def test_conll_readwrite_sentence_extra_whitespace():
sentence = Sentence.from_conll(conll_example2.splitlines() + ['', '', ''])
assert sentence.as_conll() == conll_example2.strip()
sentence = Sentence.from_conll(['', ''] + conll_example2.splitlines() +
['', ''])
assert sentence.as_conll() == conll_example2.strip()
sentence = Sentence.from_conll([''] + conll_example2.splitlines())
assert sentence.as_conll() == conll_example2.strip()
def test_conll_readwrite_sentence_extra_whitespace():
sentence = Sentence.from_conll(conll_example2.splitlines() + ['', '', ''])
assert sentence.as_conll() == conll_example2.strip()
sentence = Sentence.from_conll(['', ''] + conll_example2.splitlines() +
['', ''])
assert sentence.as_conll() == conll_example2.strip()
sentence = Sentence.from_conll([''] + conll_example2.splitlines())
assert sentence.as_conll() == conll_example2.strip()
def test_conll_as_dotgraph_custom_edgeformat():
if older_than_py27: # this feature disabled in older Pythons
return
formatted_dotgraph = '''
digraph {
0 [label=root]
1 [label=Cathy]
2 -> 1 [label=su color=blue]
2 [label=zag]
0 -> 2 [label=ROOT color=red]
3 [label=hen]
2 -> 3 [label=obj1 color=blue]
4 [label=wild]
5 -> 4 [label=mod color=blue]
5 [label=zwaaien]
2 -> 5 [label=vc color=blue]
6 [label="."]
5 -> 6 [label=punct color=blue]
}
'''.strip()
sentence = Sentence.from_conll(conll_example.splitlines())
def edge_formatter(token):
if token.head == 0:
return {'color': 'red'}
else:
return {'color': 'blue'}
dotgraph = sentence.as_dotgraph(edge_formatter=edge_formatter)
assert dotgraph.source == formatted_dotgraph
def test_conll_as_dotgraph_custom_nodeformat():
if older_than_py27: # this feature disabled in older Pythons
return
formatted_dotgraph = '''
digraph {
0 [label=root color=red]
1 [label=Cathy color=green]
2 -> 1 [label=su]
2 [label=zag color=blue]
0 -> 2 [label=ROOT]
3 [label=hen]
2 -> 3 [label=obj1]
4 [label=wild]
5 -> 4 [label=mod]
5 [label=zwaaien color=green]
2 -> 5 [label=vc]
6 [label="."]
5 -> 6 [label=punct]
}
'''.strip()
def node_formatter(token):
if token is None:
return {'color': 'red'}
elif token.cpos == 'N':
return {'color': 'green'}
elif token.cpos == 'V':
return {'color': 'blue'}
else:
return {}
sentence = Sentence.from_conll(conll_example.splitlines())
dotgraph = sentence.as_dotgraph(node_formatter=node_formatter)
assert dotgraph.source == formatted_dotgraph
def test_conll_as_dotgraph_custom_edgeformat():
if older_than_py27: # this feature disabled in older Pythons
return
formatted_dotgraph = """
digraph {
0 [label=root]
1 [label=Cathy]
2 -> 1 [label=su color=blue]
2 [label=zag]
0 -> 2 [label=ROOT color=red]
3 [label=hen]
2 -> 3 [label=obj1 color=blue]
4 [label=wild]
5 -> 4 [label=mod color=blue]
5 [label=zwaaien]
2 -> 5 [label=vc color=blue]
6 [label="."]
5 -> 6 [label=punct color=blue]
}
""".strip()
sentence = Sentence.from_conll(conll_example.splitlines())
def edge_formatter(token):
if token.head == 0:
return {"color": "red"}
else:
return {"color": "blue"}
dotgraph = sentence.as_dotgraph(edge_formatter=edge_formatter)
assert dotgraph.source == formatted_dotgraph
def test_conll_as_dotgraph_custom_nodeformat():
if older_than_py27: # this feature disabled in older Pythons
return
formatted_dotgraph = """
digraph {
0 [label=root color=red]
1 [label=Cathy color=green]
2 -> 1 [label=su]
2 [label=zag color=blue]
0 -> 2 [label=ROOT]
3 [label=hen]
2 -> 3 [label=obj1]
4 [label=wild]
5 -> 4 [label=mod]
5 [label=zwaaien color=green]
2 -> 5 [label=vc]
6 [label="."]
5 -> 6 [label=punct]
}
""".strip()
def node_formatter(token):
if token is None:
return {"color": "red"}
elif token.cpos == "N":
return {"color": "green"}
elif token.cpos == "V":
return {"color": "blue"}
else:
return {}
sentence = Sentence.from_conll(conll_example.splitlines())
dotgraph = sentence.as_dotgraph(node_formatter=node_formatter)
assert dotgraph.source == formatted_dotgraph
def test_conll_as_asciitree():
asciitree_out = '''
zag [ROOT]
+-- Cathy [su]
+-- hen [obj1]
+-- zwaaien [vc]
+-- wild [mod]
+-- . [punct]'''
sentence = Sentence.from_conll(conll_example.splitlines())
assert sentence.as_asciitree().strip() == asciitree_out.strip()
def test_conll_as_asciitree():
asciitree_out = """
zag [ROOT]
+-- Cathy [su]
+-- hen [obj1]
+-- zwaaien [vc]
+-- wild [mod]
+-- . [punct]"""
sentence = Sentence.from_conll(conll_example.splitlines())
assert sentence.as_asciitree().strip() == asciitree_out.strip()
def test_conll_as_asciitree_custom():
asciitree_out = """
ROOT:zag:V
+--su:Cathy:N
+--obj1:hen:Pron
+--vc:zwaaien:N
+--mod:wild:Adj
+--punct:.:Punc"""
def str_func(token):
return "%s:%s:%s" % (token.deprel, token.form, token.cpos)
sentence = Sentence.from_conll(conll_example.splitlines())
assert sentence.as_asciitree(str_func=str_func).strip() == asciitree_out.strip()
def test_conll_as_asciitree_custom():
asciitree_out = '''
ROOT:zag:V
+--su:Cathy:N
+--obj1:hen:Pron
+--vc:zwaaien:N
+--mod:wild:Adj
+--punct:.:Punc'''
def str_func(token):
return '%s:%s:%s' % (token.deprel, token.form, token.cpos)
sentence = Sentence.from_conll(conll_example.splitlines())
assert sentence.as_asciitree(str_func=str_func).strip() == \
asciitree_out.strip()
def test_conll_as_dotgraph_custom_digraph_and_idprefix():
if older_than_py27: # this feature disabled in older Pythons
return
formatted_dotgraph = """
digraph test {
x0 [label=root]
x1 [label=Cathy]
x2 -> x1 [label=su]
x2 [label=zag]
x0 -> x2 [label=ROOT]
x3 [label=hen]
x2 -> x3 [label=obj1]
x4 [label=wild]
x5 -> x4 [label=mod]
x5 [label=zwaaien]
x2 -> x5 [label=vc]
x6 [label="."]
x5 -> x6 [label=punct]
}""".strip()
sentence = Sentence.from_conll(conll_example.splitlines())
dotgraph = sentence.as_dotgraph(id_prefix="x", digraph_kwargs={"name": "test"})
def test_conll_as_dotgraph_custom_digraph_and_idprefix():
if older_than_py27: # this feature disabled in older Pythons
return
formatted_dotgraph = '''
digraph test {
x0 [label=root]
x1 [label=Cathy]
x2 -> x1 [label=su]
x2 [label=zag]
x0 -> x2 [label=ROOT]
x3 [label=hen]
x2 -> x3 [label=obj1]
x4 [label=wild]
x5 -> x4 [label=mod]
x5 [label=zwaaien]
x2 -> x5 [label=vc]
x6 [label="."]
x5 -> x6 [label=punct]
}'''.strip()
sentence = Sentence.from_conll(conll_example.splitlines())
dotgraph = sentence.as_dotgraph(id_prefix='x',
digraph_kwargs={'name': 'test'})
def test_conll_readwrite_sentence2():
sentence = Sentence.from_conll(conll_example2.splitlines())
assert sentence.as_conll() == conll_example2.strip()
def test_conll_readwrite_sentence():
sentence = Sentence.from_conll(conll_example.splitlines())
assert sentence.as_conll() == conll_example.strip()
