Re: [sword-devel] 1.6.1 final call

2009-12-25 Thread Dmitrijs Ledkovs
2009/12/25 Matthew Talbert : > OK, I am still not understanding why there is an issue, or what the > real cause of the issue is. However, this line I think will work: > > const unsigned int MAX_CONV_SIZE = 6536 * sizeof(wchar_t) * sizeof(wchar_t); > > If somebody can come up with an actual explanat

Re: [sword-devel] 1.6.1 final call

2009-12-24 Thread Matthew Talbert
OK, I am still not understanding why there is an issue, or what the real cause of the issue is. However, this line I think will work: const unsigned int MAX_CONV_SIZE = 6536 * sizeof(wchar_t) * sizeof(wchar_t); If somebody can come up with an actual explanation for why there is a problem, and a n

Re: [sword-devel] 1.6.1 final call

2009-12-24 Thread Matthew Talbert
On Thu, Dec 24, 2009 at 10:20 PM, Karl Kleinpaste wrote: > > Matthew Talbert writes: > > We just discovered one issue with the indexing. Evidently > > MAX_CONV_SIZE = 1024 * 1024 is too big for an int on win32. > > 1024*1024 = 1048576.  That's only 20 bits' worth. You're right, of course. I real

Re: [sword-devel] 1.6.1 final call

2009-12-24 Thread Karl Kleinpaste
Matthew Talbert writes: > We just discovered one issue with the indexing. Evidently > MAX_CONV_SIZE = 1024 * 1024 is too big for an int on win32. 1024*1024 = 1048576. That's only 20 bits' worth. ___ sword-devel mailing list: sword-devel@crosswire.org

Re: [sword-devel] 1.6.1 final call

2009-12-24 Thread Matthew Talbert
Hey Troy, We just discovered one issue with the indexing. Evidently MAX_CONV_SIZE = 1024 * 1024 is too big for an int on win32. Therefore, crashes ensue. I'm not sure yet what to change it too. unsigned int still isn't big enough. I'll get back to you as soon as I figure out a real way to solve th

Re: [sword-devel] 1.6.1 final call

2009-12-24 Thread Ben Morgan
On Fri, Dec 25, 2009 at 9:53 AM, Greg Hellings wrote: > Visual Studio 9 reporting in: > > I don't have our SWORD version of ICU on-hand, nor is it particularly > important to those of us over here in the BibleTime community, since > we use Qt for our transliteration. > > regex.c throws a number o

Re: [sword-devel] 1.6.1 final call

2009-12-24 Thread Greg Hellings
Visual Studio 9 reporting in: I don't have our SWORD version of ICU on-hand, nor is it particularly important to those of us over here in the BibleTime community, since we use Qt for our transliteration. regex.c throws a number of warnings of the type: ..\..\..\src\utilfuns\regex.c(3877) : warnin

[sword-devel] 1.6.1 final call

2009-12-24 Thread Troy A. Griffitts
Thanks again to everyone who spent time so close to the holidays to help out with this. I have committed 4 very minor changes to avoid warnings presented from the FORTIFY_SOURCE=2 compiler flag, for Jonathan Marsden and the Ubuntu team. I now know of no outstanding issues that we plan to tackle f