X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/0f668000c8cacec50804cb0019dbe7d7dc1d2b36..fcbb743e3de63e93280f2fbeedea49f98430d26f:/services/keep-web/handler.go

diff --git a/services/keep-web/handler.go b/services/keep-web/handler.go
index ddbc9b1553..e1b23621af 100644
--- a/services/keep-web/handler.go
+++ b/services/keep-web/handler.go
@@ -9,8 +9,9 @@ import (
 	"net/http"
 	"net/url"
 	"os"
+	"regexp"
+	"strconv"
 	"strings"
-	"time"
 
 	"git.curoverse.com/arvados.git/sdk/go/arvadosclient"
 	"git.curoverse.com/arvados.git/sdk/go/auth"
@@ -23,20 +24,19 @@ type handler struct{}
 var (
 	clientPool         = arvadosclient.MakeClientPool()
 	trustAllContent    = false
-	anonymousTokens    []string
 	attachmentOnlyHost = ""
 )
 
 func init() {
-	flag.BoolVar(&trustAllContent, "trust-all-content", false,
-		"Serve non-public content from a single origin. Dangerous: read docs before using!")
 	flag.StringVar(&attachmentOnlyHost, "attachment-only-host", "",
 		"Accept credentials, and add \"Content-Disposition: attachment\" response headers, for requests at this hostname:port. Prohibiting inline display makes it possible to serve untrusted and non-public content from a single origin, i.e., without wildcard DNS or SSL.")
+	flag.BoolVar(&trustAllContent, "trust-all-content", false,
+		"Serve non-public content from a single origin. Dangerous: read docs before using!")
 }
 
 // return a UUID or PDH if s begins with a UUID or URL-encoded PDH;
 // otherwise return "".
-func parseCollectionIdFromDNSName(s string) string {
+func parseCollectionIDFromDNSName(s string) string {
 	// Strip domain.
 	if i := strings.IndexRune(s, '.'); i >= 0 {
 		s = s[:i]
@@ -60,7 +60,7 @@ var urlPDHDecoder = strings.NewReplacer(" ", "+", "-", "+")
 
 // return a UUID or PDH if s is a UUID or a PDH (even if it is a PDH
 // with "+" replaced by " " or "-"); otherwise return "".
-func parseCollectionIdFromURL(s string) string {
+func parseCollectionIDFromURL(s string) string {
 	if arvadosclient.UUIDMatch(s) {
 		return s
 	}
@@ -100,6 +100,15 @@ func (h *handler) ServeHTTP(wOrig http.ResponseWriter, r *http.Request) {
 		return
 	}
 
+	if r.Header.Get("Origin") != "" {
+		// Allow simple cross-origin requests without user
+		// credentials ("user credentials" as defined by CORS,
+		// i.e., cookies, HTTP authentication, and client-side
+		// SSL certificates. See
+		// http://www.w3.org/TR/cors/#user-credentials).
+		w.Header().Set("Access-Control-Allow-Origin", "*")
+	}
+
 	arv := clientPool.Get()
 	if arv == nil {
 		statusCode, statusText = http.StatusInternalServerError, "Pool failed: "+clientPool.Err().Error()
@@ -109,7 +118,7 @@ func (h *handler) ServeHTTP(wOrig http.ResponseWriter, r *http.Request) {
 
 	pathParts := strings.Split(r.URL.Path[1:], "/")
 
-	var targetId string
+	var targetID string
 	var targetPath []string
 	var tokens []string
 	var reqTokens []string
@@ -124,24 +133,24 @@ func (h *handler) ServeHTTP(wOrig http.ResponseWriter, r *http.Request) {
 		attachment = true
 	}
 
-	if targetId = parseCollectionIdFromDNSName(r.Host); targetId != "" {
+	if targetID = parseCollectionIDFromDNSName(r.Host); targetID != "" {
 		// http://ID.collections.example/PATH...
 		credentialsOK = true
 		targetPath = pathParts
 	} else if len(pathParts) >= 2 && strings.HasPrefix(pathParts[0], "c=") {
 		// /c=ID/PATH...
-		targetId = parseCollectionIdFromURL(pathParts[0][2:])
+		targetID = parseCollectionIDFromURL(pathParts[0][2:])
 		targetPath = pathParts[1:]
 	} else if len(pathParts) >= 3 && pathParts[0] == "collections" {
 		if len(pathParts) >= 5 && pathParts[1] == "download" {
 			// /collections/download/ID/TOKEN/PATH...
-			targetId = pathParts[2]
+			targetID = pathParts[2]
 			tokens = []string{pathParts[3]}
 			targetPath = pathParts[4:]
 			pathToken = true
 		} else {
 			// /collections/ID/PATH...
-			targetId = pathParts[1]
+			targetID = pathParts[1]
 			tokens = anonymousTokens
 			targetPath = pathParts[2:]
 		}
@@ -149,7 +158,24 @@ func (h *handler) ServeHTTP(wOrig http.ResponseWriter, r *http.Request) {
 		statusCode = http.StatusNotFound
 		return
 	}
-	if t := r.FormValue("api_token"); t != "" {
+
+	formToken := r.FormValue("api_token")
+	if formToken != "" && r.Header.Get("Origin") != "" && attachment && r.URL.Query().Get("api_token") == "" {
+		// The client provided an explicit token in the POST
+		// body. The Origin header indicates this *might* be
+		// an AJAX request, in which case redirect-with-cookie
+		// won't work: we should just serve the content in the
+		// POST response. This is safe because:
+		//
+		// * We're supplying an attachment, not inline
+		//   content, so we don't need to convert the POST to
+		//   a GET and avoid the "really resubmit form?"
+		//   problem.
+		//
+		// * The token isn't embedded in the URL, so we don't
+		//   need to worry about bookmarks and copy/paste.
+		tokens = append(tokens, formToken)
+	} else if formToken != "" {
 		// The client provided an explicit token in the query
 		// string, or a form in POST body. We must put the
 		// token in an HttpOnly cookie, and redirect to the
@@ -179,12 +205,21 @@ func (h *handler) ServeHTTP(wOrig http.ResponseWriter, r *http.Request) {
 
 		http.SetCookie(w, &http.Cookie{
 			Name:     "arvados_api_token",
-			Value:    auth.EncodeTokenCookie([]byte(t)),
+			Value:    auth.EncodeTokenCookie([]byte(formToken)),
 			Path:     "/",
-			Expires:  time.Now().AddDate(10, 0, 0),
 			HttpOnly: true,
 		})
-		redir := (&url.URL{Host: r.Host, Path: r.URL.Path}).String()
+
+		// Propagate query parameters (except api_token) from
+		// the original request.
+		redirQuery := r.URL.Query()
+		redirQuery.Del("api_token")
+
+		redir := (&url.URL{
+			Host:     r.Host,
+			Path:     r.URL.Path,
+			RawQuery: redirQuery.Encode(),
+		}).String()
 
 		w.Header().Add("Location", redir)
 		statusCode, statusText = http.StatusSeeOther, redir
@@ -228,7 +263,7 @@ func (h *handler) ServeHTTP(wOrig http.ResponseWriter, r *http.Request) {
 	collection := make(map[string]interface{})
 	found := false
 	for _, arv.ApiToken = range tokens {
-		err := arv.Get("collections", targetId, nil, &collection)
+		err := arv.Get("collections", targetID, nil, &collection)
 		if err == nil {
 			// Success
 			found = true
@@ -295,8 +330,10 @@ func (h *handler) ServeHTTP(wOrig http.ResponseWriter, r *http.Request) {
 	}
 	defer rdr.Close()
 
-	// One or both of these can be -1 if not found:
 	basenamePos := strings.LastIndex(filename, "/")
+	if basenamePos < 0 {
+		basenamePos = 0
+	}
 	extPos := strings.LastIndex(filename, ".")
 	if extPos > basenamePos {
 		// Now extPos is safely >= 0.
@@ -304,14 +341,56 @@ func (h *handler) ServeHTTP(wOrig http.ResponseWriter, r *http.Request) {
 			w.Header().Set("Content-Type", t)
 		}
 	}
-	w.Header().Set("Content-Length", fmt.Sprintf("%d", rdr.Len()))
-	if attachment {
-		w.Header().Set("Content-Disposition", "attachment")
+	if rdr, ok := rdr.(keepclient.ReadCloserWithLen); ok {
+		w.Header().Set("Content-Length", fmt.Sprintf("%d", rdr.Len()))
 	}
 
-	w.WriteHeader(http.StatusOK)
-	_, err = io.Copy(w, rdr)
+	applyContentDispositionHdr(w, r, filename[basenamePos:], attachment)
+	rangeRdr, statusCode := applyRangeHdr(w, r, rdr)
+
+	w.WriteHeader(statusCode)
+	_, err = io.Copy(w, rangeRdr)
 	if err != nil {
 		statusCode, statusText = http.StatusBadGateway, err.Error()
 	}
 }
+
+var rangeRe = regexp.MustCompile(`^bytes=0-([0-9]*)$`)
+
+func applyRangeHdr(w http.ResponseWriter, r *http.Request, rdr keepclient.ReadCloserWithLen) (io.Reader, int) {
+	w.Header().Set("Accept-Ranges", "bytes")
+	hdr := r.Header.Get("Range")
+	fields := rangeRe.FindStringSubmatch(hdr)
+	if fields == nil {
+		return rdr, http.StatusOK
+	}
+	rangeEnd, err := strconv.ParseInt(fields[1], 10, 64)
+	if err != nil {
+		// Empty or too big for int64 == send entire content
+		return rdr, http.StatusOK
+	}
+	if uint64(rangeEnd) >= rdr.Len() {
+		return rdr, http.StatusOK
+	}
+	w.Header().Set("Content-Length", fmt.Sprintf("%d", rangeEnd+1))
+	w.Header().Set("Content-Range", fmt.Sprintf("bytes %d-%d/%d", 0, rangeEnd, rdr.Len()))
+	return &io.LimitedReader{R: rdr, N: rangeEnd + 1}, http.StatusPartialContent
+}
+
+func applyContentDispositionHdr(w http.ResponseWriter, r *http.Request, filename string, isAttachment bool) {
+	disposition := "inline"
+	if isAttachment {
+		disposition = "attachment"
+	}
+	if strings.ContainsRune(r.RequestURI, '?') {
+		// Help the UA realize that the filename is just
+		// "filename.txt", not
+		// "filename.txt?disposition=attachment".
+		//
+		// TODO(TC): Follow advice at RFC 6266 appendix D
+		disposition += "; filename=" + strconv.QuoteToASCII(filename)
+	}
+	if disposition != "inline" {
+		w.Header().Set("Content-Disposition", disposition)
+	}
+}