Re: advice / help needed for reading formatted text (textscan, strread, & textread)

[Ben Abbott]

On Nov 19, 2010, at 8:39 PM, John W. Eaton wrote:

> On 16-Nov-2010, John W. Eaton wrote:
> 
> | On 16-Nov-2010, Ben Abbott wrote:
> | 
> | | John, reading this again it looks to me like someone needs to take
> | | the lead. Give my lacking for c/c++ skill, should I enter a "new
> | | feature" request in the tracker?
> | 
> | I started working to add the new escape features to scanf but then I
> | got distracted and haven't finished it.  I'll try to finish it this
> | week.
> 
> After looking at this again, I think it might be best to provide a
> separate textscan function in C++ instead of trying to overload scanf
> to also do the work of textscan as I had originally planned.  Some
> code could probably be shared with scanf, but it will take more work
> than I first thought.
> 
> jwe

John,

To better understand who textscan works I've put together a script that runs on 
Matlab.

Eventually, I'd like to convert these to tests for Octave's textscan (many 
would be xtests).

In the meantime, if you return to working on this, these tests may be useful in 
understanding how ML behaves.

Ben

test_textscan.m
Description: Binary data