delete the repetetion letters in a string php - php

Question

I'm trying to parse a string and delete the adjacent letters that are same. I want to return the count of number of deletions and output the resulted string after the deletions are made. Say I have
$str = "aaabbbcc";
As you can see, you need to do 5 deletions to make the adjacent letters not same. The $output string is "abc" and the number of deletions is five.
function str_deletions ($str)
{
$prev = $str[0];
$length = strlen($str);
$deletions = 0;
$output = "";
for ($i=1 ; $i < $length; $i++)
{
if ($str[$i]== $prev)
{
$deletions++;
$prev = $str[$i]; // not sure here ?
}
}
echo $output; // ?
return $deletions;
}
$str = "aabbcc";
echo str_deletions ($str);
EDIT
This is an interview question, I'm not supposed to use any built-in functions like regex or array_count_values
Thanks for your help !

Answer 1

Here is another regex solution. I use a regex to only match a word character that is repeated, and then remove each consecutive repeating character one by one, which allows me to use &$count argument with preg_replace:
count
If specified, this variable will be filled with the number of replacements done.
The regex is
(\w)(?=\1)
See demo. Note you can replace \w with . to match any character but a newline. OR if you need to match only letters, I suggest using '/(\p{L})(?=\1)/u'
See IDEONE demo:
$str = "aaabbbcc";
$cnt = -1;
$result = preg_replace('/(\w)(?=\1)/', "", $str, -1, $cnt);
echo "Result: " . $result . PHP_EOL . "Deletions: " . $cnt;
Output:
Result: abc
Deletions: 5

Answer 2

Regex solution
This is a much simpler way of doing what you're after using preg_replace():
<?php
function str_deletions($str){
$replaced = preg_replace("/(.)\\1+/", "", $str);
$length = strlen($str) - strlen($replaced);
return array("new_word" => $replaced, "chars_replaced" => $length);
}
$str = "aabbcc";
$string_deletions = str_deletions($str);
echo "New String: " . $string_deletions['new_word'] . "\n";
echo "Chars deleted: " . $string_deletions['chars_replaced'] . "\n";
?>
No inbuilt functions
For the purposes of completion (and since you updated your question with more information to say that we can't use regexes because it's an interview question), here's what I'd do:
Using count_chars():
function str_deletions($str){
$string_data['new_word'] = count_chars($str,3);
$string_data['chars_replaced'] = strlen($str) - strlen($string_data['new_word']);
return $string_data;
}
$str = "aabbcc";
echo str_deletions($str);
Note: in this example count_chars(); will return unique chars in a string, not quite remove duplicates (i.e. "aabbccaa" would still yield "abc" as an output) but your question wasn't clear what the interviewer wanted - whether it was truly a remove duplicate question or a unique char question.
Using array_unique():
Slightly slower and a bit more heavy handed:
function str_deletions($str){
$string_array = array_unique(str_split($str));
foreach($string_array as $string_cur){
$string_data['new_word'] .= $string_cur;
}
$string_data['chars_replaced'] = strlen($str) - strlen($string_data['new_word']);
return $string_data;
}
$str = "aabbcc";
echo str_deletions($str);
Note: It's worth pointing out that if I realised it was an interview question, I wouldn't have provided an answer as doing it for you kind of defeats the purpose. Still, with the amount of answers here now and the fact that I've seen something similar to this in an interview, my hope is someone will learn from these.

Answer 3

The basic algorithm (indeed $prev = $str[$i]; isn't at the good place but you wasn't far):
function str_deletion($str) {
$del = 0;
if (1 < $length = strlen($str)) { // $str has more than 1 char
$prev = $str[0];
$output = $prev;
for ($i=1; $i<$length; $i++) {
if ($prev == $str[$i]) {
$del++;
} else {
$prev = $str[$i]; // when different, change the previous character
$output .= $prev; // and append it to the output
}
}
} else {
$output = $str;
}
echo $output;
return $del;
}

Answer 4

I have changed your function
this is not returning both the output string and number of deletions
function str_deletions ($str)
{
$prev = NULL;
$deletions = 0;
$output = "";
$i=0;
while ($i < strlen($str))
{
if (substr($str,$i,1) == $prev)
{
$deletions++;
//$prev = substr($str,$i,1);/*remove this line, no need here as the same stmnt is there after ifelse*/
}else{
$output.=substr($str,$i,1);
}
$prev = substr($str,$i,1);
$i++;
}
$arr = array(
'output'=>$output,
'deletions'=>$deletions
);
return $arr;
}
$str = "aaabbcc";
print_r(str_deletions ($str));
output for above function call is
Array ( [output] => abc [deletions] => 4 )

Answer 5

Solved with no external function except count;
$str="aaavvvffccca";
$count = strlen($str);
for($i=0;$i<$count;$i++){
$array[]=$str[$i];
}
$del =0;
for($i=0;$i<$count;$i++){
$next=isset($array[$i+1])?$array[$i+1]:null;
if($array[$i]==$next)
{
$del++;
}
else
{
$newarray[]=$array[$i];
}
}
echo "Filter Text:". implode('',$newarray);
echo"Total Deleted:".$del;

Answer 6

The straight forward solution to find out the number of deletions can be
If there are N consecutive same characters delete N-1 out of those N characters.
function str_likes($str)
{
$length = strlen($str);
$del = 0;
for ($i=0 ; $i < $length ; $i++)
{
if ($str[$i] == $str[$i+1])
{
$del++;
}
}
return $del;
}
$str = "aabbccaaa";
echo str_likes($str); //4