Add utility methods and testcases to sphinx.util.images

sphinx/util/images.py

@@ -8,10 +8,16 @@
     :copyright: Copyright 2007-2017 by the Sphinx team, see AUTHORS.
     :license: BSD, see LICENSE for details.
 """
+from __future__ import absolute_import
 
+import base64
 import imghdr
 import imagesize
 from os import path
+from collections import OrderedDict
+
+from six import PY3, BytesIO, iteritems
+from typing import NamedTuple
 
 try:
     from PIL import Image        # check for the Python Imaging Library
@@ -23,13 +29,23 @@ except ImportError:
 
 if False:
     # For type annotation
-    from typing import Dict, List, Tuple  # NOQA
+    from typing import Dict, IO, List, Tuple  # NOQA
 
-mime_suffixes = {
-    '.pdf': 'application/pdf',
-    '.svg': 'image/svg+xml',
-    '.svgz': 'image/svg+xml',
-}  # type: Dict[unicode, unicode]
+if PY3:
+    unicode = str  # special alias for static typing...
+
+mime_suffixes = OrderedDict([
+    ('.gif', 'image/gif'),
+    ('.jpg', 'image/jpeg'),
+    ('.png', 'image/png'),
+    ('.pdf', 'application/pdf'),
+    ('.svg', 'image/svg+xml'),
+    ('.svgz', 'image/svg+xml'),
+])  # type: Dict[unicode, unicode]
+
+DataURI = NamedTuple('DataURI', [('mimetype', unicode),
+                                 ('charset', unicode),
+                                 ('data', bytes)])
 
 
 def get_image_size(filename):
@@ -52,15 +68,55 @@ def get_image_size(filename):
         return None
 
 
-def guess_mimetype(filename, default=None):
-    # type: (unicode, unicode) -> unicode
-    _, ext = path.splitext(filename)
-    if ext in mime_suffixes:
-        return mime_suffixes[ext]
-    else:
-        with open(filename, 'rb') as f:
-            imgtype = imghdr.what(f)
+def guess_mimetype_for_stream(stream, default=None):
+    # type: (IO, unicode) -> unicode
+    imgtype = imghdr.what(stream)
     if imgtype:
         return 'image/' + imgtype
+    else:
+        return default
+
+
+def guess_mimetype(filename='', content=None, default=None):
+    # type: (unicode, unicode, unicode) -> unicode
+    _, ext = path.splitext(filename.lower())
+    if ext in mime_suffixes:
+        return mime_suffixes[ext]
+    elif content:
+        return guess_mimetype_for_stream(BytesIO(content), default=default)
+    elif path.exists(filename):
+        with open(filename, 'rb') as f:
+            return guess_mimetype_for_stream(f, default=default)
 
     return default
+
+
+def get_image_extension(mimetype):
+    # type: (unicode) -> unicode
+    for ext, _mimetype in iteritems(mime_suffixes):
+        if mimetype == _mimetype:
+            return ext
+
+    return None
+
+
+def parse_data_uri(uri):
+    # type: (unicode) -> DataURI
+    if not uri.startswith('data:'):
+        return None
+
+    # data:[<MIME-type>][;charset=<encoding>][;base64],<data>
+    mimetype = u'text/plain'
+    charset = u'US-ASCII'
+
+    properties, data = uri[5:].split(',', 1)
+    for prop in properties.split(';'):
+        if prop == 'base64':
+            pass  # skip
+        elif prop.startswith('charset='):
+            charset = prop[8:]
+        elif prop:
+            mimetype = prop
+
+    image_data = base64.b64decode(data)  # type: ignore
+    return DataURI(mimetype, charset, image_data)

tests/test_util_images.py

@@ -0,0 +1,97 @@
+# -*- coding: utf-8 -*-
+"""
+    test_util_images
+    ~~~~~~~~~~~~~~~~
+
+    Test images util.
+
+    :copyright: Copyright 2007-2017 by the Sphinx team, see AUTHORS.
+    :license: BSD, see LICENSE for details.
+"""
+from __future__ import print_function
+
+import pytest
+
+from sphinx.util.images import (
+    get_image_size, guess_mimetype, get_image_extension, parse_data_uri
+)
+
+from util import rootdir
+
+
+GIF_FILENAME = rootdir / 'root' / 'img.gif'
+PNG_FILENAME = rootdir / 'root' / 'img.png'
+PDF_FILENAME = rootdir / 'root' / 'img.pdf'
+TXT_FILENAME = rootdir / 'root' / 'contents.txt'
+
+
+def test_get_image_size():
+    assert get_image_size(GIF_FILENAME) == (200, 181)
+    assert get_image_size(PNG_FILENAME) == (200, 181)
+    assert get_image_size(PDF_FILENAME) is None
+    assert get_image_size(TXT_FILENAME) is None
+
+
+def test_guess_mimetype():
+    # guess by filename
+    assert guess_mimetype('img.png') == 'image/png'
+    assert guess_mimetype('img.jpg') == 'image/jpeg'
+    assert guess_mimetype('img.txt') is None
+    assert guess_mimetype('img.txt', default='text/plain') == 'text/plain'
+    assert guess_mimetype('no_extension') is None
+    assert guess_mimetype('IMG.PNG') == 'image/png'
+
+    # guess by content
+    assert guess_mimetype(content=GIF_FILENAME.bytes()) == 'image/gif'
+    assert guess_mimetype(content=PNG_FILENAME.bytes()) == 'image/png'
+    assert guess_mimetype(content=PDF_FILENAME.bytes()) is None
+    assert guess_mimetype(content=TXT_FILENAME.bytes()) is None
+    assert guess_mimetype(content=TXT_FILENAME.bytes(), default='text/plain') == 'text/plain'
+
+    # the priority of params: filename > content > default
+    assert guess_mimetype('img.png',
+                          content=GIF_FILENAME.bytes(),
+                          default='text/plain') == 'image/png'
+    assert guess_mimetype('no_extension',
+                          content=GIF_FILENAME.bytes(),
+                          default='text/plain') == 'image/gif'
+    assert guess_mimetype('no_extension',
+                          content=TXT_FILENAME.bytes(),
+                          default='text/plain') == 'text/plain'
+
+
+def test_get_image_extension():
+    assert get_image_extension('image/png') == '.png'
+    assert get_image_extension('image/jpeg') == '.jpg'
+    assert get_image_extension('image/svg+xml') == '.svg'
+    assert get_image_extension('text/plain') is None
+
+
+def test_parse_data_uri():
+    # standard case
+    uri = ("data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUAAAAFCAYAAACNbyblAAAAHElEQVQI12P4"
+           "//8/w38GIAXDIBKE0DHxgljNBAAO9TXL0Y4OHwAAAABJRU5ErkJggg==")
+    image = parse_data_uri(uri)
+    assert image is not None
+    assert image.mimetype == 'image/png'
+    assert image.charset == 'US-ASCII'
+
+    # no mimetype
+    uri = ("data:charset=utf-8,iVBORw0KGgoAAAANSUhEUgAAAAUAAAAFCAYAAACNbyblAAAAHElE"
+           "QVQI12P4//8/w38GIAXDIBKE0DHxgljNBAAO9TXL0Y4OHwAAAABJRU5ErkJggg==")
+    image = parse_data_uri(uri)
+    assert image is not None
+    assert image.mimetype == 'text/plain'
+    assert image.charset == 'utf-8'
+
+    # non data URI
+    uri = ("image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAUAAAAFCAYAAACNbyblAAAAHElEQVQI12P4"
+           "//8/w38GIAXDIBKE0DHxgljNBAAO9TXL0Y4OHwAAAABJRU5ErkJggg==")
+    image = parse_data_uri(uri)
+    assert image is None
+
+    # invalid data URI (no properties)
+    with pytest.raises(ValueError):
+        uri = ("data:iVBORw0KGgoAAAANSUhEUgAAAAUAAAAFCAYAAACNbyblAAAAHElEQVQI12P4"
+               "//8/w38GIAXDIBKE0DHxgljNBAAO9TXL0Y4OHwAAAABJRU5ErkJggg==")
+        parse_data_uri(uri)