def tokenize(data):
for text in data['items']:
if not reader.in_2008(text):
continue
txt = re.sub(r'<[^>]*>', '', text['textContent'])
txt = re.sub(r'-\n', '', txt)
found = re.findall(r'\b[a-zA-ZąęłćżźóńśŚĄĘŁĆŻŹÓŃ]+\b', txt)
found = [x.lower() for x in found]
words.extend(found)
def tag_judgment(data):
global count
for text in data['items']:
if not reader.in_2008(text):
continue
count += 1
response = requests.post(url, text['textContent'].encode('utf-8'))
with open('tagged.txt', 'a') as f:
f.writelines(response.text)
def find_references(data):
for item in data['items']:
if not reader.in_2008(item):
continue
for reference in item['referencedRegulations']:
found = re.findall(
r'(\bUstawa\b\s+\bz\b\s+\bdnia\b\s+\b23\b\s+\bkwietnia\b\s+\b1964\b\s+r\.\s+-\s+\bKodeks\b\s+\bcywilny\b)'
r'[\S\s]+(\bart.\s+\b445\b)', reference['text'])
f.extend(found)
def find_szkoda(data):
for text in data['items']:
if not reader.in_2008(text):
continue
found = re.findall(r'\b[Ss]zk[oó]d\w{0,3}\b', text['textContent'])
for word in found:
if proper(word):
words.append(word)
break
def find_values(data):
for text in data['items']:
if not reader.in_2008(text):
continue
# liczba taka jak '1203120' lub '123,1541.123,11321' lub '123 2100 3210 3210'
# jednostka 'mld' | 'mln' | 'tys' lub kwota slownie '(kwota slownie)' lub stare w postaci: '(starych|stare) lub 'starych|stare' wystepujace max 3 razy
# zlotych lub zlote na samym koncu
found = re.findall(
r'((\b\d+[\d,]*\b)|(\b\d+\s?[\d.]*,?\d{2}?\b)|(\b\d+[\d\s]+,?\d{2}?\b))\s?'
r'((\bmld\b\s?)|(\bmln\b\s?)|(\btys\b\s?)|(\b\(?star\w{1,3}\)?\b\s?)|(\([\w\s]+\)\s?)){0,3}'
r'((\bzłot\w{1,3}\b)|(\bzł\.?\b))', text['textContent'])
for matched in found:
value = parse_value(matched)
money.append(value)
def save_in_es(data):
global id
for text in data['items']:
if not reader.in_2008(text):
continue
judges = []
for judge in text['judges']:
judges.append(judge['name'])
judgment_doc = {
'text': text['textContent'],
'judgmentDate': parse(text['judgmentDate']),
'caseNumber': text['courtCases'][0]['caseNumber'],
'judges.name': judges
}
print(judges)
res = es.index(index='lab',
doc_type='judgment',
body=judgment_doc,
id=id)
print(res)
id += 1
def load_judgments(data):
for jsondoc in data['items']:
if not reader.in_2008(jsondoc):
continue
all_jugdments.append(jsondoc)