diff --git a/README.md b/README.md
index 513ace8..8b80f43 100644
--- a/README.md
+++ b/README.md
@@ -20,15 +20,16 @@ try using cs and its right at the top
 
 https://github.com/BurntSushi/ripgrep/issues/95
 
-active bugs
-search for cs --hidden --no-gitignore --no-ignore 英文 has highlight issues due to multiple byte ending matches
+```
+BUGS
+search for cs --hidden --no-gitignore --no-ignore 英文 cuts in the middle of a rune
+clicking on file in PDF mode shows the binary not the extracted text
 
 TODO
-search by filename
+search by filename (currently only ranks by it) as addtional OR search
 clean up parser so multiple spaces aren't tokens or flag em to be ignored
-if someone enables the --pdf flag not in terminal mode should spawn background process to extract the text so its pre cached
-if in http mode with --pdf need to pull the extracted text when displaying and possibly a link to raw file
-
+add "raw" link on the http page
+```
 
 Snippet generation
 
@@ -55,11 +56,6 @@ https://github.com/sourcegraph/src-cli
 
 https://arxiv.org/pdf/1904.03061.pdf
 
-hyperfine './cs "/(i?)test/"' './cs test' 'rg -i test' 'cs test' 'ag -i test'
-hyperfine './cs "/([A-Z][a-z]+)\s+([A-Z][a-z]+)/"' 'rg -uu "([A-Z][a-z]+)\s+([A-Z][a-z]+)"'
-hyperfine './cs "/[ab]+/"' 'rg -uu "[ab]+"'
-
-
 
 
 https://www.researchgate.net/publication/4004411_Topic_extraction_from_news_archive_using_TFPDF_algorithm
diff --git a/processor/http.go b/processor/http.go
index 3deae78..ce3cbc0 100644
--- a/processor/http.go
+++ b/processor/http.go
@@ -3,6 +3,7 @@ package processor
 import (
 	"crypto/md5"
 	"encoding/hex"
+	"errors"
 	"fmt"
 	"github.com/boyter/cs/file"
 	str "github.com/boyter/cs/string"
@@ -49,6 +50,18 @@ type facet struct {
 }
 
 func StartHttpServer() {
+	http.HandleFunc("/file/raw/", func(w http.ResponseWriter, r *http.Request) {
+		path := strings.Replace(r.URL.Path, "/file/raw/", "", 1)
+
+		log.Info().
+			Str("unique_code", "f24a4b1d").
+			Str("path", path).
+			Msg("raw page")
+
+		http.ServeFile(w, r, path)
+		return
+	})
+
 	http.HandleFunc("/file/", func(w http.ResponseWriter, r *http.Request) {
 		startTime := makeTimestampMilli()
 		startPos := tryParseInt(r.URL.Query().Get("sp"), 0)
@@ -63,7 +76,22 @@ func StartHttpServer() {
 			Str("path", path).
 			Msg("file view page")
 
-		content, err := ioutil.ReadFile(path)
+		var content []byte
+		var err error
+
+		// if its a PDF we should go to the cache to fetch it
+		extension := file.GetExtension(path)
+		if strings.ToLower(extension) == "pdf" {
+			c, ok := __pdfCache[path]
+			if ok {
+				content = []byte(c)
+			} else {
+				err = errors.New("")
+			}
+		} else {
+			content, err = ioutil.ReadFile(path)
+		}
+
 		if err != nil {
 			log.Error().
 				Str("unique_code", "d063c1fd").
@@ -71,7 +99,7 @@ func StartHttpServer() {
 				Int("endpos", endPos).
 				Str("path", path).
 				Msg("error reading file")
-			panic(err)
+			http.Redirect(w, r, "/", http.StatusTemporaryRedirect)
 		}
 
 		// Create a random string to define where the start and end of
@@ -130,6 +158,7 @@ func StartHttpServer() {
 		</div>
 		<div>
 			<h4>{{ .Location }}</h4>
+			<small>[<a href="/file/raw/{{ .Location }}">raw file</a>]</small>
 			<pre>{{ .Content }}</pre>
 		</div>
 	</body>
@@ -149,6 +178,7 @@ func StartHttpServer() {
 			panic(err)
 		}
 
+		return
 	})
 
 	http.HandleFunc("/", func(w http.ResponseWriter, r *http.Request) {
@@ -384,7 +414,7 @@ func StartHttpServer() {
 		if err != nil {
 			panic(err)
 		}
-
+		return
 	})
 
 	log.Info().