How can I parse this textfile with textscan? Delimiter \t not working

I'd like to parse this static_vehicle_characteristics_svc.svc file into three groups of blocks (General, Front Sus, Rear Sus) and split the group of blocks into 5 headers (Parameter, Units, Average, Left, Right) and store data into each header.

Here's my code, I was trying to make this as a function. (input : variable name, suspension type / output : data of input variable)

If I put input as function('General Characteristcs', 'Total weight'), I want this function to return all three values of total, left, right stored in structure or something like that so I could access to use the values in other script. I will make this into a function on my own if someone helps me parsing data.

clc
clear all

file_name = 'Static_Vehicle_Characteristics_svc.txt'; % 
str = fileread(file_name);

% Regular Expression to split strings into general, fr, rr character blocks
xpr = '[A-Z].+?(CHARACTERISTICS).+?(\r\n){3}';
blocks = regexp(str, xpr, 'match');
num_blocks = length(blocks);

% For each block I'd like to parse text into variable name, units, total,
% left, right
for i = (2:num_blocks)
    block(i-1).data = textscan(blocks{i}, '%s%s%s%s%s', 'Delimiter', '\t', 'HeaderLine', 6);
%     block(i-1).data = regexp(blocks{i}, '[A-Z].+?\r\n', 'Match'); %breaks into lines
%     block(i-1).data(cellfun(@isempty, block(i-1).data)) = []; %remove empty lines
%     block(i-1).each_line = textscan([block(i-1).data{:}], '%s', 'Delimiter', ' ');
%     block(i-1).each_line(cellfun(@isempty, block(i-1).each_line)) = [];
end

At first I thought, each block would be split into cell array of 5 columns if I use delimiter \t. But the result was not what I expected.

The output I got just splits each block into lines.

I tried to split each line with whitespaces, but it returns so many lines and doesn't look smart.

What did I do wrong and how can I fix this?

Answer 1

[Stephen23]

"At first I thought, each block would be split into cell array of 5 columns if I use delimiter \t. But the result was not what I expected."

That text file does not contain one single tab character, so it is unclear to me how you thought that specifying a non-existent delimiter character would help to import that file. The file is basically fixed-width, not delimited....

But unfortunately that file is not quite a fixed-width file (e.g. see lines 33, 44), but we might be able to define a broad enough specification anyway, that can import those numeric values correctly:

fnm = 'Static_Vehicle_Characteristics_svc.txt';
opt = detectImportOptions(fnm, 'FileType','fixedwidth', 'ReadVariableNames',false,...
    'Range',22, 'VariableWidths',[34,11,11,11,99]);
opt.VariableNames = {'PARAMETER','UNITS','TOTAL','LEFT','RIGHT'};
tbl = readtable(fnm, opt)

tbl = 158×5 table
            PARAMETER               UNITS          TOTAL      LEFT    RIGHT
    _________________________    ____________    _________    ____    _____

    {'Total weight'         }    {'N'       }        10000     NaN     NaN 
    {'Front ground reaction'}    {'N'       }         1000    1000    1000 
    {'Rear  ground reaction'}    {'N'       }         1000    1000    1000 
    {'Total roll  inertia'  }    {'kg mm**2'}    1.006e+08     NaN     NaN 
    {'Total pitch inertia'  }    {'kg mm**2'}        1e+09     NaN     NaN 
    {'Total yaw   inertia'  }    {'kg mm**2'}        1e+09     NaN     NaN 
    {'Total product Ixy'    }    {'kg mm**2'}       2000.4     NaN     NaN 
    {'Total product Ixz'    }    {'kg mm**2'}        1e+06     NaN     NaN 
    {'Total product Iyz'    }    {'kg mm**2'}        -1000     NaN     NaN 
    {'Sprung mass'          }    {'kg'      }         1000     NaN     NaN 
    {'Sprung roll  inertia' }    {'kg mm**2'}    1.004e+08     NaN     NaN 
    {'Sprung pitch inertia' }    {'kg mm**2'}        1e+09     NaN     NaN 
    {'Sprung yaw   inertia' }    {'kg mm**2'}        1e+09     NaN     NaN 
    {'Sprung product Ixy'   }    {'kg mm**2'}         1000     NaN     NaN 
    {'Sprung product Ixz'   }    {'kg mm**2'}       -1e+06     NaN     NaN 
    {'Sprung product Iyz'   }    {'kg mm**2'}        -1000     NaN     NaN 

Obviously you need to check that all values are imported as expected, adjust the import settings (e.g. variable widths) as required. After that you will then need some post-processing to remove some of the invalid data lines, identify the blocks, etc.

[Munho Noh]

Thanks a lot. It just looked like it had a delimiter. I will try this out.