home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.tcl
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!sdd.hp.com!think.com!ames!pacbell.com!att-out!cbfsb!cbnews!cbnewsm!gah
- From: gah@att.com (George A. Howlett)
- Subject: Re: Advice wanted on math functions for Tcl 7.0
- Reply-To: george.howlett@att.com
- Organization: AT&T Bell Laboratories
- Date: Tue, 22 Dec 1992 20:12:13 GMT
- Message-ID: <1992Dec22.201213.26543@cbnewsm.cb.att.com>
- X-Newsreader: TIN [version 1.1 PL8]
- References: <ljek3aINN7pu@news.bbn.com>
- Sender: news@cbnewsm.cb.att.com (NetNews Administrator)
- Nntp-Posting-Host: grenache.cnet.att.com
- Lines: 45
-
- Jeff Palmucci (jpalmucc@bbn.com) wrote:
-
- : The most important feature of TCL to me is its size (no one should be
- : using TCL for its speed). In the PC world, TCL is already pretty big.
- : Please don't make it any bigger. If performance were an issue, I'd
- : rethink my app/TCL boundary, and move more code into the C-side.
-
- This will depend on the size of your math library routines and if you
- have floating point hardware, etc. You need to add very little C code
- to link in the math functions. Do you see a bloat when you link in
- math functions routines into a normal C program? Of course for those
- with shared libraries, this is not a problem.
-
- : Someone will probably flame me, but I don't see any great use for
- : these functions in application script writing or GUI layout. (circular
- : windows?) Even though I have lots of stuff loaded into my emacs, no
- : one bothers to define sin or cos.
-
- Clearly, numerically oriented applications will benefit greatly. But
- I also found it invaluable when using the canvas widget (simple trigonometry
- like calculating the distance between two points, rotations, etc).
-
- Brad Morrison (brad@NeoSoft.com) wrote:
-
- : But (+, *, etc.) are operators, not functions. I am for defining the
- : math functions as intrinsic Tcl functions. I will admit, the efficiency
- : *is* tempting, but what about other math functions? Should expr just
- : keep growing to accomodate them, too? This sort of approach, with an emphasis
- : toward efficiency, with syntax changes and inconsistencies considered as
- : secondary, reminds me a bit too much of Perl.
-
- The set of intrinsic math functions needed is fairly well defined. The
- difference in actual code for either implementation is probably neglible.
-
- I don't understand where it is syntactically inconsistent. But it won't
- break extended Tcl code. People here have been using it for the last
- year or better with no problems.
-
- --gah
-
-
-
-
-
-
-