读取行并更改不包含某些单词但不以点结尾的行

I wanna read some text files in a folder line by line. for example of 1 txt :

Fast and Effective Text Mining Using Linear-time Document Clustering
Bjornar Larsen WORD2 Chinatsu Aone
SRA International AK, Inc.
4300 Fair Lakes Cow-l Fairfax, VA 22033

{bjornar-larsen, WORD1

I wanna remove line that does not contain of words = word, word2, word3, and does not end with dot .

so. from the example, the result will be :

Bjornar Larsen WORD2 Chinatsu Aone
SRA International, Inc.
{bjornar-larsen, WORD1

I am confused, hw to remove the line? it that possible? or can we replace them with a space?

here's the code :

$url = glob($savePath.'*.txt');
foreach ($url as $file => $files) {
    $handle = fopen($files, "r") or die ('can not open file');
    $ori_content= file_get_contents($files);
    foreach(preg_split("/((?
)|(
?))/", $ori_content) as $buffer){
        $pos1 = stripos($buffer, $word1);
        $pos2 = stripos($buffer, $word2);
        $pos3 = stripos($buffer, $word3);
        $last = $str[strlen($buffer)-1];//read the las character
        if (true !== $pos1 OR true !== $pos2 OR true !==$pos3 && $last != '.'){
        //how to remove
        }
    }
}

please help me, thank you so much :)

You're using a !== true comparison to test the return-value of the stripos. !== true means "is not absolutely equal-to the boolean value true". The return-value of stripos is numeric, unless the word doesn't exist, in which case it's false. In other words, your condition is always false.

Try updating it to use === false instead. Also, you're using OR in between each; Your example shows that it needs to only contain 1 of them - so if you're checking that "none of them were found", you'll need to use && for everything:

if (($pos1 === false) && ($pos2 === false) && ($pos3 === false) && ($last != '.'))

Regarding "how to remove the line", you'll need to keep a list of all lines you want to keep. This means, we'll actually want to flip the condition above to use !== false and an || between everything (because we want to keep all lines that match any rule).

Try something like this:

$url = glob($savePath.'*.txt');
foreach ($url as $file => $files) {
    $handle = fopen($files, "r") or die ('can not open file');
    $ori_content= file_get_contents($files);
    $linesToKeep = array(); // list of all lines that match our rules
    foreach(preg_split("/((?
)|(
?))/", $ori_content) as $buffer){
        $pos1 = stripos($buffer, $word1);
        $pos2 = stripos($buffer, $word2);
        $pos3 = stripos($buffer, $word3);
        $last = $str[strlen($buffer)-1];

        if (($pos1 !== false) || ($pos2 !== false) || ($pos3 !== false) || ($last == '.')) {
            $linesToKeep[] = $buffer; // save this line
        }
    }
    // process list of lines for this file;
    // file_put_contents($files, join("
", $linesToKeep)); // write back to file
    // $lines = join("
", $linesToKeep); // convert to string to manipulate
}

Now, you'll have every line that matches your ruleset in the $linesToKeep array. You can convert this back to a string with $lines = join(" ", $linesToKeep);, or iterate through it and process it however you'd like.

You'll need to create a secondary buffer.

$url = glob($savePath.'*.txt');
foreach ($url as $file => $files) {
    $handle = fopen($files, "r") or die ('can not open file');
    $ori_content= file_get_contents($files);
    /* Create our second buffer */
    $buffer2 = "";
    foreach(preg_split("/((?
)|(
?))/", $ori_content) as $buffer){
        $pos1 = stripos($buffer, $word1);
        $pos2 = stripos($buffer, $word2);
        $pos3 = stripos($buffer, $word3);
        $last = $str[strlen($buffer)-1];//read the last character
        /* This will only execute if the three words and a trailing period are _not_ found */
        if ($pos1 === false && $pos2 === false && $pos3 === false && $last != '.') {
            $buffer2 .= $buffer . PHP_EOL;
        }
    }
}
echo $buffer2;

Nice approach... But you can use arrays to read in your file and put it your file. Till now it is fine.

PS: There can be better ways to do...

$url = glob($savePath.'*.txt');
foreach ($url as $file => $files) {
    $handle = fopen($files, "r") or die ('can not open file');
    $ori_content= file_get_contents($files);

    # Declare a variable array to store the contents.
    $fileContents = array();

    foreach(preg_split("/((?
)|(
?))/", $ori_content) as $buffer){
        $pos1 = stripos($buffer, $word1);
        $pos2 = stripos($buffer, $word2);
        $pos3 = stripos($buffer, $word3);
        $last = $str[strlen($buffer)-1];//read the las character
        if (($pos1 !== false) || ($pos2 !== false) || ($pos3 !== false) || ($last == '.')){
            $fileContents[] = $buffer;
        }
    }

    # Put the contents
    file_put_contents($file, implode(PHP_EOL, $fileContents);

}

I would just use explode:

$handle = fopen($files, "r") or die ('can not open file');
$ori_content = file_get_contents($files);

$lines = explode ( '
' , $ori_content );

foreach ( $lines AS $line )
{
 if (strpos ( $line , 'word' ) !== false OR strpos ( $line , 'word2' ) !== false OR strpos ( $line , 'word3' ) !== false OR substr ( $line , -1 ) == '.')
  {
   $newParagraph = $line . '
';
  }
}

echo $newParagraph;

Much simpler than what you were trying to do.

Try

$url = glob($savePath.'*.txt');
foreach ($url as $file => $files) {
  $lines = file($files);
  foreach ($lines as $key=>$line) {
    if (!preg_match('/(word|word2|word3)/i', $line) && substr($line, -1) != '.') {
      unset($lines[$key]);
    }
  }
  $ori_content = implode("
", $lines);
}