def test_name_equality():
# Who needs transitivity?
# While this is less than ideal ('/Foo' != b'/Foo') it allows for slightly
# sloppy tests like if colorspace == '/Indexed' without requiring
# Name('/Indexed') everywhere
assert Name('/Foo') == '/Foo'
assert Name('/Foo') == b'/Foo'
def _from_pil_image(cls, *, pdf, page, name, image): # pragma: no cover
"""Insert a PIL image into a PDF (rudimentary)
Args:
pdf (pikepdf.Pdf): the PDF to attach the image to
page (pikepdf.Object): the page to attach the image to
name (str or pikepdf.Name): the name to set the image
image (PIL.Image.Image): the image to insert
"""
data = image.tobytes()
imstream = Stream(pdf, data)
imstream.Type = Name('/XObject')
imstream.Subtype = Name('/Image')
if image.mode == 'RGB':
imstream.ColorSpace = Name('/DeviceRGB')
elif image.mode in ('1', 'L'):
imstream.ColorSpace = Name('/DeviceGray')
imstream.BitsPerComponent = 1 if image.mode == '1' else 8
imstream.Width = image.width
imstream.Height = image.height
page.Resources.XObject[name] = imstream
return cls(imstream)
def test_image_roundtrip(outdir, w, h, pixeldata, cs, bpc):
pdf = Pdf.new()
image_data = pixeldata * (w * h)
image = Stream(pdf, image_data)
image.Type = Name('/XObject')
image.Subtype = Name('/Image')
image.ColorSpace = Name(cs)
image.BitsPerComponent = bpc
image.Width = w
image.Height = h
xobj = {'/Im1': image}
resources = {'/XObject': xobj}
mediabox = [0, 0, 100, 100]
stream = b'q 100 0 0 100 0 0 cm /Im1 Do Q'
contents = Stream(pdf, stream)
page_dict = {
'/Type': Name('/Page'),
'/MediaBox': mediabox,
'/Contents': contents,
'/Resources': resources,
}
page = pdf.make_indirect(page_dict)
pdf.pages.append(page)
outfile = outdir / f'test{w}{h}{cs[1:]}{bpc}.pdf'
pdf.save(
outfile, compress_streams=False, stream_decode_level=StreamDecodeLevel.none
)
with Pdf.open(outfile) as p2:
pim = PdfImage(p2.pages[0].Resources.XObject['/Im1'])
assert pim.bits_per_component == bpc
assert pim.colorspace == cs
assert pim.width == w
assert pim.height == h
if cs == '/DeviceRGB':
assert pim.mode == 'RGB'
elif cs == '/DeviceGray' and bpc == 8:
assert pim.mode == 'L'
elif cs == '/DeviceCMYK':
assert pim.mode == 'CMYK'
elif bpc == 1:
assert pim.mode == '1'
assert not pim.palette
assert pim.filters == []
assert pim.read_bytes() == pixeldata
outstream = BytesIO()
pim.extract_to(stream=outstream)
outstream.seek(0)
im = Image.open(outstream)
assert pim.mode == im.mode
def test_image_replace(congress, outdir):
pdfimage = PdfImage(congress[0])
pillowimage = pdfimage.as_pil_image()
grayscale = pillowimage.convert('L')
congress[0].write(zlib.compress(grayscale.tobytes()), Name("/FlateDecode"), Null())
congress[0].ColorSpace = Name("/DeviceGray")
pdf = congress[1]
pdf.save(outdir / 'congress_gray.pdf')
def test_image_replace(congress, outdir):
pdfimage = PdfImage(congress[0])
pillowimage = pdfimage.as_pil_image()
grayscale = pillowimage.convert('L')
grayscale = grayscale.resize((4, 4)) # So it is not obnoxious on error
congress[0].write(zlib.compress(grayscale.tobytes()), filter=Name("/FlateDecode"))
congress[0].ColorSpace = Name("/DeviceGray")
pdf = congress[1]
pdf.save(outdir / 'congress_gray.pdf')
def test_lowlevel_replace_jpeg(congress, outdir):
# This test will modify the PDF so needs its own image
raw_bytes = congress[0].read_raw_bytes()
im = Image.open(BytesIO(raw_bytes))
grayscale = im.convert('L')
congress[0].write(zlib.compress(grayscale.tobytes()), Name("/FlateDecode"), Null())
congress[0].ColorSpace = Name('/DeviceGray')
pdf = congress[1]
pdf.save(outdir / 'congress_gray.pdf')
def test_bytes():
b = b'\x79\x78\x77\x76'
qs = String(b)
assert bytes(qs) == b
s = 'é'
qs = String(s)
assert str(qs) == s
assert Name('/xyz') == b'/xyz'
with pytest.raises(TypeError, match='should be str'):
Name(b'/bytes')
def test_lowlevel_replace_jpeg(congress, outdir):
# This test will modify the PDF so needs its own image
raw_bytes = congress[0].read_raw_bytes()
im = Image.open(BytesIO(raw_bytes))
grayscale = im.convert('L')
grayscale = grayscale.resize((4, 4)) # So it is not obnoxious on error
congress[0].write(zlib.compress(grayscale.tobytes()[:10]),
filter=Name("/FlateDecode"))
congress[0].ColorSpace = Name('/DeviceGray')
pdf = congress[1]
pdf.save(outdir / 'congress_gray.pdf')
def test_new_item(resources, title, page_num, page_loc):
doc = Pdf.open(resources / 'outlines.pdf')
kwargs = dict.fromkeys(ALL_PAGE_LOCATION_KWARGS, 100)
page_ref = doc.pages[page_num]
new_item = OutlineItem(title, page_num, page_loc, **kwargs)
with doc.open_outline() as outline:
outline.root.append(new_item)
if isinstance(page_loc, PageLocation):
loc_str = page_loc.name
else:
loc_str = page_loc
if loc_str == 'FitR':
kwarg_len = 4
elif loc_str == 'XYZ':
kwarg_len = 3
elif loc_str in ('FitH', 'FitV', 'FitBH', 'FitBV'):
kwarg_len = 1
else:
kwarg_len = 0
expected_dest = [page_ref, Name('/{0}'.format(loc_str))]
expected_dest.extend(repeat(100, kwarg_len))
assert new_item.destination == expected_dest
new_obj = new_item.obj
assert new_obj.Title == title
assert new_obj.Dest == expected_dest
assert new_obj.is_indirect is True
def test_new_item(resources, title, page_num, page_loc):
# @given precludes use of outlines_doc fixture - causes hypothesis health check to
# fail
with Pdf.open(resources / 'outlines.pdf') as doc:
kwargs = dict.fromkeys(ALL_PAGE_LOCATION_KWARGS, 100)
page_ref = doc.pages[page_num]
new_item = OutlineItem(title, page_num, page_loc, **kwargs)
with doc.open_outline() as outline:
outline.root.append(new_item)
if isinstance(page_loc, PageLocation):
loc_str = page_loc.name
else:
loc_str = page_loc
if loc_str == 'FitR':
kwarg_len = 4
elif loc_str == 'XYZ':
kwarg_len = 3
elif loc_str in ('FitH', 'FitV', 'FitBH', 'FitBV'):
kwarg_len = 1
else:
kwarg_len = 0
expected_dest = [page_ref, Name(f'/{loc_str}')]
expected_dest.extend(repeat(100, kwarg_len))
assert new_item.destination == expected_dest
new_obj = new_item.obj
assert new_obj.Title == title
assert new_obj.Dest == expected_dest
assert new_obj.is_indirect is True
def test_page_destination(resources, page_num, page_loc, kwargs):
# @given precludes use of outlines_doc fixture - causes hypothesis health check to
# fail
with Pdf.open(resources / 'outlines.pdf') as doc:
page_ref = doc.pages[page_num]
if page_loc == 'invalid':
with pytest.raises(ValueError, match='unsupported page location'):
make_page_destination(doc, page_num, page_loc, **kwargs)
return
dest = make_page_destination(doc, page_num, page_loc, **kwargs)
if isinstance(page_loc, PageLocation):
loc_str = page_loc.name
else:
loc_str = page_loc
if loc_str == 'XYZ':
args = 'left', 'top', 'zoom'
elif loc_str == 'FitH':
args = ('top', )
elif loc_str == 'FitV':
args = ('left', )
elif loc_str == 'FitR':
args = 'left', 'bottom', 'right', 'top'
elif loc_str == 'FitBH':
args = ('top', )
elif loc_str == 'FitBV':
args = ('left', )
else:
args = ()
expected_dest = [page_ref, Name(f'/{loc_str}')]
expected_dest.extend(kwargs.get(k, 0) for k in args)
assert dest == expected_dest
def make_page_destination(
pdf: Pdf,
page_num: int,
page_location: Optional[Union[PageLocation, str]] = None,
**kwargs,
) -> Array:
"""
Creates a destination ``Array`` with reference to a Pdf document's page number.
Arguments:
pdf: PDF document object.
page_num: Page number (zero-based).
page_location: Optional page location, as a string or :enum:`PageLocation`.
kwargs: Optional keyword arguments for the page location, e.g. ``top``.
"""
res = [pdf.pages[page_num]]
if page_location:
if isinstance(page_location, PageLocation):
loc_key = page_location
loc_str = loc_key.name
else:
loc_str = page_location
try:
loc_key = PageLocation[loc_str]
except KeyError:
raise ValueError(
f"Invalid or unsupported page location type {loc_str}")
res.append(Name(f'/{loc_str}'))
dest_arg_names = PAGE_LOCATION_ARGS.get(loc_key)
if dest_arg_names:
res.extend(kwargs.get(k, 0) for k in dest_arg_names)
else:
res.append(Name.Fit)
return Array(res)
def _make_page_destination(
pdf: Pdf,
page_num: int,
page_location: Optional[Union[PageLocation, str]] = None,
**kwargs,
) -> Array:
kwargs = {k: v for k, v in kwargs.items() if v is not None}
res = [pdf.pages[page_num]]
if page_location:
if isinstance(page_location, PageLocation):
loc_key = page_location
loc_str = loc_key.name
else:
loc_str = page_location
try:
loc_key = PageLocation[loc_str]
except KeyError:
raise ValueError(
f"Invalid or unsupported page location type {loc_str}"
) from None
res.append(Name(f'/{loc_str}'))
dest_arg_names = PAGE_LOCATION_ARGS.get(loc_key)
if dest_arg_names:
res.extend(kwargs.get(k, 0) for k in dest_arg_names)
else:
res.append(Name.Fit)
return Array(res)
def test_page_destination(resources, page_num, page_loc, kwargs):
doc = Pdf.open(resources / 'outlines.pdf')
page_ref = doc.pages[page_num]
dest = make_page_destination(doc, page_num, page_loc, **kwargs)
if isinstance(page_loc, PageLocation):
loc_str = page_loc.name
else:
loc_str = page_loc
if loc_str == 'XYZ':
args = 'left', 'top', 'zoom'
elif loc_str == 'FitH':
args = ('top', )
elif loc_str == 'FitV':
args = ('left', )
elif loc_str == 'FitR':
args = 'left', 'bottom', 'right', 'top'
elif loc_str == 'FitBH':
args = ('top', )
elif loc_str == 'FitBV':
args = ('left', )
else:
args = ()
expected_dest = [page_ref, Name('/{0}'.format(loc_str))]
expected_dest.extend(kwargs.get(k, 0) for k in args)
assert dest == expected_dest
def test_list_apis():
a = pikepdf.Array([1, 2, 3])
a[1] = None
assert a[1] is None
assert len(a) == 3
del a[1]
assert len(a) == 2
a[-1] = Name('/Foo')
def test_create_pdf(outdir):
pdf = Pdf.new()
font = pdf.make_indirect(
Object.parse(b"""
<<
/Type /Font
/Subtype /Type1
/Name /F1
/BaseFont /Helvetica
/Encoding /WinAnsiEncoding
>>"""))
width, height = 100, 100
image_data = b"\xff\x7f\x00" * (width * height)
image = Stream(pdf, image_data)
image.stream_dict = Object.parse(b"""
<<
/Type /XObject
/Subtype /Image
/ColorSpace /DeviceRGB
/BitsPerComponent 8
/Width 100
/Height 100
>>""")
rfont = {'/F1': font}
xobj = {'/Im1': image}
resources = {
'/Font': rfont,
'/XObject': xobj
}
mediabox = [0, 0, 612, 792]
stream = b"""
BT /F1 24 Tf 72 720 Td (Hi there) Tj ET
q 144 0 0 144 234 324 cm /Im1 Do Q
"""
contents = Stream(pdf, stream)
page_dict = {
'/Type': Name('/Page'),
'/MediaBox': mediabox,
'/Contents': contents,
'/Resources': resources
}
qpdf_page_dict = page_dict
page = pdf.make_indirect(qpdf_page_dict)
pdf.pages.append(page)
pdf.save(outdir / 'hi.pdf')
def test_list_apis():
a = pikepdf.Array([1, 2, 3])
a[1] = None
assert a[1] is None
assert len(a) == 3
del a[1]
assert len(a) == 2
a[-1] = Name('/Foo')
with pytest.raises(IndexError):
a[-5555] = Name.Foo
def test_bytes():
b = b'\x79\x78\x77\x76'
qs = String(b)
assert bytes(qs) == b
s = 'é'
qs = String(s)
assert str(qs) == s
assert Name('/xyz') == b'/xyz'
def test_repr_scalar(self):
scalars = [
False, 666,
Decimal('3.14'),
String('scalar'),
Name('/Bob'),
Operator('Q')
]
for s in scalars:
assert eval(repr(s)) == s
def test_list_apis():
a = pikepdf.Array([1, 2, 3])
a[1] = None
assert a[1] is None
assert len(a) == 3
del a[1]
assert len(a) == 2
a[-1] = Name('/Foo')
with pytest.raises(IndexError):
a[-5555] = Name.Foo
assert a == pikepdf.Array([1, Name.Foo])
a.append(4)
assert a == pikepdf.Array([1, Name.Foo, 4])
a.extend([42, 666])
assert a == pikepdf.Array([1, Name.Foo, 4, 42, 666])
def set_pagelabels(doc, page_labels):
arr = []
for label in page_labels:
pn = label['start'] - 1 # page index 1-based -> 0-based
d = {}
if 'style' in label and label['style'] != 'none':
d['/S'] = Name('/' + label['style'])
if 'prefix' in label:
d['/P'] = label['prefix']
if 'initial_count' in label:
d['/St'] = label['initial_count']
obj = Dictionary(d)
arr.append(pn)
arr.append(obj)
obj = Dictionary({'/Nums': Array(arr)})
doc.root[Name.PageLabels] = obj
def test_list_apis():
a = pikepdf.Array([1, 2, 3])
a[1] = None
assert a[1] is None
assert len(a) == 3
del a[1]
assert len(a) == 2
a[-1] = Name('/Foo')
with pytest.raises(IndexError):
a[-5555] = Name.Foo
assert a == pikepdf.Array([1, Name.Foo])
a.append(4)
assert a == pikepdf.Array([1, Name.Foo, 4])
a.extend([42, 666])
assert a == pikepdf.Array([1, Name.Foo, 4, 42, 666])
with pytest.raises(ValueError, match='object is not a dictionary'):
del a.ImaginaryKey
with pytest.raises(TypeError, match=r"items\(\) not available"):
a.items()
def _remove_simple_filters(obj, filters):
"""Remove simple lossless compression where it appears.
Args:
obj (pikepdf.Stream): the compressed object
filters (list of str): all files on the data
"""
COMPLEX_FILTERS = {
'/DCTDecode',
'/JPXDecode',
'/JBIG2Decode',
'/CCITTFaxDecode',
}
idx = [n for n, item in enumerate(filters) if item in COMPLEX_FILTERS]
if idx:
if len(idx) > 1:
raise NotImplementedError(
f"Object {obj.objgen} has compound complex filters: {filters}. "
"We cannot decompress this."
)
simple_filters = filters[: idx[0]]
complex_filters = filters[idx[0] :]
else:
simple_filters = filters
complex_filters = []
if not simple_filters:
return obj.read_raw_bytes(), complex_filters
original_filters = obj.Filter
try:
obj.Filter = Array([Name(s) for s in simple_filters])
data = obj.read_bytes(StreamDecodeLevel.specialized)
finally:
obj.Filter = original_filters
return data, complex_filters
def test_contains(self):
d = Dictionary({'/Monty': 'Python', '/Flying': 'Circus'})
assert Name.Flying in d
assert Name('/Monty') in d
assert Name.Brian not in d
def test_name(self):
self.check(Name.This, Name('/This'))
def test_unequal_but_similar(self):
assert Name('/Foo') != String('/Foo')
def test_empty_name():
with pytest.raises(ValueError):
Name('')
with pytest.raises(ValueError):
Name('/')
def test_unslashed_name():
with pytest.raises(ValueError, match='must begin with'):
Name('Monty') not in [] # pylint: disable=expression-not-assigned
def test_wrap_array():
assert Name('/Foo').wrap_in_array() == Array([Name('/Foo')])
assert Array([42]).wrap_in_array() == Array([42])
def get_pdf():
form = UserForm()
if form.validate_on_submit():
profile = {}
profile["firstname"] = form.firstname.data
profile["lastname"] = form.lastname.data
profile["birthday"] = form.birthday.data.strftime("%d/%m/%Y")
profile["address"] = form.address.data
profile["city"] = form.city.data
profile["zipcode"] = form.zipcode.data
profile["reason"] = form.reason.data
profile["delay"] = form.delay.data
now, date, hour = get_date(profile["delay"])
if profile["reason"] == "sport" and (19 <= now.hour or now.hour < 6):
profile["reason"] = "animaux"
elif profile["reason"] == "animaux" and (19 > now.hour or now.hour >= 6):
profile["reason"] = "sport"
if profile["reason"] == "achats_culte_culturel" and (19 <= now.hour or now.hour < 6):
flash("Des achats pendant le couvre-feu ? T'es magique toi !")
return redirect("/")
fp = webdriver.FirefoxProfile()
fp.set_preference("browser.download.folderList", 2)
fp.set_preference("browser.download.dir", str(PureWindowsPath(ATTEST_PATH)) if platform.system() == "Windows" else ATTEST_PATH)
fp.set_preference("browser.helperApps.neverAsk.saveToDisk", "application/pdf, attachment/pdf")
fp.set_preference("browser.download.manager.showWhenStarting",False)
fp.set_preference("browser.helperApps.neverAsk.openFile","application/pdf")
fp.set_preference("browser.helperApps.alwaysAsk.force", False)
fp.set_preference("browser.download.manager.useWindow", False)
fp.set_preference("browser.download.manager.focusWhenStarting", False)
fp.set_preference("browser.download.manager.alertOnEXEOpen", False)
fp.set_preference("browser.download.manager.showAlertOnComplete", False)
fp.set_preference("browser.download.manager.closeWhenDone", True)
fp.set_preference("pdfjs.disabled", True)
options = webdriver.firefox.options.Options()
options.add_argument("--headless")
try:
driver = webdriver.Firefox(options=options, firefox_profile=fp)
except Exception as e:
app.logger.error("Error: %s" % str(e))
flash("Woopsy, une erreur s'est produite...")
return redirect("/")
# get page
driver.get("https://media.interieur.gouv.fr/attestation-deplacement-derogatoire-covid-19/")
html = driver.page_source
soup = BeautifulSoup(html, features="lxml")
# pass the update button
wait = WebDriverWait(driver, DELAY)
try:
update_button = wait.until(EC.element_to_be_clickable((By.ID, 'reload-btn')))
update_button = driver.find_element_by_id("reload-btn").click()
except:
pass
files_nb = len(os.listdir(ATTEST_PATH))
# fill form
fill_form(driver, profile, now, date, hour)
# wait file to get there
i = 5
while i:
if len(os.listdir(ATTEST_PATH)) <= files_nb:
i -= 1
time.sleep(DELAY)
else:
break
if i == 0:
app.logger.error("Error: file not downloaded.")
flash("Woopsy, une erreur s'est produite...")
return redirect("/")
# rename file to avoid duplicates between users
try:
filepath = max([os.path.join(ATTEST_PATH, f) for f in os.listdir(ATTEST_PATH)], key=os.path.getctime)
filename = profile["lastname"] + "_" + profile["firstname"] + os.path.basename(filepath)
new_filename = os.path.join(ATTEST_PATH, filename)
shutil.move(filepath, new_filename)
except Exception as e:
app.logger.error("Error: %s" % str(e))
flash("Woopsy, une erreur s'est produite...")
return redirect("/")
# make QRcode and edit PDF
img = make_qr_code(profile, now, date, hour).convert('RGB').resize((590, 590))
pdf_file = Pdf.open(new_filename, allow_overwriting_input=True)
page = pdf_file.pages[1]
page_image = list(page.images.keys())
rawimage = page.images[page_image[0]]
rawimage.write(zlib.compress(img.tobytes()), filter=Name("/FlateDecode"))
rawimage.Width, rawimage.Height = 590, 590
page.Resources.XObject[page_image[0]] = rawimage
pdf_file.save(new_filename)
# program file cleaner
file_handle = open(new_filename, 'rb')
# This *replaces* the `remove_file` + @after_this_request code above
def stream_and_remove_file():
yield from file_handle
file_handle.close()
os.remove(new_filename)
return app.response_class(
stream_and_remove_file(),
mimetype='application/pdf',
headers={'Content-Disposition': 'attachment', 'filename': filename}
)
return redirect("/")
