def save(date, name_newspaper, words):
for a_word in words:
obj_newspaper = Newspaper.objects.get(name=name_newspaper)
obj = Word(
newspaper=obj_newspaper,
_datetime=date,
word=a_word,
)
obj.save()
def wordPop(LANGUAGE_TO_POPULATE):
''' Populate the -Words- and -Lemmas- '''
fileName = ''.join(['data/langs/', LANGUAGE_TO_POPULATE, '.txt'])
languageObject = Language.objects.get(name=LANGUAGE_TO_POPULATE)
with open(fileName, "r", encoding="utf8") as wordData:
for line in wordData:
# Lemma Word Tagset - delimiter ('/t')
rowContent = line.split('\t')
# print(rowContent)
if (len(rowContent) >= 3): # checks if line is valid
tagsetName = rowContent[-1].rstrip()
tagSetObject, created = TagSet.objects.get_or_create(
name=tagsetName)
lemmaName = rowContent[0]
wordName = rowContent[1]
allLabels = tagsetName.split(
";") # last block of words corrensponds to allLabels
for currLabel in allLabels:
try:
currFeature = findFeature[currLabel.upper()]
featObject = Feature.objects.get(name=currFeature)
tagSetObject.features.add(featObject)
except KeyError:
print(f'{currLabel} - not exist')
posName = findFeature[allLabels[0].upper()].rstrip()
posObject = POS.objects.get(name=posName)
# If lemma exists
try:
lemmaObject = Lemma.objects.get(name=lemmaName,
pos=posObject.id,
language=languageObject.id)
# If not create a new one
except Lemma.DoesNotExist:
lemmaObject = Lemma(name=lemmaName)
lemmaObject.language = languageObject
lemmaObject.pos = posObject
lemmaObject.save()
# Finally create the word
finally:
wordObject = Word(name=wordName)
wordObject.lemma = lemmaObject
wordObject.tagset = tagSetObject
wordObject.language = languageObject
wordObject.save()
def undo_lowercase_words(_, __):
Word.objects.all().delete()
tweets = Tweet.objects.all()
all_tokens = []
for tweet in tweets:
tokens_from_tweet = word_tokenize(tweet.content)
for token in tokens_from_tweet:
if token not in all_tokens:
all_tokens.append(token)
for token in all_tokens:
word = Word(word=token, frequency=-1)
word.save()
def create_snippet(word, rawUrl, url, starttime, endtime):
try:
word_object = Word.objects.get(value=word)
except models.Word.DoesNotExist:
word_object = Word(value=word)
word_object.save()
audio_object = Audio.objects.get(url=rawUrl)
new_Snippet = Snippet(word=word_object,
audio=audio_object,
start=starttime,
end=endtime,
url=url)
new_Snippet.save()
print "Saved \"%s\" Snippet to database" % word
def populate_words(_, __):
tweets = Tweet.objects.all()
all_tokens = []
for tweet in tweets:
tokens_from_tweet = word_tokenize(tweet.content)
for token in tokens_from_tweet:
if token not in all_tokens:
all_tokens.append(token)
for token in all_tokens:
word = Word(
word=token,
)
word.save()
def undo_stem_words(_, __):
Word.objects.all().delete()
tweets = Tweet.objects.all()
all_tokens = []
for tweet in tweets:
tokens_from_tweet = word_tokenize(str(tweet.content).lower())
for token in tokens_from_tweet:
if token not in all_tokens:
all_tokens.append(token)
for token in all_tokens:
word = Word(
word=token,
frequency=-1,
)
if Word.objects.filter(word=token).count() is 0:
word.save()
def mutate_and_get_payload(cls, root, info):
word = Word()
word.save()
return IntroduceWord(word=word)