| 1 |
lars |
1 |
<?php
|
|
|
2 |
/* vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: */
|
|
|
3 |
|
|
|
4 |
/**
|
|
|
5 |
* File::CSV
|
|
|
6 |
*
|
|
|
7 |
* PHP versions 4 and 5
|
|
|
8 |
*
|
|
|
9 |
* Copyright (c) 2002-2008,
|
|
|
10 |
* Tomas V.V.Cox <cox@idecnet.com>,
|
|
|
11 |
* Helgi Þormar Þorbjörnsson <helgi@php.net>
|
|
|
12 |
* All rights reserved.
|
|
|
13 |
*
|
|
|
14 |
* Redistribution and use in source and binary forms, with or without
|
|
|
15 |
* modification, are permitted provided that the following conditions are met:
|
|
|
16 |
*
|
|
|
17 |
* * Redistributions of source code must retain the above copyright notice,
|
|
|
18 |
* this list of conditions and the following disclaimer.
|
|
|
19 |
* * Redistributions in binary form must reproduce the above copyright
|
|
|
20 |
* notice, this list of conditions and the following disclaimer in the
|
|
|
21 |
* documentation and/or other materials provided with the distribution.
|
|
|
22 |
*
|
|
|
23 |
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
|
|
24 |
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
25 |
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
|
|
26 |
* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE
|
|
|
27 |
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
28 |
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
|
|
|
29 |
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
|
|
|
30 |
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
|
|
31 |
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
|
|
32 |
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
|
33 |
*
|
|
|
34 |
* @category File
|
|
|
35 |
* @package File
|
|
|
36 |
* @author Tomas V.V.Cox <cox@idecnet.com>
|
|
|
37 |
* @author Helgi Þormar Þorbjörnsson <helgi@php.net>
|
|
|
38 |
* @copyright 2004-2011 The Authors
|
|
|
39 |
* @license http://www.opensource.org/licenses/bsd-license.php New BSD License
|
|
|
40 |
* @version CVS: $Id: CSV.php 309245 2011-03-15 02:02:41Z dufuz $
|
|
|
41 |
* @link http://pear.php.net/package/File
|
|
|
42 |
*/
|
|
|
43 |
|
|
|
44 |
require_once 'PEAR.php';
|
|
|
45 |
require_once 'File.php';
|
|
|
46 |
|
|
|
47 |
/**
|
|
|
48 |
* File class for handling CSV files (Comma Separated Values), a common format
|
|
|
49 |
* for exchanging data.
|
|
|
50 |
*
|
|
|
51 |
* TODO:
|
|
|
52 |
* - Usage example and Doc
|
|
|
53 |
* - Use getPointer() in discoverFormat
|
|
|
54 |
* - Add a line counter for being able to output better error reports
|
|
|
55 |
* - Store the last error in GLOBALS and add File_CSV::getLastError()
|
|
|
56 |
*
|
|
|
57 |
* Wish:
|
|
|
58 |
* - Other methods like readAll(), writeAll(), numFields(), numRows()
|
|
|
59 |
* - Try to detect if a CSV has header or not in discoverFormat() (not possible with CSV)
|
|
|
60 |
*
|
|
|
61 |
* Known Bugs:
|
|
|
62 |
* (they has been analyzed but for the moment the impact in the speed for
|
|
|
63 |
* properly handle this uncommon cases is too high and won't be supported)
|
|
|
64 |
* - A field which is composed only by a single quoted separator (ie -> ;";";)
|
|
|
65 |
* is not handled properly
|
|
|
66 |
* - When there is exactly one field minus than the expected number and there
|
|
|
67 |
* is a field with a separator inside, the parser will throw the "wrong count" error
|
|
|
68 |
*
|
|
|
69 |
* Info about CSV and links to other sources
|
|
|
70 |
* http://rfc.net/rfc4180.html
|
|
|
71 |
*
|
|
|
72 |
* @author Tomas V.V.Cox <cox@idecnet.com>
|
|
|
73 |
* @author Helgi Þormar Þorbjörnsson <helgi@php.net>
|
|
|
74 |
* @package File
|
|
|
75 |
* @license http://www.opensource.org/licenses/bsd-license.php New BSD License
|
|
|
76 |
*/
|
|
|
77 |
class File_CSV
|
|
|
78 |
{
|
|
|
79 |
/**
|
|
|
80 |
* This raiseError method works in a different way. It will always return
|
|
|
81 |
* false (an error occurred) but it will call PEAR::raiseError() before
|
|
|
82 |
* it. If no default PEAR global handler is set, will trigger an error.
|
|
|
83 |
*
|
|
|
84 |
* @param string $error The error message
|
|
|
85 |
* @return bool always false
|
|
|
86 |
*/
|
|
|
87 |
function raiseError($error)
|
|
|
88 |
{
|
|
|
89 |
// If a default PEAR Error handler is not set trigger the error
|
|
|
90 |
// XXX Add a PEAR::isSetHandler() method?
|
|
|
91 |
if ($GLOBALS['_PEAR_default_error_mode'] == PEAR_ERROR_RETURN) {
|
|
|
92 |
PEAR::raiseError($error, null, PEAR_ERROR_TRIGGER, E_USER_WARNING);
|
|
|
93 |
} else {
|
|
|
94 |
PEAR::raiseError($error);
|
|
|
95 |
}
|
|
|
96 |
return false;
|
|
|
97 |
}
|
|
|
98 |
|
|
|
99 |
/**
|
|
|
100 |
* Checks the configuration given by the user
|
|
|
101 |
*
|
|
|
102 |
* @access private
|
|
|
103 |
* @param string &$error The error will be written here if any
|
|
|
104 |
* @param array &$conf The configuration assoc array
|
|
|
105 |
* @return string error Returns a error message
|
|
|
106 |
*/
|
|
|
107 |
function _conf(&$error, &$conf)
|
|
|
108 |
{
|
|
|
109 |
// check conf
|
|
|
110 |
if (!is_array($conf)) {
|
|
|
111 |
return $error = 'Invalid configuration';
|
|
|
112 |
}
|
|
|
113 |
|
|
|
114 |
if (!isset($conf['fields']) || !(int)$conf['fields']) {
|
|
|
115 |
return $error = 'The number of fields must be numeric (the "fields" key)';
|
|
|
116 |
}
|
|
|
117 |
|
|
|
118 |
if (isset($conf['sep'])) {
|
|
|
119 |
if (strlen($conf['sep']) !== 1) {
|
|
|
120 |
return $error = 'Separator can only be one char';
|
|
|
121 |
}
|
|
|
122 |
} elseif ($conf['fields'] > 1) {
|
|
|
123 |
return $error = 'Missing separator (the "sep" key)';
|
|
|
124 |
} else {
|
|
|
125 |
// to avoid undefined index notices
|
|
|
126 |
$conf['sep'] = ',';
|
|
|
127 |
}
|
|
|
128 |
|
|
|
129 |
if (isset($conf['quote'])) {
|
|
|
130 |
if (strlen($conf['quote']) !== 1) {
|
|
|
131 |
return $error = 'The quote char must be one char (the "quote" key)';
|
|
|
132 |
}
|
|
|
133 |
} else {
|
|
|
134 |
$conf['quote'] = '"';
|
|
|
135 |
}
|
|
|
136 |
|
|
|
137 |
if (!isset($conf['crlf'])) {
|
|
|
138 |
$conf['crlf'] = "\n";
|
|
|
139 |
}
|
|
|
140 |
|
|
|
141 |
if (!isset($conf['eol2unix'])) {
|
|
|
142 |
$conf['eol2unix'] = true;
|
|
|
143 |
}
|
|
|
144 |
}
|
|
|
145 |
|
|
|
146 |
/**
|
|
|
147 |
* Return or create the file descriptor associated with a file
|
|
|
148 |
*
|
|
|
149 |
* @param string $file The name of the file
|
|
|
150 |
* @param array &$conf The configuration
|
|
|
151 |
* @param string $mode The open node (ex: FILE_MODE_READ or FILE_MODE_WRITE)
|
|
|
152 |
* @param boolean $reset if passed as true and resource for the file exists
|
|
|
153 |
* than the file pointer will be moved to the beginning
|
|
|
154 |
*
|
|
|
155 |
* @return mixed A file resource or false
|
|
|
156 |
*/
|
|
|
157 |
function getPointer($file, &$conf, $mode = FILE_MODE_READ, $reset = false)
|
|
|
158 |
{
|
|
|
159 |
static $resources = array();
|
|
|
160 |
if (isset($resources[$file][$mode])) {
|
|
|
161 |
if ($reset) {
|
|
|
162 |
fseek($resources[$file][$mode], 0);
|
|
|
163 |
}
|
|
|
164 |
|
|
|
165 |
return $resources[$file][$mode];
|
|
|
166 |
}
|
|
|
167 |
|
|
|
168 |
File_CSV::_conf($error, $conf);
|
|
|
169 |
if ($error) {
|
|
|
170 |
return File_CSV::raiseError($error);
|
|
|
171 |
}
|
|
|
172 |
|
|
|
173 |
PEAR::pushErrorHandling(PEAR_ERROR_RETURN);
|
|
|
174 |
$fp = File::_getFilePointer($file, $mode);
|
|
|
175 |
PEAR::popErrorHandling();
|
|
|
176 |
if (PEAR::isError($fp)) {
|
|
|
177 |
return File_CSV::raiseError($fp);
|
|
|
178 |
}
|
|
|
179 |
|
|
|
180 |
$resources[$file][$mode] = $fp;
|
|
|
181 |
return $fp;
|
|
|
182 |
}
|
|
|
183 |
|
|
|
184 |
/**
|
|
|
185 |
* Unquote data
|
|
|
186 |
*
|
|
|
187 |
* @param array|string $data The data to unquote
|
|
|
188 |
* @param string $quote The quote char
|
|
|
189 |
* @return string the unquoted data
|
|
|
190 |
*/
|
|
|
191 |
function unquote($data, $quote)
|
|
|
192 |
{
|
|
|
193 |
$isString = false;
|
|
|
194 |
if (!is_array($data)) {
|
|
|
195 |
$data = array($data);
|
|
|
196 |
$isString = true;
|
|
|
197 |
}
|
|
|
198 |
|
|
|
199 |
// Get rid of escaped quotes
|
|
|
200 |
$data = str_replace($quote.$quote, $quote, $data);
|
|
|
201 |
|
|
|
202 |
$tmp = array();
|
|
|
203 |
foreach ($data as $key => $field) {
|
|
|
204 |
// Trim first the string.
|
|
|
205 |
$field = trim($field);
|
|
|
206 |
|
|
|
207 |
// Incase null fields (form: ;;)
|
|
|
208 |
$field_len = strlen($field);
|
|
|
209 |
if (!$field_len) {
|
|
|
210 |
if ($isString) {
|
|
|
211 |
return $field;
|
|
|
212 |
}
|
|
|
213 |
$tmp[$key] = $field;
|
|
|
214 |
continue;
|
|
|
215 |
}
|
|
|
216 |
|
|
|
217 |
// excel compat
|
|
|
218 |
if ($field[0] === '=' && $field[1] === '"') {
|
|
|
219 |
$field = str_replace('="', '"', $field);
|
|
|
220 |
--$field_len;
|
|
|
221 |
}
|
|
|
222 |
|
|
|
223 |
if ($field[0] === $quote && $field[$field_len - 1] === $quote) {
|
|
|
224 |
// Get rid of quotes around the field
|
|
|
225 |
$field = substr($field, 1, -1);
|
|
|
226 |
}
|
|
|
227 |
|
|
|
228 |
if ($isString) {
|
|
|
229 |
$tmp = $field;
|
|
|
230 |
} else {
|
|
|
231 |
$tmp[$key] = $field;
|
|
|
232 |
}
|
|
|
233 |
}
|
|
|
234 |
|
|
|
235 |
return $tmp;
|
|
|
236 |
}
|
|
|
237 |
|
|
|
238 |
/**
|
|
|
239 |
* Reads a row of data as an array from a CSV file. It's able to
|
|
|
240 |
* read memo fields with multiline data.
|
|
|
241 |
*
|
|
|
242 |
* @param string $file The filename where to write the data
|
|
|
243 |
* @param array &$conf The configuration of the dest CSV
|
|
|
244 |
*
|
|
|
245 |
* @return mixed Array with the data read or false on error/no more data
|
|
|
246 |
*/
|
|
|
247 |
function readQuoted($file, &$conf)
|
|
|
248 |
{
|
|
|
249 |
if (!$fp = File_CSV::getPointer($file, $conf, FILE_MODE_READ)) {
|
|
|
250 |
return false;
|
|
|
251 |
}
|
|
|
252 |
|
|
|
253 |
$buff = $old = $prev = $c = '';
|
|
|
254 |
$ret = array();
|
|
|
255 |
$fields = 1;
|
|
|
256 |
$in_quote = false;
|
|
|
257 |
$quote = $conf['quote'];
|
|
|
258 |
$f = (int)$conf['fields'];
|
|
|
259 |
$sep = $conf['sep'];
|
|
|
260 |
while (false !== $ch = fgetc($fp)) {
|
|
|
261 |
$old = $prev;
|
|
|
262 |
$prev = $c;
|
|
|
263 |
$c = $ch;
|
|
|
264 |
|
|
|
265 |
// Common case
|
|
|
266 |
if ($c != $quote && $c != $sep && $c != "\n" && $c != "\r") {
|
|
|
267 |
$buff .= $c;
|
|
|
268 |
continue;
|
|
|
269 |
}
|
|
|
270 |
|
|
|
271 |
// Start quote.
|
|
|
272 |
if (
|
|
|
273 |
$in_quote === false &&
|
|
|
274 |
$quote && $c == $quote &&
|
|
|
275 |
(
|
|
|
276 |
$prev == $sep || $prev == "\n" || $prev === null ||
|
|
|
277 |
$prev == "\r" || $prev == '' || $prev == ' '
|
|
|
278 |
|| $prev == '=' //excel compat
|
|
|
279 |
)
|
|
|
280 |
) {
|
|
|
281 |
$in_quote = true;
|
|
|
282 |
// excel compat, removing the = part but only if we are in a quote
|
|
|
283 |
if ($prev == '=') {
|
|
|
284 |
$buff{strlen($buff) - 1} = '';
|
|
|
285 |
}
|
|
|
286 |
}
|
|
|
287 |
|
|
|
288 |
if ($in_quote) {
|
|
|
289 |
// When does the quote end, make sure it's not double quoted
|
|
|
290 |
if ($c == $sep && $prev == $quote && $old != $quote) {
|
|
|
291 |
$in_quote = false;
|
|
|
292 |
} elseif ($c == $sep && $buff == $quote.$quote) {
|
|
|
293 |
// In case we are dealing with double quote but empty value
|
|
|
294 |
$in_quote = false;
|
|
|
295 |
} elseif ($c == "\n" || $c == "\r") {
|
|
|
296 |
$sub = ($prev == "\r") ? 2 : 1;
|
|
|
297 |
$buff_len = strlen($buff);
|
|
|
298 |
if (
|
|
|
299 |
$buff_len >= $sub &&
|
|
|
300 |
$buff[$buff_len - $sub] == $quote
|
|
|
301 |
) {
|
|
|
302 |
$in_quote = false;
|
|
|
303 |
}
|
|
|
304 |
}
|
|
|
305 |
}
|
|
|
306 |
|
|
|
307 |
if (!$in_quote && ($c == $sep || $c == "\n" || $c == "\r")) {
|
|
|
308 |
$return = File_CSV::_readQuotedFillers($fp, $f, $fields, $ret,
|
|
|
309 |
$buff, $quote, $c, $sep);
|
|
|
310 |
if ($return !== false) {
|
|
|
311 |
return $return;
|
|
|
312 |
}
|
|
|
313 |
|
|
|
314 |
if ($prev == "\r") {
|
|
|
315 |
$buff = substr($buff, 0, -1);
|
|
|
316 |
}
|
|
|
317 |
|
|
|
318 |
// Convert EOL character to Unix EOL (LF).
|
|
|
319 |
if ($conf['eol2unix']) {
|
|
|
320 |
$buff = preg_replace('/(\r\n|\r)$/', "\n", $buff);
|
|
|
321 |
// Below replaces things everywhere not just EOL
|
|
|
322 |
//$buff = str_replace(array("\r\n", "\r"), "\n", $buff);
|
|
|
323 |
}
|
|
|
324 |
|
|
|
325 |
$ret[] = File_CSV::unquote($buff, $quote);
|
|
|
326 |
if (count($ret) === $f) {
|
|
|
327 |
return $ret;
|
|
|
328 |
}
|
|
|
329 |
$buff = '';
|
|
|
330 |
++$fields;
|
|
|
331 |
continue;
|
|
|
332 |
}
|
|
|
333 |
|
|
|
334 |
$buff .= $c;
|
|
|
335 |
}
|
|
|
336 |
|
|
|
337 |
/* If it's the end of the file and we still have something in buffer
|
|
|
338 |
* then we process it since files can have no CL/FR at the end
|
|
|
339 |
*/
|
|
|
340 |
$feof = feof($fp);
|
|
|
341 |
if ($feof && strlen($buff) > 0 && !in_array($buff, array("\r", "\n"))) {
|
|
|
342 |
$ret[] = File_CSV::unquote($buff, $quote);
|
|
|
343 |
if (count($ret) == $f) {
|
|
|
344 |
return $ret;
|
|
|
345 |
}
|
|
|
346 |
}
|
|
|
347 |
|
|
|
348 |
if ($feof && count($ret) !== $f) {
|
|
|
349 |
$return = File_CSV::_readQuotedFillers($fp, $f, $fields, $ret,
|
|
|
350 |
$buff, $quote, $c, $sep);
|
|
|
351 |
if ($return !== false) {
|
|
|
352 |
return $return;
|
|
|
353 |
}
|
|
|
354 |
}
|
|
|
355 |
|
|
|
356 |
return !$feof ? $ret : false;
|
|
|
357 |
}
|
|
|
358 |
|
|
|
359 |
/**
|
|
|
360 |
* Adds missing fields (empty ones)
|
|
|
361 |
*
|
|
|
362 |
* @param resource $fp the file resource
|
|
|
363 |
* @param string $f
|
|
|
364 |
* @param integer $fields the field count
|
|
|
365 |
* @param array $ret the processed fields in a array
|
|
|
366 |
* @param string $buff the buffer before it gets put through unquote
|
|
|
367 |
* @param string $quote Quote in use
|
|
|
368 |
* @param string $c the char currently being worked with
|
|
|
369 |
* @param string $sep Separator in use
|
|
|
370 |
*
|
|
|
371 |
* @access private
|
|
|
372 |
* @return array | boolean returns false if no data should return out.
|
|
|
373 |
*/
|
|
|
374 |
function _readQuotedFillers($fp, $f, $fields, $ret, $buff, $quote, &$c, $sep)
|
|
|
375 |
{
|
|
|
376 |
// More fields than expected
|
|
|
377 |
if ($c == $sep && (count($ret) + 1) === $f) {
|
|
|
378 |
// Seek the pointer into linebreak character.
|
|
|
379 |
while (true) {
|
|
|
380 |
$c = fgetc($fp);
|
|
|
381 |
if ($c == "\n" || $c == "\r" || $c == '') {
|
|
|
382 |
break;
|
|
|
383 |
}
|
|
|
384 |
}
|
|
|
385 |
|
|
|
386 |
// Insert last field value.
|
|
|
387 |
$ret[] = File_CSV::unquote($buff, $quote);
|
|
|
388 |
return $ret;
|
|
|
389 |
}
|
|
|
390 |
|
|
|
391 |
// Less fields than expected
|
|
|
392 |
if (($c == "\n" || $c == "\r") && $fields !== $f) {
|
|
|
393 |
// Insert last field value.
|
|
|
394 |
$ret[] = File_CSV::unquote($buff, $quote);
|
|
|
395 |
if (count($ret) === 1 && empty($ret[0])) {
|
|
|
396 |
return array();
|
|
|
397 |
}
|
|
|
398 |
|
|
|
399 |
// Pair the array elements to fields count. - inserting empty values
|
|
|
400 |
$ret_count = count($ret);
|
|
|
401 |
$sum = ($f - 1) - ($ret_count - 1);
|
|
|
402 |
$data = array_merge($ret, array_fill($ret_count, $sum, ''));
|
|
|
403 |
return $data;
|
|
|
404 |
}
|
|
|
405 |
|
|
|
406 |
return false;
|
|
|
407 |
}
|
|
|
408 |
|
|
|
409 |
/**
|
|
|
410 |
* Reads a "row" from a CSV file and return it as an array
|
|
|
411 |
*
|
|
|
412 |
* @param string $file The CSV file
|
|
|
413 |
* @param array &$conf The configuration of the dest CSV
|
|
|
414 |
*
|
|
|
415 |
* @return mixed Array or false
|
|
|
416 |
*/
|
|
|
417 |
function read($file, &$conf)
|
|
|
418 |
{
|
|
|
419 |
if (!$fp = File_CSV::getPointer($file, $conf, FILE_MODE_READ)) {
|
|
|
420 |
return false;
|
|
|
421 |
}
|
|
|
422 |
|
|
|
423 |
// The size is limited to 4K
|
|
|
424 |
if (!$line = fgets($fp, 4096)) {
|
|
|
425 |
return false;
|
|
|
426 |
}
|
|
|
427 |
|
|
|
428 |
if ($conf['fields'] === 1) {
|
|
|
429 |
$fields = array($line);
|
|
|
430 |
$field_count = 1;
|
|
|
431 |
} else {
|
|
|
432 |
$fields = explode($conf['sep'], $line);
|
|
|
433 |
$field_count = count($fields);
|
|
|
434 |
}
|
|
|
435 |
|
|
|
436 |
$real_field_count = $field_count - 1;
|
|
|
437 |
$check_char = $fields[$real_field_count];
|
|
|
438 |
if ($check_char === "\n" || $check_char === "\r") {
|
|
|
439 |
array_pop($fields);
|
|
|
440 |
--$field_count;
|
|
|
441 |
}
|
|
|
442 |
|
|
|
443 |
$last =& $fields[$real_field_count];
|
|
|
444 |
if (
|
|
|
445 |
$field_count !== $conf['fields'] || $conf['quote']
|
|
|
446 |
&& (
|
|
|
447 |
$last !== ''
|
|
|
448 |
&& (
|
|
|
449 |
($last[0] === $conf['quote'] && $last[strlen(rtrim($last)) - 1] !== $conf['quote'])
|
|
|
450 |
// excel support
|
|
|
451 |
|| ($last[0] === '=' && $last[1] === $conf['quote'])
|
|
|
452 |
// if the row has spaces or other extra chars before the quote
|
|
|
453 |
//|| preg_match('|^\s+\\' . $conf['quote'] .'|', $last)
|
|
|
454 |
)
|
|
|
455 |
)
|
|
|
456 |
// XXX perhaps there is a separator inside a quoted field
|
|
|
457 |
// || preg_match("|{$conf['quote']}.*{$conf['sep']}.*{$conf['quote']}|", $line)
|
|
|
458 |
// The regex above is really slow
|
|
|
459 |
|| ((count(explode(',', $line))) > $field_count)
|
|
|
460 |
) {
|
|
|
461 |
fseek($fp, -1 * strlen($line), SEEK_CUR);
|
|
|
462 |
$fields = File_CSV::readQuoted($file, $conf);
|
|
|
463 |
$fields = File_CSV::_processHeaders($fields, $conf);
|
|
|
464 |
return $fields;
|
|
|
465 |
}
|
|
|
466 |
|
|
|
467 |
$fields = File_CSV::unquote($fields, $conf['quote']);
|
|
|
468 |
|
|
|
469 |
if ($field_count != $conf['fields']) {
|
|
|
470 |
File_CSV::raiseError("Read wrong fields number count: '". $field_count .
|
|
|
471 |
"' expected ".$conf['fields']);
|
|
|
472 |
return true;
|
|
|
473 |
}
|
|
|
474 |
|
|
|
475 |
$fields = File_CSV::_processHeaders($fields, $conf);
|
|
|
476 |
return $fields;
|
|
|
477 |
}
|
|
|
478 |
|
|
|
479 |
/**
|
|
|
480 |
* Process the field array being passed in and map the array over to
|
|
|
481 |
* the header values if the configuration is set on.
|
|
|
482 |
*
|
|
|
483 |
* @param array $fields The CSV row columns
|
|
|
484 |
* @param array $conf File_CSV configuration
|
|
|
485 |
*
|
|
|
486 |
* @return array Processed array
|
|
|
487 |
*/
|
|
|
488 |
function _processHeaders($fields, &$conf)
|
|
|
489 |
{
|
|
|
490 |
static $headers = array();
|
|
|
491 |
|
|
|
492 |
if (isset($conf['header']) && $conf['header'] == true && empty($headers)) {
|
|
|
493 |
// read the first row and assign to $headers
|
|
|
494 |
$headers = $fields;
|
|
|
495 |
return $headers;
|
|
|
496 |
}
|
|
|
497 |
|
|
|
498 |
if (!empty($headers)) {
|
|
|
499 |
$tmp = array();
|
|
|
500 |
foreach ($fields as $k => $v) {
|
|
|
501 |
if (isset($headers[$k])) {
|
|
|
502 |
$tmp[$headers[$k]] = $v;
|
|
|
503 |
}
|
|
|
504 |
}
|
|
|
505 |
$fields = $tmp;
|
|
|
506 |
}
|
|
|
507 |
|
|
|
508 |
return $fields;
|
|
|
509 |
}
|
|
|
510 |
|
|
|
511 |
/**
|
|
|
512 |
* Internal use only, will be removed in the future
|
|
|
513 |
*
|
|
|
514 |
* @param string $str The string to debug
|
|
|
515 |
* @access private
|
|
|
516 |
*/
|
|
|
517 |
function _dbgBuff($str)
|
|
|
518 |
{
|
|
|
519 |
if (strpos($str, "\r") !== false) {
|
|
|
520 |
$str = str_replace("\r", "_r_", $str);
|
|
|
521 |
}
|
|
|
522 |
if (strpos($str, "\n") !== false) {
|
|
|
523 |
$str = str_replace("\n", "_n_", $str);
|
|
|
524 |
}
|
|
|
525 |
if (strpos($str, "\t") !== false) {
|
|
|
526 |
$str = str_replace("\t", "_t_", $str);
|
|
|
527 |
}
|
|
|
528 |
if ($str === null) {
|
|
|
529 |
$str = '_NULL_';
|
|
|
530 |
}
|
|
|
531 |
if ($str === '') {
|
|
|
532 |
$str = 'Empty string';
|
|
|
533 |
}
|
|
|
534 |
echo "buff: ($str)\n";
|
|
|
535 |
}
|
|
|
536 |
|
|
|
537 |
/**
|
|
|
538 |
* Writes a struc (array) in a file as CSV
|
|
|
539 |
*
|
|
|
540 |
* @param string $file The filename where to write the data
|
|
|
541 |
* @param array $fields Ordered array with the data
|
|
|
542 |
* @param array &$conf The configuration of the dest CSV
|
|
|
543 |
*
|
|
|
544 |
* @return bool True on success false otherwise
|
|
|
545 |
*/
|
|
|
546 |
function write($file, $fields, &$conf)
|
|
|
547 |
{
|
|
|
548 |
if (!$fp = File_CSV::getPointer($file, $conf, FILE_MODE_WRITE)) {
|
|
|
549 |
return false;
|
|
|
550 |
}
|
|
|
551 |
|
|
|
552 |
$field_count = count($fields);
|
|
|
553 |
if ($field_count != $conf['fields']) {
|
|
|
554 |
File_CSV::raiseError("Wrong fields number count: '". $field_count .
|
|
|
555 |
"' expected ".$conf['fields']);
|
|
|
556 |
return true;
|
|
|
557 |
}
|
|
|
558 |
|
|
|
559 |
$write = '';
|
|
|
560 |
$quote = $conf['quote'];
|
|
|
561 |
for ($i = 0; $i < $field_count; ++$i) {
|
|
|
562 |
// Write a single field
|
|
|
563 |
|
|
|
564 |
$quote_field = false;
|
|
|
565 |
// Only quote this field in the following cases:
|
|
|
566 |
if (is_numeric($fields[$i])) {
|
|
|
567 |
// Numeric fields should not be quoted
|
|
|
568 |
} elseif (isset($conf['sep']) && (strpos($fields[$i], $conf['sep']) !== false)) {
|
|
|
569 |
// Separator is present in field
|
|
|
570 |
$quote_field = true;
|
|
|
571 |
} elseif (strpos($fields[$i], $quote) !== false) {
|
|
|
572 |
// Quote character is present in field
|
|
|
573 |
$quote_field = true;
|
|
|
574 |
} elseif (
|
|
|
575 |
strpos($fields[$i], "\n") !== false
|
|
|
576 |
|| strpos($fields[$i], "\r") !== false
|
|
|
577 |
) {
|
|
|
578 |
// Newline is present in field
|
|
|
579 |
$quote_field = true;
|
|
|
580 |
} elseif (!is_numeric($fields[$i]) && (substr($fields[$i], 0, 1) == " " || substr($fields[$i], -1) == " ")) {
|
|
|
581 |
// Space found at beginning or end of field value
|
|
|
582 |
$quote_field = true;
|
|
|
583 |
}
|
|
|
584 |
|
|
|
585 |
if ($quote_field) {
|
|
|
586 |
// Escape the quote character within the field (e.g. " becomes "")
|
|
|
587 |
$quoted_value = str_replace($quote, $quote.$quote, $fields[$i]);
|
|
|
588 |
|
|
|
589 |
$write .= $quote . $quoted_value . $quote;
|
|
|
590 |
} else {
|
|
|
591 |
$write .= $fields[$i];
|
|
|
592 |
}
|
|
|
593 |
|
|
|
594 |
$write .= ($i < ($field_count - 1)) ? $conf['sep']: $conf['crlf'];
|
|
|
595 |
}
|
|
|
596 |
|
|
|
597 |
if (!fwrite($fp, $write, strlen($write))) {
|
|
|
598 |
return File_CSV::raiseError('Can not write to file');
|
|
|
599 |
}
|
|
|
600 |
|
|
|
601 |
return true;
|
|
|
602 |
}
|
|
|
603 |
|
|
|
604 |
/**
|
|
|
605 |
* Discover the format of a CSV file (the number of fields, the separator
|
|
|
606 |
* and if it quote string fields)
|
|
|
607 |
*
|
|
|
608 |
* @param string the CSV file name
|
|
|
609 |
* @param array extra separators that should be checked for.
|
|
|
610 |
* @return mixed Assoc array or false
|
|
|
611 |
*/
|
|
|
612 |
function discoverFormat($file, $extraSeps = array())
|
|
|
613 |
{
|
|
|
614 |
if (!$fp = @fopen($file, 'rb')) {
|
|
|
615 |
return File_CSV::raiseError("Could not open file: $file");
|
|
|
616 |
}
|
|
|
617 |
|
|
|
618 |
// Set auto detect line ending for Mac EOL support
|
|
|
619 |
$oldini = ini_get('auto_detect_line_endings');
|
|
|
620 |
if ($oldini != '1') {
|
|
|
621 |
ini_set('auto_detect_line_endings', '1');
|
|
|
622 |
}
|
|
|
623 |
|
|
|
624 |
// Take the first 30 lines and store the number of occurrences
|
|
|
625 |
// for each separator in each line
|
|
|
626 |
$lines = '';
|
|
|
627 |
for ($i = 0; $i < 30 && !feof($fp) && $line = fgets($fp, 4096); $i++) {
|
|
|
628 |
$lines .= $line;
|
|
|
629 |
}
|
|
|
630 |
fclose($fp);
|
|
|
631 |
|
|
|
632 |
if ($oldini != '1') {
|
|
|
633 |
ini_set('auto_detect_line_endings', $oldini);
|
|
|
634 |
}
|
|
|
635 |
|
|
|
636 |
$seps = array("\t", ';', ':', ',');
|
|
|
637 |
$seps = array_merge($seps, $extraSeps);
|
|
|
638 |
$matches = array();
|
|
|
639 |
$quotes = '"\'';
|
|
|
640 |
|
|
|
641 |
while ($lines != ($newLines = preg_replace('|((["\'])[^"]*(\2))|', '\2_\2', $lines))) {
|
|
|
642 |
$lines = $newLines;
|
|
|
643 |
}
|
|
|
644 |
|
|
|
645 |
$eol = strpos($lines, "\r") ? "\r" : "\n";
|
|
|
646 |
$lines = explode($eol, $lines);
|
|
|
647 |
foreach ($lines as $line) {
|
|
|
648 |
$orgLine = $line;
|
|
|
649 |
foreach ($seps as $sep) {
|
|
|
650 |
$line = preg_replace("|^[$quotes$sep]*$sep*([$quotes][^$quotes]*[$quotes])|sm", '_', $orgLine);
|
|
|
651 |
// Find all seps that are within qoutes
|
|
|
652 |
///FIXME ... counts legitimit lines as bad ones
|
|
|
653 |
|
|
|
654 |
// In case there's a whitespace infront the field
|
|
|
655 |
$regex = '|\s*?';
|
|
|
656 |
// Match the first quote (optional), also optionally match = since it's excel stuff
|
|
|
657 |
$regex.= "(?:\=?[$quotes])";
|
|
|
658 |
$regex.= '(.*';
|
|
|
659 |
// Don't match a sep if we are inside a quote
|
|
|
660 |
// also don't accept the sep if it has a quote on the either side
|
|
|
661 |
///FIXME has to be possible if we are inside a quote! (tests fail because of this)
|
|
|
662 |
$regex.= "(?:[^$quotes])$sep(?:[^$quotes])";
|
|
|
663 |
$regex.= '.*)';
|
|
|
664 |
// Close quote (if it's present) and the sep (optional, could be end of line)
|
|
|
665 |
$regex.= "(?:[$quotes](?:$sep?))|Ums";
|
|
|
666 |
|
|
|
667 |
preg_match_all($regex, $line, $match);
|
|
|
668 |
// Finding all seps, within quotes or not
|
|
|
669 |
$sep_count = substr_count($line, $sep);
|
|
|
670 |
// Real count
|
|
|
671 |
$matches[$sep][] = $sep_count - count($match[0]);
|
|
|
672 |
}
|
|
|
673 |
}
|
|
|
674 |
|
|
|
675 |
$final = array();
|
|
|
676 |
// Group the results by amount of equal ocurrences
|
|
|
677 |
foreach ($matches as $sep => $res) {
|
|
|
678 |
$times = array();
|
|
|
679 |
$times[0] = 0;
|
|
|
680 |
foreach ($res as $k => $num) {
|
|
|
681 |
if ($num > 0) {
|
|
|
682 |
$times[$num] = isset($times[$num]) ? $times[$num] + $num : 1;
|
|
|
683 |
}
|
|
|
684 |
}
|
|
|
685 |
arsort($times);
|
|
|
686 |
|
|
|
687 |
// Use max fields count.
|
|
|
688 |
$fields[$sep] = max(array_flip($times));
|
|
|
689 |
$amount[$sep] = $times[key($times)];
|
|
|
690 |
}
|
|
|
691 |
|
|
|
692 |
arsort($amount);
|
|
|
693 |
$sep = key($amount);
|
|
|
694 |
|
|
|
695 |
$conf['fields'] = $fields[$sep] + 1;
|
|
|
696 |
$conf['sep'] = $sep;
|
|
|
697 |
|
|
|
698 |
// Test if there are fields with quotes around in the first 30 lines
|
|
|
699 |
$quote = null;
|
|
|
700 |
|
|
|
701 |
$string = implode('', $lines);
|
|
|
702 |
foreach (array('"', '\'') as $q) {
|
|
|
703 |
if (preg_match_all("|$sep(?:\s*?)(\=?[$q]).*([$q])$sep?|Us", $string, $match)) {
|
|
|
704 |
if ($match[1][0] == $match[2][0]) {
|
|
|
705 |
$quote = $match[1][0];
|
|
|
706 |
break;
|
|
|
707 |
}
|
|
|
708 |
}
|
|
|
709 |
|
|
|
710 |
if (
|
|
|
711 |
preg_match_all("|^(\=?[$q]).*([$q])$sep{0,1}|Ums", $string, $match)
|
|
|
712 |
|| preg_match_all("|(\=?[$q]).*([$q])$sep\s$|Ums", $string, $match)
|
|
|
713 |
) {
|
|
|
714 |
if ($match[1][0] == $match[2][0]) {
|
|
|
715 |
$quote = $match[1][0];
|
|
|
716 |
break;
|
|
|
717 |
}
|
|
|
718 |
}
|
|
|
719 |
}
|
|
|
720 |
|
|
|
721 |
$conf['quote'] = $quote;
|
|
|
722 |
return $conf;
|
|
|
723 |
}
|
|
|
724 |
|
|
|
725 |
/**
|
|
|
726 |
* Front to call getPointer and moving the resource to the
|
|
|
727 |
* beginning of the file
|
|
|
728 |
* Reset it if you like.
|
|
|
729 |
*
|
|
|
730 |
* @param string $file The name of the file
|
|
|
731 |
* @param array &$conf The configuration
|
|
|
732 |
* @param string $mode The open node (ex: FILE_MODE_READ or FILE_MODE_WRITE)
|
|
|
733 |
*
|
|
|
734 |
* @return boolean true on success false on failure
|
|
|
735 |
*/
|
|
|
736 |
function resetPointer($file, &$conf, $mode)
|
|
|
737 |
{
|
|
|
738 |
if (!File_CSV::getPointer($file, $conf, $mode, true)) {
|
|
|
739 |
return false;
|
|
|
740 |
}
|
|
|
741 |
|
|
|
742 |
return true;
|
|
|
743 |
}
|
|
|
744 |
}
|