Fix TypeError on RSS downloads and preserve Unicode in titles

.github/workflows/tasks.yml

@@ -17,8 +17,8 @@ jobs:
   tests:
     uses: nzbgetcom/nzbget-extensions/.github/workflows/python-tests.yml@main
     with:
-      python-versions: "3.6 3.7 3.8 3.9 3.10 3.11 3.12"
-      supported-python-versions: "3.8 3.9 3.10 3.11 3.12"
+      python-versions: "3.8 3.9 3.10 3.11 3.12 3.13 3.14"
+      supported-python-versions: "3.8 3.9 3.10 3.11 3.12 3.13 3.14"
       test-script: testsort.py
       debug: true
 

main.py

@@ -32,12 +32,7 @@
 import shutil
 import guessit
 import difflib
-import locale
 
-try:
-    unicode
-except NameError:
-    unicode = str
 
 # Exit codes used by NZBGet
 POSTPROCESS_SUCCESS = 93
@@ -424,6 +419,8 @@ def get_titles(name, titleing=False):
     """
 
     # make valid filename
+    if isinstance(name, list):
+        name = ' '.join(name)
     title = re.sub(r"[\"\:\?\*\\\/\<\>\|]", " ", name)
 
     if titleing:
@@ -461,18 +458,8 @@ def get_titles(name, titleing=False):
 
 
 def titler(p):
-    """title() replacement
-    Python's title() fails with Latin-1, so use Unicode detour.
-    """
-    if isinstance(p, unicode):
-        return p.title()
-    elif gUTF:
-        try:
-            return p.decode("utf-8").title().encode("utf-8")
-        except:
-            return p.decode("latin-1", "replace").title().encode("latin-1", "replace")
-    else:
-        return p.decode("latin-1", "replace").title().encode("latin-1", "replace")
+    """Title-case a string, preserving Unicode characters."""
+    return p.title()
 
 
 def replace_word(input, one, two):
@@ -553,16 +540,6 @@ def strip_all(x):
     return os.path.normpath("/".join([strip_all(x) for x in f]))
 
 
-gUTF = False
-try:
-    if sys.platform == "darwin":
-        gUTF = True
-    else:
-        gUTF = locale.getlocale()[1] == "UTF-8"
-except:
-    # Incorrect locale implementation, assume the worst
-    gUTF = False
-
 # END * From SABnzbd+ * END
 
 
@@ -955,23 +932,23 @@ def guess_info(filename):
         print("Guessing: %s" % guessfilename)
 
     guess = guessit.api.guessit(
-        unicode(guessfilename), {"allowed_languages": [], "allowed_countries": []}
+        guessfilename, {"allowed_languages": [], "allowed_countries": []}
     )
 
     if verbose:
         print(guess)
 
+    # workaround for titles returned as lists by guessit (e.g., RSS downloads)
+    if "title" in guess and isinstance(guess["title"], list):
+        guess["title"] = ' '.join(guess["title"])
+
     # workaround for titles starting with numbers (part 2)
     if pad_start_digits:
         guess["title"] = guess["title"][1:]
         if guess["title"] == "":
             guess["title"] = os.path.splitext(os.path.basename(guessfilename))[0][1:]
             if verbose:
                 print("use filename as title for recovery")
-
-    # workaround for titles with 2 numbers with a dash between them (which guessit has problems with)
-    if "title" in guess and isinstance(guess["title"], list):
-        guess["title"] = ' '.join(guess["title"])
 
     # fix some strange guessit guessing:
     # if guessit doesn't find a year in the file name it thinks it is episode,

testdata.json

@@ -223,6 +223,18 @@
     "NZBPO_SERIESFORMAT": "%sn\\Season %0s\\%sn - S%0sE%0e - %en [%qss][%qf][%qrg]",
     "NZBPO_MULTIPLEEPISODES": "range",
     "NZBPO_EPISODESEPARATOR": "-E"
+  },
+  {
+    "id": "spec-char-jp",
+    "INPUTFILE": "進撃の巨人.S01E01.720p.BluRay.x264-SiNNERS.mkv",
+    "OUTPUTFILE": "/series/進撃の巨人/Season 1/進撃の巨人 - S01E01.mkv",
+    "NZBPO_SERIESFORMAT": "%sn/Season %s/%sn - S%0sE%0e.%ext"
+  },
+  {
+    "id": "spec-char-accents",
+    "INPUTFILE": "José.S01E01.720p.BluRay.x264-SiNNERS.mkv",
+    "OUTPUTFILE": "/series/José/Season 1/José - S01E01.mkv",
+    "NZBPO_SERIESFORMAT": "%sn/Season %s/%sn - S%0sE%0e.%ext"
   }
 ]