1
1
mirror of https://github.com/go-gitea/gitea synced 2024-12-23 17:14:27 +00:00
gitea/modules/typesniffer/typesniffer.go
KN4CK3R 8e262104c2
Add Image Diff for SVG files (#14867)
* Added type sniffer.

* Switched content detection from base to typesniffer.

* Added GuessContentType to Blob.

* Moved image info logic to client.
Added support for SVG images in diff.

* Restore old blocked svg behaviour.

* Added missing image formats.

* Execute image diff only when container is visible.

* add margin to spinner

* improve BIN tag on image diffs

* Default to render view.

* Show image diff on incomplete diff.

Co-authored-by: silverwind <me@silverwind.io>
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: Lauris BH <lauris@nix.lv>
2021-06-05 15:32:19 +03:00

97 lines
2.7 KiB
Go

// Copyright 2021 The Gitea Authors. All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package typesniffer
import (
"fmt"
"io"
"net/http"
"regexp"
"strings"
)
// Use at most this many bytes to determine Content Type.
const sniffLen = 1024
// SvgMimeType MIME type of SVG images.
const SvgMimeType = "image/svg+xml"
var svgTagRegex = regexp.MustCompile(`(?si)\A\s*(?:(<!--.*?-->|<!DOCTYPE\s+svg([\s:]+.*?>|>))\s*)*<svg[\s>\/]`)
var svgTagInXMLRegex = regexp.MustCompile(`(?si)\A<\?xml\b.*?\?>\s*(?:(<!--.*?-->|<!DOCTYPE\s+svg([\s:]+.*?>|>))\s*)*<svg[\s>\/]`)
// SniffedType contains informations about a blobs type.
type SniffedType struct {
contentType string
}
// IsText etects if content format is plain text.
func (ct SniffedType) IsText() bool {
return strings.Contains(ct.contentType, "text/")
}
// IsImage detects if data is an image format
func (ct SniffedType) IsImage() bool {
return strings.Contains(ct.contentType, "image/")
}
// IsSvgImage detects if data is an SVG image format
func (ct SniffedType) IsSvgImage() bool {
return strings.Contains(ct.contentType, SvgMimeType)
}
// IsPDF detects if data is a PDF format
func (ct SniffedType) IsPDF() bool {
return strings.Contains(ct.contentType, "application/pdf")
}
// IsVideo detects if data is an video format
func (ct SniffedType) IsVideo() bool {
return strings.Contains(ct.contentType, "video/")
}
// IsAudio detects if data is an video format
func (ct SniffedType) IsAudio() bool {
return strings.Contains(ct.contentType, "audio/")
}
// IsRepresentableAsText returns true if file content can be represented as
// plain text or is empty.
func (ct SniffedType) IsRepresentableAsText() bool {
return ct.IsText() || ct.IsSvgImage()
}
// DetectContentType extends http.DetectContentType with more content types. Defaults to text/unknown if input is empty.
func DetectContentType(data []byte) SniffedType {
if len(data) == 0 {
return SniffedType{"text/unknown"}
}
ct := http.DetectContentType(data)
if len(data) > sniffLen {
data = data[:sniffLen]
}
if (strings.Contains(ct, "text/plain") || strings.Contains(ct, "text/html")) && svgTagRegex.Match(data) ||
strings.Contains(ct, "text/xml") && svgTagInXMLRegex.Match(data) {
// SVG is unsupported. https://github.com/golang/go/issues/15888
ct = SvgMimeType
}
return SniffedType{ct}
}
// DetectContentTypeFromReader guesses the content type contained in the reader.
func DetectContentTypeFromReader(r io.Reader) (SniffedType, error) {
buf := make([]byte, sniffLen)
n, err := r.Read(buf)
if err != nil && err != io.EOF {
return SniffedType{}, fmt.Errorf("DetectContentTypeFromReader io error: %w", err)
}
buf = buf[:n]
return DetectContentType(buf), nil
}