func main() {
if len(os.Args) < 2 {
log.Fatalf("Usage: %s TOPIC", os.Args[0])
}
url := "http://l1sp.org/" + os.Args[1]
request, err := http.NewRequest("GET", url, nil)
if err != nil {
log.Fatal(err)
}
client := &http.Client{}
resp, err := client.Do(request)
if err != nil {
log.Fatal(err)
}
defer resp.Body.Close()
contents, err := ioutil.ReadAll(resp.Body)
if err != nil {
log.Fatal(err)
}
text, err := html2text.FromString(string(contents))
if err != nil {
log.Fatal(err)
}
fmt.Println(text)
}
func (bt *berthaTransformer) authorToPerson(a author) (person, error) {
uuid := uuid.NewMD5(uuid.UUID{}, []byte(a.TmeIdentifier)).String()
plainDescription, err := html2text.FromString(a.Biography)
if err != nil {
return person{}, err
}
altIds := alternativeIdentifiers{
UUIDS: []string{uuid},
TME: []string{a.TmeIdentifier},
}
p := person{
Uuid: uuid,
Name: a.Name,
PrefLabel: a.Name,
EmailAddress: a.Email,
TwitterHandle: a.TwitterHandle,
FacebookProfile: a.FacebookProfile,
LinkedinProfile: a.LinkedinProfile,
Description: plainDescription,
DescriptionXML: a.Biography,
ImageUrl: a.ImageUrl,
AlternativeIdentifiers: altIds,
}
return p, err
}
func (self *Wiki) IndexArticle(path string, article *Article) error {
html, err := article.GetHtml()
if err != nil {
return err
}
text, err := html2text.FromString(html)
if err != nil {
return err
}
self.Index.Index(path, ArticleData{Name: article.Name, Content: text})
return nil
}
func completeTags(episode *Episode) {
logger.Debug.Println("Tag update : " + episode.Podcast.feedPodcast.Title + " - " + episode.feedEpisode.Title + " : " + episode.file())
tag, err := taglib.Read(episode.file())
if err != nil {
logger.Warning.Println("Cannot complete episode tags for "+episode.Podcast.feedPodcast.Title+" - "+episode.feedEpisode.Title, err)
return
}
defer tag.Close()
var replaceArtist string
if episode.feedEpisode.Author.Name != "" {
replaceArtist = episode.feedEpisode.Author.Name
} else {
replaceArtist = episode.Podcast.feedPodcast.Title
}
//use the podcast title for now
replaceArtist = episode.Podcast.feedPodcast.Title
completeTag(taglib.Artist, replaceArtist, tag)
completeTag(taglib.Album, episode.Podcast.feedPodcast.Title, tag)
plaintextDescription, err := html2text.FromString(episode.feedEpisode.Description)
if err == nil {
episode.feedEpisode.Description = plaintextDescription
}
if len(episode.feedEpisode.Description) > maxCommentSize+5 {
episode.feedEpisode.Description = episode.feedEpisode.Description[:maxCommentSize] + " ..."
}
completeTag(taglib.Comments, episode.feedEpisode.Description, tag)
completeTag(taglib.Title, episode.feedEpisode.Title+" "+episode.formattedPubDate(dateFormat), tag)
completeTag(taglib.Genre, "Podcast", tag)
pubdate, err := episode.feedEpisode.ParsedPubDate()
if err == nil {
completeTag(taglib.Year, strconv.Itoa(pubdate.Year()), tag)
}
logger.Debug.Println("Tag Write Start for : " + episode.file())
err = tag.Save()
//setAlbumArtist(episode.Podcast.feedPodcast.Title, episode.file())
logger.Debug.Println("Tag Write End for : " + episode.file())
if err != nil {
logger.Warning.Println(episode.Podcast.feedPodcast.Title+" - "+episode.feedEpisode.Title+" : Cannot save the modified tags", err)
}
logger.Debug.Println("Tag update END : " + episode.Podcast.feedPodcast.Title + " - " + episode.feedEpisode.Title)
}
//formatMessage make into text if contains html
func formatMessage(message string) string {
isHTML := strings.Contains(message, "<html>")
if isHTML {
text, err := html2text.FromString(message)
if err != nil {
log.Println("failed to convert to text " + err.Error())
}
return text
}
return message
}
// NewMessageFrom creates new mail message object with custom From header.
func NewMessageFrom(to []string, from, subject, htmlBody string) *Message {
msg := gomail.NewMessage()
msg.SetHeader("From", from)
msg.SetHeader("To", to...)
msg.SetHeader("Subject", subject)
msg.SetDateHeader("Date", time.Now())
body, err := html2text.FromString(htmlBody)
if err != nil {
log.Error(4, "html2text.FromString: %v", err)
msg.SetBody("text/html", htmlBody)
} else {
msg.SetBody("text/plain", body)
if setting.MailService.EnableHTMLAlternative {
msg.AddAlternative("text/html", htmlBody)
}
}
return &Message{
Message: msg,
}
}
func (c *Context) View(i int) {
var email protocol.Encrypted
request, err := http.NewRequest("GET", fmt.Sprintf("%v/inbox/%v", c.host, i), nil)
request.SetBasicAuth("user", c.password)
response, err := c.client.Do(request)
if err != nil {
log.Fatal(err)
}
if response.StatusCode == http.StatusOK {
data, err := ioutil.ReadAll(response.Body)
response.Body.Close()
if err != nil {
log.Fatal(err)
}
err = proto.Unmarshal(data, &email)
if err != nil {
log.Fatal(err)
}
key, err := rsa.DecryptPKCS1v15(rand.Reader, &c.key, email.Key)
if err != nil {
log.Fatal(err)
}
cipher, err := aes.NewCipher(key)
if err != nil {
log.Fatal(err)
}
cipher.Decrypt(email.Data, email.Data)
decrypted := protocol.Email{}
err = proto.Unmarshal(email.Data, &decrypted)
if err != nil {
log.Fatal(err)
}
process := func(mediaType string, body []byte) {
fmt.Println(mediaType)
if mediaType == "text/plain" {
fmt.Println(string(body))
} else if mediaType == "text/html" {
text, err := html2text.FromString(string(body))
if err != nil {
log.Fatal(err)
}
fmt.Println(text)
}
}
message, err := goemail.ParseMessage(strings.NewReader(*decrypted.Mail))
if err != nil {
log.Fatal(err)
}
if message.HasBody() {
mediaType, _, _ := message.Header.ContentType()
process(mediaType, message.Body)
} else {
for _, part := range message.MessagesAll() {
mediaType, _, err := part.Header.ContentType()
if err != nil {
log.Fatal(err)
}
process(mediaType, part.Body)
}
}
}
}
// ParseMIMEBody parses the body of the message object into a tree of MIMEPart
// objects, each of which is aware of its content type, filename and headers.
// If the part was encoded in quoted-printable or base64, it is decoded before
// being stored in the MIMEPart object.
func ParseMIMEBody(mailMsg *mail.Message) (*MIMEBody, error) {
mimeMsg := &MIMEBody{
IsTextFromHTML: false,
header: mailMsg.Header,
}
if !IsMultipartMessage(mailMsg) {
// Attachment only?
if IsBinaryBody(mailMsg) {
return binMIME(mailMsg)
}
// Parse as text only
bodyBytes, err := decodeSection(mailMsg.Header.Get("Content-Transfer-Encoding"),
mailMsg.Body)
if err != nil {
return nil, fmt.Errorf("Error decoding text-only message: %v", err)
}
// Handle plain ASCII text, content-type unspecified
mimeMsg.Text = string(bodyBytes)
// Process top-level content-type
ctype := mailMsg.Header.Get("Content-Type")
if ctype != "" {
if mediatype, mparams, err := mime.ParseMediaType(ctype); err == nil {
if mparams["charset"] != "" {
// Convert plain text to UTF8 if content type specified a charset
newStr, err := ConvertToUTF8String(mparams["charset"], bodyBytes)
if err != nil {
return nil, err
}
mimeMsg.Text = newStr
} else if mediatype == "text/html" {
// charset is empty, look in HTML body for charset
charset, err := charsetFromHTMLString(mimeMsg.Text)
if charset != "" && err == nil {
newStr, err := ConvertToUTF8String(charset, bodyBytes)
if err == nil {
mimeMsg.Text = newStr
}
}
}
if mediatype == "text/html" {
mimeMsg.HTML = mimeMsg.Text
// Empty Text will trigger html2text conversion below
mimeMsg.Text = ""
}
}
}
} else {
// Parse top-level multipart
ctype := mailMsg.Header.Get("Content-Type")
mediatype, params, err := mime.ParseMediaType(ctype)
if err != nil {
return nil, fmt.Errorf("Unable to parse media type: %v", err)
}
if !strings.HasPrefix(mediatype, "multipart/") {
return nil, fmt.Errorf("Unknown mediatype: %v", mediatype)
}
boundary := params["boundary"]
if boundary == "" {
return nil, fmt.Errorf("Unable to locate boundary param in Content-Type header")
}
// Root Node of our tree
root := NewMIMEPart(nil, mediatype)
mimeMsg.Root = root
err = parseParts(root, mailMsg.Body, boundary)
if err != nil {
return nil, err
}
// Locate text body
if mediatype == "multipart/altern" {
match := BreadthMatchFirst(root, func(p MIMEPart) bool {
return p.ContentType() == "text/plain" && p.Disposition() != "attachment"
})
if match != nil {
if match.Charset() != "" {
newStr, err := ConvertToUTF8String(match.Charset(), match.Content())
if err != nil {
return nil, err
}
mimeMsg.Text += newStr
} else {
mimeMsg.Text += string(match.Content())
}
}
} else {
// multipart is of a mixed type
match := DepthMatchAll(root, func(p MIMEPart) bool {
return p.ContentType() == "text/plain" && p.Disposition() != "attachment"
})
for i, m := range match {
if i > 0 {
mimeMsg.Text += "\n--\n"
}
if m.Charset() != "" {
newStr, err := ConvertToUTF8String(m.Charset(), m.Content())
if err != nil {
return nil, err
}
mimeMsg.Text += newStr
} else {
mimeMsg.Text += string(m.Content())
}
}
}
// Locate HTML body
match := BreadthMatchFirst(root, func(p MIMEPart) bool {
return p.ContentType() == "text/html" && p.Disposition() != "attachment"
})
if match != nil {
if match.Charset() != "" {
newStr, err := ConvertToUTF8String(match.Charset(), match.Content())
if err != nil {
return nil, err
}
mimeMsg.HTML += newStr
} else {
mimeMsg.HTML = string(match.Content())
}
}
// Locate attachments
mimeMsg.Attachments = BreadthMatchAll(root, func(p MIMEPart) bool {
return p.Disposition() == "attachment" || p.ContentType() == "application/octet-stream"
})
// Locate inlines
mimeMsg.Inlines = BreadthMatchAll(root, func(p MIMEPart) bool {
return p.Disposition() == "inline"
})
// Locate others parts not handled in "Attachments" and "inlines"
mimeMsg.OtherParts = BreadthMatchAll(root, func(p MIMEPart) bool {
if strings.HasPrefix(p.ContentType(), "multipart/") {
return false
}
if p.Disposition() != "" {
return false
}
if p.ContentType() == "application/octet-stream" {
return false
}
return p.ContentType() != "text/plain" && p.ContentType() != "text/html"
})
}
// Down-convert HTML to text if necessary
if mimeMsg.Text == "" && mimeMsg.HTML != "" {
mimeMsg.IsTextFromHTML = true
var err error
if mimeMsg.Text, err = html2text.FromString(mimeMsg.HTML); err != nil {
// Fail gently
mimeMsg.Text = ""
return mimeMsg, err
}
}
return mimeMsg, nil
}
func main() {
domain := flag.String("domain", "", "mailgun domain")
privateKey := flag.String("private-key", "", "secret mailgun api key")
publicKey := flag.String("public-key", "", "mailgun public api key")
dry := flag.Bool("dry", false, "do not update cache & only print to stdout (no e-mail)")
verbose := flag.Bool("verbose", false, "Print detailed output per monitored url.")
flag.Parse()
if *domain == "" || *privateKey == "" || *publicKey == "" {
log.Fatalln("domain, private-key and public-key flags are required")
}
gun := m.NewMailgun(*domain, *privateKey, *publicKey)
urls := loadUrls()
for _, url := range urls {
body, err := requestURL(url)
if err != nil {
log.Println("Skipped URL because of error requesting it:", url)
continue
}
filename := getExecFolder() + "/cache/" + getMD5Hash(url) + ".html"
cached, err := ioutil.ReadFile(filename)
if err != nil {
if *dry == false {
updateCache(filename, body)
}
switch err := err.(type) {
case *os.PathError:
fmt.Printf("This URL will now be monitored: %s\n\n", url)
default:
log.Fatalf("Fatal errors type %T\n", err)
}
} else {
if reflect.DeepEqual(cached, body) {
if *verbose {
fmt.Printf("This URL didn't change: %s\n\n", url)
}
} else {
cachedDoc := getGoqueryDoc(cached)
currentDoc := getGoqueryDoc(body)
cachedContent, _ := cachedDoc.Find("#content").Html()
currentContent, _ := currentDoc.Find("#content").Html()
if cachedContent == currentContent {
if *verbose {
fmt.Println("The website changed, but the content stayed the same.")
}
} else {
cachedText, _ := html2text.FromString(cachedContent)
currentText, _ := html2text.FromString(currentContent)
diff := difflib.Diff(strings.Split(cachedText, "\n"), strings.Split(currentText, "\n"))
msg := createMessage(diff, url)
htmlMsg := createHTMLMessage(diff, url)
if *dry {
fmt.Println(msg)
} else {
sendEmails(gun, msg, htmlMsg)
}
}
if *dry == false {
updateCache(filename, body)
}
}
}
}
}
// ReadMessage reads message from r.
// Using Send() on read messages can result in garbage in headers,
// make sure to remove unnecessary ones, before sending.
func ReadMessage(r io.Reader) (*Message, error) {
rawmsg, err := mail.ReadMessage(r)
if err != nil {
return nil, err
}
m := new(Message)
// MessageID
if id := rawmsg.Header.Get("Message-Id"); id != "" {
m.ID = id
} else {
m.ID = makeID()
}
// Date
if date, err := rawmsg.Header.Date(); err == nil {
m.Date = date
} else {
m.Date = time.Now()
}
// Subject
if subject, err := decodeHeader(rawmsg.Header.Get("Subject")); err == nil && subject != "" {
m.Subject = subject
} else if err != nil {
return nil, fmt.Errorf("decode header: %v", err)
} else {
m.Subject = "No subject"
}
// Return-Path
if h := rawmsg.Header.Get("Return-Path"); h != "" {
retpath, err := DecodeAddress(h)
if err != nil {
return nil, fmt.Errorf("parse return-path: %v", err)
}
if len(retpath) > 0 {
m.ReturnPath = retpath[0]
}
}
// From
if h := rawmsg.Header.Get("From"); h != "" {
from, err := DecodeAddress(h)
if err != nil {
return nil, fmt.Errorf("parse from: %v", err)
}
if len(from) > 0 {
m.From = from[0]
}
}
// To
if h := rawmsg.Header.Get("To"); h != "" {
to, err := DecodeAddress(h)
if err != nil {
return nil, fmt.Errorf("parse to: %v", err)
}
m.To = to
} else {
m.To = make([]string, 0)
}
// CC
if h := rawmsg.Header.Get("Cc"); h != "" {
cc, err := DecodeAddress(h)
if err != nil {
return nil, fmt.Errorf("parse cc: %v", err)
}
m.CC = cc
} else {
m.CC = make([]string, 0)
}
// If return-path is unset, set it using from.
if m.ReturnPath == "" {
m.ReturnPath = m.From
}
// Decode rest of the headers.
headers := make(map[string]string)
for k, v := range rawmsg.Header {
switch k {
case "Message-Id", "Subject", "Date", "Return-Path", "From", "To", "Cc":
continue
}
for _, w := range v {
decoded, err := decodeHeader(w)
if err != nil {
return nil, fmt.Errorf("decode header: %v", err)
}
str := headers[k]
if str != "" {
str += " "
}
headers[k] = str + decoded
}
}
m.Headers = headers
// Decode body.
if err := m.decodeBody(rawmsg.Body, textproto.MIMEHeader(rawmsg.Header)); err != nil {
return nil, fmt.Errorf("decode body: %v", err)
}
if len(m.HTML) > 0 {
m.Body, err = html2text.FromString(m.HTML)
if err != nil {
return nil, err
}
m.IsHTML = true
}
return m, nil
}
