def test_addboxes(self):
box1 = Textbox([Textelement("hey ", tag=None)],
fontid=None,
top=0,
left=0,
width=50,
height=10,
lines=1)
box2 = Textbox([Textelement("ho", tag=None)],
fontid=None,
top=0,
left=50,
width=40,
height=10,
lines=1)
combinedbox = box1 + box2
want = """
<Textbox bottom="10" fontid="0" height="10" left="0" lineheight="0" lines="1" right="90" top="0" width="90">
<Textelement>hey ho</Textelement>
</Textbox>
"""
self.assertEqual(want[1:], serialize(combinedbox))
# make sure __iadd__ performs like __add__
box1 += box2
self.assertEqual(want[1:], serialize(box1))
def test_leading_tag(self):
body = Textbox([
Textelement("bold", tag="b"),
Textelement("normal", tag=None),
],
top=0,
left=0,
width=100,
height=100,
fontid=0)
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px"><b>bold</b>normal</p>
"""
self._test_asxhtml(want, body)
def test_superscripts(self):
body = Textbox([
Textelement("1", tag="sup"),
Textelement("2", tag="is"),
Textelement("3", tag="bis")
],
top=0,
left=0,
width=100,
height=100,
fontid=0)
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px"><sup>1</sup><i><sup>2</sup></i><b><i><sup>3</sup></i></b></p>
"""
self._test_asxhtml(want, body)
def test_add_different_types(self):
box1 = Textbox([Textelement("hey", tag=None)],
fontid=None,
top=0,
left=0,
width=50,
height=10,
lines=1)
box2 = Textbox([LinkedTextelement("1", tag="s", uri="foo.html")],
fontid=None,
top=0,
left=50,
width=5,
height=10,
lines=1)
combinedbox = box1 + box2
want = """
<Textbox bottom="10" fontid="0" height="10" left="0" lineheight="0" lines="1" right="55" top="0" width="55">
<Textelement>hey</Textelement>
<LinkedTextelement tag="s" uri="foo.html">1</LinkedTextelement>
</Textbox>
"""
self.assertEqual(want[1:], serialize(combinedbox))
# make sure __iadd__ performs like __add__
box1 += box2
self.assertEqual(want[1:], serialize(box1))
def test_tag_merge(self):
body = Textbox([
Textelement("identical ", tag=None),
Textelement("tags ", tag=None),
Textelement("should ", tag="b"),
Textelement("merge", tag="b"),
],
top=0,
left=0,
width=100,
height=100,
fontid=0)
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px">identical tags <b>should merge</b></p>
"""
self._test_asxhtml(want, body)
def test_elements_with_tags(self):
body = Textbox([
Textelement("normal", tag=None),
Textelement("bold", tag="b"),
Textelement("italic", tag="i"),
Textelement("both", tag="bi")
],
top=0,
left=0,
width=100,
height=100,
fontid=0)
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px">normal<b>bold</b><i>italic</i><b><i>both</i></b></p>
"""
self._test_asxhtml(want, body)
def test_linkelements(self):
body = Textbox([
Textelement("normal", tag=None),
LinkedTextelement("link", uri="http://example.org/", tag=None),
Textelement("footnote marker", tag="sup"),
LinkedTextelement(
"linked footnote marker", uri="http://example.org/", tag="s")
],
top=0,
left=0,
width=100,
height=100,
fontid=0)
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px">normal<a href="http://example.org/">link</a><sup>footnote marker</sup><a href="http://example.org/"><sup>linked footnote marker</sup></a></p>
"""
self._test_asxhtml(want, body)
def test_basic(self):
body = Textbox([Textelement("test", tag=None)],
top=0,
left=0,
width=100,
height=100,
fontid=0)
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px">test</p>
"""
self._test_asxhtml(want, body)
def test_empty_removal(self):
body = Textbox([
LinkedTextelement(" ", uri="index.html#24", tag=None),
Textelement("23", tag=None)
],
top=0,
left=0,
width=100,
height=100,
fontid=0)
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px">23</p>
"""
self._test_asxhtml(want, body)
def __call__(self, textbox, fontspecs):
if 'encoding' not in fontspecs[
textbox.fontid]: # only for some testcases
return textbox
if fontspecs[textbox.fontid]['encoding'] != "Custom":
return textbox
# NOTE: This weird checking for occurrences of 'i'
# tags is needed for functionalSources.
# TestPropRegeringen.test_parse_1999_2000_17 to pass
# (and matches encoding usage in practice)
decode_all = not ('i' in [getattr(x, 'tag', None) for x in textbox])
for idx, subpart in enumerate(textbox):
if (isinstance(subpart, Textelement)
and (decode_all or subpart.tag == 'i')):
textbox[idx] = Textelement(self.decode_string(
subpart, self.map),
tag=subpart.tag)
return textbox
def test_other_elements(self):
body = Textbox([
Textelement("plaintext ", tag=None),
LinkSubject("link",
uri="http://example.org/",
predicate="dcterms:references"), " raw string"
],
top=0,
left=0,
width=100,
height=100,
fontid=0)
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px">plaintext <a href="http://example.org/" rel="dcterms:references">link</a> raw string</p>
"""
self._test_asxhtml(want, body)
# remove the last str so that the linksubject becomes the last item
body[:] = body[:-1]
want = """
<p xmlns="http://www.w3.org/1999/xhtml" class="textbox fontspec0" style="top: 0px; left: 0px; height: 100px; width: 100px">plaintext <a href="http://example.org/" rel="dcterms:references">link</a></p>
"""
self._test_asxhtml(want, body)
def __call__(self, textbox, fontspecs):
if fontspecs[textbox.fontid]['encoding'] != "Custom":
return textbox
if textbox.font.family == "Times.New.Roman.Fet0100":
boundary = None
# extra special hack for prop 1997/98:44 which has
# textelements marked as having a font with custom
# encoding, but where only the bolded part (which
# isn't marked up...) is encoded, while the rest is
# unencoded. The "g" is a encoded section sign, which
# in these cases is the last encoded char.
if (len(textbox[0].split(" ", 2)) == 3
and textbox[0].split(" ", 2)[1] == "g"):
boundary = textbox[0].index(" ", textbox[0].index(" ") + 1)
# a similar situation with paragraphs with leading bold
# type, where the bold text is any of 3-4 fixed strings
# (Note: the xml data doesn't contain any information
# about the text being bold, or rather that the following
# text is non-bold)
else:
m = self.re_fixedleaders.match(textbox[0])
if m:
boundary = m.end()
if boundary:
orig = str(textbox[0])
textbox[0] = Textelement(self.decode_string(
orig[:boundary], self.map),
tag="b")
textbox.insert(1, Textelement(orig[boundary:], tag=None))
# Find the id for the "real" non-bold font. I think
# that in every known case the fontid should simply be
# the default font (id=0). Maybe we could hardcode
# that right away, like we hardcode the font family
# name right now.
newfontid = self.find_fontid(fontspecs, "Times-Roman",
textbox.font.size)
expected_length = 2
else:
textbox[0] = Textelement(self.decode_string(
textbox[0], self.map),
tag=textbox[0].tag)
expected_length = 1
newfontid = textbox.fontid
if len(
textbox
) > expected_length: # the <text> element contained subelements
# save and remove the 1-2 textelements we've processed
decoded = textbox[:expected_length]
textbox[:] = textbox[expected_length:]
# do the default decoding
textbox = super(OffsetDecoder20,
self).__call__(textbox, fontspecs)
# then add the previously procesed elements
textbox[:] = decoded + textbox[:]
if newfontid != textbox.fontid:
# invalidate the cached property
del textbox.__dict__['font']
textbox.fontid = newfontid
else:
textbox = super(OffsetDecoder20, self).__call__(textbox, fontspecs)
# again, if one or more textelements have an "i" tag, the
# font for the entire textbox probably shouldn't be
# specced as an italic ("Kursiv")
if textbox.font.family == "Times.New.Roman.Kursiv0104" and "i" in [
x.tag for x in textbox
]:
newfontid = self.find_fontid(fontspecs, "Times-Roman",
textbox.font.size)
# invalidate the cached property
del textbox.__dict__['font']
textbox.fontid = newfontid
return textbox