* Use in_progress workflow_job example, not queued
### tl;dr
This changes the example `workflow_job` payload to be an `in_progress` event instead of `queued` event.
### why
In #22174, I added some keys to the example workflow_job json. This was correct, but by adding non-null _values_, **I introduced a subtle bug**.
It's a bug because a `queued` workflow is—by definition—_not_ yet assigned to a runner. So all the `runner_*` attributes will be `null` in the real world.
After the workflow becomes `in_progress`, all the values are guaranteed to exist. So let's change the example to be an `in_progress` event, since it's a better example for documentation.
* Explain more about the runner_* attributes in "workflow_job_properties.md"
* Consistentize naming
Co-authored-by: Martin Lopes <martin389@github.com>
Co-authored-by: hubwriter <hubwriter@github.com>
* search improvements
* expose search result score on debug=1 qs (#22452)
* Search index improvements (#22428)
* concatenate headings for search ranking (#22445)
* search index improvements
* update index
* add data-search property used to create search records
* direct match title and normalize results to one per article
* the search index is getting huge
* add chinese lang, intro, and boost
* create a article level record with no heading for each article
* remove chinese language and exact match, boost breadcrumbs
* break out debug mode into separate branch
* break adding chinese out into another branch
* one record per article remove limits
* revert added spaces
* revert adding property on article page
* concatenate headings for search ranking
* remove slug property, fix test
* update record format in tests
* revert adding this property back
* scrape article-body not contents
* adding heading property to tests
* add headings to content property
* Update script/search/parse-page-sections-into-records.js
Co-authored-by: Peter Bengtsson <peterbe@github.com>
* Update script/search/parse-page-sections-into-records.js
Co-authored-by: Peter Bengtsson <peterbe@github.com>
Co-authored-by: Peter Bengtsson <peterbe@github.com>
* search by popularity (#22496)
* search by popularity
* adjust factors based on length of query and normalize Lunr score
* Apply suggestions from code review
Co-authored-by: Rachael Sewell <rachmari@github.com>
Co-authored-by: Rachael Sewell <rachmari@github.com>
* turn html into text with appropriate spaces (#22528)
* turn HTML into text with appropriate spaces
Part of #1141
* fix unit test
* improve tests
* small refactor
* slice the search results .content (#22514)
* search by popularity
* adjust factors based on length of query and normalize Lunr score
* slice the search result .content
Part of #1142
* Update lib/search/lunr-search.js
Co-authored-by: Rachael Sewell <rachmari@github.com>
Co-authored-by: Rachael Sewell <rachmari@github.com>
* truncate translated search index records (#22475)
* truncate translated search index records
* page entire page object
* revert search indexes
* try reverting search indexes again...
* out of sync with main in parent branch so reverting last commit again
* rever deleted indexes
* remove debug lines
* correct package-lock.json
* slice content string better (#22586)
Part of #1149
* include home pages in the search (#22568)
* Search remove hyphen as token (#22584)
* slice better when not present in content (#22672)
* fix missing search breadcrumbs (#22681)
* fix missing search breadcrumbs
* hopefully fix tests
Co-authored-by: Peter Bengtsson <mail@peterbe.com>
* Highlight based on combined matchData (#22692)
* highlight based on combined matchData
* refactoring
* remove results when lang or version changes (#22697)
* search with AND on title first and favor that more
* commented out code
* Update lib/search/lunr-search.js
* only apply the wildcard to the title search if the last word is short
Co-authored-by: Rachael Sewell <rachmari@github.com>
