1//
  2// Copyright (c) 2019 Ted Unangst <tedu@tedunangst.com>
  3//
  4// Permission to use, copy, modify, and distribute this software for any
  5// purpose with or without fee is hereby granted, provided that the above
  6// copyright notice and this permission notice appear in all copies.
  7//
  8// THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
  9// WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
 10// MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
 11// ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 12// WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
 13// ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 14// OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 15
 16package main
 17
 18import (
 19	"fmt"
 20	"io"
 21	"log"
 22	"net/http"
 23	"os"
 24	"regexp"
 25	"strings"
 26
 27	"github.com/andybalholm/cascadia"
 28	"golang.org/x/net/html"
 29	"humungus.tedunangst.com/r/webs/htfilter"
 30)
 31
 32var tweetsel = cascadia.MustCompile("p.tweet-text")
 33var linksel = cascadia.MustCompile("a.tweet-timestamp")
 34var replyingto = cascadia.MustCompile(".ReplyingToContextBelowAuthor")
 35var imgsel = cascadia.MustCompile("div.js-adaptive-photo img")
 36var authorregex = regexp.MustCompile("twitter.com/([^/]+)")
 37
 38var re_hoots = regexp.MustCompile(`hoot: ?https://\S+`)
 39var re_removepics = regexp.MustCompile(`pic\.twitter\.com/[[:alnum:]]+`)
 40
 41func hootextractor(r io.Reader, url string, seen map[string]bool) string {
 42	root, err := html.Parse(r)
 43	if err != nil {
 44		log.Printf("error parsing hoot: %s", err)
 45		return url
 46	}
 47
 48	url = strings.Replace(url, "mobile.twitter.com", "twitter.com", -1)
 49	wantmatch := authorregex.FindStringSubmatch(url)
 50	var wanted string
 51	if len(wantmatch) == 2 {
 52		wanted = wantmatch[1]
 53	}
 54
 55	var htf htfilter.Filter
 56	htf.Imager = func(node *html.Node) string {
 57		return fmt.Sprintf(" <img src='%s'>", htfilter.GetAttr(node, "src"))
 58	}
 59
 60	var buf strings.Builder
 61	fmt.Fprintf(&buf, "%s\n", url)
 62
 63	divs := tweetsel.MatchAll(root)
 64	for i, div := range divs {
 65		twp := div.Parent.Parent.Parent
 66		link := url
 67		alink := linksel.MatchFirst(twp)
 68		if alink == nil {
 69			if i != 0 {
 70				log.Printf("missing link")
 71				continue
 72			}
 73		} else {
 74			link = "https://twitter.com" + htfilter.GetAttr(alink, "href")
 75		}
 76		replto := replyingto.MatchFirst(twp)
 77		if replto != nil {
 78			continue
 79		}
 80		authormatch := authorregex.FindStringSubmatch(link)
 81		if len(authormatch) < 2 {
 82			log.Printf("no author?: %s", link)
 83			continue
 84		}
 85		author := authormatch[1]
 86		if wanted == "" {
 87			wanted = author
 88		}
 89		if author != wanted {
 90			continue
 91		}
 92		for img := imgsel.MatchFirst(twp); img != nil; imgsel.MatchFirst(twp) {
 93			img.Parent.RemoveChild(img)
 94			div.AppendChild(img)
 95		}
 96		text := htf.NodeText(div)
 97		text = strings.Replace(text, "\n", " ", -1)
 98		text = re_removepics.ReplaceAllString(text, "")
 99
100		if seen[text] {
101			continue
102		}
103
104		fmt.Fprintf(&buf, "> @%s: %s\n", author, text)
105		seen[text] = true
106	}
107	return buf.String()
108}
109
110func hooterize(noise string) string {
111	seen := make(map[string]bool)
112
113	hootfetcher := func(hoot string) string {
114		url := hoot[5:]
115		if url[0] == ' ' {
116			url = url[1:]
117		}
118		url = strings.Replace(url, "mobile.twitter.com", "twitter.com", -1)
119		log.Printf("hooterizing %s", url)
120		req, err := http.NewRequest("GET", url, nil)
121		if err != nil {
122			log.Printf("error: %s", err)
123			return hoot
124		}
125		req.Header.Set("User-Agent", "Bot")
126		req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8")
127		req.Header.Set("Accept-Language", "en-US,en;q=0.9")
128		resp, err := http.DefaultClient.Do(req)
129		if err != nil {
130			log.Printf("error: %s", err)
131			return hoot
132		}
133		defer resp.Body.Close()
134		if resp.StatusCode != 200 {
135			log.Printf("error getting %s: %d", url, resp.StatusCode)
136			return hoot
137		}
138		ld, _ := os.Create("lasthoot.html")
139		r := io.TeeReader(resp.Body, ld)
140		return hootextractor(r, url, seen)
141	}
142
143	return re_hoots.ReplaceAllStringFunc(noise, hootfetcher)
144}