Why am I receiving xlswrite error 0x800a03ec?

I've run into the issue where the error code

0x800a03ec

is output from xlswrite in the 'message' structure.( NOTE: I'm using Excel 2003 and Matlab 2006a )

All I'm writing to the excel sheet are double precision values. The original goal was to take notepad files and write them to excel sheets. The files are held in 3 folders with 20 files a piece each possessing 40000x2 (row*col) data points. Each folder gets its own sheet in excel. I've already read about the character limits, column limits, and row limits. I know I'm not exceeding the row or column limits, and I'm 99% certain I'm not violating the character limits. I also know I'm not exceeding the overall data limit (of excel not necessarily xlswrite() ) because I've had to "hand" generate 1.6 million cells worth of data for test cases for stress testing a very similar script. (The files I'm talking about have nothing to do with stress testing)

What's interesting is that in all of the folders it cycles through all of them consistently failing on columns AA,AD, and BF. In the excel template I'm writing too (not a true excel template, just a ' .xls ' file) AA and BF just happen to be the first columns on their respective printable pages; AD, of course, does not. All of the other files write just fine. AA, AD, and BF are the 10th, 11th, and 20th files to be written from each of the folders respectively with 3 folders existing. The data in each folder were not collected concurrently. I collected data one folder at a time; accumulating 20 files before moving to the next folder. All of the files (1-60) were collected in one test session, so it is highly unlikely the data itself is bad, especially considering the circumstances. The excel file isn't protected. I'm not writing to a range of cells, just a column at a time via a 'for' loop.

I have another fairly similar script from which my current script is based and it hasn't ever had a problem with writing before while using, more or less, same sets of data.

Does anyone have a suggestion as to what I should do or at least what may be causing this?

I politely ask I'm not given the following link. I've seen it too many times.

http://www.mathworks.com/matlabcentral/answers/101631-why-do-i-receive-an-error-error-code-0x800a03ec-when-using-xlswrite-in-matlab

Thanks in advance.

[Walter Roberson]

Have you tried using xlswrite1() from the File Exchange ?

[Cedric]

If the first data file goes through, what happens is you shunt your code so it always takes this one? There is almost nothing to do in the code but to add a line that overwrites the path/name of the dynamically generated data file with the path/name of the first file. E.g.

 %dataFile = fullfile( folderName, dirListing(k).name ) ;
 dataFile = 'Data001/File001.txt' ;

This way you will keep all the loops, try to process/stack the same number of files, but you always use a content that worked once.

This is a quick test. If it doesn't work .. well, if MATLAB 2006a and Excel 2003 support it, I would give it a try using this. It would moreover greatly improve the performance of your code.

[Image Analyst]

Can you attach one of the workbooks? Would you consider upgrading your Excel and MATLAB versions if that turned out to be the easiest, or only, way to fix it (like if I proved it worked on my system)?

[Michael VanderWal]

I tried getting a the xlswrite1() after you mentioned it, but I had difficulties to get it working.

The file path shouldn't and still doesn't appear to be an issue, so unless there is an explanation I don't really have a good reason to go through and change all of the variables to work with the suggested fullfile() method.

The work is literally and empty workbook in which I had re-sized columns only to realize later that I wasn't writing to those sheets because I was naming the sheets after the test groups (folders).

My work around for the problem ended up being taking part of Cedric Wannaz's suggestion and imported all of the files from a folder before trying to write them. When I imported the files, I stored them in a massive matrix instead of a single array or multiple arrays. I then wrote the entire matrix to excel. I have to now do a little bit of clean up afterward, but it works better than manually importing and filtering the data in excel.

I also caught an error in my thinking in the original post. I had stated that I had made over 1.6 million data points in excel before which is true, but I did poor math and that actually is smaller, not larger, than what I was trying to accomplish here which was 2.4 million data points.

I may try to fix it better in the future, so I will leave this unanswered.

Thanks for the help.