ucfirst() function for multibyte character encodings

Question

I've asked about strtolower function. But when using foreign characters it doesn't convert them into uppercase, so I must use:

 mb_strtolower($a,"utf8");

But what can I do, if I want to use ucfirst() function? I haven't found any similar function, where I can set encoding type.

Maybe this: http://www.if-not-true-then-false.com/2010/php-mb_ucfirst-make-a-strings-first-character-uppercase-multibyte-function/ — user956584, May 03 '12 at 20:48
P.S. There is a composer package with properly implemented mbUcFirst() https://github.com/cofirazak/phpMissingFunctions — cofirazak, Dec 09 '19 at 09:05

score 126 · Accepted Answer · edited Nov 12 '20 at 16:16

126

There is no mb_ucfirst function, as you've already noticed. You can fake a mb_ucfirst with two mb_substr:

function mb_ucfirst($string, $encoding)
{
    $firstChar = mb_substr($string, 0, 1, $encoding);
    $then = mb_substr($string, 1, null, $encoding);
    return mb_strtoupper($firstChar, $encoding) . $then;
}

edited Nov 12 '20 at 16:16

BenMorel

31,815
47
169
296

answered Mar 25 '10 at 17:43

zneak

130,082
41
248
315

1

how does this compares with the other users-entered customized functions? for example, http://php.net/manual/en/function.ucfirst.php#108856 **PS** I know I can TIAS but since lots of people have this problem someone may have the answer already – Felipe Sep 26 '12 at 00:07
7

you can actually pass NULL instead of $strlen-1 and then you don't need the first line. – ivanhoe Oct 19 '13 at 02:18
Really? That sounds a little weird. `null` as a third parameter to `mb_substr` means length minus one? – zneak Oct 19 '13 at 15:54
1

@zneak: ivanhoe is right. $strlen - 1 in this case means "up to the end of the string", which is what null does here. – Pavel V. Dec 19 '13 at 21:53
3

This didn't work for me. As per http://www.php.net/manual/de/function.mb-substr.php#77515: "Passing null as length will not make mb_substr use it's default, instead it will interpret it as 0." – sbichenko Feb 17 '14 at 14:52
4

@exizt that's because you're using an old PHP version. http://php.net/manual/en/function.mb-substr.php#refsect1-function.mb-substr-changelog – jurchiks Aug 24 '15 at 15:12
1

@zneak You could even make the `$encoding` parameter optional and read it from `mb_internal_encoding()` when it is not set so the function behavior would be consistent with other mb_ functions – Kyborek Aug 07 '18 at 09:55
$encoding should default to null and run mb_internal_encoding() if it is null. – hanshenrik Nov 18 '19 at 12:49
You can save an mbstring call by passing `null` instead of `$strlen - 1` in `mb_substr()`; this way, you don't need the length. **edit** oops, just saw that others mentioned this before I did. Actually, as I can see on your profile page that you're no longer contributing to StackOverflow, I took the liberty to update your answer to reflect this. – BenMorel Nov 12 '20 at 16:14

score 80 · Answer 2 · edited Feb 03 '18 at 08:25

80

This is more concise solution, although it is rather similar to ucwords function:

$final_string = mb_convert_case($your_string, MB_CASE_TITLE, 'UTF-8');

If you need to capitalize string consist of one word, it is the best solution.

edited Feb 03 '18 at 08:25

Melebius

5,590
3
34
46

answered Aug 22 '12 at 03:49

Alex Belyaev

1,277
1
11
14

2

Almost the best answer but **Koralek M. is more usefull** - it is **insensitive for change of encoding** of your www service – fider Jul 11 '13 at 09:41
1

@fider This one may also be insensitive for change of encoding if you omit 'UTF-8' parameter. As for every mb_* functions – Erdal G. Sep 29 '14 at 16:30
2

Works perfectly. For all intents and purposes this is the built-in `mb_ucfirst` equivalent in PHP. – Mahn Sep 16 '15 at 14:16
23

This does not perfectly as it changes first letters in all words to uppercase. Ucfirst is supposed to change only the first word. – emix Feb 26 '16 at 11:27
5

this will touch every word in the string, that's NOT what ucfirst() does. -1 – hanshenrik Nov 18 '19 at 12:53

score 28 · Answer 3 · edited Jun 20 '20 at 04:37

28

function mb_ucfirst($string)
{
    return mb_strtoupper(mb_substr($string, 0, 1)).mb_substr($string, 1);
}

edited Jun 20 '20 at 04:37

GKFX

1,366
1
11
29

answered Jan 04 '13 at 16:52

Koralek M.

3,053
4
25
31

22

You should not lowercase the rest of the string. – Haralan Dobrev Mar 10 '14 at 15:46
It is working ok. If it's not, check if mb_* functions do work in your environment at all. – cronfy Oct 05 '16 at 15:18
This function requires PHP/5.4.8 and greater. Also, you probably need to set a good locale and ensure that `mb_internal_encoding()` returns a proper value. – Álvaro González Jan 28 '17 at 16:41
1

this does not allow you to specify encoding – hanshenrik Nov 18 '19 at 12:54
.. and worse, it will lowercase everything else in the string, that's not what ucfirst() does... – hanshenrik Mar 04 '20 at 14:14
not lowercasing anymore - a good soul went ahead and fixed it... not sure if that's legit, though? Sort of made correct a wrong answer. – igorsantos07 Oct 28 '20 at 04:07
@igorsantos07 it's better now, but still not perfect, it does not allow you to specify encoding, unlike every other mb_ function ^^ – hanshenrik Oct 28 '20 at 09:00

score 9 · Answer 4 · answered Nov 18 '19 at 13:25

9

as of 2019-11-18, it seems nobody on stackoverflow got this right, here's how mb_ucfirst() should be implemented in userland:

function mb_ucfirst(string $str, string $encoding = null): string
{
    if ($encoding === null) {
        $encoding = mb_internal_encoding();
    }
    return mb_strtoupper(mb_substr($str, 0, 1, $encoding), $encoding) . mb_substr($str, 1, null, $encoding);
}

answered Nov 18 '19 at 13:25

hanshenrik

17,586
3
35
74

2

ps: if you're curious, i added a comment in every other answer, explaining what i think they got wrong. – hanshenrik Nov 19 '19 at 06:11
P.S. There is a composer package with properly implemented mbUcFirst() https://github.com/cofirazak/phpMissingFunctions – cofirazak Dec 09 '19 at 09:03

score 7 · Answer 5 · answered May 12 '15 at 05:49

7

if (!function_exists('mb_ucfirst'))
{
    function mb_ucfirst($value)
    {
        return mb_strtoupper(mb_substr($value, 0, 1)) . mb_substr($value, 1);
    }
}

answered May 12 '15 at 05:49

goyote

564
5
13

It is working ok. If it's not, check if mb_* functions do work in your environment at all. – cronfy Oct 05 '16 at 15:19
1

This function requires PHP/5.4.8 and greater. Also, you probably need to set a good locale and ensure that `mb_internal_encoding()` returns a proper value. – Álvaro González Jan 28 '17 at 16:40
1

this does not allow you to specify encoding =/ – hanshenrik Nov 18 '19 at 12:53

score 2 · Answer 6 · answered Apr 02 '12 at 12:56

I´m using cp1250 on webpage, and for Ú mb_ucfirst doesn´t work, so little upgrade:

  function mb_ucfirst($string)
{
    $main_encoding = "cp1250"; 
    $inner_encoding = "utf-8";
    $string = iconv($main_encoding, $inner_encoding , $string );
    $strlen = mb_strlen($string);
    $firstChar = mb_substr($string, 0, 1, $inner_encoding);
    $then = mb_substr($string, 1, $strlen - 1, $inner_encoding);
    return $string = iconv($inner_encoding, $main_encoding , mb_strtoupper($firstChar, $inner_encoding) . $then );
}

For more general use, $main_encoding should be either set as an optional parameter or by mb_internal_encoding(). Otherwise good. — Pavel V., Dec 19 '13 at 22:00

score 0 · Answer 7 · answered Feb 12 '14 at 15:15

0

/*This worked correctly for me*/
function mb_ucfirst($string, $encoding='UTF-8')
{
    $firstChar = mb_substr($string, 0, 1, $encoding);
    $then = mb_substr($string, 1, mb_strlen($string, $encoding)-1, $encoding);
    return mb_strtoupper($firstChar, $encoding) . $then;
}

answered Feb 12 '14 at 15:15

user3302248

9
1

This function requires PHP/5.4.8 and greater. Also, you probably need to set a good locale and ensure that `mb_internal_encoding()` returns a proper value. – Álvaro González Jan 28 '17 at 16:41
This advice was provided earlier. This code-only answer is redundant. – mickmackusa May 05 '19 at 15:22
1

default should be mb_internal_encoding() , not UTF-8 =/ – hanshenrik Nov 18 '19 at 12:54
there's no need to do `mb_strlen($string, $encoding)-1` - if you want the rest of the string, no matter how big it is, just specify the length as `null` – hanshenrik Aug 08 '20 at 18:42

score -1 · Answer 8 · edited May 21 '19 at 11:47

-1

$string = trim(preg_replace('/\s+/', ' ', $string));
$string_ar = explode(' ', mb_strtolower($string,'utf-8'));

foreach($string_ar as $key => $value {
  $string_str .= mb_convert_case(mb_substr(trim($value), 0, 1), MB_CASE_TITLE, 'utf-8')
    . mb_substr(trim($value),1)
    . ' ';
}

$string = trim($string_str);

edited May 21 '19 at 11:47

Stphane

3,276
5
31
41

answered May 21 '19 at 11:14

Paris Z

21
1

please elaborate more on your answer – Hichem BOUSSETTA May 21 '19 at 11:28
1

this will uppercase every first word after spaces, and then it will trim the result, thats NOT what ucfirst() does. -1 – hanshenrik Nov 19 '19 at 06:17

ucfirst() function for multibyte character encodings

8 Answers8

Linked

Related