Issue with string comparison in PHP

Question

I have two strings with seemingly the same values. One is stored as a key in an array, the other a value in another different array. I compare the two using ==, ===, and strcmp. All treat them as different strings. I do a var_dump and this is what I get.

string(17) "Valentine’s Day" 
string(15) "Valentine's Day"

Does anyone have any idea why the first string would be 17 characters and the second 15?

Update: This is slightly more obvious when I pasted this out of my editor whose font made the two different apostrophe's almost indistinguishable.

looks like an issue with different character encodings, notice that the apostrophe is different in each case — GSto
– GSto, Commented Jan 3, 2011 at 21:23

Mark Biek · Accepted Answer · 2011-01-03 21:40:04Z

8

The first string contains a Unicode character for the apostrophe while the second string just has a regular ASCII ' character.

The Unicode character takes up more space.

If you run the PHP ord() function on each of those characters you'll see that you get different values for each:

echo ord("’"); //226 This is just the first 2 bytes (see comments below for details from ircmaxell)
echo ord("'"); //27

edited Jan 3, 2011 at 21:40

answered Jan 3, 2011 at 21:25

Mark Biek

152k54 gold badges160 silver badges201 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Phoenix Over a year ago

Bingo. It's all in the apostrophe.

simshaun Over a year ago

Just to clarify a little, "takes up more space" means that that unicode character takes multiple bytes to represent. PHP is not currently multibyte safe. For that, there are mb_* functions.

ircmaxell Over a year ago

Pedantic note: the 226 result is only the first of the 3 bytes that make up that character (226 is the starting block for a 3 byte block). So there's 2 more bytes that ord silently discards (which is why you see the extra 2 characters)...

netcoder · Accepted Answer · 2011-01-03 21:36:36Z

1

As a complement to @Mark answer above which is right (the ’ is a multi-byte character, most probably UTF-8, while ' is not). You can easily convert it to ASCII (or ISO-8859-1) using iconv, per example:

echo iconv('utf-8', 'ascii//TRANSLIT', $str);

Note: Not all characters can be transformed from multi-byte to ASCII or latin1. You can use //IGNORE to have them removed from the resulting string.

edited Jan 3, 2011 at 21:36

answered Jan 3, 2011 at 21:30

netcoder

68k19 gold badges129 silver badges142 bronze badges

Comments

dqhendricks · Accepted Answer · 2011-01-03 21:27:59Z

0

’ != '

mainly. if you want this not to be an issue, you could do something like this.

if (str_replace('’', '\'', "Valentine’s Day") == "Valentine's Day") {

answered Jan 3, 2011 at 21:27

dqhendricks

19.2k11 gold badges53 silver badges87 bronze badges

Collectives™ on Stack Overflow

Issue with string comparison in PHP

3 Answers 3

3 Comments

Comments

Comments

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

3 Comments

Comments

Comments

Related