securemark 0.229.0 → 0.231.0

package/src/parser/inline/autolink/email.ts

@@ -6,6 +6,6 @@ import { html } from 'typed-dom';
 // https://html.spec.whatwg.org/multipage/input.html
 
 export const email: AutolinkParser.EmailParser = creator(rewrite(verify(
-  str(/^[0-9A-Za-z]+(?:[.+_-][0-9A-Za-z]+)*@[0-9A-Za-z](?:[0-9A-Za-z-]{0,61}[0-9A-Za-z])?(?:\.[0-9A-Za-z](?:[0-9A-Za-z-]{0,61}[0-9A-Za-z])?)*/),
+  str(/^[0-9A-Za-z]+(?:[.+_-][0-9A-Za-z]+)*@[0-9A-Za-z](?:(?:[0-9A-Za-z]|-(?=\w)){0,61}[0-9A-Za-z])?(?:\.[0-9A-Za-z](?:(?:[0-9A-Za-z]|-(?=\w)){0,61}[0-9A-Za-z])?)*(?![0-9A-Za-z])/),
   ([source]) => source.indexOf('@') <= 64 && source.length <= 255),
   source => [[html('a', { class: 'email', href: `mailto:${source}` }, source)], '']));

package/src/parser/inline/autolink/hashnum.test.ts

@@ -35,6 +35,7 @@ describe('Unit: parser/inline/autolink/hashnum', () => {
       assert.deepStrictEqual(inspect(parser('あ#1')), [['あ#1'], '']);
       assert.deepStrictEqual(inspect(parser(' #1')), undefined);
       assert.deepStrictEqual(inspect(parser('#12345678901234567')), [['#12345678901234567'], '']);
+      assert.deepStrictEqual(inspect(parser(`#${'1'.repeat(128)}a`)), [[`#${'1'.repeat(128)}a`], '']);
     });
 
     it('valid', () => {

package/src/parser/inline/autolink/hashnum.ts

@@ -1,11 +1,12 @@
 import { AutolinkParser } from '../../inline';
 import { union, rewrite, context, open, convert, fmap, lazy } from '../../../combinator';
 import { link } from '../link';
+import { emoji } from './hashtag';
 import { str } from '../../source';
 import { define } from 'typed-dom';
 
 export const hashnum: AutolinkParser.HashnumParser = lazy(() => fmap(rewrite(
-  open('#', str(/^[0-9]{1,16}(?![0-9A-Za-z'_]|[^\x00-\x7F\s])/)),
+  open('#', str(new RegExp(String.raw`^[0-9]{1,16}(?![^\p{C}\p{S}\p{P}\s]|${emoji}|['_])`, 'u'))),
   context({ syntax: { inline: {
     link: true,
     autolink: false,

package/src/parser/inline/autolink/hashtag.test.ts

@@ -23,9 +23,6 @@ describe('Unit: parser/inline/autolink/hashtag', () => {
       assert.deepStrictEqual(inspect(parser(`#'`)), [[`#'`], '']);
       assert.deepStrictEqual(inspect(parser(`#a''`)), [[`#a''`], '']);
       assert.deepStrictEqual(inspect(parser('#_')), [['#_'], '']);
-      assert.deepStrictEqual(inspect(parser('#a_')), [['#a_'], '']);
-      assert.deepStrictEqual(inspect(parser('#a__b')), [['#a__b'], '']);
-      assert.deepStrictEqual(inspect(parser(`#a_'b`)), [[`#a_'b`], '']);
       assert.deepStrictEqual(inspect(parser('#(a)')), [['#'], '(a)']);
       assert.deepStrictEqual(inspect(parser('#{}')), [['#'], '{}']);
       assert.deepStrictEqual(inspect(parser('#{{}')), [['#'], '{{}']);
@@ -38,6 +35,9 @@ describe('Unit: parser/inline/autolink/hashtag', () => {
       assert.deepStrictEqual(inspect(parser('a##1')), [['a##1'], '']);
       assert.deepStrictEqual(inspect(parser('a##b')), [['a##b'], '']);
       assert.deepStrictEqual(inspect(parser('あ#b')), [['あ#b'], '']);
+      assert.deepStrictEqual(inspect(parser(`#${'1'.repeat(127)}_`)), [[`#${'1'.repeat(127)}_`], '']);
+      assert.deepStrictEqual(inspect(parser(`#${'1'.repeat(127)}_a`)), [[`#${'1'.repeat(127)}_a`], '']);
+      assert.deepStrictEqual(inspect(parser(`#${'1'.repeat(127)}_'a`)), [[`#${'1'.repeat(127)}_'a`], '']);
       assert.deepStrictEqual(inspect(parser(' #a')), undefined);
     });
 
@@ -50,21 +50,22 @@ describe('Unit: parser/inline/autolink/hashtag', () => {
       assert.deepStrictEqual(inspect(parser('#a\\\n')), [['<a href="/hashtags/a" class="hashtag">#a</a>'], '\\\n']);
       assert.deepStrictEqual(inspect(parser('#a)')), [['<a href="/hashtags/a" class="hashtag">#a</a>'], ')']);
       assert.deepStrictEqual(inspect(parser('#a(b')), [['<a href="/hashtags/a" class="hashtag">#a</a>'], '(b']);
-      assert.deepStrictEqual(inspect(parser('#a(b)')), [['<a href="/hashtags/a(b)" class="hashtag">#a(b)</a>'], '']);
-      assert.deepStrictEqual(inspect(parser('#a((b))')), [['<a href="/hashtags/a" class="hashtag">#a</a>'], '((b))']);
-      assert.deepStrictEqual(inspect(parser(`#a'`)), [[`<a href="/hashtags/a'" class="hashtag">#a'</a>`], '']);
-      assert.deepStrictEqual(inspect(parser(`#a(b')`)), [[`<a href="/hashtags/a(b')" class="hashtag">#a(b')</a>`], '']);
-      assert.deepStrictEqual(inspect(parser(`#a(b'')`)), [[`<a href="/hashtags/a" class="hashtag">#a</a>`], `(b'')`]);
-      assert.deepStrictEqual(inspect(parser(`#a('b)`)), [['<a href="/hashtags/a" class="hashtag">#a</a>'], `('b)`]);
+      assert.deepStrictEqual(inspect(parser('#a(b)')), [['<a href="/hashtags/a" class="hashtag">#a</a>'], '(b)']);
+      assert.deepStrictEqual(inspect(parser('#a_')), [['<a href="/hashtags/a" class="hashtag">#a</a>'], '_']);
       assert.deepStrictEqual(inspect(parser('#a_b')), [['<a href="/hashtags/a_b" class="hashtag">#a_b</a>'], '']);
-      assert.deepStrictEqual(inspect(parser(`#a'_b`)), [[`<a href="/hashtags/a'_b" class="hashtag">#a'_b</a>`], '']);
-      assert.deepStrictEqual(inspect(parser('#a(b_c)')), [['<a href="/hashtags/a(b_c)" class="hashtag">#a(b_c)</a>'], '']);
-      assert.deepStrictEqual(inspect(parser('#a(b__c)')), [['<a href="/hashtags/a" class="hashtag">#a</a>'], '(b__c)']);
+      assert.deepStrictEqual(inspect(parser(`#a_'b`)), [['<a href="/hashtags/a" class="hashtag">#a</a>'], `_'b`]);
+      assert.deepStrictEqual(inspect(parser('#a__b')), [['<a href="/hashtags/a" class="hashtag">#a</a>'], '__b']);
       assert.deepStrictEqual(inspect(parser('#あ')), [['<a href="/hashtags/あ" class="hashtag">#あ</a>'], '']);
+      assert.deepStrictEqual(inspect(parser('#👩')), [['<a href="/hashtags/👩" class="hashtag">#👩</a>'], '']);
       assert.deepStrictEqual(inspect(parser('#1a')), [['<a href="/hashtags/1a" class="hashtag">#1a</a>'], '']);
       assert.deepStrictEqual(inspect(parser('#1あ')), [['<a href="/hashtags/1あ" class="hashtag">#1あ</a>'], '']);
+      assert.deepStrictEqual(inspect(parser('#1👩')), [['<a href="/hashtags/1👩" class="hashtag">#1👩</a>'], '']);
       assert.deepStrictEqual(inspect(parser('#domain/a')), [['<a href="https://domain/hashtags/a" target="_blank" class="hashtag">#domain/a</a>'], '']);
       assert.deepStrictEqual(inspect(parser('#domain.co.jp/a')), [['<a href="https://domain.co.jp/hashtags/a" target="_blank" class="hashtag">#domain.co.jp/a</a>'], '']);
+      // Reserved
+      assert.deepStrictEqual(inspect(parser(`#a'`)), [[`#a'`], '']);
+      assert.deepStrictEqual(inspect(parser(`#a'b`)), [[`#a'b`], '']);
+      assert.deepStrictEqual(inspect(parser(`#a'_b`)), [[`#a'_b`], '']);
     });
 
   });

package/src/parser/inline/autolink/hashtag.ts

@@ -6,15 +6,22 @@ import { define } from 'typed-dom';
 
 // https://example/hashtags/a must be a hashtag page or a redirect page going there.
 
+// https://github.com/tc39/proposal-regexp-unicode-property-escapes#matching-emoji
+export const emoji = String.raw`\p{Emoji_Modifier_Base}\p{Emoji_Modifier}?|\p{Emoji_Presentation}|\p{Emoji}\uFE0F`;
+
 export const hashtag: AutolinkParser.HashtagParser = lazy(() => fmap(rewrite(
   open(
     '#',
     tails([
       verify(
-        str(/^[0-9A-Za-z](?:[0-9A-Za-z-]{0,61}[0-9A-Za-z])?(?:\.[0-9A-Za-z](?:[0-9A-Za-z-]{0,61}[0-9A-Za-z])?)*\//),
+        str(/^[0-9A-Za-z](?:(?:[0-9A-Za-z]|-(?=\w)){0,61}[0-9A-Za-z])?(?:\.[0-9A-Za-z](?:(?:[0-9A-Za-z]|-(?=\w)){0,61}[0-9A-Za-z])?)*\//),
         ([source]) => source.length <= 253 + 1),
       verify(
-        str(/^(?=(?:[0-9]{1,127}_?)?(?:[A-Za-z]|[^\x00-\x7F\s]))(?:[0-9A-Za-z]|[^\x00-\x7F\s]|'(?!')|_(?=[0-9A-Za-z]|[^\x00-\x7F\s])){1,128}(?:_?\((?=(?:[0-9]{1,127}_?)?(?:[A-Za-z]|[^\x00-\x7F\s]))(?:[0-9A-Za-z]|[^\x00-\x7F\s]|'(?!')|_(?=[0-9A-Za-z]|[^\x00-\x7F\s])){1,125}\))?(?![0-9A-Za-z'_]|[^\x00-\x7F\s])/),
+        str(new RegExp(['^',
+          String.raw`(?=[0-9]{0,127}_?(?:[^\d\p{C}\p{S}\p{P}\s]|${emoji}))`,
+          String.raw`(?:[^\p{C}\p{S}\p{P}\s]|${emoji}|_(?=[^\p{C}\p{S}\p{P}\s]|${emoji})){1,128}`,
+          String.raw`(?!_?(?:[^\p{C}\p{S}\p{P}\s]|${emoji})|')`,
+        ].join(''), 'u')),
         ([source]) => source.length <= 128),
     ])),
   context({ syntax: { inline: {

package/src/parser/inline/autolink.ts

@@ -4,33 +4,30 @@ import { url } from './autolink/url';
 import { email } from './autolink/email';
 import { channel } from './autolink/channel';
 import { account } from './autolink/account';
-import { hashtag } from './autolink/hashtag';
+import { hashtag, emoji } from './autolink/hashtag';
 import { hashnum } from './autolink/hashnum';
 import { anchor } from './autolink/anchor';
 import { str } from '../source';
 import { stringify } from '../util';
 
 export const autolink: AutolinkParser = fmap(
-  validate(/^(?:[@#>0-9A-Za-z]|[^\x00-\x7F\s])/,
+  validate(/^(?:[@#>0-9A-Za-z]|\S#)/,
   guard(context => context.syntax?.inline?.autolink ?? true,
   some(union([
     url,
     email,
     // Escape unmatched email-like strings.
     str(/^[0-9A-Za-z]+(?:[.+_-][0-9A-Za-z]+)*(?:@(?:[0-9A-Za-z]+(?:[.-][0-9A-Za-z]+)*)?)+/),
-    // Escape repeated symbols.
-    str(/^@+(?![0-9A-Za-z]|[^\x00-\x7F\s])/),
-    str(/^#+(?![0-9A-Za-z'_]|[^\x00-\x7F\s])/),
     channel,
     account,
     // Escape unmatched account-like strings.
-    str(/^@[0-9A-Za-z]+(?:-[0-9A-Za-z]+)*/),
+    str(/^@+[0-9A-Za-z]*(?:-[0-9A-Za-z]+)*/),
     // Escape invalid leading characters.
-    str(/^[0-9A-Za-z]+(?=#)|^[^\x00-\x7F\s]+(?=#)/),
+    str(new RegExp(String.raw`^(?:[^\p{C}\p{S}\p{P}\s]|${emoji}|['_])(?=#)`, 'u')),
     hashtag,
     hashnum,
     // Escape unmatched hashtag-like strings.
-    str(/^#(?:[0-9A-Za-z'_]|[^\x00-\x7F\s])+/),
+    str(new RegExp(String.raw`^#+(?:[^\p{C}\p{S}\p{P}\s]|${emoji}|['_])*`, 'u')),
     anchor,
   ])))),
   ns => ns.length === 1 ? ns : [stringify(ns)]);

package/src/parser/inline/link.ts

@@ -83,7 +83,7 @@ export function resolve(uri: string, host: URL | Location, source: URL | Locatio
     case uri.slice(0, 2) === '^/':
       const last = host.pathname.slice(host.pathname.lastIndexOf('/') + 1);
       return last.includes('.') // isFile
-          && /^[0-9]*[a-z][0-9a-z]*$/i.test(last.slice(last.lastIndexOf('.') + 1))
+          && /^[0-9]*[A-Za-z][0-9A-Za-z]*$/.test(last.slice(last.lastIndexOf('.') + 1))
         ? `${host.pathname.slice(0, -last.length)}${uri.slice(2)}`
         : `${host.pathname.replace(/\/?$/, '/')}${uri.slice(2)}`;
     case host.origin === source.origin

package/src/parser/inline.test.ts

@@ -168,11 +168,12 @@ describe('Unit: parser/inline', () => {
       assert.deepStrictEqual(inspect(parser('#a\nb\n#c\n[#d]')), [['<a href="/hashtags/a" class="hashtag">#a</a>', '<br>', 'b', '<br>', '<a href="/hashtags/c" class="hashtag">#c</a>', '<br>', '<a class="index" href="#index:d">d</a>'], '']);
       assert.deepStrictEqual(inspect(parser('##a')), [['##a'], '']);
       assert.deepStrictEqual(inspect(parser('a#b')), [['a#b'], '']);
-      assert.deepStrictEqual(inspect(parser('0a#b')), [['0a#b'], '']);
+      assert.deepStrictEqual(inspect(parser('0a#b')), [['0', 'a#b'], '']);
       assert.deepStrictEqual(inspect(parser('あ#b')), [['あ#b'], '']);
-      assert.deepStrictEqual(inspect(parser('あい#b')), [['あい#b'], '']);
+      assert.deepStrictEqual(inspect(parser('あい#b')), [['あ', 'い#b'], '']);
       assert.deepStrictEqual(inspect(parser('0aあ#b')), [['0a', 'あ#b'], '']);
-      assert.deepStrictEqual(inspect(parser('0aあい#b')), [['0a', 'あい#b'], '']);
+      assert.deepStrictEqual(inspect(parser('0aあい#b')), [['0a', 'あ', 'い#b'], '']);
+      assert.deepStrictEqual(inspect(parser('「#あ」')), [['「', '<a href="/hashtags/あ" class="hashtag">#あ</a>', '」'], '']);
       assert.deepStrictEqual(inspect(parser('a\n#b')), [['a', '<br>', '<a href="/hashtags/b" class="hashtag">#b</a>'], '']);
       assert.deepStrictEqual(inspect(parser('a\\\n#b')), [['a', '<span class="linebreak"> </span>', '<a href="/hashtags/b" class="hashtag">#b</a>'], '']);
       assert.deepStrictEqual(inspect(parser('*a*#b')), [['<em>a</em>', '<a href="/hashtags/b" class="hashtag">#b</a>'], '']);
@@ -186,6 +187,7 @@ describe('Unit: parser/inline', () => {
     it('hashnum', () => {
       assert.deepStrictEqual(inspect(parser('#1')), [['<a class="hashnum">#1</a>'], '']);
       assert.deepStrictEqual(inspect(parser('#12345678901234567@a')), [['#12345678901234567@a'], '']);
+      assert.deepStrictEqual(inspect(parser('「#1」')), [['「', '<a class="hashnum">#1</a>', '」'], '']);
     });
 
   });

package/src/parser/source/text.ts

@@ -4,9 +4,9 @@ import { union, focus, creator } from '../../combinator';
 import { str } from './str';
 import { html } from 'typed-dom';
 
-export const separator = /[\s\x00-\x7F]|[、。！？][^\S\n]*(?=\\\n)/;
+export const separator = /[\s\x00-\x7F]|\S#|[、。！？][^\S\n]*(?=\\\n)/;
 export const nonWhitespace = /[\S\n]|$/;
-export const nonAlphanumeric = /[^0-9A-Za-z]|$/;
+export const nonAlphanumeric = /[^0-9A-Za-z]|\S#|$/;
 const repeat = str(/^(.)\1*/);
 
 export const text: TextParser = creator((source, context) => {
@@ -59,7 +59,7 @@ export const text: TextParser = creator((source, context) => {
           assert(source[0] !== '\n');
           const b = source[0].trimStart() === '';
           const i = b || isAlphanumeric(source[0])
-            ? source.search(b ? nonWhitespace : nonAlphanumeric)
+            ? source.search(b ? nonWhitespace : nonAlphanumeric) || 1
             : 1;
           assert(i > 0);
           assert(!['\\', '\n'].includes(source[0]));

package/src/parser/source/unescapable.ts

@@ -12,7 +12,7 @@ export const unescsource: UnescapableSourceParser = creator(source => {
     case 0: {
       const b = source[0] !== '\n' && source[0].trimStart() === '';
       const i = b || isAlphanumeric(source[0])
-        ? source.search(b ? nonWhitespace : nonAlphanumeric)
+        ? source.search(b ? nonWhitespace : nonAlphanumeric) || 1
         : 1;
       assert(i > 0);
       return [[source.slice(0, i)], source.slice(i)];