"""
Custom series example — build series IDs from scratch using the series module.
Run from the bls/ directory:  python3 examples/custom_series.py
"""

import sys, os
sys.path.insert(0, os.path.dirname(os.path.dirname(__file__)))

from config import BLS_API_KEY
from bls_client import BLSClient
from bls_client import series

client = BLSClient(BLS_API_KEY)

# ── Build LAUS series for any state ─────────────────────────────────────────
print("=== LAUS: unemployment rates for New York, Texas, California ===")
state_series = {
    "New York":    series.laus_state(36, "rate"),
    "Texas":       series.laus_state(48, "rate"),
    "California":  series.laus_state(6,  "rate"),
}
results = client.fetch_named(state_series, 2025, 2026)
for label, s in results.items():
    obs = client.latest_obs(s)
    if obs:
        print(f"  {label:<15} {obs['value']}%")

# ── Build CES for specific industries ───────────────────────────────────────
print("\n=== CES: employment by sector ===")
industry_series = {
    "Healthcare":      series.ces_national("education_health", "employees"),
    "Retail":          series.ces_national("retail", "employees"),
    "Manufacturing":   series.ces_national("manufacturing", "employees"),
    "Leisure/Hosp":    series.ces_national("leisure", "employees"),
}
results = client.fetch_named(industry_series, 2025, 2026)
for label, s in results.items():
    obs = client.latest_obs(s)
    if obs:
        print(f"  {label:<20} {float(obs['value']):>10,.0f}K  ({obs['periodName']} {obs['year']})")

# ── Build CPI for specific items ─────────────────────────────────────────────
print("\n=== CPI: selected item prices (index, 1982-84=100) ===")
cpi_series = {
    "All Items":   series.cpi("all_items"),
    "Shelter":     series.cpi("shelter"),
    "Gasoline":    series.cpi("gasoline", seasonal=True),
    "New Vehicles":series.cpi("new_vehicles"),
}
results = client.fetch_named(cpi_series, 2025, 2026)
for label, s in results.items():
    obs = client.latest_obs(s)
    if obs:
        print(f"  {label:<20} {obs['value']:>8}  ({obs['periodName']} {obs['year']})")

# ── JOLTS: full dashboard ────────────────────────────────────────────────────
print("\n=== JOLTS: labor market flows ===")
from bls_client.queries import employment
results = client.fetch_named(employment.jolts_dashboard(), 2025, 2026)
for label, s in results.items():
    obs = client.latest_obs(s)
    if obs:
        print(f"  {label:<25} {float(obs['value']):>8,.0f}K  ({obs['periodName']} {obs['year']})")

# ── OES: wages for specific occupations ─────────────────────────────────────
print("\n=== OES: annual median wages by occupation ===")
from bls_client.queries import wages
from bls_client.queries.wages import SOC_CODES
occ_series = {
    label.replace("_", " ").title(): wages.occupation_annual_median_wage(code)
    for label, code in list(SOC_CODES.items())[:5]
}
results = client.fetch_named(occ_series, 2023, 2024)
for label, s in results.items():
    obs = client.latest_obs(s)
    if obs:
        print(f"  {label:<35} ${float(obs['value']):>10,.0f}  ({obs['year']})")