Trying to break apart date string

Question

Cameron Power 2018-6-29

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/408099-trying-to-break-apart-date-string

评论： Cameron Power 2018-6-29

I have a table with a column that has a large string for the date, I want to isolate certain digits and separate them for example if I have date = 2.016010100000000e+11 I want yyyy = date(1:4) MM = date(5:6) dd = date(7:8) HH = date(9:12)

The only problem is that I want to do this for the entire table, is there a way to? Thank you

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Stephen23 2018-6-29

编辑：Stephen23 2018-6-29

What you have appears to be a numeric scalar, so that indexing is unrelated. How did that numeric end up in the table in the first place? Probably the easiest solution would be to fix the importing of the file...

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Stephen23 2018-6-29

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/408099-trying-to-break-apart-date-string#answer_326890

在 MATLAB Online 中打开

>> val = 2.016010100000000e+11;
>> str = sprintf('%u',val)
str = 201601010000
>> vec = datevec(str,'yyyymmddHHMM')
vec =
   2016      1      1      0      0      0

3 个评论
显示 1更早的评论隐藏 1更早的评论

Stephen23 2018-6-29

编辑：Stephen23 2018-6-29

在 MATLAB Online 中打开

"when it was recorded no delimeters were used in the dating,..."

That doesn't matter, we can handle that!

"...is there a way to implement this to fix every date in the table(every date is in this exact format)?"

One option would be to import the data properly to start with, using either readtable or textscan. Both of these have formats for reading dates, which is explained in their help, so I will not bore you with that here.

As an alternative you could simply import as numeric (e.g. using csvread), convert the first column to a character matrix, and then use datevec:

>> mat = csvread('aaab-dbase-2016-04-21.csv');
>> chr = num2str(mat(:,1));
>> chr(1:10,:)
ans =
201511180800
201511180810
201511180820
201511180830
201511180840
201511180850
201511180900
201511180910
201511180920
201511180930
>> dtv = datevec(chr,'yyyymmddHHMM');
>> dtv(1:10,:)
ans =
   2015     11     18      8      0      0
   2015     11     18      8     10      0
   2015     11     18      8     20      0
   2015     11     18      8     30      0
   2015     11     18      8     40      0
   2015     11     18      8     50      0
   2015     11     18      9      0      0
   2015     11     18      9     10      0
   2015     11     18      9     20      0
   2015     11     18      9     30      0