Kozea/WeasyPrint
1
1
Fork 0
mirror of https://github.com/Kozea/WeasyPrint.git synced 2024-10-05 00:21:15 +03:00
Code Issues Projects Releases Wiki Activity

Merge branch 'master' into table-breaks

Browse Source 
Conflicts:
	weasyprint/css/validation.py
This commit is contained in:
Simon Sapin 2012-03-21 17:06:47 +01:00
commit e23c38bfd2
9 changed files with 97 additions and 70 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
CHANGES
View File

@ -2,6 +2,7 @@ Version 0.7, released on 2012-XX-XX
===================================
* Support for the ``orphans`` and ``widows`` properties.
* Support for ``page-break-inside: avoid``
Version 0.6.1, released on 2012-03-01

9
weasyprint/css/validation.py
View File

@ -629,9 +629,7 @@ def page_break(keyword):
@single_keyword
def page_break_inside(keyword):
"""Validation for the ``page-break-inside`` property."""
if keyword == 'avoid':
raise InvalidValues('value not supported yet')
return keyword in ('auto',)
return keyword in ('auto', 'avoid')
@validator()
@ -872,7 +870,10 @@ def generic_expander(*expanded_names):
results = {}
for new_name, new_values in wrapped(name, values):
assert new_name in expanded_names, new_name
assert new_name not in results, new_name
if new_name in results:
raise InvalidValues(
'got multiple %s values in a %s shorthand'
% (new_name.strip('_'), name))
results[new_name] = new_values
for new_name in expanded_names:

58
weasyprint/images.py
View File

@ -63,10 +63,6 @@ def cairosvg_handler(file_like, uri):
This handler uses CairoSVG: http://cairosvg.org/
"""
try:
import cairosvg as _
except ImportError as exception:
return exception
from cairosvg.surface import SVGSurface
from cairosvg.parser import Tree, ParseError
@ -84,12 +80,9 @@ def cairosvg_handler(file_like, uri):
# Don’t pass data URIs to CairoSVG.
# They are useless for relative URIs anyway.
uri = None
try:
# Draw to a cairo surface but do not write to a file
tree = Tree(file_obj=file_like, url=uri)
surface = ScaledSVGSurface(tree, output=None, dpi=96)
except (ParseError, NotImplementedError) as exception:
return exception
# Draw to a cairo surface but do not write to a file
tree = Tree(file_obj=file_like, url=uri)
surface = ScaledSVGSurface(tree, output=None, dpi=96)
pattern = cairo.SurfacePattern(surface.cairo)
return pattern, surface.width, surface.height
@ -101,50 +94,27 @@ def fallback_handler(file_like, uri):
PIL supports many raster image formats and does not take a `format`
parameter, it guesses the format from the content.
"""
try:
import pystacia as _
except ImportError as exception:
return exception
from pystacia import read_blob
from pystacia.util import TinyException
try:
with contextlib.closing(read_blob(file_like.read())) as image:
png_bytes = image.get_blob('png')
except TinyException as exception:
return exception
else:
return png_handler(BytesIO(png_bytes), uri)
with contextlib.closing(read_blob(file_like.read())) as image:
png_bytes = image.get_blob('png')
return png_handler(BytesIO(png_bytes), uri)
def get_image_from_uri(uri, type_=None):
"""Get a :class:`cairo.Surface`` from an image URI."""
try:
file_like, mime_type, _charset = urlopen(uri)
except (IOError, ValueError) as exc:
LOGGER.warn('Error while fetching an image from %s : %r', uri, exc)
return None
if not type_:
type_ = mime_type # Use eg. the HTTP header
#else: the type was forced by eg. a 'type' attribute on <embed>
handler = FORMAT_HANDLERS.get(type_, fallback_handler)
exception = None
try:
image = handler(file_like, uri)
except (IOError, MemoryError) as e:
exception = e # Network or parsing error
else:
if isinstance(image, Exception):
exception = image
if not type_:
type_ = mime_type # Use eg. the HTTP header
#else: the type was forced by eg. a 'type' attribute on <embed>
handler = FORMAT_HANDLERS.get(type_, fallback_handler)
return handler(file_like, uri)
except Exception as exc:
LOGGER.warn('Error for image at %s : %r', uri, exc)
finally:
try:
file_like.close()
except Exception:
# Do not hide a more relevant exception.
# May already be closed or something. This is just cleanup anyway.
pass
if exception is None:
return image
else:
LOGGER.warn('Error while parsing an image at %s : %r', uri, exception)
return None

13
weasyprint/layout/blocks.py
View File

@ -258,7 +258,7 @@ def block_level_height(document, box, max_position_y, skip_stack,
if is_page_break:
break
else:
if new_children and not page_is_empty:
if new_children:
# between siblings, but not before the first child
# or after the last child.
break_here, next_page = forced_page_break(
@ -271,7 +271,8 @@ def block_level_height(document, box, max_position_y, skip_stack,
(new_child, resume_at, next_page, next_adjoining_margins,
collapsing_through) = block_level_layout(
document, child, max_position_y, skip_stack,
new_containing_block, device_size, page_is_empty,
new_containing_block, device_size,
page_is_empty and not new_children,
adjoining_margins)
skip_stack = None
@ -293,7 +294,8 @@ def block_level_height(document, box, max_position_y, skip_stack,
new_position_y = (
new_child.border_box_y() + new_child.border_height())
if (new_position_y > max_position_y and not page_is_empty
if (new_position_y > max_position_y and (
new_children or not page_is_empty)
and not isinstance(child, boxes.BlockBox)):
# The child overflows the page area, put it on the
# next page. (But don’t delay whole blocks if eg.
@ -314,7 +316,6 @@ def block_level_height(document, box, max_position_y, skip_stack,
# Bottom borders may overflow here
# TODO: back-track somehow when all lines fit but not borders
new_children.append(new_child)
page_is_empty = False
if resume_at is not None:
resume_at = (index, resume_at)
break
@ -322,6 +323,10 @@ def block_level_height(document, box, max_position_y, skip_stack,
else:
resume_at = None
if resume_at is not None and box.style.page_break_inside == 'avoid' \
and not page_is_empty:
return None, None, 'any', [], False
if collapsing_with_children:
if new_children and not isinstance(

31
weasyprint/tests/test_css.py
View File

@ -34,6 +34,7 @@ from .testing_utils import (
from .. import css
from ..css.computed_values import used_line_height
from ..document import PNGDocument
from ..utils import parse_data_url
from .. import HTML
@ -45,6 +46,36 @@ def parse_html(filename, **kwargs):
return PNGDocument(html.root_element, **kwargs)
@assert_no_logs
def test_data_url():
"""Test URLs with the "data:" scheme."""
def parse(url, expected_content, expected_mime_type, expected_charset):
file_like, mime_type, charset = parse_data_url(url)
assert file_like.read() == expected_content
assert mime_type == expected_mime_type
assert charset == expected_charset
parse('data:,foo', b'foo', 'text/plain', 'US-ASCII')
parse('data:,foo%22bar', b'foo"bar', 'text/plain', 'US-ASCII')
parse('data:text/plain,foo', b'foo', 'text/plain', None)
parse('data:text/html;charset=utf8,<body>', b'<body>', 'text/html', 'utf8')
parse('data:text/plain;base64,Zm9v', b'foo', 'text/plain', None)
parse('data:text/plain;base64,Zm9vbw==', b'fooo', 'text/plain', None)
parse('data:text/plain;base64,Zm9vb28=', b'foooo', 'text/plain', None)
parse('data:text/plain;base64,Zm9vb29v', b'fooooo', 'text/plain', None)
parse('data:text/plain;base64,Zm9vbw%3D%3D', b'fooo', 'text/plain', None)
parse('data:text/plain;base64,Zm9vb28%3D', b'foooo', 'text/plain', None)
# "From a theoretical point of view, the padding character is not needed,
# since the number of missing bytes can be calculated from the number
# of Base64 digits."
# https://en.wikipedia.org/wiki/Base64#Padding
# The Acid 2 test uses base64 URLs without padding.
# http://acid2.acidtests.org/
parse('data:text/plain;base64,Zm9vbw', b'fooo', 'text/plain', None)
parse('data:text/plain;base64,Zm9vb28', b'foooo', 'text/plain', None)
@assert_no_logs
def test_style_dict():
"""Test a style in a ``dict``."""

2
weasyprint/tests/test_css_properties.py
View File

@ -145,6 +145,8 @@ def test_expand_list_style():
}
with raises(ValueError):
expand_to_dict('list_style', 'red')
with raises(ValueError):
expand_to_dict('list_style', 'circle disc')
def assert_background(css, **kwargs):

4
weasyprint/tests/test_draw.py
View File

@ -909,7 +909,7 @@ def test_images():
<div><img src="inexistent1.png" alt=""></div>
''')
assert len(logs) == 1
assert 'WARNING: Error while fetching an image' in logs[0]
assert 'WARNING: Error for image' in logs[0]
assert 'inexistent1.png' in logs[0]
assert_pixels('image_no_src', 8, 8, no_image, '''
<style>
@ -944,7 +944,7 @@ def test_images():
'''),
])
assert len(logs) == 1
assert 'WARNING: Error while fetching an image' in logs[0]
assert 'WARNING: Error for image' in logs[0]
assert 'inexistent2.png' in logs[0]

24
weasyprint/tests/test_layout.py
View File

@ -641,16 +641,16 @@ def test_page_breaks():
@assert_no_logs
def test_orphans_widows():
def test_orphans_widows_avoid():
"""Test orphans and widows control."""
def line_distribution(orphans, widows):
def line_distribution(css):
pages = parse('''
<style>
@page { -weasy-size: 200px }
h1 { height: 120px }
p { line-height: 20px;
width: 1px; /* line break at each word */
orphans: %s; widows: %s }
%s }
</style>
<h1>Tasty test</h1>
<!-- There is room for 4 lines after h1 on the fist page -->
@ -663,7 +663,7 @@ def test_orphans_widows():
six
seven
</p>
''' % (orphans, widows))
''' % css)
line_counts = []
for i, page in enumerate(pages):
html, = page.children
@ -679,10 +679,13 @@ def test_orphans_widows():
line_counts.append(0)
return line_counts
assert line_distribution(orphans=2, widows=2) == [4, 3]
assert line_distribution(orphans=5, widows=2) == [0, 7]
assert line_distribution(orphans=2, widows=4) == [3, 4]
assert line_distribution(orphans=4, widows=4) == [0, 7]
assert line_distribution('orphans: 2; widows: 2') == [4, 3]
assert line_distribution('orphans: 5; widows: 2') == [0, 7]
assert line_distribution('orphans: 2; widows: 4') == [3, 4]
assert line_distribution('orphans: 4; widows: 4') == [0, 7]
assert line_distribution(
'orphans: 2; widows: 2; page-break-inside: avoid') == [0, 7]
@assert_no_logs
@ -946,10 +949,7 @@ def test_images():
with capture_logs() as logs:
page, = parse("<p><img src='%s' alt='invalid image'>" % url)
assert len(logs) == 1
if url.startswith('data:') or 'really' in url:
assert 'WARNING: Error while parsing an image' in logs[0]
else:
assert 'WARNING: Error while fetching an image' in logs[0]
assert 'WARNING: Error for image' in logs[0]
html, = page.children
body, = html.children
paragraph, = body.children

25
weasyprint/utils.py
View File

@ -65,6 +65,25 @@ def ensure_url(string):
return path2url(string.encode('utf8'))
def decode_base64(data):
"""Decode base64, padding being optional.
"From a theoretical point of view, the padding character is not needed,
since the number of missing bytes can be calculated from the number
of Base64 digits."
https://en.wikipedia.org/wiki/Base64#Padding
:param data: Base64 data as an ASCII byte string
:returns: The decoded byte string.
"""
missing_padding = 4 - len(data) % 4
if missing_padding:
data += b'='* missing_padding
return base64.decodestring(data)
def parse_data_url(url):
"""Decode URLs with the 'data' stream. urllib can handle them
in Python 2, but that is broken in Python 3.
@ -98,11 +117,9 @@ def parse_data_url(url):
charset = 'US-ASCII'
encoding = ''
data = unquote_to_bytes(data)
if encoding == 'base64':
data = data.encode('ascii')
data = base64.decodestring(data)
else:
data = unquote_to_bytes(data)
data = decode_base64(data)
return io.BytesIO(data), mime_type, charset