| [Coke] | should 'rak --ack -w someword' be expected to work? | 18:36 | |
| lizmat | "w" is not listed as a --ack argument | 18:39 | |
| what should it do ? | |||
| [Coke] | also: any way to ignore what .gitignore does? | ||
| "argument is a word" | |||
| lizmat | hmmm.. I could add it, but it wouldn't help you now | 18:40 | |
| rak §someword | 18:41 | ||
| [Coke] | -w, --word-regexp Force PATTERN to match only whole words | ||
| lizmat | would do what you want though | ||
| ah, so that would basically do the regexp, and see if the resulting match has word boundaries ? | 18:42 | ||
| Geth | App-Rak: coke++ created pull request #65: Update README.md |
||
| [Coke] | under the covers, don't know - end result is that -w word gives me hits on ' word ' but not ' someword ' | 18:43 | |
| lizmat | well, then §word would do the trick for you | 18:44 | |
| and I guess I could map -w to --type-words | |||
| [Coke] | I'm ok if it's hidden behind --ack | 18:45 | |
| I will just probably always use --ack because I've got a decade on it. | |||
| ... I still use 'ack -a' which hasn't been valid since the first big update. :) | |||
| lizmat | and that's supposed to do? | ||
| [Coke] | it was "all files" | 18:46 | |
| and then ack changed the default at some point, and decided that all was invalid, but I had so much muscle memory I still type it when it's an invalid option | |||
| (default now is text files that aren't ignorable by default, like .git) | 18:47 | ||
| which brings me to the second part of my question - is there an option to ignore files/dirs that are in .gitignore? | |||
| lizmat | perhaps --under-version-control ? | ||
| that would only search files that are under version control ? | 18:48 | ||
| ! | |||
| [Coke] | nice | 18:49 | |
| by default searches .xlsx file (and then dies because it's not valid utf8) | |||
| Sorry to hit with so many newb questions at the same time | 18:51 | ||
| "Error encoding UTF-8 string: could not encode codepoint 1190209 (0x1229A5), codepoint out of bounds. Cannot encode higher than 1114111 (0x10FFFF)" | 18:52 | ||
| (which then kills the rak that's running) | 18:53 | ||
| lizmat | that it tries it at all, is because it looks like a text file oddly enough | 18:55 | |
| according to Data::TextOrBinary semantics | 18:56 | ||
| [Coke] | weird. xlsx is basically a zip file of xml files. | ||
| lizmat | interesting, care to make an issue for it ? | 18:57 | |
| Geth | App-Rak/main: ba4f8db892 | (Will Coleda)++ (committed using GitHub Web editor) | README.md Update README.md (#65) Fix typo |
||
| lizmat | looks like --ack support is really below par... sorry | 18:58 | |
| am in the middle of some other stuff, deep-diving in App::Rak is currently not an option :-( | 18:59 | ||
| Geth | App-Rak/main: 9f5fbce088 | (Elizabeth Mattijsen)++ | 2 files Add "w" in --ack mode, and some doc fixes |
19:01 | |
| [Coke] | sure I'll see if I can make it on Data::TextOrBinary | 19:03 | |
| ... Data::TextOrBinary is not already installed. but I was running rak on the command line | 19:09 | ||
| lizmat | ah, yes, I incorporated that into path::utils, I believe | 19:10 | |
| [Coke] | doesn't look like that's actually a dep? | ||
| ok. | |||
| lizmat | or somewhere hmmm... | 19:11 | |
| raku.land/zef:lizmat/path-utils#path-is-text | 19:12 | ||
| [Coke] | path-is-text returns 0 on that file | ||
| lizmat | ah.... weird | 19:13 | |
| [Coke] | my command line was 'rak --under-version-control rc' - does that option override text? | ||
| lizmat | ah, yes... try adding --is-text ? | 19:14 | |
| [Coke] | much better. | 19:15 | |
| Probably LTA that trying to search the binary file explodes rak instead of ignoring/warning on the file. | |||
| lizmat | yeah, I think there is an issue for that already | ||
| if not, there should be one :-) | |||
| [Coke] | heard. ;) | 19:17 |