|
1
|
|
|
<?php |
|
2
|
|
|
/* |
|
3
|
|
|
* This file is part of the trefoil application. |
|
4
|
|
|
* |
|
5
|
|
|
* (c) Miguel Angel Gabriel <[email protected]> |
|
6
|
|
|
* |
|
7
|
|
|
* For the full copyright and license information, please view the LICENSE |
|
8
|
|
|
* file that was distributed with this source code. |
|
9
|
|
|
*/ |
|
10
|
|
|
|
|
11
|
|
|
namespace Trefoil\Helpers; |
|
12
|
|
|
|
|
13
|
|
|
use Easybook\Parsers\ParserInterface; |
|
14
|
|
|
|
|
15
|
|
|
/** |
|
16
|
|
|
* This class: |
|
17
|
|
|
* - For headless tables, transforms the <td> cells in first column |
|
18
|
|
|
* within <strong> tags into <th> cells (making a vertical head). |
|
19
|
|
|
* - Allows multiline cells. |
|
20
|
|
|
* |
|
21
|
|
|
* Multiline cells: |
|
22
|
|
|
* --------------- |
|
23
|
|
|
* |
|
24
|
|
|
* If a line inside a cell ends with '+' char (plus sign) it will be joined |
|
25
|
|
|
* with the next line. |
|
26
|
|
|
* |
|
27
|
|
|
* Automatic head cells for headless tables: |
|
28
|
|
|
* ---------------------------------------- |
|
29
|
|
|
* |
|
30
|
|
|
* For a headless table (i.e. without headings in first row), cells in first |
|
31
|
|
|
* column that are all bold (i.e. surrounded by "**") will be rendered |
|
32
|
|
|
* as "<th>" tags instead of normal "<td>" tags, allowing formatting. |
|
33
|
|
|
* |
|
34
|
|
|
*/ |
|
35
|
|
|
class TableExtra |
|
36
|
|
|
{ |
|
37
|
|
|
/** @var ParserInterface */ |
|
38
|
|
|
protected $markdownParser; |
|
39
|
|
|
|
|
40
|
|
|
/** |
|
41
|
|
|
* @param ParserInterface $markdownParser |
|
42
|
|
|
*/ |
|
43
|
1 |
|
public function setMarkdownParser($markdownParser) |
|
44
|
|
|
{ |
|
45
|
1 |
|
$this->markdownParser = $markdownParser; |
|
46
|
1 |
|
} |
|
47
|
|
|
|
|
48
|
|
|
/** |
|
49
|
|
|
* Processes all tables in the html string |
|
50
|
|
|
* |
|
51
|
|
|
* @param string $htmlString |
|
52
|
|
|
* |
|
53
|
|
|
* @return string |
|
54
|
|
|
*/ |
|
55
|
3 |
|
public function processAllTables($htmlString) |
|
56
|
|
|
{ |
|
57
|
3 |
|
$regExp = '/'; |
|
58
|
3 |
|
$regExp .= '(?<table><table.*<\/table>)'; |
|
59
|
3 |
|
$regExp .= '/Ums'; // Ungreedy, multiline, dotall |
|
60
|
|
|
|
|
61
|
|
|
// PHP 5.3 compat |
|
62
|
3 |
|
$me = $this; |
|
63
|
|
|
|
|
64
|
3 |
|
$callback = function ($matches) use ($me) { |
|
65
|
3 |
|
$table = new Table(); |
|
66
|
3 |
|
$table->fromHtml($matches['table']); |
|
67
|
|
|
|
|
68
|
3 |
|
if ($table->isEmpty()) { |
|
69
|
|
|
return $matches[0]; |
|
70
|
|
|
} |
|
71
|
|
|
|
|
72
|
3 |
|
$table = $me->internalProcessExtraTable($table); |
|
73
|
3 |
|
$html = $table->toHtml(); |
|
74
|
|
|
|
|
75
|
3 |
|
return $html; |
|
76
|
3 |
|
}; |
|
77
|
|
|
|
|
78
|
3 |
|
$output = preg_replace_callback($regExp, $callback, $htmlString); |
|
79
|
|
|
|
|
80
|
3 |
|
return $output; |
|
81
|
|
|
} |
|
82
|
|
|
|
|
83
|
|
|
/** |
|
84
|
|
|
* @param Table $table |
|
85
|
|
|
* |
|
86
|
|
|
* @return Table |
|
87
|
|
|
* @internal Should be protected but made public for PHP 5.3 compat |
|
88
|
|
|
*/ |
|
89
|
3 |
|
public function internalProcessExtraTable(Table $table) |
|
90
|
|
|
{ |
|
91
|
|
|
// process and adjusts table definition |
|
92
|
|
|
|
|
93
|
3 |
|
$headless = true; |
|
94
|
3 |
|
foreach ($table['thead'] as $row) { |
|
95
|
2 |
|
foreach ($row as $cell) { |
|
96
|
2 |
|
if (empty($cell['contents'])) { |
|
97
|
2 |
|
$headless = true; |
|
98
|
2 |
|
break 2; |
|
99
|
|
|
} |
|
100
|
2 |
|
} |
|
101
|
3 |
|
} |
|
102
|
|
|
|
|
103
|
|
|
// headless table |
|
104
|
3 |
|
if ($headless && $table['tbody']) { |
|
105
|
3 |
|
$table['tbody'] = $this->processFirstColumnCells($table['tbody']); |
|
106
|
3 |
|
} |
|
107
|
|
|
|
|
108
|
|
|
// table with head and body |
|
109
|
3 |
|
if ($table['thead'] || $table['tbody']) { |
|
110
|
|
|
|
|
111
|
3 |
|
$table['thead'] = $this->processMultilineCells($table['thead']); |
|
112
|
3 |
|
$table['tbody'] = $this->processMultilineCells($table['tbody']); |
|
113
|
|
|
|
|
114
|
3 |
|
return $table; |
|
115
|
|
|
} |
|
116
|
|
|
|
|
117
|
|
|
// table without head or body |
|
118
|
|
|
$table['table'] = $this->processMultilineCells($table['table']); |
|
119
|
|
|
|
|
120
|
|
|
return $table; |
|
121
|
|
|
} |
|
122
|
|
|
|
|
123
|
|
|
/** |
|
124
|
|
|
* Join the cells that belong to multiline cells. |
|
125
|
|
|
* |
|
126
|
|
|
* @param $rows |
|
127
|
|
|
* |
|
128
|
|
|
* @return array Processed table rows |
|
129
|
|
|
*/ |
|
130
|
3 |
|
protected function processMultilineCells(array $rows) |
|
131
|
|
|
{ |
|
132
|
3 |
|
$newRows = $rows; |
|
133
|
3 |
|
foreach ($newRows as $rowIndex => $row) { |
|
134
|
|
|
|
|
135
|
3 |
|
foreach ($row as $colIndex => $col) { |
|
136
|
3 |
|
$cell = $newRows[$rowIndex][$colIndex]; |
|
137
|
3 |
|
$cellText = rtrim($cell['contents']); |
|
138
|
|
|
|
|
139
|
3 |
|
if (substr($cellText, -1, 1) === '+') { |
|
140
|
|
|
// continued cell |
|
141
|
1 |
|
$newCell = array(); |
|
142
|
1 |
|
$newCell[] = substr($cellText, 0, -1); |
|
143
|
|
|
|
|
144
|
|
|
// find all the continuation cells (same col) |
|
145
|
1 |
|
for ($nextRowIndex = $rowIndex + 1; $nextRowIndex < count($newRows); $nextRowIndex++) { |
|
|
|
|
|
|
146
|
|
|
|
|
147
|
1 |
|
$nextCell = $newRows[$nextRowIndex][$colIndex]; |
|
148
|
1 |
|
$cellText = rtrim($nextCell['contents']); |
|
149
|
|
|
|
|
150
|
1 |
|
$newRows[$nextRowIndex][$colIndex]['contents'] = ''; |
|
151
|
|
|
|
|
152
|
|
|
// continued cell? |
|
153
|
1 |
|
$continued = (substr($cellText, -1, 1) === '+'); |
|
154
|
|
|
|
|
155
|
1 |
|
if ($continued) { |
|
156
|
|
|
// clean the ending (+) |
|
157
|
1 |
|
$cellText = substr($cellText, 0, -1); |
|
158
|
1 |
|
} |
|
159
|
|
|
|
|
160
|
|
|
// save cleaned text |
|
161
|
1 |
|
$newCell[] = $cellText; |
|
162
|
|
|
|
|
163
|
1 |
|
if (!$continued) { |
|
164
|
|
|
// no more continuations |
|
165
|
1 |
|
break; |
|
166
|
|
|
} |
|
167
|
1 |
|
} |
|
168
|
|
|
|
|
169
|
1 |
|
if ($this->markdownParser) { |
|
170
|
1 |
|
$parsedCell = $this->markdownParser->transform(join("\n\n", $newCell)); |
|
171
|
1 |
|
} else { |
|
172
|
|
|
// safe default |
|
173
|
|
|
$parsedCell = join("<br/>", $newCell); |
|
174
|
|
|
} |
|
175
|
|
|
|
|
176
|
1 |
|
$newRows[$rowIndex][$colIndex]['contents'] = $parsedCell; |
|
177
|
1 |
|
} |
|
178
|
3 |
|
} |
|
179
|
3 |
|
} |
|
180
|
|
|
|
|
181
|
|
|
// remove empty rows left by the process |
|
182
|
3 |
|
$newRows2 = array(); |
|
183
|
3 |
|
foreach ($newRows as $rowIndex => $row) { |
|
184
|
|
|
|
|
185
|
3 |
|
$emptyRow = true; |
|
186
|
3 |
|
foreach ($row as $colIndex => $col) { |
|
187
|
3 |
|
$cellText = trim($col['contents']); |
|
188
|
3 |
|
if (!empty($cellText)) { |
|
189
|
3 |
|
$emptyRow = false; |
|
190
|
3 |
|
} |
|
191
|
3 |
|
} |
|
192
|
|
|
|
|
193
|
3 |
|
if (!$emptyRow) { |
|
194
|
3 |
|
$newRows2[] = $row; |
|
195
|
3 |
|
} |
|
196
|
3 |
|
} |
|
197
|
|
|
|
|
198
|
3 |
|
return $newRows2; |
|
199
|
|
|
} |
|
200
|
|
|
|
|
201
|
|
|
/** |
|
202
|
|
|
* Converts <td> cells into <th> for first column cells that are fully bold. |
|
203
|
|
|
* |
|
204
|
|
|
* @param array $rows |
|
205
|
|
|
* |
|
206
|
|
|
* @return array Processed rows |
|
207
|
|
|
*/ |
|
208
|
3 |
|
protected function processFirstColumnCells(array $rows) |
|
209
|
|
|
{ |
|
210
|
3 |
|
$newRows = $rows; |
|
211
|
3 |
|
foreach ($newRows as $rowIndex => $row) { |
|
212
|
|
|
|
|
213
|
|
|
// examine first cell ir row |
|
214
|
3 |
|
$cell = $newRows[$rowIndex][0]; |
|
215
|
3 |
|
$cellText = rtrim($cell['contents']); |
|
216
|
|
|
|
|
217
|
3 |
|
$regExp = '/^<strong>.*<\/strong>$/Us'; |
|
218
|
3 |
|
if (preg_match($regExp, $cellText, $matches)) { |
|
219
|
1 |
|
$cell['tag'] = 'th'; |
|
220
|
|
|
|
|
221
|
|
|
// change cell to <th> |
|
222
|
1 |
|
$newRows[$rowIndex][0]['tag'] = 'th'; |
|
223
|
1 |
|
} |
|
224
|
3 |
|
} |
|
225
|
|
|
|
|
226
|
3 |
|
return $newRows; |
|
227
|
|
|
} |
|
228
|
|
|
|
|
229
|
|
|
} |
|
230
|
|
|
|
If the size of the collection does not change during the iteration, it is generally a good practice to compute it beforehand, and not on each iteration: