The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

Conflicts

Reviewers, this pull request conflicts with the following ones:

• #16519 (guix: Change manifest to use channels and inferiors by dongcarl)

If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.

Concept ACK, but from now on, stop opening multiple PRs for everything. If you need to figure out how github works, please experiment with a repository of your own instead of our production repo.

These words could be dropped from ignore list:

mut could be dropped with renaming s/mut/m_mutex/ in threadinterrupt.{h|cpp}.

useable, unparseable, cachable could be dropped with proper fixes in the code base. However, I'm not sure about other devs opinion ;)

homogenous could be dropped with a proper fix in tinyformat.h. Also please see #12716 (comment).

ACK ac6d97f modulo addressing of the feedback from @hebasto :)

Here are some recent similar PRs:

• doc/lint: Fix spelling errors identified by codespell 1.15.0 #16186
• test: Check for codespell in lint-spelling.sh #16767

Maybe it is better to make such a PR once only per major and/or minor release (as part of release process)?

@laanwj - I apologize for the PR noise. For some reason I can't reopen PRs. It is probably something I did. (forced push setting in your admin settings) - I have since figured out my workflow. So if it is an admin setting on your side - I apologize and have since implemented my own build environment and will only create PRs once I have tested my changes personally FIRST! Again sincere apology. ditto to @MarcoFalke and the rest of the members - @hebasto @practicalswift @dongcarl @promag etc...

These words could be dropped from ignore list:

bitcoin/test/lint/lint-spelling.ignore-words.txt

Line 1 in 9641366

cas

bitcoin/test/lint/lint-spelling.ignore-words.txt

Line 6 in 9641366

unselect

mut could be dropped with renaming s/mut/m_mutex/ in threadinterrupt.{h|cpp}.

@hebasto,

I left the words you were unsure about out until we have more feed back.

I also removed /bitcoin/contrib/guix/manifest.scm from the commit.

If @dongcarl's change gets merged he will have killed two birds.
REF: d4e0694

useable, unparseable, cachable could be dropped with proper fixes in the code base. However, I'm not sure about other devs opinion ;)

homogenous could be dropped with a proper fix in tinyformat.h. Also please see #12716 (comment).

ACK d17d888

Tested d17d888:

$ ./test/lint/lint-spelling.sh | grep mut
src/threadinterrupt.cpp:25: mut  ==> must, mutt, moot
src/threadinterrupt.cpp:33: mut  ==> must, mutt, moot
src/threadinterrupt.h:32: mut  ==> must, mutt, moot

@RandyMcMillan Did you forget to rename mut to m_mutex as suggested?

I also removed /bitcoin/contrib/guix/manifest.scm from the commit.

Why? The goal of the mentioned commit is orthogonal to the goal of this PR.
I'd also prefer a complete solution over partial one, if it is feasible.

There is a misspelling in the commit message "doc:Fix some mispellings" ;)

$ ./test/lint/lint-spelling.sh | grep mut

@hebasto - speaking of grep - we need to add an alias to ggrep in the Mac environment. Brew doesnt do it automatically - this will take care on a linting issue in the Mac build.

Concept ACK, no longer draft?

CI failing with, for instance:

Running test/util/bitcoin-util-test.py...
/usr/bin/python3.6 ../test/util/bitcoin-util-test.py
2019-11-03 20:06:22,888 - ERROR - Error mismatch:
Expected: error: Uncompressed pubkeys are not usable for SegWit outputs
Received: error: Uncompressed pubkeys are not useable for SegWit outputs

NOTE: There are misspellings built into the functional tests.

That is a whole job with in itself. I had to pull some changes out.

Sorry about the extra builds.

I guess that is why people don't try to fix the linting.

Preliminary test build:
https://travis-ci.org/RandyMcMillan/bitcoin/builds/606965245

@hebasto @promag Yes, I believe it is ready.

I appreciate your patience and feedback.

@RandyMcMillan Please do not touch files in doc/man/ as they are generated by a script, and historical release notes in doc/release-notes/.

@hebasto - done. But can I make the point - by NOT fixing "historical" spelling issues in the release notes - we are also choosing to allow them and NEW ERRORS to persist into the future? What is the value of letting them persist? If we fix the old issues NOW we can enforce better policy and linting moving forward? Couldn't we even implement a build step that focuses solely on these types of issues or remove any rules that ignore documentation? (that was the thought behind editing the release notes to begin with)

Thanks! 👍

But can I make the point - by NOT fixing "historical" spelling issues in the release notes - we are also choosing to allow them and NEW ERRORS to persist into the future?

They are historical documents, which have been published to many places besides this repository. Fixing errors there would be like going to a museum and scribbling in a 12th century bible, for example 😄 It's no reason to allow new errors in new files, just, what has been done has been done.