Perl 6 - the future is here, just unevenly distributed

IRC log for #rosettacode, 2014-08-08

| Channels | #rosettacode index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:01 mwn3d1 joined #rosettacode
00:01 mwndsphonesucks joined #rosettacode
01:37 bug-magnet joined #rosettacode
01:46 ilbot3 joined #rosettacode
01:46 Topic for #rosettacode is now Rosetta Code: Coding in  (∞ - 1) languages ; http://irclog.perlgeek.de/rosettacode/today ; Pop  a tab and stick around! ; Account creation issues? Send an email with  your desired username to bureaucrats@rosettacode.org.
03:42 mwn3d joined #rosettacode
06:31 bug-magnet joined #rosettacode
09:35 bug-magnet joined #rosettacode
10:25 mikemol joined #rosettacode
14:33 bug-magnet joined #rosettacode
15:37 mwn3d1 joined #rosettacode
16:33 mwn3d joined #rosettacode
17:30 dchapes Does anyone know if there is a pre-existing rosetta task for demonstrating build constraints and/or things like #ifdef __GNUC__ ?
17:35 TimToady hmm, those sound like they're more about compilers than about languages
17:35 TimToady RC is more about languages than compilers...
17:35 dchapes Some languages do define it tho
17:35 ivanshmakov TimToady: This is rather about /dialects,/ as implemented by particular compilers.
17:35 dchapes I was just curious, I don't intend to add it if it doesn't exist :)
17:36 TimToady a more overarching concept would be the notion of 'pragmas'
17:36 TimToady how does the language provide compiler hints in general?
17:37 TimToady specific compiler hints are probably outside the purview of RC
17:37 TimToady unless related to a particular task
17:40 dchapes I kind of figured that was the case, I thought about this in relation to http://rosettacode.org/wiki/Read_entire_file where they mention memory mapping, which for some languages could be optionally done iff a mmap syscall is available.
17:43 TimToady sure, that's fine for a specific task, though you might want a 'works with' to say which compiler supports it, though depending on how you write the thing, the #ifdef is serving as a kind of internal 'works with', I suppose
17:43 TimToady I don't see any problem with that
17:43 TimToady #ifdef __GNUC__ is still valid C, after all :)
17:44 dchapes I was thinking about the Go example where the language defines a specific method of build constraints that must be supported. E.g. "// +build !windows" at the top of the file.
17:44 TimToady even on non-gnu compilers
17:45 * TimToady is not familiar enough with Go to know exactly what that says, but again, if it's in support of a particular task, and informative of what will or will not work, it seems acceptable
17:47 TimToady we do have some more descriptive tasks, "tell how you would do this", so certainly something along those lines might be acceptable; can always add a draft task and get opinions
17:48 TimToady we just tend to shy away from tasks that are specific to any particular implementation
17:51 TimToady basically, we have two categories of languages, those that don't mind saying 'omit' for a feature, and those that do, because they want to be completist (including, presumably, most of those languages that have done the majority of tasks)
17:51 dchapes btw, while I have your attention :) … I'm new to rosettacode but I've added a few Go sections over the last few days and I've done the wikipedia "be bold" kinda thing by including links to the "Go Playground" when I felt it's useful, for example here: http://rosettacode.org/wiki/Set_of_real_numbers#Go
17:51 TimToady the contributors of those languages don't want to see tasks proposed that have nothing to do with their language :)
17:52 dchapes Is that acceptable and desirable? To include a link to a site that has the whole example in a web-runnable thingy?
17:53 TimToady usually such links are put into the description of the language rather than individual tasks, but we don't prohibit informative links on the tasks
17:54 dchapes I was thinking that since the playground link can link to the specific example with the code already there it avoids a copy-n-paste.
17:54 TimToady spam links, of course, are another matter
17:54 dchapes e.g. in the above the link is http://play.golang.org/p/YQ2GRBM4af  rather than just a link to play.golang.org.
17:55 TimToady well, that might also be a generalizable feature, since lots of people might want to copy the entries out to some such site
17:55 dchapes Hmm... interesting, something like an extended {{out}} template that knew how to make those links for languages that support it do you mean?
17:56 TimToady another problems is that, since this is a wiki, someone can change the RC version but not the YQ2GRBM4af version
17:56 TimToady and that would cause vast confustion
17:56 TimToady not to mention confusion...
17:56 dchapes Yes, that did concern me. I was hoping Go people would know enough to at minimum remove the playground link in such cases.
17:57 dchapes Anyway, I was just wondering if I should continue with the playground links or stop making them?
17:58 TimToady well, the idea is sound, even if the implementation is flawed
17:58 TimToady would want to think about the security implications of doing anything more active, as well
17:58 TimToady not that you can prevent people from copying whatever they like into the playground anyway...
17:59 dchapes Yeah, someone reviewing my submissions would either have to take my word for it or double check that the playground link doesn't include some arbitrary Go program text.
17:59 TimToady so some mechanism that somehow copied the current entry into the playground would be less confusing, presuming it isn't broken
17:59 TimToady if the current entry is broken, then it'd be more confusing :)
18:01 TimToady another problem is if the playground goes away; in my experience, such websites tend to rot
18:02 dchapes So far the go team has been really good about improving and keeping it up to date. (btw that link is run by the same people doing go releases, not some random third party)
18:03 dchapes I think it makes sense if I update the Category:Go page to include a brief description of and link /link to the playground and stop adding new specific playground links.
18:04 TimToady maybe someone could do a mashup that would copy/paste for you
18:04 dchapes Opps, I just realized I never read the (rather large) category_talk:Go page, watch this be discussed already in there :-D
18:05 ivanshmakov dchapes: I don’t know about RC in particular, but per my experience, the primary purpose of the Category: pages is to tell what belongs to the category, – not to provide some information on the topic.
18:06 ivanshmakov dchapes: That is: one gets to the Category: page to check whether some other page should or should not be in the category.
18:07 TimToady I don't think RC makes a strict distinction of metadata from data like that :)
18:07 dchapes On RC the Category:C, Category:Prolong, etc pages appear to be descriptions of each programming language and are all in http://rosettacode.org/wiki/C​ategory:Programming_Languages
18:08 TimToady RC is a social construct, not a mathematical construct :)
18:08 ivanshmakov dchapes, TimToady: As I’ve said, – I’m not familiar with RC per se.
18:10 ivanshmakov But generally, my adivce would be to use, say, /wiki/Stay (or /wiki/Language:Stay) for the description of the Stay language, – and have /that/ page belong to the eponymous category, too.
18:11 dchapes I just discovered that wiki/C, wiki/Go, wiki/<your programming language here> are just redirected to wiki/Category:C,Go,<yourlang> :)
18:12 ivanshmakov dchapes: It certainly differs from the way followed at, say, Wikibooks.
18:13 ivanshmakov Speaking of Wikibooks, – https://en.wikibooks.org/wi​ki/Algorithm_Implementation.
18:13 dchapes I can see why they do it that way at RC though, they combine the description, links and list of tasks in a language all in one place.
19:43 ttmrichter joined #rosettacode
21:08 mwn3d1 joined #rosettacode
21:27 mwn3d joined #rosettacode
23:13 mwn3d1 joined #rosettacode

| Channels | #rosettacode index | Today | | Search | Google Search | Plain-Text | summary