Hi Levente,

<[email protected] substituted for octave-maintain...@...>

Levente Torok wrote:
> On Fri, Nov 19, 2010 at 6:45 PM, Philip Nienhuis<[email protected]>  
> wrote:
>> Nov. 17, 2010 Levente Torok wrote:
>>>
>>> Dear Philip, Carlo and octave-dev,
>>>
>>> I had a chance to polish things according to your commands.
>>> However I have a few questions.
>>
>> A while later, Levente Torok wrote:
>> <snip>
>>>
>>> I don't fix myself to ML compatibility at all.
>>> IO package will be just fine place to have these functions but
>>> according to our concept FPL would be even more appropriate.
>>> So if you can adjust it to the standards I would be with happy with this.
>>
>> Hi Levente,
>>
>> I went through your functions, they look very good. Thank you very much!
>>
>> Yet I added and changed a few things. Please check if you find this
>> acceptable. See below...
>> Carlo, if Levente agrees with my fixes/additions, will you put them in fpl
>> pkg or shall I do it?
>>
>>
>> But first, a question to Levente:
>> (I know nothing about vtk, so apologies if this is a stupid question)
>>
>> The array X in the example in savevtk's texinfo header is a real array, yet
>> the .vtk file that is produced only contains integer values (as the fprintf
>> statements invoke %d format specifiers rather than %f).
>> Is that intended?
>>
>>
>> savevtk:
>> ========
>> - Fixed texinfo
>> - Added "See also" (savevtkvector)
>> - Fixed small typo (adoptatoin)
>> - Added a return value to the function statement (so one can test in scripts
>> if savevtk did its job OK). If you don't want to see "ans = 1" you need to
>> put a semicolon after the savevtk() statement.
>> - Added a default filename "vtkout.vtk". Just remove the '=vtkout.vtk' if
>> you don't like this
>> - Added basic input argument checking (incl. filename type - should be
>> character)
>> - Added a try-catch around the fprintf statements to catch disk full errors
>> - and you don't need to count dimensions yourself, Octave has a function for
>> that: ndims()
>> - Minor style changes (spaces etc).
>>
>> savevtkvector
>> =============
>> - Fixed texinfo
>> - Added "See also" (savevtkvector)
>> - Fixed small typo (adoptatoin)
>> - Added a return value to the function statement (so one can test in scripts
>> if savevtk did its job OK)
>> - Added a default filename "vtkvectorout.vtk". Just remove the
>> '=vtkvectorout.vtk' if you don't like this
>> - Added basic input argument checking
>> - Added a try-catch around the fprintf statements to catch disk full errors
>> - I had to replace wrong accents (a.o., ASCII 226, skewed) with ' (straight)
>> in almost all fprintf statements - this may be caused by editing witha word
>> processor rather than a program editor
>> - Minor style changes (spaces etc).
>>
>> See attached files.
>>
>> Please tell us if you're OK with the changes.
>>
>>
>> Philip
>>
>
> Dear Philip and Others
>
> The description of the file format is a but bogus for me however
> according to my tests, not only integers %d but floating points can go
> into the vtk file.

%d format specifier means integer output.

Now,
savevtk has %d         ===> writes integers
savevtkvector.m has %f ===> writes floats.

Indeed, my tests show that savevtk's output contains an array of 
integers rather than floats.

> I support the idea of default filenames. ( I didn't know if default
> parameters can be given in this way. This is grate! )

Told you so, Octave's script language is much better than ML's :-)

> I wonder what is the overhead of using try catch. In c++ it could be
> really considerable this is why I use it in a place where it is really
> the best to use.

Considering we're using script language and not compiled code, + we're 
writing to (potentially slow) storage devices, the overhead from 
try...catch is very probably negligible.
However, the advantages in terms of meaningful error catching are 
indispensable.

> I agree fpl is an appropriate place for these files.
> Plus, thanks guys for being so precise in corrections.

Thank you for being positive about my (admittedly somewhat picky) 
suggestions.

Carlo / Levente, what do we do with the %d format specifiers in savevtk.m?

Oh and Very Soon Now I'll wipe savevtk.m/savevtkvector.m from the io 
package in svn.

Best wishes,

Philip

------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to