Loading every n-th column of a csv with varying formats in row using textscan

Hi, i am trying to load a csv file with semicolon (;) delimiter.

Example:

150501;190722;ms_since_start=;30001276;temp=;31.97;IT=;147753;spec num=;1000;(here i have 512 floating number repetitions and ';;' to indicate the end of line)

this pattern repeats for 1000 rows.

I have been trying to use textscan but only get empty cells with the following code

formatSpec = ['%s%s%*s%*s%*s%*s%*s%*s%*s%*s%*s' repmat('%f', [1,512]) '%*[^;;]']
M = textscan(dirtmp, formatSpec, 'Delimiter', ';')

The goal is to get the first 2 columns, skip 9, get the remaining 512 columns and repeat this for 1000 rows.

Any help is highly appreciated

Answer 1

[Star Strider]

You may need to change the last part of your ‘formatSpec’ line deleting the '%*[^;;]' and including 'MultipleDelimsAsOne',1 and 'EndOfLine','\r\n' or perhaps something else if you have a different end-of-line indicator.

Also consider this as the first part of your ‘formatSpec’:

'%f;%f;ms_since_start=;%f;temp=;%f;IT=;%f;spec num=;%f'

No promises these will solve your problems, but I would experiment with those and other options until I got it working.