from inei_microdatos import load_catalog, download_modules, read_module
from inei_microdatos.catalog import filter_catalog

# Cargar el catálogo incluido (viene con el paquete, sin configuración)
catalog = load_catalog()

# Filtrar lo que necesitas
endes_2024 = filter_catalog(catalog, survey="endes", year_min=2024)

# Descargar
download_modules(endes_2024, dest="./data/", fmt="CSV", workers=4)

# Leer en DataFrames
dfs = read_module("./data/ENDES/2024/Unico/968-Modulo1629.zip")
for name, df in dfs.items():
    print(f"{name}: {df.shape}")
# RECH0: (37390, 44)
# RECH1: (135045, 36)
# RECH4: (135045, 22)
# RECHM: (3002, 8)

from inei_microdatos import search_variables, search_across_years

# Buscar variables
results = search_variables("ingreso")
for r in results[:5]:
    print(f'{r["variable"]:<20} {r["label"]:<50} {r["year"]}')

# Rastrear cambios a través de los años
by_year = search_across_years("p21", survey="enaho")
for year, matches in by_year.items():
    print(f'{year}: {matches[0]["label"][:60]}')

from inei_microdatos import load_catalog
from inei_microdatos.catalog import filter_catalog, catalog_stats, catalog_age

# Cargar catálogo incluido (sin configuración)
catalog = load_catalog()

# Verificar cuándo se generó
print(catalog_age())  # "2026-03-31T16:00:31+00:00"

# Estadísticas
print(catalog_stats(catalog))
# {'surveys': 67, 'survey_years': 295, 'modules': 5932, ...}

# Filtrar por nombre de encuesta (o alias), rango de años, período
enaho = filter_catalog(catalog, survey="enaho", year_min=2020, period="Anual")

from inei_microdatos import download_modules, download_docs

# Descargar con fallback de formato (CSV preferido, cae a STATA/SPSS si no hay)
result = download_modules(catalog, dest="./data/", fmt="CSV", workers=4)
# {'ok': 13, 'skipped': 0, 'failed': 0, 'bad_zip': 0}

# Formato estricto (sin fallback)
result = download_modules(catalog, dest="./data/", fmt="STATA", fallback=False)

# Vista previa sin descargar
result = download_modules(catalog, dest="./data/", fmt="CSV", dry_run=True)

# Documentación
result = download_docs(catalog, dest="./docs/", workers=4)

from inei_microdatos import read_module, read_catalog_entry, list_tables

# Desde un ZIP descargado
dfs = read_module("./data/968-Modulo1629.zip")

# Leer solo tablas específicas
dfs = read_module("./data/968-Modulo1629.zip", tables=["RECH0", "RECH1"])

# Desde un código de descarga (descarga a directorio temporal automáticamente)
dfs = read_module("968-Modulo1629")

# Directo desde catálogo (descarga + lee en un paso)
dfs = read_catalog_entry(catalog[0], year="2024", module="Hogar")

# Inspeccionar sin leer
tables = list_tables("./data/968-Modulo1629.zip")
# [{'name': 'RECH0', 'format': 'csv', 'size_bytes': 6598376, ...}, ...]

from inei_microdatos import search_variables, search_across_years

# Buscar por nombre o descripción
results = search_variables("ingreso")
# [{'survey': '...', 'year': '2024', 'variable': 'p21', 'label': '...', ...}, ...]

# Búsqueda exacta por nombre de variable
results = search_variables("p21", exact=True)

# Filtrar por encuesta
results = search_variables("pobreza", survey="enaho")

# Rastrear una variable a través de los años
by_year = search_across_years("ubigeo")
# {'2004': [...], '2005': [...], ..., '2024': [...]}

from inei_microdatos import INEIClient

client = INEIClient()
surveys = client.get_surveys()
years = client.get_years(surveys[0])
periods = client.get_periods(surveys[0], years[0])
modules = client.get_modules(surveys[0], years[0], periods[0])

print(modules[0].download_url("STATA"))
# https://proyectos.inei.gob.pe/iinei/srienaho/descarga/STATA/966-Modulo01.zip

from inei_microdatos import INEIClient
from inei_microdatos.catalog import build_catalog, save_catalog

client = INEIClient()
catalog = build_catalog(client)  # ~10 minutos
save_catalog(catalog, "~/.inei-microdatos/catalog.json")

from inei_microdatos import load_catalog, download_modules, read_module
from inei_microdatos.catalog import filter_catalog

# Load the bundled catalog (ships with the package, no setup needed)
catalog = load_catalog()

# Filter to what you need
endes_2024 = filter_catalog(catalog, survey="endes", year_min=2024)

# Download
download_modules(endes_2024, dest="./data/", fmt="CSV", workers=4)

# Read into DataFrames
dfs = read_module("./data/ENDES/2024/Unico/968-Modulo1629.zip")
for name, df in dfs.items():
    print(f"{name}: {df.shape}")
# RECH0: (37390, 44)
# RECH1: (135045, 36)
# RECH4: (135045, 22)
# RECHM: (3002, 8)

from inei_microdatos import search_variables, search_across_years

# Search variables
results = search_variables("ingreso")
for r in results[:5]:
    print(f'{r["variable"]:<20} {r["label"]:<50} {r["year"]}')

# Track changes across years
by_year = search_across_years("p21", survey="enaho")
for year, matches in by_year.items():
    print(f'{year}: {matches[0]["label"][:60]}')

from inei_microdatos import load_catalog
from inei_microdatos.catalog import filter_catalog, catalog_stats, catalog_age

# Load bundled catalog (zero setup)
catalog = load_catalog()

# Check when it was crawled
print(catalog_age())  # "2026-03-31T16:00:31+00:00"

# Stats
print(catalog_stats(catalog))
# {'surveys': 67, 'survey_years': 295, 'modules': 5932, ...}

# Filter by survey name (or alias), year range, period
enaho = filter_catalog(catalog, survey="enaho", year_min=2020, period="Anual")

from inei_microdatos import download_modules, download_docs

# Download with format fallback (CSV preferred, falls back to STATA/SPSS)
result = download_modules(catalog, dest="./data/", fmt="CSV", workers=4)
# {'ok': 13, 'skipped': 0, 'failed': 0, 'bad_zip': 0}

# Strict format (no fallback)
result = download_modules(catalog, dest="./data/", fmt="STATA", fallback=False)

# Dry run (preview without downloading)
result = download_modules(catalog, dest="./data/", fmt="CSV", dry_run=True)

# Documentation
result = download_docs(catalog, dest="./docs/", workers=4)

from inei_microdatos import read_module, read_catalog_entry, list_tables

# From a downloaded ZIP
dfs = read_module("./data/968-Modulo1629.zip")

# Read specific tables only
dfs = read_module("./data/968-Modulo1629.zip", tables=["RECH0", "RECH1"])

# From a download code (downloads to temp dir automatically)
dfs = read_module("968-Modulo1629")

# Directly from catalog (downloads + reads in one step)
dfs = read_catalog_entry(catalog[0], year="2024", module="Hogar")

# Inspect without reading
tables = list_tables("./data/968-Modulo1629.zip")
# [{'name': 'RECH0', 'format': 'csv', 'size_bytes': 6598376, ...}, ...]

from inei_microdatos import search_variables, search_across_years

# Search by name or description
results = search_variables("ingreso")
# [{'survey': '...', 'year': '2024', 'variable': 'p21', 'label': '...', ...}, ...]

# Exact match on variable name
results = search_variables("p21", exact=True)

# Filter by survey
results = search_variables("pobreza", survey="enaho")

# Track a variable across years
by_year = search_across_years("ubigeo")
# {'2004': [...], '2005': [...], ..., '2024': [...]}

from inei_microdatos import INEIClient

client = INEIClient()
surveys = client.get_surveys()
years = client.get_years(surveys[0])
periods = client.get_periods(surveys[0], years[0])
modules = client.get_modules(surveys[0], years[0], periods[0])

print(modules[0].download_url("STATA"))
# https://proyectos.inei.gob.pe/iinei/srienaho/descarga/STATA/966-Modulo01.zip

from inei_microdatos import INEIClient
from inei_microdatos.catalog import build_catalog, save_catalog

client = INEIClient()
catalog = build_catalog(client)  # ~10 minutes
save_catalog(catalog, "~/.inei-microdatos/catalog.json")