Simplify the script extracting long format from specificities measured in TXM

visualisation/LongFormatFromTXM.py

@@ -4,19 +4,16 @@ from GEODE import tabular
 import pandas
 import sys
 
-def extractColumn(rows, name):
-    rows['Spécificité'] = rows[name]
-    rows['lemme'] = name
-    return rows[['Spécificité', 'lemme']]
+def toLongFormat(txmData):
+    lemmas = txmData.set_index(txmData.columns[0]).transpose()
+    lemmas.index.name = lemmas.columns.name
+    lemmas.columns.name = 'lemme'
+    specificities = lemmas.stack()
+    specificities.name = 'Spécificité'
+    return pandas.DataFrame(specificities)
 
 def convert(inputTSV, outputTSV):
-    txmData = tabular(inputTSV)
-    rows = txmData.set_index(txmData.columns[0]).transpose()
-    rows.index.name = rows.columns.name
-    rows.columns.name = None
-    longFormat = pandas.concat([extractColumn(rows.copy(), c)
-                                for c in rows.columns], axis=0)
-    longFormat.to_csv(outputTSV, sep='\t')
+    toLongFormat(tabular(inputTSV)).to_csv(outputTSV, sep='\t')
 
 if __name__ == '__main__':
     convert(*sys.argv[1:])