Howdy helpful folks...

I have a large amount of data I need to convert and I'm hoping to build a regex or two to help convert something like this:

I love-love-love what [User:5138532:dgoldstein] and his [Tag:738533:erongo] kids have done with our [Tag:800657:connected] blog!  (You all should go and leave lots of comments for them!) http://countryconnection.edublogs.org/
    My guys will do some catching up once our exams are out of the way!
    (ping [User:1392073:lisaroberson], [User:1546074:jeanesmith1])

to this:

I love-love-love what {user: 14321} and his #erongo kids have done with our #connected blog!  (You all should go and leave lots of comments for them!) http://countryconnection.edublogs.org/
    My guys will do some catching up once our exams are out of the way!
    (ping {user: 2332}, {user: 12332}) 

the number that follows user: is going to be the result of a query to lookup the old id and replace it with the new one.

I'm trying to use preg_replace_callback() in php, with limited success.  

Any help would be very much appreciated.  You've saved my bacon twice before...THANKS!

I don't have access to a platform that lets me use the PHP callback function, but this may get you on the right path.

Separating the 2 types of tag, I have used a pattern of:

\[tag:[^:]*:([^\]]*)\]

and a replacement string of

#$1

with the "ignore case" option set to locate and replace the "Tag:" items; and the pattern of:

\[user:(\d+)[^\]]*\]

and the replacement string of:

{User:$1}

(again with the "ignore case" option set) to do the same thing with the "user" items - in this case I simply put back the same number to make sure that the rest was working.

Of course, you can combine these patterns into:

\[tag:[^:]*:([^\]]*)\]|\[user:(\d+)[^\]]*\]

(these can be factored but that doesn't change the rest of the processing). Now, if something is captured in match group #1, then you know that this was a "tag" entry and can construct the suitable replacement string in the callback. Similarly, if characters are captured in match group #2, then you can use the digits to perform your lookup and again build the required replacement string.

The callback function code then becomes a "simple matter of programming" which can be "left as an exercise for the reader".

Hope this helps

Susan

PS: The resulting string from the first replacement is:

I love-love-love what [User:5138532:dgoldstein] and his #erongo kids have done with our #connected blog!  (You all should go and leave lots of comments for them!) http://countryconnection.edublogs.org/
    My guys will do some catching up once our exams are out of the way!
    (ping [User:1392073:lisaroberson], [User:1546074:jeanesmith1])

and the second replacement is:

I love-love-love what {User:5138532} and his [Tag:738533:erongo] kids have done with our [Tag:800657:connected] blog!  (You all should go and leave lots of comments for them!) http://countryconnection.edublogs.org/
    My guys will do some catching up once our exams are out of the way!
    (ping {User:1392073}, {User:1546074})

<?php
  $str = 'I love-love-love what [User:5138532:dgoldstein] and his [Tag:738533:erongo] kids have done with our [Tag:800657:connected] blog!  (You all should go and leave lots of comments for them!) http://countryconnection.edublogs.org/ 
    My guys will do some catching up once our exams are out of the way! 
    (ping [User:1392073:lisaroberson], [User:1546074:jeanesmith1])';
  $str = preg_replace_callback('/\[([^:]+):([^:]+):([^\]]+)\]/', 'mycallback', $str);
  echo $str;
  function mycallback($input)
  {
      //here you would use $input[1] to access the first capture group etc...
  }
?>