jprdonnelly/docs
1
0
Fork 0
mirror of synced 2026-01-06 06:02:35 -05:00
Code Issues Projects Releases Wiki Activity

Merge remote-tracking branch 'private/main'

Browse Source
This commit is contained in:
Kevin Heis
2025-05-20 14:07:30 -07:00
parent a776b846f7 32a8d91b16
commit 7349298ab3
16 changed files with 535 additions and 130 deletions
Download Patch File Download Diff File Expand all files Collapse all files

45
.github/workflows/sync-audit-logs.yml vendored
View File

@@ -51,34 +51,38 @@ jobs:
# Needed for gh
GITHUB_TOKEN: ${{ secrets.DOCS_BOT_PAT_BASE }}
run: |
# If nothing to commit, exit now. It's fine. No orphans.
changes=$(git diff --name-only | wc -l)
untracked=$(git status --untracked-files --short | wc -l)
filesChanged=$(git diff --name-only)
# There will always be at least one file changed:
# src/audit-logs/lib/config.json
# If the config file is the only file changed, exit.
if [[ $changes -eq 1 ]] && [[ $untracked -eq 1 ]] && [[ $filesChanged == *lib/config.json ]]; then
echo "There are no changes to commit or untracked files. Exiting..."
exit 0
fi
git config --global user.name "docs-bot"
git config --global user.email "77750099+docs-bot@users.noreply.github.com"
echo "Creating a new branch if needed..."
branchname=audit-logs-schema-update-${{ steps.audit-log-allowlists.outputs.COMMIT_SHA }}
remotesha=$(git ls-remote --heads origin $branchname)
if [ -n "$remotesha" ]; then
# output is not empty, it means the remote branch exists
echo "Branch $branchname already exists in 'github/docs-internal'. Exiting..."
exit 0
fi
git checkout -b $branchname
git add .
echo "Created a new branch $branchname"
echo "Preparing commit..."
git config --global user.name "docs-bot"
git config --global user.email "77750099+docs-bot@users.noreply.github.com"
git add -A .
echo "Prepared commit"
echo "Check if there are changes..."
if git diff-index --cached --quiet HEAD -- . ':(exclude)src/audit-logs/lib/config.json'
then
echo "No real changes (only the SHA in config.json moved). Exiting…"
exit 0
fi
echo "Changes detected, proceeding"
echo "Creating commit..."
git commit -m "Add updated audit log event data"
echo "Created commit"
echo "Pushing commit..."
git push origin $branchname
echo "Pushed commit"
echo "Creating pull request..."
gh pr create \
@@ -89,16 +93,21 @@ jobs:
--repo github/docs-internal \
--label audit-log-pipeline \
--head=$branchname
echo "Created pull request"
# can't approve your own PR, approve with Actions
echo "Approving pull request..."
unset GITHUB_TOKEN
gh auth login --with-token <<< "${{ secrets.GITHUB_TOKEN }}"
gh pr review --approve
echo "Approved pull request"
# Actions can't merge the PR so back to docs-bot to merge the PR
echo "Setting pull request to auto merge..."
unset GITHUB_TOKEN
gh auth login --with-token <<< "${{ secrets.DOCS_BOT_PAT_BASE }}"
gh pr merge --auto
echo "Set pull request to auto merge"
- uses: ./.github/actions/slack-alert
if: ${{ failure() && github.event_name != 'workflow_dispatch' }}

1
.github/workflows/sync-openapi.yml vendored
View File

@@ -49,6 +49,7 @@ jobs:
repository: github/models-gateway
path: models-gateway
ref: main
token: ${{ secrets.DOCS_BOT_PAT_BASE }}
- uses: ./.github/actions/node-npm-setup

2
data/reusables/contributing/content-linter-rules.md
View File

@@ -65,4 +65,4 @@
| GHD041 | third-party-action-pinning | Code examples that use third-party actions must always pin to a full length commit SHA | error | feature, actions |
| GHD042 | liquid-tag-whitespace | Liquid tags should start and end with one whitespace. Liquid tag arguments should be separated by only one whitespace. | error | liquid, format |
| GHD043 | link-quotation | Internal link titles must not be surrounded by quotations | error | links, url |
| GHD022 | liquid-ifversion-versions | Liquid `ifversion`, `elsif`, and `else` tags should be valid and not contain unsupported versions. | error | liquid, versioning |
| GHD044 | octicon-aria-labels | Octicons should always have an aria-label attribute even if aria-hidden. | warning | accessibility, octicons |

3
src/content-linter/lib/linting-rules/index.js
View File

@@ -33,6 +33,7 @@ import { tableLiquidVersioning } from './table-liquid-versioning.js'
import { thirdPartyActionPinning } from './third-party-action-pinning.js'
import { liquidTagWhitespace } from './liquid-tag-whitespace.js'
import { linkQuotation } from './link-quotation.js'
import { octiconAriaLabels } from './octicon-aria-labels.js'
import { liquidIfversionVersions } from './liquid-ifversion-versions.js'
const noDefaultAltText = markdownlintGitHub.find((elem) =>
@@ -82,6 +83,6 @@ export const gitHubDocsMarkdownlint = {
thirdPartyActionPinning,
liquidTagWhitespace,
linkQuotation,
liquidIfversionVersions,
octiconAriaLabels,
],
}

56
src/content-linter/lib/linting-rules/octicon-aria-labels.js Normal file
View File

@@ -0,0 +1,56 @@
import { TokenKind } from 'liquidjs'
import { getLiquidTokens, getPositionData } from '../helpers/liquid-utils.js'
import { addFixErrorDetail } from '../helpers/utils.js'
/*
Octicons should always have an aria-label attribute even if aria hidden. For example:
DO use aria-label
{% octicon "alert" aria-label="alert" %}
{% octicon "alert" aria-label="alert" aria-hidden="true" %}
{% octicon "alert" aria-label="alert" aria-hidden="true" class="foo" %}
This is necessary for copilot to be able to recognize the svgs correctly when using our API.
*/
export const octiconAriaLabels = {
names: ['GHD044', 'octicon-aria-labels'],
description: 'Octicons should always have an aria-label attribute even if aria-hidden.',
tags: ['accessibility', 'octicons'],
parser: 'markdownit',
function: (params, onError) => {
const content = params.lines.join('\n')
const tokens = getLiquidTokens(content)
.filter((token) => token.kind === TokenKind.Tag)
.filter((token) => token.name === 'octicon')
for (const token of tokens) {
const { lineNumber, column, length } = getPositionData(token, params.lines)
const hasAriaLabel = token.args.includes('aria-label=')
if (!hasAriaLabel) {
const range = [column, length]
const octiconNameMatch = token.args.match(/["']([^"']+)["']/)
const octiconName = octiconNameMatch ? octiconNameMatch[1] : 'icon'
const originalContent = token.content
const fixedContent = originalContent + ` aria-label="${octiconName}"`
addFixErrorDetail(
onError,
lineNumber,
`octicon should have an aria-label even if aria hidden. Try using 'aria-label=${octiconName}'`,
token.content,
range,
{
lineNumber,
editColumn: column,
deleteCount: length,
insertText: `{% ${fixedContent} %}`,
},
)
}
}
},
}

6
src/content-linter/style/github-docs.js
View File

@@ -168,6 +168,12 @@ const githubDocsConfig = {
'partial-markdown-files': true,
'yml-files': true,
},
'octicon-aria-labels': {
// GHD044
severity: 'warning',
'partial-markdown-files': true,
'yml-files': true,
},
}
export const githubDocsFrontmatterConfig = {

160
src/content-linter/tests/unit/octicon-aria-labels.js Normal file
View File

@@ -0,0 +1,160 @@
import { describe, expect, test } from 'vitest'
import { octiconAriaLabels } from '../../lib/linting-rules/octicon-aria-labels.js'
describe('octicon-aria-labels', () => {
const rule = octiconAriaLabels
test('detects octicon without aria-label', () => {
const errors = []
const onError = (errorInfo) => {
errors.push(errorInfo)
}
const content = ['This is a test with an octicon:', '{% octicon "alert" %}', 'Some more text.']
rule.function({ lines: content }, onError)
expect(errors.length).toBe(1)
expect(errors[0].lineNumber).toBe(2)
expect(errors[0].detail).toContain('aria-label=alert')
expect(errors[0].fixInfo.insertText).toContain('aria-label="alert"')
})
test('ignores octicons with aria-label', () => {
const errors = []
const onError = (errorInfo) => {
errors.push(errorInfo)
}
const content = [
'This is a test with a proper octicon:',
'{% octicon "alert" aria-label="alert" %}',
'Some more text.',
]
rule.function({ lines: content }, onError)
expect(errors.length).toBe(0)
})
test('detects multiple octicons without aria-label', () => {
const errors = []
const onError = (errorInfo) => {
errors.push(errorInfo)
}
const content = [
'This is a test with multiple octicons:',
'{% octicon "alert" %}',
'Some text in between.',
'{% octicon "check" %}',
'More text.',
]
rule.function({ lines: content }, onError)
expect(errors.length).toBe(2)
expect(errors[0].lineNumber).toBe(2)
expect(errors[0].detail).toContain('aria-label=alert')
expect(errors[1].lineNumber).toBe(4)
expect(errors[1].detail).toContain('aria-label=check')
})
test('ignores non-octicon liquid tags', () => {
const errors = []
const onError = (errorInfo) => {
errors.push(errorInfo)
}
const content = [
'This is a test with non-octicon tags:',
'{% data foo.bar %}',
'{% ifversion fpt %}',
'Some text.',
'{% endif %}',
]
rule.function({ lines: content }, onError)
expect(errors.length).toBe(0)
})
test('suggests correct fix for octicon with other attributes', () => {
const errors = []
const onError = (errorInfo) => {
errors.push(errorInfo)
}
const content = [
'This is a test with an octicon with other attributes:',
'{% octicon "plus" aria-hidden="true" class="foo" %}',
'Some more text.',
]
rule.function({ lines: content }, onError)
expect(errors.length).toBe(1)
expect(errors[0].lineNumber).toBe(2)
expect(errors[0].fixInfo.insertText).toContain('aria-label="plus"')
expect(errors[0].fixInfo.insertText).toContain('aria-hidden="true"')
expect(errors[0].fixInfo.insertText).toContain('class="foo"')
})
test('handles octicons with unusual spacing', () => {
const errors = []
const onError = (errorInfo) => {
errors.push(errorInfo)
}
const content = [
'This is a test with unusual spacing:',
'{% octicon "x" %}',
'Some more text.',
]
rule.function({ lines: content }, onError)
expect(errors.length).toBe(1)
expect(errors[0].lineNumber).toBe(2)
expect(errors[0].detail).toContain('aria-label=x')
})
test('handles octicons split across multiple lines', () => {
const errors = []
const onError = (errorInfo) => {
errors.push(errorInfo)
}
const content = [
'This is a test with a multi-line octicon:',
'{% octicon "chevron-down"',
' class="dropdown-menu-icon"',
'%}',
'Some more text.',
]
rule.function({ lines: content }, onError)
expect(errors.length).toBe(1)
expect(errors[0].detail).toContain('aria-label=chevron-down')
})
test('falls back to "icon" when octicon name cannot be determined', () => {
const errors = []
const onError = (errorInfo) => {
errors.push(errorInfo)
}
const content = [
'This is a test with a malformed octicon:',
'{% octicon variable %}',
'Some more text.',
]
rule.function({ lines: content }, onError)
expect(errors.length).toBe(1)
expect(errors[0].detail).toContain('aria-label=icon')
expect(errors[0].fixInfo.insertText).toContain('aria-label="icon"')
})
})

4
src/frame/lib/page.js
View File

@@ -8,11 +8,11 @@ import { getAlertTitles } from '#src/languages/lib/get-alert-titles.ts'
import getTocItems from './get-toc-items.js'
import Permalink from './permalink.js'
import { renderContent } from '#src/content-render/index.js'
import processLearningTracks from '#src/learning-track/lib/process-learning-tracks.js'
import processLearningTracks from '#src/learning-track/lib/process-learning-tracks'
import { productMap } from '#src/products/lib/all-products.ts'
import slash from 'slash'
import readFileContents from './read-file-contents.js'
import getLinkData from '#src/learning-track/lib/get-link-data.js'
import getLinkData from '#src/learning-track/lib/get-link-data'
import getDocumentType from '#src/events/lib/get-document-type.ts'
import { allTools } from '#src/tools/lib/all-tools.ts'
import { renderContentWithFallback } from '#src/languages/lib/render-with-fallback.js'

18
src/landings/middleware/featured-links.ts
View File

@@ -1,8 +1,8 @@
import type { Response, NextFunction } from 'express'
import type { ExtendedRequest, FeaturedLinkExpanded } from '@/types'
import getLinkData from '@/learning-track/lib/get-link-data.js'
import { renderContent } from '@/content-render/index.js'
import getLinkData from '@/learning-track/lib/get-link-data'
import { renderContent } from '@/content-render/index'
/**
* This is the max. number of featured links, by any category, that we
@@ -73,12 +73,20 @@ export default async function featuredLinks(
if (!(key in req.context.page.featuredLinks))
throw new Error('featureLinks key not found in Page')
const pageFeaturedLink = req.context.page.featuredLinks[key]
req.context.featuredLinks[key] = (await getLinkData(
pageFeaturedLink,
// Handle different types of featuredLinks by converting to string array
const stringLinks = Array.isArray(pageFeaturedLink)
? pageFeaturedLink.map((item) => (typeof item === 'string' ? item : item.href))
: []
const linkData = await getLinkData(
stringLinks,
req.context,
{ title: true, intro: true, fullTitle: true },
MAX_FEATURED_LINKS,
)) as FeaturedLinkExpanded[] // Remove ones `getLinkData` is TS
)
// We need to use a type assertion here because the Page interfaces are incompatible
// between our local types and the global types, but the actual runtime objects are compatible
req.context.featuredLinks[key] = (linkData || []) as unknown as FeaturedLinkExpanded[]
}
}

57
src/learning-track/lib/get-link-data.js → src/learning-track/lib/get-link-data.ts
View File

@@ -1,32 +1,34 @@
import path from 'path'
import findPage from '#src/frame/lib/find-page.js'
import nonEnterpriseDefaultVersion from '#src/versions/lib/non-enterprise-default-version.js'
import removeFPTFromPath from '#src/versions/lib/remove-fpt-from-path.js'
import { renderContent } from '#src/content-render/index.js'
import { executeWithFallback } from '#src/languages/lib/render-with-fallback.js'
import findPage from '@/frame/lib/find-page'
import nonEnterpriseDefaultVersion from '@/versions/lib/non-enterprise-default-version'
import removeFPTFromPath from '@/versions/lib/remove-fpt-from-path'
import { renderContent } from '@/content-render/index'
import { executeWithFallback } from '@/languages/lib/render-with-fallback'
import { Context, LinkOptions, ProcessedLink } from './types'
// rawLinks is an array of paths: [ '/foo' ]
// we need to convert it to an array of localized objects: [ { href: '/en/foo', title: 'Foo', intro: 'Description here' } ]
export default async (
rawLinks,
context,
option = { title: true, intro: true, fullTitle: false },
export default async function getLinkData(
rawLinks: string[] | string | undefined,
context: Context,
options: LinkOptions = { title: true, intro: true, fullTitle: false },
maxLinks = Infinity,
) => {
if (!rawLinks) return
): Promise<ProcessedLink[] | undefined> {
if (!rawLinks) return undefined
if (typeof rawLinks === 'string') {
return await processLink(rawLinks, context, option)
const processedLink = await processLink(rawLinks, context, options)
return processedLink ? [processedLink] : undefined
}
const links = []
const links: ProcessedLink[] = []
// Using a for loop here because the async work is not network or
// disk bound. It's CPU bound.
// And if we use a for-loop we can potentially bail early if
// the `maxLinks` is reached. That's instead of computing them all,
// and then slicing the array. So it avoids wasted processing.
for (const link of rawLinks) {
const processedLink = await processLink(link, context, option)
const processedLink = await processLink(link, context, options)
if (processedLink) {
links.push(processedLink)
if (links.length >= maxLinks) {
@@ -38,9 +40,13 @@ export default async (
return links
}
async function processLink(link, context, option) {
const opts = { textOnly: true }
const linkHref = link.href || link
async function processLink(
link: string | { href: string },
context: Context,
options: LinkOptions,
): Promise<ProcessedLink | null> {
const opts: { textOnly: boolean; preferShort?: boolean } = { textOnly: true }
const linkHref = typeof link === 'string' ? link : link.href
// Parse the link in case it includes Liquid conditionals
const linkPath = linkHref.includes('{')
? await executeWithFallback(
@@ -55,10 +61,13 @@ async function processLink(link, context, option) {
if (!linkPath) return null
const version =
context.currentVersion === 'homepage' ? nonEnterpriseDefaultVersion : context.currentVersion
const href = removeFPTFromPath(path.join('/', context.currentLanguage, version, linkPath))
(context.currentVersion === 'homepage'
? nonEnterpriseDefaultVersion
: context.currentVersion) || 'free-pro-team@latest'
const currentLanguage = context.currentLanguage || 'en'
const href = removeFPTFromPath(path.join('/', currentLanguage, version, linkPath))
const linkedPage = findPage(href, context.pages, context.redirects)
const linkedPage = findPage(href, context.pages || {}, context.redirects || {})
if (!linkedPage) {
// This can happen when the link depends on Liquid conditionals,
// like...
@@ -66,18 +75,18 @@ async function processLink(link, context, option) {
return null
}
const result = { href, page: linkedPage }
const result: ProcessedLink = { href, page: linkedPage }
if (option.title) {
if (options.title) {
result.title = await linkedPage.renderTitle(context, opts)
}
if (option.fullTitle) {
if (options.fullTitle) {
opts.preferShort = false
result.fullTitle = await linkedPage.renderTitle(context, opts)
}
if (option.intro) {
if (options.intro) {
result.intro = await linkedPage.renderProp('intro', context, opts)
}
return result

32
src/learning-track/lib/process-learning-tracks.js → src/learning-track/lib/process-learning-tracks.ts
View File

@@ -1,15 +1,19 @@
import getLinkData from './get-link-data.js'
import getApplicableVersions from '#src/versions/lib/get-applicable-versions.js'
import { getDataByLanguage } from '#src/data-directory/lib/get-data.js'
import { renderContent } from '#src/content-render/index.js'
import { executeWithFallback } from '#src/languages/lib/render-with-fallback.js'
import getLinkData from './get-link-data'
import getApplicableVersions from '@/versions/lib/get-applicable-versions'
import { getDataByLanguage } from '@/data-directory/lib/get-data'
import { renderContent } from '@/content-render/index'
import { executeWithFallback } from '@/languages/lib/render-with-fallback'
import { Context, TrackGuide, LearningTrack, ProcessedLearningTracks } from './types'
const renderOpts = { textOnly: true }
// This module returns an object that contains a single featured learning track
// and an array of all the other learning tracks for the current version.
export default async function processLearningTracks(rawLearningTracks, context) {
const learningTracks = []
export default async function processLearningTracks(
rawLearningTracks: string[],
context: Context,
): Promise<ProcessedLearningTracks> {
const learningTracks: LearningTrack[] = []
if (!context.currentProduct) {
throw new Error(`Missing context.currentProduct value.`)
@@ -59,7 +63,7 @@ export default async function processLearningTracks(rawLearningTracks, context)
// we need to have the English `title` and `description` to
// fall back to.
//
let enTrack
let enTrack: any
if (context.currentLanguage !== 'en') {
enTrack = getDataByLanguage(
`learning-tracks.${context.currentProduct}.${renderedTrackName}`,
@@ -86,26 +90,28 @@ export default async function processLearningTracks(rawLearningTracks, context)
const title = await executeWithFallback(
context,
() => renderContent(track.title, context, renderOpts),
(enContext) => renderContent(enTrack.title, enContext, renderOpts),
(enContext: any) => renderContent(enTrack.title, enContext, renderOpts),
)
const description = await executeWithFallback(
context,
() => renderContent(track.description, context, renderOpts),
(enContext) => renderContent(enTrack.description, enContext, renderOpts),
(enContext: any) => renderContent(enTrack.description, enContext, renderOpts),
)
const learningTrack = {
const guides = (await getLinkData(track.guides, context)) || []
const learningTrack: LearningTrack = {
trackName: renderedTrackName,
trackProduct: context.currentProduct || null,
title,
description,
// getLinkData respects versioning and only returns guides available in the current version;
// if no guides are available, the learningTrack.guides property will be an empty array.
guides: await getLinkData(track.guides, context),
guides: guides as TrackGuide[],
}
// Only add the track to the array of tracks if there are guides in this version and it's not the featured track.
if (learningTrack.guides.length) {
if (Array.isArray(learningTrack.guides) && learningTrack.guides.length > 0) {
learningTracks.push(learningTrack)
}
}

122
src/learning-track/lib/types.ts Normal file
View File

@@ -0,0 +1,122 @@
/**
* Common types used across learning track components
*/
/**
* Basic context interface for rendering operations
*/
export interface Context {
currentProduct?: string
currentLanguage?: string
currentVersion?: string
pages?: any
redirects?: any
// Additional properties that may be needed for rendering
[key: string]: any
}
/**
* Options for retrieving link data
*/
export interface LinkOptions {
title?: boolean
intro?: boolean
fullTitle?: boolean
}
/**
* Result of processing a link
*/
export interface ProcessedLink {
href: string
page: Page
title?: string
fullTitle?: string
intro?: string
}
/**
* Definitions for featured links data
*/
export interface FeaturedLink {
title: string
href: string
}
export interface PageFeaturedLinks {
[key: string]: string[] | FeaturedLink[]
}
/**
* Page interface for basic page properties
*/
export interface Page {
renderTitle: (context: Context, opts: any) => Promise<string>
renderProp: (prop: string, context: Context, opts: any) => Promise<string>
}
/**
* Guide in a learning track
*/
export interface TrackGuide {
href: string
page: Page
title: string
intro?: string
}
/**
* A processed learning track
*/
export interface LearningTrack {
trackName: string
trackProduct: string | null
title: string
description: string
guides: TrackGuide[]
}
/**
* Learning track metadata with guides
*/
export interface LearningTrackMetadata {
title: string
description: string
guides: string[]
versions?: any
}
/**
* Collection of learning tracks by product and track name
*/
export interface LearningTracks {
[productId: string]: {
[trackName: string]: LearningTrackMetadata
}
}
/**
* Return type for processLearningTracks function
*/
export interface ProcessedLearningTracks {
learningTracks: LearningTrack[]
}
/**
* Learning track data for the current guide
*/
export interface CurrentLearningTrack {
trackName: string
trackProduct: string
trackTitle: string
numberOfGuides?: number
currentGuideIndex?: number
nextGuide?: {
href: string
title: string | undefined
}
prevGuide?: {
href: string
title: string | undefined
}
}

28
src/learning-track/middleware/learning-track.ts
View File

@@ -1,28 +1,22 @@
import type { Response, NextFunction } from 'express'
import type {
Context,
ExtendedRequest,
LearningTrack,
LearningTracks,
TrackGuide,
Page,
} from '@/types'
import type { ExtendedRequest, LearningTracks } from '@/types'
import type { Context, CurrentLearningTrack, TrackGuide } from '../lib/types'
import { getPathWithoutLanguage, getPathWithoutVersion } from '@/frame/lib/path-utils.js'
import getLinkData from '../lib/get-link-data.js'
import getLinkData from '../lib/get-link-data'
import { renderContent } from '@/content-render/index.js'
import { executeWithFallback } from '@/languages/lib/render-with-fallback.js'
import { getDeepDataByLanguage } from '@/data-directory/lib/get-data.js'
export default async function learningTrack(
req: ExtendedRequest,
req: ExtendedRequest & { context: Context },
res: Response,
next: NextFunction,
) {
if (!req.context) throw new Error('request is not contextualized')
const noTrack = () => {
req.context!.currentLearningTrack = null
req.context.currentLearningTrack = null
return next()
}
@@ -94,12 +88,12 @@ export default async function learningTrack(
() => '', // todo use english track.title
)
const currentLearningTrack: LearningTrack = { trackName, trackProduct, trackTitle }
const currentLearningTrack: CurrentLearningTrack = { trackName, trackProduct, trackTitle }
const guidePath = getPathWithoutLanguage(getPathWithoutVersion(req.pagePath))
// The raw track.guides will return all guide paths, need to use getLinkData
// so we only get guides available in the current version
const trackGuides = (await getLinkData(track.guides, req.context)) as TrackGuide[]
const trackGuides = ((await getLinkData(track.guides, req.context)) || []) as TrackGuide[]
const trackGuidePaths = trackGuides.map((guide) => {
return getPathWithoutLanguage(getPathWithoutVersion(guide.href))
@@ -137,8 +131,8 @@ export default async function learningTrack(
intro: false,
fullTitle: false,
})
if (!resultData) return noTrack()
const result = resultData as { href: string; page: Page; title: string }
if (!resultData || !resultData.length) return noTrack()
const result = resultData[0]
const href = result.href
const title = result.title
@@ -152,8 +146,8 @@ export default async function learningTrack(
intro: false,
fullTitle: false,
})
if (!resultData) return noTrack()
const result = resultData as { href: string; page: Page; title: string }
if (!resultData || !resultData.length) return noTrack()
const result = resultData[0]
const href = result.href
const title = result.title

120
src/search/components/input/AskAIResults.tsx
View File

@@ -232,18 +232,75 @@ export function AskAIResults({
const decoder = new TextDecoder('utf-8')
const reader = response.body.getReader()
let done = false
let leftover = '' // <= carry‑over buffer
setInitialLoading(false)
const processLine = (parsedLine: any) => {
switch (parsedLine.chunkType) {
// A conversation ID will still be sent when a question cannot be answered
case 'CONVERSATION_ID':
conversationIdBuffer = parsedLine.conversation_id
setConversationId(parsedLine.conversation_id)
break
case 'NO_CONTENT_SIGNAL':
// Serve canned response. A question that cannot be answered was asked
handleAICannotAnswer(conversationIdBuffer, 200)
break
case 'SOURCES':
if (!isCancelled) {
sourcesBuffer = uniqBy(
sourcesBuffer.concat(parsedLine.sources as AIReference[]),
'url',
)
setReferences(sourcesBuffer)
}
break
case 'MESSAGE_CHUNK':
if (!isCancelled) {
messageBuffer += parsedLine.text
setMessage(messageBuffer)
}
break
case 'INPUT_CONTENT_FILTER':
// Serve canned response. A spam question was asked
handleAICannotAnswer(
conversationIdBuffer,
200,
t('search.ai.responses.invalid_query'),
)
break
}
if (!isCancelled) setAnnouncement('Copilot Response Loading...')
}
while (!done && !isCancelled) {
const { value, done: readerDone } = await reader.read()
done = readerDone
// The sources JSON chunk may be sent in multiple parts, so we need to decode it with a leftover buffer so that it can be parsed all at once
// So when we say "incomplete" or "leftover" we mean that the JSON is not complete yet, not that the message is incomplete
if (value) {
const chunkStr = decoder.decode(value, { stream: true })
const chunkLines = chunkStr.split('\n').filter((line) => line.trim() !== '')
for (const line of chunkLines) {
let parsedLine
// 1 append this chunk's text to whatever was left over
leftover += decoder.decode(value, { stream: true })
// 2 split on newline
const lines = leftover.split('\n')
// 3 keep the *last* item (maybe incomplete) for next round
leftover = lines.pop() ?? ''
// 4 parse all complete lines
for (const raw of lines) {
if (!raw.trim()) continue
let parsedLine: any
try {
parsedLine = JSON.parse(line)
// If midstream there is an error, like a connection reset / lost, our backend will send an error JSON
parsedLine = JSON.parse(raw)
if (parsedLine?.errors) {
sendAISearchResultEvent({
sources: [],
@@ -255,50 +312,25 @@ export function AskAIResults({
setAISearchError()
return
}
} catch (e) {
console.warn(
'Failed to parse JSON:',
e,
'Line:',
line,
'Typeof line: ',
typeof line,
)
} catch (err) {
console.warn('Failed to parse JSON line:', raw, err)
continue
}
// A conversation ID will still be sent when a question cannot be answered
if (parsedLine.chunkType === 'CONVERSATION_ID') {
conversationIdBuffer = parsedLine.conversation_id
setConversationId(parsedLine.conversation_id)
} else if (parsedLine.chunkType === 'NO_CONTENT_SIGNAL') {
// Serve canned response. A question that cannot be answered was asked
handleAICannotAnswer(conversationIdBuffer, 200)
} else if (parsedLine.chunkType === 'SOURCES') {
if (!isCancelled) {
sourcesBuffer = sourcesBuffer.concat(parsedLine.sources)
sourcesBuffer = uniqBy(sourcesBuffer, 'url')
setReferences(sourcesBuffer)
}
} else if (parsedLine.chunkType === 'MESSAGE_CHUNK') {
if (!isCancelled) {
messageBuffer += parsedLine.text
setMessage(messageBuffer)
}
} else if (parsedLine.chunkType === 'INPUT_CONTENT_FILTER') {
// Serve canned response. A spam question was asked
handleAICannotAnswer(
conversationIdBuffer,
200,
t('search.ai.responses.invalid_query'),
)
}
if (!isCancelled) {
setAnnouncement('Copilot Response Loading...')
}
processLine(parsedLine)
}
}
}
// 5 flush whatever remains after the stream ends
if (!isCancelled && leftover.trim()) {
try {
const tail = JSON.parse(leftover)
processLine(tail)
} catch (err) {
console.warn('Failed to parse tail JSON:', leftover, err)
}
}
} catch (error: any) {
if (!isCancelled) {
console.error('Failed to fetch search results:', error)

7
src/search/components/input/SearchOverlay.tsx
View File

@@ -482,6 +482,9 @@ export function SearchOverlay({
}
}
} else if (event.key === 'Enter') {
if (searchLoading) {
return
}
event.preventDefault()
let pressedGroupKey = SEARCH_OVERLAY_EVENT_GROUP
let pressedGroupId = searchEventGroupId
@@ -494,9 +497,7 @@ export function SearchOverlay({
pressedGroupId = askAIEventGroupId
sendKeyboardEvent(event.key, pressedOnContext, pressedGroupId, pressedGroupKey)
aiSearchOptionOnSelect({ term: urlSearchInputQuery } as AutocompleteSearchHit)
}
if (
} else if (
combinedOptions.length > 0 &&
selectedIndex >= 0 &&
selectedIndex < combinedOptions.length

4
src/types.ts
View File

@@ -185,11 +185,11 @@ export type LearningTrack = {
currentGuideIndex?: number
nextGuide?: {
href: string
title: string
title: string | undefined
}
prevGuide?: {
href: string
title: string
title: string | undefined
}
}