alanorth/csv-metadata-quality
1
0
Fork 0
mirror of https://github.com/ilri/csv-metadata-quality.git synced 2024-11-25 23:28:18 +01:00
Code Releases Activity

More ISO 639-1 and ISO 639-3 fixes

Browse Source 
ISO 639-1 uses two-letter codes and ISO 639-3 uses three-letter codes.
Technically there ISO 639-2/T and ISO 639-2/B, which also uses three
letter codes, but those are not supported by the pycountry library
so I won't even worry about them.

See: https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes
This commit is contained in:
Alan Orth 2019-09-26 07:44:39 +03:00
parent ddbe970342
commit 86d4623fd3
Signed by: alanorth
GPG Key ID: 0FB860CC9C45B1B9
2 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
csv_metadata_quality/check.py
View File

@ -165,7 +165,7 @@ def suspicious_characters(field, field_name):
def language(field): def language(field):
"""Check if a language is valid ISO 639-1 or ISO 639-2. """Check if a language is valid ISO 639-1 (alpha 2) or ISO 639-3 (alpha 3).
Prints the value if it is invalid. Prints the value if it is invalid.
""" """
@ -182,7 +182,7 @@ def language(field):
for value in field.split("||"): for value in field.split("||"):
# After splitting, check if language value is 2 or 3 characters so we # After splitting, check if language value is 2 or 3 characters so we
# can check it against ISO 639-1 or ISO 639-2 accordingly. # can check it against ISO 639-1 or ISO 639-3 accordingly.
if len(value) == 2: if len(value) == 2:
if not languages.get(alpha_2=value): if not languages.get(alpha_2=value):
print(f"Invalid ISO 639-1 language: {value}") print(f"Invalid ISO 639-1 language: {value}")
@ -190,7 +190,7 @@ def language(field):
pass pass
elif len(value) == 3: elif len(value) == 3:
if not languages.get(alpha_3=value): if not languages.get(alpha_3=value):
print(f"Invalid ISO 639-2 language: {value}") print(f"Invalid ISO 639-3 language: {value}")
pass pass
else: else:

14
tests/test_check.py
View File

@ -129,7 +129,7 @@ def test_check_suspicious_characters(capsys):
def test_check_valid_iso639_1_language(): def test_check_valid_iso639_1_language():
'''Test valid ISO 639-1 language.''' '''Test valid ISO 639-1 (alpha 2) language.'''
value = 'ja' value = 'ja'
@ -138,8 +138,8 @@ def test_check_valid_iso639_1_language():
assert result == value assert result == value
def test_check_valid_iso639_2_language(): def test_check_valid_iso639_3_language():
'''Test invalid ISO 639-2 language.''' '''Test valid ISO 639-3 (alpha 3) language.'''
value = 'eng' value = 'eng'
@ -149,7 +149,7 @@ def test_check_valid_iso639_2_language():
def test_check_invalid_iso639_1_language(capsys): def test_check_invalid_iso639_1_language(capsys):
'''Test invalid ISO 639-1 language.''' '''Test invalid ISO 639-1 (alpha 2) language.'''
value = 'jp' value = 'jp'
@ -159,15 +159,15 @@ def test_check_invalid_iso639_1_language(capsys):
assert captured.out == f'Invalid ISO 639-1 language: {value}\n' assert captured.out == f'Invalid ISO 639-1 language: {value}\n'
def test_check_invalid_iso639_2_language(capsys): def test_check_invalid_iso639_3_language(capsys):
'''Test invalid ISO 639-2 language.''' '''Test invalid ISO 639-3 (alpha 3) language.'''
value = 'chi' value = 'chi'
check.language(value) check.language(value)
captured = capsys.readouterr() captured = capsys.readouterr()
assert captured.out == f'Invalid ISO 639-2 language: {value}\n' assert captured.out == f'Invalid ISO 639-3 language: {value}\n'
def test_check_invalid_language(capsys): def test_check_invalid_language(capsys):