@atproto/api 0.1.3 → 0.2.0

package/tests/errors.test.ts

@@ -3,7 +3,7 @@ import {
   runTestServer,
   TestServerInfo,
 } from '@atproto/pds/tests/_util'
-import { AtpAgent, ComAtprotoAccountCreate } from '..'
+import { AtpAgent, ComAtprotoServerCreateAccount } from '..'
 
 describe('errors', () => {
   let server: TestServerInfo
@@ -23,13 +23,13 @@ describe('errors', () => {
   })
 
   it('constructs the correct error instance', async () => {
-    const res = client.api.com.atproto.account.create({
-      handle: 'admin',
+    const res = client.api.com.atproto.server.createAccount({
+      handle: 'admin.blah',
       email: 'admin@test.com',
       password: 'password',
     })
     await expect(res).rejects.toThrow(
-      ComAtprotoAccountCreate.InvalidHandleError,
+      ComAtprotoServerCreateAccount.UnsupportedDomainError,
     )
   })
 })

package/tests/rich-text-detection.test.ts

@@ -0,0 +1,229 @@
+import { AtpAgent, RichText, RichTextSegment } from '../src'
+
+describe('detectFacets', () => {
+  const agent = new AtpAgent({ service: 'http://localhost' })
+  agent.resolveHandle = ({ handle }: { handle: string }) => {
+    return Promise.resolve({
+      success: true,
+      headers: {},
+      data: { did: 'did:fake:' + handle },
+    })
+  }
+
+  const inputs = [
+    'no mention',
+    '@handle.com middle end',
+    'start @handle.com end',
+    'start middle @handle.com',
+    '@handle.com @handle.com @handle.com',
+    '@full123-chars.test',
+    'not@right',
+    '@handle.com!@#$chars',
+    '@handle.com\n@handle.com',
+    'parenthetical (@handle.com)',
+    '👨‍👩‍👧‍👧 @handle.com 👨‍👩‍👧‍👧',
+
+    'start https://middle.com end',
+    'start https://middle.com/foo/bar end',
+    'start https://middle.com/foo/bar?baz=bux end',
+    'start https://middle.com/foo/bar?baz=bux#hash end',
+    'https://start.com/foo/bar?baz=bux#hash middle end',
+    'start middle https://end.com/foo/bar?baz=bux#hash',
+    'https://newline1.com\nhttps://newline2.com',
+    '👨‍👩‍👧‍👧 https://middle.com 👨‍👩‍👧‍👧',
+
+    'start middle.com end',
+    'start middle.com/foo/bar end',
+    'start middle.com/foo/bar?baz=bux end',
+    'start middle.com/foo/bar?baz=bux#hash end',
+    'start.com/foo/bar?baz=bux#hash middle end',
+    'start middle end.com/foo/bar?baz=bux#hash',
+    'newline1.com\nnewline2.com',
+
+    'not.. a..url ..here',
+    'e.g.',
+    'something-cool.jpg',
+    'website.com.jpg',
+    'e.g./foo',
+    'website.com.jpg/foo',
+
+    'Classic article https://socket3.wordpress.com/2018/02/03/designing-windows-95s-user-interface/',
+    'Classic article https://socket3.wordpress.com/2018/02/03/designing-windows-95s-user-interface/ ',
+    'https://foo.com https://bar.com/whatever https://baz.com',
+    'punctuation https://foo.com, https://bar.com/whatever; https://baz.com.',
+    'parenthentical (https://foo.com)',
+    'except for https://foo.com/thing_(cool)',
+  ]
+  const outputs: string[][][] = [
+    [['no mention']],
+    [['@handle.com', 'did:fake:handle.com'], [' middle end']],
+    [['start '], ['@handle.com', 'did:fake:handle.com'], [' end']],
+    [['start middle '], ['@handle.com', 'did:fake:handle.com']],
+    [
+      ['@handle.com', 'did:fake:handle.com'],
+      [' '],
+      ['@handle.com', 'did:fake:handle.com'],
+      [' '],
+      ['@handle.com', 'did:fake:handle.com'],
+    ],
+    [['@full123-chars.test', 'did:fake:full123-chars.test']],
+    [['not@right']],
+    [['@handle.com', 'did:fake:handle.com'], ['!@#$chars']],
+    [
+      ['@handle.com', 'did:fake:handle.com'],
+      ['\n'],
+      ['@handle.com', 'did:fake:handle.com'],
+    ],
+    [['parenthetical ('], ['@handle.com', 'did:fake:handle.com'], [')']],
+    [['👨‍👩‍👧‍👧 '], ['@handle.com', 'did:fake:handle.com'], [' 👨‍👩‍👧‍👧']],
+
+    [['start '], ['https://middle.com', 'https://middle.com'], [' end']],
+    [
+      ['start '],
+      ['https://middle.com/foo/bar', 'https://middle.com/foo/bar'],
+      [' end'],
+    ],
+    [
+      ['start '],
+      [
+        'https://middle.com/foo/bar?baz=bux',
+        'https://middle.com/foo/bar?baz=bux',
+      ],
+      [' end'],
+    ],
+    [
+      ['start '],
+      [
+        'https://middle.com/foo/bar?baz=bux#hash',
+        'https://middle.com/foo/bar?baz=bux#hash',
+      ],
+      [' end'],
+    ],
+    [
+      [
+        'https://start.com/foo/bar?baz=bux#hash',
+        'https://start.com/foo/bar?baz=bux#hash',
+      ],
+      [' middle end'],
+    ],
+    [
+      ['start middle '],
+      [
+        'https://end.com/foo/bar?baz=bux#hash',
+        'https://end.com/foo/bar?baz=bux#hash',
+      ],
+    ],
+    [
+      ['https://newline1.com', 'https://newline1.com'],
+      ['\n'],
+      ['https://newline2.com', 'https://newline2.com'],
+    ],
+    [['👨‍👩‍👧‍👧 '], ['https://middle.com', 'https://middle.com'], [' 👨‍👩‍👧‍👧']],
+
+    [['start '], ['middle.com', 'https://middle.com'], [' end']],
+    [
+      ['start '],
+      ['middle.com/foo/bar', 'https://middle.com/foo/bar'],
+      [' end'],
+    ],
+    [
+      ['start '],
+      ['middle.com/foo/bar?baz=bux', 'https://middle.com/foo/bar?baz=bux'],
+      [' end'],
+    ],
+    [
+      ['start '],
+      [
+        'middle.com/foo/bar?baz=bux#hash',
+        'https://middle.com/foo/bar?baz=bux#hash',
+      ],
+      [' end'],
+    ],
+    [
+      [
+        'start.com/foo/bar?baz=bux#hash',
+        'https://start.com/foo/bar?baz=bux#hash',
+      ],
+      [' middle end'],
+    ],
+    [
+      ['start middle '],
+      ['end.com/foo/bar?baz=bux#hash', 'https://end.com/foo/bar?baz=bux#hash'],
+    ],
+    [
+      ['newline1.com', 'https://newline1.com'],
+      ['\n'],
+      ['newline2.com', 'https://newline2.com'],
+    ],
+
+    [['not.. a..url ..here']],
+    [['e.g.']],
+    [['something-cool.jpg']],
+    [['website.com.jpg']],
+    [['e.g./foo']],
+    [['website.com.jpg/foo']],
+
+    [
+      ['Classic article '],
+      [
+        'https://socket3.wordpress.com/2018/02/03/designing-windows-95s-user-interface/',
+        'https://socket3.wordpress.com/2018/02/03/designing-windows-95s-user-interface/',
+      ],
+    ],
+    [
+      ['Classic article '],
+      [
+        'https://socket3.wordpress.com/2018/02/03/designing-windows-95s-user-interface/',
+        'https://socket3.wordpress.com/2018/02/03/designing-windows-95s-user-interface/',
+      ],
+      [' '],
+    ],
+    [
+      ['https://foo.com', 'https://foo.com'],
+      [' '],
+      ['https://bar.com/whatever', 'https://bar.com/whatever'],
+      [' '],
+      ['https://baz.com', 'https://baz.com'],
+    ],
+    [
+      ['punctuation '],
+      ['https://foo.com', 'https://foo.com'],
+      [', '],
+      ['https://bar.com/whatever', 'https://bar.com/whatever'],
+      ['; '],
+      ['https://baz.com', 'https://baz.com'],
+      ['.'],
+    ],
+    [['parenthentical ('], ['https://foo.com', 'https://foo.com'], [')']],
+    [
+      ['except for '],
+      ['https://foo.com/thing_(cool)', 'https://foo.com/thing_(cool)'],
+    ],
+  ]
+  it('correctly handles a set of text inputs', async () => {
+    for (let i = 0; i < inputs.length; i++) {
+      const input = inputs[i]
+      const rt = new RichText({ text: input })
+      await rt.detectFacets(agent)
+      expect(Array.from(rt.segments(), segmentToOutput)).toEqual(outputs[i])
+    }
+  })
+})
+
+function segmentToOutput(segment: RichTextSegment): string[] {
+  if (segment.facet) {
+    return [
+      segment.text,
+      segment.facet?.features.map((f) => {
+        if (f.did) {
+          return String(f.did)
+        }
+        if (f.uri) {
+          return String(f.uri)
+        }
+        return undefined
+      })?.[0] || '',
+    ]
+  }
+  return [segment.text]
+}

package/tests/rich-text-sanitization.test.ts

@@ -0,0 +1,211 @@
+import { RichText, sanitizeRichText, Facet, UnicodeString } from '../src'
+
+describe('sanitizeRichText: cleanNewlines', () => {
+  it('removes more than two consecutive new lines', () => {
+    const input = new RichText({
+      text: 'test\n\n\n\n\ntest\n\n\n\n\n\n\ntest\n\n\n\n\n\n\ntest\n\n\n\n\n\n\ntest',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test\n\ntest\n\ntest\n\ntest\n\ntest',
+    )
+  })
+
+  it('removes more than two consecutive new lines w/fat unicode', () => {
+    const input = new RichText({
+      text: 'test👨‍👩‍👧‍👧\n\n\n\n\n👨‍👩‍👧‍👧test\n\n\n\n\n\n\ntest👨‍👩‍👧‍👧\n\n\n\n\n\n\ntest\n\n\n\n\n\n\n👨‍👩‍👧‍👧test',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test👨‍👩‍👧‍👧\n\n👨‍👩‍👧‍👧test\n\ntest👨‍👩‍👧‍👧\n\ntest\n\n👨‍👩‍👧‍👧test',
+    )
+  })
+
+  it('removes more than two consecutive new lines with spaces', () => {
+    const input = new RichText({
+      text: 'test\n\n\n\n\ntest\n \n \n \n \n\n\ntest\n\n\n\n\n\n\ntest\n\n\n\n\n  \n\ntest',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test\n\ntest\n\ntest\n\ntest\n\ntest',
+    )
+  })
+
+  it('returns original string if there are no consecutive new lines', () => {
+    const input = new RichText({ text: 'test\n\ntest\n\ntest\n\ntest\n\ntest' })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(String(input.unicodeText))
+  })
+
+  it('returns original string if there are no new lines', () => {
+    const input = new RichText({ text: 'test test          test test test' })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(String(input.unicodeText))
+  })
+
+  it('returns empty string if input is empty', () => {
+    const input = new RichText({ text: '' })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual('')
+  })
+
+  it('works with different types of new line characters', () => {
+    const input = new RichText({
+      text: 'test\r\ntest\n\rtest\rtest\n\n\n\ntest\n\r \n \n \n \n\n\ntest',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test\r\ntest\n\rtest\rtest\n\ntest\n\ntest',
+    )
+  })
+
+  it('removes more than two consecutive new lines with zero width space', () => {
+    const input = new RichText({
+      text: 'test\n\n\n\n\ntest\n\u200B\u200B\n\n\n\ntest\n \u200B\u200B \n\n\n\ntest\n\n\n\n\n\n\ntest',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test\n\ntest\n\ntest\n\ntest\n\ntest',
+    )
+  })
+
+  it('removes more than two consecutive new lines with zero width non-joiner', () => {
+    const input = new RichText({
+      text: 'test\n\n\n\n\ntest\n\u200C\u200C\n\n\n\ntest\n \u200C\u200C \n\n\n\ntest\n\n\n\n\n\n\ntest',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test\n\ntest\n\ntest\n\ntest\n\ntest',
+    )
+  })
+
+  it('removes more than two consecutive new lines with zero width joiner', () => {
+    const input = new RichText({
+      text: 'test\n\n\n\n\ntest\n\u200D\u200D\n\n\n\ntest\n \u200D\u200D \n\n\n\ntest\n\n\n\n\n\n\ntest',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test\n\ntest\n\ntest\n\ntest\n\ntest',
+    )
+  })
+
+  it('removes more than two consecutive new lines with soft hyphen', () => {
+    const input = new RichText({
+      text: 'test\n\n\n\n\ntest\n\u00AD\u00AD\n\n\n\ntest\n \u00AD\u00AD \n\n\n\ntest\n\n\n\n\n\n\ntest',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test\n\ntest\n\ntest\n\ntest\n\ntest',
+    )
+  })
+
+  it('removes more than two consecutive new lines with word joiner', () => {
+    const input = new RichText({
+      text: 'test\n\n\n\n\ntest\n\u2060\u2060\n\n\n\ntest\n \u2060\u2060 \n\n\n\ntest\n\n\n\n\n\n\ntest',
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(String(output.unicodeText)).toEqual(
+      'test\n\ntest\n\ntest\n\ntest\n\ntest',
+    )
+  })
+})
+
+describe('sanitizeRichText w/facets: cleanNewlines', () => {
+  it('preserves entities as expected', () => {
+    const input = new RichText({
+      text: 'test\n\n\n\n\ntest\n\n\n\n\n\n\ntest\n\n\n\n\n\n\ntest\n\n\n\n\n\n\ntest',
+      facets: [
+        { index: { byteStart: 0, byteEnd: 13 }, features: [{ $type: '' }] },
+        { index: { byteStart: 13, byteEnd: 24 }, features: [{ $type: '' }] },
+        { index: { byteStart: 9, byteEnd: 15 }, features: [{ $type: '' }] },
+        { index: { byteStart: 4, byteEnd: 9 }, features: [{ $type: '' }] },
+      ],
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(facetToStr(String(input.unicodeText), input.facets?.[0])).toEqual(
+      'test\n\n\n\n\ntest',
+    )
+    expect(facetToStr(String(input.unicodeText), input.facets?.[1])).toEqual(
+      '\n\n\n\n\n',
+    )
+    expect(facetToStr(String(input.unicodeText), input.facets?.[2])).toEqual(
+      'test\n\n',
+    )
+    expect(facetToStr(String(input.unicodeText), input.facets?.[3])).toEqual(
+      '\n\n\n\n\n\n\ntest',
+    )
+    expect(String(output.unicodeText)).toEqual(
+      'test\n\ntest\n\ntest\n\ntest\n\ntest',
+    )
+    expect(facetToStr(String(output.unicodeText), output.facets?.[0])).toEqual(
+      'test\n\ntest',
+    )
+    expect(facetToStr(String(output.unicodeText), output.facets?.[1])).toEqual(
+      'test',
+    )
+    expect(facetToStr(String(output.unicodeText), output.facets?.[2])).toEqual(
+      'test',
+    )
+    expect(output.facets?.[3]).toEqual(undefined)
+  })
+
+  it('preserves entities as expected w/fat unicode', () => {
+    const str = new UnicodeString(
+      '👨‍👩‍👧‍👧test\n\n\n\n\n👨‍👩‍👧‍👧test\n\n\n\n\n👨‍👩‍👧‍👧test\n\n\n\n\n👨‍👩‍👧‍👧test\n\n\n\n\n👨‍👩‍👧‍👧test\n\n\n\n\n👨‍👩‍👧‍👧test\n\n\n\n\n👨‍👩‍👧‍👧test\n\n\n\n\n',
+    )
+    let lastI = 0
+    const makeFacet = (match: string) => {
+      const i = str.utf16.indexOf(match, lastI)
+      lastI = i + match.length
+      const byteStart = str.utf16IndexToUtf8Index(i)
+      const byteEnd = byteStart + new UnicodeString(match).length
+      return {
+        index: { byteStart, byteEnd },
+        features: [{ $type: '' }],
+      }
+    }
+
+    const input = new RichText({
+      text: str.utf16,
+      facets: [
+        makeFacet('👨‍👩‍👧‍👧test\n\n\n\n\n👨‍👩‍👧‍👧test'),
+        makeFacet('\n\n\n\n\n👨‍👩‍👧‍👧test'),
+        makeFacet('👨‍👩‍👧‍👧test\n\n'),
+        makeFacet('\n\n'),
+      ],
+    })
+    const output = sanitizeRichText(input, { cleanNewlines: true })
+    expect(facetToStr(String(input.unicodeText), input.facets?.[0])).toEqual(
+      '👨‍👩‍👧‍👧test\n\n\n\n\n👨‍👩‍👧‍👧test',
+    )
+    expect(facetToStr(String(input.unicodeText), input.facets?.[1])).toEqual(
+      '\n\n\n\n\n👨‍👩‍👧‍👧test',
+    )
+    expect(facetToStr(String(input.unicodeText), input.facets?.[2])).toEqual(
+      '👨‍👩‍👧‍👧test\n\n',
+    )
+    expect(facetToStr(String(input.unicodeText), input.facets?.[3])).toEqual(
+      '\n\n',
+    )
+    expect(String(output.unicodeText)).toEqual(
+      '👨‍👩‍👧‍👧test\n\n👨‍👩‍👧‍👧test\n\n👨‍👩‍👧‍👧test\n\n👨‍👩‍👧‍👧test\n\n👨‍👩‍👧‍👧test\n\n👨‍👩‍👧‍👧test\n\n👨‍👩‍👧‍👧test\n\n',
+    )
+    expect(facetToStr(String(output.unicodeText), output.facets?.[0])).toEqual(
+      '👨‍👩‍👧‍👧test\n\n👨‍👩‍👧‍👧test',
+    )
+    expect(facetToStr(String(output.unicodeText), output.facets?.[1])).toEqual(
+      '👨‍👩‍👧‍👧test',
+    )
+    expect(facetToStr(String(output.unicodeText), output.facets?.[2])).toEqual(
+      '👨‍👩‍👧‍👧test',
+    )
+    expect(output.facets?.[3]).toEqual(undefined)
+  })
+})
+
+function facetToStr(str: string, ent?: Facet) {
+  if (!ent) {
+    return ''
+  }
+  return new UnicodeString(str).slice(ent.index.byteStart, ent.index.byteEnd)
+}