chore: auto-commit (7 archivos)

- python/functions/core/__init__.py - python/functions/pipelines/metabase_bulk_add_users_to_group.md - python/functions/pipelines/metabase_bulk_add_users_to_group.py - cpp/apps/ - python/functions/core/clean_email.md - python/functions/core/clean_email.py - python/functions/core/clean_email_test.py Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-06-01 11:39:08 +02:00
parent 876020addf
commit 7eef442444
8 changed files with 186 additions and 4 deletions
@@ -1,5 +1,6 @@
 """Core functional programming utilities."""

+from .clean_email import clean_email
 from .core import (
    all_of,
    any_of,
@@ -25,6 +26,7 @@ __all__ = [
    "all_of",
    "any_of",
    "chunk",
+    "clean_email",
    "compose",
    "drop",
    "filter_list",
@@ -0,0 +1,59 @@
+---
+name: clean_email
+kind: function
+lang: py
+domain: core
+version: "1.0.0"
+purity: pure
+signature: "def clean_email(raw: str) -> str"
+description: "Normaliza un email crudo que puede provenir de fuentes sucias (TSVs, CSVs, hojas de calculo). Elimina whitespace, angle brackets y comillas, convierte a minusculas y valida la forma basica. Lanza ValueError si el resultado no es un email plausible."
+tags: [email, sanitize, validation, core]
+uses_functions: []
+uses_types: []
+returns: []
+returns_optional: false
+error_type: ""
+imports: []
+tested: true
+tests:
+  - "angle brackets y whitespace se limpian"
+  - "comillas dobles y mayusculas se normalizan"
+  - "whitespace puro se limpia y lowercase"
+  - "email sin arroba lanza ValueError"
+  - "string vacio lanza ValueError"
+  - "email sin punto en dominio lanza ValueError"
+test_file_path: "python/functions/core/clean_email_test.py"
+file_path: "python/functions/core/clean_email.py"
+params:
+  - name: raw
+    desc: "String crudo con la direccion de email, potencialmente rodeado de whitespace, angle brackets (<foo@bar.com>) o comillas simples/dobles."
+output: "Email limpio en minusculas con la forma 'local@dominio.tld'. Lanza ValueError si no es un email plausible tras la limpieza."
+---
+
+## Ejemplo
+
+```python
+from core.clean_email import clean_email
+
+# Caso tipico de TSV/Outlook con angle brackets y whitespace
+clean_email(" <eaznarez@mutuamadmotor.com>")
+# -> "eaznarez@mutuamadmotor.com"
+
+# Comillas dobles y mayusculas desde spreadsheet
+clean_email('"BAR@BAZ.com"')
+# -> "bar@baz.com"
+
+# Caso invalido: lanza ValueError
+import pytest
+with pytest.raises(ValueError):
+    clean_email("not-an-email")
+```
+
+## Cuando usarla
+
+Antes de enviar una direccion de email a una API que valida estrictamente (ej. Metabase `POST /api/user`, SendGrid, etc.) cuando el origen del dato es un CSV, TSV, Excel, o campo pegado a mano. Usar esta funcion como paso de limpieza antes de cualquier llamada de creacion de usuarios o envio de correos.
+
+## Gotchas
+
+- NO valida registros MX ni que el dominio exista realmente — solo comprueba forma basica (un `@`, local part no vacia, al menos un `.` en el dominio).
+- NO decodifica quoted-printable ni dominios IDN/internacionales — solo limpieza de caracteres ASCII de envoltura.
@@ -0,0 +1,53 @@
+"""Normalize a raw email string from messy sources (TSVs, CSVs, spreadsheets)."""
+
+
+def clean_email(raw: str) -> str:
+    """Strip whitespace, angle brackets, surrounding quotes; lowercase; validate basic shape.
+
+    Args:
+        raw: Raw email string that may contain surrounding whitespace, angle brackets
+             (Outlook/RFC 5322 display-name form), or surrounding quotes.
+
+    Returns:
+        Cleaned, lowercased email address.
+
+    Raises:
+        ValueError: If the cleaned result is not a plausible email (must contain
+                    exactly one '@', non-empty local part, and at least one '.'
+                    in the domain part).
+    """
+    cleaned = raw.strip()
+
+    # Strip surrounding angle brackets: <foo@bar.com> -> foo@bar.com
+    if cleaned.startswith("<") and cleaned.endswith(">"):
+        cleaned = cleaned[1:-1].strip()
+
+    # Strip surrounding double quotes: "foo@bar.com" -> foo@bar.com
+    if len(cleaned) >= 2 and cleaned[0] == '"' and cleaned[-1] == '"':
+        cleaned = cleaned[1:-1].strip()
+    # Strip surrounding single quotes: 'foo@bar.com' -> foo@bar.com
+    elif len(cleaned) >= 2 and cleaned[0] == "'" and cleaned[-1] == "'":
+        cleaned = cleaned[1:-1].strip()
+
+    cleaned = cleaned.lower()
+
+    # Validate basic shape
+    at_count = cleaned.count("@")
+    if at_count != 1:
+        raise ValueError(
+            f"invalid email after cleaning: {raw!r} -> {cleaned!r}"
+        )
+
+    local, domain = cleaned.split("@", 1)
+
+    if not local:
+        raise ValueError(
+            f"invalid email after cleaning: {raw!r} -> {cleaned!r}"
+        )
+
+    if "." not in domain:
+        raise ValueError(
+            f"invalid email after cleaning: {raw!r} -> {cleaned!r}"
+        )
+
+    return cleaned
@@ -0,0 +1,43 @@
+"""Tests para clean_email."""
+import sys
+import os
+
+sys.path.insert(0, os.path.dirname(__file__))
+
+import pytest
+from clean_email import clean_email
+
+
+def test_angle_brackets_y_whitespace_se_limpian():
+    assert clean_email(" <foo@bar.com>") == "foo@bar.com"
+
+
+def test_comillas_dobles_y_mayusculas_se_normalizan():
+    assert clean_email('"BAR@BAZ.com"') == "bar@baz.com"
+
+
+def test_whitespace_puro_se_limpia_y_lowercase():
+    assert clean_email("  Mixed.Case@Example.COM  ") == "mixed.case@example.com"
+
+
+def test_email_sin_arroba_lanza_valueerror():
+    with pytest.raises(ValueError):
+        clean_email("not-an-email")
+
+
+def test_string_vacio_lanza_valueerror():
+    with pytest.raises(ValueError):
+        clean_email("")
+
+
+def test_email_sin_punto_en_dominio_lanza_valueerror():
+    with pytest.raises(ValueError):
+        clean_email("a@b")
+
+
+def test_comillas_simples_se_limpian():
+    assert clean_email("'user@domain.es'") == "user@domain.es"
+
+
+def test_angle_brackets_con_espacio_interior():
+    assert clean_email("< user@example.com >") == "user@example.com"