smarter automatic collation of textless versions

dlibrary/dlibrary.py

@@ -24,6 +24,8 @@ DLSITE_ID_REGEX = re.compile('^[BR]J[0-9]+$')
 FANZA_ID_REGEX = re.compile('^d_[0-9]+$')
 FAKKU_ID_REGEX = re.compile('.*_FAKKU$')
 
+TEXTLESS_REGEX = re.compile('(台詞|セリフ)(な|無)し|notext|textless')
+
 IMAGE_FILE_EXTENSIONS = ['.png', '.jpg', '.jpeg', '.gif', '.tiff']
 
 IGNOREABLE_FILES = ['Thumbs.db', '__MACOSX', '.DS_Store']
@@ -191,6 +193,9 @@ def link_pdf(src, dest, start_index=0):
                 f.write(image["image"])
 
 def complete_prefix_number_ordering(entries):
+    if len(entries) == 1:
+        return entries
+
     matches = reversed(list(NUMBER_REGEX.finditer(entries[0].name)))
     for m in matches:
         pos = m.start()
@@ -266,6 +271,19 @@ def collate(args):
 
         if all(entry.is_file() and entry.suffix.lower() in IMAGE_FILE_EXTENSIONS for entry in entries):
             ordering = complete_prefix_number_ordering(entries)
+            if not ordering:
+                with_text = []
+                textless = []
+                for entry in entries:
+                    if TEXTLESS_REGEX.search(entry.name):
+                        textless.append(entry)
+                    else:
+                        with_text.append(entry)
+                if with_text and textless:
+                    with_text_ordering = complete_prefix_number_ordering(with_text)
+                    textless_ordering = complete_prefix_number_ordering(textless)
+                    if with_text_ordering and textless_ordering:
+                        ordering = with_text_ordering + textless_ordering
             if ordering:
                 print(f'Symlinking image files for {work_id}')
                 link_ordered_files(ordering, collation_dir)