Time and date fields not correclty imported using readmatrix

42 views (last 30 days)
I am trying to import csv and dat files that have columns with date and time. Here is a snapshot of csv file:
And dat file:
I am using "readmatrix" method for import. However, upon import, the date and time fields are converted to NaN. Is there a way to read these files while preserving time format? I intend to do some timestamp based filtering and plotting, so the resloution of timestamp is important.

Accepted Answer

Cris LaPierre
Cris LaPierre on 4 Apr 2021
Edited: Cris LaPierre on 4 Apr 2021
You cannot mix data types when you use readmatrix, meaning all the data you import must be of the same data type. That is why you are getting NaN and not NaT.
Use readtable or readtimetable instead. In a table, each column can be a different data type.
  4 Comments
Cris LaPierre
Cris LaPierre on 4 Apr 2021
You would have to use a second read command to capture just row 3. Perhaps textscan? I'm not feeling motivated enough to transcribe your image for testing, so here's a best guess at what that might be.
fid = fopen('filename.dat');
% Skip the first string, then read in time as a duration
tm = textscan(fid,'%*s %{hh:mm:ss.SSS}T',1,'HeaderLines',2)
fclose(fid);
tm will be a cell array. You can extract the time value from the cell array like this.
tm=tm{:}
Then use your readtable command to get the matrix of data.
With all your dates as datetimes or durations, you can perform Date and Time arithmetic.
Rakshith Badarinath
Rakshith Badarinath on 4 Apr 2021
Thanks. I ended up using fopen with fegetl and regexp to extract the time cell array and converted it to duration. textscan seems compact, will give it a try.

Sign in to comment.

More Answers (0)

Products


Release

R2020b

Community Treasure Hunt

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

Start Hunting!