Automatic log file comparison between parsers

tetras_extraction/macao_12/script/extract.py

+import filecmp
 from pprint import pprint
 from typing import Optional
 
@@ -122,18 +123,32 @@ def parse_manifest_rec(
         extract_mosetp.parse_mosetp(graph, f"{SOURCE_DIR}/sco/{id}.html", id)
 
 
-import extract_page
+def compare_files(f1, f2):
+    print(
+        "Files {} and {} {}.".format(
+            f1, f2, "are identical" if filecmp.cmp(f1, f2) else "differ"
+        )
+    )
 
 
 def main():
     g = create_graph()
+
+    # Create or reset debug log files for all activity parsers, to compare their
+    # results afterwards
+    parsers = ("Match", "Xpath", "Regex")
+    logfiles = [f"/tmp/{p}Parser_debuglog.txt" for p in parsers]
+    for logfile in logfiles:
+        with open(logfile, "w") as f:
+            print("", file=f)
+
     parse_manifest(g)
     export_graph(g)
-    # extract_page.parse_page(
-    #     g,
-    #     f"{SOURCE_DIR}/contenu/pages/pg60.html",
-    #     "pg60",
-    # )
+
+    # Compare log files 2 by 2
+    compare_files(logfiles[0], logfiles[1])
+    compare_files(logfiles[0], logfiles[2])
+    compare_files(logfiles[1], logfiles[2])
 
 
 if __name__ == "__main__":

tetras_extraction/macao_12/script/extract_page.py

@@ -301,7 +301,7 @@ def parse_page(graph: Graph, filepath: str, id: str):
 
     # Try different parsers, each writing to a different file to compare their results
     for parser in [XpathParser(), MatchParser(graph, id), RegexParser(graph, id)]:
-        with open(f"/tmp/{str(parser)}.txt", "a") as f:
+        with open(f"/tmp/{str(parser)}_debuglog.txt", "a") as f:
             print(f"{id:8}", end="", file=f)
             try:
                 parser.parse(js, output=f)