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(".time a.tweet-timestamp")
 34var authorregex = regexp.MustCompile("twitter.com/([^/]+)")
 35
 36var re_hoots = regexp.MustCompile(`hoot: ?https://\S+`)
 37
 38func hooterize(noise string) string {
 39	seen := make(map[string]bool)
 40
 41	hootfixer := func(r io.Reader, url string) string {
 42		root, err := html.Parse(r)
 43		if err != nil {
 44			log.Printf("error parsing hoot: %s", err)
 45			return url
 46		}
 47		divs := tweetsel.MatchAll(root)
 48
 49		wantmatch := authorregex.FindStringSubmatch(url)
 50		if len(wantmatch) < 2 {
 51			log.Printf("no wanted author?")
 52		}
 53		wanted := wantmatch[1]
 54		var buf strings.Builder
 55
 56		var htf htfilter.Filter
 57		fmt.Fprintf(&buf, "%s\n", url)
 58		for _, div := range divs {
 59			twp := div.Parent.Parent.Parent
 60			alink := linksel.MatchFirst(twp)
 61			if alink == nil {
 62				log.Printf("missing link")
 63				continue
 64			}
 65			link := "https://twitter.com" + htfilter.GetAttr(alink, "href")
 66			authormatch := authorregex.FindStringSubmatch(link)
 67			if len(authormatch) < 2 {
 68				log.Printf("no author?")
 69				continue
 70			}
 71			author := authormatch[1]
 72			if author != wanted {
 73				continue
 74			}
 75			text := htf.TextOnly(div)
 76			text = strings.Replace(text, "\n", " ", -1)
 77			text = strings.Replace(text, "pic.twitter.com", "https://pic.twitter.com", -1)
 78
 79			if seen[text] {
 80				continue
 81			}
 82
 83			fmt.Fprintf(&buf, "> @%s: %s\n", author, text)
 84			seen[text] = true
 85		}
 86		return buf.String()
 87	}
 88
 89	hootfetcher := func(hoot string) string {
 90		url := hoot[5:]
 91		if url[0] == ' ' {
 92			url = url[1:]
 93		}
 94		url = strings.Replace(url, "mobile.twitter.com", "twitter.com", -1)
 95		log.Printf("hooterizing %s", url)
 96		req, err := http.NewRequest("GET", url, nil)
 97		if err != nil {
 98			log.Printf("error: %s", err)
 99			return hoot
100		}
101		req.Header.Set("User-Agent", "OpenBSD ftp")
102		req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8")
103		req.Header.Set("Accept-Language", "en-US,en;q=0.9")
104		resp, err := http.DefaultClient.Do(req)
105		if err != nil {
106			log.Printf("error: %s", err)
107			return hoot
108		}
109		defer resp.Body.Close()
110		if resp.StatusCode != 200 {
111			log.Printf("error getting %s: %d", url, resp.StatusCode)
112			return hoot
113		}
114		ld, _ := os.Create("lasthoot.html")
115		r := io.TeeReader(resp.Body, ld)
116		return hootfixer(r, url)
117	}
118
119	return re_hoots.ReplaceAllStringFunc(noise, hootfetcher)
120}