| 6 Nov 2025 | |||
| [Coke] | No worries, appreciate all the fixes. | 21:11 | |
| arkiuat | so, the current doc says that method deepmap returns a list, but when I run the two examples in sub form, and call .^name against what they return, each example returns the same type it was given, an Array and a Hash respectively in this case | 21:21 | |
| I think this is one of those cases in which we want to say that what it returns is an Iterable | |||
| just like all those methods that have been kind of half-changed from documented as returning Lists to documented as returning Seqs | 21:22 | ||
| as in [Coke]'s problem-solving PR github.com/Raku/problem-solving/issues/499 | 21:23 | ||
| in fact, I'll add a comment to this effect to that issue before I move forward with this commit | 21:27 | ||
| Iterable is the most specific role that Array and Hash have in common | |||
| comment added | 21:37 | ||
| also, it would be easier to write this clearly if I just pretended that both forms of deepmap used &block for its Code argument (method uses &block, sub uses &op) | 21:38 | ||
| &reduce and method .reduce both use &with for its Code parameter | 21:39 | ||
| just for example | |||
| or I could follow the example of the reduce doc and just refer to it as "the argument", there we go | 21:42 | ||
| Geth | doc: arkiuat++ created pull request #4704: changing method to routine for deepmap |
22:13 | |
| arkiuat | i'll do duckmap and nodemap separately. This was trickier than I expected. | 22:14 | |
| [Coke] | (iterable) yup, makes sense | 23:46 | |
| 7 Nov 2025 | |||
| Geth | doc/main: e5f8fb00c5 | (Eric Forste)++ (committed using GitHub Web editor) | doc/Type/Any.rakudoc changing method to routine for deepmap, in type/Any (#4704) * changing method to routine for deepmap Adding documentation for the sub form of deepmap as per docs issue github.com/Raku/doc/issues/4560 * add missing semicolons |
00:02 | |
| doc/schultzdavid-patch-11: 323acff94a | (David Schultz)++ | doc/Language/py-nutshell.rakudoc fix output being outside of code block |
15:43 | ||
| doc/schultzdavid-patch-11: 5e6f26fc32 | (David Schultz)++ | doc/Language/py-nutshell.rakudoc change en dash to em dash |
|||
| doc/schultzdavid-patch-12: 7b005084fa | schultzdavid++ (committed using GitHub Web editor) | doc/Language/structures.rakudoc whitespace |
15:47 | ||
| 8 Nov 2025 | |||
| lizmat just notices mojibake at docs.raku.org/language/pod#Indexing_terms | 19:52 | ||
| [Coke]_ | pretty sure that's been reported. one se | 21:02 | |
| [Coke] | github.com/Raku/doc/issues/4613 | ||
| 9 Nov 2025 | |||
| lizmat notices use of superstitious parens in docs.raku.org/language/operators#infix_==%3E : my @array = (1, 2, 3, 4, 5); | 09:49 | ||
| that only works because of the single arg rule | |||
| librasteve | hmmm this is odd, the docs say docs.raku.org/language/variables#I...assignment my $b = 1,2,3; # item assignment to Scalar (same as preceding example) say $b; # OUTPUT: «1» | 18:09 | |
| but my testing has my $b = 1,2,3; say $b; # OUTPUT (1 2 3) | |||
| are the docs wrong? | 18:10 | ||
| oh - the docs are right - I did my testing in the repl, so maybe this is a repl issue? is it worthy of filing? if so, where? | 18:12 | ||
| arkiuat | however, if I do this in the REPL | 19:25 | |
| my $a = 1, 2, 3; .say for $a | |||
| I get output of just: | 19:26 | ||
| 1 | |||
| m: my $a = 1, 2, 3; .say for $a | |||
| camelia | WARNINGS for <tmp>: 1 Useless use of constant integer 3 in sink context (lines 1, 1) Useless use of constant integer 2 in sink context (lines 1, 1) |
||
| arkiuat | if I put the say (after a semicolon) on the same line as the assigment in the REPL, it works as documented | 19:29 | |
| only if you make the assignment in the REPL on a line by itself does it assign the whole list to the scalar | |||
| This *seems* like an intentional UI feature to save keystrokes | 19:32 | ||