Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

add html utf8 bom signature #268

Merged
merged 8 commits into from
Apr 17, 2022
Merged

add html utf8 bom signature #268

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
8ae06f3
add html utf8 bom signature
napalu Mar 31, 2022
bdb9f8f
Add utf8 BOM+whitespace testcase for html detection
gabriel-vasile Apr 3, 2022
0d5037a
Merge branch 'gabriel-vasile:master' into bugfix/html_utf8_bom
napalu Apr 5, 2022
bc9575b
account for whitespace after BOM in html
napalu Apr 5, 2022
7abfb09
strip BOM in magic.go
napalu Apr 6, 2022
29effd9
update FromHTML comment with modified behaviour
napalu Apr 6, 2022
e4f5c6c
remove superfluous BOM spec in markup
napalu Apr 7, 2022
b0140e7
update comment on UTF-8 BOM stripping
napalu Apr 10, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions internal/charset/charset.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -160,6 +160,9 @@ func fromXML(content []byte) string {
// FromHTML returns the charset of an HTML document. It relies on the meta tag
// <meta charset="UTF-8"> and falls back on the plain text content.
func FromHTML(content []byte) string {
if cset := FromBOM(content); cset != "" {
return cset
}
if cset := fromHTML(content); cset != "" {
return cset
}
Expand Down
30 changes: 30 additions & 0 deletions internal/charset/charset_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,29 @@ const htmlDoc = `<!DOCTYPE html>
<div class="container footer">さ</div>
</body>
</html>`
const htmlDocWithIncorrectCharset = `<!DOCTYPE html>
<!--
Some comment

-->
<html dir="ltr" mozdisallowselectionprint>
<head>
<meta charset="ISO-8859-16">
<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1">
<meta name="some name" content="notranslate">
<title>test</title>


<link rel="stylesheet" href="html.utf8bom.css">



</head>

<body tabindex="1">
<div id="printContainer"></div>
</body>
</html>`

func TestFromXML(t *testing.T) {
charset := FromXML([]byte(xmlDoc))
Expand All @@ -39,6 +62,13 @@ func TestFromHTML(t *testing.T) {
}
}

func TestFromHTMLWithBOM(t *testing.T) {
charset := FromHTML(append([]byte{0xEF, 0xBB, 0xBF}, []byte(htmlDocWithIncorrectCharset)...))
if charset != "utf-8" {
t.Errorf("expected: utf-8; got: %s", charset)
}
}

func TestFromPlain(t *testing.T) {
tcases := []struct {
raw []byte
Expand Down
2 changes: 1 addition & 1 deletion internal/magic/magic.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,7 +107,7 @@ func markup(sigs ...[]byte) Detector {
if bytes.HasPrefix(raw, []byte{0xEF, 0xBB, 0xBF}) {
// we restore BOM after stripping WS so its presence can be used
napalu marked this conversation as resolved.
Show resolved Hide resolved
// in subsequent functions
raw = append([]byte{0xEF, 0xBB, 0xBF}, trimLWS(raw[3:])...)
raw = trimLWS(raw[3:])
} else {
raw = trimLWS(raw)
}
Expand Down
159 changes: 80 additions & 79 deletions mimetype_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,85 +18,86 @@ const testDataDir = "testdata"

// test files sorted by the file name in alphabetical order.
var files = map[string]string{
"3g2.3g2": "video/3gpp2",
"3gp.3gp": "video/3gpp",
"3mf.3mf": "application/vnd.ms-package.3dmanufacturing-3dmodel+xml",
"7z.7z": "application/x-7z-compressed",
"a.a": "application/x-archive",
"aac.aac": "audio/aac",
"aaf.aaf": "application/octet-stream",
"accdb.accdb": "application/x-msaccess",
"aiff.aiff": "audio/aiff",
"amf.amf": "application/x-amf",
"amr.amr": "audio/amr",
"ape.ape": "audio/ape",
"apng.png": "image/vnd.mozilla.apng",
"asf.asf": "video/x-ms-asf",
"atom.atom": "application/atom+xml",
"au.au": "audio/basic",
"avi.avi": "video/x-msvideo",
"avif.avif": "image/avif",
"avifsequence.avif": "image/avif",
"bmp.bmp": "image/bmp",
"bpg.bpg": "image/bpg",
"bz2.bz2": "application/x-bzip2",
"cab.cab": "application/vnd.ms-cab-compressed",
"class.class": "application/x-java-applet",
"crx.crx": "application/x-chrome-extension",
"csv.csv": "text/csv",
"cpio.cpio": "application/x-cpio",
"dae.dae": "model/vnd.collada+xml",
"dbf.dbf": "application/x-dbf",
"dcm.dcm": "application/dicom",
"deb.deb": "application/vnd.debian.binary-package",
"djvu.djvu": "image/vnd.djvu",
"doc.doc": "application/msword",
"docx.1.docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
"docx.docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
"drpm.rpm": "application/x-rpm",
"dwg.1.dwg": "image/vnd.dwg",
"dwg.dwg": "image/vnd.dwg",
"eot.eot": "application/vnd.ms-fontobject",
"epub.epub": "application/epub+zip",
"exe.exe": "application/vnd.microsoft.portable-executable",
"fdf.fdf": "application/vnd.fdf",
"fits.fits": "application/fits",
"flac.flac": "audio/flac",
"flv.flv": "video/x-flv",
"gbr.gbr": "image/x-gimp-gbr",
"geojson.1.geojson": "application/geo+json",
"geojson.geojson": "application/geo+json",
"gif.gif": "image/gif",
"glb.glb": "model/gltf-binary",
"gml.gml": "application/gml+xml",
"gpx.gpx": "application/gpx+xml",
"gz.gz": "application/gzip",
"har.har": "application/json",
"hdr.hdr": "image/vnd.radiance",
"heic.single.heic": "image/heic",
"heif.heif": "image/heif",
"html.html": "text/html; charset=utf-8",
"html.iso88591.html": "text/html; charset=iso-8859-1",
"html.svg.html": "text/html; charset=utf-8",
"html.usascii.html": "text/html; charset=us-ascii",
"html.utf8.html": "text/html; charset=utf-8",
"html.utf8bom.html": "text/html; charset=utf-8",
"html.utf8bomws.html": "text/html; charset=utf-8",
"html.withbr.html": "text/html; charset=utf-8",
"ico.ico": "image/x-icon",
"ics.dos.ics": "text/calendar",
"ics.ics": "text/calendar",
"iso88591.txt": "text/plain; charset=iso-8859-1",
"jar.jar": "application/jar",
"jp2.jp2": "image/jp2",
"jpf.jpf": "image/jpx",
"jpg.jpg": "image/jpeg",
"jpm.jpm": "image/jpm",
"jxl.jxl": "image/jxl",
"xpm.xpm": "image/x-xpixmap",
"js.js": "application/javascript",
"json.json": "application/json",
"json.lowascii.json": "application/json",
"3g2.3g2": "video/3gpp2",
"3gp.3gp": "video/3gpp",
"3mf.3mf": "application/vnd.ms-package.3dmanufacturing-3dmodel+xml",
"7z.7z": "application/x-7z-compressed",
"a.a": "application/x-archive",
"aac.aac": "audio/aac",
"aaf.aaf": "application/octet-stream",
"accdb.accdb": "application/x-msaccess",
"aiff.aiff": "audio/aiff",
"amf.amf": "application/x-amf",
"amr.amr": "audio/amr",
"ape.ape": "audio/ape",
"apng.png": "image/vnd.mozilla.apng",
"asf.asf": "video/x-ms-asf",
"atom.atom": "application/atom+xml",
"au.au": "audio/basic",
"avi.avi": "video/x-msvideo",
"avif.avif": "image/avif",
"avifsequence.avif": "image/avif",
"bmp.bmp": "image/bmp",
"bpg.bpg": "image/bpg",
"bz2.bz2": "application/x-bzip2",
"cab.cab": "application/vnd.ms-cab-compressed",
"class.class": "application/x-java-applet",
"crx.crx": "application/x-chrome-extension",
"csv.csv": "text/csv",
"cpio.cpio": "application/x-cpio",
"dae.dae": "model/vnd.collada+xml",
"dbf.dbf": "application/x-dbf",
"dcm.dcm": "application/dicom",
"deb.deb": "application/vnd.debian.binary-package",
"djvu.djvu": "image/vnd.djvu",
"doc.doc": "application/msword",
"docx.1.docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
"docx.docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
"drpm.rpm": "application/x-rpm",
"dwg.1.dwg": "image/vnd.dwg",
"dwg.dwg": "image/vnd.dwg",
"eot.eot": "application/vnd.ms-fontobject",
"epub.epub": "application/epub+zip",
"exe.exe": "application/vnd.microsoft.portable-executable",
"fdf.fdf": "application/vnd.fdf",
"fits.fits": "application/fits",
"flac.flac": "audio/flac",
"flv.flv": "video/x-flv",
"gbr.gbr": "image/x-gimp-gbr",
"geojson.1.geojson": "application/geo+json",
"geojson.geojson": "application/geo+json",
"gif.gif": "image/gif",
"glb.glb": "model/gltf-binary",
"gml.gml": "application/gml+xml",
"gpx.gpx": "application/gpx+xml",
"gz.gz": "application/gzip",
"har.har": "application/json",
"hdr.hdr": "image/vnd.radiance",
"heic.single.heic": "image/heic",
"heif.heif": "image/heif",
"html.html": "text/html; charset=utf-8",
"html.iso88591.html": "text/html; charset=iso-8859-1",
"html.svg.html": "text/html; charset=utf-8",
"html.usascii.html": "text/html; charset=us-ascii",
"html.utf8.html": "text/html; charset=utf-8",
"html.utf8bom.html": "text/html; charset=utf-8",
"html.utf8bomws.html": "text/html; charset=utf-8",
"html.utf8bomdetect.html": "text/html; charset=utf-8",
"html.withbr.html": "text/html; charset=utf-8",
"ico.ico": "image/x-icon",
"ics.dos.ics": "text/calendar",
"ics.ics": "text/calendar",
"iso88591.txt": "text/plain; charset=iso-8859-1",
"jar.jar": "application/jar",
"jp2.jp2": "image/jp2",
"jpf.jpf": "image/jpx",
"jpg.jpg": "image/jpeg",
"jpm.jpm": "image/jpm",
"jxl.jxl": "image/jxl",
"xpm.xpm": "image/x-xpixmap",
"js.js": "application/javascript",
"json.json": "application/json",
"json.lowascii.json": "application/json",
// json.{int,float,string}.txt contain a single JSON value. They are valid JSON
// documents, but they should not be detected as application/json. This mimics
// the behaviour of the file utility and seems the correct thing to do.
Expand Down
24 changes: 24 additions & 0 deletions testdata/html.utf8bomdetect.html
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,24 @@

<!DOCTYPE html>
<!--
Some comment

-->
<html dir="ltr" mozdisallowselectionprint>
<head>
<meta charset="ISO-8859-16">
<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1">
<meta name="some name" content="notranslate">
<title>test</title>


<link rel="stylesheet" href="html.utf8bom.css">



</head>

<body tabindex="1">
<div id="printContainer"></div>
</body>
</html>