Normalize UTF-8 BOM Marks and ps1 Indention

[petercrabtree]

Problem

The repo mixes UTF-8 with and without a byte order mark (BOM). Some tools in my workflow write using their own defaults instead of preserving a file’s existing encoding, so routine edits flip the BOM and create noisy diffs. This PR establishes one policy, aligns scripts to it, then applies a mechanical cleanup.

Why you may not encounter this: if your tools preserve a file’s current encoding (as most of mine do...), you won’t hit churn. The noise appears only with non-preserving writers.

Changes

No functional/runtime changes.

Commit 1 — policy + tooling

• .editorconfig: default to UTF-8 without BOM; .resx stays UTF-8 with BOM. The ResX toolchain (ResXResourceWriter, Visual Studio designer, resgen) emits a BOM and ignores .editorconfig, so we configure .resx accordingly to keep editors and build tools aligned.
• BuildTools/update-assemblyinfo.ps1: write UTF-8 without BOM to match policy. On PowerShell pre-v6, Out-File writes a BOM by default; this prevents that.
• New: BuildTools/bom-strip.ps1: remove all BOMs (except resx)
• New: BuildTools/bom-classify-encoding.ps1: inspect BOM presence; also used to derive the extension list consumed by bom-strip.

dotnet format honours .editorconfig and checks the charset (including BOM), so this should prevent BOM churn going forward by hooking into the existing formatting enforcement.

Commit 2 — mechanical application

• Normalize UTF-8 BOMs across the repository by removing them from all text files other than resx

Review notes

Commit 2 is the mechanical output of running buildtools/bom-strip.ps1 after Commit 1. Its diff should contain only BOM removals.

It was done this way to make verifying the giant (>1000 files) commit easier--I suggest parsing the diff/patch or running buildtools/bom-strip.ps1 on commit 1 if you want to make sure for yourself that there are no other changes in there.

[Copilot]

Pull Request Overview

This pull request normalizes UTF-8 byte order mark (BOM) usage across the repository to eliminate noise from tools that don't preserve existing encoding. The PR establishes a "UTF-8 without BOM" policy, updates build tooling to respect this policy, and mechanically removes BOMs from over 1000 files.

Key changes:

• Establishes UTF-8 without BOM as the default encoding policy
• Updates PowerShell build scripts to write UTF-8 without BOM
• Adds new PowerShell utilities for BOM management and inspection

Reviewed Changes

Copilot reviewed 300 out of 1355 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
.editorconfig	Adds charset configuration with UTF-8 without BOM default, except for .resx files
BuildTools/update-assemblyinfo.ps1	Modified to write UTF-8 without BOM using explicit encoding
BuildTools/bom-strip.ps1	New utility script to remove BOMs from text files
BuildTools/bom-classify-encodings.ps1	New utility script to classify files by encoding type
All other files	Mechanical BOM removal from text files (over 1000 files)

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[siegfriedpammer]

indentation in ps1 should be spaces, I guess?

[siegfriedpammer]

could be added to .editorconfig as well :)

[petercrabtree]

Given that the existing ps1 files are a total mishmash of tabs vs spaces (even on the same line!), I think we should just choose.

And given that the project bucks the trend and uses tabs for .cs files (and I like tabs...), I've gone ahead and just standardized the ps1 files on tabs.

[siegfriedpammer]

Thank you very much!