How to seperate column correctly using readtable?

3 views (last 30 days)
JamJan
JamJan on 3 Jun 2021
Edited: Stephen on 10 Jun 2021
Hi,
I have been using readtable for quite some while, but for some reason I started running into problems a few days ago. Attached find an example of a .csv file. I would like to split the values from:
FLIP001,111111.1234567890,"2001-10-29 01:01:00","2001-10-29 01:01:00",7
to a table consisting of:
Var1; Var2; Var3; Var4; Var5
FLIP001; 111111.1234567890; "2001-10-29 01:01:00"; "2001-10-29 01:01:00"; 7
(Semicolons are seperating the columns here)
For some reason I do not manage to use the delimiter correctly. Can someone help me to split these variables correctly using, for instance, readtable or any other way?
Thank you!
  2 Comments
Stephen
Stephen on 10 Jun 2021
Original question by JamJam retrieved from Bing Cache (the CSV file could not be retrieved):
How to seperate column correctly using readtable?
I have been using readtable for quite some while, but for some reason I started running into problems a few days ago. Attached find an example of a .csv file. I would like to split the values from:
FLIP001,111111.1234567890,"2001-10-29 01:01:00","2001-10-29 01:01:00",7
to a table consisting of:
Var1; Var2; Var3; Var4; Var5
FLIP001; 111111.1234567890; "2001-10-29 01:01:00"; "2001-10-29 01:01:00"; 7
(Semicolons are seperating the columns here)
For some reason I do not manage to use the delimiter correctly. Can someone help me to split these variables correctly using, for instance, readtable or any other way?
Thank you!

Sign in to comment.

Accepted Answer

Star Strider
Star Strider on 3 Jun 2021
This appears to work —
C1 = readcell('https://www.mathworks.com/matlabcentral/answers/uploaded_files/640960/FakeCsv.csv');
CS1 = cellfun(@(x)strsplit(x,','), C1, 'Unif',0);
cellrows = cellfun(@(x)size(x,2),CS1);
CS2 = reshape([CS1{:}], cellrows(1), []).';
T1 = cell2table(CS2,'VariableNames',compose('Var%d',1:size(CS2,2)))
T1 = 5×5 table
Var1 Var2 Var3 Var4 Var5 ___________ _____________________ _________________________ _________________________ _______ {'FLIP001'} {'111111.1234567890'} {'"2001-10-29 01:01:00"'} {'"2001-10-29 01:01:00"'} {'7' } {'FLIP001'} {'111111.1234567890'} {'"2001-10-29 01:01:00"'} {'"2001-10-29 01:01:00"'} {'8' } {'FLIP001'} {'111111.1234567890'} {'"2001-10-29 01:01:00"'} {'"2001-10-29 01:01:00"'} {'16' } {'FLIP001'} {'111111.1234567890'} {'"2001-10-29 01:01:00"'} {'"2001-10-29 01:01:00"'} {'128'} {'FLIP001'} {'111111.1234567890'} {'"2001-10-29 01:01:00"'} {'"2001-10-29 01:01:00"'} {'7' }
Completing this with the conversions:
T1.Var2 = str2double(T1.Var2);
T1.Var3 = datetime(T1.Var3, 'InputFormat','"yyyy-MM-dd HH:mm:ss"');
T1.Var4 = datetime(T1.Var4, 'InputFormat','"yyyy-MM-dd HH:mm:ss"', 'InputFormat','"yyyy-MM-dd HH:mm:ss"');
T1.Var5 = str2double(T1.Var5)
T1 = 5×5 table
Var1 Var2 Var3 Var4 Var5 ___________ __________ ____________________ ____________________ ____ {'FLIP001'} 1.1111e+05 29-Oct-2001 01:01:00 29-Oct-2001 01:01:00 7 {'FLIP001'} 1.1111e+05 29-Oct-2001 01:01:00 29-Oct-2001 01:01:00 8 {'FLIP001'} 1.1111e+05 29-Oct-2001 01:01:00 29-Oct-2001 01:01:00 16 {'FLIP001'} 1.1111e+05 29-Oct-2001 01:01:00 29-Oct-2001 01:01:00 128 {'FLIP001'} 1.1111e+05 29-Oct-2001 01:01:00 29-Oct-2001 01:01:00 7
There does not appear to be a more straightforward way to parse that. The detectImportOptions options may work, however I have found those to be difficult to get right. so I went for a more direct approach.
.

More Answers (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by