Handling of files and data containing ® (registered trademark symbol)

**swguy** · 9 Aug 2012, 01:40 AM

Zen Cart does not handle this well right now. For instance, a ® symbol in a file in html_includes will yield this error when the defines pages editor is used from admin:

[09-Aug-2012 00:35:33 UTC] PHP Warning: htmlspecialchars(): Invalid multibyte sequence in argument in /var/www/html/mysite/myadmin/define_pages_editor.php on line 188

... and the define page is blank in admin (since the default behavior of htmlspecialchars is to return nothing in the event of an error.)

This can be resolved by using ENT_IGNORE in the htmlspecialchars call in admin/define_pages_editor.php line 194.

Of course, there are a number of htmlspecialchars calls ... and unfortunately a lot of the cart's descriptions contain the registered trademark symbol.

Any other suggestions?

**DrByte** · 9 Aug 2012, 01:57 AM

What PHP version?
What CHARSET and DB_CHARSET ?

**swguy** · 9 Aug 2012, 02:01 AM

PHP 5.3.10 (local) and 5.2.17 (on server);
CHARSET/DB_CHARSET utf8.

**DrByte** · 9 Aug 2012, 02:08 AM

I'm not seeing any such characters in a fresh 1.5.0 install.
And inserting ® via the admin, saving, re-opening, saving, re-opening, etc, doesn't trigger any problems here. PHP 5.3.3 on this particular site.

Did your files come from an older version, based on iso-8859-1 character set?

**swguy** · 9 Aug 2012, 02:09 AM

Yes. Add an ® to a file in html_includes/YOUR_TEMPLATE using a text editor, then try to open it from defines pages editor.

**DrByte** · 9 Aug 2012, 04:16 AM

Okay, when I use a text-editor and tell it to save the file as iso-8859-1 and add a multibyte character to it and save and then load it to my server and try to open it in Zen Cart which is set to UTF8, then I get the same problem.

And that's to be expected.

But if I set my editor to use UTF8 just like my Zen Cart is set to use UTF8, then I don't have any problem. And of course one should always use the same character-set / encoding consistently across all editing/storage platforms.

**swguy** · 9 Aug 2012, 05:24 AM

Fair enough. So really, what I should have done is just keep the whole system at iso-8859-1.

**chadderuski** · 22 Aug 2012, 05:34 PM

Originally Posted by swguy

Fair enough. So really, what I should have done is just keep the whole system at iso-8859-1.

Hey swguy!

I find it easiest with (r) and (c) and a number of symbols to just use the html equivalents. This really simplifies things for me.

Code:

&reg; for (r)
&copy; for (c)
&trade; for TM

-chadd

**swguy** · 22 Aug 2012, 05:41 PM

I quite agree if you're starting from scratch - but in this case, the data already had these symbols littered throughout many many fields in the database.

**torvista** · 23 Aug 2012, 08:34 AM

what I should have done is just keep the whole system at iso-8859-1

Why? Can't you set the text editor to UTF-8?

I use Dreamweaver to generate/edit complex html for html_includes or Notepad++ or Sublime Text and all can be set to UTF-8 as default.