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 hootextractor(r io.Reader, url string, seen map[string]bool) string {
 39	root, err := html.Parse(r)
 40	if err != nil {
 41		log.Printf("error parsing hoot: %s", err)
 42		return url
 43	}
 44	divs := tweetsel.MatchAll(root)
 45
 46	var wanted string
 47	wantmatch := authorregex.FindStringSubmatch(url)
 48	if len(wantmatch) == 2 {
 49		wanted = wantmatch[1]
 50	}
 51	var buf strings.Builder
 52
 53	fmt.Fprintf(&buf, "%s\n", url)
 54	var htf htfilter.Filter
 55	for _, div := range divs {
 56		twp := div.Parent.Parent.Parent
 57		alink := linksel.MatchFirst(twp)
 58		if alink == nil {
 59			log.Printf("missing link")
 60			continue
 61		}
 62		link := "https://twitter.com" + htfilter.GetAttr(alink, "href")
 63		authormatch := authorregex.FindStringSubmatch(link)
 64		if len(authormatch) < 2 {
 65			log.Printf("no author?")
 66			continue
 67		}
 68		author := authormatch[1]
 69		if wanted == "" {
 70			wanted = author
 71		}
 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
 89func hooterize(noise string) string {
 90	seen := make(map[string]bool)
 91
 92	hootfetcher := func(hoot string) string {
 93		url := hoot[5:]
 94		if url[0] == ' ' {
 95			url = url[1:]
 96		}
 97		url = strings.Replace(url, "mobile.twitter.com", "twitter.com", -1)
 98		log.Printf("hooterizing %s", url)
 99		req, err := http.NewRequest("GET", url, nil)
100		if err != nil {
101			log.Printf("error: %s", err)
102			return hoot
103		}
104		req.Header.Set("User-Agent", "OpenBSD ftp")
105		req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8")
106		req.Header.Set("Accept-Language", "en-US,en;q=0.9")
107		resp, err := http.DefaultClient.Do(req)
108		if err != nil {
109			log.Printf("error: %s", err)
110			return hoot
111		}
112		defer resp.Body.Close()
113		if resp.StatusCode != 200 {
114			log.Printf("error getting %s: %d", url, resp.StatusCode)
115			return hoot
116		}
117		ld, _ := os.Create("lasthoot.html")
118		r := io.TeeReader(resp.Body, ld)
119		return hootextractor(r, url, seen)
120	}
121
122	return re_hoots.ReplaceAllStringFunc(noise, hootfetcher)
123}