replace (regular expressions) question

ChicoMendez

hi

i have a massive csv file in this format :

'age , (Jap.) Deep-fried. '

(spaces included, quotes not, see file attched)
i would like to change this to :

'age;(Jap.) Deep-fried.'


Does anyone know how i would do so (i want to remove the spaces also!)
Tnx

seamus

In php, a preg_replace would work with this regular expression:
/\s+,\s+/

That's assuming that all lines fit this format?

ChicoMendez

seamus wrote:

In php, a preg_replace would work with this regular expression:
/\s+,\s+/

That's assuming that all lines fit this format?

tnx .. anyway of doing it in notepad,dreamweaver or the like as i need to do a few thousand of them in one command....

Serbian

ChicoMendez wrote:

tnx .. anyway of doing it in notepad,dreamweaver or the like as i need to do a few thousand of them in one command....

Dreamweaver supports Find and Replace with Regular Expressions.

seamus

I've gotten some mixed results with Dreamweaver. Try it on a copy of the file first.

ChicoMendez

seamus wrote:

I've gotten some mixed results with Dreamweaver. Try it on a copy of the file first.

ya thats my issue really . DW - does nothing fo rthe command you gave me above

/\s+,\s+/

Serbian

You could try TextPad. It also supports Find and Replace via RegEx and is a very solid editor to boot.

seamus

ChicoMendez wrote:

ya thats my issue really . DW - does nothing fo rthe command you gave me above

/\s+,\s+/

Nah, that's a PHP regular expression. DW's regexp is similar though. Remove the forward slashes at the start and end of what I gave you and try it.

ChicoMendez

i may as well juts to it in php

i have no problem using a preg_replace

but how would i read in my current csv file and run through it line by line?


(i can then do the preg_replace and print the result on the screen and copy it into a new csv!)

Tnx

seamus

www.php.net is your friend.

Search for the fopen() function, then have a look at the submitted code examples under the function.

daymobrew

Using sed (Stream EDitor), a Unix command line utility, you could do it like:

sed -e 's/ *, */,/g' -e 's/ *$//' replace.txt >new.txt

The part removes spaces on either side of the comma, the second one removing any trailing spaces at the end of the line.

It didn't work on the file you attached because it appears to be in Unicode.

PM if anyone needs an explanation about what is going on in the command.