Move colon handling when renaming to the MetadataFormatter class

Fixes #356

README.md

@@ -2,6 +2,7 @@
 [![GitHub release (latest by date)](https://img.shields.io/github/downloads/comictagger/comictagger/latest/total)](https://github.com/comictagger/comictagger/releases/latest)
 [![PyPI](https://img.shields.io/pypi/v/comictagger)](https://pypi.org/project/comictagger/)
 [![PyPI - Downloads](https://img.shields.io/pypi/dm/comictagger)](https://pypistats.org/packages/comictagger)
+[![Chocolatey package](https://img.shields.io/chocolatey/dt/comictagger?color=blue&label=chocolatey)](https://community.chocolatey.org/packages/comictagger)
 [![PyPI - License](https://img.shields.io/pypi/l/comictagger)](https://opensource.org/licenses/Apache-2.0)
 
 [![GitHub Discussions](https://img.shields.io/github/discussions/comictagger/comictagger)](https://github.com/comictagger/comictagger/discussions)
@@ -48,6 +49,12 @@ A pip package is provided, you can install it with:
 
 There are two optional dependencies GUI and CBR. You can install the optional dependencies by specifying one or more of `GUI`,`CBR` or `all` in braces e.g. `comictagger[CBR,GUI]`
 
+### Chocolatey installation (Windows only)
+
+A [Chocolatey package](https://community.chocolatey.org/packages/comictagger), maintained by @Xav83, is provided, you can install it with:
+```powershell
+choco install comictagger
+```
 ### From source
 
  1. Ensure you have python 3.9 installed

comicapi/utils.py

@@ -121,13 +121,9 @@ def remove_articles(text: str) -> str:
         "the",
         "the",
         "with",
-        "ms",
-        "mrs",
-        "mr",
-        "dr",
     ]
     new_text = ""
-    for word in text.split(" "):
+    for word in text.split():
         if word not in articles:
             new_text += word + " "
 
@@ -139,19 +135,16 @@ def remove_articles(text: str) -> str:
 def sanitize_title(text: str, basic: bool = False) -> str:
     # normalize unicode and convert to ascii. Does not work for everything eg ½ to 1⁄2 not 1/2
     text = unicodedata.normalize("NFKD", text).casefold()
-    if basic:
-        # comicvine keeps apostrophes a part of the word
-        text = text.replace("'", "")
-        text = text.replace('"', "")
-    else:
+    # comicvine keeps apostrophes a part of the word
+    text = text.replace("'", "")
+    text = text.replace('"', "")
+    if not basic:
         # comicvine ignores punctuation and accents
         # remove all characters that are not a letter, separator (space) or number
         # replace any "dash punctuation" with a space
         # makes sure that batman-superman and self-proclaimed stay separate words
         text = "".join(
-            c if not unicodedata.category(c) in ("Pd",) else " "
-            for c in text
-            if unicodedata.category(c)[0] in "LZN" or unicodedata.category(c) in ("Pd",)
+            c if unicodedata.category(c)[0] not in "P" else " " for c in text if unicodedata.category(c)[0] in "LZNP"
         )
         # remove extra space and articles and all lower case
         text = remove_articles(text).strip()

comictaggerlib/filerenamer.py

@@ -45,7 +45,7 @@ class MetadataFormatter(string.Formatter):
     def __init__(self, smart_cleanup: bool = False, platform: str = "auto") -> None:
         super().__init__()
         self.smart_cleanup = smart_cleanup
-        self.platform = platform
+        self.platform = platform.casefold()
 
     def format_field(self, value: Any, format_spec: str) -> str:
         if value is None or value == "":
@@ -72,6 +72,10 @@ class MetadataFormatter(string.Formatter):
                 if lstrip:
                     literal_text = literal_text.lstrip("-_)}]#")
                 if self.smart_cleanup:
+                    if self.platform in ["universal", "windows"] or sys.platform.casefold() in ["windows"]:
+                        # colons get special treatment
+                        literal_text = literal_text.replace(": ", " - ")
+                        literal_text = literal_text.replace(":", "-")
                     lspace = literal_text[0].isspace() if literal_text else False
                     rspace = literal_text[-1].isspace() if literal_text else False
                     literal_text = " ".join(literal_text.split())
@@ -179,13 +183,6 @@ class FileRenamer:
 
         new_basename = ""
         for component in pathlib.PureWindowsPath(template).parts:
-            if (
-                self.platform.casefold() in ["universal", "windows"] or sys.platform.casefold() in ["windows"]
-            ) and self.smart_cleanup:
-                # colons get special treatment
-                component = component.replace(": ", " - ")
-                component = component.replace(":", "-")
-
             new_basename = str(
                 sanitize_filename(fmt.vformat(component, args=[], kwargs=Default(md_dict)), platform=self.platform)
             ).strip()

testing/filenames.py

@@ -757,6 +757,13 @@ rnames = [
         "Cory Doctorow's Futuristic Tales of the Here and Now #001 - Anda's Game (2007).cbz",
         does_not_raise(),
     ),
+    (
+        "{series} #{issue} - {title} {volume:02} ({year})",  # Ensure format specifier works
+        False,
+        "universal",
+        "Cory Doctorow's Futuristic Tales of the Here and Now #001 - Anda's Game 01 (2007).cbz",
+        does_not_raise(),
+    ),
     (
         "{series} #{issue} - {title} ({year})({price})",  # price should be none, test no  space between ')('
         False,

tests/utils_test.py

@@ -108,3 +108,19 @@ titles = [
 @pytest.mark.parametrize("value, result", titles)
 def test_titles_match(value, result):
     assert comicapi.utils.titles_match(value[0], value[1]) == result
+
+
+titles_2 = [
+    ("", ""),
+    ("鋼の錬金術師", "鋼の錬金術師"),
+    ("Conan el Bárbaro", "Conan el Barbaro"),
+    ("The Batman's Grave", "batmans grave"),
+    ("A+X", "ax"),
+    ("ms. marvel", "ms marvel"),
+    ("spider-man/deadpool", "spider man deadpool"),
+]
+
+
+@pytest.mark.parametrize("value, result", titles_2)
+def test_sanitize_title(value, result):
+    assert comicapi.utils.sanitize_title(value) == result.casefold()