2017-01-02 20:20:17 +01:00
package iris
import (
2017-03-13 00:40:57 +01:00
"fmt"
"io"
2017-03-01 18:17:32 +01:00
"mime"
2017-03-13 00:40:57 +01:00
"mime/multipart"
2017-01-02 20:20:17 +01:00
"net/http"
2017-03-13 00:40:57 +01:00
"net/textproto"
"net/url"
2017-01-02 20:20:17 +01:00
"os"
2017-03-13 00:40:57 +01:00
"path"
2017-02-16 21:19:44 +01:00
"path/filepath"
2017-03-13 00:40:57 +01:00
"sort"
"strconv"
2017-01-02 20:20:17 +01:00
"strings"
"sync"
2017-03-13 00:40:57 +01:00
"time"
"github.com/kataras/go-errors"
2017-01-02 20:20:17 +01:00
)
2017-03-14 01:58:56 +01:00
// StaticHandler returns a new Handler which is ready
// to serve all kind of static files.
//
// Developers can wrap this handler using the `iris.StripPrefix`
// for a fixed static path when the result handler is being, finally, registered to a route.
//
//
// Usage:
// app := iris.New()
// ...
// fileserver := iris.StaticHandler("./static_files", false, false)
// h := iris.StripPrefix("/static", fileserver)
// /* http://mydomain.com/static/css/style.css */
// app.Get("/static", h)
// ...
//
func StaticHandler ( systemPath string , showList bool , enableGzip bool , exceptRoutes ... RouteInfo ) HandlerFunc {
return NewStaticHandlerBuilder ( systemPath ) .
Listing ( showList ) .
Gzip ( enableGzip ) .
Except ( exceptRoutes ... ) .
Build ( )
}
2017-01-02 20:20:17 +01:00
// StaticHandlerBuilder is the web file system's Handler builder
// use that or the iris.StaticHandler/StaticWeb methods
type StaticHandlerBuilder interface {
Gzip ( enable bool ) StaticHandlerBuilder
Listing ( listDirectoriesOnOff bool ) StaticHandlerBuilder
2017-02-14 04:54:11 +01:00
Except ( r ... RouteInfo ) StaticHandlerBuilder
2017-01-02 20:20:17 +01:00
Build ( ) HandlerFunc
}
2017-03-01 18:17:32 +01:00
// +------------------------------------------------------------+
// | |
// | Static Builder |
// | |
// +------------------------------------------------------------+
2017-02-14 04:54:11 +01:00
type fsHandler struct {
2017-01-02 20:20:17 +01:00
// user options, only directory is required.
directory http . Dir
gzip bool
listDirectories bool
// these are init on the Build() call
filesystem http . FileSystem
once sync . Once
2017-02-14 04:54:11 +01:00
exceptions [ ] RouteInfo
2017-01-02 20:20:17 +01:00
handler HandlerFunc
}
func toWebPath ( systemPath string ) string {
// winos slash to slash
webpath := strings . Replace ( systemPath , "\\" , slash , - 1 )
// double slashes to single
webpath = strings . Replace ( webpath , slash + slash , slash , - 1 )
// remove all dots
webpath = strings . Replace ( webpath , "." , "" , - 1 )
return webpath
}
2017-02-16 21:19:44 +01:00
// abs calls filepath.Abs but ignores the error and
2017-02-17 09:45:47 +01:00
// returns the original value if any error occurred.
2017-02-16 21:19:44 +01:00
func abs ( path string ) string {
absPath , err := filepath . Abs ( path )
if err != nil {
return path
}
return absPath
}
2017-01-02 20:20:17 +01:00
// NewStaticHandlerBuilder returns a new Handler which serves static files
// supports gzip, no listing and much more
// Note that, this static builder returns a Handler
// it doesn't cares about the rest of your iris configuration.
//
// Use the iris.StaticHandler/StaticWeb in order to serve static files on more automatic way
// this builder is used by people who have more complicated application
// structure and want a fluent api to work on.
func NewStaticHandlerBuilder ( dir string ) StaticHandlerBuilder {
2017-02-14 04:54:11 +01:00
return & fsHandler {
2017-02-16 21:19:44 +01:00
directory : http . Dir ( abs ( dir ) ) ,
2017-01-02 20:20:17 +01:00
// gzip is disabled by default
gzip : false ,
// list directories disabled by default
listDirectories : false ,
}
}
// Gzip if enable is true then gzip compression is enabled for this static directory
// Defaults to false
2017-02-14 04:54:11 +01:00
func ( w * fsHandler ) Gzip ( enable bool ) StaticHandlerBuilder {
2017-01-02 20:20:17 +01:00
w . gzip = enable
return w
}
// Listing turn on/off the 'show files and directories'.
// Defaults to false
2017-02-14 04:54:11 +01:00
func ( w * fsHandler ) Listing ( listDirectoriesOnOff bool ) StaticHandlerBuilder {
2017-01-02 20:20:17 +01:00
w . listDirectories = listDirectoriesOnOff
return w
}
2017-01-12 09:24:27 +01:00
// Except add a route exception,
// gives priority to that Route over the static handler.
2017-02-14 04:54:11 +01:00
func ( w * fsHandler ) Except ( r ... RouteInfo ) StaticHandlerBuilder {
2017-01-12 09:24:27 +01:00
w . exceptions = append ( w . exceptions , r ... )
return w
}
2017-01-02 20:20:17 +01:00
type (
noListFile struct {
http . File
}
)
// Overrides the Readdir of the http.File in order to disable showing a list of the dirs/files
func ( n noListFile ) Readdir ( count int ) ( [ ] os . FileInfo , error ) {
return nil , nil
}
// Implements the http.Filesystem
// Do not call it.
2017-02-14 04:54:11 +01:00
func ( w * fsHandler ) Open ( name string ) ( http . File , error ) {
2017-01-02 20:20:17 +01:00
info , err := w . filesystem . Open ( name )
if err != nil {
return nil , err
}
if ! w . listDirectories {
return noListFile { info } , nil
}
return info , nil
}
// Build the handler (once) and returns it
2017-02-14 04:54:11 +01:00
func ( w * fsHandler ) Build ( ) HandlerFunc {
2017-01-02 20:20:17 +01:00
// we have to ensure that Build is called ONLY one time,
// one instance per one static directory.
w . once . Do ( func ( ) {
2017-02-14 04:54:11 +01:00
w . filesystem = w . directory
2017-01-02 20:20:17 +01:00
2017-03-14 01:58:56 +01:00
fileserver := func ( ctx * Context ) {
2017-03-13 00:40:57 +01:00
upath := ctx . Request . URL . Path
if ! strings . HasPrefix ( upath , "/" ) {
upath = "/" + upath
ctx . Request . URL . Path = upath
}
2017-01-02 20:20:17 +01:00
2017-03-13 00:40:57 +01:00
// Note the request.url.path is changed but request.RequestURI is not
// so on custom errors we use the requesturi instead.
// this can be changed
_ , prevStatusCode := serveFile ( ctx ,
w . filesystem ,
path . Clean ( upath ) ,
false ,
w . listDirectories ,
( w . gzip && ctx . clientAllowsGzip ( ) ) ,
)
2017-01-02 20:20:17 +01:00
2017-03-13 00:40:57 +01:00
// check for any http errors after the file handler executed
if prevStatusCode >= 400 { // error found (404 or 400 or 500 usually)
if writer , ok := ctx . ResponseWriter . ( * gzipResponseWriter ) ; ok && writer != nil {
writer . ResetBody ( )
writer . Disable ( )
// ctx.ResponseWriter.Header().Del(contentType) // application/x-gzip sometimes lawl
// remove gzip headers
headers := ctx . ResponseWriter . Header ( )
headers [ contentType ] = nil
headers [ "X-Content-Type-Options" ] = nil
headers [ varyHeader ] = nil
headers [ contentEncodingHeader ] = nil
headers [ contentLength ] = nil
}
// execute any custom error handler (per-party or global, if not found then it creates a new one and fires it)
ctx . Framework ( ) . Router . Errors . Fire ( prevStatusCode , ctx )
return
2017-01-02 20:20:17 +01:00
}
2017-03-13 00:40:57 +01:00
// go to the next middleware
if ctx . Pos < len ( ctx . Middleware ) - 1 {
ctx . Next ( )
}
}
2017-01-12 09:24:27 +01:00
if len ( w . exceptions ) > 0 {
middleware := make ( Middleware , len ( w . exceptions ) + 1 )
for i := range w . exceptions {
middleware [ i ] = Prioritize ( w . exceptions [ i ] )
}
2017-03-13 00:40:57 +01:00
middleware [ len ( w . exceptions ) ] = HandlerFunc ( fileserver )
2017-01-12 09:24:27 +01:00
w . handler = func ( ctx * Context ) {
ctx . Middleware = append ( middleware , ctx . Middleware ... )
ctx . Do ( )
}
2017-03-13 00:40:57 +01:00
return
2017-01-12 09:24:27 +01:00
}
2017-03-13 00:40:57 +01:00
w . handler = fileserver
2017-01-02 20:20:17 +01:00
} )
return w . handler
}
2017-01-12 09:24:27 +01:00
// StripPrefix returns a handler that serves HTTP requests
// by removing the given prefix from the request URL's Path
// and invoking the handler h. StripPrefix handles a
// request for a path that doesn't begin with prefix by
// replying with an HTTP 404 not found error.
2017-03-14 01:58:56 +01:00
//
// Usage:
// fileserver := iris.StaticHandler("./static_files", false, false)
// h := iris.StripPrefix("/static", fileserver)
// app.Get("/static", h)
//
2017-01-12 09:24:27 +01:00
func StripPrefix ( prefix string , h HandlerFunc ) HandlerFunc {
if prefix == "" {
return h
}
2017-03-14 01:58:56 +01:00
// here we separate the path from the subdomain (if any), we care only for the path
// fixes a bug when serving static files via a subdomain
fixedPrefix := prefix
if dotWSlashIdx := strings . Index ( fixedPrefix , subdomainIndicator ) ; dotWSlashIdx > 0 {
fixedPrefix = fixedPrefix [ dotWSlashIdx + 1 : ]
}
fixedPrefix = toWebPath ( fixedPrefix )
2017-01-12 09:24:27 +01:00
return func ( ctx * Context ) {
2017-03-14 01:58:56 +01:00
if p := strings . TrimPrefix ( ctx . Request . URL . Path , fixedPrefix ) ; len ( p ) < len ( ctx . Request . URL . Path ) {
2017-01-12 09:24:27 +01:00
ctx . Request . URL . Path = p
h ( ctx )
} else {
ctx . NotFound ( )
}
}
}
2017-03-01 18:17:32 +01:00
2017-03-13 00:40:57 +01:00
// +------------------------------------------------------------+
// | |
// | serve file handler |
// | edited from net/http/fs.go in order to support GZIP with |
// | custom iris http errors and fallback to non-compressed data|
// | when not supported. |
// | |
// +------------------------------------------------------------+
var htmlReplacer = strings . NewReplacer (
"&" , "&" ,
"<" , "<" ,
">" , ">" ,
// """ is shorter than """.
` " ` , """ ,
// "'" is shorter than "'" and apos was not in HTML until HTML5.
"'" , "'" ,
)
func dirList ( ctx * Context , f http . File ) ( string , int ) {
dirs , err := f . Readdir ( - 1 )
if err != nil {
// TODO: log err.Error() to the Server.ErrorLog, once it's possible
// for a handler to get at its Server via the http.ResponseWriter. See
// Issue 12438.
return "Error reading directory" , StatusInternalServerError
}
sort . Slice ( dirs , func ( i , j int ) bool { return dirs [ i ] . Name ( ) < dirs [ j ] . Name ( ) } )
ctx . ResponseWriter . Header ( ) . Set ( contentType , contentHTML + "; charset=" + ctx . Framework ( ) . Config . Charset )
fmt . Fprintf ( ctx . ResponseWriter , "<pre>\n" )
for _ , d := range dirs {
name := d . Name ( )
if d . IsDir ( ) {
name += "/"
}
// name may contain '?' or '#', which must be escaped to remain
// part of the URL path, and not indicate the start of a query
// string or fragment.
url := url . URL { Path : name }
fmt . Fprintf ( ctx . ResponseWriter , "<a href=\"%s\">%s</a>\n" , url . String ( ) , htmlReplacer . Replace ( name ) )
}
fmt . Fprintf ( ctx . ResponseWriter , "</pre>\n" )
return "" , StatusOK
}
// errSeeker is returned by ServeContent's sizeFunc when the content
// doesn't seek properly. The underlying Seeker's error text isn't
// included in the sizeFunc reply so it's not sent over HTTP to end
// users.
var errSeeker = errors . New ( "seeker can't seek" )
// errNoOverlap is returned by serveContent's parseRange if first-byte-pos of
// all of the byte-range-spec values is greater than the content size.
var errNoOverlap = errors . New ( "invalid range: failed to overlap" )
// The algorithm uses at most sniffLen bytes to make its decision.
const sniffLen = 512
// 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.
// The sizeFunc is called at most once. Its error, if any, is sent in the HTTP response.
func serveContent ( ctx * Context , name string , modtime time . Time , sizeFunc func ( ) ( int64 , error ) , content io . ReadSeeker , gzip bool ) ( string , int ) /* we could use the TransactionErrResult but prefer not to create new objects for each of the errors on static file handlers*/ {
setLastModified ( ctx , modtime )
done , rangeReq := checkPreconditions ( ctx , modtime )
if done {
return "" , StatusNotModified
}
code := StatusOK
// If Content-Type isn't set, use the file's extension to find it, but
// if the Content-Type is unset explicitly, do not sniff the type.
ctypes , haveType := ctx . ResponseWriter . Header ( ) [ contentType ]
var ctype string
if ! haveType {
ctype = typeByExtension ( filepath . Ext ( name ) )
if ctype == "" {
// read a chunk to decide between utf-8 text and binary
var buf [ sniffLen ] byte
n , _ := io . ReadFull ( content , buf [ : ] )
ctype = http . DetectContentType ( buf [ : n ] )
_ , err := content . Seek ( 0 , io . SeekStart ) // rewind to output whole file
if err != nil {
return "seeker can't seek" , StatusInternalServerError
}
}
ctx . ResponseWriter . Header ( ) . Set ( contentType , ctype )
} else if len ( ctypes ) > 0 {
ctype = ctypes [ 0 ]
}
size , err := sizeFunc ( )
if err != nil {
return err . Error ( ) , StatusInternalServerError
}
// handle Content-Range header.
sendSize := size
var sendContent io . Reader = content
if gzip {
// set the "Accept-Encoding" here in order to prevent the content-length header to be setted later on.
ctx . SetHeader ( contentEncodingHeader , "gzip" )
ctx . ResponseWriter . Header ( ) . Add ( varyHeader , acceptEncodingHeader )
gzipResWriter := acquireGzipResponseWriter ( ctx . ResponseWriter )
ctx . ResponseWriter = gzipResWriter
}
if size >= 0 {
ranges , err := parseRange ( rangeReq , size )
if err != nil {
if err == errNoOverlap {
ctx . ResponseWriter . Header ( ) . Set ( "Content-Range" , fmt . Sprintf ( "bytes */%d" , size ) )
}
return err . Error ( ) , StatusRequestedRangeNotSatisfiable
}
if sumRangesSize ( ranges ) > size {
// The total number of bytes in all the ranges
// is larger than the size of the file by
// itself, so this is probably an attack, or a
// dumb client. Ignore the range request.
ranges = nil
}
switch {
case len ( ranges ) == 1 :
// RFC 2616, Section 14.16:
// "When an HTTP message includes the content of a single
// range (for example, a response to a request for a
// single range, or to a request for a set of ranges
// that overlap without any holes), this content is
// transmitted with a Content-Range header, and a
// Content-Length header showing the number of bytes
// actually transferred.
// ...
// A response to a request for a single range MUST NOT
// be sent using the multipart/byteranges media type."
ra := ranges [ 0 ]
if _ , err := content . Seek ( ra . start , io . SeekStart ) ; err != nil {
return err . Error ( ) , StatusRequestedRangeNotSatisfiable
}
sendSize = ra . length
code = StatusPartialContent
ctx . ResponseWriter . Header ( ) . Set ( "Content-Range" , ra . contentRange ( size ) )
case len ( ranges ) > 1 :
sendSize = rangesMIMESize ( ranges , ctype , size )
code = StatusPartialContent
pr , pw := io . Pipe ( )
mw := multipart . NewWriter ( pw )
ctx . ResponseWriter . Header ( ) . Set ( contentType , "multipart/byteranges; boundary=" + mw . Boundary ( ) )
sendContent = pr
defer pr . Close ( ) // cause writing goroutine to fail and exit if CopyN doesn't finish.
go func ( ) {
for _ , ra := range ranges {
part , err := mw . CreatePart ( ra . mimeHeader ( ctype , size ) )
if err != nil {
pw . CloseWithError ( err )
return
}
if _ , err := content . Seek ( ra . start , io . SeekStart ) ; err != nil {
pw . CloseWithError ( err )
return
}
if _ , err := io . CopyN ( part , content , ra . length ) ; err != nil {
pw . CloseWithError ( err )
return
}
}
mw . Close ( )
pw . Close ( )
} ( )
}
ctx . ResponseWriter . Header ( ) . Set ( "Accept-Ranges" , "bytes" )
if ctx . ResponseWriter . Header ( ) . Get ( contentEncodingHeader ) == "" {
ctx . ResponseWriter . Header ( ) . Set ( contentLength , strconv . FormatInt ( sendSize , 10 ) )
}
}
ctx . SetStatusCode ( code )
if ctx . Method ( ) != MethodHead {
io . CopyN ( ctx . ResponseWriter , sendContent , sendSize )
}
return "" , code
}
// scanETag determines if a syntactically valid ETag is present at s. If so,
// the ETag and remaining text after consuming ETag is returned. Otherwise,
// it returns "", "".
func scanETag ( s string ) ( etag string , remain string ) {
s = textproto . TrimString ( s )
start := 0
if strings . HasPrefix ( s , "W/" ) {
start = 2
}
if len ( s [ start : ] ) < 2 || s [ start ] != '"' {
return "" , ""
}
// ETag is either W/"text" or "text".
// See RFC 7232 2.3.
for i := start + 1 ; i < len ( s ) ; i ++ {
c := s [ i ]
switch {
// Character values allowed in ETags.
case c == 0x21 || c >= 0x23 && c <= 0x7E || c >= 0x80 :
case c == '"' :
return string ( s [ : i + 1 ] ) , s [ i + 1 : ]
default :
break
}
}
return "" , ""
}
// etagStrongMatch reports whether a and b match using strong ETag comparison.
// Assumes a and b are valid ETags.
func etagStrongMatch ( a , b string ) bool {
return a == b && a != "" && a [ 0 ] == '"'
}
// etagWeakMatch reports whether a and b match using weak ETag comparison.
// Assumes a and b are valid ETags.
func etagWeakMatch ( a , b string ) bool {
return strings . TrimPrefix ( a , "W/" ) == strings . TrimPrefix ( b , "W/" )
}
// condResult is the result of an HTTP request precondition check.
// See https://tools.ietf.org/html/rfc7232 section 3.
type condResult int
const (
condNone condResult = iota
condTrue
condFalse
)
func checkIfMatch ( ctx * Context ) condResult {
im := ctx . Request . Header . Get ( "If-Match" )
if im == "" {
return condNone
}
for {
im = textproto . TrimString ( im )
if len ( im ) == 0 {
break
}
if im [ 0 ] == ',' {
im = im [ 1 : ]
continue
}
if im [ 0 ] == '*' {
return condTrue
}
etag , remain := scanETag ( im )
if etag == "" {
break
}
if etagStrongMatch ( etag , ctx . ResponseWriter . Header ( ) . Get ( "Etag" ) ) {
return condTrue
}
im = remain
}
return condFalse
}
func checkIfUnmodifiedSince ( ctx * Context , modtime time . Time ) condResult {
ius := ctx . Request . Header . Get ( "If-Unmodified-Since" )
if ius == "" || isZeroTime ( modtime ) {
return condNone
}
if t , err := http . ParseTime ( ius ) ; err == nil {
// The Date-Modified header truncates sub-second precision, so
// use mtime < t+1s instead of mtime <= t to check for unmodified.
if modtime . Before ( t . Add ( 1 * time . Second ) ) {
return condTrue
}
return condFalse
}
return condNone
}
func checkIfNoneMatch ( ctx * Context ) condResult {
inm := ctx . Request . Header . Get ( "If-None-Match" )
if inm == "" {
return condNone
}
buf := inm
for {
buf = textproto . TrimString ( buf )
if len ( buf ) == 0 {
break
}
if buf [ 0 ] == ',' {
buf = buf [ 1 : ]
}
if buf [ 0 ] == '*' {
return condFalse
}
etag , remain := scanETag ( buf )
if etag == "" {
break
}
if etagWeakMatch ( etag , ctx . ResponseWriter . Header ( ) . Get ( "Etag" ) ) {
return condFalse
}
buf = remain
}
return condTrue
}
func checkIfModifiedSince ( ctx * Context , modtime time . Time ) condResult {
if ctx . Method ( ) != MethodGet && ctx . Method ( ) != MethodHead {
return condNone
}
ims := ctx . Request . Header . Get ( "If-Modified-Since" )
if ims == "" || isZeroTime ( modtime ) {
return condNone
}
t , err := http . ParseTime ( ims )
if err != nil {
return condNone
}
// The Date-Modified header truncates sub-second precision, so
// use mtime < t+1s instead of mtime <= t to check for unmodified.
if modtime . Before ( t . Add ( 1 * time . Second ) ) {
return condFalse
}
return condTrue
}
func checkIfRange ( ctx * Context , modtime time . Time ) condResult {
if ctx . Method ( ) != MethodGet {
return condNone
}
ir := ctx . Request . Header . Get ( "If-Range" )
if ir == "" {
return condNone
}
etag , _ := scanETag ( ir )
if etag != "" {
if etagStrongMatch ( etag , ctx . ResponseWriter . Header ( ) . Get ( "Etag" ) ) {
return condTrue
}
return condFalse
}
// The If-Range value is typically the ETag value, but it may also be
// the modtime date. See golang.org/issue/8367.
if modtime . IsZero ( ) {
return condFalse
}
t , err := http . ParseTime ( ir )
if err != nil {
return condFalse
}
if t . Unix ( ) == modtime . Unix ( ) {
return condTrue
}
return condFalse
}
var unixEpochTime = time . Unix ( 0 , 0 )
// isZeroTime reports whether t is obviously unspecified (either zero or Unix()=0).
func isZeroTime ( t time . Time ) bool {
return t . IsZero ( ) || t . Equal ( unixEpochTime )
}
func setLastModified ( ctx * Context , modtime time . Time ) {
if ! isZeroTime ( modtime ) {
ctx . SetHeader ( "Last-Modified" , modtime . UTC ( ) . Format ( ctx . Framework ( ) . Config . TimeFormat ) )
}
}
func writeNotModified ( ctx * Context ) {
// RFC 7232 section 4.1:
// a sender SHOULD NOT generate representation metadata other than the
// above listed fields unless said metadata exists for the purpose of
// guiding cache updates (e.g., Last-Modified might be useful if the
// response does not have an ETag field).
h := ctx . ResponseWriter . Header ( )
delete ( h , contentType )
delete ( h , contentLength )
if h . Get ( "Etag" ) != "" {
delete ( h , "Last-Modified" )
}
ctx . SetStatusCode ( StatusNotModified )
}
// checkPreconditions evaluates request preconditions and reports whether a precondition
// resulted in sending StatusNotModified or StatusPreconditionFailed.
func checkPreconditions ( ctx * Context , modtime time . Time ) ( done bool , rangeHeader string ) {
// This function carefully follows RFC 7232 section 6.
ch := checkIfMatch ( ctx )
if ch == condNone {
ch = checkIfUnmodifiedSince ( ctx , modtime )
}
if ch == condFalse {
ctx . SetStatusCode ( StatusPreconditionFailed )
return true , ""
}
switch checkIfNoneMatch ( ctx ) {
case condFalse :
if ctx . Method ( ) == MethodGet || ctx . Method ( ) == MethodHead {
writeNotModified ( ctx )
return true , ""
}
ctx . SetStatusCode ( StatusPreconditionFailed )
return true , ""
case condNone :
if checkIfModifiedSince ( ctx , modtime ) == condFalse {
writeNotModified ( ctx )
return true , ""
}
}
rangeHeader = ctx . Request . Header . Get ( "Range" )
if rangeHeader != "" {
if checkIfRange ( ctx , modtime ) == condFalse {
rangeHeader = ""
}
}
return false , rangeHeader
}
// name is '/'-separated, not filepath.Separator.
func serveFile ( ctx * Context , fs http . FileSystem , name string , redirect bool , showList bool , gzip bool ) ( string , int ) {
const indexPage = "/index.html"
// redirect .../index.html to .../
// can't use Redirect() because that would make the path absolute,
// which would be a problem running under StripPrefix
if strings . HasSuffix ( ctx . Request . URL . Path , indexPage ) {
localRedirect ( ctx , "./" )
return "" , StatusMovedPermanently
}
f , err := fs . Open ( name )
if err != nil {
msg , code := toHTTPError ( err )
return msg , code
}
defer f . Close ( )
d , err := f . Stat ( )
if err != nil {
msg , code := toHTTPError ( err )
return msg , code
}
if redirect {
// redirect to canonical path: / at end of directory url
// ctx.Request.URL.Path always begins with /
url := ctx . Request . URL . Path
if d . IsDir ( ) {
if url [ len ( url ) - 1 ] != '/' {
localRedirect ( ctx , path . Base ( url ) + "/" )
return "" , StatusMovedPermanently
}
} else {
if url [ len ( url ) - 1 ] == '/' {
localRedirect ( ctx , "../" + path . Base ( url ) )
return "" , StatusMovedPermanently
}
}
}
// redirect if the directory name doesn't end in a slash
if d . IsDir ( ) {
url := ctx . Request . URL . Path
if url [ len ( url ) - 1 ] != '/' {
localRedirect ( ctx , path . Base ( url ) + "/" )
return "" , StatusMovedPermanently
}
}
// use contents of index.html for directory, if present
if d . IsDir ( ) {
index := strings . TrimSuffix ( name , "/" ) + indexPage
ff , err := fs . Open ( index )
if err == nil {
defer ff . Close ( )
dd , err := ff . Stat ( )
if err == nil {
name = index
d = dd
f = ff
}
}
}
// Still a directory? (we didn't find an index.html file)
if d . IsDir ( ) {
if ! showList {
return "" , StatusForbidden
}
if checkIfModifiedSince ( ctx , d . ModTime ( ) ) == condFalse {
writeNotModified ( ctx )
return "" , StatusNotModified
}
ctx . ResponseWriter . Header ( ) . Set ( "Last-Modified" , d . ModTime ( ) . UTC ( ) . Format ( ctx . Framework ( ) . Config . TimeFormat ) )
return dirList ( ctx , f )
}
// serveContent will check modification time
sizeFunc := func ( ) ( int64 , error ) { return d . Size ( ) , nil }
return serveContent ( ctx , d . Name ( ) , d . ModTime ( ) , sizeFunc , f , gzip )
}
// toHTTPError returns a non-specific HTTP error message and status code
// for a given non-nil error value. It's important that toHTTPError does not
// actually return err.Error(), since msg and httpStatus are returned to users,
// and historically Go's ServeContent always returned just "404 Not Found" for
// all errors. We don't want to start leaking information in error messages.
func toHTTPError ( err error ) ( msg string , httpStatus int ) {
if os . IsNotExist ( err ) {
return "404 page not found" , StatusNotFound
}
if os . IsPermission ( err ) {
return "403 Forbidden" , StatusForbidden
}
// Default:
return "500 Internal Server Error" , StatusInternalServerError
}
// localRedirect gives a Moved Permanently response.
// It does not convert relative paths to absolute paths like Redirect does.
func localRedirect ( ctx * Context , newPath string ) {
if q := ctx . Request . URL . RawQuery ; q != "" {
newPath += "?" + q
}
ctx . ResponseWriter . Header ( ) . Set ( "Location" , newPath )
ctx . SetStatusCode ( StatusMovedPermanently )
}
func containsDotDot ( v string ) bool {
if ! strings . Contains ( v , ".." ) {
return false
}
for _ , ent := range strings . FieldsFunc ( v , isSlashRune ) {
if ent == ".." {
return true
}
}
return false
}
func isSlashRune ( r rune ) bool { return r == '/' || r == '\\' }
// httpRange specifies the byte range to be sent to the client.
type httpRange struct {
start , length int64
}
func ( r httpRange ) contentRange ( size int64 ) string {
return fmt . Sprintf ( "bytes %d-%d/%d" , r . start , r . start + r . length - 1 , size )
}
func ( r httpRange ) mimeHeader ( contentType string , size int64 ) textproto . MIMEHeader {
return textproto . MIMEHeader {
"Content-Range" : { r . contentRange ( size ) } ,
contentType : { contentType } ,
}
}
// parseRange parses a Range header string as per RFC 2616.
// errNoOverlap is returned if none of the ranges overlap.
func parseRange ( s string , size int64 ) ( [ ] httpRange , error ) {
if s == "" {
return nil , nil // header not present
}
const b = "bytes="
if ! strings . HasPrefix ( s , b ) {
return nil , errors . New ( "invalid range" )
}
var ranges [ ] httpRange
noOverlap := false
for _ , ra := range strings . Split ( s [ len ( b ) : ] , "," ) {
ra = strings . TrimSpace ( ra )
if ra == "" {
continue
}
i := strings . Index ( ra , "-" )
if i < 0 {
return nil , errors . New ( "invalid range" )
}
start , end := strings . TrimSpace ( ra [ : i ] ) , strings . TrimSpace ( ra [ i + 1 : ] )
var r httpRange
if start == "" {
// If no start is specified, end specifies the
// range start relative to the end of the file.
i , err := strconv . ParseInt ( end , 10 , 64 )
if err != nil {
return nil , errors . New ( "invalid range" )
}
if i > size {
i = size
}
r . start = size - i
r . length = size - r . start
} else {
i , err := strconv . ParseInt ( start , 10 , 64 )
if err != nil || i < 0 {
return nil , errors . New ( "invalid range" )
}
if i >= size {
// If the range begins after the size of the content,
// then it does not overlap.
noOverlap = true
continue
}
r . start = i
if end == "" {
// If no end is specified, range extends to end of the file.
r . length = size - r . start
} else {
i , err := strconv . ParseInt ( end , 10 , 64 )
if err != nil || r . start > i {
return nil , errors . New ( "invalid range" )
}
if i >= size {
i = size - 1
}
r . length = i - r . start + 1
}
}
ranges = append ( ranges , r )
}
if noOverlap && len ( ranges ) == 0 {
// The specified ranges did not overlap with the content.
return nil , errNoOverlap
}
return ranges , nil
}
// countingWriter counts how many bytes have been written to it.
type countingWriter int64
func ( w * countingWriter ) Write ( p [ ] byte ) ( n int , err error ) {
* w += countingWriter ( len ( p ) )
return len ( p ) , nil
}
// rangesMIMESize returns the number of bytes it takes to encode the
// provided ranges as a multipart response.
func rangesMIMESize ( ranges [ ] httpRange , contentType string , contentSize int64 ) ( encSize int64 ) {
var w countingWriter
mw := multipart . NewWriter ( & w )
for _ , ra := range ranges {
mw . CreatePart ( ra . mimeHeader ( contentType , contentSize ) )
encSize += ra . length
}
mw . Close ( )
encSize += int64 ( w )
return
}
func sumRangesSize ( ranges [ ] httpRange ) ( size int64 ) {
for _ , ra := range ranges {
size += ra . length
}
return
}
2017-03-01 18:17:32 +01:00
// typeByExtension returns the MIME type associated with the file extension ext.
// The extension ext should begin with a leading dot, as in ".html".
// When ext has no associated type, typeByExtension returns "".
//
// Extensions are looked up first case-sensitively, then case-insensitively.
//
// The built-in table is small but on unix it is augmented by the local
// system's mime.types file(s) if available under one or more of these
// names:
//
// /etc/mime.types
// /etc/apache2/mime.types
// /etc/apache/mime.types
//
// On Windows, MIME types are extracted from the registry.
//
// Text types have the charset parameter set to "utf-8" by default.
func typeByExtension ( fullfilename string ) ( t string ) {
ext := filepath . Ext ( fullfilename )
//these should be found by the windows(registry) and unix(apache) but on windows some machines have problems on this part.
if t = mime . TypeByExtension ( ext ) ; t == "" {
// no use of map here because we will have to lock/unlock it, by hand is better, no problem:
if ext == ".json" {
t = "application/json"
} else if ext == ".js" {
t = "application/javascript"
} else if ext == ".zip" {
t = "application/zip"
} else if ext == ".3gp" {
t = "video/3gpp"
} else if ext == ".7z" {
t = "application/x-7z-compressed"
} else if ext == ".ace" {
t = "application/x-ace-compressed"
} else if ext == ".aac" {
t = "audio/x-aac"
} else if ext == ".ico" { // for any case
t = "image/x-icon"
} else if ext == ".png" {
t = "image/png"
} else {
t = "application/octet-stream"
}
// mime.TypeByExtension returns as text/plain; | charset=utf-8 the static .js (not always)
} else if t == "text/plain" || t == "text/plain; charset=utf-8" {
if ext == ".js" {
t = "application/javascript"
}
}
return
}
// directoryExists returns true if a directory(or file) exists, otherwise false
func directoryExists ( dir string ) bool {
if _ , err := os . Stat ( dir ) ; os . IsNotExist ( err ) {
return false
}
return true
}