Add Python web scraper for NJC travel rates with currency extraction

- Implemented Python scraper using BeautifulSoup and pandas to automatically collect travel rates from official NJC website
- Added currency extraction from table titles (supports EUR, USD, AUD, CAD, ARS, etc.)
- Added country extraction from table titles for international rates
- Flatten pandas MultiIndex columns for cleaner data structure
- Default to CAD for domestic Canadian sources (accommodations and domestic tables)
- Created SQLite database schema (raw_tables, rate_entries, exchange_rates, accommodations)
- Successfully scraped 92 tables with 17,205 rate entries covering 25 international cities
- Added migration script to convert scraped data to Node.js database format
- Updated .gitignore for Python files (.venv/, __pycache__, *.pyc, *.sqlite3)
- Fixed city validation and currency conversion in main app
- Added comprehensive debug and verification scripts

This replaces manual JSON maintenance with automated data collection from official government source.
This commit is contained in:
2026-01-13 09:21:43 -05:00
commit 15094ac94b
84 changed files with 19859 additions and 0 deletions

208
documents/scrapers.py Normal file
View File

@@ -0,0 +1,208 @@
from __future__ import annotations
import json
import re
from dataclasses import dataclass
from typing import Any, Iterable
import pandas as pd
import requests
from bs4 import BeautifulSoup
USER_AGENT = "GovTravelScraper/1.0 (+https://example.com)"
@dataclass(frozen=True)
class SourceConfig:
name: str
url: str
SOURCES = [
SourceConfig(name="international", url="https://www.njc-cnm.gc.ca/directive/app_d.php?lang=en"),
SourceConfig(name="domestic", url="https://www.njc-cnm.gc.ca/directive/d10/v325/s978/en"),
SourceConfig(name="accommodations", url="https://rehelv-acrd.tpsgc-pwgsc.gc.ca/lth-crl-eng.aspx"),
]
def fetch_html(url: str) -> str:
response = requests.get(url, headers={"User-Agent": USER_AGENT}, timeout=60)
response.raise_for_status()
response.encoding = response.apparent_encoding
return response.text
def extract_tables(html: str) -> list[pd.DataFrame]:
return pd.read_html(html)
def _normalize_header(header: str) -> str:
return re.sub(r"\s+", " ", header.strip().lower())
def _parse_amount(value: Any) -> float | None:
if value is None:
return None
text = str(value)
match = re.search(r"-?\d+(?:[\.,]\d+)?", text)
if not match:
return None
amount_text = match.group(0).replace(",", "")
try:
return float(amount_text)
except ValueError:
return None
def _detect_currency(value: Any, fallback: str | None = None) -> str | None:
if value is None:
return fallback
text = str(value).upper()
if "CAD" in text:
return "CAD"
if "USD" in text:
return "USD"
match = re.search(r"\b[A-Z]{3}\b", text)
if match:
return match.group(0)
return fallback
def _table_title_map(html: str) -> dict[int, str]:
soup = BeautifulSoup(html, "html.parser")
titles: dict[int, str] = {}
for index, table in enumerate(soup.find_all("table")):
heading = table.find_previous(["h1", "h2", "h3", "h4", "caption"])
if heading:
titles[index] = heading.get_text(strip=True)
return titles
def scrape_tables_from_source(source: SourceConfig) -> list[dict[str, Any]]:
html = fetch_html(source.url)
tables = extract_tables(html)
title_map = _table_title_map(html)
results = []
for index, table in enumerate(tables):
data = json.loads(table.to_json(orient="records"))
results.append(
{
"table_index": index,
"title": title_map.get(index),
"data": data,
}
)
return results
def extract_rate_entries(
source: SourceConfig,
tables: Iterable[dict[str, Any]],
) -> list[dict[str, Any]]:
entries: list[dict[str, Any]] = []
for table in tables:
for row in table["data"]:
normalized = {_normalize_header(k): v for k, v in row.items()}
country = normalized.get("country") or normalized.get("country/territory")
city = normalized.get("city") or normalized.get("location")
province = normalized.get("province") or normalized.get("province/territory")
currency = _detect_currency(normalized.get("currency"))
effective_date = normalized.get("effective date") or normalized.get("effective")
for key, value in normalized.items():
if key in {"country", "country/territory", "city", "location", "province", "province/territory", "currency", "effective", "effective date"}:
continue
amount = _parse_amount(value)
if amount is None:
continue
entry_currency = _detect_currency(value, fallback=currency)
entries.append(
{
"source": source.name,
"source_url": source.url,
"country": country,
"city": city,
"province": province,
"currency": entry_currency,
"rate_type": key,
"rate_amount": amount,
"unit": None,
"effective_date": effective_date,
"raw": row,
}
)
return entries
def extract_exchange_rates(
source: SourceConfig,
tables: Iterable[dict[str, Any]],
) -> list[dict[str, Any]]:
entries: list[dict[str, Any]] = []
for table in tables:
for row in table["data"]:
normalized = {_normalize_header(k): v for k, v in row.items()}
currency = (
normalized.get("currency")
or normalized.get("currency code")
or normalized.get("code")
)
rate = (
normalized.get("exchange rate")
or normalized.get("rate")
or normalized.get("cad rate")
or normalized.get("rate to cad")
)
rate_amount = _parse_amount(rate)
if not currency or rate_amount is None:
continue
entries.append(
{
"source": source.name,
"source_url": source.url,
"currency": _detect_currency(currency),
"rate_to_cad": rate_amount,
"effective_date": normalized.get("effective date") or normalized.get("date"),
"raw": row,
}
)
return entries
def extract_accommodations(
source: SourceConfig,
tables: Iterable[dict[str, Any]],
) -> list[dict[str, Any]]:
entries: list[dict[str, Any]] = []
for table in tables:
for row in table["data"]:
normalized = {_normalize_header(k): v for k, v in row.items()}
property_name = (
normalized.get("property")
or normalized.get("hotel")
or normalized.get("accommodation")
or normalized.get("name")
)
if not property_name and not normalized.get("city"):
continue
rate_amount = _parse_amount(
normalized.get("rate")
or normalized.get("room rate")
or normalized.get("daily rate")
)
currency = _detect_currency(normalized.get("rate"))
entries.append(
{
"source": source.name,
"source_url": source.url,
"property_name": property_name,
"address": normalized.get("address"),
"city": normalized.get("city") or normalized.get("location"),
"province": normalized.get("province") or normalized.get("province/territory"),
"phone": normalized.get("phone") or normalized.get("telephone"),
"rate_amount": rate_amount,
"currency": currency,
"effective_date": normalized.get("effective date") or normalized.get("effective"),
"raw": row,
}
)
return entries