Newer
Older
// Module to generate a bunch of CATALOG.json files which
// have arbitrary but realistic data
//
// TODO: generate some random text files and add them
// as well
const ORGANISATION = {
'id': 'https://examples.edu',
'name': 'Examples University'
};
const EMAIL_DOMAIN = 'examples.edu';
const OWNER = 'owner@examples.edu';
const APPROVER = 'approver@examples.edu';
const NAME_MIN = 3;
const NAME_MAX = 10;
const KEYWORD_MIN = 3;
const KEYWORD_MAX = 12;
const WORD_MIN = 2;
const WORD_MAX = 14;
const SENTENCE_MIN = 3;
const SENTENCE_MAX = 30;
const PARA_MIN = 1;
const PARA_MAX = 10;
const N_KEYWORD_MIN = 2;
const N_KEYWORD_MAX = 10;
const N_PEOPLE_MIN = 1;
const N_PEOPLE_MAX = 5;
const HONORIFICS = [ 'Dr', 'A/Prof', 'Prof', 'Dr', 'Dr', 'Dr', 'Mr', 'Ms' ];
const datacrate = require('./catalog.js');
const _ = require('lodash');
const fs = require('fs-extra');
const randdict = require('random-word');
const path = require('path');
const uuidv4 = require('uuid/v4');
const ArgumentParser = require('argparse').ArgumentParser;
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
function randrange(min, max) {
return Math.floor(Math.random() * (max - min)) + min;
}
function randoms(n, fn) {
return Array(n).fill(0).map(fn);
}
function randname() {
return _.upperFirst(randdict());
}
function randperson() {
const honorific = _.sample(HONORIFICS);
const surname = randname();
const givenname = randname();
const name = givenname + ' ' + surname;
const email = givenname + '.' + surname + '@' + EMAIL_DOMAIN;
const id = uuidv4();
return {
'dc:identifier': id,
'text_full_name': name,
'full_name_honorific': honorific + ' ' + name,
'email': email
}
}
function randkeyword() {
return randdict();
}
function randsentence() {
const nwords = randrange(SENTENCE_MIN, SENTENCE_MAX);
const s = randoms(nwords, randdict).join(' ') + '.';
return _.upperFirst(s);
}
function randtext() {
const nsentences = randrange(PARA_MIN, PARA_MAX);
return randoms(nsentences, randsentence).join(' ') + '\n';
}
function randdatapub(keywords, people) {
const k = _.sampleSize(keywords, randrange(N_KEYWORD_MIN, N_KEYWORD_MAX));
const title = _.startCase(_.camelCase(randsentence()));
const desc = randtext();
const creators = _.clone(_.sampleSize(people, randrange(N_PEOPLE_MIN, N_PEOPLE_MAX)));
const contributors = _.clone(creators);
const collabs = contributors.splice(1);
contributors[0].role = 'Chief Investigator';
return {
keywords: k,
contributor_ci: contributors[0],
contributors: collabs,
creators: creators,
title: title,
description: desc
}
}
function randdatapubs(n) {
const keywords = randoms(Math.floor(n / 2), randkeyword);
const people = randoms(n * 2, randperson);
return randoms(n, () => randdatapub(keywords, people))
}
async function makedatacrate(dest, datapub) {
const id = uuidv4();
await fs.ensureDir(path.join(dest, id));
const catalog = await datacrate.datapub2catalog({
id: id,
datapub: datapub,
organisation: ORGANISATION,
owner: OWNER,
approver: APPROVER
});
const catfile = path.join(dest, id, 'CATALOG.json');
await fs.writeFile(catfile, JSON.stringify(catalog, null, 2));
}
async function randomdatacrates(dest, n) {
const datapubs = randdatapubs(n);
Promise.all(
datapubs.map(async p => {
await makedatacrate(dest, p);
})
);
console.log("Done");
}
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
var parser = new ArgumentParser({
version: '1.0.0',
addHelp: true,
description: 'Generates a bunch of plausible random DataCrates'
});
parser.addArgument(
['-d', '--directory'],
{
help: "Directory in which to write DataCrates. Will be created if it doesn't exist",
defaultValue: './output/'
}
);
parser.addArgument(
['-n', '--number'],
{
help: 'Number of DataCrates to generate.',
type: 'int',
defaultValue: 10
}
);
var args = parser.parseArgs();
console.log(`Generating ${args['number']} random DataCrates in ${args['directory']}`);
randomdatacrates(args['directory'], args['number']);