> On Aug 2, 2015, at 5:25 PM, Greg Reagle <greg.rea...@umbc.edu> wrote: > > On Sat, Aug 01, 2015 at 01:38:02PM -0600, @lbutlr wrote: >> I was hoping the list was more active > > FYI, the GitHub site at https://github.com/fish-shell/fish-shell/issues is > very > active with bug reports and feature requests.
Yes, but I am looking more for “How-to” or “getting -started”. I haven’t used it enough to request anything or even know if something might be a bug. > Anyway, is there any sort of writeup anyone’s done on something like “fish >> for bash users” or similar? > > The comparison at http://hyperpolyglot.org/unix-shells is good. Also, have > you > read the FAQ at http://fishshell.com/docs/current/faq.html and the tutorial at > http://fishshell.com/docs/current/tutorial.html? I’ve read the tutorial, and skimmed the FAQ. >> fish seems to have some different ideas about how to do things (preferring >> functions in individual files over aliases or functions in a single rcfile >> for one) > > That's the default way, but you don't have to do it that way. You can have > all > of your functions in one file if you like. You could put them all in > config.fish. Fish supports the source command, so you can group them into > files however you like. Right. I meant to say that it might be better to try fish “the fish way” as opposed to just trying to duplicate my bash setup. >> Also, I’m not sure what the advantage is to creating a whole lot of files >> each containing one to three lines in the functions folder. > > My guess is: The functions can be auto-loaded on first use that way, a > performance gain over loading them all at start-up no matter whether they get > used. That makes sense. >> function gbz () { >> for i in `ls | grep ".bz2"` ; do bzgrep -H "$*" $i; done >> } >> >> If I am understanding the documentation, this would be rewritten into >> …/functions/gbz.fish as >> >> function gbz >> for i in *.bz2 >> bzgrep -H $argv $i >> end >> end > > That looks right to me, but I haven't checked it. FYI, you can have the for > loop in fish in one line if you'd like--use semicolons. E.g. > > for i in (seq 5) ; echo $i; end > >> Yes? How to I pass all the args properly escaped or quoted? Just “$argv” like >> in the original? > > In fish, you just use $argv unquoted. This is exemplified in the tutorial. That’s what i thought, but if I’m using bash and type % gbz text1 text2 I get results in the bz2 files for “text1 text2” while with fish I get: > gbz text1 text2 bzip2: Can't open input file text2: No such file or directory. and then it greps the files for text1. > >> What if I wanted to check that there was only one argv? Or similarly what if >> I wanted to parse argv[1] and argv[2] separately? > > Use count. Again, see the tutorial. Example > for i in (seq (count $PATH)); echo $i : $PATH[$i]; end Ah. The tutorial just has count $PATH to count the number of elements in the $Patch variable. the reliance on web to config some settings is a bit of an issue. A lot of my work in the shell is to remote machines, so setting up Things via a web interface simply doesn’t work. -- You start a conversation you can't even finish it You're talkin' a lot, but you're not sayin' anything When I have nothing to say, my lips are sealed Say something once, why say it again? ------------------------------------------------------------------------------ _______________________________________________ Fish-users mailing list Fish-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fish-users
