File size: 17,014 Bytes
f4456af 2332381 f4456af c5ea37c f4456af 3079197 f4456af c5ea37c f4456af 3079197 2d09c38 3079197 f4456af 3079197 f4456af 3079197 028acc4 3079197 f4456af 3079197 1eb186a f4456af 028fe40 f4456af c5ea37c f4456af c5ea37c f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af d0db329 f4456af 3079197 f4456af 3079197 f4456af a8294f2 c5ea37c 9fe9fc4 c5ea37c c372afe f4456af 3079197 f4456af 1eb186a f4456af 1eb186a d0db329 1eb186a 3079197 1eb186a f4456af c5ea37c f4456af 3079197 3198faf f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 f4456af 3079197 2d09c38 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 |
import re
import json
import time
import copy
import elasticsearch
from elastic_transport import ConnectionTimeout
from elasticsearch import Elasticsearch
from elasticsearch_dsl import UpdateByQuery, Search, Index
from rag.settings import es_logger
from rag import settings
from rag.utils import singleton
es_logger.info("Elasticsearch version: "+str(elasticsearch.__version__))
@singleton
class ESConnection:
def __init__(self):
self.info = {}
self.conn()
self.idxnm = settings.ES.get("index_name", "")
if not self.es.ping():
raise Exception("Can't connect to ES cluster")
def conn(self):
for _ in range(10):
try:
self.es = Elasticsearch(
settings.ES["hosts"].split(","),
basic_auth=(settings.ES["username"], settings.ES["password"]) if "username" in settings.ES and "password" in settings.ES else None,
verify_certs=False,
timeout=600
)
if self.es:
self.info = self.es.info()
es_logger.info("Connect to es.")
break
except Exception as e:
es_logger.error("Fail to connect to es: " + str(e))
time.sleep(1)
def version(self):
v = self.info.get("version", {"number": "5.6"})
v = v["number"].split(".")[0]
return int(v) >= 7
def health(self):
return dict(self.es.cluster.health())
def upsert(self, df, idxnm=""):
res = []
for d in df:
id = d["id"]
del d["id"]
d = {"doc": d, "doc_as_upsert": "true"}
T = False
for _ in range(10):
try:
if not self.version():
r = self.es.update(
index=(
self.idxnm if not idxnm else idxnm),
body=d,
id=id,
doc_type="doc",
refresh=True,
retry_on_conflict=100)
else:
r = self.es.update(
index=(
self.idxnm if not idxnm else idxnm),
body=d,
id=id,
refresh=True,
retry_on_conflict=100)
es_logger.info("Successfully upsert: %s" % id)
T = True
break
except Exception as e:
es_logger.warning("Fail to index: " +
json.dumps(d, ensure_ascii=False) + str(e))
if re.search(r"(Timeout|time out)", str(e), re.IGNORECASE):
time.sleep(3)
continue
self.conn()
T = False
if not T:
res.append(d)
es_logger.error(
"Fail to index: " +
re.sub(
"[\r\n]",
"",
json.dumps(
d,
ensure_ascii=False)))
d["id"] = id
d["_index"] = self.idxnm
if not res:
return True
return False
def bulk(self, df, idx_nm=None):
ids, acts = {}, []
for d in df:
id = d["id"] if "id" in d else d["_id"]
ids[id] = copy.deepcopy(d)
ids[id]["_index"] = self.idxnm if not idx_nm else idx_nm
if "id" in d:
del d["id"]
if "_id" in d:
del d["_id"]
acts.append(
{"update": {"_id": id, "_index": ids[id]["_index"]}, "retry_on_conflict": 100})
acts.append({"doc": d, "doc_as_upsert": "true"})
res = []
for _ in range(100):
try:
if elasticsearch.__version__[0] < 8:
r = self.es.bulk(
index=(
self.idxnm if not idx_nm else idx_nm),
body=acts,
refresh=False,
timeout="600s")
else:
r = self.es.bulk(index=(self.idxnm if not idx_nm else
idx_nm), operations=acts,
refresh=False, timeout="600s")
if re.search(r"False", str(r["errors"]), re.IGNORECASE):
return res
for it in r["items"]:
if "error" in it["update"]:
res.append(str(it["update"]["_id"]) +
":" + str(it["update"]["error"]))
return res
except Exception as e:
es_logger.warn("Fail to bulk: " + str(e))
if re.search(r"(Timeout|time out)", str(e), re.IGNORECASE):
time.sleep(3)
continue
self.conn()
return res
def bulk4script(self, df):
ids, acts = {}, []
for d in df:
id = d["id"]
ids[id] = copy.deepcopy(d["raw"])
acts.append({"update": {"_id": id, "_index": self.idxnm}})
acts.append(d["script"])
es_logger.info("bulk upsert: %s" % id)
res = []
for _ in range(10):
try:
if not self.version():
r = self.es.bulk(
index=self.idxnm,
body=acts,
refresh=False,
timeout="600s",
doc_type="doc")
else:
r = self.es.bulk(
index=self.idxnm,
body=acts,
refresh=False,
timeout="600s")
if re.search(r"False", str(r["errors"]), re.IGNORECASE):
return res
for it in r["items"]:
if "error" in it["update"]:
res.append(str(it["update"]["_id"]))
return res
except Exception as e:
es_logger.warning("Fail to bulk: " + str(e))
if re.search(r"(Timeout|time out)", str(e), re.IGNORECASE):
time.sleep(3)
continue
self.conn()
return res
def rm(self, d):
for _ in range(10):
try:
if not self.version():
r = self.es.delete(
index=self.idxnm,
id=d["id"],
doc_type="doc",
refresh=True)
else:
r = self.es.delete(
index=self.idxnm,
id=d["id"],
refresh=True,
doc_type="_doc")
es_logger.info("Remove %s" % d["id"])
return True
except Exception as e:
es_logger.warn("Fail to delete: " + str(d) + str(e))
if re.search(r"(Timeout|time out)", str(e), re.IGNORECASE):
time.sleep(3)
continue
if re.search(r"(not_found)", str(e), re.IGNORECASE):
return True
self.conn()
es_logger.error("Fail to delete: " + str(d))
return False
def search(self, q, idxnm=None, src=False, timeout="2s"):
if not isinstance(q, dict):
q = Search().query(q).to_dict()
for i in range(3):
try:
res = self.es.search(index=(self.idxnm if not idxnm else idxnm),
body=q,
timeout=timeout,
# search_type="dfs_query_then_fetch",
track_total_hits=True,
_source=src)
if str(res.get("timed_out", "")).lower() == "true":
raise Exception("Es Timeout.")
return res
except Exception as e:
es_logger.error(
"ES search exception: " +
str(e) +
"【Q】：" +
str(q))
if str(e).find("Timeout") > 0:
continue
raise e
es_logger.error("ES search timeout for 3 times!")
raise Exception("ES search timeout.")
def sql(self, sql, fetch_size=128, format="json", timeout="2s"):
for i in range(3):
try:
res = self.es.sql.query(body={"query": sql, "fetch_size": fetch_size}, format=format, request_timeout=timeout)
return res
except ConnectionTimeout as e:
es_logger.error("Timeout【Q】：" + sql)
continue
except Exception as e:
raise e
es_logger.error("ES search timeout for 3 times!")
raise ConnectionTimeout()
def get(self, doc_id, idxnm=None):
for i in range(3):
try:
res = self.es.get(index=(self.idxnm if not idxnm else idxnm),
id=doc_id)
if str(res.get("timed_out", "")).lower() == "true":
raise Exception("Es Timeout.")
return res
except Exception as e:
es_logger.error(
"ES get exception: " +
str(e) +
"【Q】：" +
doc_id)
if str(e).find("Timeout") > 0:
continue
raise e
es_logger.error("ES search timeout for 3 times!")
raise Exception("ES search timeout.")
def updateByQuery(self, q, d):
ubq = UpdateByQuery(index=self.idxnm).using(self.es).query(q)
scripts = ""
for k, v in d.items():
scripts += "ctx._source.%s = params.%s;" % (str(k), str(k))
ubq = ubq.script(source=scripts, params=d)
ubq = ubq.params(refresh=False)
ubq = ubq.params(slices=5)
ubq = ubq.params(conflicts="proceed")
for i in range(3):
try:
r = ubq.execute()
return True
except Exception as e:
es_logger.error("ES updateByQuery exception: " +
str(e) + "【Q】：" + str(q.to_dict()))
if str(e).find("Timeout") > 0 or str(e).find("Conflict") > 0:
continue
self.conn()
return False
def updateScriptByQuery(self, q, scripts, idxnm=None):
ubq = UpdateByQuery(
index=self.idxnm if not idxnm else idxnm).using(
self.es).query(q)
ubq = ubq.script(source=scripts)
ubq = ubq.params(refresh=True)
ubq = ubq.params(slices=5)
ubq = ubq.params(conflicts="proceed")
for i in range(3):
try:
r = ubq.execute()
return True
except Exception as e:
es_logger.error("ES updateByQuery exception: " +
str(e) + "【Q】：" + str(q.to_dict()))
if str(e).find("Timeout") > 0 or str(e).find("Conflict") > 0:
continue
self.conn()
return False
def deleteByQuery(self, query, idxnm=""):
for i in range(3):
try:
r = self.es.delete_by_query(
index=idxnm if idxnm else self.idxnm,
refresh = True,
body=Search().query(query).to_dict())
return True
except Exception as e:
es_logger.error("ES updateByQuery deleteByQuery: " +
str(e) + "【Q】：" + str(query.to_dict()))
if str(e).find("NotFoundError") > 0: return True
if str(e).find("Timeout") > 0 or str(e).find("Conflict") > 0:
continue
return False
def update(self, id, script, routing=None):
for i in range(3):
try:
if not self.version():
r = self.es.update(
index=self.idxnm,
id=id,
body=json.dumps(
script,
ensure_ascii=False),
doc_type="doc",
routing=routing,
refresh=False)
else:
r = self.es.update(index=self.idxnm, id=id, body=json.dumps(script, ensure_ascii=False),
routing=routing, refresh=False) # , doc_type="_doc")
return True
except Exception as e:
es_logger.error(
"ES update exception: " + str(e) + " id：" + str(id) + ", version:" + str(self.version()) +
json.dumps(script, ensure_ascii=False))
if str(e).find("Timeout") > 0:
continue
return False
def indexExist(self, idxnm):
s = Index(idxnm if idxnm else self.idxnm, self.es)
for i in range(3):
try:
return s.exists()
except Exception as e:
es_logger.error("ES updateByQuery indexExist: " + str(e))
if str(e).find("Timeout") > 0 or str(e).find("Conflict") > 0:
continue
return False
def docExist(self, docid, idxnm=None):
for i in range(3):
try:
return self.es.exists(index=(idxnm if idxnm else self.idxnm),
id=docid)
except Exception as e:
es_logger.error("ES Doc Exist: " + str(e))
if str(e).find("Timeout") > 0 or str(e).find("Conflict") > 0:
continue
return False
def createIdx(self, idxnm, mapping):
try:
if elasticsearch.__version__[0] < 8:
return self.es.indices.create(idxnm, body=mapping)
from elasticsearch.client import IndicesClient
return IndicesClient(self.es).create(index=idxnm,
settings=mapping["settings"],
mappings=mapping["mappings"])
except Exception as e:
es_logger.error("ES create index error %s ----%s" % (idxnm, str(e)))
def deleteIdx(self, idxnm):
try:
return self.es.indices.delete(idxnm, allow_no_indices=True)
except Exception as e:
es_logger.error("ES delete index error %s ----%s" % (idxnm, str(e)))
def getTotal(self, res):
if isinstance(res["hits"]["total"], type({})):
return res["hits"]["total"]["value"]
return res["hits"]["total"]
def getDocIds(self, res):
return [d["_id"] for d in res["hits"]["hits"]]
def getSource(self, res):
rr = []
for d in res["hits"]["hits"]:
d["_source"]["id"] = d["_id"]
d["_source"]["_score"] = d["_score"]
rr.append(d["_source"])
return rr
def scrollIter(self, pagesize=100, scroll_time='2m', q={
"query": {"match_all": {}}, "sort": [{"updated_at": {"order": "desc"}}]}):
for _ in range(100):
try:
page = self.es.search(
index=self.idxnm,
scroll=scroll_time,
size=pagesize,
body=q,
_source=None
)
break
except Exception as e:
es_logger.error("ES scrolling fail. " + str(e))
time.sleep(3)
sid = page['_scroll_id']
scroll_size = page['hits']['total']["value"]
es_logger.info("[TOTAL]%d" % scroll_size)
# Start scrolling
while scroll_size > 0:
yield page["hits"]["hits"]
for _ in range(100):
try:
page = self.es.scroll(scroll_id=sid, scroll=scroll_time)
break
except Exception as e:
es_logger.error("ES scrolling fail. " + str(e))
time.sleep(3)
# Update the scroll ID
sid = page['_scroll_id']
# Get the number of results that we returned in the last scroll
scroll_size = len(page['hits']['hits'])
ELASTICSEARCH = ESConnection()
|