feat: make filename escaping configurable

README.md

@@ -21,7 +21,7 @@
 - [ ] Nested categories
 - [ ] Share the fun somewhere
 - [x] Edge cases: mark as read during sync (if marked as read on server or not)
-- [ ] Proper filename escaping
+- [x] Proper filename escaping
 - [ ] Command to force regenerate all HTML files (incl. recompute datetimes & paths)
 - [ ] Handle item updates
 

config.default.toml

@@ -44,6 +44,8 @@ template = '''
 filename_template = "{{ published_formatted }}\t[{{ origin_title }}]\t{{ title }}.html"
 # Maximum allowed filename length (in bytes assuming UTF-8 encoding) before truncating. Depending on your filesystem filename's limits it may be possible to increase the value, ask Wikipedia for details.
 max_filename_length = 250
+# Table mapping characters to what they will be replaced with in filenames. Useful to remove/replace characters that are not allowed in filename by your filesystem. The default should be fine for most Unix filesystems.
+filename_replacement = { "/" = "⧸", "\u0000" = "" }
 
 [time]
 # Which timezone to use when writing date and time.

main.py

@@ -46,6 +46,7 @@ class Config:
         self.item_template: Template = Template(get_config("html", "template"), autoescape=True)
         self.item_filename_template: Template = Template(get_config("html", "filename_template"), autoescape=False)
         self.max_filename_length: int = int(get_config("html", "max_filename_length"))
+        self.filename_translation = str.maketrans(get_config("html", "filename_replacement"))
 
         # Computed config fields
         self.update_lock = self.json_root / "update.lock"
@@ -109,8 +110,8 @@ def mark_deleted_as_read(config, client_session):
 
     print(f"Marked {marked_as_read} items as read")
 
-def escape_filename(filename):
-    return filename.replace("/", "-")
+def escape_filename(config, filename):
+    return filename.translate(config.filename_translation)
 
 def truncate_filename(config, filename):
     max_filename_length = config.max_filename_length
@@ -124,10 +125,10 @@ def truncate_filename(config, filename):
         return filename[:cutoff] + '…' + suffix
 
 def get_html_path(config, item_json):
-    folder_directory = config.html_root / escape_filename(item_json["folder"])
+    folder_directory = config.html_root / escape_filename(config, item_json["folder"])
     folder_directory.mkdir(exist_ok=True)
 
-    html_name = truncate_filename(config, escape_filename(config.item_filename_template.render(item_json)))
+    html_name = truncate_filename(config, escape_filename(config, config.item_filename_template.render(item_json)))
 
     return folder_directory / html_name