mirror of
https://github.com/stablyai/orca.git
synced 2026-10-17 08:03:14 +00:00
78519d8dce
* fix: address pr-bug-scan validated finding from #6315 Protect fenced/inline code spans from the HTML stripTags pass in normalizeMobileMarkdownPreviewHtml; blocks stripTags(next) at mobile-markdown-preview-html.ts:95 from deleting <...> inside code. * Preserve non-tag angle brackets in markdown preview HTML parser Only strip standard/known HTML tags instead of matching any `<...>` pattern. This prevents false-positive stripping of generic types (e.g., `Array<string>`) or math comparisons (e.g., `1 < 2`) in prose. Additionally, robustify markdown code block preservation by using a Private Use Area prefix (`\uE000`) and dynamically padding it to avoid collisions with literal input content. * Protect code spans when stripping HTML tags in mobile preview Ensure HTML-like content within code spans is not erroneously removed by the tag stripping logic. We now protect markdown code blocks and restore them after stripping tags and decoding HTML entities. --------- Co-authored-by: orca-bug-scan-bot <orca-bug-scan-bot@stably.ai> Co-authored-by: Jinjing <6427696+AmethystLiang@users.noreply.github.com>
95 lines
3.6 KiB
TypeScript
95 lines
3.6 KiB
TypeScript
import { describe, expect, it } from 'vitest'
|
|
import { normalizeMobileMarkdownPreviewHtml } from './mobile-markdown-preview-html'
|
|
import { parseMobileMarkdown } from './mobile-markdown-parser'
|
|
|
|
describe('parseMobileMarkdown', () => {
|
|
it('parses GFM tables into table blocks', () => {
|
|
expect(parseMobileMarkdown('| Name | State |\n| --- | --- |\n| Orca | Open |')).toEqual([
|
|
{
|
|
type: 'table',
|
|
headers: ['Name', 'State'],
|
|
rows: [['Orca', 'Open']]
|
|
}
|
|
])
|
|
})
|
|
|
|
it('parses standalone HTTPS images without folding them into paragraphs', () => {
|
|
expect(parseMobileMarkdown('')).toEqual([
|
|
{
|
|
type: 'image',
|
|
alt: 'Screenshot',
|
|
url: 'https://example.com/screen.png'
|
|
}
|
|
])
|
|
})
|
|
|
|
it('normalizes common README HTML into readable Markdown preview text', () => {
|
|
const normalized = normalizeMobileMarkdownPreviewHtml(`
|
|
<h1 align="center">
|
|
<a href="https://onOrca.dev"><img src="resources/build/icon.png" alt="Orca" width="64" /></a>
|
|
Orca
|
|
</h1>
|
|
|
|
<p align="center">
|
|
<a href="https://github.com/stablyai/orca/stargazers"><img src="https://badgen.net/github/stars/stablyai/orca" alt="GitHub stars" /></a>
|
|
<strong>The AI Orchestrator</strong><br/>
|
|
Run Codex side-by-side.
|
|
</p>
|
|
`)
|
|
|
|
expect(normalized).toContain('# [Orca](https://onOrca.dev)')
|
|
expect(normalized).toContain('[GitHub stars](https://github.com/stablyai/orca/stargazers)')
|
|
expect(normalized).toContain('**The AI Orchestrator**')
|
|
expect(normalized).not.toContain('<h1')
|
|
expect(normalized).not.toContain('<img')
|
|
})
|
|
|
|
it('preserves documented HTML entities while normalizing preview HTML', () => {
|
|
expect(
|
|
normalizeMobileMarkdownPreviewHtml('<p>Use <code>&lt;button&gt;</code></p>')
|
|
).toBe('Use `<button>`')
|
|
})
|
|
|
|
it('preserves angle brackets and generics inside fenced and inline code', () => {
|
|
expect(normalizeMobileMarkdownPreviewHtml('```html\n<div>x</div>\n```')).toBe(
|
|
'```html\n<div>x</div>\n```'
|
|
)
|
|
expect(normalizeMobileMarkdownPreviewHtml('```ts\nconst x: Array<string> = []\n```')).toBe(
|
|
'```ts\nconst x: Array<string> = []\n```'
|
|
)
|
|
expect(normalizeMobileMarkdownPreviewHtml('```ts\nconst x: Array<string> = []')).toBe(
|
|
'```ts\nconst x: Array<string> = []'
|
|
)
|
|
expect(normalizeMobileMarkdownPreviewHtml('Use `Array<string>` here')).toBe(
|
|
'Use `Array<string>` here'
|
|
)
|
|
})
|
|
|
|
it('preserves non-tag angle bracket prose while stripping known HTML tags', () => {
|
|
expect(normalizeMobileMarkdownPreviewHtml('1 < 2 and 3 > 1')).toBe('1 < 2 and 3 > 1')
|
|
expect(normalizeMobileMarkdownPreviewHtml('Array<string> in prose')).toBe(
|
|
'Array<string> in prose'
|
|
)
|
|
expect(normalizeMobileMarkdownPreviewHtml('<https://example.com>')).toBe(
|
|
'<https://example.com>'
|
|
)
|
|
expect(normalizeMobileMarkdownPreviewHtml('<div>Readable text</div>')).toBe('Readable text')
|
|
})
|
|
|
|
it('preserves angle brackets inside Markdown code produced from HTML code tags', () => {
|
|
expect(normalizeMobileMarkdownPreviewHtml('<p>Use <code>Array<string></code></p>')).toBe(
|
|
'Use `Array<string>`'
|
|
)
|
|
expect(
|
|
normalizeMobileMarkdownPreviewHtml('<p>Use <code><div>x</div></code></p>')
|
|
).toBe('Use `<div>x</div>`')
|
|
})
|
|
|
|
it('does not replace literal code placeholder text in markdown prose', () => {
|
|
const literalPlaceholder = '\uE000ORCA_MD_CODE_0\uE000'
|
|
expect(normalizeMobileMarkdownPreviewHtml(`${literalPlaceholder} and \`Array<string>\``)).toBe(
|
|
`${literalPlaceholder} and \`Array<string>\``
|
|
)
|
|
})
|
|
})
|