crawler

Documentation

Overview

Package crawler is a distributed web crawler.

Index

• Variables
• type Backend
• type Crawler
  • func New(cfg config.Provider) *Crawler
  • func (c *Crawler) Close()
  • func (c *Crawler) Start(t time.Duration) error
• type ElasticSearch
  • func (e *ElasticSearch) CrawledAndCount(u, domain string) (time.Time, int, error)
  • func (e *ElasticSearch) Upsert(doc *document.Document) error
• type Stats
  • func (s *Stats) Elapsed() *Stats
  • func (s *Stats) String() string
  • func (s *Stats) Update(code int)
• type UserAgent

Variables

var RobotsPath, _ = url.Parse("/robots.txt")

RobotsPath is robots.txt path

Types

type Backend

type Backend interface {
	Setup() error
	CrawledAndCount(u, domain string) (time.Time, int, error) // gotta be a better name for this
	Upsert(*document.Document) error
}

Backend outlines methods to save documents and count the docs a domain has

type Crawler

type Crawler struct {
	HTTPClient *http.Client
	UserAgent

	Robots robots.Cacher
	Queue  queue.Queuer

	Backend
	// contains filtered or unexported fields
}

Crawler holds crawler settings for our UserAgent, Seed URLs, etc.

func New

func New(cfg config.Provider) *Crawler

New creates a Crawler from a config Provider

func (*Crawler) Close

func (c *Crawler) Close()

Close the crawler

func (*Crawler) Start

func (c *Crawler) Start(t time.Duration) error

Start the crawler

type ElasticSearch

type ElasticSearch struct {
	*document.ElasticSearch
	Bulk *elastic.BulkProcessor
	sync.Mutex
}

ElasticSearch satisfies the crawler's Backend interface

func (*ElasticSearch) CrawledAndCount

func (e *ElasticSearch) CrawledAndCount(u, domain string) (time.Time, int, error)

CrawledAndCount returns the crawled date of the url (if any) and the total number of links a domain has

func (*ElasticSearch) Upsert

func (e *ElasticSearch) Upsert(doc *document.Document) error

Upsert updates a document or inserts it if it doesn't exist NOTE: Elasticsearch has a 512-byte limit on an insert operation. Upsert does not have that limit.

type Stats

type Stats struct {
	sync.Mutex
	Start time.Time

	StatusCodes map[int]int64
	// contains filtered or unexported fields
}

Stats keeps track of time elapsed & status codes

func (*Stats) Elapsed

func (s *Stats) Elapsed() *Stats

Elapsed will set the total time the crawler has been running

func (*Stats) String

func (s *Stats) String() string

Print our stats into human-readable

func (*Stats) Update

func (s *Stats) Update(code int)

Update our stats from a document's results

type UserAgent

type UserAgent struct {
	Full  string
	Short string
}

UserAgent holds the full and short version of the crawler's useragent