style: Refactoring imports, adapting MongoConnector to different connection_strings

This commit is contained in:
TrisNol
2023-06-30 20:36:03 +02:00
parent 3cd8860312
commit e44385ce3a
6 changed files with 145 additions and 8 deletions

View File

@ -4199,6 +4199,102 @@
" ) as export_file:\n",
" json.dump(dataclasses.asdict(company), export_file, ensure_ascii=False)"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"import sys\n",
"\n",
"module_path = os.path.abspath(os.path.join(\"..\"))\n",
"if module_path not in sys.path:\n",
" sys.path.append(module_path)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"mongodb://root:pR0R0v2e2@trisnol.tech:27017\n"
]
}
],
"source": [
"from News.utils.mongodb.mongo import MongoConnector\n",
"from Unternehmensregister.utils.CompanyMongoService import CompanyMongoService\n",
"\n",
"connector = MongoConnector(\n",
" hostname=\"trisnol.tech\",\n",
" database=\"transparenzregister\",\n",
" username=\"root\",\n",
" password=\"pR0R0v2e2\",\n",
" port=27017,\n",
")\n",
"\n",
"service = CompanyMongoService(connector)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
" 0%| | 0/3147 [00:00<?, ?it/s]"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|██████████| 3147/3147 [00:30<00:00, 102.30it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Inserted documents: 0\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"from tqdm import tqdm\n",
"import glob\n",
"import json\n",
"from Unternehmensregister.models.Company import Company\n",
"\n",
"num_inserted = 0\n",
"for file in tqdm(glob.glob1(\"./data/Unternehmensregister/transformed\", \"*.json\")):\n",
" path = os.path.join(\"./data/Unternehmensregister/transformed\", file)\n",
" with open(path, \"r\", encoding=\"utf-8\") as file_object:\n",
" data = json.loads(file_object.read())\n",
" company: Company = Company(**data)\n",
"\n",
" company_db = service.get_by_id(company.id)\n",
" if company_db is None:\n",
" service.insert(company)\n",
" num_inserted += 1\n",
"print(f\"Inserted documents: {num_inserted}\")"
]
}
],
"metadata": {