wrange

An assignment.

Implemented in OCaml & Node.js, via BuckleScript; a decision which cost me quite a bit of time, about a year ago.

Installation

For ease of use, I've shipped compiled JavaScript sources directly in the Git repo. This would normally be a bad idea ... but I'm pretty sure the target audience hereof isn't likely to have an up-to-date OCaml compiler just sitting around. (In an ideal world, that would probably mean "resort to shipping a Dockerfile", but, again. time constraints.)

tl;dr:

# reasonably recent versions of node and npm are required
$ brew install node # or similar; you want >=v6
$ npm install --global npm # you want >= v5.2

$ git clone https://github.com/ELLIOTTCABLE/wrange.git
$ cd wrange
$ npm install
# or `npm install --global` to actually install the `wrange` command; see below

Usage

If you're unfamiliar with the Node ecosystem: you can run commands in the root directory of this project as if you'd installed it globally, by replacing the command-name with npx . in the following commands.

Usage is fairly-thoroughly documented via the built-in command-line help & manpage functionality. Try, for example,

$ wrange --help
# or `npx . --help`, for example

$ wrange print --help
$ wrange serve --help

For the printing-CLI, invocations that produce the requested outputs are as follows:

$ wrange print example_data/* \
   --sort gender,asc --sort last,asc --fields last,first,gender
Abbet, Fey, Female
Abelevitz, Flora, Female
Ablett, Emelina, Female
...

$ wrange print example_data/* \
   --sort birth,asc --fields last,first,birth
Kelsey, Sigismond, 1/5/1990
Waistell, Marys, 1/16/1990
Beall, Jozef, 1/17/1990
...

$ wrange print example_data/* \
   --sort last,desc --fields last,first
deLloyd, Agnola
deGlanville, Keen
deChastelain, Jackquelin
...

As for the REST server, start it with the serve command, and then feed it some requests. For example, if you have HTTPie installed,

$ wrange serve example_data/* &
Serving 3000 records;
Listening at http://127.0.0.1:3000
[1] 2530

$ http POST :3000/v1/records Content-Type:text/plain \
   <<< "Cable | Elliott | Male | Green | 1989-01-01"
HTTP/1.1 201 Created
{   "status": "success",
    "value": [ {
            "birthday": "1989-01-01T00:00:00.000Z",
            "favourite_colour": "Green",
            "first_name": "Elliott",
...

$ http GET :3000/v1/records/last/ascending | jq '.value[0:3] | .[] | .last_name'
"Abbet"
"Abelevitz"
"Ablett"

Contributing / Building

To make modifications to, or build from, ML sources; as long as you make no changes to the parser or lexer components (which you shouldn't need to, as they're very simple), the dependencies and scripts included in the package.json should get you pretty far out-of-the-box:

$ git clone https://github.com/ELLIOTTCABLE/wrange.git
$ cd wrange
$ npm install
# make your changes to, e.g. src/person.ml

$ npm run build
$ npm run test

If you want to modify parser.mly or lexer.mll, though, you'll need a full OCaml development environment, Menhir, and ocamllex. Once you've figured that out, uncomment the relevant entries in the "generators" section of bsconfig.json; thereafter, npm run build (which invokes bsb) will handle regeneration of parser.ml and lexer.ml.