From c8a847bd544efcdec80a5860a3a01b56297cbfdc Mon Sep 17 00:00:00 2001
From: Sascha Cowley <16543535+SaschaCowley@users.noreply.github.com>
Date: Mon, 12 Aug 2024 14:44:22 +1000
Subject: [PATCH] Update Pre-Commit hooks (#16983)

Summary of the issue:
Continuing work on our pre-commit hooks, there are still several we don't use but could.

Description of user facing changes
None.

Description of development approach
Tested pre-commit hooks available at pre-commit/pre-commit-hooks. Added those that work as expected and do not require major changes to our code base. Added:

Automatic checking of yaml, toml and XML syntax
Checks that no Python debug statements are included in commits
Remove BOM on Python, C/C++ and Batch files, if present
Ensure that unit test files are named according to unittest rules.
Ensure that any include VCS links are permalinks.
---
 .pre-commit-config.yaml | 44 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 5cbfec49d8d..2d506f6dcf2 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -17,16 +17,60 @@ repos:
     # Prevents commits to certain branches
     - id: no-commit-to-branch
       args: ["--branch", "master", "--branch", "beta", "--branch", "rc"]
+    # Checks that large files have not been added. Default cut-off for "large" files is 500kb.
+    - id: check-added-large-files
+      # POFiles and TTF fonts can't be made smaller
+      exclude_types: ["pofile", "ttf"]
+      # Same applies for NVDA dictionary (.dic) files and Spline Font Database (.SFD) files, but these aren't recognised by the Identify library.
+      exclude: "\\.(dic|sfd)$"
     # Checks python syntax
     - id: check-ast
       # File uses encoding not supported in Linux
       exclude: source/comInterfaces/_944DE083_8FB8_45CF_BCB7_C477ACB2F897_0_1_0.py
+    # Checks for filenames that will conflict on case insensitive filesystems (the majority of Windows filesystems, most of the time)
     - id: check-case-conflict
+    # Checks for artifacts from resolving merge conflicts.
     - id: check-merge-conflict
+    # Checks Python files for debug statements, such as python's breakpoint function, or those inserted by some IDEs.
+    - id: debug-statements
+      # File uses encoding not supported in Linux
+      exclude: source/comInterfaces/_944DE083_8FB8_45CF_BCB7_C477ACB2F897_0_1_0.py
+    # Removes trailing whitespace.
     - id: trailing-whitespace
       types_or: [python, c, c++, batch, markdown]
+    # Ensures all files end in 1 (and only 1) newline.
     - id: end-of-file-fixer
       types_or: [python, c, c++, batch, markdown]
+    # Removes the UTF-8 BOM from files that have it.
+    # See https://github.com/nvaccess/nvda/blob/master/projectDocs/dev/codingStandards.md#encoding
+    - id: fix-byte-order-marker
+      types_or: [python, c, c++, batch, markdown]
+    # Validates TOML files.
+    - id: check-toml
+    # Validates YAML files.
+    - id: check-yaml
+    # Validates XML files.
+    - id: check-xml
+    # Ensures that links to lines in files under version control point to a particular commit.
+    - id: check-vcs-permalinks
+    # Checks that tests are named test_*.py.
+    - id: name-tests-test
+      args: ["--unittest"]
+      # Exclude Python files under `tests/` that aren't unittest files.
+      # This is a Python verbose regular expression.
+      # See https://docs.python.org/3/library/re.html#re.VERBOSE
+      exclude: |
+        (?x)^tests/(
+          checkPot.py |  # Doesn't use unittest
+          system |  # Uses robot
+          unit/ (
+            # Test helpers
+            textProvider.py |
+            extensionPointTestHelpers.py |
+            objectProvider.py |
+            test_speechManager/speechManagerTestHarness.py
+          )
+        )
 
 - repo: https://github.com/asottile/add-trailing-comma
   rev: v3.1.0