bugGNU Octave - Bugs: bug #50157, fwrite input argument type

 
 

bug #50157: fwrite input argument type

Submitter:  Guillaume <gyom>
Submitted:  Thu 26 Jan 2017 04:09:40 PM UTC
   
 
Category:  None Severity:  3 - Normal
Priority:  5 - Normal Item Group:  Regression
Status:  Fixed Assigned to:  None
Originator Name:  Guillaume Open/Closed:  * Closed
Release:  * dev Operating System:  * GNU/Linux
Fixed Release:  None Planned Release:  None
* Mandatory Fields

Add a New Comment Rich Markup
   

Jump to the original submission

Fri 27 Jan 2017 10:23:52 AM UTC, comment #9: 

Many thanks for the quick fix - I can confirm it works fine now.

Guillaume <gyom>
Thu 26 Jan 2017 06:12:32 PM UTC, comment #8: 

And yes, the tests are better, but many more could be added.

The tests I just added are good for scalar/matrix types, but not for reading without conversion (result is double) or for other conversion combinations when reading and writing.  There's a lot to test for just fread/fwrite...

John W. Eaton <jwe>
Group administrator
Thu 26 Jan 2017 06:11:04 PM UTC, comment #7: 

Oh damn, I copied the test to a separate file for quick testing, then fixed the problems there and forgot to copy them back.  Checking in a fix as soon as I can verify that I haven't made some other stupid mistake.

John W. Eaton <jwe>
Group administrator
Thu 26 Jan 2017 05:49:05 PM UTC, comment #6: 

New failing test in io.tst after running 'make check':


>>>>> processing io.tst
***** test
 classes = {"int8", "int16", "int32", "int64", ...
            "uint8", "uint16", "uint32", "uint64", ...
            "single", "double"};
 nm = tempname ();
 id = fopen (nm, "wb+");
 n = numel (classes);
 for i = 1:n
   cls = classes{i};
   s_in = ones (1, 1, cls);
   m_in = ones (2, 2, cls);
   m_shape = size (m);
   fwrite (id, s, numel (s), cls);
   fwrite (id, m, numel (m), cls);
   frewind (id);
   s_out = fread (id, numel (s_in), sprintf ("%s=>%s", cls, cls));
   m_out = fread (id, numel (m_in), sprintf ("%s=>%s", cls, cls));
   m_out = reshape (m_out, m_shape);
   fclose (id);
   unlink (nm);
   assert (s_in, s_out);
   assert (m_in, m_out);
 endfor
!!!!! test failed
'm' undefined near line 13 column 20


Looks like instances of "m" should be replaced with "m_in", and "s" with "s_in".

Rik <rik5>
Group administrator
Thu 26 Jan 2017 05:40:13 PM UTC, comment #5: 

On the testing note, we still don't have 100% function coverage, let alone 100% line coverage, but I think the test suite has grown large enough to finally be valuable.  I don't think we are making so many quick fixes that lead to later bugs because more of those problems are getting caught early by the regression tests.

That's mostly a "hurray for us" statement as things are getting better.

Rik <rik5>
Group administrator
Thu 26 Jan 2017 05:25:04 PM UTC, comment #4: 

I pushed the following fix (with a test!):

  http://hg.savannah.gnu.org/hgweb/octave/rev/8aba6d170887

Reopen if this doesn't fix the problem.

The skip parameter probably should be octave_idx_type, but that's something to fix for another day...

John W. Eaton <jwe>
Group administrator
Thu 26 Jan 2017 05:13:15 PM UTC, comment #3: 

jwe's diagnosis is correct.  The problem is the difference between scalar/matrix rather than between say double/uint32.  See the following.


octave:1> fid = fopen ('file.dat', 'wb');
octave:2> fwrite (fid, uint32([1 2]), 'uint32')
ans =  2
octave:3> fwrite (fid, uint32([1]), 'uint32')
error: octave_base_value::write(): wrong type argument 'uint32 scalar'



Rik <rik5>
Group administrator
Thu 26 Jan 2017 04:56:27 PM UTC, comment #2: 

It's not the conversion routines, it's a mistake in the prototype for the write function for scalar integer octave_value types, int vs octave_idx_type.  I'll fix.

John W. Eaton <jwe>
Group administrator
Thu 26 Jan 2017 04:50:33 PM UTC, comment #1: 

Confirmed.  Octave seems to be missing a large number of conversion routines.

Rik <rik5>
Group administrator
Thu 26 Jan 2017 04:09:40 PM UTC, original submission:  

Running the follwing with current tip:


fid = fopen ('file.dat', 'wb')
fwrite (fid, int32(1), 'int32')


returns:


error: octave_base_value::write(): wrong type argument 'int32 scalar


fwrite returns 1 as expected if second argument is double, or with 4.2.

Guillaume <gyom>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:
   
   
Comment:
   

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by jwe (Posted a comment)
  • -email is unavailable- added by rik5 (Posted a comment)
  • -email is unavailable- added by gyom (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

    Only group members can vote.

     

    Follow 3 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2017-01-26 jwe StatusConfirmed Fixed
        Open/ClosedOpen Closed
    2017-01-26 rik5 StatusNone Confirmed

    Back to the top

    Powered by Savane 3.13-758e.
    Corresponding source code