Skip to content

Hand over the command where the parity page counted the tree - #364

Merged
iderex merged 1 commit into
masterfrom
docs/the-parity-page-counts-a-tree-that-tripled
Sep 5, 2026
Merged

iderex merged 1 commit into
masterfrom
docs/the-parity-page-counts-a-tree-that-tripled

Conversation

@iderex

@iderex iderex commented Sep 5, 2026

Copy link
Copy Markdown
Collaborator

Refs #363

Finishes: #363

What was wrong

git show 74f5d62:docs/gate-parity.md | sed -n '95,98p'
This repository has eight tracked files inside that glob:

    git ls-files | grep -Eic '\.(js|html|md|css|scss)$'
    8

git ls-files | grep -Eic '\.(js|html|md|css|scss)$'
24

The number was right when it was written and has not been touched since:

git log --format='%h %ad %s' --date=short 74f5d62 -S"This repository has eight tracked files inside that glob" -- docs/gate-parity.md
03dd447 2026-08-06 Quote the parity numbers from the commit under review

git ls-tree -r --name-only 03dd447 | grep -Eic '\.(js|html|md|css|scss)$'
8
git ls-tree -r --name-only 74f5d62 | grep -Eic '\.(js|html|md|css|scss)$'
24

So the tree tripled under the sentence over a month and nothing reddened.

What it says now

The count is deleted rather than corrected and the command is handed to the
reader. That is not an invented repair: the mutation section of this same file
took it three sections down, for a count of the plugin's files that drifted the
same way, and says why in its own words - nothing here reads a figure in a
document, so the command is handed over in place of its output.

The paragraph's own argument is what the number was there to illustrate, and it
survives without it: the count includes this document, which is why it was first
written as seven, and it was taken before the file holding it was added. The
row's conclusion never turned on the size of the count, only on it not being
zero, and the closing sentence now says exactly that.

The third condition, which is the other figures

Read in the same pass. None has moved.

The two required-set counts are dated to 2026-08-06 in the text and already tell
a reader to re-run the commands rather than quote them, which is the shape this
change gives the prettier count.

The sign-off row's count and the target's commit date are read at pinned refs
rather than at a branch, and both reproduce:

gh api 'repos/iderex/jellyfin-plugin-sso/contents/.github/workflows/dco.yml?ref=e9cee021e95763e5240b44b8d7af16598df609ce' --jq '.content' | base64 -d | grep -c 'expected="Signed-off-by: ${author_name} <${author_email}>"'
1
gh api repos/iderex/jellyfin-plugin-sso/commits/54873c4920a88812f65f3a7884c5830e2ce31fb5 --jq '.commit.committer.date'
2026-08-06T14:13:06Z

What holds it, which is nothing, and the residual that leaves

No leg is added and none is claimed. Nothing in this repository reads
docs/gate-parity.md at all - it is not among the documents the suite is
given - so the next figure typed into it goes stale exactly the same way. #363's
done-condition offers a derived leg as the alternative to deletion; a leg would
mean giving the suite a reading of a document whose whole subject is a comparison
against another board's tree, which is a larger decision than this repair and is
not taken here.

VSTEST_CONNECTION_TIMEOUT=300 DOTNET_CLI_UI_LANGUAGE=en dotnet test Jellyfin.Plugin.MetadataSync.sln --nologo
Failed: 0, Passed: 1696

per target, over both server lines, before the change and after it.

Reading

No second reader looked at this. The commands above are the evidence in place of
one.

The prettier row said this repository has eight tracked files inside the
glob the target lints. It has twenty-four. The number was right at 03dd447
on 2026-08-06 and untouched since, while the tree it counts tripled under
it, and nothing reddened because nothing reads a figure in a document.

The count is deleted rather than corrected and the command is handed to the
reader, which is the repair the mutation section of this same file already
took for the same class of literal. The paragraph's own argument, that the
count includes this document and was first written as seven because it was
taken before that file was added, is what the number illustrated and it
survives without it. The row's conclusion never turned on the size of the
count, only on it not being zero, and the closing sentence now says that.

The other figures in the file were read in the same pass and none has
moved: the required-set counts are dated to 2026-08-06 and already tell a
reader to re-run rather than quote, and the sign-off row's count and the
target's commit date are both read at pinned refs and reproduce.

No leg is added and none is claimed. Nothing in this repository reads this
document at all - it is not among the files the suite is given - so the
next figure typed into it goes stale the same way. That residual is what
#363 records and this does not close.

Signed-off-by: Nils Lehnen <30603423+iderex@users.noreply.github.com>
@iderex iderex added the documentation Improvements or additions to documentation label Sep 5, 2026
@iderex iderex self-assigned this Sep 5, 2026
@iderex iderex added this to the M10 Quality parity program milestone Sep 5, 2026
@iderex
iderex merged commit 009b843 into master Sep 5, 2026
14 of 15 checks passed
@iderex
iderex deleted the docs/the-parity-page-counts-a-tree-that-tripled branch September 5, 2026 08:02
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

documentation Improvements or additions to documentation

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant