Documents search

README.md

@@ -22,7 +22,8 @@ There are some special `query` values:
 - `images` - find all images on pages and output to the corresponding directory in `output_dir` (**IMPORTANT**: set `content_fetch_timeout_ms` to `0` so the images (and other content below) load fully)
 - `videos` - find and fetch files that look like videos
 - `audio` - find and fetch files that look like audio
-- `everything` - find and fetch images, audio and video
+- `documents` - find and fetch files that look like a document
+- `everything` - find and fetch images, audio, video, documents and email addresses
 
 When `is_regexp` is enabled, the `query` is treated as a regexp string and pages will be scanned for matches that satisfy it.
 

src/config/config.go

@@ -29,14 +29,16 @@ const (
 	QueryVideos     string = "videos"
 	QueryAudio      string = "audio"
 	QueryEmail      string = "email"
+	QueryDocuments  string = "documents"
 	QueryEverything string = "everything"
 )
 
 const (
-	SavePagesDir  string = "pages"
-	SaveImagesDir string = "images"
-	SaveVideosDir string = "videos"
-	SaveAudioDir  string = "audio"
+	SavePagesDir     string = "pages"
+	SaveImagesDir    string = "images"
+	SaveVideosDir    string = "videos"
+	SaveAudioDir     string = "audio"
+	SaveDocumentsDir string = "documents"
 )
 
 type Search struct {

src/main.go

@@ -39,7 +39,7 @@ import (
 	"unbewohnte/wecr/worker"
 )
 
-const version = "v0.2.4"
+const version = "v0.2.5"
 
 const (
 	defaultConfigFile           string = "conf.json"
@@ -295,6 +295,12 @@ func main() {
 		return
 	}
 
+	err = os.MkdirAll(filepath.Join(conf.Save.OutputDir, config.SaveDocumentsDir), os.ModePerm)
+	if err != nil {
+		logger.Error("Failed to create output directory for documents: %s", err)
+		return
+	}
+
 	switch conf.Search.Query {
 	case config.QueryEmail:
 		logger.Info("Looking for email addresses")
@@ -304,8 +310,15 @@ func main() {
 		logger.Info("Looking for videos (%+s)", web.VideoExtentions)
 	case config.QueryAudio:
 		logger.Info("Looking for audio (%+s)", web.AudioExtentions)
+	case config.QueryDocuments:
+		logger.Info("Looking for documents (%+s)", web.DocumentExtentions)
 	case config.QueryEverything:
-		logger.Info("Looking for email addresses, images, videos and audio (%+s - %+s - %+s)", web.ImageExtentions, web.VideoExtentions, web.AudioExtentions)
+		logger.Info("Looking for email addresses, images, videos, audio and various documents (%+s - %+s - %+s - %+s)",
+			web.ImageExtentions,
+			web.VideoExtentions,
+			web.AudioExtentions,
+			web.DocumentExtentions,
+		)
 	default:
 		if conf.Search.IsRegexp {
 			logger.Info("Looking for RegExp matches (%s)", conf.Search.Query)

src/web/documents.go

@@ -0,0 +1,100 @@
+package web
+
+import (
+	"net/url"
+	"strings"
+)
+
+func HasDocumentExtention(url string) bool {
+	for _, extention := range DocumentExtentions {
+		if strings.HasSuffix(url, extention) {
+			return true
+		}
+	}
+
+	return false
+}
+
+// Tries to find docs' URLs on the page
+func FindPageDocuments(pageBody []byte, from *url.URL) []string {
+	var urls []string
+
+	// for every element that has "src" attribute
+	for _, match := range tagSrcRegexp.FindAllString(string(pageBody), -1) {
+		var linkStartIndex int
+		var linkEndIndex int
+
+		linkStartIndex = strings.Index(match, "\"")
+		if linkStartIndex == -1 {
+			linkStartIndex = strings.Index(match, "'")
+			if linkStartIndex == -1 {
+				continue
+			}
+
+			linkEndIndex = strings.LastIndex(match, "'")
+			if linkEndIndex == -1 {
+				continue
+			}
+		} else {
+			linkEndIndex = strings.LastIndex(match, "\"")
+			if linkEndIndex == -1 {
+				continue
+			}
+		}
+
+		if linkEndIndex <= linkStartIndex+1 {
+			continue
+		}
+
+		link, err := url.Parse(match[linkStartIndex+1 : linkEndIndex])
+		if err != nil {
+			continue
+		}
+
+		linkResolved := ResolveLink(link, from.Host)
+		if HasDocumentExtention(linkResolved) {
+			urls = append(urls, linkResolved)
+		}
+	}
+
+	// for every "a" element as well
+	for _, match := range tagHrefRegexp.FindAllString(string(pageBody), -1) {
+		var linkStartIndex int
+		var linkEndIndex int
+
+		linkStartIndex = strings.Index(match, "\"")
+		if linkStartIndex == -1 {
+			linkStartIndex = strings.Index(match, "'")
+			if linkStartIndex == -1 {
+				continue
+			}
+
+			linkEndIndex = strings.LastIndex(match, "'")
+			if linkEndIndex == -1 {
+				continue
+			}
+		} else {
+			linkEndIndex = strings.LastIndex(match, "\"")
+			if linkEndIndex == -1 {
+				continue
+			}
+		}
+
+		if linkEndIndex <= linkStartIndex+1 {
+			continue
+		}
+
+		link, err := url.Parse(match[linkStartIndex+1 : linkEndIndex])
+		if err != nil {
+			continue
+		}
+
+		linkResolved := ResolveLink(link, from.Host)
+		if HasDocumentExtention(linkResolved) {
+			urls = append(urls, linkResolved)
+		}
+	}
+
+	// return discovered doc urls
+	return urls
+}

src/web/extentions.go

@@ -82,3 +82,56 @@ var VideoExtentions = []string{
 	".vob",
 	".ogv",
 }
+
+var DocumentExtentions = []string{
+	".pdf",
+	".doc",
+	".docx",
+	".epub",
+	".fb2",
+	".pub",
+	".ppt",
+	".pptx",
+	".txt",
+	".tex",
+	".odt",
+	".bib",
+	".ps",
+	".dwg",
+	".lyx",
+	".key",
+	".ott",
+	".odf",
+	".odc",
+	".ppg",
+	".xlc",
+	".latex",
+	".c",
+	".cpp",
+	".sh",
+	".go",
+	".java",
+	".cs",
+	".rs",
+	".lua",
+	".php",
+	".py",
+	".pl",
+	".lua",
+	".kt",
+	".js",
+	".rb",
+	".asm",
+	".rar",
+	".tar",
+	".db",
+	".7z",
+	".zip",
+	".gbr",
+	".tex",
+	".ttf",
+	".ttc",
+	".woff",
+	".otf",
+	".exif",
+}

src/web/images.go

@@ -64,7 +64,7 @@ func FindPageImages(pageBody []byte, from *url.URL) []string {
 			continue
 		}
 
-		link, err := url.Parse(match)
+		link, err := url.Parse(match[linkStartIndex+1 : linkEndIndex])
 		if err != nil {
 			continue
 		}

src/worker/worker.go

@@ -97,6 +97,8 @@ func (w *Worker) saveContent(links []string, pageURL *url.URL) {
 			filePath = filepath.Join(w.Conf.Save.OutputDir, config.SaveVideosDir, fileName)
 		} else if web.HasAudioExtention(link) {
 			filePath = filepath.Join(w.Conf.Save.OutputDir, config.SaveAudioDir, fileName)
+		} else if web.HasDocumentExtention(link) {
+			filePath = filepath.Join(w.Conf.Save.OutputDir, config.SaveDocumentsDir, fileName)
 		} else {
 			filePath = filepath.Join(w.Conf.Save.OutputDir, fileName)
 		}
@@ -146,11 +148,16 @@ func (w *Worker) Work() {
 		if w.Conf.VisitQueue.VisitQueue != nil {
 			w.Conf.VisitQueue.Lock.Lock()
 			newJob, err := queue.PopLastJob(w.Conf.VisitQueue.VisitQueue)
-			if err != nil || newJob == nil {
+			if err != nil {
 				logger.Error("Failed to get a new job from visit queue: %s", err)
 				w.Conf.VisitQueue.Lock.Unlock()
 				continue
 			}
+			if newJob == nil {
+				w.Conf.VisitQueue.Lock.Unlock()
+				continue
+			}
+
 			job = *newJob
 			w.Conf.VisitQueue.Lock.Unlock()
 		} else {
@@ -276,26 +283,35 @@ func (w *Worker) Work() {
 		case config.QueryImages:
 			// find image URLs, output images to the file while not saving already outputted ones
 			imageLinks := web.FindPageImages(pageData, pageURL)
-			w.saveContent(imageLinks, pageURL)
 			if len(imageLinks) > 0 {
+				w.saveContent(imageLinks, pageURL)
 				savePage = true
 			}
 
 		case config.QueryVideos:
 			// search for videos
 			// find video URLs, output videos to the files while not saving already outputted ones
 			videoLinks := web.FindPageVideos(pageData, pageURL)
-			w.saveContent(videoLinks, pageURL)
 			if len(videoLinks) > 0 {
+				w.saveContent(videoLinks, pageURL)
 				savePage = true
 			}
 
 		case config.QueryAudio:
 			// search for audio
 			// find audio URLs, output audio to the file while not saving already outputted ones
 			audioLinks := web.FindPageAudio(pageData, pageURL)
-			w.saveContent(audioLinks, pageURL)
 			if len(audioLinks) > 0 {
+				w.saveContent(audioLinks, pageURL)
+				savePage = true
+			}
+
+		case config.QueryDocuments:
+			// search for various documents
+			// find documents URLs, output docs to the file while not saving already outputted ones
+			docsLinks := web.FindPageDocuments(pageData, pageURL)
+			if len(docsLinks) > 0 {
+				w.saveContent(docsLinks, pageURL)
 				savePage = true
 			}
 
@@ -320,6 +336,7 @@ func (w *Worker) Work() {
 			contentLinks = append(contentLinks, web.FindPageImages(pageData, pageURL)...)
 			contentLinks = append(contentLinks, web.FindPageAudio(pageData, pageURL)...)
 			contentLinks = append(contentLinks, web.FindPageVideos(pageData, pageURL)...)
+			contentLinks = append(contentLinks, web.FindPageDocuments(pageData, pageURL)...)
 			w.saveContent(contentLinks, pageURL)
 
 			// email