regular expression inquiry

[vito]

I have a text file (see below please) to parse with the following code. i
don't know why it cannot match anything. thanks.


$data=fread(fop en("$dir/$file","r"),fil esize("$dir/file"));

$matchno = preg_match_all( "|.*:(.+)\n$|U" , $data, $reg);

for ($i=1; $i<$matchno; $i++) {
$line = ",'".trim($ regs[1][$i]."'");
fwrite($fp, $line);
}


GROUP: CARS R/C SERIES

PRODUCT
NAME: R/C CAR

PACKAGE: WINDOW BOX

SIZE:30.5¡Á12¡Á 10.6 cm

MEAS:84* 34*93 cm

PCS/CTN: 48 pcs

GROSS
WEIGHT: NIL

NET
WEIGHT: NIL

INNER
BOX QTY: NIL

[Palle Hansen]

Re: regular expression inquiry

vito wrote:[color=blue]
> I have a text file (see below please) to parse with the following code. i
> don't know why it cannot match anything. thanks.
>
>
> $matchno = preg_match_all( "|.*:(.+)\n$|U" , $data, $reg);[/color]

In single line mode $ will match the end of the entire text. This mode
is default. |Um will switch to multi line mode

In single line only \n is needed
In multi line only $ is needed

[vito]

Re: regular expression inquiry

>> $matchno = preg_match_all( "|.*:(.+)\n$|U" , $data, $reg);[color=blue]
>
> In single line mode $ will match the end of the entire text. This mode is
> default. |Um will switch to multi line mode
>
> In single line only \n is needed
> In multi line only $ is needed[/color]

For irregularity of files (created by users), i need to first make the file
content into a single line

preg_replace("/\n/", "", $data);

and then perform the matching.

however, i found the replacment fails. maybe that is due to microsoft
newline ^M instead of \n? if so, how to represent ^M in the pattern
matching? furthermore, the newline assumption may be wrong and how can i
know what the characters are there in the white spaces? I cannot use all
space characters since the data also contain spaces.

Thanks a lot.