Update README.md

.env.example

@@ -1,5 +1,9 @@
 REFETCH_PERIOD_H=6
+STORE_NULL_BID=True
+
 POSTGRES_USER=lenenergo
 POSTGRES_PASSWORD=lenenergo
 POSTGRES_DB=lenenergo
 POSTGRES_HOST=db
+# or
+DB_URL=postgresql://lenenergo:lenenergo@localhost:5432

.gitignore

@@ -5,3 +5,5 @@ __pycache__
 data*.csv
 .idea/
 .ipynb_checkpoints
+.vscode/
+*.odb

.vscode/launch.json

@@ -1,15 +0,0 @@
-{
-    // Use IntelliSense to learn about possible attributes.
-    // Hover to view descriptions of existing attributes.
-    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
-    "version": "0.2.0",
-    "configurations": [
-        {
-            "name": "Python: Module",
-            "type": "python",
-            "request": "launch",
-            "module": "parser",
-            "justMyCode": true,
-        }
-    ]
-}

README.md

@@ -0,0 +1,37 @@
+# Lenenergo Parser
+
+## DB columns settings
+
+Append to `runner/config.py`
+
+```python
+DB_COLUMNS_MAP["<COL_NS key>"] = "<corresponding db column name>"
+```
+
+## Running instructions
+
+```bash
+docker build . -it lenenergo_parser
+docker run -d \
+  [-e REFETCH_PERIOD_H=4] \ # Refetch period
+  [-e STORE_NULL_BID=False] \ # Store rows with null building_id
+  # DB auth variants
+  [-e POSTGRES_USER=lenenergo] \
+  [-e POSTGRES_PASSWORD=lenenergo] \
+  [-e POSTGRES_DB=lenenergo] \
+  [-e POSTGRES_HOST=localhost] \
+  [-e POSTGRES_PORT=5432] \
+  # or
+  [DB_URL=postgresql://lenenergo:lenenergo@localhost:5432/lenenergo] \
+  lenenergo_parser
+```
+
+## Dev instructions
+
+```bash
+python -m venv .venv
+
+pip install -r requirements.txt
+
+python -m runner
+```

parser/README.md

@@ -31,12 +31,18 @@ def split_addresses(df: pd.DataFrame) -> pd.DataFrame
 ```
 - `get_building_id`:
 ```python
-def get_building_id(street: str) -> Tuple[Optional[int], Optional[float], Optional[float]]
+def get_building_id(street: str) -> GeoTupleType
 ```
 - `fetch_builing_ids`:
 ```python
 def fetch_builing_ids(df: pd.DataFrame) -> pd.DataFrame
 ```
+- `async_fetch_building_id`:
+```python
+async def async_fetch_building_id(
+    session: aiohttp.ClientSession, street: str
+) -> GeoTupleType
+```
 - `async_fetch_building_ids`:
 ```python
 async def async_fetch_building_ids(df: pd.DataFrame) -> pd.DataFrame
@@ -51,11 +57,15 @@ def preprocess_df(df: pd.DataFrame) -> pd.DataFrame
 ```
 - `COL_NS`:
 ```python
-COL_NS: Dict[str, str]
+COL_NS: dict[str, str]
 ```
 - `ICOL_NS`:
 ```python
-ICOL_NS: Dict[str, str]
+ICOL_NS: dict[str, str]
+```
+- `PR_COL_NS`:
+```python
+PR_COL_NS: tuple[str]
 ```
 - `preprocess_read_df`:
 ```python
@@ -82,4 +92,4 @@ pip install -r requirements.txt
 python -m parser [<Период в часах>]
 ```
 
-Формат сохраняемых файлов: `data_%d-%m-%y_%H:%M.csv`
+Формат сохраняемых файлов: `data_%d-%m-%y_%H.%M.csv`

parser/__init__.py

@@ -1,5 +1,6 @@
 from .address import split_addresses
 from .building_id import (
+    GeoTupleType,
     async_fetch_building_id,
     async_fetch_building_ids,
     concurrent_fetch_builing_ids,
@@ -11,20 +12,23 @@ from .pipeline import pipeline
 from .preprocess import (
     COL_NS,
     ICOL_NS,
+    PR_COL_NS,
     group_by_index,
     preprocess_df,
     preprocess_read_df,
 )
 
 __all__ = (
+    "COL_NS",
+    "GeoTupleType",
+    "ICOL_NS",
+    "PR_COL_NS",
     "async_fetch_building_id",
     "async_fetch_building_ids",
-    "COL_NS",
     "concurrent_fetch_builing_ids",
     "fetch_builing_ids",
     "get_building_id",
     "group_by_index",
-    "ICOL_NS",
     "LenenergoParser",
     "pipeline",
     "preprocess_df",

parser/building_id.py

@@ -45,7 +45,10 @@ async def async_fetch_building_id(
     async with session.get(
         "https://geocode.gate.petersburg.ru/parse/eas", params={"street": street}
     ) as r:
+        try:
             res = await r.json()
+        except aiohttp.client_exceptions.ContentTypeError:
+            res = "error"
 
         if "error" in res:
             return None, None, None

parser/preprocess.py

@@ -21,6 +21,23 @@ COL_NS = {
     "lng": "Долгота",
 }
 
+PR_COL_NS = (
+    "index",
+    "region",
+    "area",
+    "town",
+    "street",
+    "branch",
+    "res",
+    "comment",
+    "building_id",
+    "lat",
+    "lng",
+    "start",
+    "finish",
+)
+
+
 ICOL_NS = dict(map(reversed, COL_NS.items()))
 
 

runner/__main__.py

@@ -1,11 +1,17 @@
 import schedule
 
-from .config import REFETCH_PERIOD_H
+from .config import REFETCH_PERIOD_H, STORE_NULL_BID
 from .job import job
 from .scheduler import run_continuously
 
 schedule.every(REFETCH_PERIOD_H).hours.do(job)
 stop_run_continuously = run_continuously()
 
+print(
+    f"Scheduled to run every {REFETCH_PERIOD_H} hour and "
+    + ("" if STORE_NULL_BID else "not ")
+    + "to store NULL building_id"
+)
+
 # First run
 job()

runner/config.py

@@ -1,6 +1,8 @@
 import os
+from parser import PR_COL_NS
 
 REFETCH_PERIOD_H = int(os.environ.get("REFETCH_PERIOD_H", "4"))
+STORE_NULL_BID = os.environ.get("STORE_NULL_BID", "False") == "True"
 
 POSTGRES_USER = os.environ.get("POSTGRES_USER", "lenenergo")
 POSTGRES_PASSWORD = os.environ.get("POSTGRES_PASSWORD", "lenenergo")
@@ -9,3 +11,9 @@ POSTGRES_HOST = os.environ.get("POSTGRES_HOST", "localhost")
 POSTGRES_PORT = int(os.environ.get("POSTGRES_PORT", "5432"))
 
 DB_URL = os.environ.get("DB_URL", None)
+
+DB_COLUMNS_MAP = dict(zip(PR_COL_NS, PR_COL_NS))
+"""
+Feel free to rewrite mapping like 
+DB_COLUMNS_MAP["<COL_NS key>"] = "<corresponding db column name>"
+"""

runner/job.py

@@ -4,23 +4,12 @@ from parser import pipeline
 import pandas as pd
 import psycopg
 
+from .config import DB_COLUMNS_MAP, STORE_NULL_BID
 from .database import db_credentials
 
-sql_statement = """COPY records (
+sql_statement = "".join(
-    index,
+    ("COPY records (", ", ".join(DB_COLUMNS_MAP.values()), ") FROM STDIN")
-    region,
+)
-    area,
-    town,
-    street,
-    branch,
-    res,
-    comment,
-    building_id,
-    lat,
-    lng,
-    start,
-    finish
-) FROM STDIN"""
 
 
 def job():
@@ -36,6 +25,8 @@ def job():
             with cursor.copy(sql_statement) as copy:
                 for _, row in parser.df.iterrows():
                     row = row.where((pd.notnull(row)), None)
-                    copy.write_row(row.to_list())
+                    if row["building_id"] is not None or STORE_NULL_BID:
+                        db_row = row.rename(DB_COLUMNS_MAP)
+                        copy.write_row(db_row.to_list())
 
     print(f"Fetched in {datetime.now() - fetch_start}\n{parser}")