I use Dreamweaver as my editor, mainly because I'm familiar with it, although I 
only use
about 1% of its capabilities. However it generally handles long files well.  
The other day
I downloaded the two shortest of Brian Dunning's sets of test data *.  I opened 
shortest in Dreamweaver, had a quick look at it, and realised I would have to 
replace the
quote, comma, quote separators with semicolons, as part of converting the files 
to my

So I thought I would do that while I was working out what else I had to do. I 
entered the
old separator and the replacement in the 'Find and replace' window, then hit 
all', expecting the job be done in a few seconds. First I thought nothing was 
but then I realised it was trudging through the file, as if it was wading 
through waist
high molasses.

So I closed the results window, and opened another file, but a few seconds 
later focus
switched back to the original file. I tried a couple of times more, but each 
time it
returned to the original window. I watched in morbid fascination for a bit, 
then decided I
would let it go, just to see how long it took.

The file contained 500 lines, and was about 80 K. It was taking five seconds to 
each line, and eventually finished in about 40 minutes.

The problem appeared to be the results processing. I have only looked at the 
results list
about twice, out of idle curiosity, but never saw any thing that I thought 
could be
remotely useful. I would like to be able to turn results logging off 
altogether, as it
wastes real estate (and time!), but this appears to be impossible.  

On this occasion the program was apparently writing a new line every time it 
replaced a
separator (9 times in each line), and then when it finished processing a line 
it would
erase all the intermediate result lines, and write a new one for the whole 
line.  At the
same time it reopened the results window if I had closed it, and return focus 
to the file
being processed.

I then wrote a PHP program to read the file, split it, clean up and re-arrange 
the various
elements, enter them into an array in my format, and finally save it as a file 
my program
could handle.

 After I had got this running on the 500 line file I used it to process the 
5000 line
file. The whole process was done in the blink of an eye -- literally a fraction 
of a

* http://www.briandunning.com/sample-data/

PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php

Reply via email to