From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-0.8 required=5.0 tests=BAYES_00,INVALID_DATE autolearn=no autolearn_force=no version=3.4.4 Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!ames!ucbcad!ucbvax!SIERRA.STANFORD.EDU!Mendal From: Mendal@SIERRA.STANFORD.EDU.UUCP Newsgroups: comp.lang.ada Subject: 11.6 at work Message-ID: <12293594852.9.MENDAL@Sierra.Stanford.EDU> Date: Sat, 11-Apr-87 03:13:26 EST Article-I.D.: Sierra.12293594852.9.MENDAL Posted: Sat Apr 11 03:13:26 1987 Date-Received: Sun, 12-Apr-87 03:08:45 EST Sender: daemon@ucbvax.BERKELEY.EDU Distribution: world Organization: The ARPA Internet List-Id: Dear Ada Fans- Concerning the Constraint_Error on the program which has the 0 .. 100 range.... Although I do not have trusty Language Maintenance Committee notes in front of me, I seem to recall on several occasions that Robert Dewar and Robert Eachus have stated that according to the ARM, 11.6, an implementation can "optimize" a portion of a program if its only "effect" is to cause the raising of a predefined exception. (I hope I have not misrepresented Dewar or Eachus here.) Taken in the context in which the original program appeared, the only "effect" of the program would be to raise Constraint_Error. By the rules granted in 11.6, a clever code generator could simply eliminate the assignment altogether. I recall the DEC and Verdix compilers doing such things. Any comments? -- gom -------