Writing to CSV file help

Question

Rodriguez Pham 2015-9-16

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/243459-writing-to-csv-file-help

关闭： MATLAB Answer Bot 2021-8-20

So I have a CSV file with a list of variables with various assignments, so for example: *note these are each in on cell each

a = 5464,,,,,,,
b = 657578,0.46454,0,85976
c = 46,,,,,, 
d = 45345
e = 0,5,6,8,,,,,,,

I want to filter the csv file so that it removes the ,,,,,, and then reformat this so that it will then become

a = 5464; b = 657578,0.46454,0,85976; c = 46; d = 45345; e = 0,5,6,8;

in once cell

I have the idea to use a regexp to filter the commas but I dont know how to filter it specifically.

2 个评论
显示无隐藏无

Walter Roberson 2015-9-17

When you say that they are in one cell each, do you mean one line each? Or do you mean that in the CSV file each of these expressions was enclosed in double-quotes "" to mark that the commas did not indicate the end of the value?

Rodriguez Pham 2015-9-17

编辑：Rodriguez Pham 2015-9-17

So when I load the csv file into the matlab workspace using uigetfile, then fopen, then textscan, the variable are in a cell type of a 5x1 vector. So when I apply my opperation the result I want is a 1x1 with all of the varibles in that specfic format where it follows the 5 variables in a line.

The goal for me is to understand the syntax to do this operation because I will have lets say a 100x1 with variables assigned but having this comma issue and I want to apply my operation so that it will end up as a 20x1 with 5 variables in each row but 'one cell' for each row which I will then write to a text file. Does that make sense?

此问题已关闭。

Answer 1

Walter Roberson 2015-9-17

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/243459-writing-to-csv-file-help#answer_192686

在 MATLAB Online 中打开

data = {'a = 5464,,,,,,,'; 'b = 657578,0.46454,0,85976'; 'c = 46,,,,,, '; 'd = 45345'; 'e = 0,5,6,8,,,,,,,'}
trimmed_data = regexprep(data, ',+\s*$', '');
joined_data = [strjoin(trimmed_data(:).', '; '), ';'];