mam/audible: null author_info or authors no longer abort the run

CHANGELOG.md

@@ -6,6 +6,12 @@ first release is 3.0.0 because exit codes and the cookie store changed in ways a
 
 ## Unreleased
 
+### Fixed
+- A MAM search hit with `author_info` null (or already a JSON object, or unparsable) aborted `getMAMBook`
+  for every later book. The same run-killing TypeError/KeyError happened in `product2Book` when Audible sent
+  `authors`/`narrators`/`category_ladders` as null or an entry without `name`. Those fields are skipped now,
+  the way `series_info` / `series.sequence` already were.
+
 ## 3.0.4 - 2026-09-22
 
 ### Fixed

myx_audible.py

@@ -132,11 +132,17 @@ def product2Book(product):
         if 'subtitle' in product: book.subtitle=str(product["subtitle"])
         if 'publisher_summary' in product: book.description=str(product["publisher_summary"])
         if 'runtime_length_min' in product: book.length=product["runtime_length_min"]
-        if 'authors' in product: 
-            for author in product["authors"]:
+        if 'authors' in product:
+            # null, a bare string, or an entry without `name` used to TypeError/KeyError in _rankAudible
+            # and abort the run; skip the bad entry the same way series already skips a missing title
+            for author in product["authors"] or []:
+                if not isinstance(author, dict) or not author.get("name"):
+                    continue
                 book.authors.append(myx_classes.Contributor(str(author["name"])))
-        if 'narrators' in product: 
-            for narrator in product["narrators"]:
+        if 'narrators' in product:
+            for narrator in product["narrators"] or []:
+                if not isinstance(narrator, dict) or not narrator.get("name"):
+                    continue
                 book.narrators.append(myx_classes.Contributor(str(narrator["name"])))
         if 'publisher_name' in product: book.publisher=str(product["publisher_name"])
         if 'publication_datetime' in product: book.publishYear=str(product["publication_datetime"])
@@ -150,9 +156,13 @@ def product2Book(product):
                 book.series.append(myx_classes.Series(str(s["title"]), s.get("sequence")))
         if 'language' in product: book.language=str(product ["language"])
         if 'category_ladders' in product:
-            for cl in product["category_ladders"]:
-                for i, item in enumerate(cl["ladder"]):
+            for cl in product["category_ladders"] or []:
+                if not isinstance(cl, dict):
+                    continue
+                for i, item in enumerate(cl.get("ladder") or []):
                     #the first one is genre, the rest are tags
+                    if not isinstance(item, dict) or not item.get("name"):
+                        continue
                     if (i==0):
                         book.genres.append(item["name"])
                     else:

myx_mam.py

@@ -183,6 +183,24 @@ def searchMAM(cfg, titleFilename, authors, extension, refresh=False):
 
     return None
 
+def _mamJsonObject(value):
+    """MAM stores author_info / narrator_info / series_info as a JSON object string. Some answers send
+    the object already decoded, or null / a number / a bare string. Return a dict, or None to skip.
+    Never raise: one bad field used to abort getMAMBook for every later book."""
+    if value is None or value == "":
+        return None
+    if isinstance(value, dict):
+        parsed = value
+    elif isinstance(value, str):
+        try:
+            parsed = json.loads(value)
+        except (TypeError, ValueError):
+            return None
+    else:
+        return None
+    return parsed if isinstance(parsed, dict) else None
+
+
 def getMAMBook(cfg, titleFilename="", authors="", extension="", refresh=False):
     books=[]
     mamBook=searchMAM(cfg, titleFilename, authors, extension, refresh=refresh)
@@ -195,36 +213,37 @@ def getMAMBook(cfg, titleFilename="", authors="", extension="", refresh=False):
                 book.asin=str(b["asin"])
             if 'title' in b: 
                 book.title=str(b["title"])
-            if 'author_info'in b:
-                #format {id:author, id:author}
-                if len(b["author_info"]):
-                    authors = json.loads(b["author_info"])
-                    for author in authors.values():
-                        book.authors.append(myx_classes.Contributor(str(author)))
-            if 'narrator_info'in b:
-                #format {id:narrator, id:narrator}
-                if ((not b["narrator_info"] is None) and len(b["narrator_info"])):
-                    narrators = json.loads(b["narrator_info"])
-                    for narrator in narrators.values():
-                        book.narrators.append(myx_classes.Contributor(str(narrator)))
-            if 'series_info'in b:
+            # format {id:author, id:author}. null / already-decoded object / bad JSON used to TypeError
+            # here (narrator_info and series_info were guarded for null in #24; author_info was not)
+            author_info = _mamJsonObject(b.get("author_info"))
+            if author_info:
+                for author in author_info.values():
+                    if author in (None, ""):
+                        continue
+                    book.authors.append(myx_classes.Contributor(str(author)))
+            narrator_info = _mamJsonObject(b.get("narrator_info"))
+            if narrator_info:
+                for narrator in narrator_info.values():
+                    if narrator in (None, ""):
+                        continue
+                    book.narrators.append(myx_classes.Contributor(str(narrator)))
+            series_info = _mamJsonObject(b.get("series_info"))
+            if series_info:
                 #format {"35598": ["Kat Dubois", "5"]}
-                if ((not b["series_info"] is None) and len(b["series_info"])):
-                    series_info = json.loads(b["series_info"])
-                    for series in series_info.values():
-                        # a value that is not a list (null, a number, a bare string that list() would split into
-                        # letters) or one with no name is not a series entry: skip it rather than abort the run
-                        if not isinstance(series, (list, tuple)):
-                            continue
-                        s=list(series)
-                        if not s or s[0] in (None, ""):
-                            continue
-                        seriesName = str(s[0])
-                        seriesName = seriesName.replace("'", "'")
-                        # a series without a part is ["Name"] or ["Name", null]; s[1] used to IndexError
-                        # and abort the run, or become "None" and file as "Series #None - Title"
-                        part = s[1] if len(s) > 1 else ""
-                        book.series.append(myx_classes.Series(seriesName, part))
+                for series in series_info.values():
+                    # a value that is not a list (null, a number, a bare string that list() would split into
+                    # letters) or one with no name is not a series entry: skip it rather than abort the run
+                    if not isinstance(series, (list, tuple)):
+                        continue
+                    s=list(series)
+                    if not s or s[0] in (None, ""):
+                        continue
+                    seriesName = str(s[0])
+                    seriesName = seriesName.replace("'", "'")
+                    # a series without a part is ["Name"] or ["Name", null]; s[1] used to IndexError
+                    # and abort the run, or become "None" and file as "Series #None - Title"
+                    part = s[1] if len(s) > 1 else ""
+                    book.series.append(myx_classes.Series(seriesName, part))
             if 'lang_code' in b:
                 book.language=myx_utilities.getLanguage((b["lang_code"]))
             if 'my_snatched' in b:

tests/test_mam_throttle.py

@@ -134,6 +134,30 @@ def test_series_info_without_a_part_does_not_abort_the_run(self):
             ("Tripwire", [("Jack Reacher", "3")]),
         ])
 
+    def test_author_info_null_or_already_decoded_does_not_abort_the_run(self):
+        # series_info and narrator_info were guarded for null in #24; author_info still did len(None)
+        # or json.loads(dict) and aborted getMAMBook for every later book. Radio / uncredited torrents
+        # send null; some answers already have the object decoded.
+        with tempfile.TemporaryDirectory() as td:
+            cfg = FakeConfig(td)
+            FakeSession.answer = _Resp(200, "x", {"data": [
+                {"id": 1, "title": "Radio Hour", "my_snatched": 1, "author_info": None},
+                {"id": 2, "title": "Killing Floor", "my_snatched": 1, "author_info": {"1": "Lee Child"}},
+                {"id": 3, "title": "Die Trying", "my_snatched": 1, "author_info": '{"1": "Lee Child", "2": null}'},
+                {"id": 4, "title": "Tripwire", "my_snatched": 1, "author_info": 5},
+                {"id": 5, "title": "The Visitor", "my_snatched": 1, "author_info": "not-json",
+                 "narrator_info": None, "series_info": {"9": ["Jack Reacher", "4"]}},
+            ], "total": 5})
+            with contextlib.redirect_stdout(io.StringIO()):
+                books = myx_mam.getMAMBook(cfg, titleFilename="T.m4b", extension='"m4b"')
+        self.assertEqual([(b.title, [a.name for a in b.authors], [s.name for s in b.series]) for b in books], [
+            ("Radio Hour", [], []),
+            ("Killing Floor", ["Lee Child"], []),
+            ("Die Trying", ["Lee Child"], []),
+            ("Tripwire", [], []),
+            ("The Visitor", [], ["Jack Reacher"]),
+        ])
+
     def test_unsnatched_answer_is_cached_but_filtered(self):
         with tempfile.TemporaryDirectory() as td:
             cfg = FakeConfig(td)

tests/test_matching.py

@@ -766,3 +766,34 @@ def test_ranking_does_not_crash_when_a_hit_omits_sequence(self):
         self.assertIsNotNone(best)
         self.assertEqual(best.asin, "B0SERIES001")
         self.assertEqual([(s.name, s.part) for s in best.series], [("Jack Reacher", "")])
+
+    def test_product_with_null_authors_or_ladders_does_not_abort_ranking(self):
+        # series.sequence was hardened in #24; authors: null, an author without `name`, or
+        # category_ladders: null still TypeError/KeyError'd in product2Book and killed the run
+        import myx_audible
+        p = product("B0AUTHOR01", "Killing Floor", ["Lee Child"], 600)
+        p["authors"] = None
+        p["narrators"] = None
+        p["category_ladders"] = None
+        book = myx_audible.product2Book(p)
+        self.assertEqual(book.title, "Killing Floor")
+        self.assertEqual(book.authors, [])
+        p = product("B0AUTHOR02", "Die Trying", ["Lee Child"], 600)
+        p["authors"] = [{"asin": "B00AUTHOR"}, {"name": "Lee Child"}]
+        p["category_ladders"] = [{"ladder": None}, {"ladder": [{"name": "Mystery"}, {"name": "Thriller"}]}]
+        book = myx_audible.product2Book(p)
+        self.assertEqual([a.name for a in book.authors], ["Lee Child"])
+        self.assertEqual(book.genres, ["Mystery"])
+        self.assertEqual(book.tags, ["Thriller"])
+        with tempfile.TemporaryDirectory() as td:
+            cfg = FakeConfig(td)
+            hit = product("B0AUTHOR01", "Killing Floor", ["Lee Child"], 600)
+            hit["authors"] = None
+            hit["category_ladders"] = None
+            client = FakeAudible(search=[hit])
+            mb = mambook("Killing Floor - Lee Child.m4b",
+                         id3_book("Killing Floor", ["Lee Child"], 600 * 60))
+            best, _ = run(mb, client, cfg)
+        # no authors on the hit: the title gate still accepts it
+        self.assertIsNotNone(best)
+        self.assertEqual(best.asin, "B0AUTHOR01")