PHP - Find/Replace Text in RTF/txt files - php

Question

I'm running into an issue with finding specific text and replacing it with alternative text.
I'm testing my code below with .rtf and .txt files only. I'm also ensuring the files are writable, from within my server.
It's a hit and miss situation, and I'm curious if my code is wrong, or if this is just weirdness of opening and manipulating files.
<?php
$filelocation = '/tmp/demo.txt';
$firstname = 'John';
$lastname = 'Smith';
$output = file_get_contents($filelocation);
$output = str_replace('[[FIRSTNAME]]', $firstname, $output);
$output = str_replace('[[LASTNAME]]', $lastname, $output);
$output = str_replace('[[TODAY]]', date('F j, Y'), $output);
// rewrite file
file_put_contents($filelocation, $output);
?>
So, inside the demo.txt file I have about a full page of text with [[FIRSTNAME]], [[LASTNAME]], and [[TODAY]] scattered around.
It's hit and miss with the find/replace. So far, [[TODAY]] is always replaced correctly, while the names aren't.
Has anyone had this same issue?
(on a side note, I've checked error logs and so far no PHP warning/error is returned from opening the file, nor writing it)

Answer 1

Hard to say for sure without seeing the contents of demo.txt. My first guess is that it might be a problem with using brackets for your pointers. I would try changing to something not used by RTF like percent signs or an asterisk. ex: %%FIRSTNAME%%, **FIRSTNAME** (this is assuming of course that you have control of the contents of demo.txt.)

Answer 2

I have had this issue too. It seems like Microsoft Word inserts formatting codes in the tags. I have made a blog post about how to get around this on my technical blog.
http://tech.humlesite.eu/2017/01/13/using-regular-expression-to-merge-database-content-into-rich-text-format-template-documents/
The PHP example is shown here:
<?php
$file = file_get_contents('mergedoc.rtf');
// To temporary get rid of the escape characters...
$mergetext = str_replace("\\", "€€", $file);
// New seven part regex with default value detection
$regex2 = '/<<((?:€€[a-z0-9]*|\}|\{|\s)*)([a-z0-9.\-\+_æøåÆØÅA-Z]*)((?:€€[a-z0-9]*|\}|\{|\s)*)([a-z0-9.\-\+_æøåÆØÅA-Z]*)((?:€€[a-z0-9]*|\}|\{|\s)*)(?:\s*:(.*?)\s*)?((?:€€[a-z0-9]*|\}|\{|\s)*)>>/';
// Find all the matches in it....
preg_match_all($regex2,$mergetext, $out, PREG_SET_ORDER);
// Lets see the result
var_dump($out);
foreach ($out as $match) {
$whole_tag = $match[0]; // The part we actually replace.
$start = $match[1]; // The start formatting that has been injected in our tag, if any
$tag = $match[2]; // The tag word itself.
if (($match[4].$match[6]) != "") { //some sec-part tag or default value?
$end = $match[5]; // The end formatting that might be inserted.
if ($end == "") {
$end = $match[7]; // No end in 5, we try 7.
}
} else {
$end = $match[3]; // No second tag or default value, we find end in match-3
}
$secPartTag = $match[4]; // Do we have inserted some formatting inside the tag word too ?
if ($secPartTag != "") {
$tag .= $secPartTag; // Put it together with the tag word.
}
$default_value = $match[6];
// Simple selection of what we do with the tag.
switch ($tag) {
case 'COMPANY_NAME':
$txt = "MY MERGE COMPANY EXAMPLE LTD";
break;
case 'SOMEOTHERTAG':
$txt = "SOME OTHER TEXT XX";
break;
case 'THISHASDEFAULT':
$txt = "";
break;
default:
$txt = "NOTAG";
}
if ($txt == "") {
$txt = $default_value;
}
// Create RTF Line breaks in text, if any.
$txt = str_replace(chr(10), chr(10)."\\line", $txt);
// Do the replace in the file.
$mergetext = str_replace($whole_tag, $start.$txt.$end, $mergetext);
}
// Put back the escape characters.
$file = str_replace("€€", "\\", $mergetext);
// Save to file. Extention .doc makes it open in Word by default.
file_put_contents("ResultDoc.doc", $file);
?>