ReadTable producing different result for 2017 and 2020A matlab version

Hi,

I have created a script which is reading data from excel in matlab 2017v.It's reading whole excel file as it is ,without adding "" for missing value ,not omitiitng any row and setting variable name = VAR1,VAR2....etc

file_name = 'TestCase.xlsx';
excel_sheet = readtable(file_name,'Sheet',1,'ReadVariableNames',false);

Now i have tried running same script in the matlab2020v,it's reading excel in a different way than 2017v.It's omitting row and reading first line as VarName.

file_name = 'TestCase.xlsx';
excel_sheet = readtable(file_name,'Sheet',1,'ReadVariableNames',false);

Update:Added Image

I want MATLAB2020A read excel in the format as it's read by 2017A

[Star Strider]

The file appears to be made up of several blocks of data.

file_name = 'https://www.mathworks.com/matlabcentral/answers/uploaded_files/982945/TestCase.xlsx';
cell_sheet = readcell(file_name)

cell_sheet = 35×8 cell array
    {'<TV>TestCase1:1'}    {1×1 missing                }    {1×1 missing                }    {1×1 missing             }    {1×1 missing                }    {1×1 missing     }    {1×1 missing           }    {1×1 missing      }
    {'Time'           }    {'Inlet_Volatge_Sensor_Diag'}    {'Inlet_Volatge_Sensor_volt'}    {'Charging_Started_Latch'}    {'Normal_Charging_Shutdown1'}    {'VCharg_Act1'   }    {'CCS_EV_Inlet_voltage'}    {'Sensor Validity'}
    {'single'         }    {'boolean'                  }    {'uint8'                    }    {'boolean'               }    {'boolean'                  }    {'fixdt(1,32,16)'}    {'fixdt(1,32,16)'      }    {'uint8'          }
    {[              0]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   0]}    {[              1]}
    {[              1]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   0]}    {[              1]}
    {[              2]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   0]}    {[              1]}
    {[              3]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   0]}    {[              1]}
    {[              4]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   0]}    {[              1]}
    {1×1 missing      }    {1×1 missing                }    {1×1 missing                }    {1×1 missing             }    {1×1 missing                }    {1×1 missing     }    {1×1 missing           }    {1×1 missing      }
    {'<TV>TestCase1:2'}    {1×1 missing                }    {1×1 missing                }    {1×1 missing             }    {1×1 missing                }    {1×1 missing     }    {1×1 missing           }    {1×1 missing      }
    {'Time'           }    {'Inlet_Volatge_Sensor_Diag'}    {'Inlet_Volatge_Sensor_volt'}    {'Charging_Started_Latch'}    {'Normal_Charging_Shutdown1'}    {'VCharg_Act1'   }    {'CCS_EV_Inlet_voltage'}    {'Sensor Validity'}
    {'single'         }    {'boolean'                  }    {'uint8'                    }    {'boolean'               }    {'boolean'                  }    {'fixdt(1,32,16)'}    {'fixdt(1,32,16)'      }    {'uint8'          }
    {[              0]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   1]}    {[              1]}
    {[              1]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   1]}    {[              1]}
    {[              2]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   1]}    {[              1]}
    {[              3]}    {[                        0]}    {[                        1]}    {[                     0]}    {[                        0]}    {[             0]}    {[                   1]}    {[              1]}

What do you want to do with it?

[Lalit Kumar]

I wanted to read whole file in the MATLAB2020 using "readTable" without omitting any information ,after that i know how to process data to convert them into TestCase.

My intend is to run same .m script in both 2017 and 2020 version.

readcell is not available in mat2017B

[Jan]

Of course you can implement branching:

if verLessThan('matlab', '9.8')
    excel_sheet = readtable(file_name,'Sheet',1,'ReadVariableNames',false);
else
    excel_sheet = readcell(file_name);
end

The general problem remains: Matlab versions are not perfectly compatible. The need to run a code in different Matlab versions requires to find workarounds.

In the lab I'm working in, we still run a virtual machine unter WindowsXP, because the reference implementation of a code for clinical decision making was developped in Matlab 6.5 (2002). A validation of the results under Matlab R2009a took 2 month, because the outcome of 1000 patients have to be compared. It took me 1 year to implement an automatic comparison of the results, such that the migration to Matlab 2018b needed 48 hours only. During the tests a bunch of incompatibilities have been found, e.g. in strncmp, if the inputs are empty and n=0, or fopen() in VAXD format.

Do not try to write a "one-code fits all Matlab versions" program, because this letexplode the complexity of maintaining the code. Rely on unit-test: Import a reference file in R2017a, save the result as a MAT file. Insert a branching as shown above and let it import the Excel file again. Now compare the results with the contents of the MAT file. If this is equivalen, it is a strong hint, that your import is working. Afterwards run these tests whenever you change the Matlab version.

My software for clinical decision making includes 100'000 lines of dull test codes only to validate the output of the subfunctions.

[dpb]

I'd probably go back a step and use an import object that matches the expected file structure and form in which want the file to be interpreted. This then should produce the desired consistent results.

I don't know what your 2017 imported file looked like; be good to attach it as a .mat file, but R2020b didn't seem to leave anything out and didn't read variable names...

[Lalit Kumar]

@dpb added image for better understanding

Answer 1

[Prasanna Konyala]

Hi @Lalit Kumar

From my understanding, you want to get the whole data in excel to be read in the same format as 2017a.

From 2020a, "readtable()" functionality is a bit different than earlier versions. It can detect data types, discard extra header lines, and fill in missing values. In the file attached, it is discarding extra header lines at the start and reading each column as numeric data. Hence, unknown or missing data is replaced by NaN.

If you want the functionality as earlier versions, you can add ('Format','auto') name-value pair argument to the readtable function in 2020a.

file_name = 'TestCase.xlsx';
excel_sheet = readtable(file_name,'Sheet',1,'ReadVariableNames',false,'Format','auto');

Please refer this document for more information about "readtable()"