Trouble with textscan
in [Matlab]
|
From: Muhlbauer on 20 Apr 2010 15:29 Hi, I have troubles reading a line in a text file using textscan. Here is the example: UTC Launch Time (y,m,d,h,m,s): 2008, 10, 28, 03:27:29 I tried several things but the problem is that MATLAB gets confused by the "," delimiters. I tried for example: tmp=textscan(fid,'%s %s %s %s %f %f %f %f'); which correctly reads all the stuff including 2008 but stops after that due to the ,. I only need the 2008 10 and 28 information. How can I do this correctly? Modifying the text line is not an option for me.
From: Walter Roberson on 20 Apr 2010 15:37 Muhlbauer wrote: > Hi, > I have troubles reading a line in a text file using textscan. > Here is the example: > > UTC Launch Time (y,m,d,h,m,s): 2008, 10, 28, 03:27:29 > > I tried several things but the problem is that MATLAB gets confused by > the "," delimiters. > > I tried for example: > > tmp=textscan(fid,'%s %s %s %s %f %f %f %f'); > > which correctly reads all the stuff including 2008 but stops after that > due to the ,. > I only need the 2008 10 and 28 information. > How can I do this correctly? Modifying the text line is not an option > for me. >> textscan('UTC Launch Time (y,m,d,h,m,s): 2008, 10, 28, 03:27:29', '%*[^:]: %f %f %f%*s', 'Delimiter', ' ,', 'MultipleDelimsAsOne', 1) ans = [2008] [10] [28]
From: dpb on 20 Apr 2010 15:36 Muhlbauer wrote: > Hi, > I have troubles reading a line in a text file using textscan. > Here is the example: > > UTC Launch Time (y,m,d,h,m,s): 2008, 10, 28, 03:27:29 > .... > I only need the 2008 10 and 28 information. .... One way assuming the text doesn't change (which doesn't look like it should)... tmp=textscan(fid,'UTC Launch Time (y,m,d,h,m,s): %d %d %*[^\n]'); I don't have textscan() in this version to test, but think that should do... --
From: Muhlbauer on 20 Apr 2010 15:59 dpb <none(a)non.net> wrote in message <hqkvrf$ufj$1(a)news.eternal-september.org>... > Muhlbauer wrote: > > Hi, > > I have troubles reading a line in a text file using textscan. > > Here is the example: > > > > UTC Launch Time (y,m,d,h,m,s): 2008, 10, 28, 03:27:29 > > > ... > > I only need the 2008 10 and 28 information. > ... > > One way assuming the text doesn't change (which doesn't look like it > should)... > > tmp=textscan(fid,'UTC Launch Time (y,m,d,h,m,s): %d %d %*[^\n]'); > > I don't have textscan() in this version to test, but think that should do... > > -- Unfortunately, your suggestion doesn't solve the problem. It ignores the text before 2008 and reads 2008 correctly but not what comes after that. The "," delimiter is screwing things up...
From: dpb on 20 Apr 2010 20:08 Muhlbauer wrote: > dpb <none(a)non.net> wrote in message > <hqkvrf$ufj$1(a)news.eternal-september.org>... >> Muhlbauer wrote: .... >> > > UTC Launch Time (y,m,d,h,m,s): 2008, 10, 28, 03:27:29 >> > ... >> > I only need the 2008 10 and 28 information. >> ... >> >> One way assuming the text doesn't change (which doesn't look like it >> should)... >> >> tmp=textscan(fid,'UTC Launch Time (y,m,d,h,m,s): %d %d %*[^\n]'); >> .... > Unfortunately, your suggestion doesn't solve the problem. It ignores the > text before 2008 and reads 2008 correctly but not what comes after that. > The "," delimiter is screwing things up... Oh, sorry, I forgotted them... tmp=textscan(fid,'UTC Launch Time (y,m,d,h,m,s): %d, %d, %*[^\n]'); altho I'd look at Walter's suggestion; it's a little slicker in both finding the ":" as well as using the ' ,' as delimiters w/ the "multiple as one" option. Either should work. Here's one I can do that doesn't rely on textscan... >> s='UTC Launch Time (y,m,d,h,m,s): 2008, 10, 28, 03:27:29'; >> sscanf(s, 'UTC Launch Time (y,m,d,h,m,s): %d, %d, %*[^\n]') ans = 2008 10 >> --
|
Pages: 1 Prev: How to get elements in a specific order from matrix Next: TreeBagger missing values |