ado@elsie.UUCP (Arthur David Olson) (11/12/86)
One possibility for emulating most of the Reiser cpp's handling of
#define CTRL(x) ('x' & 037)
in proposed draft ANSI C is:
#define CTRL_A ('A' & 037)
#define CTRL_B ('B' & 037)
...
#define CTRL_Z ('Z' & 037)
#define CTRL_a ('a' & 037)
...
#define CTRL_z ('z' & 037)
#define CTRL(x) CTRL_ ## x
A deficiency here is that the handling of constructs such as
CTRL([)
isn't emulated. . .but the above approach would handle most existing code.
Or, at any rate, it would handle most existing code if the result of catenating
preprocessing tokens with the "##" operator is itself macro expanded.
Whether the Standard currently calls for such rescanning,
whether the Standard will call for such rescanning in the future,
and whether such rescanning is advisable are all open questions.
--
UUCP: ..decvax!seismo!elsie!ado ARPA: elsie!ado@seismo.ARPA
DEC, VAX, Elsie & Ado are Digital, Borden & Ampex trademarks.jsdy@hadron.UUCP (Joseph S. D. Yao) (11/13/86)
In article <1319@sunybcs.UUCP> colonel@sunybcs.UUCP (Col. G. L. Sicherman) writes: >) >But. . .is there a way to write an ANSI cpp version of >) > #define CTRL(c) > ... What's wrong with this old trick? > #define Q " > #define CTRL(z) (Q z "[1]^0100) The XOR, (x^0100), will not generally work. Use (x&037). Also, there was some concern that "..."[x] will not work where a constant is required, such as in a case statement. -- Joe Yao hadron!jsdy@seismo.{CSS.GOV,ARPA,UUCP} jsdy@hadron.COM (not yet domainised)
karl@haddock.UUCP (Karl Heuer) (11/13/86)
In article <614@hadron.UUCP> jsdy@hadron.UUCP (Joseph S. D. Yao) writes: >The XOR, (x^0100), will not generally work. Use (x&037). True if you want the domain of the function to include lowercase. I never do; e.g. I claim that the proper name for EOT is "^D" rather than "^d". If you're willing to make this restriction, then the XOR has the advantage of allowing the use of CTRL('?') for DEL, which is a widely (though by no means universally) accepted notation. Karl W. Z. Heuer (ima!haddock!karl or karl@haddock.isc.com), The Walking Lint
jsdy@hadron.UUCP (Joseph S. D. Yao) (11/24/86)
In article <170@haddock.UUCP> karl@haddock.UUCP (Karl Heuer) writes: >In article <614@hadron.UUCP> jsdy@hadron.UUCP (Joseph S. D. Yao) writes: >>The XOR, (x^0100), will not generally work. Use (x&037). >True if you want the domain of the function to include lowercase. I never >do; e.g. I claim that the proper name for EOT is "^D" rather than "^d". Your privilege. My definition of control char, umpteen years ago, just said that no bits not in the lower five were clear. Said nothing about "proper" names for them (othere than NUL, EOT, etc.). Most people make a control-d without hitting the shift key. Further, Control-ESC is ESC, not '['; and Control-TAB is TAB, not 'I'. Yes, people are unlikely to do this in C code. But that's part of the reason I reacted as I did. >allowing the use of CTRL('?') for DEL, which is a widely (though by no means >universally) accepted notation. My surprise. I'd thought that Berkeley introduced this notation. At any rate I'd never heard or seen it before 2BSD. (My experience before that had covered a variety of hardware, software, and documentation.) If anyone has a VERIFIABLE pre-BSD reference for use of ^?, I'd appreciate your MAILing to me. Summary will be posted. (I can't remember any non-BSD-derived version of the Unix(R) operating systems that use this, just to stick my neck out further.) This cute (x^0100) trick is the first reason I'd ever seen for '^?' to mean DEL; so I have learned something new today. (I guess I'd assumed that with someone's screwy keyboard you could generate DEL as a Control-?.) -- Joe Yao hadron!jsdy@seismo.{CSS.GOV,ARPA,UUCP} jsdy@hadron.COM (not yet domainised)
maart@cs.vu.nl (Maarten Litmaath) (10/24/89)
djones@megatest.UUCP (Dave Jones) writes:
\...
\#define CNTL(c) ((c) & 037)
I prefer:
#define CTRL(x) ((x) ^ 0100)
...so you can say
CTRL('?')
to get DEL.
--
A symbolic link is a POINTER to a file, | Maarten Litmaath @ VU Amsterdam:
a hard link is the file system's GOTO. | maart@cs.vu.nl, mcsun!botter!maartdavidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/25/89)
In article <3786@solo6.cs.vu.nl>, maart@cs.vu.nl (Maarten Litmaath) writes: | | #define CTRL(x) ((x) ^ 0100) | | ...so you can say | | CTRL('?') | | to get DEL. You should use another name, perhaps, since many people assume that CTRL returns control characters. They are defined as the values from 00-37 octal, from the action of the Control key on the original Teletype terminals. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
maart@cs.vu.nl (Maarten Litmaath) (10/25/89)
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) writes: \In article <3786@solo6.cs.vu.nl>, maart@cs.vu.nl (Maarten Litmaath) writes: \| \| #define CTRL(x) ((x) ^ 0100) \| \| ...so you can say \| \| CTRL('?') \| \| to get DEL. \ \ You should use another name, perhaps, since many people assume that \CTRL returns control characters. They are defined as the values from \00-37 octal, from the action of the Control key on the original Teletype \terminals. In BSD the new tty driver as well as vi, cat -v, ... use ^? to denote DEL. Seems pretty normal to me. According to iscntrl(3) DEL is a control character too. Too bad it isn't a normal control character; it would have simplified things. -- A symbolic link is a POINTER to a file, | Maarten Litmaath @ VU Amsterdam: a hard link is the file system's GOTO. | maart@cs.vu.nl, mcsun!botter!maart
karl@haddock.ima.isc.com (Karl Heuer) (10/25/89)
In article <1422@crdos1.crd.ge.COM> davidsen@crdos1.UUCP (bill davidsen) writes: >control characters ... are defined as the values from 00-37 octal... That definition may have been the original one, but as far as C is concerned the only definition of `control character' is `what iscntrl() tests for' -- and this does include DEL on ASCII systems. Karl W. Z. Heuer (ima!haddock!karl or karl@haddock.isc.com), The Walking Lint
diamond@csl.sony.co.jp (Norman Diamond) (10/26/89)
In article <3802@solo7.cs.vu.nl> maart@cs.vu.nl (Maarten Litmaath) writes: >In BSD the new tty driver as well as vi, cat -v, ... use ^? to denote DEL. >Seems pretty normal to me. According to iscntrl(3) DEL is a control character >too. It also seems reasonable to me to think of DEL as a control character. However, it does seem to be ambiguous in the industry. That is why some manuals refer to "control characters and DEL" in order to avoid ambiguity. >Too bad it isn't a normal control character; it would have simplified >things. Well, it couldn't possibly be a normal control character. To delete a typographical error from paper tape, you had to backspace the tape and punch all 1's (holes) in that column. If someone decided to make DEL something other than all 1's, then they would have to invent another delete that would be all 1's, and it would STILL be ambiguous if the new delete was really a control character or not. -- Norman Diamond, Sony Corp. (diamond%ws.sony.junet@uunet.uu.net seems to work) Should the preceding opinions be caught or | James Bond asked his killed, the sender will disavow all knowledge | ATT rep for a source of their activities or whereabouts. | licence to "kill".