* A minor name improvement suggestion to Guile API @ 2009-12-16 13:14 Yi DAI 2009-12-17 10:46 ` Ludovic Courtès 0 siblings, 1 reply; 8+ messages in thread From: Yi DAI @ 2009-12-16 13:14 UTC (permalink / raw) To: guile-user [-- Attachment #1: Type: text/plain, Size: 566 bytes --] Hi, all, I read the following in "5.1 Overview of the Guile API" of the reference manual. - Replace <= (less than or equal) with _leq. - Replace >= (greater than or equal) with _geq. - Replace < (less than) with _less. - Replace > (greater than) with _gr. I'd like to suggest the following name convention which complies to names commonly found in assembly languages. - Replace < (less than) with *_lt*. - Replace > (greater than) with *_gt*. Should I send this to the devel-mailing list? -- DAI Yi (代 毅) [-- Attachment #2: Type: text/html, Size: 946 bytes --] ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: A minor name improvement suggestion to Guile API 2009-12-16 13:14 A minor name improvement suggestion to Guile API Yi DAI @ 2009-12-17 10:46 ` Ludovic Courtès 2009-12-17 14:26 ` Linas Vepstas 0 siblings, 1 reply; 8+ messages in thread From: Ludovic Courtès @ 2009-12-17 10:46 UTC (permalink / raw) To: guile-user Hi, Yi DAI <plm.day@gmail.com> writes: > I'd like to suggest the following name convention which complies to names > commonly found in assembly languages. > > > - Replace < (less than) with *_lt*. > - Replace > (greater than) with *_gt*. Changing conventions would mean changing the API or maintaining function name aliases, and I don’t think it’s worth it. Thanks, Ludo’. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: A minor name improvement suggestion to Guile API 2009-12-17 10:46 ` Ludovic Courtès @ 2009-12-17 14:26 ` Linas Vepstas 2009-12-17 15:00 ` Ludovic Courtès 0 siblings, 1 reply; 8+ messages in thread From: Linas Vepstas @ 2009-12-17 14:26 UTC (permalink / raw) To: Ludovic Courtès; +Cc: guile-user 2009/12/17 Ludovic Courtès <ludo@gnu.org>: > Hi, > > Yi DAI <plm.day@gmail.com> writes: > >> I'd like to suggest the following name convention which complies to names >> commonly found in assembly languages. >> >> >> - Replace < (less than) with *_lt*. >> - Replace > (greater than) with *_gt*. > > Changing conventions would mean changing the API or maintaining function > name aliases, and I don’t think it’s worth it. well, the API is controlled by the scheme steering committee, and a change would affect dozens of scheme implementations, not just guile! So this is an odd request. And any novice scheme programmer can create brand-new predicates of any sort they wish, with any name desired, so I don't understand this request at all. --linas ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: A minor name improvement suggestion to Guile API 2009-12-17 14:26 ` Linas Vepstas @ 2009-12-17 15:00 ` Ludovic Courtès 2009-12-17 19:05 ` Keith Wright 0 siblings, 1 reply; 8+ messages in thread From: Ludovic Courtès @ 2009-12-17 15:00 UTC (permalink / raw) To: linasvepstas; +Cc: guile-user Hi, Linas Vepstas <linasvepstas@gmail.com> writes: > 2009/12/17 Ludovic Courtès <ludo@gnu.org>: >> Hi, >> >> Yi DAI <plm.day@gmail.com> writes: >> >>> I'd like to suggest the following name convention which complies to names >>> commonly found in assembly languages. >>> >>> >>> - Replace < (less than) with *_lt*. >>> - Replace > (greater than) with *_gt*. >> >> Changing conventions would mean changing the API or maintaining function >> name aliases, and I don’t think it’s worth it. > > well, the API is controlled by the scheme steering committee, AIUI he’s talking about the Scheme-to-C API mapping, which is not controlled by “the committee”. Thanks, Ludo’. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: A minor name improvement suggestion to Guile API 2009-12-17 15:00 ` Ludovic Courtès @ 2009-12-17 19:05 ` Keith Wright 2009-12-17 18:44 ` parawaiter 0 siblings, 1 reply; 8+ messages in thread From: Keith Wright @ 2009-12-17 19:05 UTC (permalink / raw) To: ludo; +Cc: guile-user > From: ludo@gnu.org (Ludovic =?iso-8859-1?Q?Court=E8s?=) > Cc: guile-user@gnu.org > >> > >> Yi DAI <plm.day@gmail.com> writes: > >> > >>> I'd like to suggest the following name convention > >>> which complies to names commonly found in > >>> assembly languages. > >>> > >>> > >>> - Replace < (less than) with *_lt*. > >>> - Replace > (greater than) with *_gt*. > >> > > well, the API is controlled by the scheme steering committee, > > AIUI he's talking about the Scheme-to-C API mapping, which is not > controlled by "the committee". Furthermore, I think the proposal has been over truncated. Original: > From: Yi DAI <plm.day@gmail.com> > reference manual: > > - Replace < (less than) with _less. > - Replace > (greater than) with _gr. > > I'd like to suggest the following name convention So as I understand, the proposal is: Currently, the variables that Scheme programmers know as |<| and |>|, are called in C _less and _gr. Propose to rename the C variables so that |<| and |>| are called in C *_lt* and and *_gt*, respectively. I'm not sure whether the asterisks are meant to be part of the name. This seems good to me; what was the programmer thinking that made it seem good to have asymmetric identifers for |<| and |>|? The names .LT. and .GT. go back to the Fortran of the late fifties. Whether it is worth the trouble to change, I don't know, but if I were planning to use this more than once or twice, I would rename it myself in my own code. -- Keith ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: A minor name improvement suggestion to Guile API 2009-12-17 19:05 ` Keith Wright @ 2009-12-17 18:44 ` parawaiter 2009-12-19 2:51 ` Yi DAI 0 siblings, 1 reply; 8+ messages in thread From: parawaiter @ 2009-12-17 18:44 UTC (permalink / raw) To: Keith Wright; +Cc: ludo, guile-user On Dec 17, 2009, at 20:05 , Keith Wright wrote: > Currently, the variables that Scheme programmers know as > |<| and |>|, are called in C _less and _gr. This probably also applies to things like string<? -> string_lt() (string_lt_p ?) > Propose to rename the C variables so that |<| and |>| > are called in C *_lt* and and *_gt*, respectively. > > I'm not sure whether the asterisks are meant to > be part of the name. I think that is an artifact of a multipart mime font-change. > This seems good to me; what > was the programmer thinking that made it seem > good to have asymmetric identifers for > |<| and |>|? The names .LT. and .GT. go back > to the Fortran of the late fifties. I don't see how _less and _gr was introduced with any consideration either, except just having to have a name as they were implemented. > Whether it is worth the trouble to change, > I don't know. That's the question, and I'll leave that to you guys. lt, gt, are universally known. Even /bin/sh 'test' uses this. And as far as scheme implementations go, I've probably found mine (for my kind of C-integrated usage) in chibi-scheme, so disregard my opinion as anything but an outsider. I'm just saying that I think the original poster has a point and his suggestion is what I would expect to find when searching for the C-equivalents of the given scheme predicates... Best regards, Lars J ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: A minor name improvement suggestion to Guile API 2009-12-17 18:44 ` parawaiter @ 2009-12-19 2:51 ` Yi DAI 2009-12-19 18:10 ` Linas Vepstas 0 siblings, 1 reply; 8+ messages in thread From: Yi DAI @ 2009-12-19 2:51 UTC (permalink / raw) To: parawaiter; +Cc: guile-user, ludo [-- Attachment #1: Type: text/plain, Size: 2833 bytes --] On Fri, Dec 18, 2009 at 2:44 AM, parawaiter <parawaiter@gmail.com> wrote: > On Dec 17, 2009, at 20:05 , Keith Wright wrote: > > Currently, the variables that Scheme programmers know as > > |<| and |>|, are called in C _less and _gr. > > This probably also applies to things like string<? -> string_lt() > (string_lt_p ?) > Yes, the `_less' and `_gr' convention is universal in the Scheme-to-C API according to the manual > > > Propose to rename the C variables so that |<| and |>| > > are called in C *_lt* and and *_gt*, respectively. > > > > I'm not sure whether the asterisks are meant to > > be part of the name. > > I think that is an artifact of a multipart mime font-change. > Right, that's done by Gmail. The original was in bold. > > > This seems good to me; what > > was the programmer thinking that made it seem > > good to have asymmetric identifers for > > |<| and |>|? The names .LT. and .GT. go back > > to the Fortran of the late fifties. > > I don't see how _less and _gr was introduced with any consideration either, > except just having to have a name as they were implemented. > > > Whether it is worth the trouble to change, > > I don't know. > > That's the question, and I'll leave that to you guys. lt, gt, are > universally known. Even /bin/sh 'test' uses this. And as far as scheme > implementations go, I've probably found mine (for my kind of C-integrated > usage) in chibi-scheme, so disregard my opinion as anything but an outsider. > I'm just saying that I think the original poster has a point and his > suggestion is what I would expect to find when searching for the > C-equivalents of the given scheme predicates... > > Perhaps a little statement about my reason to suggest this seeming minor (maybe not, according to Linas Vepstas) change. There are several Scheme implementations that provides a Scheme-to-C API, as far as I know, like Gambit-C, Chicken, PLT Scheme, (so does Bigloo?). I did not look into detail of their API, solely look through their name conventions (maybe this is not a good practice, but I don't think there's too much difference between these APIs, so the name convention does matter in a sense). I cannot say they did wrong, but not that beautiful really. And I finally came to Guile after being told by some others that it has the most beautiful API. I surprising found it was true, except these two names (`_less' and `_gr'), it's far less beautiful than `_lt' and `_gt', note that we do have `_leq' and `_geq' (I'd like further suggest change them to `_le' and `_ge') with symmetry. I admit I am a perfectionist. So this proposal may not worth it. Guile's API is beautiful, but if the mentioned four names (may affect others) are fixed, it would be perfect. Thanks. -- DAI Yi (代 毅) [-- Attachment #2: Type: text/html, Size: 3716 bytes --] ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: A minor name improvement suggestion to Guile API 2009-12-19 2:51 ` Yi DAI @ 2009-12-19 18:10 ` Linas Vepstas 0 siblings, 0 replies; 8+ messages in thread From: Linas Vepstas @ 2009-12-19 18:10 UTC (permalink / raw) To: Yi DAI; +Cc: guile-user, ludo 2009/12/18 Yi DAI <plm.day@gmail.com>: > (maybe not, according to Linas Vepstas) Sorry, I had misunderstood the nature of your email. Please disregard my earlier comments. -linas ^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2009-12-19 18:10 UTC | newest] Thread overview: 8+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2009-12-16 13:14 A minor name improvement suggestion to Guile API Yi DAI 2009-12-17 10:46 ` Ludovic Courtès 2009-12-17 14:26 ` Linas Vepstas 2009-12-17 15:00 ` Ludovic Courtès 2009-12-17 19:05 ` Keith Wright 2009-12-17 18:44 ` parawaiter 2009-12-19 2:51 ` Yi DAI 2009-12-19 18:10 ` Linas Vepstas
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).