I can't get millisecond accuracy from string based timestamp - not sure what I'm doing wrong

I have a load of data that's timestamped that I'm looking to manipulate using mainly durations. Issue is that the data is in 3 digit millisecond precisions and nothing less precise will do and everytime I try to convert my string cells to timestamps, it round everything to the nearest second and adds the current date to the front. I don't actually need the date at all.

This is my raw input, its a cell array of timestamps.

My code:

> infmt = 'HH:mm:ss.SSS';

> formatted_timestamps = datetime(raw_timestamps, 'InputFormat', infmt);

My output:

I have tried appending the values with less millisecond digits with zeros to ensure they're all the same length as strings, but all that did was remove the NaT values in output. I have spent hours trying to figure it out but all the forums and support docs say that what I have done is correct and as no on e else appears to have this issue I'm assuming its perhaps something more fundamental that I'm doing wrong?

Thanks in advance!

Answer 1

[Ameer Hamza]

Also set the format property to display them in the format you want

formatted_timestamps = datetime(raw_timestamps, 'InputFormat', infmt, 'Format', infmt);

[Josh Eadie]

Immediately worked, you're a saint

[Ameer Hamza]

I am glad to be of help! :)

Answer 2

[Steven Lord]

If you don't have date information associated with your timestamps, I would use duration instead of datetime.

S = ["11:00:00.147"; "11:00:00.668"; "11:00:00.932"; "11:00:04"]

S = 4×1 string array
    "11:00:00.147"
    "11:00:00.668"
    "11:00:00.932"
    "11:00:04"

fmt = 'hh:mm:ss.SSS';
d = duration(S, 'InputFormat', fmt)

d = 4×1 duration array
   11:00:00
   11:00:00
   11:00:00
   11:00:04

d2 = duration(S, 'InputFormat', fmt, 'Format', fmt)

d2 = 4×1 duration array
   11:00:00.147
   11:00:00.668
   11:00:00.932
   11:00:04.000