Add experimental word splitting to the filename parser

Adds a global setting as well as a setting that is only in effect
during auto-tagging

comicapi/comicarchive.py

@@ -16,6 +16,7 @@
 
 import zipfile
 import os
+import os.path
 import struct
 import sys
 import tempfile
@@ -23,6 +24,7 @@ import subprocess
 import platform
 import time
 import io
+import wordninja
 
 import natsort
 from PyPDF2 import PdfFileReader
@@ -1080,11 +1082,14 @@ class ComicArchive:
                         data = self.getPage(idx)
                         p['ImageSize'] = str(len(data))
 
-    def metadataFromFilename(self, parse_scan_info=True):
+    def metadataFromFilename(self, parse_scan_info=True, split_words=False):
         metadata = GenericMetadata()
 
         fnp = FileNameParser()
-        fnp.parseFilename(self.path)
+        filename = self.path
+        if split_words:
+            filename = " ".join(wordninja.split(os.path.splitext(os.path.basename(self.path))[0]))
+        fnp.parseFilename(filename)
 
         if fnp.issue != "":
             metadata.issue = fnp.issue

comicapi/filenameparser.py

@@ -180,10 +180,10 @@ class FileNameParser:
         series = re.sub("\(.*?\)", "", series)
 
         # search for volume number
-        match = re.search('(.+)([vV]|[Vv][oO][Ll]\.?\s?)(\d+)\s*$', series)
+        match = re.search('(.+)([vV]|[Vv][oO][Ll]\.?\s?)(\d+)?\s*$', series)
         if match:
             series = match.group(1)
-            volume = match.group(3)
+            volume = match.group(3) or ""
 
         # if a volume wasn't found, see if the last word is a year in parentheses
         # since that's a common way to designate the volume
@@ -283,6 +283,9 @@ class FileNameParser:
             self.volume,
             issue_end)
 
+        if self.issue == "" and self.volume != "":
+            self.issue = self.volume
+
         if self.issue != "":
             # strip off leading zeros
             self.issue = self.issue.lstrip("0")