def get_target_phrases_item_data(target_phrases):
# make an itemData of our custom target phrases
target_phrases_item_data = itemData.itemData()
for target_phrase in target_phrases:
# create a contextItem from the target phrase
contextItemTarget = itemData.contextItem(
[target_phrase, target_phrase, target_phrase, target_phrase])
target_phrases_item_data.append(contextItemTarget)
return target_phrases_item_data
def get_items(_file):
def get_fileobj(_file):
if not urllib.parse.urlparse(_file).scheme:
_file = "file://"+_file
return urllib.request.urlopen(_file, data=None)
f0 = get_fileobj(_file)
context_items = [contextItem((d["Lex"],
d["Type"],
r"%s"%d["Regex"],
d["Direction"])) for d in yaml.load_all(f0)]
return context_items
def get_target_items(target_list):
"""
takes concepts response in json format
"""
target_items = []
if not target_list:
return target_items
target_items = [
itemData.contextItem(
(d["lex"], d["type"], r"%s" % d["regex"], d["direction"]))
for d in target_list
]
return target_items
def convertCSVtoitemData(csvFile,
encoding='utf-8',
delimiter="\t",
headerRows=1,
literalColumn=0,
categoryColumn=1,
regexColumn=2,
ruleColumn=3):
"""
takes a CSV file of itemdata rules and creates a single itemData instance.
csvFile: name of file to read items from
encoding: unicode enocidng to use; default = 'utf-8'
headerRows: number of header rows in file; default = 1
literalColumn: column from which to read the literal; default = 0
categoryColumn: column from which to read the category; default = 1
regexColumn: column from which to read the regular expression: default = 2
ruleColumn: column from which to read the rule; default = 3
"""
items = itemData.itemData() # itemData to be returned to the user
header = []
#reader, f0 = get_fileobj(csvFile)
#print csvFile
f = open(csvFile, 'rU')
reader = csv.reader(f, delimiter=delimiter)
# first grab numbe rof specified header rows
for i in range(headerRows):
row = next(reader)
header.append(row)
# now grab each itemData
for row in reader:
#print row
tmp = [
row[literalColumn], row[categoryColumn], row[regexColumn],
row[ruleColumn]
]
tmp[2] = r'''{0}'''.format(
tmp[2]) # convert the regular expression string into a raw string
item = itemData.contextItem(tmp)
items.append(item)
f.close()
return items
def test_instantiate_contextItem0(items):
for item in items:
assert itemData.contextItem(item)
def test_contextItem_rule(items):
cti = itemData.contextItem(items[1])
assert cti.getRule() == "forward"
def test_contextItem_category(items):
cti = itemData.contextItem(items[1])
assert cti.getCategory() == ["probable_negated_existence"]
iris.append(row[1])
sentences.append(row[5])
modifiers = itemData.get_items(
"https://raw.githubusercontent.com/chapmanbe/pyConTextNLP/20c752d6bd5191833f21ab81fc7f41877dca1db6/KB/pneumonia_modifiers.yml"
)
for i in range(len(sentences)):
# omg
targets = []
with open('./hpo_labels.txt') as labelfile:
reader = csv.reader(labelfile, delimiter='\t')
for row in reader:
if (row[1] == iris[i]):
targets.append(contextItem((row[0], row[1], '', '')))
s = sentences[i]
markup = pyConText.ConTextMarkup()
markup.setRawText(s.lower())
markup.cleanText()
markup.markItems(modifiers, mode="modifier")
markup.markItems(targets, mode="target")
markup.applyModifiers()
markup.pruneMarks()
markup.pruneSelfModifyingRelationships()
markup.dropInactiveModifiers()
def test_contextItem_isa2(items):
cti = itemData.contextItem(items[1])
assert cti.isA("PROBABLE_NEGATED_EXISTENCE")
def test_contextItem_getRE1(items):
cti = itemData.contextItem(items[0])
assert cti.getRE() == r"""pulmonary\s(artery )?(embol[a-z]+)"""
def test_instantiate_contextItem(self):
cit1 = itemData.contextItem(self.items[0])
assert cit1
def test_contextItem_isa(items):
cti = itemData.contextItem(items[0])
assert cti.isA("pulmonary_embolism")
def test_contextItem_getRE(items):
cti = itemData.contextItem(items[1])
assert cti.getRE() == r'\b%s\b'%items[1][0]
def test_contextItem_getRE1(items):
cti = itemData.contextItem(items[0])
assert cti.getRE() == r"""pulmonary\s(artery )?(embol[a-z]+)"""
def test_contextItem_isa2(items):
cti = itemData.contextItem(items[1])
assert cti.isA("PROBABLE_NEGATED_EXISTENCE")
def test_contextItem_isa1(items):
cti = itemData.contextItem(items[0])
assert cti.isA("PULMONARY_EMBOLISM")
def test_contextItem_isa(items):
cti = itemData.contextItem(items[0])
assert cti.isA("pulmonary_embolism")
def test_contextItem_literal(items):
cti = itemData.contextItem(items[0])
assert cti.getLiteral() == "pulmonary embolism"
def test_instantiate_itemData(self):
cit1 = itemData.contextItem(self.items[0])
it1 = itemData.itemData()
it1.append(cit1)
assert it1
def test_contextItem_category(items):
cti = itemData.contextItem(items[1])
assert cti.getCategory() == ["probable_negated_existence"]
def test_instantiate_itemData(self):
cit1 = itemData.contextItem(self.items[0])
it1 = itemData.itemData()
it1.append(cit1)
assert it1
def test_contextItem_isa1(items):
cti = itemData.contextItem(items[0])
assert cti.isA("PULMONARY_EMBOLISM")
def test_contextItem_rule(items):
cti = itemData.contextItem(items[1])
assert cti.getRule() == "forward"
def test_contextItem_getRE(items):
cti = itemData.contextItem(items[1])
assert cti.getRE() == r'\b%s\b' % items[1][0]
def test_instantiate_contextItem0(items):
for item in items:
assert itemData.contextItem(item)
def test_instantiate_contextItem(self):
cit1 = itemData.contextItem(self.items[0])
assert cit1
def test_contextItem_literal(items):
cti = itemData.contextItem(items[0])
assert cti.getLiteral() == "pulmonary embolism"