Total Hour Conversion to Date and Time

Hi,

I was having difficulties when originally trying to convert with datenum, so I manually converted all of my time values to hours(Figure 2). The format was originally in month/day/year hour:minute AM, as can be seen in Figure 1. I now have a slope value(seen in Fig 3) which I need to reconvert back to the original format; however, it signifies a time in the future. I dont know how to manually convert back as the day/month/year hour are all included in the total hour value, and I am having trouble distinguishing between a month or day. If there is an easy fix to convert back, I would appreciate the feedback.

Figure 1

Figure 2

Figure 3

[dpb]

As another case just answered on slightly different subject, let's go back and fix the original problem instead of trying to clean up a mess later.

Attach a short section of the input file with the original date strings so we don't have to make it up for ourselves...

[Felicia Lepore]

Good point.

Here is the attached input file with original date strings.

sorry! here is the code

[Star Strider]

Please also post the code to read the file and assign the variables you use.

Answer 1

[dpb]

t=readtable('hc07.xlsx');

leaves me with

>> t(1:5,:)
ans =
  5×21 table
            RECORDED            CH04_HC07_CZ_SD_Corrected_rel    CH04_HC07_3D_Movement_rel    CH04_HC07_Horizontal_Movement_rel    CH04_HC07_Elevation_Change_rel      Var6         hours       hourDiff     disDiff       du_dt       t_du_dt    Var12    Var13    Var14    Var15    Var16    Var17    Var18    Var19    Var20    Var21
    ________________________    _____________________________    _________________________    _________________________________    ______________________________    _________    __________    ________    _________    __________    _______    _____    _____    _____    _____    _____    _____    _____    _____    _____    _____
    8/2/2018 12:39:17.372 AM            0                               0                            0                                    NaN                                0    1.7683e+07    0                   0             0        0      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      ''   
    8/2/2018 2:27:57.046 AM     0.0030494                        0.081832                     0.080672                              -0.013735                        0.0030494    1.7683e+07    2           0.0030494     0.0015247    26974      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      ''   
    8/2/2018 4:41:25.817 AM     0.0047692                        0.053733                     0.053672                             -0.0025604                        0.0078186    1.7683e+07    2           0.0047692     0.0023846    42187      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      ''   
    8/2/2018 6:38:34.026 AM     0.0018315                         0.11069                      0.10569                               0.032892                        0.0096501    1.7683e+07    2           0.0018315    0.00091577    16201      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      ''   
    8/2/2018 8:36:15.693 AM     0.0032632                         0.10986                     0.095714                              -0.053936                         0.012913    1.7683e+07    2           0.0032632     0.0016316    28865      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      NaN      ''   
>> t.RECORDED(1)
ans = 
  datetime
   8/2/2018 12:39:17.372 AM
>>

readtable converted to datetime() on input for you...

If you used xlsread instead, the text dates were only to second precision...

>> [x,t,r]=xlsread('HC07.xlsx');
>> t(1:5,1)
ans =
  5×1 cell array
    {'RECORDED'            }
    {'8/2/2018 12:39:17 AM'}
    {'8/2/2018 2:27:57 AM' }
    {'8/2/2018 4:41:26 AM' }
    {'8/2/2018 6:38:34 AM' }
>> dt=datetime(t(2:5,1),'InputFormat','M/d/uuuu h:m:s a')
dt = 
  4×1 datetime array
   02-Aug-2018 00:39:17
   02-Aug-2018 02:27:57
   02-Aug-2018 04:41:26
   02-Aug-2018 06:38:34
>>

readtable read the Excel date and returned full precision; xlsread returned only the displayed text to seconds as text so the conversion has to match...

Either way, you end up w/ datetime() from the git-go...whether you need the full precision will predicate how you read the file or you'll have to add the extra precision to the display

Answer 2

[Felicia Lepore]

Thank you for all your help,

However, now there is a double array that is being divided by a duration. Is there a way to convert a duration array to a double?

[Guillaume]

Please don't use Answer this question to ask a new question.

It is trivial to convert a duration into a plain numeric type with the days, hours, minutes, seconds, etc. function depending on which unit you want for your numeric type:

>> d = duration('5:12:6:30'); %a duration of 5 days, 12 hours, 6 minutes, 30 seconds
>> years(d)  %convert to years
ans =
        0.0150708471464544
>> days(d)   %convert to days
ans =
          5.50451388888889
>> hours(d)  %convert to hours
ans =
          132.108333333333
>> minutes(d) %convert to minutes
ans =
                    7926.5
>> seconds(d) %convert to seconds
ans =
      475590