deleting part of a list of strings

 I have a column of strings, with the same number of element in each string, all comma delimited. I would like to trim all the strings after the last coordinate to have something from this:

'$SDDBT,00006.5,f,0002.0,M,0001.0,F*06,$GPGLL,5340.90756479,N,00713.79236750,E,073424.00,A,D*61' '$SDDBT,00006.8,f,0002.1,M,0001.1,F*0B,$GPGLL,5340.90609055,N,00713.79168827,E,073425.00,A,D*66' '$SDDBT,00007.8,f,0002.4,M,0001.3,F*0D,$GPGLL,5340.90462662,N,00713.79106230,E,073426.00,A,D*6C' To this:

'$SDDBT,00006.5,f,0002.0,M,0001.0,F*06,$GPGLL,5340.90756479,N,00713.79236750,E' '$SDDBT,00006.8,f,0002.1,M,0001.1,F*0B,$GPGLL,5340.90609055,N,00713.79168827,E' '$SDDBT,00007.8,f,0002.4,M,0001.3,F*0D,$GPGLL,5340.90462662,N,00713.79106230,E' I appreciate the help

Answer 1

[David Sanchez]

Since in you GPS data you want to trim right after the "E", you can try this:

str='$SDDBT,00006.5,f,0002.0,M,0001.0,F*06,$GPGLL,5340.90756479,N,00713.79236750,E,073424.00,A,D*61' ;

E_position = find(str=='E');
new_str = srt(1:E_position);

do it for each string.

Answer 2

[Thorsten]

newstr = str(:, 1:findstr(str(1,:), 'E'))

[Thorsten]

If the position of the 'E' may be different for each line, use

for i=1:size(str, 1)
 newstr(i,:) = str(:, 1:findstr(str(i,:), 'E'));
end

Answer 3

[mb1400]

E_position = find(out2=='E'); ??? Undefined function or method 'eq' for input arguments of type 'cell'.

newstr = out2(:, 1:findstr(out2(1,:), 'E')) ??? Error using ==> findstr Inputs must be character arrays.

not working in both ways

arrrrrghhhhhh.

And also. By the way. I do I extract the final result into a text file?

Answer 4

[mb1400]

The second way it did not really cut after E, that's what it returned

 newstr = out3(:, 1:findstr(out3(1,:), 'E'))

newstr =

$SDDBT,00004.9,f,0001.5,M,0000.8,F*07,$GPGLL,5340.91664087,N,00713.79587546,E $SDDBT,00004.5,f,0001.4,M,0000.7,F*05004.5,f,0001.4,M,0000.7,F*02,$GPGLL,5340 $SDDBT,00006.5,f,0002.0,M,0001.0,F*06,$GPGLL,5340.90756479,N,00713.79236750,E $SDDBT,00006.8,f,0002.1,M,0001.1,F*0B,$GPGLL,5340.90609055,N,00713.79168827,E $SDDBT,00007.8,f,0002.4,M,0001.3,F*0D,$GPGLL,5340.90462662,N,00713.79106230,E

[Thorsten]

Please see my revision above using a for loop to determine the correct position of 'E' for each line. (In your sample data in the question the 'E' appeared always at the same position.)

Answer 5

[David Sanchez]

do you have a cell array containing your strings?

If my_cell is your cell array, try:

str = char(my_cell{k});

for the k-th cell. Then:

E_position = find(str=='E');
new_str = srt(1:E_position);

Or:

E_position = findstr(str,'E')
new_str = srt(1:E_position);

Answer 6

[Matt Tearle]

Assuming you have an n-by-1 cell array of strings to start with (and you'd like an n-by-1 cell arrays of strings at the end), here are a couple of ways you can do it:

oldstr = {'$SDDBT,00006.5,f,0002.0,M,0001.0,F*06,$GPGLL,5340.90756479,N,00713.79236750,E,073424.00,A,D*61';'$SDDBT,00006.8,f,0002.1,M,0001.1,F*0B,$GPGLL,5340.90609055,N,00713.79168827,E,073425.00,A,D*66';'$SDDBT,00007.8,f,0002.4,M,0001.3,F*0D,$GPGLL,5340.90462662,N,00713.79106230,E,073426.00,A,D*6C'};

bits = regexp(oldstr,',','split');
f = @(x) strjoin(x(1:12),',');
newstr = cellfun(f,bits,'uniform',false)

This uses strjoin which is a fairly recent function. Alternatively,

oldstr = {'$SDDBT,00006.5,f,0002.0,M,0001.0,F*06,$GPGLL,5340.90756479,N,00713.79236750,E,073424.00,A,D*61';'$SDDBT,00006.8,f,0002.1,M,0001.1,F*0B,$GPGLL,5340.90609055,N,00713.79168827,E,073425.00,A,D*66';'$SDDBT,00007.8,f,0002.4,M,0001.3,F*0D,$GPGLL,5340.90462662,N,00713.79106230,E,073426.00,A,D*6C'};

bits = textscan(sprintf('%s\n',oldstr{:}),...
    [repmat('%s',1,12),repmat('%*s',1,3)],...
    'delimiter',',','collectoutput',true);
tmp = bits{1}';
newstr = regexp(deblank(sprintf([repmat('%s,',1,11),'%s\n'],tmp{:})),'\n','split')'

Answer 7

[mb1400]

I will soon try Matt's advice and get back to you. To Thorsten the problem is that the string is created from a combination of 2 strings. The original file looked something like this.

$SDDBT,00004.9,f,0001.5,M,0000.8,F*07 $SDDBS,00004.9,f,0001.5,M,0000.8,F*00 $VWVHW,,T,,M,00.0,N,,K*4A $SDRMB,A,13.74,L,00,TEST1,5330.546,N,00807.276,E,033.4,108,,*45 $GPGLL,5340.91664087,N,00713.79587546,E,073418.00,A,D*62 $GPHDT,9.421,T*3B $GPVTG,193.65,T,,M,5.82,N,10.77,K,D*0E $GPVTG,192.06,T,,M,5.54,N,10.26,K,D*05 $GPHDT,9.428,T*32 $GPGLL,5340.91360433,N,00713.79468132,E,073420.00,A,D*64 $GPHDT,9.430,T*3B .35,K,D*0F $GPHDT,9.431,T*3A $SDDPT,0001.4,000.0*62 $SDDBT,00004.5,f,0001.4,M,0000.7,F*05004.5,f,0001.4,M,0000.7,F*02 $VWVHW,,T,,M,00.0,N,,K*4A $SDRMB,A,13.74,L,00,TEST1,5330.546,N,00807.276,E,033.4,108,,*45 $GPGLL,5340.91207731,N,00713.79413391,E,073421.00,A,D*63 $GPHDT,9.433,T*38 $GPVTG,193.32,T,,M,5.69,N,10.54,K,D*08 $GPHDT,9.435,T*3E $GPGLL,5340.91056222,N,00713.79354123,E,073422.00,A,D*6E $SDDPT,0002.0,000.0*65 $SDDBT,00006.5,f,0002.0,M,0001.0,F*06 $SDDBS,00006.5,f,0002.0,M,0001.0,F*01 $VWVHW,,T,,M,00.0,N,,K*4A $SDRMB,A,13.75,L,00,TEST1,5330.546,N,00807.276,E,033.4,108,,*44 $GPHDT,9.437,T*3C

I was helped to write a script that with a loop would extract the SDDPT and GPGLL strings, it was a complicated process since strings have not always the same occurence, all I wanted was the first SDDPT and the successive associated GPGLL, all in one line. So, this is what we did.

 for j=1:length(data);
        if strncmp('$SDDBT',data(j,1), 6)==1;
            %             nuova_variabile=[nuova_variabile; vertcat(data{1,1}{j,1})];
            nuova_variabile=[nuova_variabile; (data(j,1))];
        elseif strncmp('$GPGLL',data(j,1), 6)==1; %%stanno sempre sotto?
            nuova_variabile=[nuova_variabile; (data(j,1))];
        end
    end
end

b=char(nuova_variabile)

c=cellstr(b(:,1:6)) idx=strcmp(c,'$SDDBT')'; ii=[1 diff(idx)]; out=nuova_variabile(find(ii~=0))

ne=ceil(numel(out)/2) out=cellfun(@(x,y) [x ' ' y],out(1:2:end), out(2:2:end),'un',0); expression = ' '; replace = ','; out2= regexprep(out,expression,replace); and this is what we ended up with:

'$SDDBT,00004.9,f,0001.5,M,0000.8,F*07,$GPGLL,5340.91664087,N,00713.79587546,E,073418.00,A,D*62' [1x122 char] '$SDDBT,00006.5,f,0002.0,M,0001.0,F*06,$GPGLL,5340.90756479,N,00713.79236750,E,073424.00,A,D*61' '$SDDBT,00006.8,f,0002.1,M,0001.1,F*0B,$GPGLL,5340.90609055,N,00713.79168827,E,073425.00,A,D*66' '$SDDBT,00007.8,f,0002.4,M,0001.3,F*0D,$GPGLL,5340.90462662,N,00713.79106230,E,073426.00,A,D*6C'

The problem is that the SDDBT string is not always of the same lenght. Cutting at E for the whole cell, would not be appropriate. I am completely new in matlab and I am still learning. Working with string is not that easy for me. I would immagine that I could have split the two type of string into two colums, trim and then ricombine.

MATLAB is driving me MAD!!!!!!!!!!!!!!!!!!!!!