From 022a41723b2f3e46c30dc271c80cc70eae82994f Mon Sep 17 00:00:00 2001
From: Maximilian Giller <max@giller.dev>
Date: Wed, 19 Feb 2025 06:15:54 +0100
Subject: [PATCH] Refactored parsing of search results and covering all known
 cases for know

---
 src/dvd-compare-python/dvd_compare.py | 63 ++++++++++++++++-----------
 1 file changed, 37 insertions(+), 26 deletions(-)

diff --git a/src/dvd-compare-python/dvd_compare.py b/src/dvd-compare-python/dvd_compare.py
index fd28100..e811410 100644
--- a/src/dvd-compare-python/dvd_compare.py
+++ b/src/dvd-compare-python/dvd_compare.py
@@ -57,37 +57,47 @@ class DvdCompare:
         # Parse QoL values
         result.fid = int(result.url.split("?fid=")[1])
 
-        label_matches = re.match(
-            r"^(?P<titles>(?:[^()]+\s(?:\((?:The|A|An)\))?(?:(?P<formats>(?:\([^()]+\)\s?)*)\s)?(?:\s?AKA\s?)?)+)\s?\((?P<years>(?P<year>[0-9]{4})(?:[-,][0-9]{4})?)\)$",
-            result.label.replace("–", "-"),
-            re.IGNORECASE,
-        )
+        result.formats = []
+        result.titles = []
+        for part in (
+            result.label.replace("–", "-").replace("\x96–", "-").split(" AKA ")
+        ):  # Iterate over alternative titles and collect information
+            part_matches = re.match(
+                r"^(?P<title>[^()\n]*[^()\s])(?:\s\((?P<prefix>The|A|An)\)(?P<suffix>[^()\n]+)?)?(?P<formats> ?(?:\([^()]*[a-z]+[^()]*\)\s?)+)?(?P<years>(?: ?\((?P<range>(?P<year>[0-9]{4})(?:[-,/][0-9]{1,2}(?:[0-9]{2})?)?)\)))?",
+                part,
+                re.IGNORECASE,
+            )
 
-        if not label_matches:
+            if part_matches is None:
+                raise UnknownFormatError(
+                    f"Label part in unknown format that could not be parsed: {part}"
+                )
+
+            # Title
+            title = f"{part_matches['title']}{part_matches['suffix']}".strip()
+            if part_matches["prefix"]:
+                title = f"{part_matches['prefix']} {title}"
+            result.titles.append(title)
+
+            # Formats
+            if part_matches["formats"]:
+                result.formats = [
+                    m
+                    for m in re.findall(
+                        r"\((?P<format>[^()]+)\)", part_matches["formats"]
+                    )
+                ]
+
+            # Years
+            if part_matches["years"]:
+                result.year = int(part_matches["year"])
+                result.years = part_matches["years"]
+
+        if len(result.titles) <= 0:
             raise UnknownFormatError(
                 f"Label in unknown format that could not be parsed: {result.label}"
             )
 
-        result.formats = []
-        if label_matches["formats"]:
-            result.formats = [
-                m
-                for m in re.findall(r"\((?P<format>[^()]+)\)", label_matches["formats"])
-            ]
-        result.year = int(label_matches["year"])
-        result.years = label_matches["years"]
-        result.titles = []
-        for raw_title in label_matches["titles"].strip().split(" AKA "):
-            title_match = re.match(
-                r"(?P<title>[^()]+[^() ])(?:\s\((?P<prefix>The|A|An)\))?(?:\s\([^()]+\))?",
-                raw_title,
-                re.IGNORECASE,
-            )
-            title = title_match["title"]
-            if title_match["prefix"]:
-                title = f"{title_match['prefix']} {title}"
-            result.titles.append(title)
-
         return result
 
     async def search_async(
@@ -157,6 +167,7 @@ if __name__ == "__main__":
         results = await DvdCompare.search_async("The")
         if results:
             print(results[0])
+            print(f"And {len(results)} additional results")
         else:
             print("No results found.")