Hi I'm in need of parsing EDI files and I found your library after a

Swedish characters utf8 issue about edifact HOT 7 CLOSED

php-edifact commented on May 28, 2024

Swedish characters utf8 issue

from edifact.

Comments (7)

sabas commented on May 28, 2024

Uhm, it's strange!
Does it happen also when you var_dump the array?
What's the charset in UNB?
Could you send me an example via email?

from edifact.

ArvidEnbom commented on May 28, 2024

Yes, it does also happen if I var_dump the $parser->get() array.

The EDI file I was testing with was using a file encoding of UTF-8, however this is literally the first time I work with this file format, so I have no idea what you mean by "UNB".
This EDI file gave me this error There's a not printable character on line... seven times (once for each line with a swedish character, I'm assuming)

I have two example EDI files, however, and the second one was using Windows 1252 encoding instead of UTF-8. I tried this one as well, and it no longer gave me the not printable character error, so at first I thought it worked, but then I scrolled down and found that it hadn't worked.

I'm not sure if I am allowed to send you one of the EDI files, so we'll have to hold off on that for now.

from edifact.

sabas commented on May 28, 2024

Try this, load your file with file_get_contents in a variable $text

$p = new Parser();
$p->setStripRegex("/[\x01-\x1F\x7F-\x9F]/"); //or something less restricting
$p->loadString($text);

from edifact.

ArvidEnbom commented on May 28, 2024

I'm already sort of doing that. Here's my entire debug file currently

<!DOCTYPE html>
<html lang="sv">
	<head>
		<meta charset="UTF8">
	</head>
	<body>
		<pre><?php

			require( __DIR__ . "/EDI/Parser.php" );

			$edi = file_get_contents( "in.edi" );

			$parser = new \EDI\Parser($edi);

			var_dump( $parser->get() );
		?></pre>
	</body>
</html>

I'll try your above code too, one moment

EDIT: Tried your above code, no change. Also tried changing the pattern a bunch, including these

$parser->setStripRegex("[\x01-\x1]");
$parser->setStripRegex("[\x7F-\x9F]");
$parser->setStripRegex("//");
$parser->setStripRegex("/MATCH NOTHING PLEASE/");

none of those worked.

from edifact.

sabas commented on May 28, 2024

If you var dump $edi you see the correct chars?

If you generate an anonymized version I can look at it, because with other diacritics I see it works as expected...

from edifact.

ArvidEnbom commented on May 28, 2024

Yes, I do see the correct chars then
I've asked to see if I was allowed to share a file, so I'm sending it to you via email

from edifact.

ArvidEnbom commented on May 28, 2024

To anyone else reading this: After some emails back and forth we found the solution.

use utf8_decode on the file contents

$file = utf8_decode(file_get_contents($path));
$parser = new EDI\Parser($file);
// ... and so on

And then, use utf8_encode($var) every time you read a value from the resulting array.

This will fix it.

from edifact.

Swedish characters utf8 issue about edifact HOT 7 CLOSED

Comments (7)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent