added a check for duplicate references which will prevent loading the…

… data into the web app
D-PLACE · Aug 15, 2024 · fe3229d · fe3229d
1 parent 05e38b8
commit fe3229d
Show file tree

Hide file tree

Showing 3 changed files with 7 additions and 1 deletion.
diff --git a/src/pydplace/commands/check.py b/src/pydplace/commands/check.py
@@ -51,5 +51,10 @@ def run(args):
             print(colored('{} not found'.format(fname), 'red'))
             passed = False
 
+    for dp in ds.cldf_reader().objects('ValueTable'):
+        if len(dp.cldf.source) != len(set(dp.cldf.source)):
+            print(colored('{} duplicate reference'.format(dp.cldf.source), 'red'))
+            passed = False
+
     print(colored('OK' if passed else 'FAIL', 'green' if passed else 'red', attrs={'bold'}))
     return 0
diff --git a/tests/dataset_without_societies/cldf/data.csv b/tests/dataset_without_societies/cldf/data.csv
@@ -10,4 +10,4 @@ ID,Soc_ID,Var_ID,Value,Code_ID,Comment,Source,sub_case,year,source_coded_data,ad
 9,WNAI3,WNAI12,Absent,WNAI12-1,,drucker1940;drucker1950;drucker1965;krause1956;murdock1934a;murdock1934b;murdock1936a;rosmanandrubel1971;swanton1909,,,"wn01.dat (World Cultures CD-ROM 2011, folder 10#2)",
 10,WNAI1,WNAI13,Absent,WNAI13-1,,delaguna1960;barnett1968;drucker1950;druckerandheizer1967;garfield1947;jones1914;krause1956;olson1936b;olson1967;rosmanandrubel1971;swanton1908,,,"wn01.dat (World Cultures CD-ROM 2011, folder 10#2)",
 11,WNAI2,WNAI13,Absent,WNAI13-1,,delaguna1960;barnett1968;drucker1950;druckerandheizer1967;garfield1947;jones1914;krause1956;olson1936b;olson1967;rosmanandrubel1971;swanton1908,,,"wn01.dat (World Cultures CD-ROM 2011, folder 10#2)",
-12,WNAI3,WNAI13,Absent,WNAI13-1,,drucker1940;drucker1950;drucker1965;krause1956;murdock1934a;murdock1934b;murdock1936a;rosmanandrubel1971;swanton1909,,,"wn01.dat (World Cultures CD-ROM 2011, folder 10#2)",
+12,WNAI3,WNAI13,Absent,WNAI13-1,,drucker1940;drucker1940;drucker1950;drucker1965;krause1956;murdock1934a;murdock1934b;murdock1936a;rosmanandrubel1971;swanton1909,,,"wn01.dat (World Cultures CD-ROM 2011, folder 10#2)",
diff --git a/tests/test_cli.py b/tests/test_cli.py
@@ -33,6 +33,7 @@ def test_check(capsys, dataset_with_societies, dataset_without_societies):
     out, _ = capsys.readouterr()
     assert 'must reference' in out
     assert 'Re-run' in out
+    assert 'duplicate reference' in out
 
 
 def test_readme(dataset_with_societies, dataset_without_societies):