Mandatory rules blog post

Author: fileformat

docs/_drafts/mandatory-rules.md

@@ -71,20 +71,21 @@
 		<div class="row">
 			<div class="col pt-2">
 				<div class="pb-2 d-flex align-items-center justify-content-between">
-				<h1 class="flex-grow-1">
-					{% if page.h1 %}{{page.h1}}{% else %}{{page.title}}{% endif %}
-				</h1>
-				<div class="text-end">
-					{% if page.previous.url %}
-					<a class="btn btn-outline-secondary" href="{{page.previous.url}}" title="{{page.previous.title}}">&laquo;<span
-							class="d-none d-lg-inline"> {{page.previous.title}}</span></a>
-					{% endif %}
-					{% if page.next.url %}
-					<a class="btn btn-outline-secondary" href="{{page.next.url}}" title="{{page.next.title}}"><span
-							class="d-none d-lg-inline">{{page.next.title}} </span>&raquo;</a>
-					{% endif %}
-				</div>
+					<h1 class="flex-grow-1">
+						{% if page.h1 %}{{page.h1}}{% else %}{{page.title}}{% endif %}
+					</h1>
+					<div class="text-end">
+						{% if page.previous.url %}
+						<a class="btn btn-outline-secondary" href="{{page.previous.url}}" title="{{page.previous.title}}">&laquo;<span
+								class="d-none d-lg-inline"> {{page.previous.title}}</span></a>
+						{% endif %}
+						{% if page.next.url %}
+						<a class="btn btn-outline-secondary" href="{{page.next.url}}" title="{{page.next.title}}"><span
+								class="d-none d-lg-inline">{{page.next.title}} </span>&raquo;</a>
+						{% endif %}
+					</div>
 				</div>
+{% if page.draft %}<div class="alert alert-warning">This post is a draft!</div>{% endif %}
 {{content}}
 			{% if page.url contains "blog/"-%}
 				<p class="text-center">

docs/_layouts/default.html

@@ -25,7 +25,7 @@ operating systems
 
 - [Rich Jenks' filename regex](https://richjenks.com/filename-regex/) - a regex that tests for cross-plaform issues (including safe for URLs)
 
-- [sindresorhus/filename-reserved-regex](https://github.com/sindresorhus/filename-reserved-regex) - NPM library with regexes for reserved characters and MS-DOS reserved names.
+- [sindresorhus/filename-reserved-regex](https://github.com/sindresorhus/filename-reserved-regex) - NPM library with regexes for reserved characters and MS-DOS reserved names.  Used in [sindresorhus/filenamify](https://github.com/sindresorhus/filenamify) and [sindresorhus/valid-filename](https://github.com/sindresorhus/valid-filename).
 
 - [filesan](https://github.com/BonnyAD9/filesan) - Rust, a library that sanitizes strings so they can safely be used as file names.  I.e. if someone gives you a string as a file name, this will make sure it is safe.
 
@@ -39,6 +39,8 @@ operating systems
 
 - [CrossRename](https://pypi.org/project/CrossRename/) - Python, focused on portability between Linux and Windows ([source](https://github.com/Jemeni11/CrossRename))
 
+- [flytam/filenamify](https://github.com/flytam/filenamify) - Go, focused on creating portable file names
+
 - [spf13/pathologize](https://github.com/spf13/pathologize) - Go, focused on creating portable file names
 
 - [ClaireCJS/fix_unicode_filenames](https://github.com/ClaireCJS/fix_unicode_filenames) - Python, renames files with Unicode and/or troublesome characters

docs/_posts/2024-08-17-prior-art.md

@@ -4,7 +4,7 @@ title: Wildcards are dangerous
 
 We consider wildcards to be very dangerous from a security standpoint.  It is too easy to have something strange in a filename that causes a wildcard to not match.  This made worse if there are multiple programs that evaluate wildcards: something could match in one and not in the other.
 
-We are explicitly choosing not to support wildcards when specifying top-level paths.  Ideally, you should specify the current directory (with `.`) and every file and directry should be included.
+We are explicitly choosing not to support wildcards when specifying top-level paths.  Ideally, you should specify the current directory (with `.`) and every file and directory should be included.
 
 Additionally, we want to check *every* file.  Even hidden and ignored files should obey basic sanity rules.  Thus, we have an `not-linted` rule that will flag every file that doesn't get linted.
 

docs/_posts/2025-04-12-wildcard-danger.md

@@ -0,0 +1,26 @@
+---
+title: "Mandatory Rules"
+---
+
+There are a some rules that cannot be disabled.  These rules will cause the file to fail linting and additional lints to be skipped.
+
+## UTF-8 and no nulls
+
+These two rules are mandatory:
+
+1. Must be UTF-8
+2. Must not have any null (`0`) bytes
+
+Reasoning:
+
+* If it is encoding in something other than UTF-8, we don't to be in the business of converting encodings.
+* If it isn't valid UTF-8 on a system that uses UTF-8 encoding, there then something is really wrong.
+* If it has a null byte, again, something is really wrong.
+
+## Rust conversions
+
+Rust is pretty explicit about handling potential errors when converting from paths and directory entries.  These failures can't be disabled.  In theory, the UTF-8 required rule should catch them, but if something slips through, it will still be a lint failure.
+
+## Simultaneous file changes
+
+A variety of problems can occur if you are changing files/directories while namelint is running.  Why would you do that?