Files
plantillas-proyectos/backend/api/v1/modules/a76/layouts_csv/classes/tasks.py
2026-03-03 16:45:12 -07:00

529 lines
20 KiB
Python

"""
Tareas Celery para importación CSV de Clases de Materiales.
Flujo: scan_file (validación) → insert_valid_rows (commit).
"""
import os
import base64
import csv
import json
import logging
import re
import unicodedata
from typing import Dict, Any, Optional, List, Set
from core.celery_app import celery_app
from core.database import CoreSessionLocal
from core.paths import layout_path
from .template_config import row_from_template
logger = logging.getLogger(__name__)
CLS_IMPORT_FILE_PREFIX = "cls_import_file:"
CLS_IMPORT_META_PREFIX = "cls_import_meta:"
CLS_IMPORT_ERROR_LINES_PREFIX = "cls_import_error_lines:"
CLS_IMPORT_REDIS_TTL = 3600
def _get_redis():
import redis
url = os.getenv("VALKEY_URL", os.getenv("REDIS_URL", "redis://valkey:6379/0"))
return redis.Redis.from_url(url, decode_responses=False)
def _worker_upload_dir() -> str:
return layout_path("imports", "temp")
def _ensure_worker_has_file_from_redis(job_id: str) -> Optional[str]:
r = _get_redis()
data = r.get(f"{CLS_IMPORT_FILE_PREFIX}{job_id}")
if not data:
return None
try:
raw = base64.b64decode(data)
except Exception as e:
logger.warning(f"Classes import: failed to decode file from Redis: {e}")
return None
upload_dir = _worker_upload_dir()
os.makedirs(upload_dir, exist_ok=True)
file_path = os.path.join(upload_dir, f"cls_{job_id}.csv")
with open(file_path, "wb") as f:
f.write(raw)
return file_path
def _ensure_worker_has_meta_from_redis(job_id: str, file_path: str) -> bool:
r = _get_redis()
data = r.get(f"{CLS_IMPORT_META_PREFIX}{job_id}")
if not data:
return False
try:
meta = json.loads(data.decode("utf-8"))
except Exception as e:
logger.warning(f"Classes import: failed to decode meta from Redis: {e}")
return False
meta_path = file_path.replace(".csv", ".meta.json")
with open(meta_path, "w", encoding="utf-8") as f:
json.dump(meta, f)
return True
def _delete_import_from_redis(job_id: str) -> None:
try:
r = _get_redis()
r.delete(
f"{CLS_IMPORT_FILE_PREFIX}{job_id}",
f"{CLS_IMPORT_META_PREFIX}{job_id}",
f"{CLS_IMPORT_ERROR_LINES_PREFIX}{job_id}",
)
except Exception as e:
logger.warning(f"Classes import: failed to delete Redis keys: {e}")
def normalize_header(name: Optional[str]) -> str:
if not name:
return ""
name = unicodedata.normalize("NFKD", str(name)).upper()
name = "".join(ch for ch in name if not unicodedata.combining(ch))
name = re.sub(r"[^A-Z0-9]+", " ", name)
return re.sub(r"\s+", " ", name).strip()
def _validate_row_class(
row: Dict[str, Any],
line_num: int,
valid_material_keys: Optional[Set[str]] = None,
valid_uom_codes: Optional[Set[str]] = None,
) -> Optional[Dict[str, Any]]:
class_code = (row.get("CLASE") or "").strip()
if not class_code:
return {"line": line_num, "col": "CLASE", "msg": "Requerido"}
if len(class_code) > 8:
return {"line": line_num, "col": "CLASE", "msg": "Máximo 8 caracteres"}
desc_es = (row.get("DESCRIPCIONE") or "").strip()
if desc_es and len(desc_es) > 500:
return {"line": line_num, "col": "DESCRIPCIONE", "msg": "Máximo 500 caracteres"}
desc_en = (row.get("DESCRIPCIONI") or "").strip()
if desc_en and len(desc_en) > 500:
return {"line": line_num, "col": "DESCRIPCIONI", "msg": "Máximo 500 caracteres"}
material_key = (row.get("CLAVEMAT") or "").strip()
if material_key:
if len(material_key) > 10:
return {"line": line_num, "col": "CLAVEMAT", "msg": "Máximo 10 caracteres"}
if valid_material_keys is not None and material_key not in valid_material_keys:
return {"line": line_num, "col": "CLAVEMAT", "msg": "Tipo de material no existe"}
uom = (row.get("UNIMED") or "").strip()
if uom:
if len(uom) > 5:
return {"line": line_num, "col": "UNIMED", "msg": "Máximo 5 caracteres"}
if valid_uom_codes is not None and uom not in valid_uom_codes:
return {"line": line_num, "col": "UNIMED", "msg": "Unidad de medida no existe"}
fraction = (row.get("FRACCION") or "").strip()
if fraction and len(fraction) > 20:
return {"line": line_num, "col": "FRACCION", "msg": "Máximo 20 caracteres"}
us_fraction = (row.get("FRACCIONAME") or "").strip()
if us_fraction and len(us_fraction) > 16:
return {"line": line_num, "col": "FRACCIONAME", "msg": "Máximo 16 caracteres"}
sub_key = (row.get("CLAVESUB") or "").strip()
if sub_key and len(sub_key) > 5:
return {"line": line_num, "col": "CLAVESUB", "msg": "Máximo 5 caracteres"}
iva_exempt = (row.get("FRACCIONEXENTAIVA") or "").strip()
if iva_exempt and len(iva_exempt) > 4:
return {"line": line_num, "col": "FRACCIONEXENTAIVA", "msg": "Máximo 4 caracteres"}
rev_fisica = row.get("REVFISICA")
if rev_fisica is not None and rev_fisica != "":
try:
v = int(rev_fisica)
if v < -32768 or v > 32767:
return {"line": line_num, "col": "REVFISICA", "msg": "Valor fuera de rango"}
except (ValueError, TypeError):
return {"line": line_num, "col": "REVFISICA", "msg": "Debe ser número entero"}
return None
@celery_app.task(bind=True)
def scan_file(self, job_id: str, config: str = None):
logger.info(f"Classes import: starting scan for job {job_id}")
file_path = _ensure_worker_has_file_from_redis(job_id)
if not file_path:
return {"status": "failed", "error": "Archivo no encontrado (expirado o no subido). Sube de nuevo."}
_ensure_worker_has_meta_from_redis(job_id, file_path)
error_dir = layout_path("imports", "errors")
os.makedirs(error_dir, exist_ok=True)
error_path = os.path.join(error_dir, f"cls_{job_id}.jsonl")
total_rows = 0
try:
with open(file_path, "r", encoding="utf-8-sig") as f:
total_rows = sum(1 for _ in f) - 1
except Exception as e:
return {"status": "failed", "error": str(e)}
meta_path = file_path.replace(".csv", ".meta.json")
meta = {}
if os.path.exists(meta_path):
try:
with open(meta_path, "r", encoding="utf-8") as f:
meta = json.load(f) or {}
except Exception as e:
logger.warning(f"Classes import: failed to read meta: {e}")
tenant_id = meta.get("tenant_id")
company_id = meta.get("company_id")
if not tenant_id or not company_id:
return {"status": "failed", "error": "Falta contexto (tenant/company)"}
valid_material_keys: Set[str] = set()
valid_uom_codes: Set[str] = set()
try:
with CoreSessionLocal() as session:
from api.v1.modules.public.reference_data.material_types.models import MaterialType
from api.v1.modules.a76.general_catalogs.units_of_measure.models import UnitOfMeasure
for m in session.query(MaterialType.key).all():
valid_material_keys.add(m[0])
for u in (
session.query(UnitOfMeasure.code)
.filter(
UnitOfMeasure.tenant_id == tenant_id,
UnitOfMeasure.company_id == company_id,
)
.all()
):
valid_uom_codes.add(u[0])
except Exception as e:
logger.warning(f"Classes import: could not load FK sets: {e}")
error_count = 0
processed_rows = 0
errors_detail: List[Dict[str, Any]] = []
try:
with open(file_path, "r", encoding="utf-8-sig") as f_in, open(
error_path, "w", encoding="utf-8"
) as f_err:
sample = f_in.read(2048)
f_in.seek(0)
try:
dialect = csv.Sniffer().sniff(sample, delimiters=",;\t")
except Exception:
dialect = "excel"
reader = csv.DictReader(f_in, dialect=dialect)
for i, row in enumerate(reader, start=1):
if i % 500 == 0:
self.update_state(
state="PROGRESS",
meta={"current": i, "total": total_rows, "errors": error_count},
)
row_norm = row_from_template(row, normalize_header)
err = _validate_row_class(
row_norm, i,
valid_material_keys=valid_material_keys,
valid_uom_codes=valid_uom_codes,
)
if err:
error_count += 1
f_err.write(json.dumps(err) + "\n")
if len(errors_detail) < 500:
errors_detail.append(
{"line": err["line"], "col": err.get("col", ""), "msg": err.get("msg", "")}
)
processed_rows += 1
except Exception as e:
logger.error(f"Classes import scan failed: {e}")
return {"status": "failed", "error": str(e)}
error_lines_list = []
try:
if os.path.exists(error_path):
with open(error_path, "r", encoding="utf-8") as f:
for line in f:
try:
err = json.loads(line)
if "line" in err:
error_lines_list.append(err["line"])
except Exception:
pass
if error_lines_list:
r = _get_redis()
r.set(
f"{CLS_IMPORT_ERROR_LINES_PREFIX}{job_id}",
json.dumps(error_lines_list).encode("utf-8"),
ex=CLS_IMPORT_REDIS_TTL,
)
except Exception as e:
logger.warning(f"Classes import: failed to store error lines in Redis: {e}")
return {
"status": "waiting_confirmation",
"job_id": job_id,
"total_rows": processed_rows,
"error_count": error_count,
"valid_rows": processed_rows - error_count,
"errors": errors_detail,
}
def _str_or_none(val: Any, max_len: Optional[int] = None) -> Optional[str]:
if val is None:
return None
s = str(val).strip()
if not s:
return None
if max_len and len(s) > max_len:
return s[:max_len]
return s
def _int_or_none(val: Any) -> Optional[int]:
if val is None or val == "":
return None
try:
return int(val)
except (ValueError, TypeError):
return None
@celery_app.task(bind=True)
def insert_valid_rows(self, job_id: str):
logger.info(f"Classes import: starting commit for job {job_id}")
file_path = _ensure_worker_has_file_from_redis(job_id)
if not file_path:
alt_path = os.path.join(_worker_upload_dir(), f"cls_{job_id}.csv")
if not os.path.exists(alt_path):
return {
"status": "failed",
"error": "Archivo no encontrado (expirado). Sube y confirma de nuevo.",
}
file_path = alt_path
else:
_ensure_worker_has_meta_from_redis(job_id, file_path)
error_dir = layout_path("imports", "errors")
error_path = os.path.join(error_dir, f"cls_{job_id}.jsonl")
error_lines = set()
try:
r = _get_redis()
raw = r.get(f"{CLS_IMPORT_ERROR_LINES_PREFIX}{job_id}")
if raw:
error_lines = set(json.loads(raw.decode("utf-8")))
except Exception as e:
logger.debug(f"Classes import: could not load error lines from Redis: {e}")
if not error_lines and os.path.exists(error_path):
with open(error_path, "r", encoding="utf-8") as f:
for line in f:
try:
err = json.loads(line)
error_lines.add(err["line"])
except Exception:
pass
meta_path = file_path.replace(".csv", ".meta.json")
tenant_id = None
company_id = None
meta = {}
if os.path.exists(meta_path):
try:
with open(meta_path, "r", encoding="utf-8") as f:
meta = json.load(f) or {}
tenant_id = meta.get("tenant_id")
company_id = meta.get("company_id")
except Exception:
pass
if not tenant_id or not company_id:
return {"status": "failed", "error": "Falta contexto (tenant/company)"}
from api.v1.modules.a76.classes.models import Class
valid_material_keys: Set[str] = set()
valid_uom_codes: Set[str] = set()
try:
with CoreSessionLocal() as session:
from api.v1.modules.public.reference_data.material_types.models import MaterialType
from api.v1.modules.a76.general_catalogs.units_of_measure.models import UnitOfMeasure
for m in session.query(MaterialType.key).all():
valid_material_keys.add(m[0])
for u in (
session.query(UnitOfMeasure.code)
.filter(
UnitOfMeasure.tenant_id == tenant_id,
UnitOfMeasure.company_id == company_id,
)
.all()
):
valid_uom_codes.add(u[0])
except Exception as e:
logger.warning(f"Classes import: could not load FK sets: {e}")
inserted_count = 0
skipped_invalid = 0
skipped_details: List[Dict[str, Any]] = []
response = None
try:
with CoreSessionLocal() as session:
existing_by_code: Dict[str, Class] = {}
for c in (
session.query(Class)
.filter(
Class.tenant_id == tenant_id,
Class.company_id == company_id,
)
.all()
):
existing_by_code[c.class_code] = c
with open(file_path, "r", encoding="utf-8-sig") as f:
sample = f.read(2048)
f.seek(0)
try:
dialect = csv.Sniffer().sniff(sample, delimiters=",;\t")
except Exception:
dialect = "excel"
reader = csv.DictReader(f, dialect=dialect)
for i, row in enumerate(reader, start=1):
if i in error_lines:
continue
row_norm = row_from_template(row, normalize_header)
err = _validate_row_class(
row_norm, i,
valid_material_keys=valid_material_keys,
valid_uom_codes=valid_uom_codes,
)
if err:
skipped_invalid += 1
skipped_details.append(
{"line": i, "reason": f"{err.get('col', '')}: {err.get('msg', '')}"}
)
continue
class_code = _str_or_none(row_norm.get("CLASE"), 8)
if not class_code:
skipped_invalid += 1
continue
existing = existing_by_code.get(class_code)
desc_es = _str_or_none(row_norm.get("DESCRIPCIONE"), 500)
desc_en = _str_or_none(row_norm.get("DESCRIPCIONI"), 500)
material_key = _str_or_none(row_norm.get("CLAVEMAT"), 10)
if material_key and material_key not in valid_material_keys:
material_key = None
unit_of_measure = _str_or_none(row_norm.get("UNIMED"), 5)
if unit_of_measure and unit_of_measure not in valid_uom_codes:
unit_of_measure = None
fraction = _str_or_none(row_norm.get("FRACCION"), 20)
us_fraction = _str_or_none(row_norm.get("FRACCIONAME"), 16)
sub_key = _str_or_none(row_norm.get("CLAVESUB"), 5)
physical_review = _int_or_none(row_norm.get("REVFISICA"))
iva_exempt_fraction = _str_or_none(row_norm.get("FRACCIONEXENTAIVA"), 4)
if existing:
existing.description_es = desc_es
existing.description_en = desc_en
existing.material_key = material_key
existing.unit_of_measure = unit_of_measure
existing.fraction = fraction
existing.us_fraction = us_fraction
existing.sub_key = sub_key
existing.physical_review = physical_review
existing.iva_exempt_fraction = iva_exempt_fraction
session.add(existing)
inserted_count += 1
else:
new_class = Class(
tenant_id=tenant_id,
company_id=company_id,
class_code=class_code,
description_es=desc_es,
description_en=desc_en,
material_key=material_key,
unit_of_measure=unit_of_measure,
fraction=fraction,
us_fraction=us_fraction,
sub_key=sub_key,
physical_review=physical_review,
iva_exempt_fraction=iva_exempt_fraction,
)
session.add(new_class)
existing_by_code[class_code] = new_class
inserted_count += 1
try:
session.commit()
except Exception as db_err:
session.rollback()
logger.error(f"Classes import DB error: {db_err}")
return {"status": "failed", "error": str(db_err)}
total_skipped = skipped_invalid
if inserted_count == 0 and total_skipped > 0:
response = {
"status": "warning",
"inserted": 0,
"skipped_invalid": skipped_invalid,
"skipped_missing_fk": 0,
"skipped_details": skipped_details,
"message": f"No se insertaron registros. {total_skipped} rechazados.",
}
elif inserted_count == 0:
response = {
"status": "failed",
"error": "No hay registros válidos en el archivo CSV",
"inserted": 0,
"skipped_invalid": skipped_invalid,
"skipped_missing_fk": 0,
"skipped_details": skipped_details,
}
else:
response = {
"status": "finished",
"inserted": inserted_count,
"skipped_invalid": skipped_invalid,
"skipped_missing_fk": 0,
"skipped_details": skipped_details,
}
except Exception as e:
logger.error(f"Classes import task failed: {e}")
import traceback
logger.error(traceback.format_exc())
return {"status": "failed", "error": str(e)}
try:
if file_path and os.path.exists(file_path):
os.remove(file_path)
if os.path.exists(error_path):
os.remove(error_path)
if os.path.exists(meta_path):
os.remove(meta_path)
_delete_import_from_redis(job_id)
except Exception as cleanup_err:
logger.warning(f"Classes import cleanup failed: {cleanup_err}")
if response is None:
response = {
"status": "failed",
"error": "Error inesperado",
"inserted": 0,
"skipped_invalid": skipped_invalid,
"skipped_missing_fk": 0,
"skipped_details": skipped_details,
}
return response