def build_pdf(output_dir):
"""Build PDF versions of each content page."""
for html_file in files(output_dir, '*.html', label='Building PDFs'):
html = weasyprint.HTML(filename=html_file)
pdf_path = Path(output_dir,
html_file.name.replace('.html', '.pdf'))
html.write_pdf(pdf_path)
def test_python_render():
"""Test rendering with the Python API."""
html = TestHTML(string='<body><img src=pattern.png>',
base_url=resource_filename('dummy.html'))
css = CSS(string='''
@page { margin: 2px; size: 8px; background: #fff }
body { margin: 0; font-size: 0 }
img { image-rendering: optimizeSpeed }
''')
png_bytes = html.write_png(stylesheets=[css])
pdf_bytes = html.write_pdf(stylesheets=[css])
assert png_bytes.startswith(b'\211PNG\r\n\032\n')
assert pdf_bytes.startswith(b'%PDF')
check_png_pattern(png_bytes)
# TODO: check PDF content? How?
class fake_file(object):
def __init__(self):
self.chunks = []
self.write = self.chunks.append
def getvalue(self):
return b''.join(self.chunks)
png_file = fake_file()
html.write_png(png_file, stylesheets=[css])
assert png_file.getvalue() == png_bytes
pdf_file = fake_file()
html.write_pdf(pdf_file, stylesheets=[css])
assert pdf_file.getvalue() == pdf_bytes
with temp_directory() as temp:
png_filename = os.path.join(temp, '1.png')
pdf_filename = os.path.join(temp, '1.pdf')
html.write_png(png_filename, stylesheets=[css])
html.write_pdf(pdf_filename, stylesheets=[css])
assert read_file(png_filename) == png_bytes
assert read_file(pdf_filename) == pdf_bytes
png_filename = os.path.join(temp, '2.png')
pdf_filename = os.path.join(temp, '2.pdf')
with open(png_filename, 'wb') as png_file:
html.write_png(png_file, stylesheets=[css])
with open(pdf_filename, 'wb') as pdf_file:
html.write_pdf(pdf_file, stylesheets=[css])
assert read_file(png_filename) == png_bytes
assert read_file(pdf_filename) == pdf_bytes
x2_png_bytes = html.write_png(stylesheets=[css], resolution=192)
check_png_pattern(x2_png_bytes, x2=True)
pages = list(html.get_png_pages(stylesheets=[css]))
assert pages == [(8, 8, png_bytes)]
pages = list(html.get_png_pages(stylesheets=[css], resolution=192))
assert pages == [(16, 16, x2_png_bytes)]
def test_low_level_api():
html = TestHTML(string="<body>")
css = CSS(
string="""
@page { margin: 2px; size: 8px; background: #fff }
html { background: #00f; }
body { background: #f00; width: 1px; height: 1px }
"""
)
pdf_bytes = html.write_pdf(stylesheets=[css])
assert pdf_bytes.startswith(b"%PDF")
assert html.render([css]).write_pdf() == pdf_bytes
png_bytes = html.write_png(stylesheets=[css])
document = html.render([css], enable_hinting=True)
page, = document.pages
assert page.width == 8
assert page.height == 8
assert document.write_png() == (png_bytes, 8, 8)
assert document.copy([page]).write_png() == (png_bytes, 8, 8)
surface = cairo.ImageSurface(cairo.FORMAT_ARGB32, 8, 8)
page.paint(cairo.Context(surface))
file_obj = io.BytesIO()
surface.write_to_png(file_obj)
check_png_pattern(file_obj.getvalue())
surface = cairo.ImageSurface(cairo.FORMAT_ARGB32, 8, 8)
context = cairo.Context(surface)
# Rotate at the center
context.translate(4, 4)
context.rotate(-math.pi / 2)
context.translate(-4, -4)
page.paint(context)
file_obj = io.BytesIO()
surface.write_to_png(file_obj)
check_png_pattern(file_obj.getvalue(), rotated=True)
document = html.render([css], enable_hinting=True)
page, = document.pages
assert (page.width, page.height) == (8, 8)
png_bytes, width, height = document.write_png(resolution=192)
assert (width, height) == (16, 16)
check_png_pattern(png_bytes, x2=True)
def png_size(result):
png_bytes, width, height = result
surface = cairo.ImageSurface.create_from_png(io.BytesIO(png_bytes))
assert (surface.get_width(), surface.get_height()) == (width, height)
return width, height
document = html.render([css], enable_hinting=True)
page, = document.pages
assert (page.width, page.height) == (8, 8)
# A resolution that is not multiple of 96:
assert png_size(document.write_png(resolution=145.2)) == (13, 13)
document = TestHTML(
string="""
<style>
@page:first { size: 5px 10px } @page { size: 6px 4px }
p { page-break-before: always }
</style>
<p></p>
<p></p>
"""
).render()
page_1, page_2 = document.pages
assert (page_1.width, page_1.height) == (5, 10)
assert (page_2.width, page_2.height) == (6, 4)
result = document.write_png()
# (Max of both widths, Sum of both heights)
assert png_size(result) == (6, 14)
assert document.copy([page_1, page_2]).write_png() == result
assert png_size(document.copy([page_1]).write_png()) == (5, 10)
assert png_size(document.copy([page_2]).write_png()) == (6, 4)
def test_python_render():
"""Test rendering with the Python API."""
base_url = resource_filename("dummy.html")
html_string = "<body><img src=pattern.png>"
css_string = """
@page { margin: 2px; size: 8px; background: #fff }
body { margin: 0; font-size: 0 }
img { image-rendering: optimizeSpeed }
@media screen { img { transform: rotate(-90deg) } }
"""
html = TestHTML(string=html_string, base_url=base_url)
css = CSS(string=css_string)
png_bytes = html.write_png(stylesheets=[css])
pdf_bytes = html.write_pdf(stylesheets=[css])
assert png_bytes.startswith(b"\211PNG\r\n\032\n")
assert pdf_bytes.startswith(b"%PDF")
check_png_pattern(png_bytes)
# TODO: check PDF content? How?
class fake_file(object):
def __init__(self):
self.chunks = []
def write(self, data):
self.chunks.append(bytes(data[:]))
def getvalue(self):
return b"".join(self.chunks)
png_file = fake_file()
html.write_png(png_file, stylesheets=[css])
assert png_file.getvalue() == png_bytes
pdf_file = fake_file()
html.write_pdf(pdf_file, stylesheets=[css])
assert pdf_file.getvalue() == pdf_bytes
with temp_directory() as temp:
png_filename = os.path.join(temp, "1.png")
pdf_filename = os.path.join(temp, "1.pdf")
html.write_png(png_filename, stylesheets=[css])
html.write_pdf(pdf_filename, stylesheets=[css])
assert read_file(png_filename) == png_bytes
assert read_file(pdf_filename) == pdf_bytes
png_filename = os.path.join(temp, "2.png")
pdf_filename = os.path.join(temp, "2.pdf")
with open(png_filename, "wb") as png_file:
html.write_png(png_file, stylesheets=[css])
with open(pdf_filename, "wb") as pdf_file:
html.write_pdf(pdf_file, stylesheets=[css])
assert read_file(png_filename) == png_bytes
assert read_file(pdf_filename) == pdf_bytes
x2_png_bytes = html.write_png(stylesheets=[css], resolution=192)
check_png_pattern(x2_png_bytes, x2=True)
screen_css = CSS(string=css_string, media_type="screen")
rotated_png_bytes = html.write_png(stylesheets=[screen_css])
check_png_pattern(rotated_png_bytes, rotated=True)
assert (
TestHTML(string=html_string, base_url=base_url, media_type="screen").write_png(
stylesheets=[io.BytesIO(css_string.encode("utf8"))]
)
== rotated_png_bytes
)
assert (
TestHTML(
string="<style>%s</style>%s" % (css_string, html_string), base_url=base_url, media_type="screen"
).write_png()
== rotated_png_bytes
)
def test_low_level_api():
html = TestHTML(string='<body>')
css = CSS(string='''
@page { margin: 2px; size: 8px; background: #fff }
html { background: #00f; }
body { background: #f00; width: 1px; height: 1px }
''')
pdf_bytes = html.write_pdf(stylesheets=[css])
assert pdf_bytes.startswith(b'%PDF')
assert html.render([css]).write_pdf() == pdf_bytes
png_bytes = html.write_png(stylesheets=[css])
document = html.render([css], enable_hinting=True)
page, = document.pages
assert page.width == 8
assert page.height == 8
assert document.write_png() == (png_bytes, 8, 8)
assert document.copy([page]).write_png() == (png_bytes, 8, 8)
surface = cairo.ImageSurface(cairo.FORMAT_ARGB32, 8, 8)
page.paint(cairo.Context(surface))
file_obj = io.BytesIO()
surface.write_to_png(file_obj)
check_png_pattern(file_obj.getvalue())
surface = cairo.ImageSurface(cairo.FORMAT_ARGB32, 8, 8)
context = cairo.Context(surface)
# Rotate at the center
context.translate(4, 4)
context.rotate(-math.pi / 2)
context.translate(-4, -4)
page.paint(context)
file_obj = io.BytesIO()
surface.write_to_png(file_obj)
check_png_pattern(file_obj.getvalue(), rotated=True)
document = html.render([css], enable_hinting=True)
page, = document.pages
assert (page.width, page.height) == (8, 8)
png_bytes, width, height = document.write_png(resolution=192)
assert (width, height) == (16, 16)
check_png_pattern(png_bytes, x2=True)
def png_size(result):
png_bytes, width, height = result
surface = cairo.ImageSurface.create_from_png(io.BytesIO(png_bytes))
assert (surface.get_width(), surface.get_height()) == (width, height)
return width, height
document = html.render([css], enable_hinting=True)
page, = document.pages
assert (page.width, page.height) == (8, 8)
# A resolution that is not multiple of 96:
assert png_size(document.write_png(resolution=145.2)) == (13, 13)
document = TestHTML(string='''
<style>
@page:first { size: 5px 10px } @page { size: 6px 4px }
p { page-break-before: always }
</style>
<p></p>
<p></p>
''').render()
page_1, page_2 = document.pages
assert (page_1.width, page_1.height) == (5, 10)
assert (page_2.width, page_2.height) == (6, 4)
result = document.write_png()
# (Max of both widths, Sum of both heights)
assert png_size(result) == (6, 14)
assert document.copy([page_1, page_2]).write_png() == result
assert png_size(document.copy([page_1]).write_png()) == (5, 10)
assert png_size(document.copy([page_2]).write_png()) == (6, 4)
def test_python_render():
"""Test rendering with the Python API."""
base_url = resource_filename('dummy.html')
html_string = '<body><img src=pattern.png>'
css_string = '''
@page { margin: 2px; size: 8px; background: #fff }
body { margin: 0; font-size: 0 }
img { image-rendering: pixelated }
@media screen { img { transform: rotate(-90deg) } }
'''
html = TestHTML(string=html_string, base_url=base_url)
css = CSS(string=css_string)
png_bytes = html.write_png(stylesheets=[css])
pdf_bytes = html.write_pdf(stylesheets=[css])
assert png_bytes.startswith(b'\211PNG\r\n\032\n')
assert pdf_bytes.startswith(b'%PDF')
check_png_pattern(png_bytes)
# TODO: check PDF content? How?
class fake_file(object):
def __init__(self):
self.chunks = []
def write(self, data):
self.chunks.append(bytes(data[:]))
def getvalue(self):
return b''.join(self.chunks)
png_file = fake_file()
html.write_png(png_file, stylesheets=[css])
assert png_file.getvalue() == png_bytes
pdf_file = fake_file()
html.write_pdf(pdf_file, stylesheets=[css])
assert pdf_file.getvalue() == pdf_bytes
with temp_directory() as temp:
png_filename = os.path.join(temp, '1.png')
pdf_filename = os.path.join(temp, '1.pdf')
html.write_png(png_filename, stylesheets=[css])
html.write_pdf(pdf_filename, stylesheets=[css])
assert read_file(png_filename) == png_bytes
assert read_file(pdf_filename) == pdf_bytes
png_filename = os.path.join(temp, '2.png')
pdf_filename = os.path.join(temp, '2.pdf')
with open(png_filename, 'wb') as png_file:
html.write_png(png_file, stylesheets=[css])
with open(pdf_filename, 'wb') as pdf_file:
html.write_pdf(pdf_file, stylesheets=[css])
assert read_file(png_filename) == png_bytes
assert read_file(pdf_filename) == pdf_bytes
x2_png_bytes = html.write_png(stylesheets=[css], resolution=192)
check_png_pattern(x2_png_bytes, x2=True)
screen_css = CSS(string=css_string, media_type='screen')
rotated_png_bytes = html.write_png(stylesheets=[screen_css])
check_png_pattern(rotated_png_bytes, rotated=True)
assert TestHTML(string=html_string, base_url=base_url,
media_type='screen').write_png(
stylesheets=[io.BytesIO(css_string.encode('utf8'))
]) == rotated_png_bytes
assert TestHTML(string='<style>%s</style>%s' % (css_string, html_string),
base_url=base_url,
media_type='screen').write_png() == rotated_png_bytes
def test_python_render():
"""Test rendering with the Python API."""
base_url = resource_filename('dummy.html')
html_string = '<body><img src=pattern.png>'
css_string = '''
@page { margin: 2px; size: 8px; background: #fff }
body { margin: 0; font-size: 0 }
img { image-rendering: optimizeSpeed }
@media screen { img { transform: rotate(-90deg) } }
'''
html = TestHTML(string=html_string, base_url=base_url)
css = CSS(string=css_string)
png_bytes = html.write_png(stylesheets=[css])
pdf_bytes = html.write_pdf(stylesheets=[css])
assert png_bytes.startswith(b'\211PNG\r\n\032\n')
assert pdf_bytes.startswith(b'%PDF')
check_png_pattern(png_bytes)
# TODO: check PDF content? How?
class fake_file(object):
def __init__(self):
self.chunks = []
self.write = self.chunks.append
def getvalue(self):
return b''.join(self.chunks)
png_file = fake_file()
html.write_png(png_file, stylesheets=[css])
assert png_file.getvalue() == png_bytes
pdf_file = fake_file()
html.write_pdf(pdf_file, stylesheets=[css])
assert pdf_file.getvalue() == pdf_bytes
with temp_directory() as temp:
png_filename = os.path.join(temp, '1.png')
pdf_filename = os.path.join(temp, '1.pdf')
html.write_png(png_filename, stylesheets=[css])
html.write_pdf(pdf_filename, stylesheets=[css])
assert read_file(png_filename) == png_bytes
assert read_file(pdf_filename) == pdf_bytes
png_filename = os.path.join(temp, '2.png')
pdf_filename = os.path.join(temp, '2.pdf')
with open(png_filename, 'wb') as png_file:
html.write_png(png_file, stylesheets=[css])
with open(pdf_filename, 'wb') as pdf_file:
html.write_pdf(pdf_file, stylesheets=[css])
assert read_file(png_filename) == png_bytes
assert read_file(pdf_filename) == pdf_bytes
x2_png_bytes = html.write_png(stylesheets=[css], resolution=192)
check_png_pattern(x2_png_bytes, x2=True)
screen_css = CSS(string=css_string, media_type='screen')
rotated_png_bytes = html.write_png(stylesheets=[screen_css])
check_png_pattern(rotated_png_bytes, rotated=True)
assert TestHTML(
string=html_string, base_url=base_url, media_type='screen'
).write_png(
stylesheets=[io.BytesIO(css_string.encode('utf8'))]
) == rotated_png_bytes
assert TestHTML(
string='<style>%s</style>%s' % (css_string, html_string),
base_url=base_url, media_type='screen'
).write_png() == rotated_png_bytes