Re: Random bug, OSX 10.04

This WebDNA talk-list message is from

2001

It keeps the original formatting. numero = 38778
interpreted = N
texte = Thanks for your reply John,You are correct in your assumption that I am only using it in a shownext. Myoriginal code before I tried a few different ways was with a number that Iset and it still did not work.Even this shows me different results upon clicking next and previous:[Search db=/database.db&group1field=field1+field2+field3&wagroup1datarq=[keyword]&group1word=ww&max=10&GROUP1sort=1&GROUP1sdir=ra&randseed=25]Any ideas what I should do to report this (if you agree it is a bug).I will copy this to Smith Micro.Paul> Paul Fowler wrote:>> >> I am running WebCat 4.0.2rc2a with OSX 10.0.4 and the latest iTools.>> >> Any ideas why after moving this search to an OSX server the random search>> stopped working, and when the [shownext] is hit a couple of times the>> [lastrandom] turns into a scientific notation mess.> > There have been platform specific problems with random search before.> It is dependent on the O/S for the underlying random numbers. I would> set up a page with a loop containing [random] and [lastrandom] and see> if it looks like the numbers returned are random.> >> >> I tried setting a text variable and using that as the randseed, but the>> search was still different every time. The sort was ignoring the randseed>> and returning different orders every time.> > To quote the docs:> > You can force the randomizer to return items in the same> order if you add another parameter RandSeed=879 (or any other> integer number).> ^^^^^^^^^^^^^^>> >> Could this be a random bug in the OSX version...>> >> This worked fine on a OS9, WebTen WebCat 3.8 machine.>> >> My search below.>> >> [Search>> db=/database.db&group1field=field1+field2+field3&wagroup1datarq=[keyword]&gr>> oup1word=ww&max=10&GROUP1sort=1&GROUP1sdir=ra&randseed=[lastrandom]]> > You are using this strictly in a [shownext] situation aren't you? I> can't think of another reason why you would want to force a specific> random order. And you do understand that [lastrandom] is only the last> [random] from that page, don't you? If you move this between more> than one page, you must either pass the [lastrandom] as a form var> or use [SearchString].> > HTH> > John|--------------------------------------|| Paul Fowler| Zgtec, Inc.| P.O. Box 11199| Aspen, CO 81612| paul@zgtec.com| http://www.zgtec.com|--------------------------------------|-------------------------------------------------------------This message is sent to you because you are subscribed to the mailing list .To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://search.smithmicro.com/ Associated Messages, from the most recent to the oldest:

Re: Random bug, OSX 10.04 (John Peacock 2001)
Re: Random bug, OSX 10.04 (Paul Fowler 2001)
Re: Random bug, OSX 10.04 (John Peacock 2001)
Random bug, OSX 10.04 (Paul Fowler 2001)

Thanks for your reply John,You are correct in your assumption that I am only using it in a shownext. Myoriginal code before I tried a few different ways was with a number that Iset and it still did not work.Even this shows me different results upon clicking next and previous:[Search db=/database.db&group1field=field1+field2+field3&wagroup1datarq=[keyword]&group1word=ww&max=10&GROUP1sort=1&GROUP1sdir=ra&randseed=25]Any ideas what I should do to report this (if you agree it is a bug).I will copy this to Smith Micro.Paul> Paul Fowler wrote:>> >> I am running WebCat 4.0.2rc2a with OSX 10.0.4 and the latest iTools.>> >> Any ideas why after moving this search to an OSX server the random search>> stopped working, and when the [shownext] is hit a couple of times the>> [lastrandom] turns into a scientific notation mess.> > There have been platform specific problems with random search before.> It is dependent on the O/S for the underlying random numbers. I would> set up a page with a loop containing [random] and [lastrandom] and see> if it looks like the numbers returned are random.> >> >> I tried setting a text variable and using that as the randseed, but the>> search was still different every time. The sort was ignoring the randseed>> and returning different orders every time.> > To quote the docs:> > You can force the randomizer to return items in the same> order if you add another parameter RandSeed=879 (or any other> integer number).> ^^^^^^^^^^^^^^>> >> Could this be a random bug in the OSX version...>> >> This worked fine on a OS9, WebTen WebCat 3.8 machine.>> >> My search below.>> >> [Search>> db=/database.db&group1field=field1+field2+field3&wagroup1datarq=[keyword]&gr>> oup1word=ww&max=10&GROUP1sort=1&GROUP1sdir=ra&randseed=[lastrandom]]> > You are using this strictly in a [shownext] situation aren't you? I> can't think of another reason why you would want to force a specific> random order. And you do understand that [lastrandom] is only the last> [random] from that page, don't you? If you move this between more> than one page, you must either pass the [lastrandom] as a form var> or use [SearchString].> > HTH> > John|--------------------------------------|| Paul Fowler| Zgtec, Inc.| P.O. Box 11199| Aspen, CO 81612| paul@zgtec.com| http://www.zgtec.com|--------------------------------------|-------------------------------------------------------------This message is sent to you because you are subscribed to the mailing list .To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://search.smithmicro.com/ Paul Fowler

DOWNLOAD WEBDNA NOW!

Re: Random bug, OSX 10.04

2001

Top Articles:

Related Readings: