Re: Textread from io package

[Daniel Arteaga]

Dear PetrSt,

Al 10/04/13 14:18, En/na PetrSt ha escrit:
> Daniel Arteaga-2 wrote
> > [Trial Event_Type Code
> > Time]=textread('test.log','%d%s%s%d%*[^\n]','headerlines',5);
> >
> > In Octave 3.6.2 raises an error (both tested in Ubuntu 12.10 and Windows)
>
> 1) What type of error? First error I got was related to absence of number of
> repeating. Despite the manual to textread it seems that this input argument
> is not optional, so textread(file,format,[],options), or similar, must be
> specified.

I noticed this first error. That there is a bug whenever you only 
specify 'headerlines' and not any other option.

> 2) Textread doesn't return single argument. Equal number of outputs to
> format specifier must be given instead.

I think that was a problem of automatic line wrapping in the email 
client. I meant:

Trial Event_Type Code  Time]= ...
textread('test.log','%d%s%s%d%*[^\n]','headerlines',5);

> 3) It doesn't work anyway. I tried also textscan with similar result, i.e. I
> got the output but the parsing was incorrect. Even adding 0\t to several
> first lines to got the same formating for those lines didn't help.

Same happened to me

> 4) This code worked just fine:
>
> str = fileread ('test.log');
> [s, e, te, m, t, nm] = regexp (str','(\n(\S*)\t(\S*)\t(\S*)\t(\S*)\t)');
> nl = size(t,2);
> for ii=1:nl
>         a(ii,1) = str2num(t{ii}{2});
>         b{ii,1} = t{ii}{3};
>         c{ii,1} = t{ii}{4};
>         d(ii,1) = str2num(t{ii}{5});
> end

Thank you very much! That piece of code worked fine for me.

Best,

Daniel

> Maybe the transformation of cell t to desired outputs could be done more
> effectively.
>
>
>
> --
> View this message in context: 
> http://octave.1599824.n4.nabble.com/Textread-from-io-package-tp4651648p4651682.html
> Sent from the Octave - General mailing list archive at Nabble.com.