I'm not actually sure what he is starting with. (And I am not sure I would want to replace awk - it works fine for what it does.)
It might be fun, though, to design and write some J words and phrases to represent traditional unix concepts, and then wrap J in something (a shell script initially and a compiled executable later) that lets us use J efficiently in #! scripts. Thanks, -- Raul On Sat, Feb 15, 2014 at 6:21 AM, Linda Alvord <lindaalv...@verizon.net>wrote: > What I can't understand is how to match what he is starting with which is > not 11;13;17 > > Linda > > -----Original Message----- > From: programming-boun...@forums.jsoftware.com > [mailto:programming-boun...@forums.jsoftware.com] On Behalf Of linda > Sent: Saturday, February 15, 2014 5:13 AM > To: programm...@jsoftware.com > Subject: Re: [Jprogramming] awk-like J sentences? > > I speak only J. Once Raul translated awk I could understand what it was > doing. Here is how I would think of the problem as a "native" speaker of > J. > > ]A=:11;13;15 > ---T--T--┐ > │11│13│15│ > L--+--+--- > > >A > 11 13 15 > > > 2{.>A > 11 13 > > +/2{.>A > 24 > > ]B=: ":+/2{.>A > 24 > > B;' WEEKS' > ---T------┐ > │24│ WEEKS│ > L--+------- > > B,' WEEKS' > 24 WEEKS > > f=: 13 :'":+/2{.>y' > > f A > 24 > > f > [: ": [: +/ 2 {. > > > (f A),' WEEKS' > 24 WEEKS > > Linda > > > -----Original Message----- > From: programming-boun...@forums.jsoftware.com > [mailto:programming-boun...@forums.jsoftware.com] On Behalf Of Raul Miller > Sent: Friday, February 14, 2014 11:40 PM > To: Programming forum > Subject: Re: [Jprogramming] awk-like J sentences? > > One thing to be careful about - awk is designed for a unix command line. > When using J, you should assume a J command line. With a little work, you > can use a J program at the unix command line, but I am not really > comfortable with the way that works, yet. > > Meanwhile, at the unix command line you can work with stdin, hereris > scripts, or with files. Similarly, with J, you can work with function > arguments (vaguely like stdin, but right to left, instead of left to > right), with scripts, and with files. > > In all cases, the natural unit of computation, for this problem - just like > in awk - would be a single line. I've seen some other code here, but this > is pretty simple: > > BEGIN {FS=";"} > > becomes > > ';'&. > or > ,&':' > > Where in AWK you are giving the program a directive, here you are > physically incorporating the semicolon character in the line. > > Also, when coding J, it's good to give your code some test data, so you can > verify that it's working the way you like. So, here's a first attempt: > > ": +/ 2 {. 0&".;._2 ,&';' '11;13;17' > > 24 > > > I went with the ,&';' approach. if I had used ';'&, I would replace ;. _2 > with :. _1 in that expression. (This is the modifier which chops up the > line in "fields". > > > I went with 0&". to convert string representations of numbers to character > representation. This mimics a feature which is implicit in awk. > > > I did not directly mention the field numbers in my example. I probably > should have. To fix that replace 2 {. with 0 1 { (J has 0 for the first > element where AWK uses 1). Finally, +/ inserts + between the two values and > ": converts back to text. Quite likely converting back to text is an > unnecessary step, but that's what AWK is doing so I included it here. If > you want this as a named verb, you could go like this: > > > addtwo=:verb def '": +/0 1 { 0&".;._1 '';''&, y' > > addtwo '11;13;17' > > 24 > > > That's probably not completely intuitive, but I remember taking quite some > time before I was really comfortable working with awk. Everything takes > time to learn, and it's really easy to forget that after you know it well. > > > Once you have this, here's using it with an inline script (J's variation on > hereis documents): > > > addtwo;._2(0 :0) > > 1;2;3;4;5 > > 6;7;8;9 > > 10;11; > > ) > > > Executing this gave me the result: > > 3 > > 13 > > 21 > > > If you want to run against a file, replace (0 :0) with fread filename. > > > I hope this helps, > > > -- > > Raul > > > > > > > > > > On Fri, Feb 14, 2014 at 9:51 PM, Lee Fallat <ircsurfe...@gmail.com> wrote: > > > Hey there, > > > > As new user to J (but several years experience with C and Java), I > > find it very, very interesting. The power of its one liners and > > mathematical heritage really have me hooked. I was wondering though > > if it has similar capabilities as awk. What's the equivalent to this > > awk script in J?: > > > > BEGIN { FS=";" } > > { print $1+$2 } > > > > This script sets a FieldSeparator to ;, and then for every "row", add > > the first and second column and prints it. I would like to replace awk > > with J! > > > > Thank you, > > > > Lee > > > > P.S. Excuse me if I've misidentified J sentences. (Sentences -> > > statements?) > > ---------------------------------------------------------------------- > > For information about J forums see http://www.jsoftware.com/forums.htm > > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
