[SOLVED] Re: Replacing blank cells



I would like to thanks Steven Schwartz, Jean Bréfort, Morten Welinder and Uri David Akavla for their answers.

I finally find a (rather) quick solution to replace blank cells by NA.
Gnumeric wasn't able to convert the entire .txt into .csv file as I had more than 100 000 lines.
I then used the "gedit" program under Ubuntu 6.10 Edgy Eft.
I replaced tabulations (\t) by a coma (,) and then replace double coma (,,) by (,NA,) as suggested by Steven Schwartz.

Regards,

Ghislain Vieilledent

2007/8/15, Steve Schwartz <s schwartz imperial ac uk>:

On Wed, 2007-08-15 at 13:39 -0400, Uri David Akavia wrote:
> What about doing replace
> Value to replace: (just leave blank)
> Replace with: NA

This was my thought to, but gnumeric complains that "Search string must
not be empty" I tried various alternatives with regular expressions, but
without any success.

Actually, my first thought was to save the file as a csv text file, and
then open it in a text editor (or use sed) to replace:


^,
,,
,$

(i.e., a comma at the beginning of a line, a pair of commas with nothing
inbetween, and a comma at the end of a line) by

"NA",
,"NA",
,"NA"

respectively. This, of course, loses any formatting and formulae, but
should work.

HTH

Steve

--
+-------------------------------------------------------------------+
Professor Steven J Schwartz      Phone: +44-(0)20-7594-7660
Space and Atmospheric Physics    Fax:   +44-(0)20-7594-7772
The Blackett Laboratory          E-mail: s schwartz imperial ac uk
Imperial College London          Office: Huxley 6M70
London SW7 2BW, U.K.             Web: http://www.sp.ph.ic.ac.uk/~sjs
+-------------------------------------------------------------------+




--
Ghislain Vieilledent
06 24 62 65 07

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]