match a date

[Renlok]

im trying to make a script which pulls data from a file and im trying to format it

 

and example string would be

20-11-2009, 14:21:00:: Unknown error type: [8] Undefined index: SESSION_ERROR on /home/me/public_html/error.php line 19

and i want to change it to

<b>20-11-2009, 14:21:00::</b> Unknown error type: [8] Undefined index: SESSION_ERROR on /home/me/public_html/error.php line 19

 

the code i have is

$data = preg_replace('/(0-9){2}-(0-9){2}-(0-9){4} (0-9){2}:(0-9){2}:(0-9){2}::/', '<b>\1-\2-\3 \4:\5:\6::</b>', $data);

but it doesnt work anyone have any ideas?

[JAY6390]

Use this

$result = preg_replace('/(\d{2}-\d{2}-\d{4}, \d{2}:\d{2}:\d{2}::)/s', '<b>$1</b>', $subject);

Note you should really be using <strong> tags rather than <b> tags

[cags]

To expand on why your original code doesn't work Renlok. You encase the 0-9 with normal brackets, this creates a capture group, the problem is your capture group looks for the literal 0 followed by dash followed by 9 because you have not made it a character class by also encasing them in square brackets. ([0-9]{2}) would have been the correct syntax to match 2 single digits. Another reason your pattern doesn't match is because you don't include the comma in your pattern, which is present in the input. As JAY6390 has pointed out, since you are basically just encasing the whole thing in a pair of tags, you might as well capture it all as one. Also bare in mind that \d is not strictly speaking always the same as [0-9].

[nozai]

If your date is always there at the beginning of the string, and ends with 2 colons, you could just:

 

$result = preg_replace('/^(.*?::)/s', '<b>$1</b>', $subject);

[cags]

If it's always there and always in the same format you could always use substr or similar to add in the tags, there's a fair chance it would be quicker. Point is there's dozens of different methods to use and probably 100s of different Regex patterns. For example you could probably do something along the lines of...

 

'#^(.{22})#'

 

But since the OP says the data is being pulled from another file I'd imagine it's coming from the middle of a string.

[nrg_alpha]

Use this

$result = preg_replace('/(\d{2}-\d{2}-\d{4}, \d{2}:\d{2}:\d{2}::)/s', '<b>$1</b>', $subject);

 

Just note that you don't really need that capture, as the entire pattern is already represented by $0. Also, you are using the s modifier, but is not necessary as your pattern doesn't rely on a dot_match_all (.) with the need to include newlines.

 

As cags mentions, there are numerous ways to accomplish tasks.. here is yet another alternative:

 

$result = preg_replace('#(?:\d{2}-){2}\d{4}, (?:\d{2}{3}:#', '<b>$0</b>', $subject);