@recaptime-dev's working patches + fork for Phorge, a community fork of Phabricator. (Upstream dev and stable branches are at upstream/main and upstream/stable respectively.)
hq.recaptime.dev/wiki/Phorge
phorge
phabricator
Generate more friendly anchor names for header sections in Remarkup
Summary:
Depends on D20820. Ref T13410. We currently cut anchor names in the middle, don't support emoji in anchors, and generate relatively short anchors.
Generate slightly longer anchors, allow more unicode, and try not to cut things in the middle.
Test Plan: Created a document with a variety of different anchors and saw them generate more usable names.
Maniphest Tasks: T13410
Differential Revision: https://secure.phabricator.com/D20821
+100
-24
+2
src/__phutil_library_map__.php
+2
src/__phutil_library_map__.php
···
2150
2150
'PhabricatorAlmanacApplication' => 'applications/almanac/application/PhabricatorAlmanacApplication.php',
2151
2151
'PhabricatorAmazonAuthProvider' => 'applications/auth/provider/PhabricatorAmazonAuthProvider.php',
2152
2152
'PhabricatorAmazonSNSFuture' => 'applications/metamta/future/PhabricatorAmazonSNSFuture.php',
2153
+
'PhabricatorAnchorTestCase' => 'infrastructure/markup/__tests__/PhabricatorAnchorTestCase.php',
2153
2154
'PhabricatorAnchorView' => 'view/layout/PhabricatorAnchorView.php',
2154
2155
'PhabricatorAphlictManagementDebugWorkflow' => 'applications/aphlict/management/PhabricatorAphlictManagementDebugWorkflow.php',
2155
2156
'PhabricatorAphlictManagementNotifyWorkflow' => 'applications/aphlict/management/PhabricatorAphlictManagementNotifyWorkflow.php',
···
8314
8315
'PhabricatorAlmanacApplication' => 'PhabricatorApplication',
8315
8316
'PhabricatorAmazonAuthProvider' => 'PhabricatorOAuth2AuthProvider',
8316
8317
'PhabricatorAmazonSNSFuture' => 'PhutilAWSFuture',
8318
+
'PhabricatorAnchorTestCase' => 'PhabricatorTestCase',
8317
8319
'PhabricatorAnchorView' => 'AphrontView',
8318
8320
'PhabricatorAphlictManagementDebugWorkflow' => 'PhabricatorAphlictManagementWorkflow',
8319
8321
'PhabricatorAphlictManagementNotifyWorkflow' => 'PhabricatorAphlictManagementWorkflow',
+38
src/infrastructure/markup/__tests__/PhabricatorAnchorTestCase.php
+38
src/infrastructure/markup/__tests__/PhabricatorAnchorTestCase.php
···
1
+
<?php
2
+
3
+
final class PhabricatorAnchorTestCase
4
+
extends PhabricatorTestCase {
5
+
6
+
public function testAnchors() {
7
+
8
+
$low_ascii = '';
9
+
for ($ii = 19; $ii <= 127; $ii++) {
10
+
$low_ascii .= chr($ii);
11
+
}
12
+
13
+
$snowman = "\xE2\x9B\x84";
14
+
15
+
$map = array(
16
+
'' => '',
17
+
'Bells and Whistles' => 'bells-and-whistles',
18
+
'Termination for Nonpayment' => 'termination-for-nonpayment',
19
+
$low_ascii => '0123456789-abcdefghijklmnopqrstu',
20
+
'xxxx xxxx xxxx xxxx xxxx on' => 'xxxx-xxxx-xxxx-xxxx-xxxx',
21
+
'xxxx xxxx xxxx xxxx xxxx ox' => 'xxxx-xxxx-xxxx-xxxx-xxxx-ox',
22
+
"So, You Want To Build A {$snowman}?" =>
23
+
"so-you-want-to-build-a-{$snowman}",
24
+
str_repeat($snowman, 128) => str_repeat($snowman, 32),
25
+
);
26
+
27
+
foreach ($map as $input => $expect) {
28
+
$anchor = PhutilRemarkupHeaderBlockRule::getAnchorNameFromHeaderText(
29
+
$input);
30
+
31
+
$this->assertEqual(
32
+
$expect,
33
+
$anchor,
34
+
pht('Anchor for "%s".', $input));
35
+
}
36
+
}
37
+
38
+
}
+49
-21
src/infrastructure/markup/blockrule/PhutilRemarkupHeaderBlockRule.php
+49
-21
src/infrastructure/markup/blockrule/PhutilRemarkupHeaderBlockRule.php
···
73
73
}
74
74
75
75
private function generateAnchor($level, $text) {
76
-
$anchor = strtolower($text);
77
-
$anchor = preg_replace('/[^a-z0-9]/', '-', $anchor);
78
-
$anchor = preg_replace('/--+/', '-', $anchor);
79
-
$anchor = trim($anchor, '-');
80
-
$anchor = substr($anchor, 0, 24);
81
-
$anchor = trim($anchor, '-');
82
-
$base = $anchor;
83
-
84
-
$key = self::KEY_HEADER_TOC;
85
76
$engine = $this->getEngine();
86
-
$anchors = $engine->getTextMetadata($key, array());
87
-
88
-
$suffix = 1;
89
-
while (!strlen($anchor) || isset($anchors[$anchor])) {
90
-
$anchor = $base.'-'.$suffix;
91
-
$anchor = trim($anchor, '-');
92
-
$suffix++;
93
-
}
94
77
95
78
// When a document contains a link inside a header, like this:
96
79
//
···
100
83
// header itself. We push the 'toc' state so all the link rules generate
101
84
// just names.
102
85
$engine->pushState('toc');
103
-
$text = $this->applyRules($text);
104
-
$text = $engine->restoreText($text);
86
+
$plain_text = $text;
87
+
$plain_text = $this->applyRules($plain_text);
88
+
$plain_text = $engine->restoreText($plain_text);
89
+
$engine->popState('toc');
90
+
91
+
$anchor = self::getAnchorNameFromHeaderText($plain_text);
92
+
93
+
if (!strlen($anchor)) {
94
+
return null;
95
+
}
105
96
106
-
$anchors[$anchor] = array($level, $text);
107
-
$engine->popState('toc');
97
+
$base = $anchor;
98
+
99
+
$key = self::KEY_HEADER_TOC;
100
+
$anchors = $engine->getTextMetadata($key, array());
108
101
102
+
$suffix = 1;
103
+
while (isset($anchors[$anchor])) {
104
+
$anchor = $base.'-'.$suffix;
105
+
$anchor = trim($anchor, '-');
106
+
$suffix++;
107
+
}
108
+
109
+
$anchors[$anchor] = array($level, $plain_text);
109
110
$engine->setTextMetadata($key, $anchors);
110
111
111
112
return phutil_tag(
···
157
158
}
158
159
159
160
return phutil_implode_html("\n", $toc);
161
+
}
162
+
163
+
public static function getAnchorNameFromHeaderText($text) {
164
+
$anchor = phutil_utf8_strtolower($text);
165
+
166
+
// Replace all latin characters which are not "a-z" or "0-9" with "-".
167
+
// Preserve other characters, since non-latin letters and emoji work
168
+
// fine in anchors.
169
+
$anchor = preg_replace('/[\x00-\x2F\x3A-\x60\x7B-\x7F]+/', '-', $anchor);
170
+
$anchor = trim($anchor, '-');
171
+
172
+
// Truncate the fragment to something reasonable.
173
+
$anchor = id(new PhutilUTF8StringTruncator())
174
+
->setMaximumGlyphs(32)
175
+
->setTerminator('')
176
+
->truncateString($anchor);
177
+
178
+
// If the fragment is terminated by a word which "The U.S. Government
179
+
// Printing Office Style Manual" normally discourages capitalizing in
180
+
// titles, discard it. This is an arbitrary heuristic intended to avoid
181
+
// awkward hanging words in anchors.
182
+
$anchor = preg_replace(
183
+
'/-(a|an|the|at|by|for|in|of|on|per|to|up|and|as|but|if|or|nor)\z/',
184
+
'',
185
+
$anchor);
186
+
187
+
return $anchor;
160
188
}
161
189
162
190
}
+4
src/infrastructure/markup/markuprule/PhutilRemarkupBoldRule.php
+4
src/infrastructure/markup/markuprule/PhutilRemarkupBoldRule.php
+4
src/infrastructure/markup/remarkup/PhutilRemarkupEngine.php
+4
src/infrastructure/markup/remarkup/PhutilRemarkupEngine.php
+3
-3
src/infrastructure/markup/remarkup/__tests__/remarkup/toc.txt
+3
-3
src/infrastructure/markup/remarkup/__tests__/remarkup/toc.txt
···
6
6
7
7
~~~~~~~~~~
8
8
<ul>
9
-
<li><a href="#http-www-example-com-lin">link_name</a></li>
9
+
<li><a href="#link-name">link_name</a></li>
10
10
<ul>
11
-
<li><a href="#bold"><strong>bold</strong></a></li>
11
+
<li><a href="#bold">bold</a></li>
12
12
</ul>
13
13
<li><a href="#http-www-example-com">http://www.example.com</a></li>
14
14
</ul>
15
15
16
-
<h2 class="remarkup-header"><a name="http-www-example-com-lin"></a><a href="http://www.example.com/" class="remarkup-link" target="_blank" rel="noreferrer">link_name</a></h2>
16
+
<h2 class="remarkup-header"><a name="link-name"></a><a href="http://www.example.com/" class="remarkup-link" target="_blank" rel="noreferrer">link_name</a></h2>
17
17
18
18
<h3 class="remarkup-header"><a name="bold"></a><strong>bold</strong></h3>
19
19