moved db_preparation logic to fill-database phase

This commit is contained in:
2019-04-15 02:35:50 +02:00
parent 1494d4dfed
commit c2dea0bf70
8 changed files with 107 additions and 72 deletions

View File

@@ -431,33 +431,8 @@ def api_senses_update():
# APP PREFLIGHT ---------------------.
def prepare_db():
def helper_tid_to_token(tid, tokens):
for t in tokens:
if t["tid"] == tid:
return t
return None
# update entries (add headwords and fuctors for indexing)
for corpus in CORPORA:
for e in valdb[corpus].find({}):
if e["srl_links"] is None:
e["headwords"] = []
e["functors"] = []
else:
hw_tids = list(set([x["from"] for x in e["srl_links"]]))
hw_tokens = [helper_tid_to_token(tid, e["tokens"]) for tid in hw_tids]
headwords = [(t["lemma"] if t["msd"][0] == "G" else t["lemma"] + "_") for t in hw_tokens]
e["headwords"] = headwords
functors = list(set([x["afun"] for x in e["srl_links"]]))
e["functors"] = functors
valdb[corpus].save(e)
valdb[corpus].ensure_index([("headwords", pymongo.ASCENDING)])
valdb[corpus].ensure_index([("functors", pymongo.ASCENDING)])
def prepare_app_index():
log.info("[*] preparing app_index")
# create app_index (used in frontend, left side word index)
tmp_app_index = {c: {} for c in CORPORA}
for corpus in CORPORA:
@@ -504,6 +479,7 @@ if __name__ == "__main__":
print("Starting app.py main()")
aparser = argparse.ArgumentParser(description="Arguments for app.py")
aparser.add_argument("--config-file", type=str, help="check ./conf_files/")
aparser.add_argument('--prepare-db', action="store_true", default=False)
aparser.add_argument("--dbuser", type=str)
aparser.add_argument("--dbpass", type=str)
aparser.add_argument("--dbaddr", type=str)
@@ -530,8 +506,9 @@ if __name__ == "__main__":
)
valdb = client.valdb
if bool(config["prepare_db"]):
prepare_db()
if args.prepare_db:
prepare_app_index()
sys.exit()
# app index from db
app_index = (valdb.appindex.find_one({"dockey": "appindex"}))["data"]

View File

@@ -3,5 +3,4 @@ debug: True
port: 5004
host: localhost
logfile: "/var/log/valency_backend.log"
prepare_db: False
---

View File

@@ -1,7 +0,0 @@
---
debug: True
port: 5004
host: localhost
logfile: "/var/log/valency_backend.log"
prepare_db: True
---

View File

@@ -0,0 +1,70 @@
CORPORA = ["kres", "ssj"]
if __name__ == "__main__":
valdb = None
def helper_tid_to_token(tid, tokens):
for t in tokens:
if t["tid"] == tid:
return t
return None
# update entries (add headwords and fuctors for indexing)
for corpus in CORPORA:
for e in valdb[corpus].find({}):
if e["srl_links"] is None:
e["headwords"] = []
e["functors"] = []
else:
hw_tids = list(set([x["from"] for x in e["srl_links"]]))
hw_tokens = [helper_tid_to_token(tid, e["tokens"]) for tid in hw_tids]
headwords = [(t["lemma"] if t["msd"][0] == "G" else t["lemma"] + "_") for t in hw_tokens]
e["headwords"] = headwords
functors = list(set([x["afun"] for x in e["srl_links"]]))
e["functors"] = functors
valdb[corpus].save(e)
valdb[corpus].ensure_index([("headwords", pymongo.ASCENDING)])
valdb[corpus].ensure_index([("functors", pymongo.ASCENDING)])
# create app_index (used in frontend, left side word index)
tmp_app_index = {c: {} for c in CORPORA}
for corpus in CORPORA:
res_hws = {}
res_fns = {}
for e in valdb[corpus].find({}):
if "headwords" not in e:
continue
for hw in e["headwords"]:
if hw in res_hws:
res_hws[hw] += 1
else:
res_hws[hw] = 1
if "functors" not in e:
continue
for fn in e["functors"]:
if fn in res_fns:
res_fns[fn] += 1
else:
res_fns[fn] = 1
alphabetical = {}
for k, e in res_hws.items():
fst = k[0].lower()
if fst in alphabetical:
alphabetical[fst].append((k, e))
else:
alphabetical[fst] = [(k, e)]
for k, e in alphabetical.items():
alphabetical[k] = sorted(e, key=lambda x: x[0])
tmp_app_index[corpus]["words"] = alphabetical
functors = [(k, e) for (k, e) in res_fns.items()]
functors = sorted(functors, key=lambda x: x[0])
tmp_app_index[corpus]["functors"] = functors
valdb.appindex.update({"dockey": "appindex"}, {"dockey": "appindex", "data": tmp_app_index}, upsert=True)

View File

@@ -3632,14 +3632,12 @@
"balanced-match": {
"version": "1.0.0",
"bundled": true,
"dev": true,
"optional": true
"dev": true
},
"brace-expansion": {
"version": "1.1.11",
"bundled": true,
"dev": true,
"optional": true,
"requires": {
"balanced-match": "^1.0.0",
"concat-map": "0.0.1"
@@ -3654,20 +3652,17 @@
"code-point-at": {
"version": "1.1.0",
"bundled": true,
"dev": true,
"optional": true
"dev": true
},
"concat-map": {
"version": "0.0.1",
"bundled": true,
"dev": true,
"optional": true
"dev": true
},
"console-control-strings": {
"version": "1.1.0",
"bundled": true,
"dev": true,
"optional": true
"dev": true
},
"core-util-is": {
"version": "1.0.2",
@@ -3784,8 +3779,7 @@
"inherits": {
"version": "2.0.3",
"bundled": true,
"dev": true,
"optional": true
"dev": true
},
"ini": {
"version": "1.3.5",
@@ -3797,7 +3791,6 @@
"version": "1.0.0",
"bundled": true,
"dev": true,
"optional": true,
"requires": {
"number-is-nan": "^1.0.0"
}
@@ -3812,7 +3805,6 @@
"version": "3.0.4",
"bundled": true,
"dev": true,
"optional": true,
"requires": {
"brace-expansion": "^1.1.7"
}
@@ -3820,14 +3812,12 @@
"minimist": {
"version": "0.0.8",
"bundled": true,
"dev": true,
"optional": true
"dev": true
},
"minipass": {
"version": "2.2.4",
"bundled": true,
"dev": true,
"optional": true,
"requires": {
"safe-buffer": "^5.1.1",
"yallist": "^3.0.0"
@@ -3846,7 +3836,6 @@
"version": "0.5.1",
"bundled": true,
"dev": true,
"optional": true,
"requires": {
"minimist": "0.0.8"
}
@@ -3927,8 +3916,7 @@
"number-is-nan": {
"version": "1.0.1",
"bundled": true,
"dev": true,
"optional": true
"dev": true
},
"object-assign": {
"version": "4.1.1",
@@ -3940,7 +3928,6 @@
"version": "1.4.0",
"bundled": true,
"dev": true,
"optional": true,
"requires": {
"wrappy": "1"
}
@@ -4062,7 +4049,6 @@
"version": "1.0.2",
"bundled": true,
"dev": true,
"optional": true,
"requires": {
"code-point-at": "^1.0.0",
"is-fullwidth-code-point": "^1.0.0",