Improved extract method to work properly (sync)
This commit is contained in:
Kikimanox
@@ -1,114 +1,83 @@
|
||||
import codecs
|
||||
import os
|
||||
|
||||
import connexion
|
||||
import json
|
||||
from pathlib import Path
|
||||
from swagger_server.models.izlusci_body import IzlusciBody # noqa: E501
|
||||
from swagger_server.models.izlusci_async_body import IzlusciAsyncBody # noqa: E501
|
||||
from swagger_server.models.izlusci_sync_body import IzlusciSyncBody # noqa: E501
|
||||
from swagger_server.utils import cl_utils
|
||||
from swagger_server.util import get_random_filename, create_random_file_in_tmp_folder
|
||||
import requests
|
||||
from werkzeug.utils import secure_filename
|
||||
|
||||
# ATEapi_endpoint = "http://localhost:5000/predict"
|
||||
ATEapi_endpoint = "http://localhost:5000/predict"
|
||||
|
||||
|
||||
ATEapi_endpoint = "http://ate-api:5000/predict"
|
||||
# ATEapi_endpoint = "http://ate-api:5000/predict"
|
||||
|
||||
|
||||
def get_candidates(body): # noqa: E501
|
||||
"""Izlusci terminološke kandidate iz seznama besedil v conllu obliki
|
||||
def get_candidates_async(body): # noqa: E501
|
||||
"""Izlusci terminološke kandidate iz seznama besedil v conllu obliki [asinhrono, ustvari novi job]
|
||||
|
||||
# noqa: E501
|
||||
|
||||
:param body:
|
||||
:param body:
|
||||
:type body: dict | bytes
|
||||
|
||||
:rtype: str
|
||||
"""
|
||||
if connexion.request.is_json:
|
||||
body = IzlusciAsyncBody.from_dict(connexion.request.get_json()) # noqa: E501
|
||||
return 'do some magic!'
|
||||
|
||||
|
||||
def do_izlusci(conllus, prepovedane_besede):
|
||||
tmp_file_path = ""
|
||||
try:
|
||||
big_conllu = cl_utils.multipla_conllus_to_one_from_conllus_arr(conllus)
|
||||
tmp_file_path = create_random_file_in_tmp_folder(big_conllu, ".conllu")
|
||||
fp = open(tmp_file_path, 'rb')
|
||||
try:
|
||||
files = [
|
||||
('file', ('temp_1.conllu', fp, 'application/octet-stream'))
|
||||
]
|
||||
res = requests.post(ATEapi_endpoint, files=files)
|
||||
data = json.loads(res.text)
|
||||
finally:
|
||||
fp.close()
|
||||
os.remove(tmp_file_path)
|
||||
|
||||
ret = {'terminoloski_kandidati': [
|
||||
{
|
||||
'POSoznake': tk['msd'],
|
||||
'kandidat': tk['terms'], # more to bit lemma al terms?
|
||||
'kanonicnaoblika': tk['canonical'],
|
||||
'ranking': tk['ranking'],
|
||||
'podporneutezi': [
|
||||
0.0, # ????????
|
||||
0.0 # ??????
|
||||
],
|
||||
'pogostostpojavljanja': [0, 0] # ???????
|
||||
}
|
||||
for tk in data if tk['terms'] not in prepovedane_besede
|
||||
]}
|
||||
return ret, 200
|
||||
except Exception as e:
|
||||
return str(e), 500
|
||||
|
||||
|
||||
def get_candidates_sync(body): # noqa: E501
|
||||
"""Izlusci terminološke kandidate iz seznama besedil v conllu obliki [sihrono, rezultat v sami zahtevi]
|
||||
|
||||
# noqa: E501
|
||||
|
||||
:param body:
|
||||
:type body: dict | bytes
|
||||
|
||||
:rtype: List[TerminoloskiKandidat]
|
||||
"""
|
||||
if connexion.request.is_json:
|
||||
body = IzlusciBody.from_dict(connexion.request.get_json()) # noqa: E501
|
||||
body = IzlusciSyncBody.from_dict(connexion.request.get_json()) # noqa: E501
|
||||
|
||||
# Todo: What is "conllus" in body input anyway??
|
||||
# Todo: Kaj je s prepovedanimi besedami?
|
||||
# file_ids = [10000, 10001, 10002, 10003]
|
||||
# big_conllu = cl_utils.multipla_conllus_to_one_from_file_ids(file_ids)
|
||||
|
||||
# txt = Path('../ATEapi/temp_1.conllu').read_text('utf-8') # LOCAL ONLY
|
||||
|
||||
# # Todo: does this close the file after the request?
|
||||
# files = [
|
||||
# ('file', ('temp_1.conllu', open('../ATEapi/temp_1.conllu', 'rb'),
|
||||
# 'application/octet-stream'))
|
||||
# ]
|
||||
# res = requests.post(ATEapi_endpoint, files=files)
|
||||
# # res.status_code
|
||||
# # res.text
|
||||
#
|
||||
# return res.text, res.status_code
|
||||
#
|
||||
# example_data = ""
|
||||
# with open("C:\\Users\\Kiki\\Desktop\\Untitled-2.json", "r", encoding='utf-8') as f:
|
||||
# example_data = json.loads(f.read())
|
||||
#
|
||||
# d = 0
|
||||
#
|
||||
# ret = {'terminoloski_kandidati': [
|
||||
# {
|
||||
# 'POSoznake': tk['msd'],
|
||||
# 'kandidat': tk['terms'], # more to bit lemma al terms?
|
||||
# 'kanonicnaoblika': tk['canonical'],
|
||||
# 'ranking': tk['ranking'],
|
||||
# 'podporneutezi': [
|
||||
# 6.0274563, # ????????
|
||||
# 6.0274563 # ??????
|
||||
# ],
|
||||
# 'pogostostpojavljanja': [101, 71] # ???????
|
||||
# }
|
||||
# for tk in example_data
|
||||
# ]}
|
||||
|
||||
data = {
|
||||
"terminoloski_kandidati": [
|
||||
{
|
||||
"POSoznake": "Ncmsn",
|
||||
"kandidat": "vpliv",
|
||||
"kanonicnaoblika": "vpliv",
|
||||
"nosilnautez": 0.8008282,
|
||||
"podporneutezi": [
|
||||
6.0274563,
|
||||
6.0274563
|
||||
],
|
||||
"pogostostpojavljanja": [101, 71]
|
||||
},
|
||||
{
|
||||
"POSoznake": "Agpnsg Ncnsg",
|
||||
"kandidat": "bivalen okolje",
|
||||
"kanonicnaoblika": "bivalno okolje",
|
||||
"nosilnautez": 0.60254,
|
||||
"podporneutezi": [
|
||||
3.263,
|
||||
2.134
|
||||
],
|
||||
"pogostostpojavljanja": [27, 11]
|
||||
},
|
||||
{
|
||||
"POSoznake": "Agpmsny Ncmsn Ncmpg",
|
||||
"kandidat": "motorični status otrok",
|
||||
"kanonicnaoblika": "motorični status otrok",
|
||||
"nosilnautez": 0.3324,
|
||||
"podporneutezi": [
|
||||
1.221,
|
||||
3.323
|
||||
],
|
||||
"pogostostpojavljanja": [31, 51]
|
||||
},
|
||||
{
|
||||
"POSoznake": "Agpnsn Ncnsn",
|
||||
"kandidat": "diplomsko delo",
|
||||
"kanonicnaoblika": "diplomsko delo",
|
||||
"nosilnautez": 0.2008282,
|
||||
"podporneutezi": [
|
||||
0.883,
|
||||
1.02
|
||||
],
|
||||
"pogostostpojavljanja": [1241, 111]
|
||||
}
|
||||
]
|
||||
}
|
||||
return data
|
||||
return do_izlusci(body.conllus, body.prepovedane_besede)
|
||||
|
||||
Reference in New Issue
Block a user