.0 ; python_version >= "3.9" and python_version < "3.10"
+url-normalize==1.4.3 ; python_version >= "3.9" and python_version < "4.0"
+urllib3==1.26.14 ; python_version >= "3.9" and python_version < "4"
+wcwidth==0.2.6 ; python_version >= "3.9" and python_version < "4"
+xlrd==2.0.1 ; python_version >= "3.9" and python_version < "4.0"

requirements.txt

@@ -1,25 +1,23 @@
-appdirs==1.4.4 ; python_version >= "3.8" and python_version < "4.0"
-attrs==22.1.0 ; python_version >= "3.8" and python_version < "4.0"
-cattrs==22.1.0 ; python_version >= "3.8" and python_version < "4.0"
-certifi==2022.6.15 ; python_version >= "3.8" and python_version < "4"
-charset-normalizer==2.1.1 ; python_version >= "3.8" and python_version < "4"
-colorama==0.4.5 ; python_version >= "3.8" and python_version < "4.0"
-country-converter==0.7.7 ; python_version >= "3.8" and python_version < "4.0"
-exceptiongroup==1.0.0rc9 ; python_version >= "3.8" and python_version <= "3.10"
-ftfy==6.1.1 ; python_version >= "3.8" and python_version < "4"
-idna==3.3 ; python_version >= "3.8" and python_version < "4"
-langid==1.1.6 ; python_version >= "3.8" and python_version < "4.0"
-numpy==1.23.2 ; python_version < "4.0" and python_version >= "3.8"
-pandas==1.4.4 ; python_version >= "3.8" and python_version < "4.0"
-pycountry==22.3.5 ; python_version >= "3.8" and python_version < "4"
-python-dateutil==2.8.2 ; python_version >= "3.8" and python_version < "4.0"
-python-stdnum==1.17 ; python_version >= "3.8" and python_version < "4.0"
-pytz==2022.2.1 ; python_version >= "3.8" and python_version < "4.0"
-requests-cache==0.9.6 ; python_version >= "3.8" and python_version < "4.0"
-requests==2.28.1 ; python_version >= "3.8" and python_version < "4"
-setuptools==65.3.0 ; python_version >= "3.8" and python_version < "4"
-six==1.16.0 ; python_version >= "3.8" and python_version < "4.0"
-spdx-license-list==0.5.2 ; python_version >= "3.8" and python_version < "4.0"
-url-normalize==1.4.3 ; python_version >= "3.8" and python_version < "4.0"
-urllib3==1.26.12 ; python_version >= "3.8" and python_version < "4"
-wcwidth==0.2.5 ; python_version >= "3.8" and python_version < "4"
+appdirs==1.4.4 ; python_version >= "3.9" and python_version < "4.0"
+attrs==22.2.0 ; python_version >= "3.9" and python_version < "4.0"
+cattrs==22.2.0 ; python_version >= "3.9" and python_version < "4.0"
+certifi==2022.12.7 ; python_version >= "3.9" and python_version < "4"
+charset-normalizer==3.0.1 ; python_version >= "3.9" and python_version < "4"
+colorama==0.4.6 ; python_version >= "3.9" and python_version < "4.0"
+country-converter @ git+https://github.com/alanorth/country_converter.git@myanmar-region ; python_version >= "3.9" and python_version < "4.0"
+exceptiongroup==1.1.0 ; python_version >= "3.9" and python_version < "3.11"
+ftfy==6.1.1 ; python_version >= "3.9" and python_version < "4"
+idna==3.4 ; python_version >= "3.9" and python_version < "4"
+langid==1.1.6 ; python_version >= "3.9" and python_version < "4.0"
+numpy==1.24.2 ; python_version < "4.0" and python_version >= "3.9"
+pandas==1.5.3 ; python_version >= "3.9" and python_version < "4.0"
+pycountry @ git+https://github.com/alanorth/pycountry@iso-codes-4.12.0 ; python_version >= "3.9" and python_version < "4.0"
+python-dateutil==2.8.2 ; python_version >= "3.9" and python_version < "4.0"
+python-stdnum==1.18 ; python_version >= "3.9" and python_version < "4.0"
+pytz==2022.7.1 ; python_version >= "3.9" and python_version < "4.0"
+requests-cache==0.9.8 ; python_version >= "3.9" and python_version < "4.0"
+requests==2.28.2 ; python_version >= "3.9" and python_version < "4"
+six==1.16.0 ; python_version >= "3.9" and python_version < "4.0"
+url-normalize==1.4.3 ; python_version >= "3.9" and python_version < "4.0"
+urllib3==1.26.14 ; python_version >= "3.9" and python_version < "4"
+wcwidth==0.2.6 ; python_version >= "3.9" and python_version < "4"

setup.cfg

@@ -1,6 +0,0 @@
-[isort]
-multi_line_output=3
-include_trailing_comma=True
-force_grid_wrap=0
-use_parentheses=True
-line_length=88

setup.py

@@ -14,7 +14,7 @@ install_requires = [
 
 setuptools.setup(
     name="csv-metadata-quality",
-    version="0.6.0",
+    version="0.6.1",
     author="Alan Orth",
     author_email="aorth@mjanja.ch",
     description="A simple, but opinionated CSV quality checking and fixing pipeline for CSVs in the DSpace ecosystem.",
@@ -23,7 +23,6 @@ setuptools.setup(
     long_description_content_type="text/markdown",
     url="https://github.com/alanorth/csv-metadata-quality",
     classifiers=[
-        "Programming Language :: Python :: 3.8",
         "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",
         "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",

tests/test_check.py

@@ -25,7 +25,7 @@ def test_check_valid_issn():
 
     result = check.issn(value)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_invalid_isbn(capsys):
@@ -46,7 +46,7 @@ def test_check_valid_isbn():
 
     result = check.isbn(value)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_missing_date(capsys):
@@ -102,7 +102,7 @@ def test_check_valid_date():
 
     result = check.date(value, field_name)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_suspicious_characters(capsys):
@@ -128,7 +128,7 @@ def test_check_valid_iso639_1_language():
 
     result = check.language(value)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_valid_iso639_3_language():
@@ -138,7 +138,7 @@ def test_check_valid_iso639_3_language():
 
     result = check.language(value)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_invalid_iso639_1_language(capsys):
@@ -249,7 +249,7 @@ def test_check_common_filename_extension():
 
     result = check.filename_extension(value)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_incorrect_iso_639_1_language(capsys):
@@ -305,7 +305,7 @@ def test_check_correct_iso_639_1_language():
 
     result = experimental.correct_language(series, exclude)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_correct_iso_639_3_language():
@@ -321,7 +321,7 @@ def test_check_correct_iso_639_3_language():
 
     result = experimental.correct_language(series, exclude)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_valid_spdx_license_identifier():
@@ -331,7 +331,7 @@ def test_check_valid_spdx_license_identifier():
 
     result = check.spdx_license_identifier(license)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_invalid_spdx_license_identifier(capsys):
@@ -339,7 +339,7 @@ def test_check_invalid_spdx_license_identifier(capsys):
 
     license = "CC-BY-SA"
 
-    result = check.spdx_license_identifier(license)
+    check.spdx_license_identifier(license)
 
     captured = capsys.readouterr()
     assert (
@@ -362,7 +362,7 @@ def test_check_duplicate_item(capsys):
     }
     df = pd.DataFrame(data=d)
 
-    result = check.duplicate_items(df)
+    check.duplicate_items(df)
 
     captured = capsys.readouterr()
     assert (
@@ -379,7 +379,7 @@ def test_check_no_mojibake():
 
     result = check.mojibake(field, field_name)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_mojibake(capsys):
@@ -388,7 +388,7 @@ def test_check_mojibake(capsys):
     field = "CIAT Publicaçao"
     field_name = "dcterms.isPartOf"
 
-    result = check.mojibake(field, field_name)
+    check.mojibake(field, field_name)
 
     captured = capsys.readouterr()
     assert (
@@ -411,7 +411,7 @@ def test_check_doi_field():
 
     result = check.citation_doi(series, exclude)
 
-    assert result == None
+    assert result is None
 
 
 def test_check_doi_only_in_citation(capsys):
@@ -448,7 +448,7 @@ def test_title_in_citation():
 
     result = check.title_in_citation(series, exclude)
 
-    assert result == None
+    assert result is None
 
 
 def test_title_not_in_citation(capsys):
@@ -485,7 +485,7 @@ def test_country_matches_region():
 
     result = check.countries_match_regions(series, exclude)
 
-    assert result == None
+    assert result is None
 
 
 def test_country_not_matching_region(capsys):
@@ -510,5 +510,5 @@ def test_country_not_matching_region(capsys):
     captured = capsys.readouterr()
     assert (
         captured.out
-        == f"{Fore.YELLOW}Missing region ({missing_region}): {Fore.RESET}{title}\n"
+        == f"{Fore.YELLOW}Missing region ({country} → {missing_region}): {Fore.RESET}{title}\n"
     )