MATLAB：删除每 14 行重复的 4 行标头，并在 txt 文件中保留第 2/3 列数据

Question

基本上，我在一堆文本文件中有一堆重复的数据，看起来如下（显示了两个部分）：

t0  1/20/2012  05:41:05.068750
delta t 0.100000
time    Y
1/20/2012  05:41:05.068750  0.925883
1/20/2012  05:41:05.168750  0.926540
1/20/2012  05:41:05.268750  0.926540
1/20/2012  05:41:05.368750  0.926540
1/20/2012  05:41:05.468750  0.926869
1/20/2012  05:41:05.568750  0.925225
1/20/2012  05:41:05.668750  0.926540
1/20/2012  05:41:05.768750  0.928185
1/20/2012  05:41:05.868750  0.925554
1/20/2012  05:41:05.968750  0.925225
t0  1/20/2012  05:41:06.068750
delta t 0.100000
time    Y
1/20/2012  05:41:06.068750  0.926212
1/20/2012  05:41:06.168750  0.924567
1/20/2012  05:41:06.268750  0.926540
1/20/2012  05:41:06.368750  0.925883
1/20/2012  05:41:06.468750  0.914371
1/20/2012  05:41:06.568750  0.907135
1/20/2012  05:41:06.668750  0.906806
1/20/2012  05:41:06.768750  0.903188
1/20/2012  05:41:06.868750  0.902201
1/20/2012  05:41:06.968750  0.906148

我需要在MATLAB中找到一种方法，只保留第二列和第三列中的10行数据（时间，没有日期和y值），并将其全部存储在一个变量中，这样我就可以将其写入一个新的文本或xls文件。我甚至不需要时间值，因为每个y值都是以0.1秒为增量的。任何帮助都将不胜感激。

Answer 1

批处理：

% Slurp in all lines.
f = fopen('foo.txt');
c = textscan(f,'%s','Delimiter','n');
lines = c{1};
fclose(f);
% Remove headers.
lines(1:14:end) = [];
lines(1:13:end) = [];
lines(1:12:end) = [];
lines(1:11:end) = [];
% Extract data.
output = zeros(numel(lines),2);
for i = 1:numel(lines)
    x = lines{i};
    output(i,1) = datenum(x(1:19),'mm/dd/yyyy  HH:MM:SS') + ...
        str2double(x(20:26));
    output(i,2) = str2double(x(28:end));
end

或者作为一个状态机：

f = fopen('foo.txt');
output = [];
while true
    for i = 1:4
        x = fgetl(f);
        if x == -1
            break
        end
    end
    for i = 1:10
        x = fgetl(f);
        if x == -1
            break
        end
        time = datenum(x(1:19),'mm/dd/yyyy  HH:MM:SS') + ...
            str2double(x(20:26));
        val = str2double(x(28:end));
        output(end+1,1:2) = [time,val];
    end
    if x == -1
        break
    end
end
fclose(f);