exampleData/headlessRuns/wikipediaRun.json


1
2
3
4
5
6
7
8
9
10
11
12
13
14

/*
 * This run crawls Wikipedia, starting at the main page, going to a depth
 * of 2, with a total of 20 pages maximum crawled, and with a 1000 ms delay
 * between queries (for politeness).
 */
{
  'rulePath'  : '../ruleSets/',
  'crawlType' :    '2 20 1000 *wikipedia.org*',
  'runs': [
  { 'url': 'http://en.wikipedia.org/wiki/Main_Page', 'ruleSet': 'headingRules.json' },
  { 'url': 'http://en.wikipedia.org/wiki/Main_Page', 'ruleSet': 'miscRules.json' },
  { 'url': 'http://en.wikipedia.org/wiki/Main_Page', 'ruleSet': 'textRules.json' }
  ]
}