From: Brad Fitzpatrick <bradfitz@golang.org>
Date: Thu, 9 Feb 2012 23:02:06 +0000 (+1100)
Subject: net/http: add ServeContent
X-Git-Tag: weekly.2012-02-14~175
X-Git-Url: http://www.git.cypherpunks.su/?a=commitdiff_plain;h=4539d1f307d0f8f110367bc61d11e0888feb071d;p=gostls13.git

net/http: add ServeContent

Fixes #2039

R=r, rsc, n13m3y3r, r, rogpeppe
CC=golang-dev
https://golang.org/cl/5643067
---

diff --git a/src/pkg/net/http/fs.go b/src/pkg/net/http/fs.go
index 1392ca68ad..0e192eb99c 100644
--- a/src/pkg/net/http/fs.go
+++ b/src/pkg/net/http/fs.go
@@ -17,7 +17,6 @@ import (
 	"strconv"
 	"strings"
 	"time"
-	"unicode/utf8"
 )
 
 // A Dir implements http.FileSystem using the native file
@@ -58,32 +57,6 @@ type File interface {
 	Seek(offset int64, whence int) (int64, error)
 }
 
-// Heuristic: b is text if it is valid UTF-8 and doesn't
-// contain any unprintable ASCII or Unicode characters.
-func isText(b []byte) bool {
-	for len(b) > 0 && utf8.FullRune(b) {
-		rune, size := utf8.DecodeRune(b)
-		if size == 1 && rune == utf8.RuneError {
-			// decoding error
-			return false
-		}
-		if 0x7F <= rune && rune <= 0x9F {
-			return false
-		}
-		if rune < ' ' {
-			switch rune {
-			case '\n', '\r', '\t':
-				// okay
-			default:
-				// binary garbage
-				return false
-			}
-		}
-		b = b[size:]
-	}
-	return true
-}
-
 func dirList(w ResponseWriter, f File) {
 	w.Header().Set("Content-Type", "text/html; charset=utf-8")
 	fmt.Fprintf(w, "<pre>\n")
@@ -104,6 +77,123 @@ func dirList(w ResponseWriter, f File) {
 	fmt.Fprintf(w, "</pre>\n")
 }
 
+// ServeContent replies to the request using the content in the
+// provided ReadSeeker.  The main benefit of ServeContent over io.Copy
+// is that it handles Range requests properly, sets the MIME type, and
+// handles If-Modified-Since requests.
+//
+// If the response's Content-Type header is not set, ServeContent
+// first tries to deduce the type from name's file extension and,
+// if that fails, falls back to reading the first block of the content
+// and passing it to DetectContentType.
+// The name is otherwise unused; in particular it can be empty and is
+// never sent in the response.
+//
+// If modtime is not the zero time, ServeContent includes it in a
+// Last-Modified header in the response.  If the request includes an
+// If-Modified-Since header, ServeContent uses modtime to decide
+// whether the content needs to be sent at all.
+//
+// The content's Seek method must work: ServeContent uses
+// a seek to the end of the content to determine its size.
+//
+// Note that *os.File implements the io.ReadSeeker interface.
+func ServeContent(w ResponseWriter, req *Request, name string, modtime time.Time, content io.ReadSeeker) {
+	size, err := content.Seek(0, os.SEEK_END)
+	if err != nil {
+		Error(w, "seeker can't seek", StatusInternalServerError)
+		return
+	}
+	_, err = content.Seek(0, os.SEEK_SET)
+	if err != nil {
+		Error(w, "seeker can't seek", StatusInternalServerError)
+		return
+	}
+	serveContent(w, req, name, modtime, size, content)
+}
+
+// if name is empty, filename is unknown. (used for mime type, before sniffing)
+// if modtime.IsZero(), modtime is unknown.
+// content must be seeked to the beginning of the file.
+func serveContent(w ResponseWriter, r *Request, name string, modtime time.Time, size int64, content io.ReadSeeker) {
+	if checkLastModified(w, r, modtime) {
+		return
+	}
+
+	code := StatusOK
+
+	// If Content-Type isn't set, use the file's extension to find it.
+	if w.Header().Get("Content-Type") == "" {
+		ctype := mime.TypeByExtension(filepath.Ext(name))
+		if ctype == "" {
+			// read a chunk to decide between utf-8 text and binary
+			var buf [1024]byte
+			n, _ := io.ReadFull(content, buf[:])
+			b := buf[:n]
+			ctype = DetectContentType(b)
+			_, err := content.Seek(0, os.SEEK_SET) // rewind to output whole file
+			if err != nil {
+				Error(w, "seeker can't seek", StatusInternalServerError)
+				return
+			}
+		}
+		w.Header().Set("Content-Type", ctype)
+	}
+
+	// handle Content-Range header.
+	// TODO(adg): handle multiple ranges
+	sendSize := size
+	if size >= 0 {
+		ranges, err := parseRange(r.Header.Get("Range"), size)
+		if err == nil && len(ranges) > 1 {
+			err = errors.New("multiple ranges not supported")
+		}
+		if err != nil {
+			Error(w, err.Error(), StatusRequestedRangeNotSatisfiable)
+			return
+		}
+		if len(ranges) == 1 {
+			ra := ranges[0]
+			if _, err := content.Seek(ra.start, os.SEEK_SET); err != nil {
+				Error(w, err.Error(), StatusRequestedRangeNotSatisfiable)
+				return
+			}
+			sendSize = ra.length
+			code = StatusPartialContent
+			w.Header().Set("Content-Range", fmt.Sprintf("bytes %d-%d/%d", ra.start, ra.start+ra.length-1, size))
+		}
+
+		w.Header().Set("Accept-Ranges", "bytes")
+		if w.Header().Get("Content-Encoding") == "" {
+			w.Header().Set("Content-Length", strconv.FormatInt(sendSize, 10))
+		}
+	}
+
+	w.WriteHeader(code)
+
+	if r.Method != "HEAD" {
+		if sendSize == -1 {
+			io.Copy(w, content)
+		} else {
+			io.CopyN(w, content, sendSize)
+		}
+	}
+}
+
+// modtime is the modification time of the resource to be served, or IsZero().
+// return value is whether this request is now complete.
+func checkLastModified(w ResponseWriter, r *Request, modtime time.Time) bool {
+	if modtime.IsZero() {
+		return false
+	}
+	if t, err := time.Parse(TimeFormat, r.Header.Get("If-Modified-Since")); err == nil && modtime.After(t) {
+		w.WriteHeader(StatusNotModified)
+		return true
+	}
+	w.Header().Set("Last-Modified", modtime.UTC().Format(TimeFormat))
+	return false
+}
+
 // name is '/'-separated, not filepath.Separator.
 func serveFile(w ResponseWriter, r *Request, fs FileSystem, name string, redirect bool) {
 	const indexPage = "/index.html"
@@ -148,14 +238,11 @@ func serveFile(w ResponseWriter, r *Request, fs FileSystem, name string, redirec
 		}
 	}
 
-	if t, err := time.Parse(TimeFormat, r.Header.Get("If-Modified-Since")); err == nil && !d.ModTime().After(t) {
-		w.WriteHeader(StatusNotModified)
-		return
-	}
-	w.Header().Set("Last-Modified", d.ModTime().UTC().Format(TimeFormat))
-
 	// use contents of index.html for directory, if present
 	if d.IsDir() {
+		if checkLastModified(w, r, d.ModTime()) {
+			return
+		}
 		index := name + indexPage
 		ff, err := fs.Open(index)
 		if err == nil {
@@ -174,60 +261,7 @@ func serveFile(w ResponseWriter, r *Request, fs FileSystem, name string, redirec
 		return
 	}
 
-	// serve file
-	size := d.Size()
-	code := StatusOK
-
-	// If Content-Type isn't set, use the file's extension to find it.
-	if w.Header().Get("Content-Type") == "" {
-		ctype := mime.TypeByExtension(filepath.Ext(name))
-		if ctype == "" {
-			// read a chunk to decide between utf-8 text and binary
-			var buf [1024]byte
-			n, _ := io.ReadFull(f, buf[:])
-			b := buf[:n]
-			if isText(b) {
-				ctype = "text/plain; charset=utf-8"
-			} else {
-				// generic binary
-				ctype = "application/octet-stream"
-			}
-			f.Seek(0, os.SEEK_SET) // rewind to output whole file
-		}
-		w.Header().Set("Content-Type", ctype)
-	}
-
-	// handle Content-Range header.
-	// TODO(adg): handle multiple ranges
-	ranges, err := parseRange(r.Header.Get("Range"), size)
-	if err == nil && len(ranges) > 1 {
-		err = errors.New("multiple ranges not supported")
-	}
-	if err != nil {
-		Error(w, err.Error(), StatusRequestedRangeNotSatisfiable)
-		return
-	}
-	if len(ranges) == 1 {
-		ra := ranges[0]
-		if _, err := f.Seek(ra.start, os.SEEK_SET); err != nil {
-			Error(w, err.Error(), StatusRequestedRangeNotSatisfiable)
-			return
-		}
-		size = ra.length
-		code = StatusPartialContent
-		w.Header().Set("Content-Range", fmt.Sprintf("bytes %d-%d/%d", ra.start, ra.start+ra.length-1, d.Size()))
-	}
-
-	w.Header().Set("Accept-Ranges", "bytes")
-	if w.Header().Get("Content-Encoding") == "" {
-		w.Header().Set("Content-Length", strconv.FormatInt(size, 10))
-	}
-
-	w.WriteHeader(code)
-
-	if r.Method != "HEAD" {
-		io.CopyN(w, f, size)
-	}
+	serveContent(w, r, d.Name(), d.ModTime(), d.Size(), f)
 }
 
 // localRedirect gives a Moved Permanently response.
diff --git a/src/pkg/net/http/fs_test.go b/src/pkg/net/http/fs_test.go
index feea9209e6..45059fd5f3 100644
--- a/src/pkg/net/http/fs_test.go
+++ b/src/pkg/net/http/fs_test.go
@@ -6,6 +6,7 @@ package http_test
 
 import (
 	"fmt"
+	"io"
 	"io/ioutil"
 	. "net/http"
 	"net/http/httptest"
@@ -14,6 +15,7 @@ import (
 	"path/filepath"
 	"strings"
 	"testing"
+	"time"
 )
 
 const (
@@ -56,18 +58,18 @@ func TestServeFile(t *testing.T) {
 	req.Method = "GET"
 
 	// straight GET
-	_, body := getBody(t, req)
+	_, body := getBody(t, "straight get", req)
 	if !equal(body, file) {
 		t.Fatalf("body mismatch: got %q, want %q", body, file)
 	}
 
 	// Range tests
-	for _, rt := range ServeFileRangeTests {
+	for i, rt := range ServeFileRangeTests {
 		req.Header.Set("Range", "bytes="+rt.r)
 		if rt.r == "" {
 			req.Header["Range"] = nil
 		}
-		r, body := getBody(t, req)
+		r, body := getBody(t, fmt.Sprintf("test %d", i), req)
 		if r.StatusCode != rt.code {
 			t.Errorf("range=%q: StatusCode=%d, want %d", rt.r, r.StatusCode, rt.code)
 		}
@@ -298,7 +300,6 @@ func TestServeIndexHtml(t *testing.T) {
 		if err != nil {
 			t.Fatal(err)
 		}
-		defer res.Body.Close()
 		b, err := ioutil.ReadAll(res.Body)
 		if err != nil {
 			t.Fatal("reading Body:", err)
@@ -306,17 +307,66 @@ func TestServeIndexHtml(t *testing.T) {
 		if s := string(b); s != want {
 			t.Errorf("for path %q got %q, want %q", path, s, want)
 		}
+		res.Body.Close()
+	}
+}
+
+func TestServeContent(t *testing.T) {
+	type req struct {
+		name    string
+		modtime time.Time
+		content io.ReadSeeker
+	}
+	ch := make(chan req, 1)
+	ts := httptest.NewServer(HandlerFunc(func(w ResponseWriter, r *Request) {
+		p := <-ch
+		ServeContent(w, r, p.name, p.modtime, p.content)
+	}))
+	defer ts.Close()
+
+	css, err := os.Open("testdata/style.css")
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer css.Close()
+
+	ch <- req{"style.css", time.Time{}, css}
+	res, err := Get(ts.URL)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if g, e := res.Header.Get("Content-Type"), "text/css; charset=utf-8"; g != e {
+		t.Errorf("style.css: content type = %q, want %q", g, e)
+	}
+	if g := res.Header.Get("Last-Modified"); g != "" {
+		t.Errorf("want empty Last-Modified; got %q", g)
+	}
+
+	fi, err := css.Stat()
+	if err != nil {
+		t.Fatal(err)
+	}
+	ch <- req{"style.html", fi.ModTime(), css}
+	res, err = Get(ts.URL)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if g, e := res.Header.Get("Content-Type"), "text/html; charset=utf-8"; g != e {
+		t.Errorf("style.html: content type = %q, want %q", g, e)
+	}
+	if g := res.Header.Get("Last-Modified"); g == "" {
+		t.Errorf("want non-empty last-modified")
 	}
 }
 
-func getBody(t *testing.T, req Request) (*Response, []byte) {
+func getBody(t *testing.T, testName string, req Request) (*Response, []byte) {
 	r, err := DefaultClient.Do(&req)
 	if err != nil {
-		t.Fatal(req.URL.String(), "send:", err)
+		t.Fatalf("%s: for URL %q, send error: %v", testName, req.URL.String(), err)
 	}
 	b, err := ioutil.ReadAll(r.Body)
 	if err != nil {
-		t.Fatal("reading Body:", err)
+		t.Fatalf("%s: for URL %q, reading body: %v", testName, req.URL.String(), err)
 	}
 	return r, b
 }