Added modifications to validate xsd

.gitignore

@@ -7,6 +7,17 @@ data/appindex.json
 src/frontend_vue/node_modules/
 src/frontend_vue/dist/
 dockerfiles/database/create.js
+dockerfiles/database/create_mongo.js
+dockerfiles/database/create_postgres.js
+dockerfiles/database/mongo_db.gz
+dockerfiles/database/postgres_db.tar
+dockerfiles/database/postgres_db_OLD.tar
 *__pycache__/
 env.local
 logs/*
+.idea/
+venv*
+data/
+data
+deploy_instructions/
+run.sh

.gitmodules

@@ -1,3 +1,6 @@
 [submodule "src/pkg/cjvt-corpusparser"]
 	path = src/pkg/cjvt-corpusparser
 	url = git@gitea.cjvt.si:kristjan/cjvt-corpusparser.git
+[submodule "src/pkg/luscenje_struktur"]
+	path = src/pkg/luscenje_struktur
+	url = https://gitea.cjvt.si/ozbolt/luscenje_struktur.git

Makefile

@@ -13,10 +13,11 @@ SSJ_FILE = "$(MAKE_ROOT)/data/ssj_file_link"
 # KRES_FOLDER = "$(MAKE_ROOT)/data/samples/kres_xml"
 # KRES_FOLDER = "$(MAKE_ROOT)/data/kres_xml_folder_link"
 KRES_FOLDER = "/home/kristjan/kres_data/payload/kres_xml"
+GIGAFIDA_FOLDER = "/home/lukakrsnik/cjvt-valency/data_all/giga_orig"
 # KRES_SRL_FOLDER = "$(MAKE_ROOT)/data/samples/kres_srl_json"
 # KRES_SRL_FOLDER = "$(MAKE_ROOT)/data/kres_json_folder_link"
 KRES_SRL_FOLDER = "/home/kristjan/kres_data/payload/kres_json"
-
+GIGAFIDA_SRL_FOLDER = "/home/lukakrsnik/cjvt-valency/data_all/final_json"
 # This file comes with the source code. Make sure you unpack it and name it right.
 SSKJ_WORDLIST = "$(MAKE_ROOT)/data/wordlist.json"
 SSKJ_JSON = "$(MAKE_ROOT)/data/sskj_senses.json"
@@ -26,14 +27,14 @@ APPINDEX_PATH = "$(MAKE_ROOT)/data/appindex.json"
 
 OUTPUT = "db"
 # OUTPUT = "file"
-OUTDIR = "/tmp/three"  # if you're running this in docker, make sure to mount the volume
+OUTDIR = "/project/data"  # if you're running this in docker, make sure to mount the volume
 DBADDR = "0.0.0.0:27017"  # don't use localhost
 
 # credentials from .gitignored file
 # create it from env.default
 include env.local
 
-N_CORES = 3
+N_CORES = 4
 # insert kres files into database in chunks, for fewer connections
 KRES_CHUNK_SIZE = 30
 
@@ -56,6 +57,12 @@ database-service:
 database-users:
 	cd dockerfiles/database; $(MAKE) create_users
 
+database-restore:
+	cd dockerfiles/database; $(MAKE) restore_db
+
+database-restore-postgres:
+	cd dockerfiles/database; $(MAKE) restore_postgres_db
+
 # also useful, if we want to restart the db
 database-clean:
 	cd dockerfiles/database; $(MAKE) clean_stack
@@ -69,6 +76,7 @@ python-env-install:
 	pip3 install -e src/pkg/cjvt-corpusparser/.
 	pip3 install -e src/pkg/valency/.
 	pip3 install -e src/pkg/seqparser/.
+	pip3 install -e src/pkg/luscenje_struktur/.
 
 # from inside python-env container:
 data/samples:
@@ -93,7 +101,14 @@ fill-database-kres: data/samples
 		--chunk-size $(KRES_CHUNK_SIZE) \
 		--cores $(N_CORES)
 
-
+fill-database-gigafida: data/samples
+	python3 src/pkg/cjvt-corpusparser/corpusparser/main.py --kres-folder $(GIGAFIDA_FOLDER) \
+		--corpus="gigafida" \
+		--ssj-file $(SSJ_FILE) --kres-srl-folder $(GIGAFIDA_SRL_FOLDER) \
+		--output $(OUTPUT) --outdir $(OUTDIR) --dbaddr $(DBADDR) \
+		--dbuser $(DB_USR_USER) --dbpass $(DB_USR_PASS) \
+		--chunk-size $(KRES_CHUNK_SIZE) \
+		--cores $(N_CORES)
 
 ## Frontend
 

README.md

@@ -179,3 +179,47 @@ $ mongorestore /dumps/valdb --db valdb --uri=mongodb://valuser:valuserpass@0.0.0
 ```
 
 After uploading, restart the stack with `27017` commented out.  
+
+## Script running
+
+### Environment setup
+```bash
+pip install -r requirements.txt
+pip install git+https://gitea.cjvt.si/ozbolt/luscenje_struktur.git
+pip install git+https://gitea.cjvt.si/kristjan/cjvt-corpusparser.git
+```
+
+### Running on already setup environment
+```bash
+make database-service
+```
+
+### Setting up environment for running on ramdisk
+
+```bash
+# create ramdisk
+sudo mount -t tmpfs tmpfs /mnt/tmp
+sudo mount -o remount,size=120G,noexec,nosuid,nodev,noatime /mnt/tmp
+
+# change volumes to /mnt/tmp:/data/db
+vim dockerfiles/database/valency-stack.yml
+
+# change Makefile -runStack to mkdir -p /mnt/tmp
+vim dockerfiles/database/Makefile
+
+# run service
+make database-service
+
+# run ONLY ONCE to create users and restore database
+make database-users
+make database-restore
+
+# double check if it worked
+docker exec -it ef0a /bin/bash
+
+# following steps in docker bash:
+    # check if it worked by
+    mongo --username <REGULAR USER> --password --authenticationDatabase valdb
+    db.getRoles()
+
+```

data/kres_json_folder_link

@@ -1 +0,0 @@
-/home/kristjan/workdir/final_json/

data/kres_xml_folder_link

@@ -1 +0,0 @@
-/home/kristjan/kres_mount/kres_parsed/tei/

data/ssj_file_link

@@ -1 +0,0 @@
-/home/kristjan/git/diploma/data/ssj500k-sl.TEI/ssj500k-sl.body.xml

dockerfiles/database/Dockerfile

@@ -1,5 +1,5 @@
-FROM mongo:latest
+FROM mongo:4.2.9
 
 WORKDIR /
-COPY init_inside_container.sh /.
-COPY create.js /.
+COPY init_inside_mongo_container.sh /.
+COPY create_mongo.js /.

dockerfiles/database/Makefile

@@ -2,33 +2,62 @@
 # collection names: lower case, plural
 # user names?
 
-# mongo admin -u root -p password --eval "db.getSiblingDB('vlDB').addUser('vluser', 'password')"
-
-STACKNAME = dbstack
-
-.PHONY: start_db FORCE
-
 all: build_run create_users
 
-build_run: build_mongo run_stack
+build_run: build_mongo run_docker_compose
 
-create.js: FORCE
+postgres_create_roles:
+	echo 'psql -v ON_ERROR_STOP=OFF --username $(DB_ADM_USER) <<-EOSQL' > create_postgres.js
+	echo "create user $(DB_USR_USER) with encrypted password '$(DB_USR_PASS)';" >> create_postgres.js
+	echo "create database superdb_small;" >> create_postgres.js
+	echo "grant all privileges on database superdb_small to $(DB_USR_USER);" >> create_postgres.js
+	echo "grant usage on schema public to $(DB_USR_USER);" >> create_postgres.js
+	echo "grant select on all tables in schema public to $(DB_USR_USER);" >> create_postgres.js
+	echo "EOSQL" >> create_postgres.js
+	chmod +x create_postgres.js
 
-FORCE:
-	echo 'db.auth("$(DB_ADM_USER)", "$(DB_ADM_PASS)")' > create.js
-	echo 'use valdb' >> create.js
-	echo 'db.createUser({user: "$(DB_USR_USER)", pwd: "$(DB_USR_PASS)", roles: ["readWrite"]})' >> create.js
+mongo_create_roles:
+	echo 'db.auth("$(DB_ADM_USER)", "$(DB_ADM_PASS)")' > create_mongo.js
+	echo 'use valdb' >> create_mongo.js
+	echo 'db.createUser({user: "$(DB_USR_USER)", pwd: "$(DB_USR_PASS)", roles: ["readWrite"]})' >> create_mongo.js
+	echo 'db.grantRolesToUser("$(DB_USR_USER)", [{ role: "readWrite", db: "extvaldb"}])' >> create_mongo.js
 
-build_mongo: create.js
+build_mongo: mongo_create_roles
 	docker build . -t my-mongo --no-cache
 
-clean_stack:
-	docker stack rm $(STACKNAME)
+# build_postgres: postgres_create_roles
+# 	docker build . -t my-mongo --no-cache
 
-run_stack:
-	mkdir -p ${HOME}/mongo_container/data/
-	docker stack deploy --compose-file mongodb-stack.yml $(STACKNAME)
+run_docker_compose:
+	mkdir -p ${HOME}/valency_data/mongo_container/data/
+	#docker kill $(shell ./get_mongo_container_name.sh)
+	#docker kill $(shell ./get_postgres_container_name.sh)
+	#docker-compose stop
+	docker-compose -f valency-stack.yml up -d --force-recreate
+	# docker stack deploy --compose-file mongodb-stack.yml $(STACKNAME)
 
-create_users: create.js
-	docker exec $(shell ./get_container_name.sh) /init_inside_container.sh
+create_users: create_mongo_users create_postgres_users
+
+
+create_mongo_users: mongo_create_roles
+	docker exec $(shell ./get_mongo_container_name.sh) /init_inside_mongo_container.sh
 	# rm create.js
+
+create_postgres_users: postgres_create_roles
+	docker exec $(shell ./get_postgres_container_name.sh) /scripts/init_inside_postgres_container.sh
+
+restore_db: restore_mongo_db restore_postgres_db
+
+restore_mongo_db:
+ifeq (,$(wildcard ./mongo_db.gz))
+	$(error "mongo_db.gz does not exists. Make sure to have dump of mongo db in 'dockerfiles/database/mongo_db.gz'")
+else
+	docker exec $(shell ./get_mongo_container_name.sh) sh -c 'mongorestore --gzip --archive=/scripts/mongo_db.gz --db valdb --username $(DB_USR_USER) --password $(DB_USR_PASS) --authenticationDatabase valdb'
+endif
+
+restore_postgres_db:
+ifeq (,$(wildcard ./postgres_db.tar))
+	$(error "postgres_db.tar does not exists. Make sure to have dump of postgres db in 'dockerfiles/database/postgres_db.tar'")
+else
+	docker exec $(shell ./get_postgres_container_name.sh) sh -c 'pg_restore -U $(DB_ADM_USER) --dbname=superdb_small --create --verbose /scripts/postgres_db.tar'
+endif

dockerfiles/database/get_postgres_container_name.sh

@@ -0,0 +1,2 @@
+#!/bin/bash
+docker ps | grep postgres | awk '{print $1}'

dockerfiles/database/init_inside_container.sh

@@ -1,3 +0,0 @@
-#!/bin/bash
-
-mongo admin < /create.js

dockerfiles/database/init_inside_mongo_container.sh

@@ -0,0 +1,3 @@
+#!/bin/bash
+
+mongo admin < /create_mongo.js

dockerfiles/database/init_inside_postgres_container.sh

@@ -0,0 +1,3 @@
+#!/bin/bash
+
+/scripts/create_postgres.js

dockerfiles/database/mongodb-stack.yml

@@ -1,26 +0,0 @@
-version: '3.1'
-
-services:
-
-  my_mongo:
-    image: my-mongo
-    restart: always
-    ports:
-      - 27017:27017
-    environment:
-      MONGO_INITDB_ROOT_USERNAME: ${DB_ADM_USER}
-      MONGO_INITDB_ROOT_PASSWORD: ${DB_ADM_PASS}
-    volumes:
-      - ${HOME}/mongo_container/data/:/data/db
-
-  mongo_express:
-    image: mongo-express
-    restart: always
-    ports:
-      - 8087:8081
-    environment:
-      ME_CONFIG_BASICAUTH_USERNAME: ${MONGOEXPRESS_USER}
-      ME_CONFIG_BASICAUTH_PASSWORD: ${MONGOEXPRESS_PASS}
-      ME_CONFIG_MONGODB_ADMINUSERNAME: ${DB_ADM_USER}
-      ME_CONFIG_MONGODB_ADMINPASSWORD: ${DB_ADM_PASS}
-      ME_CONFIG_MONGODB_SERVER: my_mongo

dockerfiles/database/valency-stack.yml

@@ -0,0 +1,27 @@
+version: '3.1'
+
+services:
+
+  my_mongo:
+    image: my-mongo
+    restart: always
+    ports:
+      - 127.0.0.1:27017:27017
+    environment:
+      MONGO_INITDB_ROOT_USERNAME: ${DB_ADM_USER}
+      MONGO_INITDB_ROOT_PASSWORD: ${DB_ADM_PASS}
+    volumes:
+      - ${HOME}/valency_data/mongo_container/data/:/data/db
+      - ./:/scripts
+
+  my_postgres:
+    image: postgres
+    restart: always
+    ports:
+      - 127.0.0.1:5432:5432
+    environment:
+      POSTGRES_USER: ${DB_ADM_USER}
+      POSTGRES_PASSWORD: ${DB_ADM_PASS}
+    volumes:
+      - ${HOME}/valency_data/postgres_container/data/:/var/lib/postgresql/data
+      - ./:/scripts

dockerfiles/python-env/Dockerfile

@@ -6,7 +6,8 @@ vim \
 python3 \
 python3-pip \
 sshfs \
-curl
+curl \
+locales
 
 RUN pip3 install --upgrade pip
 
@@ -21,6 +22,16 @@ RUN pip3 install \
 	flask_cors \
 	pymongo \
 	flask-pymongo \
-	gunicorn
+	gunicorn \
+	SQLAlchemy \
+	tqdm \
+	psycopg2-binary
+
+# Set the locale
+RUN sed -i -e 's/# en_US.UTF-8 UTF-8/en_US.UTF-8 UTF-8/' /etc/locale.gen && \
+    locale-gen
+ENV LANG en_US.UTF-8  
+ENV LANGUAGE en_US:en  
+ENV LC_ALL en_US.UTF-8
 
 ENV PYTHONIOENCODING UTF-8

nginx.conf

@@ -23,4 +23,7 @@ server {
         proxy_set_header Host $http_host;
         proxy_pass http://backend_flask:8084;
     }
-}
+}
+
+https://vezljivostni.cjvt.si/api/* -> http://vezljivostni-host.cjvt.si:8084/api/*
+https://vezljivostni.cjvt.si/* -> http://vezljivostni-host.cjvt.si:80/*

requirements.txt

@@ -0,0 +1,37 @@
+asn1crypto==0.24.0
+beautifulsoup4==4.8.0
+bs4==0.0.1
+cffi==1.12.3
+Click==7.0
+cryptography==2.1.4
+Flask==1.1.1
+Flask-Cors==3.0.8
+Flask-PyMongo==2.3.0
+gunicorn==19.9.0
+idna==2.6
+itsdangerous==1.1.0
+Jinja2==2.10.1
+joblib==0.13.2
+keyring==10.6.0
+keyrings.alt==3.0
+lxml==4.4.0
+MarkupSafe==1.1.1
+numpy==1.17.0
+pandas==0.25.0
+pathlib==1.0.1
+psycopg2==2.8.4
+pycparser==2.19
+pycrypto==2.6.1
+pymongo==3.8.0
+python-dateutil==2.8.0
+pytz==2019.2
+pyxdg==0.25
+PyYAML==5.1.2
+scikit-learn==0.21.3
+scipy==1.3.0
+SecretStorage==2.3.1
+six==1.11.0
+sklearn==0.0
+soupsieve==1.9.3
+SQLAlchemy==1.3.12
+Werkzeug==0.15.5

scripts/extract_keywords.py

@@ -0,0 +1,189 @@
+import copy
+import csv
+from xml.etree import ElementTree
+import re
+import sys
+import logging
+import argparse
+import pickle
+import time
+import gc
+import subprocess
+import concurrent.futures
+import tempfile
+
+
+def read_gigafida(path):
+    words = {}
+    with open(path) as tsvfile:
+        reader = csv.reader(tsvfile, delimiter='\t')
+        for row in reader:
+            words[row[0]] = int(row[2])
+    return words
+
+
+def read_sloleks(path):
+    words = set()
+    with open(path) as tsvfile:
+        reader = csv.reader(tsvfile, delimiter='\t')
+        for row in reader:
+            words.add(row[1])
+    return words
+
+
+def read_zele(path):
+    with open(path) as f:
+        content = f.readlines()
+        # fix content
+        content[0] = content[0][1:]
+    # a = content[2]
+    # a = content[2].split()
+    # a = content[2].split()[0].split('<IZT>')[1]
+    # a = content[2].split()[0].split('<IZT>')[1].split('</IZT>')[0]
+    content = [x.split()[0].split('<IZT>')[1].split('</IZT>')[0] for x in content]
+    # content = [x.split() for x in content]
+    return set(content)
+
+
+def read_wordlist(path):
+    with open(path) as f:
+        content = [line[:-1] for line in f.readlines()]
+    print(content[-1])
+    return set(content)
+
+
+def filter_gigafida(gigafida_raw, min_limit, max_limit):
+    return {word[0]: word[1] for word in gigafida_raw.items() if (word[0][-2:] == 'ti' or word[0][-2:] == 'či') and word[1] > min_limit and word[1] <= max_limit}
+
+
+def set_list_intersection(gigafida_filtered, sloleks):
+    intersection = {}
+    for word, num in gigafida_filtered.items():
+        if word in sloleks:
+            intersection[word] = num
+    return intersection
+
+
+def list_list_union(list1, list2):
+    union = copy.copy(list1)
+    for w, n in list2.items():
+        if w not in list1:
+            union[w] = list2[w]
+    return union
+
+
+def list_list_subtraction(list1, list2):
+    subtraction = {}
+    for w, n in list2.items():
+        # if w == 'dejati':
+        #     print('here')
+        if w not in list1:
+            subtraction[w] = n
+    return subtraction
+
+
+def set_set_subtraction(set1, set2):
+    subtraction = {}
+    for w in set2:
+        if w not in set1:
+            subtraction[w] = -1
+    return subtraction
+
+
+def create_document(list1, path):
+    with open(path, "w") as text_file:
+        for w, n in list1.items():
+            text_file.write("%s\t%d\n" % (w, n))
+
+
+def create_document_set(list1, path):
+    with open(path, "w") as text_file:
+        for w in sorted(list(list1)):
+            text_file.write("%s\n" % w)
+
+
+def gigafida_merge(sloleks, zele, gigafida_raw, giga_min, giga_max):
+    gigafida_filtered = filter_gigafida(gigafida_raw, giga_min, giga_max)
+    sloleks_gf_intersect = set_list_intersection(gigafida_filtered, sloleks)
+    gigafida_filtered1 = filter_gigafida(gigafida_raw, 1, sys.maxsize)
+    zele_gf_intersect = set_list_intersection(gigafida_filtered1, zele)
+    sloleks_zele_union = list_list_union(sloleks_gf_intersect, zele_gf_intersect)
+    sloleks_zele_subtraction = list_list_subtraction(sloleks_zele_union, gigafida_filtered)
+    return sloleks_zele_subtraction
+
+
+def main(args):
+    gigafida_raw = read_gigafida(args.gigafida_verb_list)
+    sloleks = read_sloleks(args.sloleks)
+    zele = read_zele(args.zele)
+    if args.wordlist is not None:
+        sloleks_wordlist = set()
+        # sloleks_wordlist = set()
+        for el in sloleks:
+            if el in gigafida_raw:
+                sloleks_wordlist.add(el)
+        filtered_wordlist = read_wordlist(args.wordlist)
+
+        # sloleks_wordlist = set()
+        for el in sloleks:
+            if el in gigafida_raw:
+                filtered_wordlist.add(el)
+
+        create_document_set(filtered_wordlist, 'wordlist.tsv')
+    # gigafida_merge(sloleks, zele, gigafida_raw, 3, sys.maxsize)
+    gigafida_filtered3 = filter_gigafida(gigafida_raw, 2, sys.maxsize)
+    sloleks_gf_intersect = set_list_intersection(gigafida_filtered3, sloleks)
+
+    nouns_sloleks_gf_intersect = sorted(sloleks_gf_intersect.items(), key=lambda x: x[1], reverse=True)
+    res = [el[0] for el in nouns_sloleks_gf_intersect]
+
+    gigafida_filtered1 = filter_gigafida(gigafida_raw, 0, sys.maxsize)
+    zele_gf_intersect = set_list_intersection(gigafida_filtered1, zele)
+    sloleks_zele_union = list_list_union(sloleks_gf_intersect, zele_gf_intersect)
+    sloleks_zele_subtraction = set_set_subtraction(sloleks, zele)
+    create_document(gigafida_filtered3, 'gigafida_3+.tsv')
+    # create_document(sloleks_gf_intersect, 'gigafida_3+-sloleks-presek.tsv')
+    create_document(sloleks_zele_union, 'gigafida_3+-sloleks_zele-presek.tsv')
+    create_document(sloleks_zele_subtraction, 'sloleks-zele-razlika.tsv')
+
+    # gigafida_filtered = filter_gigafida(gigafida_raw, 10, sys.maxsize)
+    # sloleks_zele_subtraction = list_list_subtraction(sloleks_zele_union, gigafida_filtered)
+    gigafida_10 = gigafida_merge(sloleks, zele, gigafida_raw, 10, sys.maxsize)
+    create_document(gigafida_10, 'gigafida_10+-sloleks_zele-razlika.tsv')
+
+    # gigafida_filtered = filter_gigafida(gigafida_raw, 3, 10)
+    # sloleks_zele_subtraction = list_list_subtraction(sloleks_zele_union, gigafida_filtered)
+    gigafida_3_10 = gigafida_merge(sloleks, zele, gigafida_raw, 2, 10)
+    create_document(gigafida_3_10, 'gigafida_3-10-sloleks_zele-razlika.tsv')
+    # pass
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(
+        description='Extract keywords from multiple lists.')
+    parser.add_argument('gigafida_verb_list',
+                        help='Path to gigafida list of verbs in tsv format.')
+    parser.add_argument('sloleks',
+                        help='Path to Sloleks in tsv format.')
+    parser.add_argument('--zele',
+                        help='Path to zele valency dictionary.')
+    parser.add_argument('--wordlist', default=None,
+                        help='Path to filtered wordlist.')
+    parser.add_argument('--handchecked_words', default=None,
+                        help='Path to handchecked words.')
+    # parser.add_argument('--min_limit',
+    #                     help='Limit min number of ocurrences',
+    #                     type=int, default=0)
+    # parser.add_argument('--max_limit',
+    #                     help='Limit max number of ocurrences',
+    #                     type=int, default=sys.maxsize)
+    parser.add_argument('--verbose', help='Enable verbose output to stderr',
+                        choices=["warning", "info", "debug"], default="info",
+                        const="info", nargs='?')
+
+    args = parser.parse_args()
+    logging.basicConfig(stream=sys.stderr, level=args.verbose.upper())
+
+    start = time.time()
+    main(args)
+    logging.info("TIME: {}".format(time.time() - start))

scripts/form_csv.py

@@ -0,0 +1,117 @@
+import argparse
+import csv
+import os
+
+from lxml import etree, objectify, html
+
+
+def write_general_statistics(path, out_list):
+    if len(out_list) == 0:
+        return
+    with open(path, 'w') as csvfile:
+        writer = csv.writer(csvfile, delimiter='\t',
+                                quotechar='"')
+        writer.writerow(['Semantic role', 'Valency pattern ratio', 'Valency sentence ratio'])
+        for line in out_list:
+            writer.writerow(line)
+
+
+def write_statistics(path, out_list):
+    if len(out_list) == 0:
+        return
+    with open(path, 'w') as csvfile:
+        writer = csv.writer(csvfile, delimiter='\t',
+                            quotechar='"')
+        writer.writerow(['Valency pattern id', 'Frequency all GF', 'Semantic role', 'Pattern representation', 'Corpus example'])
+        for line in out_list:
+            writer.writerow(line)
+
+
+def main(args):
+    for file in sorted(os.listdir(args.input)):
+        path = os.path.join(args.input, file)
+        tree = etree.parse(path)
+        gf_output = []
+        ssj_output = []
+        head = next(tree.iter('head'))
+        headword = head.find('headword').find('lemma').text
+        #for div in root.iterfind('.//div'):
+        for elem in tree.iter('statisticsContainer'):
+        # for element in tree.iterfind('statisticsContainer'):
+        # for element in tree.find('statisticsContainer'):
+            semRole = elem.find('semanticRole').text
+            gf_pattern = None
+            gf_sentence = None
+            ssj_pattern = None
+            ssj_sentence = None
+            measure = elem.find('measureList')
+            for el in measure:
+                if el.attrib['type'] == 'valency_pattern_ratio' and el.attrib['source'] == 'Gigafida 2.0':
+                    gf_pattern = el.text
+                if el.attrib['type'] == 'valency_sentence_ratio' and el.attrib['source'] == 'Gigafida 2.0':
+                    gf_sentence = el.text
+                if el.attrib['type'] == 'valency_pattern_ratio' and el.attrib['source'] == 'ssj500k 2.2':
+                    ssj_pattern = el.text
+                if el.attrib['type'] == 'valency_sentence_ratio' and el.attrib['source'] == 'ssj500k 2.2':
+                    ssj_sentence = el.text
+            if gf_pattern is not None and gf_sentence is not None:
+                gf_output.append([semRole, gf_pattern, gf_sentence])
+            if ssj_pattern is not None and ssj_sentence is not None:
+                ssj_output.append([semRole, ssj_pattern, ssj_sentence])
+
+        print(file)
+
+        analyze_output = []
+        for elem in tree.iter('valencyPattern'):
+            valency_pattern_id = elem.attrib['id']
+
+            # get frequency
+            measure = ''
+            for measure_el in elem.find('measureList').findall('measure'):
+                if measure_el.attrib['source'] == 'Gigafida 2.0':
+                    measure = measure_el.text
+
+            # get semantic roles
+            semantic_roles_list = []
+            for semantic_rol_con in elem.find('semanticRoleContainerList').findall('semanticRoleContainer'):
+                semantic_roles_list.append(semantic_rol_con.find('semanticRole').text)
+            semantic_roles = '_'.join(semantic_roles_list)
+
+            # pattern representation
+            pattern_representation = elem.find('patternRepresentation').text
+
+            # corpus example
+            if elem.find('exampleContainerList') is not None and elem.find('exampleContainerList').find('exampleContainer') is not None and elem.find('exampleContainerList').find('exampleContainer').find('corpusExample') is not None:
+                corpus_example_text = html.tostring(elem.find('exampleContainerList').find('exampleContainer').find('corpusExample'), encoding='unicode')
+
+            else:
+                continue
+
+            # ugly postprocessing to remove xmlns:xsi=... duh..
+            root = etree.fromstring(corpus_example_text)
+
+            # Remove namespace prefixes
+            for elem in root.getiterator():
+                elem.tag = etree.QName(elem).localname
+            # Remove unused namespace declarations
+            etree.cleanup_namespaces(root)
+
+            corpus_example = etree.tostring(root, encoding='unicode')
+
+            print(f"Valency pattern {valency_pattern_id}")
+
+
+            analyze_output.append([valency_pattern_id, measure, semantic_roles, pattern_representation, corpus_example])
+
+        write_general_statistics(os.path.join(args.output, headword + '_gf_stats.tsv'), gf_output)
+        write_general_statistics(os.path.join(args.output, headword + '_ssj_stats.tsv'), ssj_output)
+        write_statistics(os.path.join(args.output, headword + '_patterns.tsv'), analyze_output)
+
+if __name__ == '__main__':
+    arg_parser = argparse.ArgumentParser(description='Export and validate collocation data from DDD database.')
+    arg_parser.add_argument('--input', type=str, help='Input directory')
+    arg_parser.add_argument('--output', type=str, help='Output directory')
+
+    args = arg_parser.parse_args()
+
+    main(args)

scripts/valency

@@ -0,0 +1 @@
+../src/pkg/valency/valency

scripts/xsd_checker.py

@@ -0,0 +1,8 @@
+from lxml import etree as lxml
+
+with open('../data/inventory.xsd') as f:
+    xmlschema_doc = lxml.parse(f)
+    xmlschema = lxml.XMLSchema(xmlschema_doc)
+    with open('../data/xmls/output.xml') as op:
+        doc = lxml.parse(op)
+        print(xmlschema.validate(doc))

src/backend_flask/app.py

@@ -37,7 +37,8 @@ app = Flask(__name__)
 app.config.from_object("db_config")
 mongo = PyMongo(app)
 
-app.config["CORPORA"] = ["ssj", "kres"]
+# app.config["CORPORA"] = ["ssj", "kres", "gigafida"]
+app.config["CORPORA"] = ["gigafida"]
 app.config["BANNED_HEADWORDS"] = ["biti"]
 app.config["QUERY_LIMIT"] = 1000
 
@@ -247,20 +248,23 @@ def api_get_frames():
     if corpus not in app.config["CORPORA"]:
         return json.dumps({"error": "cor={kres,ssj}"})
 
+    log.info("Test1")
     cur = mongo.db[corpus].find({"headwords": hw})
+    log.info("Test2")
     frames = []
     for ent in cur[:app.config["QUERY_LIMIT"]]:
         frames += frames_from_db_entry(ent)  # pre-process this step for prod TODO
     cur.close()
-
+    log.info("Test3")
     # filter by relevant hw
     frames = [x for x in frames if x.hw == hw]
 
     ret_frames = RF(frames, mongo.db.sensemap)
-
+    log.info("Test3")
     json_ret = {"frames": []}
     for frame in ret_frames:
         json_ret["frames"].append(frame.to_json())
+    log.info("Test4")
     return json.dumps(json_ret)
     # return prepare_frames(ret_frames)
 
@@ -444,7 +448,7 @@ def _is_banned(hw):
         banned = False
     return banned
 
-def prepare_app_index(appindex_json, sskj_wordlist):
+def prepare_app_index(appindex_json):
     log.info("[*] preparing app_index")
     # create app_index (used in frontend, left side word index)
     tmp_app_index = {c: {} for c in app.config["CORPORA"]}
@@ -452,6 +456,14 @@ def prepare_app_index(appindex_json, sskj_wordlist):
         res_hws = {}
         res_fns = {}
 
+        # print('CORPUS...!!...')
+        # print(corpus)
+        # a = mongo.db[corpus]
+        # print('TEST_OK')
+        # print(a)
+        # print(mongo.db)
+        # a = mongo.db.list_collection_names()
+        # print('TEST_OK2')
         nentries = mongo.db[corpus].count()
         idx = 0
         for e in mongo.db[corpus].find({}):
@@ -484,6 +496,7 @@ def prepare_app_index(appindex_json, sskj_wordlist):
 
         for letter, words in alphabetical.items():
             filtered_words = [x for x in words if not _is_banned(x[0])]
+            # filtered_words = [x for x in words]
             alphabetical[letter] = sorted(filtered_words, key=lambda x: x[0])
 
         tmp_app_index[corpus]["words"] = alphabetical
@@ -560,12 +573,16 @@ if __name__ == "__main__":
     if args.prepare_db:
         with Path(args.sskj_wordlist).open("r") as fp:
             sskj_wordlist = json.load(fp)
-        prepare_app_index(args.appindex_json, sskj_wordlist)
+        prepare_app_index(args.appindex_json)
         sys.exit()
 
     # app index from db
     with Path(args.appindex_json).open("r") as fp:
         app.config["app_index"] = json.load(fp)
+        # a = app.config["app_index"]
+        # b = app.config["app_index"]["kres"]
+        # c = app.config["app_index"]["kres"]["words"]
+        # print('HERE')
 
     # log.info("[*] Starting app.py with config:\n%s".format(config))
     log.info("[*] Starting app.py with config:\n{}".format(config))

src/backend_flask/build_app_index.py

@@ -0,0 +1,106 @@
+import argparse
+import json
+
+from flask import Flask
+from flask_pymongo import PyMongo
+from pathlib import Path
+
+app = Flask(__name__)
+
+app.config.from_object("db_config")
+mongo = PyMongo(app)
+
+app.config["BANNED_HEADWORDS"] = ["biti"]
+
+def _is_banned(hw):
+    banned = True
+    if hw in app.config["BANNED_HEADWORDS"]:
+        banned = True
+    elif hw in sskj_wordlist["wordlist"]:
+        banned = False
+    elif (hw + " se") in sskj_wordlist["wordlist"]:
+        banned = False
+    return banned
+
+
+def prepare_app_index(appindex_json, corporas, previous_json=None):
+    if previous_json:
+        with Path(previous_json).open("r") as fp:
+            tmp_app_index = json.load(fp)
+    else:
+        tmp_app_index = {}
+    # create app_index (used in frontend, left side word index)
+    for c in corporas:
+        tmp_app_index[c] = {}
+
+    for corpus in corporas:
+        res_hws = {}
+        res_fns = {}
+
+        # print('CORPUS...!!...')
+        # print(corpus)
+        # a = mongo.db[corpus]
+        # print('TEST_OK')
+        # print(a)
+        # print(mongo.db)
+        # a = mongo.db.list_collection_names()
+        # print('TEST_OK2')
+        nentries = mongo.db[corpus].count()
+        idx = 0
+        for e in mongo.db[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
+            idx += 1
+            if idx % 10000 == 0:
+                print("indexing {}: {}/{}".format(
+                    corpus, idx, nentries))
+
+        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 letter, words in alphabetical.items():
+            filtered_words = [x for x in words if not _is_banned(x[0])]
+            # filtered_words = [x for x in words]
+            alphabetical[letter] = sorted(filtered_words, 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
+
+    with Path(appindex_json).open("w") as fp:
+        json.dump(tmp_app_index, fp)
+
+if __name__ == "__main__":
+    print("Starting app.py main()")
+    aparser = argparse.ArgumentParser(description="Arguments for app.py")
+    aparser.add_argument("--previous-json", type=str, default=None)
+    aparser.add_argument("--appindex-json", type=str)
+    aparser.add_argument("--sskj-wordlist", type=str)
+    args = aparser.parse_args()
+
+    corporas = ['gigafida']
+
+    with Path(args.sskj_wordlist).open("r") as fp:
+        sskj_wordlist = json.load(fp)
+
+    prepare_app_index(args.appindex_json, corporas, args.previous_json)

src/backend_flask/db_config.py

@@ -1,2 +1,2 @@
-MONGO_URI = "mongodb://sizif:p5e3r4u8t7@my_mongo:27017/valdb"
+MONGO_URI = "mongodb://user:user@0.0.0.0:27017/valdb"
 MONGO_AUTH_SOURCE = 'admin'

src/backend_flask/get_sentence_ids.py

@@ -0,0 +1,18 @@
+import json
+import os
+
+input_dir = "/media/luka/Portable Disk/Datasets/gigafida_jos/final_json"
+output_file = "../../all_sentences.json"
+
+results = {}
+filenames = os.listdir(input_dir)
+len(filenames)
+for i, filename in enumerate(filenames):
+    if filename.endswith(".json"):
+        with open(os.path.join(input_dir, filename)) as json_file:
+            data = json.load(json_file)
+            results[filename.split('-')[0]] = list(data.keys())
+        print('Progress: %.2f %%' % (i/len(filenames)))
+
+with open(output_file, 'w') as f:
+    json.dump(results, f)

src/frontend_vue/config/config.json

@@ -1,3 +1,3 @@
 {
-    "api_addr": "http://193.2.76.103:8084"
+    "api_addr": "http://0.0.0.0:8084"
 }

src/frontend_vue/package-lock.json

@@ -3513,14 +3513,12 @@
         "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"
@@ -3535,20 +3533,17 @@
         "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",
@@ -3665,8 +3660,7 @@
         "inherits": {
           "version": "2.0.3",
           "bundled": true,
-          "dev": true,
-          "optional": true
+          "dev": true
         },
         "ini": {
           "version": "1.3.5",
@@ -3678,7 +3672,6 @@
           "version": "1.0.0",
           "bundled": true,
           "dev": true,
-          "optional": true,
           "requires": {
             "number-is-nan": "^1.0.0"
           }
@@ -3693,7 +3686,6 @@
           "version": "3.0.4",
           "bundled": true,
           "dev": true,
-          "optional": true,
           "requires": {
             "brace-expansion": "^1.1.7"
           }
@@ -3701,14 +3693,12 @@
         "minimist": {
           "version": "0.0.8",
           "bundled": true,
-          "dev": true,
-          "optional": true
+          "dev": true
         },
         "minipass": {
           "version": "2.3.5",
           "bundled": true,
           "dev": true,
-          "optional": true,
           "requires": {
             "safe-buffer": "^5.1.2",
             "yallist": "^3.0.0"
@@ -3727,7 +3717,6 @@
           "version": "0.5.1",
           "bundled": true,
           "dev": true,
-          "optional": true,
           "requires": {
             "minimist": "0.0.8"
           }
@@ -3808,8 +3797,7 @@
         "number-is-nan": {
           "version": "1.0.1",
           "bundled": true,
-          "dev": true,
-          "optional": true
+          "dev": true
         },
         "object-assign": {
           "version": "4.1.1",
@@ -3821,7 +3809,6 @@
           "version": "1.4.0",
           "bundled": true,
           "dev": true,
-          "optional": true,
           "requires": {
             "wrappy": "1"
           }
@@ -3943,7 +3930,6 @@
           "version": "1.0.2",
           "bundled": true,
           "dev": true,
-          "optional": true,
           "requires": {
             "code-point-at": "^1.0.0",
             "is-fullwidth-code-point": "^1.0.0",

src/frontend_vue/src/components/Nav.vue

@@ -62,7 +62,7 @@ export default {
     name: "Nav",
     props: ["appState"],
     data() {return {
-        optCorpora: ["kres", "ssj"],
+        optCorpora: ["kres", "ssj", "gigafida"],
         optIndexes: [
             {key: "besede", val: "words"},
             {key: "udeleženske vloge", val: "functors"},

src/pkg/cjvt-corpusparser

@@ -1 +1 @@
-Subproject commit 01adf47b9b63b43f86bff52429792b0de2327ddd
+Subproject commit 92b3ac4ea3a73b93c25b363b5b9cb096d4d011cd

src/pkg/luscenje_struktur

@@ -0,0 +1 @@
+Subproject commit 8c87d07b8a3ca73faac2fac30c39969bc5f97d45

src/pkg/valency/valency/Frame.py

@@ -3,6 +3,41 @@ from corpusparser import enriched_lemma
 
 log = logging.getLogger(__name__)
 
+def frames_from_db_entry_headword(dbent, headword):
+    def _full_tid(tid):
+        return ".".join([dbent["sid"], str(tid)])
+
+    token_dict = {str(x["tid"]): x for x in dbent["tokens"]}
+
+    frames = []
+    if "srl_links" not in dbent:
+        return []
+    srldict = {}
+    for srl in dbent["srl_links"]:
+        key = str(srl["from"])
+        if enriched_lemma(token_dict[key]) != headword:
+            continue
+        if key not in srldict:
+            srldict[key] = [srl]
+        else:
+            srldict[key] += [srl]
+    for hwtid, srlarr in srldict.items():
+        frames += [Frame(
+            hw_lemma=enriched_lemma(token_dict[hwtid]),
+            tids=[_full_tid(hwtid)],
+            slots=[
+                Slot(
+                    functor=srl["afun"],
+                    tids=[_full_tid(srl["to"])]
+                ) for srl in srlarr
+            ],
+            # sentences=[(dbent["sid"], dbent["tokens"])],
+            sentences=[
+                [(_full_tid(t["tid"]), t) for t in dbent["tokens"]],
+            ]
+        )]
+    return frames
+
 def frames_from_db_entry(dbent):
     def _full_tid(tid):
         return ".".join([dbent["sid"], str(tid)])