← Back to team overview

cuneiform team mailing list archive

Re: Project status & windows port progress

 

On Wed, Oct 22, 2008 at 1:36 PM, Ladwig, Mike <mike.ladwig@xxxxxxx> wrote:

> I've made some progress; the only change necessary to get past the
> dictionary file read
> was to modify a string read method TE_handle_fgets to break (without
> character skip)
> when a '\n' character was found.

Please post patches to the bug tracker so your work is not needlessly
duplicated.

https://bugs.launchpad.net/cuneiform-linux/+bug/267684

> Currently, I am looking for a problem in the reformatting code,
> OpenFullOutTiger, where
> I am getting a segfault reading back in a internally generated file
> (internal.vit).  Looks
> to be in the ns loop.  I think the problem is that the file needs to be
> opened explicitly
> to read in binary mode, but haven't found how to do that yet (there are some
> odd redefinitions
> of file modes that I haven't tracked down yet).

If I remember correctly, the code uses sizeof(int) when writing the
file and sizeof(int16) when reading it. Or the other way around. Or
something completely different. I had a hell of a time fixing it,
since int is 16 bits on Windows and 32 bits everywhere else. You might
be hitting regressions of this conversion.



Follow ups

References