1
0
mirror of https://github.com/ilri/csv-metadata-quality.git synced 2024-12-22 12:12:18 +01:00
csv-metadata-quality/tests/test_check.py

515 lines
13 KiB
Python
Raw Normal View History

# SPDX-License-Identifier: GPL-3.0-only
2020-01-15 10:41:31 +01:00
import pandas as pd
from colorama import Fore
2020-01-15 10:41:31 +01:00
import csv_metadata_quality.check as check
import csv_metadata_quality.experimental as experimental
2019-07-28 16:47:28 +02:00
def test_check_invalid_issn(capsys):
2019-09-26 13:02:51 +02:00
"""Test checking invalid ISSN."""
2019-09-26 13:02:51 +02:00
value = "2321-2302"
check.issn(value)
2019-07-28 16:47:28 +02:00
captured = capsys.readouterr()
assert captured.out == f"{Fore.RED}Invalid ISSN: {Fore.RESET}{value}\n"
def test_check_valid_issn():
2019-09-26 13:02:51 +02:00
"""Test checking valid ISSN."""
2019-09-26 13:02:51 +02:00
value = "0024-9319"
result = check.issn(value)
assert result is None
def test_check_invalid_isbn(capsys):
2019-09-26 13:02:51 +02:00
"""Test checking invalid ISBN."""
2019-09-26 13:02:51 +02:00
value = "99921-58-10-6"
check.isbn(value)
captured = capsys.readouterr()
assert captured.out == f"{Fore.RED}Invalid ISBN: {Fore.RESET}{value}\n"
def test_check_valid_isbn():
2019-09-26 13:02:51 +02:00
"""Test checking valid ISBN."""
2019-09-26 13:02:51 +02:00
value = "99921-58-10-7"
result = check.isbn(value)
assert result is None
def test_check_missing_date(capsys):
2019-09-26 13:02:51 +02:00
"""Test checking missing date."""
value = None
2019-09-26 13:02:51 +02:00
field_name = "dc.date.issued"
2019-08-21 14:34:52 +02:00
check.date(value, field_name)
captured = capsys.readouterr()
assert captured.out == f"{Fore.RED}Missing date ({field_name}).{Fore.RESET}\n"
def test_check_multiple_dates(capsys):
2019-09-26 13:02:51 +02:00
"""Test checking multiple dates."""
2019-09-26 13:02:51 +02:00
value = "1990||1991"
2019-09-26 13:02:51 +02:00
field_name = "dc.date.issued"
2019-08-21 14:34:52 +02:00
check.date(value, field_name)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.RED}Multiple dates not allowed ({field_name}): {Fore.RESET}{value}\n"
)
def test_check_invalid_date(capsys):
2019-09-26 13:02:51 +02:00
"""Test checking invalid ISO8601 date."""
2019-09-26 13:02:51 +02:00
value = "1990-0"
2019-09-26 13:02:51 +02:00
field_name = "dc.date.issued"
2019-08-21 14:34:52 +02:00
check.date(value, field_name)
captured = capsys.readouterr()
assert (
captured.out == f"{Fore.RED}Invalid date ({field_name}): {Fore.RESET}{value}\n"
)
def test_check_valid_date():
2019-09-26 13:02:51 +02:00
"""Test checking valid ISO8601 date."""
2019-09-26 13:02:51 +02:00
value = "1990"
2019-09-26 13:02:51 +02:00
field_name = "dc.date.issued"
2019-08-21 14:34:52 +02:00
result = check.date(value, field_name)
assert result is None
def test_check_suspicious_characters(capsys):
2019-09-26 13:02:51 +02:00
"""Test checking for suspicious characters."""
2019-09-26 13:02:51 +02:00
value = "foreˆt"
2019-09-26 13:02:51 +02:00
field_name = "dc.contributor.author"
check.suspicious_characters(value, field_name)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.YELLOW}Suspicious character ({field_name}): {Fore.RESET}ˆt\n"
)
def test_check_valid_iso639_1_language():
2019-09-26 13:02:51 +02:00
"""Test valid ISO 639-1 (alpha 2) language."""
2019-09-26 13:02:51 +02:00
value = "ja"
result = check.language(value)
assert result is None
def test_check_valid_iso639_3_language():
2019-09-26 13:02:51 +02:00
"""Test valid ISO 639-3 (alpha 3) language."""
2019-09-26 13:02:51 +02:00
value = "eng"
result = check.language(value)
assert result is None
def test_check_invalid_iso639_1_language(capsys):
2019-09-26 13:02:51 +02:00
"""Test invalid ISO 639-1 (alpha 2) language."""
2019-09-26 13:02:51 +02:00
value = "jp"
check.language(value)
captured = capsys.readouterr()
assert (
captured.out == f"{Fore.RED}Invalid ISO 639-1 language: {Fore.RESET}{value}\n"
)
def test_check_invalid_iso639_3_language(capsys):
2019-09-26 13:02:51 +02:00
"""Test invalid ISO 639-3 (alpha 3) language."""
2019-09-26 13:02:51 +02:00
value = "chi"
check.language(value)
captured = capsys.readouterr()
assert (
captured.out == f"{Fore.RED}Invalid ISO 639-3 language: {Fore.RESET}{value}\n"
)
def test_check_invalid_language(capsys):
2019-09-26 13:02:51 +02:00
"""Test invalid language."""
2019-09-26 13:02:51 +02:00
value = "Span"
check.language(value)
captured = capsys.readouterr()
assert captured.out == f"{Fore.RED}Invalid language: {Fore.RESET}{value}\n"
def test_check_invalid_agrovoc(capsys):
"""Test invalid AGROVOC subject. Invalid values *will not* be dropped."""
valid_agrovoc = "LIVESTOCK"
invalid_agrovoc = "FOREST"
value = f"{valid_agrovoc}||{invalid_agrovoc}"
field_name = "dcterms.subject"
drop = False
new_value = check.agrovoc(value, field_name, drop)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.RED}Invalid AGROVOC ({field_name}): {Fore.RESET}{invalid_agrovoc}\n"
)
assert new_value == value
def test_check_invalid_agrovoc_dropped(capsys):
"""Test invalid AGROVOC subjects. Invalid values *will* be dropped."""
valid_agrovoc = "LIVESTOCK"
invalid_agrovoc = "FOREST"
value = f"{valid_agrovoc}||{invalid_agrovoc}"
field_name = "dcterms.subject"
drop = True
new_value = check.agrovoc(value, field_name, drop)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.GREEN}Dropping invalid AGROVOC ({field_name}): {Fore.RESET}{invalid_agrovoc}\n"
)
assert new_value == valid_agrovoc
def test_check_valid_agrovoc():
2019-09-26 13:02:51 +02:00
"""Test valid AGROVOC subject."""
2019-09-26 13:02:51 +02:00
value = "FORESTS"
field_name = "dcterms.subject"
drop = False
result = check.agrovoc(value, field_name, drop)
assert result == "FORESTS"
def test_check_uncommon_filename_extension(capsys):
2019-09-26 13:02:51 +02:00
"""Test uncommon filename extension."""
2019-09-26 13:02:51 +02:00
value = "file.pdf.lck"
check.filename_extension(value)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.YELLOW}Filename with uncommon extension: {Fore.RESET}{value}\n"
)
def test_check_common_filename_extension():
2019-09-26 13:02:51 +02:00
"""Test common filename extension."""
2019-09-26 13:02:51 +02:00
value = "file.pdf"
result = check.filename_extension(value)
assert result is None
def test_check_incorrect_iso_639_1_language(capsys):
2019-09-26 13:02:51 +02:00
"""Test incorrect ISO 639-1 language, as determined by comparing the item's language field with the actual language predicted in the item's title."""
2019-09-26 13:02:51 +02:00
title = "A randomised vaccine field trial in Kenya demonstrates protection against wildebeest-associated malignant catarrhal fever in cattle"
language = "es"
exclude = []
# Create a dictionary to mimic Pandas series
2019-09-26 13:02:51 +02:00
row = {"dc.title": title, "dc.language.iso": language}
series = pd.Series(row)
experimental.correct_language(series, exclude)
captured = capsys.readouterr()
2019-09-26 13:02:51 +02:00
assert (
captured.out
== f"{Fore.YELLOW}Possibly incorrect language {language} (detected en): {Fore.RESET}{title}\n"
2019-09-26 13:02:51 +02:00
)
def test_check_incorrect_iso_639_3_language(capsys):
2019-09-26 13:02:51 +02:00
"""Test incorrect ISO 639-3 language, as determined by comparing the item's language field with the actual language predicted in the item's title."""
2019-09-26 13:02:51 +02:00
title = "A randomised vaccine field trial in Kenya demonstrates protection against wildebeest-associated malignant catarrhal fever in cattle"
language = "spa"
exclude = []
# Create a dictionary to mimic Pandas series
2019-09-26 13:02:51 +02:00
row = {"dc.title": title, "dc.language.iso": language}
series = pd.Series(row)
experimental.correct_language(series, exclude)
captured = capsys.readouterr()
2019-09-26 13:02:51 +02:00
assert (
captured.out
== f"{Fore.YELLOW}Possibly incorrect language {language} (detected eng): {Fore.RESET}{title}\n"
2019-09-26 13:02:51 +02:00
)
def test_check_correct_iso_639_1_language():
2019-09-26 13:02:51 +02:00
"""Test correct ISO 639-1 language, as determined by comparing the item's language field with the actual language predicted in the item's title."""
2019-09-26 13:02:51 +02:00
title = "A randomised vaccine field trial in Kenya demonstrates protection against wildebeest-associated malignant catarrhal fever in cattle"
language = "en"
exclude = []
# Create a dictionary to mimic Pandas series
2019-09-26 13:02:51 +02:00
row = {"dc.title": title, "dc.language.iso": language}
series = pd.Series(row)
result = experimental.correct_language(series, exclude)
assert result is None
def test_check_correct_iso_639_3_language():
2019-09-26 13:02:51 +02:00
"""Test correct ISO 639-3 language, as determined by comparing the item's language field with the actual language predicted in the item's title."""
2019-09-26 13:02:51 +02:00
title = "A randomised vaccine field trial in Kenya demonstrates protection against wildebeest-associated malignant catarrhal fever in cattle"
language = "eng"
exclude = []
# Create a dictionary to mimic Pandas series
2019-09-26 13:02:51 +02:00
row = {"dc.title": title, "dc.language.iso": language}
series = pd.Series(row)
result = experimental.correct_language(series, exclude)
assert result is None
def test_check_valid_spdx_license_identifier():
"""Test valid SPDX license identifier."""
license = "CC-BY-SA-4.0"
result = check.spdx_license_identifier(license)
assert result is None
def test_check_invalid_spdx_license_identifier(capsys):
"""Test invalid SPDX license identifier."""
license = "CC-BY-SA"
check.spdx_license_identifier(license)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.YELLOW}Non-SPDX license identifier: {Fore.RESET}{license}\n"
)
def test_check_duplicate_item(capsys):
"""Test item with duplicate title, type, and date."""
item_title = "Title"
item_type = "Report"
item_date = "2021-03-17"
d = {
"dc.title": [item_title, item_title],
"dcterms.type": [item_type, item_type],
"dcterms.issued": [item_date, item_date],
}
df = pd.DataFrame(data=d)
check.duplicate_items(df)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.YELLOW}Possible duplicate (dc.title): {Fore.RESET}{item_title}\n"
)
def test_check_no_mojibake():
"""Test string with no mojibake."""
field = "CIAT Publicaçao"
field_name = "dcterms.isPartOf"
result = check.mojibake(field, field_name)
assert result is None
def test_check_mojibake(capsys):
"""Test string with mojibake."""
field = "CIAT Publicaçao"
field_name = "dcterms.isPartOf"
check.mojibake(field, field_name)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.YELLOW}Possible encoding issue ({field_name}): {Fore.RESET}{field}\n"
)
def test_check_doi_field():
"""Test an item with a DOI field."""
doi = "https://doi.org/10.1186/1743-422X-9-218"
citation = "Orth, A. 2021. Testing all the things. doi: 10.1186/1743-422X-9-218"
# Emulate a column in a transposed dataframe (which is just a series), with
2021-12-05 14:58:25 +01:00
# the citation and a DOI field.
2021-10-06 21:10:26 +02:00
d = {"cg.identifier.doi": doi, "dcterms.bibliographicCitation": citation}
series = pd.Series(data=d)
exclude = []
result = check.citation_doi(series, exclude)
assert result is None
def test_check_doi_only_in_citation(capsys):
"""Test an item with a DOI in its citation, but no DOI field."""
citation = "Orth, A. 2021. Testing all the things. doi: 10.1186/1743-422X-9-218"
exclude = []
# Emulate a column in a transposed dataframe (which is just a series), with
# an empty DOI field and a citation containing a DOI.
2021-10-06 21:10:26 +02:00
d = {"cg.identifier.doi": None, "dcterms.bibliographicCitation": citation}
series = pd.Series(data=d)
check.citation_doi(series, exclude)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.YELLOW}DOI in citation, but missing a DOI field: {Fore.RESET}{citation}\n"
)
def test_title_in_citation():
"""Test an item with its title in the citation."""
title = "Testing all the things"
citation = "Orth, A. 2021. Testing all the things."
exclude = []
# Emulate a column in a transposed dataframe (which is just a series), with
# the title and citation.
d = {"dc.title": title, "dcterms.bibliographicCitation": citation}
series = pd.Series(data=d)
result = check.title_in_citation(series, exclude)
assert result is None
def test_title_not_in_citation(capsys):
"""Test an item with its title missing from the citation."""
title = "Testing all the things"
citation = "Orth, A. 2021. Testing all teh things."
exclude = []
# Emulate a column in a transposed dataframe (which is just a series), with
# the title and citation.
d = {"dc.title": title, "dcterms.bibliographicCitation": citation}
series = pd.Series(data=d)
check.title_in_citation(series, exclude)
captured = capsys.readouterr()
assert (
captured.out
== f"{Fore.YELLOW}Title is not present in citation: {Fore.RESET}{title}\n"
)
def test_country_matches_region():
"""Test an item with regions matching its country list."""
country = "Kenya"
region = "Eastern Africa"
exclude = []
# Emulate a column in a transposed dataframe (which is just a series)
d = {"cg.coverage.country": country, "cg.coverage.region": region}
series = pd.Series(data=d)
result = check.countries_match_regions(series, exclude)
assert result is None
def test_country_not_matching_region(capsys):
"""Test an item with regions not matching its country list."""
title = "Testing an item with no matching region."
country = "Kenya"
region = ""
missing_region = "Eastern Africa"
exclude = []
# Emulate a column in a transposed dataframe (which is just a series)
d = {
"dc.title": title,
"cg.coverage.country": country,
"cg.coverage.region": region,
}
series = pd.Series(data=d)
check.countries_match_regions(series, exclude)
captured = capsys.readouterr()
assert (
captured.out
2022-11-28 20:12:17 +01:00
== f"{Fore.YELLOW}Missing region ({country} → {missing_region}): {Fore.RESET}{title}\n"
)