How can I read in a txt file, which is being created by a different programm, and plot the numeric data into a contourf plot continously?
显示 更早的评论
Hi there,
So far I have written one function to read in any kind of txt file, which is created by another software (Example is attached, yet in the future these files will be a lot larger). Then, the data was added into the workspace in Matlab. The second script, which I wrote, uses the data, which was created, and plots it continously. Both of these scripts worked independently real well.
Now, I tried to merge these two scripts, so that the data shall be read from the txt file and be plotted at the same time (thats why I start the function with a while cycle).
The txt file, which is created in another program, appends a new row of data every 3 seconds. Ideally, my function will read and update the plot at the same rate.
This is my code:
%first, make sure the data which is read in should be errorless
%Moli: read file and cut the temperature data from raw data(.txt)
% 2nd, write in command window : addpath ('D:\matlab-test'); to make sure
%the directory is correct
%addpath eg ('E:\matlab_daten');
% 3rd, start function with defining dataset and writing name of function
%eg "TD23111501 = ReadTempData ('23.11.20150_1.txt') into
% command window
% then dataset TD23111501 was created and function was read
function [data,Nrows] = ReadPlotTempData(FileName)
count=0;% starts cycle with reading and plotting
while (1)%keeplooping
fid = fopen(FileName); % open data
Nrows = numel(textread(FileName,'%1c%*[^\n]')) % find the number of rows
%4th, data rows 1-15 is only read but not used
%Moli:explore data unitl 'data begin'
for i = 1:15
tline = fgetl(fid);
end
% 5th, a new matrix called eg "data" or "TD23111501" is created, leaving
%the first 15 rows out and using 20 columns
data = zeros(Nrows-15,20); %create matrix (# of rows, # of columns eg 20)
for i = 1:Nrows-15 %ab row 16 daten werden verwendet
tline = fgetl(fid); % read the line
k = strfind(tline, ','); %find positions of comma (,) in each line
restline = tline((k(29)+1):(k(49)));
% %Moli: cut the data, from real data (first record of temperature) to end
% the restline ... ab komma 29-49 mit jeweils 1 leerzeichen nach dem komma
data(i,:) = str2num(restline);
end
fclose(FileName)
%end of the ReadTempData script!
%script: contourf_script_while_cycle starts now
figure (1)
[hc hc] = contourf(data,100) % 100 gibt die genauheit der interpolation an
colorbar; % add aditional things eg
caxis auto %([4 12])%%bestimmt die Temperaturintervall
% welches angezeigt werden soll, 'auto' legt das Intervall selbst fest
legend ('Temperatur in C°')
set (hc,'Linestyle','none');
set (gca,'fontsize',14);
xlabel ('Anzahl der Sensoren', 'fontsize',16);
ylabel ('Zeitreihe', 'fontsize', 16);
colormap jet; colorbar;
drawnow
count=count+1
end
end
My problem / question now is: In the first part of the function, I have to define the dataset, which will be created (TD23111501). Later, when creating the plot, I want to use this data TD23111501 and plot it. However, it doesn't work. Is there another way of defining the dataset and retrieving the data within the same function? Or is it even not possible at all? Thank you for your help
1 个评论
adnan ali awan
2016-1-26
The way I would have done is simply by:
1) Calling and Reading the file;
2) Know what's the type of data! if the original data is in matrix form then use mat2dataset!
3) Use a timer function that calls the reading function
4) For Plotting use hold on with drawnow() ....
采纳的回答
Ingrid
2016-1-26
0 个投票
first of all, you call fopen without a corresponding fclose. This will get you into trouble so always close the file again after you are done reading it.
Secondly, you want to use TD23111501 but you do not create it anywhere. There is no example data attached so I an only guess what you are trying to do, but I guess you want the "data" to be used. If so you need to make sure that both variables have the same name. It is however not clear how many lines you read in per time, if only the last line is changed each time, you should not read in the whole file each time but only that time. If your program runs for a long time this will save you much time.
you will also need a timer of some sort (e.g. wait 3 seconds each loop) or check to see when the file is updated (<http://stackoverflow.com/questions/17522463/matlab-get-the-last-modification-time-of-a-file>)
5 个评论
Catharina Keim
2016-1-26
Ingrid
2016-1-27
for the timer you can use the pause function http://nl.mathworks.com/help/matlab/ref/pause.html but you have to be careful as it might lead to some problems when you run your code for a longer time as there might be a mismatch in the time counting of your code and the one of the other program. I think the safest bet is to check if the file has been updated or not before proceeding, adding a pause of 1 second to avoid excessive checking.
is think this is more what you want?
function ReadPlotTempData(FileName)
count=0;% starts cycle with reading and plotting
figure
while (1)%keeplooping
fid = fopen(FileName); % open data
formatString = '%*f %*f %*f %*f %*f %*f %*f %*s %*f %*f %*s %*f %*s %*f %*f %*f %*f %*s %*f %*s %*s %*s'; % it seems you do not want these 21 first columns so use * to discard them
for ii = 1:28
formatString = [formatString ' %f'];
end
rawData = textscan(fid,formatString,'HeaderLines',15,'Delimiter',',');
data = cell2mat(rawData);
fclose(fid);
[hc hc] = contourf(data,100); % 100 gibt die genauheit der interpolation an
colorbar; % add aditional things eg
caxis auto %([4 12])%%bestimmt die Temperaturintervall
% welches angezeigt werden soll, 'auto' legt das Intervall selbst fest
legend ('Temperatur in C°')
set (hc,'Linestyle','none');
set (gca,'fontsize',14);
xlabel ('Anzahl der Sensoren', 'fontsize',16);
ylabel ('Zeitreihe', 'fontsize', 16);
colormap jet; colorbar;
drawnow
count=count+1
pause(3);
end
Catharina Keim
2016-1-28
Ingrid
2016-1-28
simple, just use %*f instead of %f for the last column. The formatString is just a variable name that I have chosen. For more information on this you can look at the documentation of textscan. It just seemed to me that this is the way to go for you instead of doing all the parsing of your data manually after reading it in line by line
Catharina Keim
2016-1-28
更多回答（0 个）
类别
在 帮助中心 和 File Exchange 中查找有关 Large Files and Big Data 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
