Weird characters appear after I use php's mb_substr() on a string

Question

I'm developing a web site with PHP (5.3.5, Ubuntu) and all the content is in Spanish. I would like to cut the text when it doesn't fit the space designated for it. I have the following meta tag in the php file where I want to do this: <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />.

The text comes from a MySQL database where charset is latin1 and collation latin1_spanish_ci. I'm trying to cut the text with the mb_substr() function. But it isn't working correctly. For example, let's say I want to cut Short Psicodélico to Short Psicodéli, the function would be:

mb_substr('Short Psicodélico', 0, 15, 'ISO-8859-1');

But the result is something like this: Short Psicod&ea. The e with the diacritic is transformed in &ea and I don't know why. I think it has something to do with the character encoding but I don't know exactly how. If I don't use this function the characters appear as they should, instead of Short Psicod&ea it shows Short Psicodélico.

Answer 1

The text is encoded in the database as "Short Psicodélico". You will need to scrub your database to remove the encoding, as well as fix your input routines to make certain that text is not saved to the database encoded.