[PHP Function] Make Microsoft Word documents understandable

[TimeBomb]

I just finished up project for a client of mine that involved, amongst other things migrating a blog from one area to another.

Without going into too much detail, one problem I had was that a lot of their blog entries seemed to have been copy pasted from Microsoft Word.
This was even more problematic when migrating them, as the blog system I was migrating to displayed the Microsoft word characters (ex. curly quotation marks, AKA smart quotes), as messy blocky symbols - like it didn't know what symbol it was.
This had to do with the character encoding and what not, but it came down to that I needed the posts to use normal characters. " instead of ” and ' instead of ’.

There were also some very mysterious spaces that I could just not get rid of. I had to deal with those too.

Overall, this problem was frustrating for me. It cost me 2 hours of research and testing. I wanted to release this function to perhaps help anyone who ever gets into this problematic scenario.

function fixMessyString($string) // Works exceptionally well with screwy Microsoft Word strings
    $find[] = '“';  // left side double smart quote
    $find[] = '”';  // right side double smart quote
    $find[] = '‘';  // left side single smart quote
    $find[] = '’';  // right side single smart quote
    $find[] = '…';  // elipsis
    $find[] = '—';  // em dash
    $find[] = '–';  // en dash
    $find[] = html_entity_decode('<p>&nbsp;</p>'); // Messy lines at the end
    $find[] = chr(160); // Mysterious space
    $find[] = '  '; // We fixed the mysterious space above, so now we will make sure there are no double spaces

    $replace[] = '"';
    $replace[] = '"';
    $replace[] = "'";
    $replace[] = "'";
    $replace[] = "...";
    $replace[] = "-";
    $replace[] = "-";
    $replace[] = "";
    $replace[] = " ";
    $replace[] = "  ";
    
    $string = str_replace($find,  $replace, $string);
    $string = str_replace(utf8_encode('Â'), '', utf8_encode($string)); // Finish removing odd double spaces
    
    return $string;
} 

The second str_replace is what I found to be necessary to completely remove all traces of the mysterious space, which is apparently chr(160) (in PHP, anyways). I could not figure out any other way to completely kill it off.

If anyone has any comments or suggestions as to how I could better deal with this scenario in future cases, then please do speak up.

G'luck all.

[MentaL]

[George SS]

did you try

string trim ( string $str [, string $charlist ] ) 

This function returns a string with whitespace stripped from the beginning and end of str. Without the second parameter, trim() will strip these characters:

" " (ASCII 32 (0x20)), an ordinary space.
"\t" (ASCII 9 (0x09)), a tab.
"\n" (ASCII 10 (0x0A)), a new line (line feed).
"\r" (ASCII 13 (0x0D)), a carriage return.
"\0" (ASCII 0 (0x00)), the NUL-byte.
"\x0B" (ASCII 11 (0x0B)), a vertical tab.

[TimeBomb]

did you try

string trim ( string $str [, string $charlist ] ) 

This function returns a string with whitespace stripped from the beginning and end of str. Without the second parameter, trim() will strip these characters:

" " (ASCII 32 (0x20)), an ordinary space.
"\t" (ASCII 9 (0x09)), a tab.
"\n" (ASCII 10 (0x0A)), a new line (line feed).
"\r" (ASCII 13 (0x0D)), a carriage return.
"\0" (ASCII 0 (0x00)), the NUL-byte.
"\x0B" (ASCII 11 (0x0B)), a vertical tab.

Heh, yes. That was one of the first few things I tried. It did absolutely nothing ;).

[George SS]

How about

str_replace(" ","",$string); 

??

[holthelper]

How about

str_replace(" ","",$string); 

??

look closer at his function