Sometimes, I have to use the function strpos() to find out if e.g. the string "Where is my bag?" contains "bag"
or anything similar...
Table of contents:
|
|
Definition: int strpos (string haystack, string needle [, int offset])
Now, see what strpos() does (starting with an example):
|
|
|
<?php
$MyString = "abcdefg";
echo strpos($MyString, "cde");
?>
|
This example finds the substr (needle) "cde" in the string $MyString.
The result is: 2 (a is $MyString[0], b is $MyString[1], c is $MyString[2])
show me
|
|
|
|
|
<?php
$MyString = "abcdefg";
echo strpos($MyString, "abc");
?>
|
And this finds "abc" in $MyString.
Result is: 0 ("abc" is at position 0 in $MyString)
show me
|
|
|
|
|
<?php
$MyString = "abcdefg";
echo strpos($MyString, "abc");
echo "-";
$res = strpos($MyString, "xyz");
echo $res;
echo "<br>";
if (isset($res)) echo "Variable is set, content of Variable: $res<br>";
if (empty($res)) echo "Variable is empty";
?>
|
Before v4.03b this example displayed 0 - 0. So, when a substr is found on
the first position of the string, it displayed 0 equally to the
case when the substr is not found.
Since v4.03b this will display "0 -", the second (search for "xyz") returns
empty (but the variable is set).
This change makes it besser to distinguish, whether a substr has been found or
not. We could e.g. ask if the variable $res was empty
if "0" would not be empty (check the empty-problem).
show me
The new workaround for this is:
|
|
|
<?php
$MyString = "abcdefg";
$pos = strpos ($MyString, "abc");
if ($pos === false) // 3 equal signs !!!
echo "not found!";
else
echo "found the substr!";
?>
|
Okay, this makes it definitly better. You can successfully distinguish the two cases.
show me
But what the heck does this "===" mean? Why do I have to create such an operator? To me, it seems that
this is just a dirty workaround.
=> The "===" means identical. That way you can check whether two variables are
identical (=of same type and same content).
|
|
|
<?php
if (5=="5") echo '5 equals "5"<br>'."\n";
if (5==="5") echo 'and is identical';
else echo 'but 5 is <b>not identical</b> with "5"';
?>
|
|
show me
|
|
|
Of course, this makes sense but what do you need it for in a
language that is NOT strong-typed?
At last, you do not need it here:
In C/C++ functions that do not find anything or produce an error, return a negative value!
Something like:
|
|
|
<?php
$MyString = "abcdefg";
$pos = strpos ($MyString, "xyz");
if ($pos < 0) // you can never have a string at a position < 0,
// so it has not been found
echo "not found!";
else
echo "found the substr!";
?>
|
would make it very very much better and easier and even more readable!
|
|
|
|
There's another reason, why the behaviour of strpos() is so bad. Let me show you
using a program in C-style:
|
|
|
<?c
int strpos(String haystack, String needle) // function returns an int
// forgetting offset and using object String
// for easier programming...
{
// ... look if I can find it
if (found)
return pos_where_found
else
return (boolean) false;
// see the typecast?
// It is unnecessary, but I used it to make
// clear what happens.
}
?>
|
Do you see what happens? The function either returns an integer or a
boolean in dependency of the input-values.
You can produce functions that return different types by overloading, using
different or a different number of parameters, but here it only depends on
what you use as values for the function-parameters.
e.g. the call strpos("abc", "bc") returns an integer, but strpos("abc", "xy")
returns a boolean. Have you ever seen anything alike? And if you have, have
you ever seen anything alike where it is so very easy to make things clearer?
|
|
|
I will never understand ... :(
Please check the section PHP types, too.
|
