PHP's implementation of fgetcsv() and fputcsv() does not follow RFC 4180 or Wikipedia's description, and does not work in all cases with Excel or OO Calc, as described in bug http://bugs.php.net/bug.php?id=50686 .

Here is an RFC 4180-complient implementation in PHP. This can parse more than 1MiB of CSV text per second, so should be fast enough for most uses. It is fairly strict, in that it does not allow whitespace before the start of a quoted item, such as 1, "2" ,3 , in according to the standard, but it would be simple to make it laxer. It supports both \r\n and \n linebreaks, where the standard only supports \r\n. whitespace is just another data character, so '1, 2,3 ' becomes Array(1, " 2", "3 "). backslashes are also just another character, as backslashes have no special meaning in RFC 4180.

The code below may be used and redistributed without restrictions. You may consider buying me a beer if you find it useful.


class csv {

private static function error($error, $str, $m, $offset, $moving_offset, $csv_line) {
if ($error === "unexpected quote in unquoted field") {
$raw = _("Found an unexpected quote character in field %d of csv line %d (text line %d).".
" The first 50 chars from the start of the field are '%s'.");
} else if ($error === "unexpected quote in quoted field") {
$raw = _("Found an unescaped quote in quoted field %d of csv line %d (text line %d).".
" The first 50 chars from the start of the field are '%s'.");
} else if ($error === "unexpected text after end quote in quoted field") {
$raw = _("Unexpected test after end quote in quoted field %d of csv line %d (text line %d).".
" The first 50 chars from the start of the field are '%s'.");
} else {
die("impossible");
}

$t = sprintf($raw,
sizeof($m[$csv_line])+1,
$csv_line,
sizeof(explode("\n", substr($str, 0, $offset))),
addcslashes(substr($str, $offset, 50), "\t\r\n'\\")
);
throw new Exception($t);
}

/* returns
* array(start_line1 => Array(field1, field2,...),
* start_line2 =>
* ...);
*
* Faily speed-important, therefore:
* -With few function calls.
* -Never copy the entire (potentially MB-long) string
* -Never operate on the whole string (so therefore we use no pregs)
*
* This implementation parses a 1MB csv file in under a second,
* which is obviously slow, but should be fast enough. My first try
* was using regexp_max with regexp's offset parameter, which failed
* horribly speed-wise.
*
* The basic idea is to eat one char at a time, and append fields
* and lines to the matrix as we encounter separators and newlines.
*
* Not using PHP's fgetcsvdue to http://bugs.php.net/bug.php?id=50686
*/
public static function parse($str, $sep=',') {
if ($str === "") {
return Array();
}

$str .= "\r\n";
$offset = 0;
$item = "";
$len = strlen($str);
$csv_line = 1;
$m = Array($csv_line => Array());
$force_empty_field = false;

$moving_offset = 0;
while ($moving_offset < $len) {
$c = $str[$moving_offset];
if ($c === $sep) {
//separator
$m[$csv_line][] = $item;
$item = "";
$force_empty_field = true;
$offset = ++$moving_offset;
} else if ($c === "\n") {
//newline
if ($str[$moving_offset -1] === "\r") {
//The \r belonged to the newline
$item = substr($item, 0, -1);
}
if ($item !== "" || $force_empty_field) {
$m[$csv_line][] = $item;
$item = "";
}
$offset = ++$moving_offset;

//end of string
if ($offset === $len) {
return $m;
}

$m[++$csv_line] = Array();
$force_empty_field = false;
} else if ($c === '"') {
//quoted item

if ($item !== "") {
//throws exception
self::error("unexpected quote in unquoted field",
$str, $m, $offset, $moving_offset, $csv_line);
}

//eat quote
$moving_offset++;

//read until end quote
while (true) {
$c = $str[$moving_offset];

if ($moving_offset >= $len) {
//throws exception
self::error("unexpected quote in quoted field",
$str, $m, $offset, $moving_offset, $csv_line);
}
if ($c === '"') {
if ($str[$moving_offset+1] === '"') {
//escaped quote
$item .= '"';
$moving_offset += 2; //eat doubled quotes
} else {
//end of item
$moving_offset++; //eat end quote
break;
}
} else {
$item .= $c;
$moving_offset++;
}
}

//eat separator
if ($str[$moving_offset] === $sep) {
$force_empty_field = true;
$moving_offset++;
} else if ( ($str[$moving_offset] === "\r" && $str[$moving_offset+1] === "\n")
|| $str[$moving_offset] === "\n") {
$force_empty_field = false;
} else {
self::error("unexpected text after end quote in quoted field",
$str, $m, $offset, $moving_offset, $csv_line);
}

//add field and reset for next field
$m[$csv_line][] = $item;
$item = "";
$offset = $moving_offset;
} else {
$moving_offset++;
$item .= $c;
}
}

die("impossible, since the last char is a \n, and the newline handling should catch that");
}

public static function matrix_to_csv(Array $matrix, $sep=',') {
$str = "";
foreach ($matrix as $row) {
if ($str !== "") {
$str .= "\r\n"; //not just \n, according to RFC 4180
}

$row = array_map(Array(__CLASS__, "escape"), $row);
$row_str = implode($sep, $row);
$str .= $row_str;
}

return $str;
}
}

?>

Kommentarer

  1. Thank you very much!

    SvarSlet
  2. If you need to parse a RFC4180-style CSV file, but don't want to read the whole file into memory first, go to
    http://www.ubercart.org/project/uc_stock_update-2x
    and grab the latest uc_stock_update-*.tar.gz

    In that tarball you'll find uc_stock_update/ReadCSV.inc which uses a finite state machine to parse a CSV file. I think it handles RFC4180-compliant files correctly, although it currently tolerates many error cases. There's a little test suite in uc_stock_update/tests/ . Feedback welcome (see the download page).

    SvarSlet
  3. I've read your bug report on php.net and totally agree on why php should adapt to the RFC. Too bad Iliaa hasn't responded on your last comment.

    SvarSlet

Send en kommentar

Populære opslag fra denne blog

Status for Universal Power Adapter for Mobile Devices (2019)

Making the Trust 1200 Wireless tablet work on Linux