scope/search/web/google.go

/*
 *	Scope - A simple and minimal metasearch engine
 *	Copyright (C) 2021 Arsen Musayelyan
 *
 *	This program is free software: you can redistribute it and/or modify
 *	it under the terms of the GNU Affero General Public License as published by
 *	the Free Software Foundation, either version 3 of the License, or
 *	(at your option) any later version.
 *
 *	This program is distributed in the hope that it will be useful,
 *	but WITHOUT ANY WARRANTY; without even the implied warranty of
 *	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *	GNU Affero General Public License for more details.
 *
 *	You should have received a copy of the GNU Affero General Public License
 *	along with this program.  If not, see <https://www.gnu.org/licenses/>.
 */

package web

import (
	"net/http"
	"net/url"
	"strconv"

	"github.com/PuerkitoBio/goquery"
)

var googleURL = urlMustParse("https://www.google.com/search")

// Google represents the Google search engine
type Google struct {
	keyword   string
	userAgent string
	page      int
	doc       *goquery.Document
	initDone  bool
	baseSel   *goquery.Selection
}

// SetKeyword sets the keyword for searching
func (g *Google) SetKeyword(keyword string) {
	g.keyword = keyword
}

// SetPage sets the page number for searching
func (g *Google) SetPage(page int) {
	g.page = page * 10
}

// SetUserAgent sets the user agent for the request
func (g *Google) SetUserAgent(ua string) {
	g.userAgent = ua
}

// Init runs requests for the Google search engine
func (g *Google) Init() error {
	// Copy URL so that it can be changed
	initURL := copyURL(googleURL)
	// Get query parameters
	query := initURL.Query()
	// Set query
	query.Set("q", g.keyword)
	// Set starting result (page number)
	query.Set("start", strconv.Itoa(g.page))
	// Update URL query
	initURL.RawQuery = query.Encode()

	// Create new request for modified URL
	req, err := http.NewRequest(
		http.MethodGet,
		initURL.String(),
		nil,
	)
	if err != nil {
		return err
	}
	// If user agent empty, use default
	if g.userAgent == "" {
		g.userAgent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/74.0.3729.169 Safari/537.36"
	}
	// Set user agent of request
	req.Header.Set("User-Agent", g.userAgent)

	// Perform request
	res, err := http.DefaultClient.Do(req)
	if err != nil {
		return err
	}
	defer res.Body.Close()

	// Create goquery document from reader
	doc, err := goquery.NewDocumentFromReader(res.Body)
	if err != nil {
		return err
	}
	g.doc = doc
	g.baseSel = doc.Find(`a > h3`)
	g.initDone = true
	return nil
}

// Each runs eachCb with the index of each search result
func (g *Google) Each(eachCb func(int) error) error {
	for i := 0; i < g.baseSel.Length(); i++ {
		err := eachCb(i)
		if err != nil {
			return err
		}
	}
	return nil
}

// Title returns the title of the search result corresponding to i
func (g *Google) Title(i int) (string, error) {
	return get(g.baseSel, i).Text(), nil
}

// Link returns the link to the search result corresponding to i
func (g *Google) Link(i int) (string, error) {
	return get(g.baseSel, i).Parent().AttrOr("href", ""), nil
}

// Desc returns the description of the search result corresponding to i
func (g *Google) Desc(i int) (string, error) {
	return get(g.baseSel, i).Parent().Parent().Next().Text(), nil
}

// Name returns "google"
func (g *Google) Name() string {
	return "google"
}

// get gets an element and given index from given selection
func get(sel *goquery.Selection, i int) *goquery.Selection {
	return sel.Slice(i, i+1)
}

// Parse url ignoring error
func urlMustParse(urlStr string) *url.URL {
	out, _ := url.Parse(urlStr)
	return out
}

// copyURL makes a copy of the url and returns it
func copyURL(orig *url.URL) *url.URL {
	newURL := new(url.URL)
	*newURL = *orig
	return newURL
}