Combine time related columns into timestamp for data in csv

Question

Ant376 2021-9-7

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1448439-combine-time-related-columns-into-timestamp-for-data-in-csv

评论： Ant376 2021-9-7

Hi, I have a number of csv files where I would like change the way date and time data are recorded so that it is timestamped.

Time and data are currently stored as below:

year,month,day,hour,minute,second,data1,data2,data3

2009,1,1,0,0,0,2.9,12.1,7.1

I would like the CSV to read:

2009-01-01 00:00:00, 2.9, 12.1,7.1

I have tried the following script, which I appreciate has a few inefficiencies:

for y = 1:5 
    yStr = num2str(y); 
    
        
    C = csvread([folderPath,'c',yStr,'.csv']);
    t = string(datestr(datetime(C(:,1),C(:,2),C(:,3),C(:,4),C(:,5),C(:,6)), 'yyyy-mm-dd HH:MM:SS'));
    D = [t,C(:,7),C(:,8),C(:,9)];
    
    csvfilename = 'd'+string(y)+'.csv';
    csvwrite(csvfilename,D);
   
end

However the output I get has each individual letter in the string seperated by a comma.

How do I get the script to record the timestamp in my preferred format? The data stored in the variable 't' looks correct when viewed

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Dave B 2021-9-7

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1448439-combine-time-related-columns-into-timestamp-for-data-in-csv#answer_782629

编辑：Dave B 2021-9-7

在 MATLAB Online 中打开

The datestr is perhaps adding confusion?

This page has some good information on how to format datetimes: https://www.mathworks.com/help/matlab/matlab_prog/set-display-format-of-date-and-time-arrays.html

%% Generate some data that looks like your C
somerandomdates = datetime('now') - days(rand(100,1)*10);
C = [somerandomdates.Year somerandomdates.Month somerandomdates.Day somerandomdates.Hour somerandomdates.Minute somerandomdates.Second];
%% convert back to datetime, set the format, convert to string
dt = datetime(C); % Or you could use C(:,1:6) if you have more columns in your matrix
dt.Format = 'yyyy-MM-dd hh:mm:ss'; % note that MM is month, mm is minute
string(dt)
ans = 100×1 string array
    "2021-09-05 04:52:05"
    "2021-08-29 12:29:16"
    "2021-08-31 09:44:02"
    "2021-09-02 02:55:05"
    "2021-08-29 05:57:31"
    "2021-09-05 04:09:57"
    "2021-08-30 04:52:39"
    "2021-08-28 07:53:39"
    "2021-09-03 02:28:25"
    "2021-09-02 07:51:07"
    "2021-08-31 11:15:13"
    "2021-08-29 07:52:29"
    "2021-08-30 02:04:30"
    "2021-08-30 03:50:26"
    "2021-09-06 02:32:11"
    "2021-08-31 06:28:24"
    "2021-09-01 04:57:40"
    "2021-09-01 11:04:14"
    "2021-09-05 09:00:17"
    "2021-09-07 04:41:36"
    "2021-09-02 09:40:03"
    "2021-09-04 02:09:25"
    "2021-09-03 07:26:21"
    "2021-08-30 06:32:53"
    "2021-09-01 04:19:05"
    "2021-08-29 12:43:07"
    "2021-09-06 06:31:58"
    "2021-08-29 12:55:13"
    "2021-08-30 01:19:35"
    "2021-08-29 05:53:59"

I'd also recommend considering writematrix instead of csvwrite, or you might consider writecell/writetable for your heterogenous data. If you load the data with readtable then writetable will work particularly well.

writematrix([string(dt) string(rand(size(dt)))], 'foo.csv')

3 个评论
显示 1更早的评论隐藏 1更早的评论

Dave B 2021-9-7

No worries, if you have writematrix then by all means use it instead of csvwrite, it's better in every way. It sounds like this resolves your issue, but if you're still struggling or need to use csvwrite for some reason please let me know.

If you can use readtable/writetable instead, you may find it's even easier - tables work naturally with different types of variables, so you don't have to pretend that the values are all strings.

Ant376 2021-9-7