[00:02] *** lucs_ is now known as lucs [00:50] *** oodani left [00:50] *** oodani joined [00:54] Matching nested "bracketed" strings question: https://gist.github.com/lucs/fd59d6d5c513a21dc3528bac6ca9505f [00:57] (I tried posing this in stackoverflow, but it wants info I consider irrelevant, f it. [01:05] *** hulk joined [01:06] *** kylese left [01:11] Which is sad, because raiph, absent here, often has great insights.) [01:20] Jeebus, my question shows wrong stuff. Deleting, will rework. [02:01] *** kylese joined [02:03] *** hulk left [02:15] *** kylese left [02:17] *** kylese joined [05:12] *** Sgeo left [05:14] *** Sgeo joined [05:41] *** kjp left [05:46] *** kjp joined [07:31] *** dicit left [08:28] *** belluzj joined [08:33] *** belluzj left [08:44] *** Sgeo left [12:59] @lucs Please, post the SO link at some point. [13:44] Turns out I didn't need to post. I asked an AI assistant to show me how to do what I was looking for and got a good answer. [13:45] I wanted a regex to match possibly nested arbitrarily bracketed strings. [13:46] Cleaned up example solution: [13:48] my token beg { '{-' } my token end { '-}' } my token haskl-cmnt { [ <&haskl-cmnt> | | > . ]* } [14:30] *** vasko4535586160 joined [14:30] *** vasko453558616 left [14:30] *** vasko4535586160 is now known as vasko453558616 [14:38] *** vasko4535586169 joined [14:40] *** vasko453558616 left [14:40] *** vasko4535586169 is now known as vasko453558616 [15:20] *** abraxxa joined [15:38] *** vasko4535586163 joined [15:39] *** vasko453558616 left [15:39] *** vasko4535586163 is now known as vasko453558616 [15:43] *** vasko4535586167 joined [15:43] *** vasko453558616 left [15:43] *** vasko4535586167 is now known as vasko453558616 [15:45] *** oodani left [15:46] *** oodani joined [16:09] *** human-blip left [16:11] *** human-blip joined [16:15] @scullucs that doesn't appear to be the Raku way, which uses FAILGOAL. Search for FAILGOAL on this page and also read the section that follows on Recursive Regexes: https://docs.raku.org/language/regexes [16:20] *** stanrifkin joined [16:21] Does the Statistics module work for anyone? [16:23] It fails to compile here with `[Statistics] Failed to get the directory contents of '/tmp/.zef.1791648945.3849788/0b1db89f7e634edbc8e8836496743ed02191b1f4.tar.gz/dist/src':' [16:24] confirmed [16:24] OK [16:25] guess it's time for an issue [16:25] *** MasterDuke joined [17:03] *** stanrifkin left [17:16] *** stanrifkin joined [17:25] *** Sgeo joined [17:34] *** stanrifkin left [17:48] I don't use "Statistics", I use "Stats". [17:50] "Statistics" always seemed to me "non standard" and somewhat cumbersome. [17:51] *** MasterDuke left [17:51] Oh, yeah, uses stupid R names! [18:30] *** oodani left [18:31] *** oodani joined [18:37] I also use Stats, long identifiers are annoying to read and to write [18:46] *** stanrifkin joined [19:06] *** johnjay left [19:39] Do tell which one is clearer: UniformDistribution or runif ? [19:46] *** stanrifkin left [19:57] UniformDist would be more sufficient in the context [19:57] I suppose runif is taken from Dist(r)ibution(Unif)orm, interesting choice of letter [19:57] normal people would pick first letter [19:58] actually, in the code I write I'd even prefer DistUniform [19:58] because Dist would give me choices [20:28] In R the is a family of uniform distributions: runif, dunif, etc. The letter "r" in runif stands for "random points" or something... [20:32] In the package "Statistics::Distributions" the distributions are defined in the way you prefer, e.g. Statistics::Distributions::Defined::Benford but are exported in CamelCase I mentioned, i.e. Benford::Distribution. [20:35] kinda backwards imo [20:36] in fact, with `Distribution::Benford` if function or method expected distribution, one could only say `distribution=Benford` or something, skipping the package/module/class whatever [20:37] sadly no language I know is able to do this [20:40] The main use case for those distributions is code of the form: random-variate(BenfordDistribution.new(:p(0.3)), 1000) [20:40] For the discrete distributions this code makes sense: random-variate(BernoulliDistribution.new(:p(0.3)), 1000).BagHash.Hash; [20:43] But what you say/describe can be supported for the built in distributions of that package. For example, as: random-variate(dist => 'bernoulli', :p(0.3), size => 1000) [20:44] But that requires more explanations / descriptions. [20:54] *** johnjay joined [21:04] *** abraxxa left [21:19] *** erooke left [21:29] *** erooke joined [22:16] *** vrurg_ joined [22:17] *** vrurg left [22:21] Oh, interesting. Thanks for the tip, will look those up. [22:24] (Will do later. Leaving for a concert: The Rite of Spring, arranged for wind quintet!) [22:25] https://pentaedre.com/en