achim@ruso.UUCP (Achim Sturm) (04/28/89)
Hello World
Does someone work with CTREE ?
my problem is :
mix different indices for searching .
for example :
Data_file with length 10 ,
indices over each byte ;
searching :
first index between 0x00 0x10
second index between 0x21 0x45
fifth index between 0x67 0xa9
other indices between 0x00 0xff
The problem now is , to solve this in an acceptable time .
Achim
uucp: ...!uunet!mcvax!unido!ruso!achimchardin@cbnewsk.att.com (christopher.hardin) (04/04/91)
Greetings, I am having difficulty getting Ctree Plus from Faircom to work on an AT&T 3B2/1000. Has anyone compiled the Version 6.0 code and managed to get the sample programs to work? In particular, creating files with incremental ISAM structures causes a memory fault and drops core. I'm 99.44% sure that I installed it correctly. As an aside, can anyone give me their opinion on [cdr]tree? How well do they work in a multi-user environment (AT&T UNIX System V Release 3.2.2). If you think ctree is a bad product, please recommend a low end database that will work in my environment (no MS-DOS please). Any insight would be greatly appreciated. -- TTFN (Ta ta for now) -- Tigger Chris Hardin chris@green.att.com ...!att!green!chris 404.750.8921
jhagen@TALOS.UUCP (Jarom Hagen) (04/05/91)
In comp.databases you write: > I am having difficulty getting Ctree Plus from Faircom to work on an >AT&T 3B2/1000. Has anyone compiled the Version 6.0 code and managed to get >the sample programs to work? We got Ctree Plus to work on a DEC 5400 running Ultrix 4.0. We also got C-tree 4.3 to run on a 386 running System V 3.2 and a Pyramid. There are several #defines that need to be used to get things to work. Did you run cttest? Cttest shows you what several #defines should be. Also, remember that you have to change the parameter files that come with the sample programs to 4 bytes alignment. Personally, I found the ctree errors not that helpful in finding problems. > As an aside, can anyone give me their opinion on [cdr]tree? C-tree seems to be a good product, although we haven't tried stressing it with several users at the same time yet. Jarom -- ------------------------------------------------------------------------------- *Not paid for and/or endorsed by National Political Resources Incorporated. 602 Cameron St, Alexandria VA 22314 (UUCP: ...uunet!uupsi!pbs!npri6!jhagen)
cohill@vtserf.cc.vt.edu (Andrew M. Cohill) (04/05/91)
We have been working with ctree, dtree, and rtree for about three months now, and have mixed feelings about it. We liked the concept of coupling the display stuff (dtree) with the database stuff (ctree), and indeed, if you use dtree you don't have to make any database calls at all. None. Nada. Dtree takes care of everything for you, including little niceties like checking for duplicate keys and letting you know about. By contrast, if you were using db_vista and Vermont Views, you would have to hand code all that stuff. Not terribly hard, but still involves a lot of low level grunt work. We have ctree running on a Mac, DOS machines, and a DEC 5810 running Ultrix. Now the downside: The new release of dtree, although it includes lots of screen goodies, is extremely buggy. All one programmer here has done for the last six weeks is find bugs for Faircom (no, we aren't beta testers, but we should be). The ctree upgrade appears to be okay, but we really have not had much chance to test it, bogged down as we are with dtree. Rtree is really big; it may be problematic to use on DOS machines because adding it makes the executables so large. Small programs with cdr-tree libraries check in at 400K+ before you add your own code, and a big chunk of that is just rtree. Rtree appears to be buggy with the upgrade; we aren't really sure. While we had little trouble moving ctree from platform to platform, dtree is supposed to work on both DOS and Unix. We could not get it work reliably on Ultrix. Faircom is moderately responsive; they use an answering machine a lot but do call back within 24 hours. They have sent us two sets of bug patches in the last month, and are open about admitting problems, which goes a long way with us. I much prefer that to vendors who try to pretend there aren't any problems. Their biggest problem, I think, is that they have a good product but are a little too small to support it properly, and are under a lot of pressure from competitors to get stuff out the door. But then, that's the history of software... I'd be interested in hearing from other Faircom users working with the upgrade. Andy Cohill -- | ...we have to look for routes of power our teachers never | imagined, or were encouraged to avoid. T. Pynchon | |Andy Cohill cohill@vtserf.cc.vt.edu VPI&SU