Refactoring

renamer.py

@@ -65,11 +65,32 @@ def main():
 
 
 def process_dir(dir_path):
+    """
+    Process dir with media files.
+    If this dir is season of some tv show -> process them as season.
+    Otherwise -> process each file individually.
+    """
     for fname in os.listdir(dir_path):
         fpath = os.path.join(dir_path, fname)
         process_file(fpath)
 
 
+def generate_pretty_name(parsed_title):
+    """ Create file name from parsed chunks. """
+    chunk_order = [k for k, _ in PATTERNS]
+    chunk_order = ["name"] + chunk_order
+    ep_idx = chunk_order.index("episode") + 1
+    chunk_order = chunk_order[:ep_idx] + ["episode_name"] + chunk_order[ep_idx:]
+
+    result = []
+    for chunk_type in chunk_order:
+        if not parsed_title.get(chunk_type, []):
+            continue
+        result.append(".".join(parsed_title[chunk_type]))
+    result = ".".join(result)
+    return result
+
+
 def process_file(fpath):
     # process only files
     if not os.path.isfile(fpath):
@@ -85,25 +106,15 @@ def process_file(fpath):
         return
 
     parsed_title = parse_title(title)
+    pretty_title = generate_pretty_name(parsed_title)
+    pretty_title += ".%s" % ext
 
-    # create file name from parsed chunks
+    if pretty_title != fname:
-    chunk_order = [k for k, _ in PATTERNS]
+        _lg.warning("%s -> %s", fname, pretty_title)
-    chunk_order = ["name"] + chunk_order
-    episode_idx = chunk_order.index("episode") + 1
-    chunk_order = chunk_order[:episode_idx] + ["episode_name"] + chunk_order[episode_idx:]
-    result = []
-    for chunk_type in chunk_order:
-        if not parsed_title.get(chunk_type, []):
-            continue
-        result.append(".".join(parsed_title[chunk_type]))
-    result.append(ext)
-    result = ".".join(result)
-
-    if result != fname:
-        _lg.warning("%s -> %s", fname, result)
 
 
 def _get_parsed_title_dict(chunk_list, chunk_map):
+    """ Get {chunk_type: [chunk_value_1, ...,  chunk_value_n]} dictionary. """
     p_title = collections.defaultdict(list)
     for idx, chunk in enumerate(chunk_list):
         chunk_type = chunk_map[idx]
@@ -112,7 +123,7 @@ def _get_parsed_title_dict(chunk_list, chunk_map):
 
 
 def _guess_combined(chunk_values, chunk_map):
-    """ Try to combine unknown chunks in pairs and parse them """
+    """ Try to combine unknown chunks in pairs and parse them. """
     is_changed = False
     p_title = _get_parsed_title_dict(chunk_values, chunk_map)
     if len(p_title["unknown"]) < 2:
@@ -165,14 +176,14 @@ def parse_title(title):
     # remove non-word chunks (like single hyphens), but leave ampersands (&)
     chunk_values = list(filter(lambda ch: re.search(r"(\w|&)+", ch), chunk_values))
 
+    chunk_map = []  # list of chunk_types
     # parse each chunk
-    chunk_map = []
     for ch_value in chunk_values:
         chunk_map.append(guess_part(ch_value))
 
     _, chunk_values, chunk_map = _guess_combined(chunk_values, chunk_map)
 
-    # # try to parse unknown chunks, replacing all hyphens in them with dots
+    # try to parse unknown chunks, replacing all hyphens in them with dots
     p_title = _get_parsed_title_dict(chunk_values, chunk_map)
     is_changed = False
     if p_title.get("unknown"):
@@ -230,11 +241,12 @@ def parse_title(title):
     return dict(p_title)
 
 
-def guess_part(fname_part):
+def guess_part(chunk_value):
-    for pat_type, pattern in PATTERNS:
+    """ Return chunk type for given chunk value. """
+    for chunk_type, pattern in PATTERNS:
         full_match_pat = r"^" + pattern + r"$"
-        if re.match(full_match_pat, fname_part, flags=re.I):
+        if re.match(full_match_pat, chunk_value, flags=re.I):
-            return pat_type
+            return chunk_type
     raise RuntimeError("unhandled pattern type")