Use ASCII strings in source files

[am11]

This is to align with coding style pt. 16.

Used https://gist.github.com/am11/f35c61783ad7b7c304b3d6e8bd1dcab4 as:

git config core.quotepath off
git ls-files ':/src/*cs' ':/src/*vb' ':/src/*fs' | xargs -I{} conv {} {}

followed by some manual fixes.

[ghost]

Tagging subscribers to this area: @dotnet/area-meta
See info in area-owners.md if you want to be subscribed.

Issue Details

---

This is to align with coding style pt. 16.

Used https://gist.github.com/am11/f35c61783ad7b7c304b3d6e8bd1dcab4 as:

git config core.quotepath off
git ls-files ':/src/*cs' ':/src/*vb' ':/src/*fs' | xargs -I{} conv {} {}

followed by some manual fixes.

Author:	am11
Assignees:	-
Labels:	area-Meta, community-contribution
Milestone:	-

[danmoseley]

How do we feel about BOMs - should we remove them all? I guess some would just come back, so it might just be making unnecessary diffs.

[am11]

Yes, there are plenty of files with BOM (1514 files with ':/src/*cs' ':/src/*vb' ':/src/*fs' filter alone). It will create unnecessary diff. For now, I've preserved it. We can remove them in a separate PR. Once sanitized, we can also add a small github workflow for PRs, which checks for its presence in modified files (and fail the job to protect it from regressing).

There are two BOM related changes in this PR delta:

• XmlSchema.Writer TrimmingTests needed a manual fix with preamble bytes to string, using UTF-8 encoder.
• TraceSourceConfigurationTests.cs had invalid UTF-8 BOM (I double checked that it wasn't a UTF-16 BOM but first two of three bytes of UTF8 BOM were repeated after the valid BOM).

[am11]

@danmoseley, was there any other comment / concern?

[jeffhandley]

Thanks for the attention to detail on this @am11 (including making author name/email comments consistent). I've spot-checked the characters throughout and all was matching up. Thanks for sharing how you approached this too; that was helpful.

I'm rerunning the failed checks for good measure...

[danmoseley]

Thanks @am11. This definitely has made our tests more robust. The case I was concerned about is tests that have a magic input and end up verifying it matches the output. If those get garbelized probably nobody would have noticed.