Compare commits

..

1 Commits

Author SHA1 Message Date
egutierrez a69d14d38e feat(eda): capítulo TIMESERIES del AutomaticEDA (evolución + análisis de serie)
Capítulo nuevo build_timeseries(profile, ctx) -> Chapter|None del motor
AutomaticEDA. Cuando la tabla tiene columna de fecha/datetime, grafica la
evolución de cada columna numérica por periodo (valor agregado + conteo de filas)
y los paneles de descomposición STL y autocorrelación (ACF), con el análisis de
la serie: estacionariedad (ADF+KPSS), autocorrelación (Ljung-Box), fuerzas de
tendencia/estacionalidad (Hyndman) y la transformación sugerida (retornos o
diferencias) para evitar correlaciones espurias. Sin columna temporal devuelve
None. Consolida series OHLC casi idénticas en un único gráfico conservando el
análisis de cada columna.

La serie cruda llega por ctx['timeseries_raw'] (mismo patrón que modelos con
raw_numeric); las figuras son perezosas (Figure.make) y el paginador del núcleo
garantiza no-corte en PDF y PPTX. CHAPTER_VERSION 1.0.0.

Cubre los MUST del diseño (report 2043): MUST-9.1 (línea valor-vs-tiempo + conteo
por periodo), MUST-9.2 (paneles STL + ACF), MUST-9.3 (perfil datetime +
consolidación OHLC).

Funciones nuevas del registry (grupo eda), delegadas a fn-constructor, no inline:
- detect_time_column (pure): detecta la columna temporal y las numéricas
- profile_datetime (pure): rango/frecuencia/regularidad/huecos de la fecha
- resample_timeseries (pure): agrega la serie por periodo + conteo
- extract_timeseries_raw (impure): lee la serie cruda ordenada de DuckDB/PG

Verificación: 69 tests verdes (capítulo 9 + funciones 28 + núcleo/renderers);
golden real sobre seattle-weather (estacional) y aapl (OHLC) con PDF+PPTX sin
cortar nada (cols_cortadas=[]).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-30 15:35:42 +02:00
29 changed files with 2324 additions and 2994 deletions
+8 -6
View File
@@ -25,7 +25,6 @@ from .describe_numeric import describe_numeric
from .summarize_categorical import summarize_categorical
from .infer_semantic_type import infer_semantic_type
from .column_quality_score import column_quality_score
from .select_groupby_keys import select_groupby_keys
from .render_eda_markdown import render_eda_markdown
from .detect_distribution_type import detect_distribution_type
from .spearman_corr import spearman_corr
@@ -37,8 +36,6 @@ from .infer_fk_containment_duckdb import infer_fk_containment_duckdb
from .build_join_graph import build_join_graph
from .association_matrix import association_matrix
from .correlation_matrix_duckdb import correlation_matrix_duckdb
from .pivot_table_duckdb import pivot_table_duckdb
from .groupby_stats_duckdb import groupby_stats_duckdb
from .pca_explained import pca_explained
from .kmeans_segments import kmeans_segments
from .isolation_forest_outliers import isolation_forest_outliers
@@ -60,8 +57,16 @@ from .exploratory_caveats import exploratory_caveats
from .render_eda_pdf import render_eda_pdf, render_eda_pdf_relational
from .render_automatic_eda_pdf import render_automatic_eda_pdf
from .render_automatic_eda_pptx import render_automatic_eda_pptx
from .detect_time_column import detect_time_column
from .extract_timeseries_raw import extract_timeseries_raw
from .profile_datetime import profile_datetime
from .resample_timeseries import resample_timeseries
__all__ = [
"detect_time_column",
"extract_timeseries_raw",
"profile_datetime",
"resample_timeseries",
"render_automatic_eda_pdf",
"render_automatic_eda_pptx",
"decode_qr_image",
@@ -85,8 +90,6 @@ __all__ = [
"build_join_graph",
"association_matrix",
"correlation_matrix_duckdb",
"pivot_table_duckdb",
"groupby_stats_duckdb",
"pca_explained",
"kmeans_segments",
"isolation_forest_outliers",
@@ -101,7 +104,6 @@ __all__ = [
"summarize_categorical",
"infer_semantic_type",
"column_quality_score",
"select_groupby_keys",
"render_eda_markdown",
"detect_distribution_type",
"pull_gsc_search_analytics",
@@ -1,592 +0,0 @@
"""Aggregation chapter (AGREGACION) — group analysis / OLAP of the EDA.
This chapter is the group-by / pivot ("OLAP") section of an AutomaticEDA report
and is meant to be present **whenever the dataset has at least one low-cardinality
categorical column to group by**. For the most interesting categoricals (chosen
by their cardinality/relevance, optionally with an LLM) it renders, as blocks the
core paginator never cuts:
1. **Per-group statistics** (split-apply-combine) — for each interesting
categorical key, the count of rows per group and, for each numeric measure,
its mean/median/std/min/max. One compact summary table (mean of every measure
per group) plus a per-measure detail table.
2. **Bar charts** — a vertical bar chart of a measure's mean per group, bars from
zero (Tufte Lie-Factor = 1).
3. **Pivot tables** — categorical A x categorical B -> aggregate of a measure,
limited to the top rows/cols so it fits a mobile page/slide, with a grouped
bar chart of the same pivot.
The raw data needed to aggregate is **not** in the TableProfile, so — exactly
like ``modelos`` reads its cluster projection from ``ctx`` — this chapter gets
the aggregation results in one of two ways and degrades honestly when neither is
available:
ctx keys this chapter consumes (all optional):
aggregations : dict — pre-computed results, used directly (offline / tests /
forward-compatible with a calculation phase). Shape::
{"groupby": [{"group_by": str, "measures": [str], "why": str,
"result": <groupby_stats_duckdb-shaped dict>}],
"pivots": [{"index": str, "columns": str, "value": str, "agg": str,
"why": str, "result": <pivot_table_duckdb-shaped dict>}]}
db_path, table : str — when ``aggregations`` is absent, the chapter selects
the interesting keys (``select_groupby_keys``), optionally asks an LLM
which to show (``suggest_aggregations_llm`` when ``run_agg_llm`` is True)
and computes the group-by/pivot results live via the push-down registry
functions ``groupby_stats_duckdb`` / ``pivot_table_duckdb``.
run_agg_llm : bool — when True (and ``db_path``/``table`` present), let the
LLM pick the interesting aggregations; otherwise the deterministic
quantitative selection is used.
agg_llm_model : str — model id for the optional LLM selection.
agg_max_keys, agg_max_card, agg_max_measures, agg_top_n : int — limits.
agg_insights : list — optional pre-computed micro-analysis entries
(``[{"title": str, "text": str}]``) rendered as an interpretation section.
Contract: build_<id>(profile, ctx) -> Chapter | None ; CHAPTER_VERSION = "x.y.z".
Reads everything defensively (``.get``) and never raises: anything missing
degrades to a note instead of aborting the chapter; the chapter returns ``None``
only when the dataset has no categorical column to group by.
"""
from __future__ import annotations
from .. import model
# Pure/impure registry functions (group ``eda``) this chapter composes. Imported
# defensively so the chapter still builds (degrading the affected part to a note)
# if a function is somehow unavailable / not indexed yet.
try:
from datascience.select_groupby_keys import select_groupby_keys
except Exception: # noqa: BLE001 — keep the chapter importable no matter what.
select_groupby_keys = None # type: ignore[assignment]
try:
from datascience.groupby_stats_duckdb import groupby_stats_duckdb
except Exception: # noqa: BLE001
groupby_stats_duckdb = None # type: ignore[assignment]
try:
from datascience.pivot_table_duckdb import pivot_table_duckdb
except Exception: # noqa: BLE001
pivot_table_duckdb = None # type: ignore[assignment]
try:
from datascience.suggest_aggregations_llm import suggest_aggregations_llm
except Exception: # noqa: BLE001
suggest_aggregations_llm = None # type: ignore[assignment]
CHAPTER_VERSION = "1.0.0"
CHAPTER_ID = "agregacion"
CHAPTER_TITLE = "Agregación por grupos"
# Tableau-10 palette — stable colours for the pivot's grouped-bar series.
_SERIES_COLORS = [
"#4e79a7", "#f28e2b", "#e15759", "#76b7b2", "#59a14f",
"#edc948", "#b07aa1", "#ff9da7", "#9c755f", "#bab0ac",
]
# Defaults for the live selection/aggregation (overridable via ctx).
_DEF_MAX_KEYS = 3
_DEF_MAX_CARD = 20
_DEF_MAX_MEASURES = 4
_DEF_TOP_N = 12
# --------------------------------------------------------------------------- #
# Formatting helpers (mirror the other chapters' defensive style).
# --------------------------------------------------------------------------- #
def _fmt_num(value, decimals: int = 3) -> str:
if value is None:
return ""
if isinstance(value, bool):
return "" if value else "no"
if isinstance(value, int):
return f"{value:,}".replace(",", ".")
if isinstance(value, float):
if value != value: # NaN
return "NaN"
if value in (float("inf"), float("-inf")):
return str(value)
text = f"{value:.{decimals}f}".rstrip("0").rstrip(".")
return text if text else "0"
return model._safe_str(value)
def _is_dict(v) -> bool:
return isinstance(v, dict)
def _measure_mean(group: dict, measure: str):
"""Pull the mean of one measure out of a groupby-result group entry."""
stats = group.get("stats") if _is_dict(group.get("stats")) else {}
ms = stats.get(measure) if _is_dict(stats.get(measure)) else {}
return ms.get("mean")
# --------------------------------------------------------------------------- #
# Plan + data resolution. Either a pre-computed ctx['aggregations'] is used
# verbatim, or the plan is selected and the results are computed live.
# --------------------------------------------------------------------------- #
def _resolve_candidates(profile: dict, ctx: dict) -> dict:
"""Return {group_keys, measures, pivots, note} of interesting columns."""
pre = ctx.get("agg_candidates")
if _is_dict(pre) and pre.get("group_keys") is not None:
return pre
if select_groupby_keys is not None:
try:
out = select_groupby_keys(
profile,
max_keys=int(ctx.get("agg_max_keys", _DEF_MAX_KEYS)),
max_card=int(ctx.get("agg_max_card", _DEF_MAX_CARD)),
max_measures=int(ctx.get("agg_max_measures", _DEF_MAX_MEASURES)),
)
if _is_dict(out):
return out
except Exception: # noqa: BLE001 — fall through to the inline fallback.
pass
return _inline_candidates(profile, ctx)
def _inline_candidates(profile: dict, ctx: dict) -> dict:
"""Minimal defensive selection when select_groupby_keys is unavailable."""
max_card = int(ctx.get("agg_max_card", _DEF_MAX_CARD))
max_keys = int(ctx.get("agg_max_keys", _DEF_MAX_KEYS))
max_measures = int(ctx.get("agg_max_measures", _DEF_MAX_MEASURES))
keys = profile.get("key_candidates") or []
group_keys, measures = [], []
for col in profile.get("columns") or []:
if not _is_dict(col):
continue
name = col.get("name")
it = col.get("inferred_type")
flags = col.get("flags") or []
dc = col.get("distinct_count")
if it in ("categorical", "boolean") and name not in keys:
if ("possible_id" not in flags and "high_cardinality" not in flags
and "constant" not in flags
and isinstance(dc, int) and 2 <= dc <= max_card):
group_keys.append({"col": name, "cardinality": dc, "score": 0.0})
elif it == "numeric":
num = col.get("numeric") or {}
if num.get("std") not in (None, 0) and not (
"possible_id" in flags and (col.get("unique_pct") or 0) >= 0.99):
measures.append(name)
group_keys = group_keys[:max_keys]
measures = measures[:max_measures]
pivots = []
if len(group_keys) >= 2:
pivots.append({"index": group_keys[0]["col"],
"columns": group_keys[1]["col"],
"value": measures[0] if measures else None})
return {"group_keys": group_keys, "measures": measures, "pivots": pivots,
"note": "selección cuantitativa básica"}
def _resolve_plan(profile: dict, ctx: dict, candidates: dict) -> dict:
"""Return {aggregations:[{group_by,measures,why}], pivots:[...], source}."""
group_keys = candidates.get("group_keys") or []
measures = candidates.get("measures") or []
if ctx.get("run_agg_llm") and suggest_aggregations_llm is not None:
try:
plan = suggest_aggregations_llm(
profile, candidates,
max_aggs=int(ctx.get("agg_max_keys", _DEF_MAX_KEYS)),
model=ctx.get("agg_llm_model", "claude-haiku-4-5-20251001"))
if _is_dict(plan) and plan.get("aggregations"):
return {"aggregations": plan.get("aggregations") or [],
"pivots": plan.get("pivots") or [],
"source": plan.get("source", "llm")}
except Exception: # noqa: BLE001 — fall back to the quantitative plan.
pass
aggregations = [{
"group_by": gk.get("col"),
"measures": measures,
"why": f"categórica de {_fmt_num(gk.get('cardinality'))} niveles",
} for gk in group_keys if _is_dict(gk) and gk.get("col")]
pivots = []
for pv in candidates.get("pivots") or []:
if _is_dict(pv) and pv.get("index") and pv.get("columns"):
pivots.append({"index": pv.get("index"), "columns": pv.get("columns"),
"value": pv.get("value") or (measures[0] if measures else None),
"agg": "mean", "why": "cruce de dos categóricas"})
return {"aggregations": aggregations, "pivots": pivots, "source": "quantitative"}
def _live_groupby(ctx: dict, group_by: str, measures: list, top_n: int):
"""Compute one group-by result live via the push-down registry function."""
db_path = ctx.get("db_path")
table = ctx.get("table")
if not db_path or not table or groupby_stats_duckdb is None:
return None
try:
out = groupby_stats_duckdb(db_path, table, group_by, list(measures or []),
top_n=top_n)
if _is_dict(out) and out.get("status") == "ok":
return out
except Exception: # noqa: BLE001
return None
return None
def _live_pivot(ctx: dict, index: str, columns: str, value, agg: str):
"""Compute one pivot live via the push-down registry function."""
db_path = ctx.get("db_path")
table = ctx.get("table")
if not db_path or not table or pivot_table_duckdb is None or not value:
return None
try:
out = pivot_table_duckdb(db_path, table, index, columns, value,
agg=agg or "mean")
if _is_dict(out) and out.get("status") == "ok":
return out
except Exception: # noqa: BLE001
return None
return None
# --------------------------------------------------------------------------- #
# Figure builders (lazy: matplotlib only imported when the renderer draws them).
# --------------------------------------------------------------------------- #
def _make_group_bars(group_by: str, measure: str, groups: list):
"""Vertical bars: mean of ``measure`` per group, bars from zero."""
labels, values = [], []
for g in groups:
if not _is_dict(g):
continue
mean = _measure_mean(g, measure)
if mean is None:
continue
labels.append(model._safe_str(g.get("key")))
values.append(float(mean))
if not labels:
return None
def _draw():
import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt
fig, ax = plt.subplots(figsize=(6.6, 3.6))
xs = list(range(len(labels)))
ax.bar(xs, values, color="#4e79a7", alpha=0.9, edgecolor="#2f4d6e",
linewidth=0.4)
ax.set_xticks(xs)
short = [(s[:18] + "") if len(s) > 19 else s for s in labels]
rot = 30 if max((len(s) for s in short), default=0) > 6 else 0
ax.set_xticklabels(short, rotation=rot, ha="right" if rot else "center",
fontsize=7)
ax.set_ylabel(f"media de {measure}", fontsize=8)
ax.set_xlabel(group_by, fontsize=8)
ax.set_title(f"Media de «{measure}» por «{group_by}»", fontsize=10)
ax.grid(axis="y", color="#dddddd", linewidth=0.6)
for spine in ("top", "right"):
ax.spines[spine].set_visible(False)
# Value labels above each bar.
vmax = max(values) if values else 0
for x, v in zip(xs, values):
ax.text(x, v + (abs(vmax) * 0.01 if vmax else 0.01),
_fmt_num(v, 2), ha="center", va="bottom", fontsize=6.5)
fig.tight_layout()
return fig
return _draw
def _make_pivot_bars(pivot: dict):
"""Grouped bars of a pivot: x = row_labels, one series per col_label."""
row_labels = pivot.get("row_labels") or []
col_labels = pivot.get("col_labels") or []
matrix = pivot.get("matrix") or []
if not row_labels or not col_labels or not matrix:
return None
def _draw():
import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt
n_rows = len(row_labels)
n_cols = len(col_labels)
fig, ax = plt.subplots(figsize=(6.8, 3.8))
total_w = 0.8
bar_w = total_w / max(n_cols, 1)
base = list(range(n_rows))
for j, clabel in enumerate(col_labels):
offs = [b - total_w / 2 + bar_w * (j + 0.5) for b in base]
vals = []
for i in range(n_rows):
cell = matrix[i][j] if (i < len(matrix) and j < len(matrix[i])) else None
vals.append(float(cell) if isinstance(cell, (int, float)) else 0.0)
color = _SERIES_COLORS[j % len(_SERIES_COLORS)]
ax.bar(offs, vals, width=bar_w, color=color, alpha=0.9,
label=model._safe_str(clabel))
ax.set_xticks(base)
short = [(s[:16] + "") if len(s) > 17 else s
for s in (model._safe_str(r) for r in row_labels)]
rot = 30 if max((len(s) for s in short), default=0) > 6 else 0
ax.set_xticklabels(short, rotation=rot, ha="right" if rot else "center",
fontsize=7)
ax.set_xlabel(model._safe_str(pivot.get("index")), fontsize=8)
ax.set_ylabel(f"{pivot.get('agg','mean')} de {pivot.get('value')}",
fontsize=8)
ax.set_title(f"{pivot.get('index')} × {pivot.get('columns')}", fontsize=10)
ax.grid(axis="y", color="#dddddd", linewidth=0.6)
ax.legend(title=model._safe_str(pivot.get("columns")), fontsize=6.5,
title_fontsize=7, frameon=True, framealpha=0.9, loc="best")
for spine in ("top", "right"):
ax.spines[spine].set_visible(False)
fig.tight_layout()
return fig
return _draw
def _group_bars_maker(group_by: str, measure: str, groups: list):
"""Bind per-aggregation args so the lazy closure is loop-safe."""
def _make():
return _make_group_bars(group_by, measure, groups)()
return _make
def _pivot_bars_maker(pivot: dict):
def _make():
return _make_pivot_bars(pivot)()
return _make
# --------------------------------------------------------------------------- #
# Section builders. Each returns a list of blocks (possibly empty).
# --------------------------------------------------------------------------- #
def _groupby_section(group_by: str, measures: list, result: dict, why: str) -> list:
"""Build the blocks for one group-by aggregation, or [] if unusable."""
if not _is_dict(result) or not result.get("groups"):
return []
groups = [g for g in result.get("groups") or [] if _is_dict(g)]
if not groups:
return []
eff_measures = result.get("measures") or measures or []
blocks = [model.Heading(text=f"Agrupado por «{group_by}»", level=2)]
intro = f"**{why}.** " if why else ""
intro += (f"{_fmt_num(result.get('n_groups') or len(groups))} grupos"
f"{' (top por tamaño)' if result.get('truncated') else ''}.")
blocks.append(model.Markdown(text=intro))
# Summary table: one row per group, count + mean of every measure.
header = ["Grupo", "n"] + [f"{m} (media)" for m in eff_measures]
rows = []
for g in groups:
row = [model._safe_str(g.get("key")), _fmt_num(g.get("n"))]
for m in eff_measures:
row.append(_fmt_num(_measure_mean(g, m), 2))
rows.append(row)
blocks.append(model.DataTable(
header=header, rows=rows, title=f"Resumen por «{group_by}»",
note="Conteo de filas y media de cada medida por grupo."))
if not eff_measures:
return blocks
# Primary measure: a bar chart + a detail table (mean/median/std/min/max).
primary = eff_measures[0]
bars = _make_group_bars(group_by, primary, groups)
if bars is not None:
blocks.append(model.Figure(
make=_group_bars_maker(group_by, primary, groups),
caption=f"Media de «{primary}» por «{group_by}» (barras desde cero)."))
det_header = ["Grupo", "n", "media", "mediana", "σ", "mín", "máx"]
det_rows = []
for g in groups:
stats = g.get("stats") if _is_dict(g.get("stats")) else {}
ms = stats.get(primary) if _is_dict(stats.get(primary)) else {}
det_rows.append([
model._safe_str(g.get("key")), _fmt_num(g.get("n")),
_fmt_num(ms.get("mean"), 2), _fmt_num(ms.get("median"), 2),
_fmt_num(ms.get("std"), 2), _fmt_num(ms.get("min"), 2),
_fmt_num(ms.get("max"), 2),
])
blocks.append(model.DataTable(
header=det_header, rows=det_rows,
title=f"Detalle de «{primary}» por «{group_by}»"))
return blocks
def _pivot_section(pivot_spec: dict, result: dict) -> list:
"""Build the blocks for one pivot table, or [] if unusable."""
if not _is_dict(result) or not result.get("row_labels"):
return []
row_labels = result.get("row_labels") or []
col_labels = result.get("col_labels") or []
matrix = result.get("matrix") or []
if not row_labels or not col_labels or not matrix:
return []
index = result.get("index") or pivot_spec.get("index")
columns = result.get("columns") or pivot_spec.get("columns")
value = result.get("value") or pivot_spec.get("value")
agg = result.get("agg") or pivot_spec.get("agg") or "mean"
why = pivot_spec.get("why") or ""
blocks = [model.Heading(text=f"Pivot: «{index}» × «{columns}»", level=2)]
intro = f"**{why}.** " if why else ""
intro += (f"{agg} de «{value}» cruzando «{index}» (filas) y «{columns}» "
f"(columnas).")
if result.get("truncated_rows") or result.get("truncated_cols"):
intro += " Limitado a las filas/columnas más frecuentes."
blocks.append(model.Markdown(text=intro))
header = [model._safe_str(index)] + [model._safe_str(c) for c in col_labels]
rows = []
for i, rlabel in enumerate(row_labels):
row = [model._safe_str(rlabel)]
cells = matrix[i] if i < len(matrix) else []
for j in range(len(col_labels)):
cell = cells[j] if j < len(cells) else None
row.append(_fmt_num(cell, 2))
rows.append(row)
blocks.append(model.DataTable(
header=header, rows=rows,
title=f"{agg} de «{value}»",
note=f"Cada celda es {agg} de «{value}» para esa combinación."))
fig_pivot = {"row_labels": row_labels, "col_labels": col_labels,
"matrix": matrix, "index": index, "columns": columns,
"value": value, "agg": agg}
if _make_pivot_bars(fig_pivot) is not None:
blocks.append(model.Figure(
make=_pivot_bars_maker(fig_pivot),
caption=f"{agg} de «{value}» por «{index}» y «{columns}» "
f"(barras agrupadas)."))
return blocks
def _insights_section(ctx: dict) -> list:
"""Optional pre-computed micro-analysis of the aggregations (SHOULD-11.4)."""
entries = ctx.get("agg_insights")
if not isinstance(entries, list) or not entries:
return []
blocks = [model.Heading(text="Interpretación de los grupos", level=2)]
for e in entries:
if not _is_dict(e):
continue
title = model._safe_str(e.get("title"))
text = model._safe_str(e.get("text"))
line = (f"**{title}.** " if title else "") + text
if line.strip():
blocks.append(model.Markdown(text=line))
return blocks if len(blocks) > 1 else []
# --------------------------------------------------------------------------- #
# Pre-computed path: ctx['aggregations'] already carries the results.
# --------------------------------------------------------------------------- #
def _sections_from_precomputed(agg: dict) -> list:
sections = []
for entry in agg.get("groupby") or []:
if not _is_dict(entry):
continue
sections += _groupby_section(
entry.get("group_by"), entry.get("measures") or [],
entry.get("result") or {}, entry.get("why") or "")
for entry in agg.get("pivots") or []:
if not _is_dict(entry):
continue
sections += _pivot_section(entry, entry.get("result") or {})
return sections
# --------------------------------------------------------------------------- #
# Live path: select keys, pick a plan, compute results via push-down functions.
# --------------------------------------------------------------------------- #
def _sections_live(profile: dict, ctx: dict, candidates: dict) -> list:
top_n = int(ctx.get("agg_top_n", _DEF_TOP_N))
plan = _resolve_plan(profile, ctx, candidates)
sections = []
for agg in plan.get("aggregations") or []:
if not _is_dict(agg) or not agg.get("group_by"):
continue
result = _live_groupby(ctx, agg.get("group_by"),
agg.get("measures") or [], top_n)
if result is not None:
sections += _groupby_section(agg.get("group_by"),
agg.get("measures") or [], result,
agg.get("why") or "")
for pv in plan.get("pivots") or []:
if not _is_dict(pv) or not pv.get("index") or not pv.get("columns"):
continue
result = _live_pivot(ctx, pv.get("index"), pv.get("columns"),
pv.get("value"), pv.get("agg") or "mean")
if result is not None:
sections += _pivot_section(pv, result)
return sections
# --------------------------------------------------------------------------- #
# Entry point.
# --------------------------------------------------------------------------- #
def _intro_blocks() -> list:
text = (
"Este capítulo analiza la tabla **por grupos** (split-apply-combine): "
"elige las columnas categóricas más informativas — por su cardinalidad "
"y relevancia, no todas contra todas, para no inflar comparaciones "
"espurias — y resume las variables numéricas dentro de cada grupo "
"(conteo, media, mediana, desviación). Las **tablas dinámicas** (pivot) "
"cruzan dos categóricas sobre una medida, y los **gráficos de barras** "
"(siempre desde cero) comparan los grupos de un vistazo."
)
return [model.Heading(text=CHAPTER_TITLE, level=1),
model.Markdown(text=text)]
def build_agregacion(profile: dict, ctx: dict):
"""Build the AGREGACION Chapter, or None if the dataset can't be grouped.
Args:
profile: the ``eda`` group TableProfile dict.
ctx: presentation context (see module docstring for the keys consumed).
Returns:
A ``model.Chapter`` with per-group stats, pivots and bar charts; or
``None`` when the dataset has no low-cardinality categorical column to
group by (the chapter does not apply).
"""
profile = profile or {}
ctx = ctx or {}
if not isinstance(profile, dict):
return None
# Pre-computed results take precedence (offline / tests / forward-compat).
pre = ctx.get("aggregations")
if _is_dict(pre) and (pre.get("groupby") or pre.get("pivots")):
sections = _sections_from_precomputed(pre)
if not sections:
return None
blocks = _intro_blocks() + sections + _insights_section(ctx)
return model.Chapter(id=CHAPTER_ID, title=CHAPTER_TITLE,
version=CHAPTER_VERSION, blocks=blocks)
# Live path: needs at least one categorical key to group by.
candidates = _resolve_candidates(profile, ctx)
if not _is_dict(candidates) or not (candidates.get("group_keys")):
return None # chapter does not apply: nothing to group by.
sections = _sections_live(profile, ctx, candidates)
if not sections:
# Applies (there are categorical keys) but no aggregation data is
# reachable: emit an honest note instead of fabricating numbers.
keys = ", ".join(model._safe_str((k or {}).get("col"))
for k in candidates.get("group_keys") or []
if _is_dict(k))
note = model.Note(
"No se pudo calcular la agregación: el capítulo necesita los datos "
"crudos. Pasa ctx['db_path'] + ctx['table'] (para el cálculo "
"push-down en DuckDB) o ctx['aggregations'] ya precalculado. "
f"Columnas categóricas candidatas: {keys or ''}.")
blocks = _intro_blocks() + [note] + _insights_section(ctx)
return model.Chapter(id=CHAPTER_ID, title=CHAPTER_TITLE,
version=CHAPTER_VERSION, blocks=blocks)
blocks = _intro_blocks() + sections + _insights_section(ctx)
return model.Chapter(id=CHAPTER_ID, title=CHAPTER_TITLE,
version=CHAPTER_VERSION, blocks=blocks)
@@ -1,256 +0,0 @@
"""Tests for the AGREGACION chapter — DoD: golden + edges + error/no-cut path.
Self-contained and deterministic: no DuckDB and no LLM. The aggregation results
are passed pre-computed via ``ctx['aggregations']`` (the same shape the push-down
registry functions ``groupby_stats_duckdb`` / ``pivot_table_duckdb`` produce), so
the chapter's rendering logic is exercised without touching disk or the network.
Live push-down + LLM selection are covered separately by the golden script.
Verifies:
- Golden: a profile with categoricals + numerics builds a Chapter with per-group
stats tables, a pivot table and bar-chart figures, and it renders to PDF AND
PPTX showing the group keys, values and pivot — nothing cut.
- Edges: a dataset with no low-cardinality categorical returns None; an empty
profile returns None; a profile that *could* be grouped but has no reachable
data degrades to an honest note instead of raising.
- No-cut: many groups (30) + a long interpretation paragraph survive intact in
the rendered PDF (table split by rows, text wrapped whole).
"""
import os
import re
import tempfile
from pptx import Presentation
from pypdf import PdfReader
from datascience.automatic_eda.chapters.agregacion import build_agregacion
from datascience.automatic_eda.model import Chapter
from datascience.render_automatic_eda_pdf import render_automatic_eda_pdf
from datascience.render_automatic_eda_pptx import render_automatic_eda_pptx
# --------------------------------------------------------------------------- #
# Synthetic fixtures.
# --------------------------------------------------------------------------- #
def _profile() -> dict:
"""A titanic-like profile: 2 categoricals + 2 numeric measures + 1 id."""
return {
"table": "titanic",
"source": "/data/titanic.csv",
"n_rows": 891,
"n_cols": 5,
"key_candidates": ["passenger_id"],
"columns": [
{"name": "passenger_id", "inferred_type": "numeric",
"unique_pct": 1.0, "flags": ["possible_id"],
"numeric": {"mean": 446.0, "std": 257.0}},
{"name": "sex", "inferred_type": "categorical", "distinct_count": 2,
"flags": [], "categorical": {"n_distinct": 2, "imbalance": 0.1,
"top": [{"value": "male", "count": 577}]}},
{"name": "pclass", "inferred_type": "categorical", "distinct_count": 3,
"flags": [], "categorical": {"n_distinct": 3, "imbalance": 0.2}},
{"name": "fare", "inferred_type": "numeric", "flags": [],
"numeric": {"mean": 32.2, "std": 49.7, "cv": 1.54}},
{"name": "age", "inferred_type": "numeric", "flags": [],
"numeric": {"mean": 29.7, "std": 14.5, "cv": 0.49}},
],
}
def _groupby_result(group_by: str, keys_n: list) -> dict:
"""A groupby_stats_duckdb-shaped result for `fare` and `age`."""
groups = []
for i, (key, n) in enumerate(keys_n):
groups.append({
"key": key, "n": n,
"stats": {
"fare": {"mean": 20.0 + i * 15, "median": 10.0 + i * 8,
"std": 40.0 + i, "min": 0.0, "max": 512.3},
"age": {"mean": 28.0 + i, "median": 27.0 + i, "std": 14.0,
"min": 0.42, "max": 80.0},
},
})
return {"status": "ok", "group_by": group_by, "measures": ["fare", "age"],
"aggs": ["count", "mean", "median", "std", "min", "max"],
"n_groups": len(groups), "truncated": False, "groups": groups}
def _pivot_result() -> dict:
return {"status": "ok", "index": "sex", "columns": "pclass", "value": "fare",
"agg": "mean", "row_labels": ["male", "female"],
"col_labels": ["1", "2", "3"],
"matrix": [[62.0, 19.0, 12.0], [110.0, 22.0, 15.0]],
"truncated_rows": False, "truncated_cols": False}
def _ctx_precomputed() -> dict:
return {
"aggregations": {
"groupby": [
{"group_by": "sex", "measures": ["fare", "age"],
"why": "sexo del pasajero",
"result": _groupby_result("sex", [("male", 577), ("female", 314)])},
{"group_by": "pclass", "measures": ["fare", "age"],
"why": "clase del billete",
"result": _groupby_result(
"pclass", [("3", 491), ("1", 216), ("2", 184)])},
],
"pivots": [
{"index": "sex", "columns": "pclass", "value": "fare",
"agg": "mean", "why": "tarifa por sexo y clase",
"result": _pivot_result()},
],
},
"agg_insights": [
{"title": "Tarifa por sexo",
"text": "Las mujeres pagaron de media casi el doble que los hombres."},
],
}
def _pdf_text(path: str) -> str:
txt = "".join((pg.extract_text() or "") for pg in PdfReader(path).pages)
return re.sub(r"\s+", " ", txt)
def _pptx_text(path: str) -> str:
prs = Presentation(path)
parts = []
for sl in prs.slides:
for sh in sl.shapes:
if sh.has_text_frame:
parts.append(sh.text_frame.text)
if sh.has_table:
tb = sh.table
for r in range(len(tb.rows)):
for c in range(len(tb.columns)):
parts.append(tb.cell(r, c).text)
return re.sub(r"\s+", " ", " ".join(parts))
# --------------------------------------------------------------------------- #
# Golden: builds a Chapter and renders to both formats.
# --------------------------------------------------------------------------- #
def test_golden_chapter_blocks_present():
ch = build_agregacion(_profile(), _ctx_precomputed())
assert isinstance(ch, Chapter)
assert ch.id == "agregacion"
kinds = [b.kind for b in ch.blocks]
assert "heading" in kinds
assert kinds.count("data_table") >= 3 # 2 group summaries + pivot (+details)
assert "figure" in kinds # at least one bar chart.
# Headings mention the group keys and the pivot.
htext = " ".join(b.text for b in ch.blocks if b.kind == "heading")
assert "sex" in htext and "pclass" in htext and "Pivot" in htext
def test_golden_render_pdf():
ch = build_agregacion(_profile(), _ctx_precomputed())
with tempfile.TemporaryDirectory() as d:
out = os.path.join(d, "agg.pdf")
res = render_automatic_eda_pdf([ch], out, {"write_manifest": False})
assert res["path"] == out and os.path.exists(out)
assert res["n_pages"] >= 1
txt = _pdf_text(out)
assert "Agregación por grupos" in txt
assert "male" in txt and "female" in txt # group + pivot labels.
assert "Pivot" in txt
assert "mediana" in txt # per-measure detail.
assert "casi el doble" in txt # interpretation kept.
def test_golden_render_pptx():
ch = build_agregacion(_profile(), _ctx_precomputed())
with tempfile.TemporaryDirectory() as d:
out = os.path.join(d, "agg.pptx")
res = render_automatic_eda_pptx([ch], out, {"write_manifest": False})
assert res["path"] == out and os.path.exists(out)
assert res["n_slides"] >= 1
txt = _pptx_text(out)
assert "male" in txt and "pclass" in txt
assert "Pivot" in txt or "sex" in txt
# --------------------------------------------------------------------------- #
# Edges.
# --------------------------------------------------------------------------- #
def test_edge_no_categorical_returns_none():
# Only numerics + an id: nothing to group by -> chapter does not apply.
prof = {
"table": "t", "n_rows": 100, "key_candidates": ["id"],
"columns": [
{"name": "id", "inferred_type": "numeric", "unique_pct": 1.0,
"flags": ["possible_id"], "numeric": {"std": 10.0}},
{"name": "x", "inferred_type": "numeric", "flags": [],
"numeric": {"mean": 1.0, "std": 2.0}},
],
}
assert build_agregacion(prof, {}) is None
def test_edge_empty_profile_returns_none():
assert build_agregacion({}, {}) is None
assert build_agregacion(None, None) is None
def test_edge_high_cardinality_only_returns_none():
# The single categorical is id-like (high cardinality) -> not groupable.
prof = {
"table": "t", "n_rows": 100, "key_candidates": ["uuid"],
"columns": [
{"name": "uuid", "inferred_type": "categorical", "distinct_count": 100,
"flags": ["high_cardinality", "possible_id"]},
{"name": "x", "inferred_type": "numeric", "flags": [],
"numeric": {"mean": 1.0, "std": 2.0}},
],
}
assert build_agregacion(prof, {}) is None
def test_live_without_data_degrades_to_note():
# Has a categorical to group by but no db_path / no precomputed results:
# must NOT raise and must emit an honest note (chapter still applies).
prof = {
"table": "t", "n_rows": 100, "key_candidates": [],
"columns": [
{"name": "grp", "inferred_type": "categorical", "distinct_count": 3,
"flags": [], "categorical": {"n_distinct": 3}},
{"name": "v", "inferred_type": "numeric", "flags": [],
"numeric": {"mean": 1.0, "std": 2.0}},
],
}
ch = build_agregacion(prof, {})
assert isinstance(ch, Chapter)
notes = [b.text for b in ch.blocks if b.kind == "note"]
assert any("datos crudos" in n for n in notes)
# --------------------------------------------------------------------------- #
# No-cut: many groups + long text survive intact in the PDF.
# --------------------------------------------------------------------------- #
def test_anti_corte_muchos_grupos_y_texto_largo():
keys_n = [(f"grupo_{i:02d}", 30 - (i % 5)) for i in range(30)]
long_text = " ".join(f"palabra{i}" for i in range(120))
ctx = {
"aggregations": {
"groupby": [
{"group_by": "cat", "measures": ["fare"], "why": "muchos niveles",
"result": _groupby_result("cat", keys_n)},
],
"pivots": [],
},
"agg_insights": [{"title": "Nota larga", "text": long_text}],
}
ch = build_agregacion(_profile(), ctx)
with tempfile.TemporaryDirectory() as d:
out = os.path.join(d, "big.pdf")
res = render_automatic_eda_pdf([ch], out, {"write_manifest": False})
assert res["path"] == out
assert res["n_pages"] > 1 # 30-row table + figure spill across pages.
txt = _pdf_text(out)
# First and last group labels both survive (table not truncated).
assert "grupo_00" in txt and "grupo_29" in txt
# First, middle and last words of the long paragraph all present.
for i in (0, 60, 119):
assert f"palabra{i}" in txt
@@ -0,0 +1,613 @@
"""Time-series chapter (TIMESERIES) for AutomaticEDA.
This chapter applies **only when the table has a date/datetime column**. When it
does, it draws — exactly the user requirement — the evolution of the data over
time (the value of each numeric column aggregated per period *and* the count of
rows per period) plus the statistical analysis of the series (stationarity,
autocorrelation, trend and seasonality). When there is no temporal column
``build_timeseries`` returns ``None``.
Data sources, read defensively and never recomputed here:
- ``profile['columns']`` — to detect the time column and the numeric columns.
Delegated to the pure registry function ``detect_time_column`` (group ``eda``).
- ``profile['series'][col]`` — the per-column time-series analysis already
produced by ``profile_table(run_series=True)``: ``stationarity`` (ADF+KPSS),
``acf_pacf`` (ACF/PACF + Ljung-Box), ``stl`` (trend/seasonal/resid +
Hyndman strengths) and the levels/returns suggestion.
- ``ctx['timeseries_raw']`` (or ``profile['timeseries_raw']``) — the *raw* ordered
series ``{time_col, t:[iso...], series:{col:[float|None]}}`` needed to draw the
value-vs-time line and the per-period row count. Exactly like ``modelos`` reads
``raw_numeric`` from ``ctx``, this chapter looks for the raw series there and
degrades honestly when it is absent (it still renders the textual analysis).
The raw series is aggregated per period with the pure registry function
``resample_timeseries`` and the datetime header is built with ``profile_datetime``
(both group ``eda``). Every figure is emitted as a lazy ``Figure`` so the
renderers rasterize and scale it to fit a whole page/slide; tables go through
``DataTable``/``KVTable`` so the paginator splits them repeating the header. No
content is ever cut.
ctx keys this chapter consumes (all optional):
timeseries_raw : dict — ``{time_col, t:[...], series:{col:[...]}}`` raw
ordered series used to draw the value-vs-time line and the row-count
panel. When absent the chapter omits those figures (with a note) and
renders only the analysis available in ``profile['series']``.
Contract: build_<id>(profile, ctx) -> Chapter | None ; CHAPTER_VERSION = "x.y.z".
Reads everything defensively (``.get``) and never raises.
"""
from __future__ import annotations
from .. import model
# Pure/impure registry functions (group ``eda``) consumed by this chapter,
# imported defensively so the chapter still builds (degrading the affected
# section to a note) if any of them is somehow unavailable.
try:
from datascience.detect_time_column import detect_time_column
except Exception: # noqa: BLE001 — keep the chapter importable no matter what.
detect_time_column = None # type: ignore[assignment]
try:
from datascience.profile_datetime import profile_datetime
except Exception: # noqa: BLE001
profile_datetime = None # type: ignore[assignment]
try:
from datascience.resample_timeseries import resample_timeseries
except Exception: # noqa: BLE001
resample_timeseries = None # type: ignore[assignment]
CHAPTER_VERSION = "1.0.0"
CHAPTER_ID = "timeseries"
CHAPTER_TITLE = "Series temporales"
# Plain-Spanish gloss for the stationarity verdict of adf_kpss_stationarity.
_VERDICT_GLOSS = {
"stationary": "estacionaria: media y varianza estables en el tiempo; se "
"puede modelar directamente.",
"non_stationary": "no estacionaria: tiene tendencia o varianza cambiante "
"(raíz unitaria). Correlacionar o modelar sus niveles "
"produce relaciones espurias (Granger-Newbold); conviene "
"diferenciar o pasar a retornos.",
"inconclusive": "resultado no concluyente (ADF y KPSS discrepan): tratar con "
"cautela, probablemente cerca de la no estacionariedad.",
}
# OHLC-style name fragments used to collapse near-identical financial series.
_OHLC_HINTS = ("open", "high", "low", "close", "adj", "price", "vwap")
def _fmt_num(value, decimals: int = 3) -> str:
"""Compact, defensive number formatting shared with the other chapters."""
if value is None:
return ""
if isinstance(value, bool):
return "" if value else "no"
if isinstance(value, int):
return f"{value:,}".replace(",", ".")
if isinstance(value, float):
if value != value: # NaN
return "NaN"
if value in (float("inf"), float("-inf")):
return str(value)
text = f"{value:.{decimals}f}".rstrip("0").rstrip(".")
return text if text else "0"
return model._safe_str(value)
def _is_dict(v) -> bool:
return isinstance(v, dict)
# --------------------------------------------------------------------------- #
# Detection: which column is the time axis and which numeric columns to chart.
# --------------------------------------------------------------------------- #
def _detect(cols: list) -> dict:
"""Return ``{time_col, numeric_cols, ...}`` via the registry function.
Falls back to an inline scan (datetime inferred_type / datetime semantic
types) when ``detect_time_column`` is unavailable, so the chapter still works.
"""
if detect_time_column is not None:
try:
res = detect_time_column(cols)
if _is_dict(res):
return res
except Exception: # noqa: BLE001 — degrade to the inline scan.
pass
time_col = None
numeric_cols = []
for c in cols or []:
if not _is_dict(c):
continue
it = c.get("inferred_type")
sem = c.get("semantic_type")
if time_col is None and (
it == "datetime" or sem in ("datetime_iso", "date_eu")):
time_col = c.get("name")
if it == "numeric":
numeric_cols.append(c.get("name"))
return {"time_col": time_col, "numeric_cols": numeric_cols,
"time_semantic": "", "reason": "inline fallback"}
def _raw_series_for(raw: dict, col: str):
"""Return (t_list, v_list) for a column from the raw bundle, or (None, None)."""
if not _is_dict(raw):
return None, None
t = raw.get("t")
series = raw.get("series") if _is_dict(raw.get("series")) else {}
v = series.get(col)
if isinstance(t, list) and isinstance(v, list) and t and len(t) == len(v):
return t, v
return None, None
def _ohlc_groups(numeric_cols: list, raw: dict) -> dict:
"""Map each numeric column to a representative to collapse OHLC duplicates.
When several numeric columns are near-identical financial level series
(open/high/low/close/adj close), charting each one repeats the same figure
four times. We keep the first OHLC-looking column as the representative for
the *figures* and list the collapsed ones in a note; the textual analysis is
still produced for every column. Detection is by name only (cheap, no extra
data dependency) and conservative: only collapses when >=2 OHLC-like names
are present.
"""
ohlc = [c for c in numeric_cols
if isinstance(c, str) and any(h in c.lower() for h in _OHLC_HINTS)]
if len(ohlc) < 2:
return {}
representative = ohlc[0]
return {c: representative for c in ohlc if c != representative}
# --------------------------------------------------------------------------- #
# Datetime header (MUST-9.3): range / frequency / regularity / gaps.
# --------------------------------------------------------------------------- #
def _datetime_header(time_col: str, raw: dict) -> list:
"""Build the datetime profile header from the raw time axis, when present."""
blocks: list = []
t, _ = (raw.get("t"), None) if _is_dict(raw) else (None, None)
if not (isinstance(t, list) and t and profile_datetime is not None):
return blocks
try:
dt = profile_datetime(t)
except Exception: # noqa: BLE001
return blocks
if not _is_dict(dt):
return blocks
freq_gloss = {
"daily": "diaria", "weekly": "semanal", "monthly": "mensual",
"quarterly": "trimestral", "yearly": "anual",
"irregular": "irregular", "unknown": "indeterminada",
}
rows = [
("Columna de fecha", model._safe_str(time_col)),
("Rango", f"{model._safe_str(dt.get('min'))}"
f"{model._safe_str(dt.get('max'))}"),
("Observaciones", _fmt_num(dt.get("n"))),
("Fechas distintas", _fmt_num(dt.get("n_distinct"))),
("Frecuencia", freq_gloss.get(dt.get("freq"), model._safe_str(dt.get("freq")))),
("Regular", "" if dt.get("is_regular") else "no"),
]
span = dt.get("span_days")
if span is not None:
rows.append(("Duración (días)", _fmt_num(span, 1)))
n_gaps = dt.get("n_gaps")
if n_gaps is not None:
rows.append(("Huecos en la rejilla", _fmt_num(n_gaps)))
blocks.append(model.KVTable(rows=rows, title="Perfil temporal"))
note = dt.get("note")
if note:
blocks.append(model.Note(model._safe_str(note)))
return blocks
# --------------------------------------------------------------------------- #
# Figure builders (lazy: matplotlib only imported when the renderer draws them).
# --------------------------------------------------------------------------- #
def _parse_dates(labels: list):
"""Parse a list of ISO-ish strings/dates to datetime, dropping unparseable.
Returns (dates, kept_index) so callers can align the values list.
"""
from datetime import date, datetime
out = []
keep = []
for i, lab in enumerate(labels):
if isinstance(lab, datetime):
out.append(lab)
keep.append(i)
continue
if isinstance(lab, date):
out.append(datetime(lab.year, lab.month, lab.day))
keep.append(i)
continue
s = model._safe_str(lab).strip()
if not s:
continue
s2 = s.replace("T", " ")
parsed = None
for fmt in ("%Y-%m-%d %H:%M:%S", "%Y-%m-%d %H:%M", "%Y-%m-%d"):
try:
parsed = datetime.strptime(s2[:len(fmt) + 4] if False else s2, fmt)
break
except ValueError:
continue
if parsed is None:
try:
parsed = datetime.fromisoformat(s.replace("T", " "))
except ValueError:
continue
out.append(parsed)
keep.append(i)
return out, keep
def _make_evolution_figure(name: str, rs: dict):
"""Lazy callable: value-vs-time line + per-period row-count panel (MUST-9.1)."""
def _draw():
import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
t_labels = rs.get("t") or []
v = rs.get("v") or []
counts = rs.get("count") or []
dates, keep = _parse_dates(t_labels)
vv = [v[i] if i < len(v) else None for i in keep]
cc = [counts[i] if i < len(counts) else 0 for i in keep]
fig, (ax_v, ax_c) = plt.subplots(
2, 1, figsize=(7.0, 4.6), sharex=True,
gridspec_kw={"height_ratios": [3.0, 1.2], "hspace": 0.12})
# Top: value aggregated per period (line; gaps where the value is None).
xs = [d for d, val in zip(dates, vv) if val is not None]
ys = [val for val in vv if val is not None]
if xs and ys:
ax_v.plot(xs, ys, color="#4e79a7", linewidth=1.4, zorder=3)
ax_v.fill_between(xs, ys, min(ys), color="#9ec6df", alpha=0.18,
zorder=1)
else:
ax_v.text(0.5, 0.5, "(sin valores numéricos)", ha="center",
va="center", fontsize=9, color="#8a8a8a",
transform=ax_v.transAxes)
ax_v.set_ylabel(name, fontsize=8)
ax_v.tick_params(labelsize=7)
ax_v.grid(axis="y", color="#eeeeee", linewidth=0.6)
for spine in ("top", "right"):
ax_v.spines[spine].set_visible(False)
# Bottom: number of observations per period (density / gaps).
if dates and cc:
# Bar width ~ median spacing so bars do not overlap nor leave gaps.
width = 1.0
if len(dates) > 1:
deltas = sorted((dates[i + 1] - dates[i]).days
for i in range(len(dates) - 1))
width = max(deltas[len(deltas) // 2] * 0.8, 1.0)
ax_c.bar(dates, cc, width=width, color="#59a14f", alpha=0.75,
align="center")
ax_c.set_ylabel("nº filas", fontsize=8)
ax_c.tick_params(labelsize=7)
ax_c.grid(axis="y", color="#eeeeee", linewidth=0.6)
for spine in ("top", "right"):
ax_c.spines[spine].set_visible(False)
ax_c.xaxis.set_major_locator(mdates.AutoDateLocator())
ax_c.xaxis.set_major_formatter(mdates.ConciseDateFormatter(
ax_c.xaxis.get_major_locator()))
freq = rs.get("freq")
suptitle = f"{name} — evolución temporal"
if freq:
suptitle += f" (agregado {freq})"
fig.suptitle(suptitle, fontsize=10, fontweight="bold", x=0.02, ha="left")
return fig
return _draw
def _make_stl_figure(stl: dict):
"""Lazy callable: the STL trend/seasonal/resid panels, or None if no values.
``stl_decompose`` only carries the component *values* for short series; for
long ones it returns just summary stats (``note``). In that case there is
nothing to plot and we return None (the caller renders the strengths as text).
"""
def _component_values(comp):
if _is_dict(comp):
vals = comp.get("values")
if isinstance(vals, list) and vals:
return [x for x in vals]
return None
trend = _component_values(stl.get("trend"))
seasonal = _component_values(stl.get("seasonal"))
resid = _component_values(stl.get("resid"))
if not any([trend, seasonal, resid]):
return None
def _draw():
import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt
panels = [("Tendencia", trend, "#4e79a7"),
("Estacional", seasonal, "#59a14f"),
("Resto", resid, "#e15759")]
panels = [(lbl, vals, col) for lbl, vals, col in panels if vals]
fig, axes = plt.subplots(len(panels), 1, figsize=(7.0, 1.4 * len(panels) + 0.6),
sharex=True)
if len(panels) == 1:
axes = [axes]
for ax, (lbl, vals, col) in zip(axes, panels):
ax.plot(range(len(vals)), vals, color=col, linewidth=1.2)
ax.set_ylabel(lbl, fontsize=8)
ax.tick_params(labelsize=7)
ax.grid(axis="y", color="#eeeeee", linewidth=0.6)
for spine in ("top", "right"):
ax.spines[spine].set_visible(False)
axes[-1].set_xlabel("índice temporal", fontsize=8)
fig.suptitle("Descomposición STL", fontsize=10, fontweight="bold",
x=0.02, ha="left")
fig.tight_layout(rect=(0, 0, 1, 0.96))
return fig
return _draw
def _make_acf_figure(acf_pacf: dict):
"""Lazy callable: the ACF stem plot with ±1.96/√n bands, or None."""
acf = acf_pacf.get("acf")
n = acf_pacf.get("n")
if not (isinstance(acf, list) and len(acf) > 1 and isinstance(n, int) and n > 0):
return None
def _draw():
import math
import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt
lags = list(range(len(acf)))
fig, ax = plt.subplots(figsize=(7.0, 3.2))
ax.vlines(lags, 0, acf, color="#4e79a7", linewidth=1.4)
ax.plot(lags, acf, "o", color="#4e79a7", markersize=3)
band = 1.96 / math.sqrt(n)
ax.axhspan(-band, band, color="#cccccc", alpha=0.3,
label="banda ±1.96/√n (ruido blanco)")
ax.axhline(0, color="#888888", linewidth=0.8)
ax.set_xlabel("retardo (lag)", fontsize=8)
ax.set_ylabel("ACF", fontsize=8)
ax.tick_params(labelsize=7)
ax.legend(fontsize=7, loc="upper right", framealpha=0.85)
ax.set_title("Autocorrelación (ACF): lags fuera de la banda = "
"correlación significativa", fontsize=9)
fig.tight_layout()
return fig
return _draw
# --------------------------------------------------------------------------- #
# Per-column textual analysis from profile['series'][col].
# --------------------------------------------------------------------------- #
def _analysis_markdown(sblock: dict) -> str:
"""One markdown block summarizing stationarity / autocorrelation / STL."""
parts: list = []
stat = sblock.get("stationarity") if _is_dict(sblock.get("stationarity")) else {}
verdict = stat.get("verdict")
if verdict:
adf = stat.get("adf") if _is_dict(stat.get("adf")) else {}
kpss = stat.get("kpss") if _is_dict(stat.get("kpss")) else {}
line = (f"**Estacionariedad:** {_VERDICT_GLOSS.get(verdict, verdict)} "
f"(ADF p={_fmt_num(adf.get('p_value'), 4)}, "
f"KPSS p={_fmt_num(kpss.get('p_value'), 4)}).")
warning = stat.get("warning")
if warning:
line += f"{model._safe_str(warning)}"
parts.append(line)
acf = sblock.get("acf_pacf") if _is_dict(sblock.get("acf_pacf")) else {}
if acf:
is_auto = acf.get("is_autocorrelated")
lb = acf.get("ljung_box") if _is_dict(acf.get("ljung_box")) else {}
sig = acf.get("significant_acf_lags") or []
if is_auto is True:
ac_line = ("**Autocorrelación:** la serie está autocorrelada "
"(Ljung-Box rechaza independencia, "
f"p={_fmt_num(lb.get('p_value'), 4)}): los valores dependen "
"de su pasado, no es ruido blanco.")
if sig:
shown = ", ".join(str(x) for x in sig[:8])
more = "" if len(sig) > 8 else ""
ac_line += f" Lags significativos: {shown}{more}."
elif is_auto is False:
ac_line = ("**Autocorrelación:** no se detecta autocorrelación "
"significativa (compatible con ruido blanco, Ljung-Box "
f"p={_fmt_num(lb.get('p_value'), 4)}).")
else:
ac_line = "**Autocorrelación:** no evaluable (datos insuficientes)."
parts.append(ac_line)
stl = sblock.get("stl") if _is_dict(sblock.get("stl")) else {}
if stl:
ts = stl.get("trend_strength")
ss = stl.get("seasonal_strength")
if ts is not None or ss is not None:
parts.append(
"**Descomposición STL:** fuerza de tendencia "
f"{_fmt_num(ts, 2)} y fuerza estacional {_fmt_num(ss, 2)} "
"(escala 01 de Hyndman: cuanto más alto, más marcada la "
"componente).")
elif stl.get("note"):
parts.append(f"**Descomposición STL:** {model._safe_str(stl.get('note'))}")
if sblock.get("levels_suggested"):
reason = sblock.get("levels_reason")
kind = sblock.get("levels_kind")
tr = sblock.get("to_returns") if _is_dict(sblock.get("to_returns")) else None
line = "**Transformación sugerida:** "
line += "pasar a retornos" if kind == "returns" else "diferenciar la serie"
if reason:
line += f"{model._safe_str(reason)}"
if tr and tr.get("mean") is not None:
line += (f" (retornos: media {_fmt_num(tr.get('mean'), 5)}, "
f"σ {_fmt_num(tr.get('std'), 5)}).")
parts.append(line)
return "\n\n".join(parts)
# --------------------------------------------------------------------------- #
# Per-column section.
# --------------------------------------------------------------------------- #
def _column_section(name: str, sblock: dict, raw: dict, collapsed_into) -> list:
"""Blocks for one numeric column: evolution figure + STL + ACF + analysis."""
blocks = [model.Heading(text=model._safe_str(name), level=2)]
# --- Value-vs-time line + per-period row count (MUST-9.1). ---
drew_evolution = False
if collapsed_into is None: # skip the figure for collapsed OHLC duplicates.
t, v = _raw_series_for(raw, name)
if t is not None and resample_timeseries is not None:
try:
rs = resample_timeseries(t, v)
except Exception: # noqa: BLE001
rs = None
if _is_dict(rs) and rs.get("t"):
blocks.append(model.Figure(
make=_make_evolution_figure(name, rs),
caption=f"Evolución de «{name}» por periodo y nº de "
f"observaciones (conteo de filas)."))
drew_evolution = True
else:
blocks.append(model.Note(
f"Serie casi idéntica a «{collapsed_into}» (grupo OHLC): se omite el "
"gráfico para no repetirlo; el análisis estadístico se mantiene."))
if not drew_evolution and collapsed_into is None:
blocks.append(model.Note(
"Gráfico de evolución temporal no disponible: falta la serie cruda "
"(pásala en ctx['timeseries_raw'] = {time_col, t, series}). Se "
"muestra solo el análisis estadístico."))
# --- STL panels (MUST-9.2). ---
stl = sblock.get("stl") if _is_dict(sblock.get("stl")) else {}
if collapsed_into is None and stl:
stl_fig = _make_stl_figure(stl)
if stl_fig is not None:
blocks.append(model.Figure(
make=stl_fig,
caption=f"Descomposición STL de «{name}»: tendencia, componente "
f"estacional y resto."))
# --- ACF figure (autocorrelation structure). ---
acf = sblock.get("acf_pacf") if _is_dict(sblock.get("acf_pacf")) else {}
if collapsed_into is None and acf:
acf_fig = _make_acf_figure(acf)
if acf_fig is not None:
blocks.append(model.Figure(
make=acf_fig,
caption=f"Función de autocorrelación de «{name}»."))
# --- Textual analysis (always, even for collapsed duplicates). ---
analysis = _analysis_markdown(sblock)
if analysis:
blocks.append(model.Markdown(text=analysis))
return blocks
# --------------------------------------------------------------------------- #
# Entry point.
# --------------------------------------------------------------------------- #
def build_timeseries(profile: dict, ctx: dict):
"""Build the TIMESERIES Chapter, or ``None`` if the table has no date column.
Args:
profile: the ``eda`` group TableProfile dict.
ctx: presentation context; ``ctx['timeseries_raw']`` (optional) carries
the raw ordered series used to draw the value-vs-time line and the
per-period row count.
Returns:
A ``model.Chapter`` with, per numeric column, the value-vs-time evolution
+ row-count figure, the STL panels, the ACF figure and the statistical
analysis; or ``None`` when there is no temporal column (the chapter does
not apply).
"""
profile = profile or {}
if not _is_dict(profile):
profile = {}
ctx = ctx or {}
cols = profile.get("columns") or []
det = _detect(cols)
time_col = det.get("time_col")
if not time_col:
return None # no date/datetime column -> chapter does not apply.
numeric_cols = det.get("numeric_cols") or []
series_map = profile.get("series") if _is_dict(profile.get("series")) else {}
raw = ctx.get("timeseries_raw") or profile.get("timeseries_raw")
raw = raw if _is_dict(raw) else {}
# Which columns can the chapter say anything about: those with a series
# analysis block and/or a raw series to chart. Preserve the profile order.
chartable = []
for name in numeric_cols:
has_analysis = _is_dict(series_map.get(name))
has_raw, _ = _raw_series_for(raw, name)
if has_analysis or has_raw is not None:
chartable.append(name)
if not chartable:
# A date column exists but nothing numeric to chart/analyse: still a
# valid (small) chapter — show just the datetime header if we have it.
header = _datetime_header(time_col, raw)
if not header:
return None
intro = (
f"La tabla tiene una columna temporal («{time_col}») pero no hay "
"columnas numéricas con serie analizable.")
blocks = [model.Heading(text=CHAPTER_TITLE, level=1),
model.Markdown(text=intro)] + header
return model.Chapter(id=CHAPTER_ID, title=CHAPTER_TITLE,
version=CHAPTER_VERSION, blocks=blocks)
collapsed = _ohlc_groups(chartable, raw)
intro = (
"Este capítulo analiza la evolución de la tabla en el tiempo usando la "
f"columna de fecha «{time_col}». Para cada columna numérica se muestra su "
"**evolución por periodo** (valor agregado) junto al **número de filas por "
"periodo** (densidad de observaciones), su **descomposición STL** "
"(tendencia / estacionalidad / resto) y la **función de autocorrelación**; "
"debajo, el análisis de la serie: estacionariedad (ADF + KPSS), "
"autocorrelación (Ljung-Box) y, cuando procede, la transformación "
"sugerida (retornos o diferencias) para evitar correlaciones espurias.")
blocks = [model.Heading(text=CHAPTER_TITLE, level=1),
model.Markdown(text=intro)]
blocks += _datetime_header(time_col, raw)
if collapsed:
reps = sorted(set(collapsed.values()))
collapsed_names = ", ".join(sorted(collapsed.keys()))
blocks.append(model.Note(
f"Series OHLC casi idénticas detectadas ({collapsed_names}): se "
f"grafican consolidadas en «{', '.join(reps)}» para no repetir el "
"mismo gráfico; cada columna conserva su análisis estadístico."))
for name in chartable:
sblock = series_map.get(name) if _is_dict(series_map.get(name)) else {}
blocks += _column_section(name, sblock, raw, collapsed.get(name))
return model.Chapter(id=CHAPTER_ID, title=CHAPTER_TITLE,
version=CHAPTER_VERSION, blocks=blocks)
@@ -0,0 +1,244 @@
"""Tests for the TIMESERIES chapter — DoD: golden + edges + anti-cut.
Self-contained: builds synthetic ``series`` blocks (shaped like
``profile_table(run_series=True)`` output) and a raw ``timeseries_raw`` bundle,
with no DuckDB, so the suite is fast and deterministic. Verifies that the chapter:
- returns ``None`` when there is no date/datetime column (the user requirement);
- never raises on ``None``/empty/garbage input;
- with a date column + raw series emits, per numeric column, the value-vs-time +
row-count evolution figure, the STL panels, the ACF figure and the textual
analysis (stationarity / autocorrelation / suggested transform);
- collapses near-identical OHLC series into one chart while keeping every
column's analysis;
- renders without cutting anything in both PDF and PPTX (every column heading
survives in the rendered output).
"""
import math
import os
import re
import tempfile
from pypdf import PdfReader
from datascience.automatic_eda.chapters.timeseries import (
build_timeseries, CHAPTER_VERSION, _VERDICT_GLOSS,
)
from datascience.render_automatic_eda_pdf import render_automatic_eda_pdf
from datascience.render_automatic_eda_pptx import render_automatic_eda_pptx
# --------------------------------------------------------------------------- #
# Synthetic fixtures shaped like the real profile_table(run_series=True) output.
# --------------------------------------------------------------------------- #
def _dates(n: int) -> list:
"""n consecutive daily ISO date strings starting 2021-01-01."""
from datetime import date, timedelta
start = date(2021, 1, 1)
return [(start + timedelta(days=i)).isoformat() for i in range(n)]
def _series_block(n=120, verdict="non_stationary", autocorr=True, levels=True,
with_stl_values=True):
"""A synthetic ``series`` block like _build_series_block produces."""
trend = [float(i) for i in range(n)]
seasonal = [math.sin(i / 6.0) for i in range(n)]
resid = [0.1 * ((-1) ** i) for i in range(n)]
acf = [1.0] + [max(0.0, 0.9 - 0.05 * k) for k in range(1, 21)]
block = {
"order_col": "fecha",
"ordered": True,
"n": n,
"stationarity": {
"n": n, "verdict": verdict,
"adf": {"p_value": 0.42, "stationary": False},
"kpss": {"p_value": 0.01, "stationary": False},
"warning": ("serie no estacionaria: riesgo de correlación espuria"
if verdict != "stationary" else None),
},
"acf_pacf": {
"n": n, "nlags": 20, "acf": acf,
"significant_acf_lags": [1, 2, 3, 4, 5],
"ljung_box": {"stat": 123.4, "p_value": 0.0 if autocorr else 0.7,
"lags": 20},
"is_autocorrelated": autocorr,
},
"period_source": "datetime_freq",
"stl": {
"n": n, "period": 7, "period_inferred": False, "robust": False,
"trend": {"values": trend} if with_stl_values else {
"note": "serie larga: solo estadisticos", "mean": 60.0},
"seasonal": {"values": seasonal} if with_stl_values else {"mean": 0.0},
"resid": {"values": resid} if with_stl_values else {"mean": 0.0},
"trend_strength": 0.95, "seasonal_strength": 0.42,
},
}
if levels:
block["levels_suggested"] = True
block["levels_kind"] = "returns"
block["levels_reason"] = ("columna financiera no estacionaria: usar "
"retornos evita correlación espuria.")
block["to_returns"] = {"method": "log", "mean": 0.001, "std": 0.02}
else:
block["levels_suggested"] = False
return block
def _profile(numeric_names=("precio",), n=120, with_stl_values=True):
cols = [{"name": "fecha", "inferred_type": "datetime",
"semantic_type": "datetime_iso"}]
series_map = {}
for nm in numeric_names:
cols.append({"name": nm, "inferred_type": "numeric",
"numeric": {"min": 1.0, "max": 200.0, "mean": 100.0,
"median": 95.0, "std": 40.0}})
series_map[nm] = _series_block(n=n, with_stl_values=with_stl_values)
return {"table": "cotizaciones", "n_rows": n, "n_cols": len(cols),
"columns": cols, "series": series_map}
def _ctx_raw(numeric_names=("precio",), n=120):
t = _dates(n)
series = {}
for j, nm in enumerate(numeric_names):
series[nm] = [float(100 + i + 5 * j) for i in range(n)]
return {"timeseries_raw": {"time_col": "fecha", "t": t, "series": series}}
def _pdf_text(path: str) -> str:
txt = "".join((pg.extract_text() or "") for pg in PdfReader(path).pages)
return re.sub(r"\s+", " ", txt)
# --------------------------------------------------------------------------- #
# Golden.
# --------------------------------------------------------------------------- #
def test_golden_estructura_y_figuras():
ch = build_timeseries(_profile(("precio",)), _ctx_raw(("precio",)))
assert ch is not None
assert ch.id == "timeseries"
assert ch.version == CHAPTER_VERSION
kinds = [b.kind for b in ch.blocks]
assert kinds[0] == "heading" # chapter title
assert kinds[1] == "markdown" # intro
assert "kv_table" in kinds # datetime profile header (MUST-9.3)
# Per column: evolution figure + STL figure + ACF figure + analysis markdown.
figs = [b for b in ch.blocks if b.kind == "figure"]
assert len(figs) >= 3, "evolución + STL + ACF esperadas"
# Lazy makers must produce real matplotlib figures.
import matplotlib.pyplot as plt
for f in figs:
fig = f.make()
assert fig is not None
plt.close(fig)
def test_golden_evolucion_tiene_dos_paneles_valor_y_conteo():
# MUST-9.1: the evolution figure has a value panel + a row-count panel.
ch = build_timeseries(_profile(("precio",)), _ctx_raw(("precio",)))
figs = [b for b in ch.blocks if b.kind == "figure"]
import matplotlib.pyplot as plt
fig = figs[0].make() # first figure is the evolution one.
assert len(fig.axes) == 2, "panel de valor + panel de conteo de filas"
plt.close(fig)
def test_golden_analisis_textual_presente():
ch = build_timeseries(_profile(("precio",)), _ctx_raw(("precio",)))
md = " ".join(b.text for b in ch.blocks if b.kind == "markdown")
assert "Estacionariedad" in md
assert "Autocorrelación" in md
assert "STL" in md
# Verdict gloss surfaced for the non-stationary preset.
assert _VERDICT_GLOSS["non_stationary"].split(":")[0] in md
# Levels/returns suggestion surfaced.
assert "retornos" in md.lower()
# --------------------------------------------------------------------------- #
# Edges.
# --------------------------------------------------------------------------- #
def test_edge_sin_columna_fecha_devuelve_none():
prof = {"columns": [
{"name": "precio", "inferred_type": "numeric", "numeric": {"mean": 1.0}},
{"name": "ciudad", "inferred_type": "categorical",
"categorical": {"top": []}},
], "series": {"precio": _series_block()}}
assert build_timeseries(prof, {}) is None
def test_edge_none_y_vacio_no_revienta():
assert build_timeseries(None, None) is None
assert build_timeseries({}, {}) is None
assert build_timeseries({"columns": []}, {}) is None
# Date column but nothing numeric/series and no raw -> None (nothing to say).
assert build_timeseries(
{"columns": [{"name": "fecha", "inferred_type": "datetime"}]}, {}) is None
def test_edge_sin_raw_degrada_pero_mantiene_analisis():
# No ctx['timeseries_raw']: the chapter must still build (STL/ACF/analysis
# from the profile) and note that the evolution chart is unavailable.
ch = build_timeseries(_profile(("precio",)), {})
assert ch is not None
notes = " ".join(b.text for b in ch.blocks if b.kind == "note")
assert "evolución temporal no disponible" in notes
md = " ".join(b.text for b in ch.blocks if b.kind == "markdown")
assert "Estacionariedad" in md
def test_edge_stl_solo_estadisticos_no_dibuja_panel_pero_no_revienta():
# Long series: STL carries only stats (no 'values') -> no STL figure, but the
# strengths still surface in the textual analysis.
ch = build_timeseries(_profile(("precio",), with_stl_values=False),
_ctx_raw(("precio",)))
assert ch is not None
md = " ".join(b.text for b in ch.blocks if b.kind == "markdown")
assert "STL" in md
# --------------------------------------------------------------------------- #
# OHLC consolidation (MUST-9.3).
# --------------------------------------------------------------------------- #
def test_ohlc_consolidacion():
names = ("Open", "High", "Low", "Close")
ch = build_timeseries(_profile(names), _ctx_raw(names))
assert ch is not None
notes = " ".join(b.text for b in ch.blocks if b.kind == "note")
assert "OHLC" in notes
# Only the representative draws the evolution figure; the other 3 are collapsed
# so there are fewer evolution figures than columns.
captions = [b.caption or "" for b in ch.blocks if b.kind == "figure"]
evo = [c for c in captions if "Evolución" in c]
assert len(evo) < len(names), "las series OHLC deben consolidarse"
# Every column still has its analysis markdown (one heading per column).
headings = [b.text for b in ch.blocks if b.kind == "heading" and b.level == 2]
for nm in names:
assert nm in headings
# --------------------------------------------------------------------------- #
# Anti-cut: PDF + PPTX.
# --------------------------------------------------------------------------- #
def test_anti_corte_pdf_y_pptx():
names = tuple(f"serie_{i}" for i in range(6))
prof = _profile(names, n=90)
ctx = _ctx_raw(names, n=90)
ch = build_timeseries(prof, ctx)
col_headings = [b.text for b in ch.blocks if b.kind == "heading" and b.level == 2]
assert len(col_headings) == 6
with tempfile.TemporaryDirectory() as d:
pdf = os.path.join(d, "ts.pdf")
res_pdf = render_automatic_eda_pdf(
prof, pdf, {"ctx": ctx, "write_manifest": False})
assert res_pdf["path"] == pdf
txt = _pdf_text(pdf)
for nm in col_headings:
assert nm in txt, f"columna '{nm}' cortada/ausente en el PDF"
pptx = os.path.join(d, "ts.pptx")
res_pptx = render_automatic_eda_pptx(
prof, pptx, {"ctx": ctx, "write_manifest": False})
assert res_pptx["path"] == pptx
assert res_pptx["n_slides"] >= 6
@@ -0,0 +1,68 @@
---
name: detect_time_column
kind: function
lang: py
domain: datascience
version: "1.0.0"
purity: pure
signature: "def detect_time_column(columns: list) -> dict"
description: "Detecta, a partir de la lista de ColumnProfile de un TableProfile del grupo eda, cual es la columna de orden temporal y que columnas numericas hay para graficar una serie en el tiempo. Una columna es temporal si inferred_type=='datetime' o semantic_type in {datetime_iso, date_eu}; time_col es la primera temporal en orden. Es la pieza que usa el capitulo TIMESERIES del AutomaticEDA para decidir si aplica. Lectura defensiva dict-no-throw: nunca lanza, siempre devuelve las mismas claves."
tags: [eda, timeseries, datetime, profiling, column-detection, automatic-eda, datascience, python]
uses_functions: []
uses_types: []
returns: []
returns_optional: false
error_type: ""
imports: []
params:
- name: columns
desc: "lista de ColumnProfile dict de un TableProfile del grupo eda. Cada elemento suele tener name, inferred_type, semantic_type y numeric. Elementos que no sean dict se ignoran; None/no-lista/vacia -> dict 'no aplica'."
output: "dict SIEMPRE con: time_col (str|None, columna temporal elegida = primera temporal), time_semantic (str, semantic_type de la temporal o ''), numeric_cols (list[str], columnas con inferred_type=='numeric' en orden), n_datetime_cols (int), datetime_cols (list[str], todas las temporales en orden de aparicion), reason (str en espanol explicando la eleccion). Nunca lanza excepcion."
tested: true
tests: ["test_golden_datetime_y_numericas", "test_deteccion_por_semantic_type_date_eu", "test_sin_columna_temporal", "test_columns_none_no_revienta", "test_columns_vacia_no_revienta", "test_columns_no_lista_no_revienta", "test_elementos_basura_se_ignoran", "test_varias_datetime_elige_la_primera"]
test_file_path: "python/functions/datascience/detect_time_column_test.py"
file_path: "python/functions/datascience/detect_time_column.py"
---
## Ejemplo
```python
from datascience import detect_time_column
columns = [
{"name": "fecha", "inferred_type": "datetime", "semantic_type": "datetime_iso"},
{"name": "ventas", "inferred_type": "numeric"},
{"name": "unidades", "inferred_type": "numeric"},
{"name": "region", "inferred_type": "text"},
]
res = detect_time_column(columns)
res["time_col"] # -> "fecha"
res["numeric_cols"] # -> ["ventas", "unidades"]
res["n_datetime_cols"] # -> 1
# Sin columna temporal: el capitulo TIMESERIES no aplica.
detect_time_column([{"name": "id", "inferred_type": "numeric"}])["time_col"] # -> None
```
## Cuando usarla
Cuando el capitulo TIMESERIES del AutomaticEDA recibe un TableProfile y necesita
decidir si la tabla admite analisis de serie temporal: si `time_col` es None no
hay eje de tiempo y el capitulo se salta; si hay `time_col` y `numeric_cols`,
úsalas como eje X (orden cronologico) y series Y. Tambien sirve para enrutar el
resto del pipeline (acf_pacf / stl_decompose / adf_kpss_stationarity) sobre las
columnas numericas detectadas.
## Gotchas
- Es pura y stdlib-only (sin numpy ni DuckDB): segura de llamar en cualquier paso.
- `time_col` se elige por ORDEN de aparicion en la lista, no por "mejor candidata".
Si hay varias columnas datetime y quieres otra, filtra `datetime_cols` tu mismo.
- Solo mira metadatos del perfil (`inferred_type`/`semantic_type`); no parsea ni
valida los valores reales de la columna. La calidad de la deteccion depende de
que el profiler (summarize_table_duckdb / infer_semantic_type) haya inferido bien.
- Las claves del semantic_type son exactamente las del profiler: `datetime_iso`
(ISO 8601) y `date_eu` (DD/MM/AAAA). Otros formatos de fecha no se detectan por
semantic_type salvo que `inferred_type` ya sea `"datetime"`.
- `numeric_cols` se basa en `inferred_type == "numeric"` (no en "integer"/"float");
si tu profiler usa otra etiqueta, normalizala antes.
@@ -0,0 +1,112 @@
"""Detecta la columna temporal y las columnas numericas de un TableProfile (grupo eda).
Funcion pura y determinista: a partir de la lista de columnas de un TableProfile
producido por el grupo de capacidad `eda` (cada elemento es un ColumnProfile dict),
decide cual es la columna de orden temporal y que columnas numericas hay disponibles
para graficar una serie en el tiempo. Es la pieza que usa el capitulo TIMESERIES del
AutomaticEDA para decidir si la tabla admite analisis de serie temporal.
Lectura 100% defensiva al estilo "dict-no-throw" del grupo eda: nunca lanza
excepcion, siempre devuelve el mismo conjunto de claves.
"""
# semantic_type que el profiler (infer_semantic_type) emite para fechas/datetimes.
_DATETIME_SEMANTICS = ("datetime_iso", "date_eu")
def detect_time_column(columns: list) -> dict:
"""Detecta la columna temporal y las numericas de una lista de ColumnProfile.
Recorre los ColumnProfile de un TableProfile y clasifica cada columna como
temporal o numerica leyendo de forma defensiva sus claves. Una columna es
temporal si su ``inferred_type == "datetime"`` o si su ``semantic_type`` esta
en {``"datetime_iso"``, ``"date_eu"``}. La columna temporal elegida
(``time_col``) es la PRIMERA temporal en el orden de la lista. Las numericas
(``numeric_cols``) son las de ``inferred_type == "numeric"``, en orden.
Funcion pura: no hace I/O, no muta el input, es determinista.
Args:
columns: lista de ColumnProfile dict del grupo eda. Cada elemento suele
tener claves como ``name``, ``inferred_type``, ``semantic_type`` y
``numeric``. Los elementos que no sean dict se ignoran. Si ``columns``
es None, no es lista o esta vacia, se devuelve el dict "no aplica".
Returns:
Siempre un dict con las mismas claves::
{
"time_col": str | None, # columna temporal elegida (None si no hay)
"time_semantic": str, # semantic_type de la temporal ("" si no aplica)
"numeric_cols": [str, ...], # columnas con inferred_type == "numeric"
"n_datetime_cols": int, # nº de columnas temporales detectadas
"datetime_cols": [str, ...],# todas las temporales, en orden de aparicion
"reason": str, # frase corta (en espanol) que explica la eleccion
}
"""
# Caso "no aplica": entrada invalida o vacia.
if not isinstance(columns, list) or not columns:
return {
"time_col": None,
"time_semantic": "",
"numeric_cols": [],
"n_datetime_cols": 0,
"datetime_cols": [],
"reason": "no se detecto columna de fecha/datetime",
}
datetime_cols: list[str] = []
datetime_semantics: list[str] = []
numeric_cols: list[str] = []
for col in columns:
# Ignora elementos que no sean dict sin fallar.
if not isinstance(col, dict):
continue
name = col.get("name")
if name is None:
name = ""
else:
name = str(name)
inferred_type = col.get("inferred_type") or ""
semantic_type = col.get("semantic_type") or ""
is_datetime = inferred_type == "datetime" or semantic_type in _DATETIME_SEMANTICS
if is_datetime:
datetime_cols.append(name)
datetime_semantics.append(semantic_type)
if inferred_type == "numeric":
numeric_cols.append(name)
if not datetime_cols:
return {
"time_col": None,
"time_semantic": "",
"numeric_cols": numeric_cols,
"n_datetime_cols": 0,
"datetime_cols": [],
"reason": "no se detecto columna de fecha/datetime",
}
time_col = datetime_cols[0]
time_semantic = datetime_semantics[0]
if len(datetime_cols) == 1:
reason = f"columna temporal '{time_col}' detectada"
else:
reason = (
f"{len(datetime_cols)} columnas temporales; se elige la primera "
f"'{time_col}'"
)
return {
"time_col": time_col,
"time_semantic": time_semantic,
"numeric_cols": numeric_cols,
"n_datetime_cols": len(datetime_cols),
"datetime_cols": datetime_cols,
"reason": reason,
}
@@ -0,0 +1,102 @@
"""Tests para detect_time_column (grupo eda). Self-contained, sin DuckDB."""
from detect_time_column import detect_time_column
def test_golden_datetime_y_numericas():
columns = [
{"name": "fecha", "inferred_type": "datetime", "semantic_type": "datetime_iso"},
{"name": "ventas", "inferred_type": "numeric"},
{"name": "unidades", "inferred_type": "numeric"},
{"name": "region", "inferred_type": "text"},
]
res = detect_time_column(columns)
assert res["time_col"] == "fecha"
assert res["time_semantic"] == "datetime_iso"
assert res["numeric_cols"] == ["ventas", "unidades"]
assert res["n_datetime_cols"] == 1
assert res["datetime_cols"] == ["fecha"]
assert isinstance(res["reason"], str) and res["reason"]
def test_deteccion_por_semantic_type_date_eu():
# inferred_type no es datetime, pero semantic_type date_eu => temporal.
columns = [
{"name": "id", "inferred_type": "numeric"},
{"name": "dia", "inferred_type": "text", "semantic_type": "date_eu"},
{"name": "importe", "inferred_type": "numeric"},
]
res = detect_time_column(columns)
assert res["time_col"] == "dia"
assert res["time_semantic"] == "date_eu"
assert res["numeric_cols"] == ["id", "importe"]
assert res["n_datetime_cols"] == 1
assert res["datetime_cols"] == ["dia"]
def test_sin_columna_temporal():
columns = [
{"name": "id", "inferred_type": "numeric"},
{"name": "nombre", "inferred_type": "text"},
{"name": "activo", "inferred_type": "boolean"},
]
res = detect_time_column(columns)
assert res["time_col"] is None
assert res["time_semantic"] == ""
assert res["numeric_cols"] == ["id"]
assert res["n_datetime_cols"] == 0
assert res["datetime_cols"] == []
assert res["reason"] == "no se detecto columna de fecha/datetime"
def test_columns_none_no_revienta():
res = detect_time_column(None)
assert res["time_col"] is None
assert res["time_semantic"] == ""
assert res["numeric_cols"] == []
assert res["n_datetime_cols"] == 0
assert res["datetime_cols"] == []
assert res["reason"] == "no se detecto columna de fecha/datetime"
def test_columns_vacia_no_revienta():
res = detect_time_column([])
assert res["time_col"] is None
assert res["numeric_cols"] == []
assert res["n_datetime_cols"] == 0
def test_columns_no_lista_no_revienta():
# Un dict (no lista) tambien debe caer en el caso "no aplica".
res = detect_time_column({"name": "fecha", "inferred_type": "datetime"})
assert res["time_col"] is None
assert res["numeric_cols"] == []
def test_elementos_basura_se_ignoran():
columns = [
None,
"no soy un dict",
42,
{"name": "ts", "inferred_type": "datetime"},
{"name": "valor", "inferred_type": "numeric"},
]
res = detect_time_column(columns)
assert res["time_col"] == "ts"
assert res["numeric_cols"] == ["valor"]
assert res["n_datetime_cols"] == 1
def test_varias_datetime_elige_la_primera():
columns = [
{"name": "created_at", "inferred_type": "datetime", "semantic_type": "datetime_iso"},
{"name": "metric", "inferred_type": "numeric"},
{"name": "updated_at", "inferred_type": "datetime", "semantic_type": "datetime_iso"},
{"name": "fecha_baja", "inferred_type": "text", "semantic_type": "date_eu"},
]
res = detect_time_column(columns)
assert res["time_col"] == "created_at"
assert res["time_semantic"] == "datetime_iso"
assert res["n_datetime_cols"] == 3
assert res["datetime_cols"] == ["created_at", "updated_at", "fecha_baja"]
assert res["numeric_cols"] == ["metric"]
@@ -0,0 +1,92 @@
---
name: extract_timeseries_raw
kind: function
lang: py
domain: datascience
version: "1.0.0"
purity: impure
signature: "def extract_timeseries_raw(query_fn, table: str, time_col: str, value_cols: list, max_rows: int = 5000) -> dict"
description: "Extrae la serie temporal CRUDA (fechas + una o varias columnas numericas) de una tabla, ordenada cronologicamente, para alimentar el render del capitulo TIMESERIES de AutomaticEDA (linea valor-vs-tiempo + conteo por periodo). Recibe un lector read-only inyectado `query_fn(sql) -> dict` (mismo contrato que duckdb_query_readonly / pg_query / el `_q` de profile_table) y NO abre ninguna conexion por su cuenta. Construye UNA sola query con identificadores escapados, ORDER BY por la columna temporal y LIMIT. Devuelve dict dict-no-throw: t (fechas ISO string), series (lista paralela float|None por columna) y n. El capitulo no toca la BD: recibe esto en ctx['timeseries_raw']. Reutilizable tambien por profile_table en una fase futura."
tags: [eda, timeseries, datascience, automatic-eda, extraction, read-only, duckdb, postgres, python]
uses_functions: []
uses_types: []
returns: []
returns_optional: false
error_type: "error_go_core"
imports: [datetime]
params:
- name: query_fn
desc: "callable lector read-only del backend activo. Recibe un string SQL y devuelve un dict {'status':'ok','rows':[{col:val,...},...]} (mismo contrato que duckdb_query_readonly o el `_q` de profile_table). NO se abre ninguna conexion dentro de la funcion: toda la lectura pasa por query_fn. Si es None -> error."
- name: table
desc: "nombre de la tabla de la que extraer la serie. Se escapa con comillas dobles en la query."
- name: time_col
desc: "nombre de la columna de orden temporal. Se usa en ORDER BY (cronologico ascendente) y se filtra IS NOT NULL. Sus valores se devuelven en `t` como string ISO."
- name: value_cols
desc: "lista de nombres de columnas numericas a extraer. Cada una produce una entrada en `series` con una lista paralela a `t`. Vacia o None -> status error."
- name: max_rows
desc: "limite de filas a leer (clausula LIMIT). Default 5000. Protege el render frente a tablas enormes."
output: "dict (nunca lanza). En exito: {'status':'ok','time_col':str,'t':[str,...] (fechas ISO en orden),'series':{col:[float|None,...],...} (paralela a t por value_col, None si el valor no es convertible a float),'n':int}. En error (sin lanzar): {'status':'error','error':str,'time_col':str,'t':[],'series':{},'n':0}. Errores: query_fn None, value_cols vacia, table/time_col vacios, o query_fn devuelve status!='ok' (se propaga su error)."
tested: true
tests: ["test_golden_t_y_series_alineadas", "test_valor_no_convertible_da_none", "test_value_cols_vacia_status_error", "test_query_fn_status_error_propaga", "test_query_fn_none_da_error_sin_reventar", "test_sql_contiene_order_by_y_limit"]
test_file_path: "python/functions/datascience/extract_timeseries_raw_test.py"
file_path: "python/functions/datascience/extract_timeseries_raw.py"
---
## Ejemplo
```python
import sys, os
sys.path.insert(0, os.path.join("python", "functions"))
from datascience import extract_timeseries_raw
from infra import duckdb_query_readonly
# El lector read-only se inyecta como closure (igual que el `_q` de profile_table).
db = "data/ventas.duckdb"
def _q(sql):
return duckdb_query_readonly(db, sql)
res = extract_timeseries_raw(_q, "ventas_diarias", "fecha", ["importe", "unidades"])
# res == {
# "status": "ok",
# "time_col": "fecha",
# "t": ["2024-01-01", "2024-01-02", ...],
# "series": {"importe": [1234.5, 980.0, ...], "unidades": [12.0, 9.0, ...]},
# "n": 365,
# }
# Se entrega al capitulo TIMESERIES sin que este toque la BD:
ctx = {"timeseries_raw": res}
```
## Cuando usarla
Cuando el capitulo TIMESERIES de AutomaticEDA necesita pintar una serie
valor-vs-tiempo (o conteo por periodo) y NO debe abrir la base de datos por su
cuenta: extraes aqui las fechas + columnas numericas ordenadas y se las pasas en
`ctx['timeseries_raw']`. Usala tambien siempre que quieras la secuencia cruda
ordenada cronologicamente de una o varias columnas para alimentar otros
contrastes de serie (ADF/KPSS, ACF/PACF, STL) reutilizando un unico lector
read-only inyectado, en vez de hacer N muestreos a mano.
## Gotchas
- **Impura**: lee de la base de datos a traves de `query_fn`. No abre conexiones
por su cuenta — depende por completo del lector inyectado. Sigue el estilo
dict-no-throw del grupo `eda`: nunca lanza; ante cualquier fallo devuelve
`{"status":"error","error":...}` con `t=[]`, `series={}`, `n=0`.
- **`error_type` en el frontmatter es `error_go_core` por convencion del registry**
(toda funcion impura debe declararlo y el indexer lo exige), pero el codigo
NO lanza esa excepcion: degrada al dict de error. Es metadata, no comportamiento.
- **No loguear los datos crudos**: `t`/`series` pueden contener datos sensibles
(igual que un HAR). No volcar el dict completo a logs ni a telemetria; en
trazas usa solo `n` y los nombres de columna.
- **Alineacion por fila**: `series[col][i]` corresponde a `t[i]`. Un valor no
convertible a float se guarda como `None` (no se descarta la fila) para no
romper la alineacion temporal.
- **Orden**: el orden cronologico depende del `ORDER BY "time_col"` del backend.
Si `time_col` esta guardada como texto con formato no lexicograficamente
ordenable (p.ej. `DD/MM/YYYY`), el orden no sera el real — normaliza la columna
a date/timestamp antes, o pasa una columna ya ordenable.
- **`max_rows`**: con LIMIT, si la tabla supera `max_rows` obtienes solo el primer
tramo cronologico, no un muestreo uniforme. Sube `max_rows` si necesitas el rango
completo.
@@ -0,0 +1,122 @@
"""extract_timeseries_raw — extrae la serie temporal CRUDA de una tabla.
Lector read-only inyectado: recibe `query_fn(sql) -> dict` con el mismo contrato
que duckdb_query_readonly / pg_query (y que el `_q` de profile_table):
`{"status": "ok", "rows": [{col: val, ...}, ...]}`. Esta funcion NO abre ninguna
conexion por su cuenta — solo usa `query_fn`. Construye UNA sola query ordenada
por la columna temporal y devuelve las fechas (`t`) mas cada columna numerica en
listas paralelas (`series`), listas para alimentar el render del capitulo
TIMESERIES de AutomaticEDA (linea valor-vs-tiempo + conteo por periodo) sin que
el capitulo toque la base de datos: recibe esto en `ctx['timeseries_raw']`.
Estilo dict-no-throw del grupo `eda`: nunca lanza; captura cualquier excepcion y
degrada a `{"status": "error", "error": str, ...}`.
"""
from datetime import date, datetime
def _to_float(value):
"""Convierte un valor a float de forma defensiva. None si no es convertible."""
if value is None:
return None
if isinstance(value, bool):
# Un bool es subclase de int en Python; no es un valor de serie valido.
return None
if isinstance(value, (int, float)):
return float(value)
s = str(value).strip()
if not s:
return None
try:
return float(s)
except (TypeError, ValueError):
return None
def _to_iso(value):
"""Convierte un valor temporal a string ISO conservando el orden de la query.
date/datetime -> isoformat(); cualquier otro valor (string, etc.) -> str().
None se preserva como None.
"""
if value is None:
return None
if isinstance(value, (datetime, date)):
return value.isoformat()
return str(value)
def extract_timeseries_raw(query_fn, table, time_col, value_cols, max_rows=5000):
"""Extrae la serie temporal cruda (fechas + columnas numericas) de una tabla.
Args:
query_fn: callable lector read-only del backend activo. Recibe un string
SQL y devuelve un dict {"status": "ok", "rows": [{col: val, ...}]}
(mismo contrato que duckdb_query_readonly / el `_q` de profile_table).
No se abre ninguna conexion aqui: toda la lectura pasa por query_fn.
table: nombre de la tabla.
time_col: nombre de la columna de orden temporal.
value_cols: lista de nombres de columnas numericas a extraer.
max_rows: limite de filas (LIMIT). Default 5000.
Returns:
dict (nunca lanza):
{
"status": "ok" | "error",
"error": str, # solo si status == "error"
"time_col": str,
"t": [str, ...], # time_col como ISO string, en orden
"series": {col: [float|None, ...], ...}, # paralela a t por columna
"n": int # nº de filas devueltas
}
"""
base = {"status": "ok", "time_col": time_col, "t": [], "series": {}, "n": 0}
try:
if query_fn is None:
return {**base, "status": "error", "error": "query_fn es None"}
if not value_cols:
return {**base, "status": "error", "error": "value_cols vacío"}
if not table or not time_col:
return {
**base,
"status": "error",
"error": "table y time_col son obligatorios",
}
# Identificadores escapados con comillas dobles (como hace profile_table)
# para tolerar nombres con mayusculas/espacios/palabras reservadas.
cols_sql = ", ".join(f'"{c}"' for c in value_cols)
sql = (
f'SELECT "{time_col}", {cols_sql} FROM "{table}" '
f'WHERE "{time_col}" IS NOT NULL '
f'ORDER BY "{time_col}" '
f"LIMIT {int(max_rows)}"
)
q = query_fn(sql)
if not isinstance(q, dict) or q.get("status") != "ok":
err = (
q.get("error", "query_fn fallo")
if isinstance(q, dict)
else "query_fn no devolvio un dict"
)
return {**base, "status": "error", "error": err}
rows = q.get("rows", []) or []
t = []
series = {c: [] for c in value_cols}
for row in rows:
t.append(_to_iso(row.get(time_col)))
for c in value_cols:
series[c].append(_to_float(row.get(c)))
return {
"status": "ok",
"time_col": time_col,
"t": t,
"series": series,
"n": len(t),
}
except Exception as e: # noqa: BLE001 - dict-no-throw: degradar, nunca lanzar
return {**base, "status": "error", "error": str(e)}
@@ -0,0 +1,109 @@
"""Tests para extract_timeseries_raw.
No usa DuckDB real: inyecta un query_fn FAKE (closure) que devuelve filas
predefinidas y, opcionalmente, captura el SQL recibido para verificar la query
generada (ORDER BY por la columna temporal + LIMIT). Asi el test es
autocontenido y no depende de ningun backend.
"""
import os
import sys
sys.path.insert(0, os.path.dirname(__file__))
from extract_timeseries_raw import extract_timeseries_raw
def _fake_query(rows, captured=None, status="ok", error=None):
"""Crea un query_fn FAKE.
`captured` (lista opcional) recibe el SQL ejecutado para poder inspeccionarlo.
`status`/`error` permiten simular un fallo del backend.
"""
def _q(sql):
if captured is not None:
captured.append(sql)
if status != "ok":
return {"status": "error", "error": error or "boom"}
return {"status": "ok", "rows": rows}
return _q
def test_golden_t_y_series_alineadas():
"""Golden: t y series alineadas, floats convertidos, n correcto."""
rows = [
{"fecha": "2024-01-01", "ventas": "10", "stock": 5},
{"fecha": "2024-01-02", "ventas": "20.5", "stock": 7},
{"fecha": "2024-01-03", "ventas": 30, "stock": 9},
]
res = extract_timeseries_raw(_fake_query(rows), "t", "fecha", ["ventas", "stock"])
assert res["status"] == "ok"
assert res["n"] == 3
assert res["time_col"] == "fecha"
assert res["t"] == ["2024-01-01", "2024-01-02", "2024-01-03"]
assert res["series"]["ventas"] == [10.0, 20.5, 30.0]
assert res["series"]["stock"] == [5.0, 7.0, 9.0]
def test_valor_no_convertible_da_none():
"""Valor no convertible a float -> None en la serie (alineacion preservada)."""
rows = [
{"fecha": "2024-01-01", "ventas": "abc"},
{"fecha": "2024-01-02", "ventas": None},
{"fecha": "2024-01-03", "ventas": "12.5"},
]
res = extract_timeseries_raw(_fake_query(rows), "t", "fecha", ["ventas"])
assert res["status"] == "ok"
assert res["series"]["ventas"] == [None, None, 12.5]
assert res["n"] == 3
def test_value_cols_vacia_status_error():
"""value_cols vacia -> status error con t/series/n vacios."""
res = extract_timeseries_raw(_fake_query([]), "t", "fecha", [])
assert res["status"] == "error"
assert "value_cols" in res["error"]
assert res["t"] == []
assert res["series"] == {}
assert res["n"] == 0
def test_query_fn_status_error_propaga():
"""query_fn que devuelve status != ok -> se propaga como error."""
res = extract_timeseries_raw(
_fake_query([], status="error", error="db locked"),
"t",
"fecha",
["ventas"],
)
assert res["status"] == "error"
assert "db locked" in res["error"]
assert res["n"] == 0
def test_query_fn_none_da_error_sin_reventar():
"""query_fn None -> error degradado, sin excepcion."""
res = extract_timeseries_raw(None, "t", "fecha", ["ventas"])
assert res["status"] == "error"
assert res["t"] == []
assert res["n"] == 0
def test_sql_contiene_order_by_y_limit():
"""La query generada ordena por time_col y aplica el LIMIT sobre la tabla."""
captured = []
rows = [{"fecha": "2024-01-01", "ventas": 1}]
extract_timeseries_raw(
_fake_query(rows, captured),
"ventas_tbl",
"fecha",
["ventas"],
max_rows=123,
)
assert len(captured) == 1
sql = captured[0]
assert 'ORDER BY "fecha"' in sql
assert "LIMIT 123" in sql
assert 'FROM "ventas_tbl"' in sql
@@ -1,87 +0,0 @@
---
name: groupby_stats_duckdb
kind: function
lang: py
domain: datascience
version: "1.0.0"
purity: impure
signature: "def groupby_stats_duckdb(db_path: str, table: str, group_by: str, measures: list, aggs: list = None, top_n: int = 15) -> dict"
description: "Agregaciones GROUP BY con push-down SQL en DuckDB: para cada measure numerica calcula mean/median/std/min/max por grupo (split-apply-combine en el motor), trayendo solo una fila por grupo. Nucleo de un capitulo de agregacion/OLAP de un EDA. count = tamanio del grupo, independiente de measures."
tags: [eda, groupby, aggregation, olap, duckdb, datascience, push-down, split-apply-combine]
uses_functions: [duckdb_query_readonly_py_infra]
uses_types: []
returns: []
returns_optional: false
error_type: "error_go_core"
imports: []
params:
- name: db_path
desc: "Ruta al archivo DuckDB. Debe existir; el modo read_only NO crea la base. Path inexistente -> {status:'error'} sin lanzar."
- name: table
desc: "Nombre de la tabla. Se interpola citado con dobles comillas (soporta nombres con espacios; las comillas internas se escapan)."
- name: group_by
desc: "Columna por la que agrupar. Se interpola citada. Sus valores distintos son las claves de los grupos."
- name: measures
desc: "Lista de columnas numericas a agregar. Lista vacia es valida: cada grupo trae solo su tamanio `n` y `stats` vacio."
- name: aggs
desc: "Lista de agregaciones. None (default) = ['count','mean','median','std','min','max']. Validas: count (tamanio del grupo, va a `n`), mean->avg, median, std->stddev_samp, min, max (estas cinco por measure). Agg desconocido -> error."
- name: top_n
desc: "Maximo de grupos a devolver, ordenados por tamanio de grupo descendente (default 15). Internamente se piden top_n+1 para detectar truncado."
output: "dict. En exito {status:'ok', group_by, measures:[...], aggs:[...], n_groups:int, truncated:bool, groups:[{key:<valor grupo>, n:int, stats:{<measure>:{mean,median,std,min,max}}}], note:str}. Las estadisticas son float o None (p.ej. std de un grupo de 1 fila -> NULL -> None). En error {status:'error', error:str} (no lanza)."
tested: true
tests: ["agrega por grupo con valores conocidos", "db inexistente devuelve error sin lanzar", "measures vacias agrega solo count", "columna con espacio agrupa bien"]
test_file_path: "python/functions/datascience/groupby_stats_duckdb_test.py"
file_path: "python/functions/datascience/groupby_stats_duckdb.py"
---
## Ejemplo
```python
import duckdb
from datascience import groupby_stats_duckdb
# Cargar el titanic en una tabla DuckDB de prueba.
db = "/tmp/titanic.duckdb"
con = duckdb.connect(db)
con.execute(
"CREATE TABLE titanic AS "
"SELECT * FROM read_csv_auto('https://raw.githubusercontent.com/"
"datasciencedojo/datasets/master/titanic.csv')"
)
con.close()
# Agrupar por sexo midiendo edad y tarifa.
res = groupby_stats_duckdb(db, "titanic", "Sex", ["Age", "Fare"])
print(res["status"]) # ok
print(res["n_groups"]) # 2 (male, female)
for g in res["groups"]:
print(g["key"], g["n"], round(g["stats"]["Fare"]["mean"], 2))
# female 314 44.48
# male 577 25.52
```
## Cuando usarla
Cuando en un EDA necesitas el clasico split-apply-combine: "para cada categoria de X,
¿cuanto vale en media/mediana/desviacion/min/max la metrica Y?". Es el nucleo de un
capitulo de agregacion/OLAP. Usala antes de pintar barras o boxplots por grupo, para
detectar segmentos con comportamiento distinto, o para resumir una tabla grande sin
traer las filas a RAM: todo el GROUP BY ocurre push-down en el motor de DuckDB y solo
viaja una fila por grupo. `top_n` te deja quedarte con los grupos mas poblados.
## Gotchas
- Funcion impura: lee un archivo DuckDB del disco (read_only, nunca lo modifica). La
tabla debe existir ya en el `.db` (no carga CSV; para eso crea la tabla antes).
- Identificadores (tabla, group_by, measures) se interpolan citados con dobles comillas
y escapando las internas: soporta nombres con espacios y evita inyeccion. No pases
expresiones SQL como group_by/measure — solo nombres de columna.
- `count` es el tamanio del grupo (`COUNT(*)`), independiente de las measures: se
refleja en el campo `n` de cada grupo, NO como clave dentro de `stats`. Las claves de
`stats[measure]` son las measure-aggs efectivas (mean/median/std/min/max menos count).
- `std` usa `stddev_samp` (muestral, n-1): un grupo con una sola fila da `NULL` -> `None`.
Las measures pueden contener NULLs; cada agregada los ignora segun la semantica de DuckDB.
- `truncated:True` indica que habia mas grupos que `top_n` (se devolvieron los `top_n`
mayores por tamanio). Sube `top_n` si necesitas todos los grupos.
- Si `measures` esta vacio, cada grupo trae solo `n` y `stats == {}` (valido, util para
un simple conteo por categoria).
@@ -1,184 +0,0 @@
"""groupby_stats_duckdb — agregaciones GROUP BY con push-down SQL en DuckDB.
Funcion impura: lee de disco a traves de DuckDB (via la primitiva read-only
`duckdb_query_readonly` del grupo `duckdb`). Pertenece al grupo de capacidad `eda`.
Ejecuta un `GROUP BY <group_by>` en el motor de DuckDB (split-apply-combine con
push-down) calculando, para cada columna numerica de `measures`, las agregaciones
pedidas (mean/median/std/min/max). Solo trae al cliente una fila por grupo, nunca
las filas crudas: apto para tablas grandes. Es el nucleo de un capitulo de
agregacion/OLAP de un EDA.
Estilo dict-no-throw del grupo duckdb: nunca lanza; captura cualquier error y
devuelve {status:'error', error:str}.
"""
from infra import duckdb_query_readonly
# Mapeo agg -> funcion agregada SQL de DuckDB. `count` se trata aparte: es
# COUNT(*) (tamanio del grupo), independiente de las measures.
_AGG_SQL = {
"mean": "avg",
"median": "median",
"std": "stddev_samp",
"min": "min",
"max": "max",
}
# Aggs por defecto cuando aggs=None. count primero (tamanio del grupo) + las
# cinco estadisticas por measure.
_DEFAULT_AGGS = ["count", "mean", "median", "std", "min", "max"]
def _quote_ident(ident: str) -> str:
"""Cita un identificador SQL con dobles comillas, escapando las internas.
Soporta nombres con espacios o caracteres especiales y evita inyeccion: dentro
de un identificador entrecomillado el unico caracter peligroso es la propia
comilla doble, que se duplica ("") segun el estandar SQL. DuckDB no admite
parametros posicionales para nombres de tabla/columna, asi que esta es la via
segura de interpolarlos.
"""
return '"' + str(ident).replace('"', '""') + '"'
def groupby_stats_duckdb(
db_path: str,
table: str,
group_by: str,
measures: list,
aggs: list = None,
top_n: int = 15,
) -> dict:
"""GROUP BY con agregaciones por measure, todo push-down en DuckDB.
Args:
db_path: ruta al archivo DuckDB. Debe existir; el modo read_only NO crea la
base. Un path inexistente devuelve {status:'error', ...} sin lanzar.
table: nombre de la tabla. Se interpola citado con dobles comillas (soporta
nombres con espacios).
group_by: columna por la que agrupar. Se interpola citada.
measures: lista de columnas numericas a agregar. Lista vacia es valida:
cada grupo trae solo su tamanio `n` y `stats` vacio.
aggs: lista de agregaciones a calcular. None (default) =
["count", "mean", "median", "std", "min", "max"]. Valores validos:
count (tamanio del grupo, va a `n`), mean, median, std, min, max
(estas cinco se calculan por cada measure). Un agg desconocido devuelve
error.
top_n: numero maximo de grupos a devolver, ordenados por tamanio de grupo
descendente (default 15). Se pide top_n+1 internamente para detectar si
habia mas grupos y marcar `truncated`.
Returns:
dict. En exito:
{status:'ok',
group_by:str,
measures:[...],
aggs:[...], # las efectivas (incluye count si se pidio)
n_groups:int, # nº de grupos devueltos (<= top_n)
truncated:bool, # True si habia mas de top_n grupos
groups:[{key:<valor grupo>, n:int,
stats:{<measure>:{mean,median,std,min,max}}}, ...],
note:str}
Las estadisticas son float o None (p.ej. stddev_samp de un grupo de una
sola fila -> NULL -> None). En error (sin lanzar): {status:'error', error:str}.
"""
try:
# 1. Validar entradas.
if not isinstance(table, str) or table == "":
return {"status": "error", "error": "table must be a non-empty string"}
if not isinstance(group_by, str) or group_by == "":
return {"status": "error", "error": "group_by must be a non-empty string"}
if measures is None:
measures = []
if not isinstance(measures, list):
return {"status": "error", "error": "measures must be a list"}
for m in measures:
if not isinstance(m, str) or m == "":
return {
"status": "error",
"error": f"invalid measure identifier: {m!r}",
}
if aggs is None:
aggs = list(_DEFAULT_AGGS)
if not isinstance(aggs, list) or len(aggs) == 0:
return {
"status": "error",
"error": "aggs must be a non-empty list or None",
}
for a in aggs:
if a != "count" and a not in _AGG_SQL:
return {
"status": "error",
"error": f"unknown agg {a!r}; valid: count, "
+ ", ".join(_AGG_SQL),
}
if not isinstance(top_n, int) or isinstance(top_n, bool) or top_n < 1:
return {"status": "error", "error": "top_n must be a positive int"}
# 2. Aggs por measure = todas menos count (count es el tamanio del grupo,
# se mapea siempre a la columna `n`).
measure_aggs = [a for a in aggs if a != "count"]
# 3. Construir el SELECT. grp y n primero; luego un termino por measure x agg
# con alias posicional (m{idx}_{agg}) para no chocar con nombres de columna
# que lleven espacios o caracteres raros.
select_terms = [f"{_quote_ident(group_by)} AS grp", "COUNT(*) AS n"]
agg_index = [] # (measure_name, agg_name, alias)
for mi, m in enumerate(measures):
for a in measure_aggs:
alias = f"m{mi}_{a}"
fn = _AGG_SQL[a]
select_terms.append(f"{fn}({_quote_ident(m)}) AS {alias}")
agg_index.append((m, a, alias))
# Pedimos top_n+1 grupos para detectar truncado (habia mas que top_n).
sql = (
f"SELECT {', '.join(select_terms)} "
f"FROM {_quote_ident(table)} "
f"GROUP BY {_quote_ident(group_by)} "
f"ORDER BY n DESC "
f"LIMIT {top_n + 1}"
)
# 4. Ejecutar push-down. sandbox=True (default) basta: la tabla ya existe en
# el .db, no necesitamos read_csv/read_blob ni acceso al filesystem.
result = duckdb_query_readonly(db_path, sql, max_rows=top_n + 1)
if result.get("status") != "ok":
return {
"status": "error",
"error": "groupby query failed: "
+ str(result.get("error", "unknown")),
}
rows = result.get("rows", [])
truncated = len(rows) > top_n
if truncated:
rows = rows[:top_n]
# 5. Reconstruir la estructura por grupo.
groups = []
for row in rows:
stats = {m: {} for m in measures}
for (m, a, alias) in agg_index:
stats[m][a] = row.get(alias)
groups.append(
{"key": row.get("grp"), "n": row.get("n"), "stats": stats}
)
return {
"status": "ok",
"group_by": group_by,
"measures": list(measures),
"aggs": list(aggs),
"n_groups": len(groups),
"truncated": truncated,
"groups": groups,
"note": f"GROUP BY {group_by}: top {len(groups)} grupos por tamanio sobre "
f"{len(measures)} measure(s)",
}
except Exception as e: # noqa: BLE001
return {"status": "error", "error": str(e)}
@@ -1,106 +0,0 @@
"""Tests para groupby_stats_duckdb."""
import os
import sys
import duckdb
# Permitir importar funciones del registry (from infra import ..., from datascience import ...).
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", ".."))
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "functions"))
from datascience.groupby_stats_duckdb import groupby_stats_duckdb
def _make_db(tmp_path, rows):
"""Crea una DuckDB con tabla t(g VARCHAR, x DOUBLE) e inserta `rows`."""
db = os.path.join(str(tmp_path), "t.duckdb")
con = duckdb.connect(db)
con.execute("CREATE TABLE t(g VARCHAR, x DOUBLE)")
con.executemany("INSERT INTO t VALUES (?, ?)", rows)
con.close()
return db
def test_agrega_por_grupo_con_valores_conocidos(tmp_path):
# Grupo a: [10, 20, 30] -> n=3, mean=20, min=10, max=30, median=20, std=10.
# Grupo b: [5, 15] -> n=2, mean=10, median=10.
# Grupo c: [100] -> n=1, mean=100, std=None (1 sola fila).
rows = [
("a", 10.0), ("a", 20.0), ("a", 30.0),
("b", 5.0), ("b", 15.0),
("c", 100.0),
]
db = _make_db(tmp_path, rows)
res = groupby_stats_duckdb(db, "t", "g", ["x"])
assert res["status"] == "ok", res
assert res["n_groups"] == 3
assert res["truncated"] is False
assert res["aggs"] == ["count", "mean", "median", "std", "min", "max"]
by_key = {g["key"]: g for g in res["groups"]}
assert set(by_key) == {"a", "b", "c"}
# Grupo a: comprobacion manual de mean/min/max/median/std.
sa = by_key["a"]["stats"]["x"]
assert by_key["a"]["n"] == 3
assert abs(sa["mean"] - 20.0) < 1e-9
assert abs(sa["min"] - 10.0) < 1e-9
assert abs(sa["max"] - 30.0) < 1e-9
assert abs(sa["median"] - 20.0) < 1e-9
assert "std" in sa and sa["std"] is not None
assert abs(sa["std"] - 10.0) < 1e-9 # stddev_samp([10,20,30]) = 10
# Grupo b: mean y median conocidas.
sb = by_key["b"]["stats"]["x"]
assert by_key["b"]["n"] == 2
assert abs(sb["mean"] - 10.0) < 1e-9
assert abs(sb["median"] - 10.0) < 1e-9
assert "median" in sb and "std" in sb
# Grupo c: una sola fila -> std None (stddev_samp NULL), mean/min/max definidos.
sc = by_key["c"]["stats"]["x"]
assert by_key["c"]["n"] == 1
assert abs(sc["mean"] - 100.0) < 1e-9
assert sc["std"] is None
def test_db_inexistente_devuelve_error_sin_lanzar(tmp_path):
db = os.path.join(str(tmp_path), "no_existe.duckdb")
res = groupby_stats_duckdb(db, "t", "g", ["x"])
assert res["status"] == "error", res
assert isinstance(res["error"], str) and res["error"]
def test_measures_vacias_agrega_solo_count(tmp_path):
rows = [("a", 1.0), ("a", 2.0), ("b", 3.0)]
db = _make_db(tmp_path, rows)
res = groupby_stats_duckdb(db, "t", "g", [])
assert res["status"] == "ok", res
by_key = {g["key"]: g for g in res["groups"]}
assert by_key["a"]["n"] == 2
assert by_key["b"]["n"] == 1
# Sin measures, stats por grupo es un dict vacio (valido).
assert by_key["a"]["stats"] == {}
assert by_key["b"]["stats"] == {}
def test_columna_con_espacio_agrupa_bien(tmp_path):
# Tabla con nombres de columna con espacios -> prueba el quoting con dobles
# comillas tanto en group_by como en la measure.
db = os.path.join(str(tmp_path), "space.duckdb")
con = duckdb.connect(db)
con.execute('CREATE TABLE t("my col" VARCHAR, "the val" DOUBLE)')
con.executemany(
'INSERT INTO t VALUES (?, ?)',
[("x", 1.0), ("x", 3.0), ("y", 10.0)],
)
con.close()
res = groupby_stats_duckdb(db, "t", "my col", ["the val"])
assert res["status"] == "ok", res
by_key = {g["key"]: g for g in res["groups"]}
assert by_key["x"]["n"] == 2
assert abs(by_key["x"]["stats"]["the val"]["mean"] - 2.0) < 1e-9
assert by_key["y"]["n"] == 1
assert abs(by_key["y"]["stats"]["the val"]["mean"] - 10.0) < 1e-9
@@ -1,92 +0,0 @@
---
name: pivot_table_duckdb
kind: function
lang: py
domain: datascience
version: "1.0.0"
purity: impure
signature: "def pivot_table_duckdb(db_path: str, table: str, index: str, columns: str, value: str, agg: str = 'mean', top_rows: int = 10, top_cols: int = 8) -> dict"
description: "Pivot table (index x columns -> agg(value)) calculada con push-down SQL en DuckDB (GROUP BY en el motor, sin traer filas a RAM) y recortada a las top_rows filas y top_cols columnas con mas observaciones para que quepa entera en un PDF movil / slide PPTX sin cortarse. Version push-down para tablas grandes de la funcion pura `pivot` (que pivota list[dict] en memoria)."
tags: [eda, pivot, duckdb, aggregate, datascience, push-down, report]
uses_functions: [duckdb_query_readonly_py_infra]
uses_types: []
returns: []
returns_optional: false
error_type: "error_go_core"
imports: []
params:
- name: db_path
desc: "Ruta al archivo DuckDB. Debe existir; el modo read_only NO crea la base."
- name: table
desc: "Nombre de la tabla a pivotar. Se interpola citado con dobles comillas (DuckDB no admite parametros para identificadores)."
- name: index
desc: "Columna cuyos valores forman las filas de la pivot (eje vertical)."
- name: columns
desc: "Columna cuyos valores forman las columnas de la pivot (eje horizontal)."
- name: value
desc: "Columna numerica a agregar en cada celda. Ignorada cuando agg='count'."
- name: agg
desc: "Funcion de agregacion: mean, sum, count, min, max, median. mean->avg(), count->COUNT(*). Otro valor devuelve {status:'error'}."
- name: top_rows
desc: "Numero maximo de filas a conservar, elegidas por mayor numero de observaciones (suma de COUNT(*) por valor de index). Default 10."
- name: top_cols
desc: "Numero maximo de columnas a conservar, elegidas por mayor numero de observaciones (suma de COUNT(*) por valor de columns). Default 8."
output: "dict. En exito {status:'ok', index, columns, value, agg, row_labels:[...], col_labels:[...], matrix:[[...]], truncated_rows:bool, truncated_cols:bool, note:str}. matrix tiene len(row_labels) filas y cada fila len(col_labels) celdas (valor agregado o None si la combinacion no existe). truncated_* indica si hubo mas filas/columnas que el top. En error {status:'error', error:str} (no lanza)."
tested: true
tests: ["pivot mean labels y celda conocida", "pivot trunca a top rows y top cols", "pivot count no necesita value real", "pivot db inexistente devuelve error sin lanzar", "pivot agg invalido devuelve error"]
test_file_path: "python/functions/datascience/pivot_table_duckdb_test.py"
file_path: "python/functions/datascience/pivot_table_duckdb.py"
---
## Ejemplo
```python
import duckdb
from datascience import pivot_table_duckdb
# Tabla DuckDB de prueba estilo titanic: sex x pclass -> mean(fare).
db = "/tmp/pivot_demo.duckdb"
con = duckdb.connect(db)
con.execute(
"CREATE TABLE titanic AS SELECT * FROM (VALUES "
"('male',1,211.3),('female',1,151.5),('male',3,7.9),"
"('female',3,16.7),('male',1,52.0),('female',2,41.6)"
") t(sex, pclass, fare)"
)
con.close()
res = pivot_table_duckdb(db, "titanic", index="sex", columns="pclass", value="fare", agg="mean")
print(res["status"]) # ok
print(res["row_labels"]) # ['female', 'male'] (orden por nº de observaciones desc; empate -> etiqueta)
print(res["col_labels"]) # [1, 3, 2] (pclass=1 tiene 3 obs, pclass=3 -> 2, pclass=2 -> 1)
print(res["matrix"]) # [[151.5, 16.7, 41.6], [131.65, 7.9, None]] (male/pclass=2 no existe -> None)
```
## Cuando usarla
Cuando quieres una pivot table (`index` x `columns` -> `agg(value)`) de una tabla
DuckDB con MUCHAS filas y necesitas que el resultado quepa entero en un informe: un
PDF abierto en el movil o un slide PPTX, donde una matriz de 50x30 se cortaria. La
agregacion se hace push-down en el motor (no traes las filas a RAM) y el resultado se
limita a las `top_rows` x `top_cols` combinaciones con mas observaciones. Encaja en el
flujo `eda` para resumir el cruce de dos categoricas (sexo x clase, region x producto)
contra una metrica. Para pivotar un `list[dict]` ya cargado en memoria usa la funcion
pura `pivot_py_datascience`; esta es la version push-down sobre disco.
## Gotchas
- Funcion impura: lee un archivo DuckDB del disco (read_only, nunca lo modifica).
- Recorta a `top_rows` x `top_cols` por numero de observaciones (suma de `COUNT(*)`),
NO por magnitud del valor agregado. Si habia mas filas/columnas, `truncated_rows` /
`truncated_cols` quedan en True y esas combinaciones NO aparecen en la matriz.
- Las celdas sin datos (combinacion `index` x `columns` que no existe en la tabla) se
rellenan con `None`, no con 0: distinguir "cero medido" de "sin observaciones".
- `agg='count'` cuenta filas por celda con `COUNT(*)` e ignora `value` (puedes pasar
cualquier nombre de columna). Para el resto de aggs, `value` debe ser una columna
numerica real o la query fallara con `{status:'error'}`.
- `agg` solo admite mean, sum, count, min, max, median; cualquier otro valor devuelve
`{status:'error'}` sin tocar la base.
- Orden de `row_labels` / `col_labels`: por numero de observaciones descendente, con
desempate estable por etiqueta. No es orden alfabetico ni el de aparicion.
- La query se ejecuta con `sandbox=False` en `duckdb_query_readonly` (uso interno
confiable: el SQL lo construye esta funcion, no un cliente externo).
@@ -1,176 +0,0 @@
"""pivot_table_duckdb — pivot table (index x columns -> agg(value)) con push-down SQL.
Funcion impura: lee de disco a traves de DuckDB reusando la primitiva read-only del
grupo `duckdb` (`duckdb_query_readonly`). Pertenece al grupo de capacidad `eda`
(exploratory data analysis).
A diferencia de la funcion pura `pivot` (que pivota un `list[dict]` ya cargado en
memoria), esta version empuja la agregacion al motor de DuckDB (push-down): el
GROUP BY lo resuelve DuckDB y solo se traen los valores agregados, nunca las filas
crudas. Esto la hace apta para tablas grandes.
Ademas reduce el resultado a las `top_rows` filas y `top_cols` columnas con mas
observaciones, de modo que la pivot quepa entera en un PDF movil / slide PPTX sin
cortarse. Marca `truncated_rows`/`truncated_cols` cuando hubo recorte.
Estilo dict-no-throw del grupo duckdb: nunca lanza; captura cualquier error y
devuelve {status:'error', error:str}.
"""
from collections import defaultdict
from infra import duckdb_query_readonly
# Funciones de agregacion permitidas y su nombre en SQL DuckDB.
# mean -> avg; el resto mapea directo. count se trata aparte (COUNT(*), sin value).
_AGG_SQL = {
"mean": "avg",
"sum": "sum",
"count": "count",
"min": "min",
"max": "max",
"median": "median",
}
def _quote_ident(ident: str) -> str:
"""Cita un identificador SQL con dobles comillas, escapando `"` -> `""`.
DuckDB no admite parametros posicionales para nombres de tabla/columna, asi que
hay que interpolarlos. El quoting con `"` y el doblado de comillas internas evita
que un nombre rompa la sentencia (mismo patron que correlation_matrix_duckdb).
"""
return '"' + str(ident).replace('"', '""') + '"'
def pivot_table_duckdb(
db_path: str,
table: str,
index: str,
columns: str,
value: str,
agg: str = "mean",
top_rows: int = 10,
top_cols: int = 8,
) -> dict:
"""Pivot table push-down en DuckDB, recortada a top_rows x top_cols.
Construye una pivot (filas = valores de `index`, columnas = valores de `columns`,
celda = `agg(value)`) agregando en el motor de DuckDB, y la reduce a las filas y
columnas con mas observaciones para que quepa en un PDF / slide.
Args:
db_path: ruta al archivo DuckDB. Debe existir (read_only NO crea la base).
table: nombre de la tabla a pivotar.
index: columna cuyos valores forman las filas de la pivot.
columns: columna cuyos valores forman las columnas de la pivot.
value: columna numerica a agregar. Ignorada cuando agg="count".
agg: funcion de agregacion. Una de: "mean", "sum", "count", "min", "max",
"median". mean se traduce a avg(); count a COUNT(*).
top_rows: numero maximo de filas a conservar, elegidas por mayor numero de
observaciones (suma de COUNT(*) por valor de index). Default 10.
top_cols: numero maximo de columnas a conservar, elegidas por mayor numero de
observaciones (suma de COUNT(*) por valor de columns). Default 8.
Returns:
dict. En exito:
{status:'ok',
index, columns, value, agg,
row_labels:[...], # valores de index, en orden de freq desc
col_labels:[...], # valores de columns, en orden de freq desc
matrix:[[...], ...], # len == len(row_labels); cada fila
# len == len(col_labels); celda = agg o None
truncated_rows:bool, truncated_cols:bool,
note:str}
En error (sin lanzar): {status:'error', error:str}.
"""
try:
if not isinstance(agg, str) or agg not in _AGG_SQL:
return {
"status": "error",
"error": "invalid agg "
+ repr(agg)
+ "; allowed: "
+ ", ".join(sorted(_AGG_SQL)),
}
# Paso 1 (push-down): agregar (index, columns) -> agg(value) + COUNT(*).
if agg == "count":
agg_expr = "COUNT(*)"
else:
agg_expr = f"{_AGG_SQL[agg]}({_quote_ident(value)})"
sql = (
f"SELECT {_quote_ident(index)} AS r, "
f"{_quote_ident(columns)} AS c, "
f"{agg_expr} AS v, "
f"COUNT(*) AS n "
f"FROM {_quote_ident(table)} "
f"GROUP BY {_quote_ident(index)}, {_quote_ident(columns)}"
)
# max_rows alto: queremos todos los grupos (index x columns) para elegir el
# top con criterio global. sandbox=False igual que correlation_matrix_duckdb,
# porque db_path es una ruta interna de confianza.
result = duckdb_query_readonly(
db_path, sql, max_rows=1_000_000, sandbox=False
)
if result.get("status") != "ok":
return {
"status": "error",
"error": "pivot query failed: "
+ str(result.get("error", "unknown")),
}
# Paso 2 (en python): contar observaciones por fila y por columna, y guardar
# el valor agregado de cada celda (r, c).
row_obs: dict = defaultdict(int)
col_obs: dict = defaultdict(int)
cell: dict = {}
for row in result.get("rows", []):
r = row.get("r")
c = row.get("c")
n = row.get("n") or 0
row_obs[r] += n
col_obs[c] += n
cell[(r, c)] = row.get("v")
def _top(obs: dict, limit: int):
# Orden: mas observaciones primero; desempate estable por etiqueta.
ranked = sorted(obs.items(), key=lambda kv: (-kv[1], str(kv[0])))
selected = [label for label, _ in ranked[:limit]]
return selected, len(ranked) > limit
row_labels, truncated_rows = _top(row_obs, top_rows)
col_labels, truncated_cols = _top(col_obs, top_cols)
# Paso 3: materializar la matriz; None donde la combinacion no existe.
matrix = [
[cell.get((r, c)) for c in col_labels] for r in row_labels
]
note = (
f"pivot {agg}({value}) reducida a {len(row_labels)}x{len(col_labels)} "
"(top por observaciones) para caber en PDF/slide"
)
if agg == "count":
note = (
f"pivot count(*) reducida a {len(row_labels)}x{len(col_labels)} "
"(top por observaciones) para caber en PDF/slide"
)
return {
"status": "ok",
"index": index,
"columns": columns,
"value": value,
"agg": agg,
"row_labels": row_labels,
"col_labels": col_labels,
"matrix": matrix,
"truncated_rows": truncated_rows,
"truncated_cols": truncated_cols,
"note": note,
}
except Exception as e: # noqa: BLE001
return {"status": "error", "error": str(e)}
@@ -1,115 +0,0 @@
"""Tests para pivot_table_duckdb."""
import os
import sys
import duckdb
# Permitir importar funciones del registry (from infra import ..., from datascience import ...).
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", ".."))
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "functions"))
from datascience.pivot_table_duckdb import pivot_table_duckdb
def _make_db(tmp_name: str) -> str:
"""Crea una DuckDB con dos categoricas (a, b) y un valor numerico conocido.
Filas:
a='x', b='y', val=10
a='x', b='y', val=20 -> mean(x,y) = 15, count(x,y) = 2
a='x', b='z', val=5 -> mean(x,z) = 5
a='w', b='y', val=100 -> mean(w,y) = 100
Observaciones por a: x=3, w=1. Por b: y=3, z=1.
La combinacion (w, z) no existe -> celda None.
"""
db = os.path.join("/tmp", tmp_name)
if os.path.exists(db):
os.remove(db)
con = duckdb.connect(db)
con.execute("CREATE TABLE t (a VARCHAR, b VARCHAR, val DOUBLE)")
con.execute(
"INSERT INTO t VALUES "
"('x','y',10),('x','y',20),('x','z',5),('w','y',100)"
)
con.close()
return db
def test_pivot_mean_labels_y_celda_conocida():
db = _make_db("pivot_test_mean.duckdb")
res = pivot_table_duckdb(db, "t", index="a", columns="b", value="val", agg="mean")
assert res["status"] == "ok", res
# Filas ordenadas por observaciones desc: x (3) antes que w (1).
assert res["row_labels"] == ["x", "w"], res["row_labels"]
# Columnas ordenadas por observaciones desc: y (3) antes que z (1).
assert res["col_labels"] == ["y", "z"], res["col_labels"]
# matrix[0][0] = mean(a='x', b='y') = (10 + 20) / 2 = 15.
assert abs(res["matrix"][0][0] - 15.0) < 1e-9, res["matrix"]
# matrix[0][1] = mean(a='x', b='z') = 5.
assert abs(res["matrix"][0][1] - 5.0) < 1e-9, res["matrix"]
# matrix[1][0] = mean(a='w', b='y') = 100.
assert abs(res["matrix"][1][0] - 100.0) < 1e-9, res["matrix"]
# (w, z) no existe -> None.
assert res["matrix"][1][1] is None, res["matrix"]
# Sin truncado con los defaults (top_rows=10, top_cols=8).
assert res["truncated_rows"] is False
assert res["truncated_cols"] is False
# La matriz es rectangular consistente con las etiquetas.
assert len(res["matrix"]) == len(res["row_labels"])
for fila in res["matrix"]:
assert len(fila) == len(res["col_labels"])
def test_pivot_trunca_a_top_rows_y_top_cols():
db = _make_db("pivot_test_trunc.duckdb")
res = pivot_table_duckdb(
db, "t", index="a", columns="b", value="val", agg="mean",
top_rows=1, top_cols=1,
)
assert res["status"] == "ok", res
# Solo la fila/columna mas frecuente sobrevive.
assert res["row_labels"] == ["x"], res["row_labels"]
assert res["col_labels"] == ["y"], res["col_labels"]
assert res["matrix"] == [[15.0]], res["matrix"]
# Habia mas de 1 fila y mas de 1 columna -> truncado en ambos ejes.
assert res["truncated_rows"] is True
assert res["truncated_cols"] is True
def test_pivot_count_no_necesita_value_real():
db = _make_db("pivot_test_count.duckdb")
# value apunta a una columna real pero count(*) la ignora; tambien valdria un
# nombre cualquiera. Verificamos que count funciona igualmente.
res = pivot_table_duckdb(
db, "t", index="a", columns="b", value="val", agg="count"
)
assert res["status"] == "ok", res
assert res["row_labels"] == ["x", "w"]
assert res["col_labels"] == ["y", "z"]
# count(a='x', b='y') = 2 observaciones.
assert res["matrix"][0][0] == 2, res["matrix"]
# count(a='x', b='z') = 1.
assert res["matrix"][0][1] == 1, res["matrix"]
# count(a='w', b='y') = 1.
assert res["matrix"][1][0] == 1, res["matrix"]
# (w, z) no existe -> None.
assert res["matrix"][1][1] is None, res["matrix"]
def test_pivot_db_inexistente_devuelve_error_sin_lanzar():
res = pivot_table_duckdb(
"/nonexistent/path/does_not_exist.duckdb",
"t", index="a", columns="b", value="val", agg="mean",
)
assert res["status"] == "error", res
assert isinstance(res["error"], str)
def test_pivot_agg_invalido_devuelve_error():
db = _make_db("pivot_test_badagg.duckdb")
res = pivot_table_duckdb(
db, "t", index="a", columns="b", value="val", agg="stddev"
)
assert res["status"] == "error", res
assert "invalid agg" in res["error"]
@@ -0,0 +1,79 @@
---
name: profile_datetime
kind: function
lang: py
domain: datascience
version: "1.0.0"
purity: pure
signature: "def profile_datetime(values: list) -> dict"
description: "Perfil minimo de una columna fecha/datetime para la cabecera del capitulo TIMESERIES de AutomaticEDA. Acepta datetime.date, datetime.datetime y strings ISO mezclados, parsea defensivamente e ignora lo no parseable (nunca lanza). Devuelve rango (min/max ISO), n, n_distinct, span_days, frecuencia inferida (daily/weekly/monthly/quarterly/yearly/irregular/unknown) a partir del paso mediano entre fechas distintas, is_regular (pasos ~constantes), n_gaps (huecos en la rejilla) y median_step_days. Solo stdlib (datetime + statistics)."
tags: [statistics, timeseries, datetime, profiling, frequency, eda, automatic_eda, python]
uses_functions: []
uses_types: []
returns: []
returns_optional: false
error_type: ""
imports: [datetime, statistics]
params:
- name: values
desc: "lista de valores fecha. Acepta datetime.date, datetime.datetime y strings ISO ('2021-06-28', '2021-06-28T00:00:00', '2021-06-28 12:00:00'). None, vacios y no parseables se ignoran; tz-aware se normaliza a naive. Si values es None o no iterable se trata como lista vacia."
output: "dict SIEMPRE presente con: 'min'/'max' (ISO date YYYY-MM-DD o None), 'n' (valores parseables), 'n_distinct' (fechas unicas), 'span_days' (float o None), 'freq' (daily|weekly|monthly|quarterly|yearly|irregular|unknown), 'is_regular' (bool), 'n_gaps' (int), 'median_step_days' (float o None) y 'note' (str). Con <2 valores o una sola fecha distinta: freq='unknown', is_regular=False, n_gaps=0, median_step_days=None y nota. Nunca lanza."
tested: true
tests: ["test_serie_diaria_regular_golden", "test_serie_mensual_freq_monthly", "test_serie_con_hueco_cuenta_gaps", "test_strings_iso_mezclados_con_datetime", "test_lista_vacia_y_none_devuelve_unknown", "test_valores_no_parseables_ignorados", "test_span_days_correcto", "test_una_sola_fecha_es_coherente"]
test_file_path: "python/functions/datascience/profile_datetime_test.py"
file_path: "python/functions/datascience/profile_datetime.py"
---
## Ejemplo
```python
from datascience import profile_datetime
from datetime import date, datetime, timedelta
# Serie diaria regular de 30 dias
fechas = [date(2021, 1, 1) + timedelta(days=i) for i in range(30)]
res = profile_datetime(fechas)
res["freq"] # -> "daily"
res["is_regular"] # -> True
res["n_gaps"] # -> 0
res["min"], res["max"] # -> ("2021-01-01", "2021-01-30")
res["span_days"] # -> 29.0
# Acepta strings ISO mezclados con objetos datetime/date; ignora lo no parseable
profile_datetime(["2021-06-28", datetime(2021, 6, 29, 12), "basura", None])["n"] # -> 2
# Columna vacia o sin fechas validas
profile_datetime([])["freq"] # -> "unknown" + note "datos insuficientes"
```
## Cuando usarla
Cuando construyes la cabecera del capitulo TIMESERIES de un EDA y necesitas
caracterizar la columna de fecha antes de modelar: que rango cubre, cada cuanto
llegan los datos (frecuencia), si la cadencia es regular y si hay huecos en la
rejilla temporal. Es el complemento de fecha al perfil numerico/categorico del
TableProfile (cierra el `datetime{}=None` pendiente). Pasale la columna de fechas
en bruto (tal cual venga de la BD: dates, datetimes o strings ISO) y usa `freq` +
`is_regular` + `n_gaps` para decidir si conviene resamplear, rellenar huecos o
desestacionalizar mas adelante.
## Gotchas
- Es pura y stdlib-only, pero la inferencia de `freq` es heuristica por bandas
sobre el **paso mediano entre fechas distintas** (se deduplica antes de medir).
Cualquier paso fuera de las bandas conocidas (incluido sub-diario, p.ej. datos
horarios) cae en `"irregular"`: no hay banda hourly.
- El analisis de frecuencia/regularidad/huecos necesita **>=2 fechas distintas**.
Con 0-1 valores parseables o una sola fecha unica, `freq="unknown"`,
`median_step_days=None` y `n_gaps=0`, pero `min`/`max`/`span_days` siguen siendo
coherentes si hay al menos una fecha.
- `min`/`max` se reportan como ISO **date** (`YYYY-MM-DD`); la hora se conserva
internamente para calcular `span_days` y `median_step_days` (que pueden ser
fraccionarios con datetimes sub-diarios) pero no aparece en min/max.
- Los datetime con zona horaria se normalizan a naive (se descarta el tzinfo) para
poder mezclarlos con fechas naive sin que las restas lancen; esto puede desplazar
la fecha en datetimes con offset grande. Para EDA es despreciable.
- `is_regular` usa tolerancia ±25% sobre el paso mediano y umbral del 80% de los
pasos dentro de banda; series de "primero de mes" (deltas 28-31) salen regulares.
- `n_gaps` solo se calcula cuando `freq` es una rejilla regular conocida; con
`freq` `"irregular"` o `"unknown"` siempre es 0.
@@ -0,0 +1,183 @@
"""Perfil minimo de una columna fecha/datetime para la cabecera TIMESERIES (grupo eda).
Funcion pura y determinista que resume una columna temporal: rango (min/max),
numero de fechas distintas, frecuencia inferida (daily/weekly/monthly/quarterly/
yearly/irregular), regularidad de los pasos, huecos respecto a la rejilla inferida
y paso mediano entre fechas consecutivas. Cierra el `datetime{}=None` que hoy deja
pendiente el TableProfile de AutomaticEDA.
Acepta valores heterogeneos (``datetime.date``, ``datetime.datetime`` y strings
ISO como ``"2021-06-28"``, ``"2021-06-28T00:00:00"`` o ``"2021-06-28 12:00:00"``),
parsea de forma defensiva, ignora lo que no se puede parsear y NUNCA lanza.
Solo usa stdlib (``datetime`` + ``statistics``).
"""
from __future__ import annotations
import statistics
from datetime import date, datetime
def _parse_one(v) -> datetime | None:
"""Parsea un valor a ``datetime`` naive, o devuelve None si no es una fecha.
Acepta ``datetime.datetime``, ``datetime.date`` y strings ISO. Cualquier
datetime con zona horaria se normaliza a naive (se descarta el tzinfo) para
poder mezclarlo con fechas naive sin que las restas lancen ``TypeError``.
"""
if v is None or isinstance(v, bool):
return None
# datetime es subclase de date: comprobar datetime primero.
if isinstance(v, datetime):
return v.replace(tzinfo=None)
if isinstance(v, date):
return datetime(v.year, v.month, v.day)
if isinstance(v, str):
s = v.strip()
if not s:
return None
try:
dt = datetime.fromisoformat(s)
except ValueError:
return None
return dt.replace(tzinfo=None)
return None
def _infer_freq(median_step_days: float) -> str:
"""Clasifica la frecuencia a partir del paso mediano (en dias) entre fechas.
Bandas con tolerancia: ~1 dia -> daily, ~7 -> weekly, 28-31 -> monthly,
89-92 -> quarterly, 360-366 -> yearly. Cualquier paso fuera de las bandas
(incluido sub-diario) -> irregular.
"""
m = median_step_days
if 0.5 <= m <= 1.5:
return "daily"
if 6.0 <= m <= 8.0:
return "weekly"
if 28.0 <= m <= 31.0:
return "monthly"
if 89.0 <= m <= 92.0:
return "quarterly"
if 360.0 <= m <= 366.0:
return "yearly"
return "irregular"
def profile_datetime(values: list) -> dict:
"""Perfila una columna de fechas para la cabecera del capitulo TIMESERIES.
Funcion pura y determinista: no hace I/O, no muta el input y nunca lanza.
El analisis de frecuencia, regularidad y huecos se hace sobre las **fechas
distintas ordenadas** (se deduplica antes de calcular los pasos): los valores
repetidos generarian pasos de 0 dias que distorsionarian el mediano y la
inferencia. ``n`` cuenta los valores parseables (con duplicados) y
``n_distinct`` las fechas unicas.
Args:
values: lista de valores fecha. Acepta ``datetime.date``,
``datetime.datetime`` y strings ISO (``"2021-06-28"``,
``"2021-06-28T00:00:00"``, ``"2021-06-28 12:00:00"``). Los valores
None, vacios o no parseables se ignoran. Si ``values`` es None o no
iterable se trata como lista vacia.
Returns:
Siempre un dict con esta forma::
{
"min": str | None, # fecha minima ISO date (YYYY-MM-DD)
"max": str | None, # fecha maxima ISO date
"n": int, # nº de valores fecha parseables
"n_distinct": int, # nº de fechas distintas
"span_days": float | None, # (max - min) en dias
"freq": str, # daily|weekly|monthly|quarterly|
# yearly|irregular|unknown
"is_regular": bool, # pasos ~constantes (tolerancia ±25%)
"n_gaps": int, # saltos > ~1.5x el paso mediano
"median_step_days": float | None, # paso mediano entre fechas
"note": str # "" o nota corta
}
Con menos de 2 valores parseables (o una sola fecha distinta) devuelve
``freq="unknown"``, ``is_regular=False``, ``n_gaps=0``,
``median_step_days=None`` y la nota correspondiente, manteniendo min/max
y span_days coherentes cuando hay al menos una fecha.
"""
base = {
"min": None,
"max": None,
"n": 0,
"n_distinct": 0,
"span_days": None,
"freq": "unknown",
"is_regular": False,
"n_gaps": 0,
"median_step_days": None,
"note": "",
}
if values is None:
values = []
try:
iterator = list(values)
except TypeError:
iterator = []
parsed: list[datetime] = []
for v in iterator:
dt = _parse_one(v)
if dt is not None:
parsed.append(dt)
n = len(parsed)
base["n"] = n
if n == 0:
base["note"] = "datos insuficientes"
return base
distinct = sorted(set(parsed))
n_distinct = len(distinct)
dt_min = min(parsed)
dt_max = max(parsed)
base["n_distinct"] = n_distinct
base["min"] = dt_min.date().isoformat()
base["max"] = dt_max.date().isoformat()
base["span_days"] = round((dt_max - dt_min).total_seconds() / 86400.0, 6)
# Sin al menos dos fechas distintas no hay pasos que medir.
if n_distinct < 2:
base["note"] = "datos insuficientes" if n < 2 else "una sola fecha distinta"
return base
steps = [
(distinct[i + 1] - distinct[i]).total_seconds() / 86400.0
for i in range(n_distinct - 1)
]
median_step = float(statistics.median(steps))
base["median_step_days"] = round(median_step, 6)
freq = _infer_freq(median_step)
base["freq"] = freq
# Regularidad: >=80% de los pasos dentro de ±25% del paso mediano.
if median_step > 0:
tol = 0.25 * median_step
within = sum(1 for s in steps if abs(s - median_step) <= tol)
base["is_regular"] = (within / len(steps)) >= 0.8
else:
base["is_regular"] = False
# Huecos: pasos que superan ~1.5x el mediano. Solo tiene sentido cuando la
# frecuencia es una rejilla regular conocida (no irregular/unknown).
if freq not in ("unknown", "irregular") and median_step > 0:
threshold = 1.5 * median_step
base["n_gaps"] = sum(1 for s in steps if s > threshold)
else:
base["n_gaps"] = 0
return base
@@ -0,0 +1,127 @@
"""Tests para profile_datetime."""
from datetime import date, datetime, timedelta
from profile_datetime import profile_datetime
def test_serie_diaria_regular_golden():
# 30 dias consecutivos: frecuencia diaria, regular, sin huecos.
fechas = [date(2021, 1, 1) + timedelta(days=i) for i in range(30)]
res = profile_datetime(fechas)
assert res["n"] == 30
assert res["n_distinct"] == 30
assert res["min"] == "2021-01-01"
assert res["max"] == "2021-01-30"
assert res["span_days"] == 29.0
assert res["freq"] == "daily"
assert res["is_regular"] is True
assert res["n_gaps"] == 0
assert res["median_step_days"] == 1.0
assert res["note"] == ""
def test_serie_mensual_freq_monthly():
# Primero de mes durante 14 meses: paso mediano ~30/31 dias -> monthly.
fechas = []
y, m = 2021, 1
for _ in range(14):
fechas.append(date(y, m, 1))
m += 1
if m > 12:
m = 1
y += 1
res = profile_datetime(fechas)
assert res["n"] == 14
assert res["freq"] == "monthly"
assert res["min"] == "2021-01-01"
assert res["max"] == "2022-02-01"
assert 28.0 <= res["median_step_days"] <= 31.0
def test_serie_con_hueco_cuenta_gaps():
# Serie diaria con un hueco de 3 dias (faltan i=7,8,9) -> n_gaps >= 1.
fechas = [
date(2021, 1, 1) + timedelta(days=i)
for i in range(20)
if i not in (7, 8, 9)
]
res = profile_datetime(fechas)
assert res["freq"] == "daily"
assert res["n_gaps"] >= 1
assert res["median_step_days"] == 1.0
def test_strings_iso_mezclados_con_datetime():
# Mezcla de strings ISO (varios formatos) y objetos datetime/date.
valores = [
"2021-06-28",
datetime(2021, 6, 29, 12, 0, 0),
"2021-06-30T00:00:00",
date(2021, 7, 1),
]
res = profile_datetime(valores)
assert res["n"] == 4
assert res["n_distinct"] == 4
assert res["min"] == "2021-06-28"
assert res["max"] == "2021-07-01"
assert res["freq"] == "daily"
assert res["note"] == ""
def test_lista_vacia_y_none_devuelve_unknown():
for entrada in ([], None):
res = profile_datetime(entrada)
assert res["n"] == 0
assert res["n_distinct"] == 0
assert res["min"] is None
assert res["max"] is None
assert res["span_days"] is None
assert res["freq"] == "unknown"
assert res["is_regular"] is False
assert res["n_gaps"] == 0
assert res["median_step_days"] is None
assert res["note"] == "datos insuficientes"
def test_valores_no_parseables_ignorados():
# Strings basura, None, ints y un date valido mezclados: ignora lo no fecha.
valores = [
"no es una fecha",
None,
"2021-01-01",
"2021-01-02",
12345,
"tampoco",
date(2021, 1, 3),
"",
]
res = profile_datetime(valores)
assert res["n"] == 3 # solo 3 fechas parseables
assert res["n_distinct"] == 3
assert res["freq"] == "daily"
assert res["min"] == "2021-01-01"
assert res["max"] == "2021-01-03"
def test_span_days_correcto():
# Dos fechas a un anio de distancia: span 365 dias -> yearly.
res = profile_datetime([date(2020, 1, 1), date(2020, 12, 31)])
assert res["n"] == 2
assert res["n_distinct"] == 2
assert res["span_days"] == 365.0
assert res["median_step_days"] == 365.0
assert res["freq"] == "yearly"
def test_una_sola_fecha_es_coherente():
# Un unico valor: min == max, span 0, freq unknown, nota datos insuficientes.
res = profile_datetime(["2021-06-28"])
assert res["n"] == 1
assert res["n_distinct"] == 1
assert res["min"] == "2021-06-28"
assert res["max"] == "2021-06-28"
assert res["span_days"] == 0.0
assert res["freq"] == "unknown"
assert res["median_step_days"] is None
assert res["note"] == "datos insuficientes"
@@ -0,0 +1,72 @@
---
name: resample_timeseries
kind: function
lang: py
domain: datascience
version: "1.0.0"
purity: pure
signature: "def resample_timeseries(t: list, v: list, freq: str = \"auto\", agg: str = \"mean\", max_points: int = 400) -> dict"
description: "Agrega una serie temporal por periodo para graficar su evolucion y el CONTEO de observaciones por bucket. Nucleo del capitulo TIMESERIES de AutomaticEDA (grupo eda): recibe las fechas y los valores YA leidos (pura, sin tocar ninguna base de datos), empareja t[i] con v[i] por indice, parsea fechas defensivamente, trunca cada fecha al inicio de su bucket (daily/weekly/monthly/quarterly/yearly), y agrega los valores numericos validos por bucket mientras cuenta TODAS las observaciones con fecha valida (densidad temporal, incluida la fila cuyo valor es None). freq='auto' infiere del delta mediano entre fechas. Si hay mas buckets que max_points hace downsampling uniforme conservando primero y ultimo. Estilo dict-no-throw: NUNCA lanza; entrada vacia o longitudes incompatibles devuelve listas vacias + note='datos insuficientes'."
tags: [eda, timeseries, resample, aggregate, profiling, datascience, time]
params:
- name: t
desc: "Lista de fechas paralela a v. Acepta strings ISO ('YYYY-MM-DD' o 'YYYY-MM-DDTHH:MM:SS', con 'Z' opcional), datetime.date o datetime.datetime. Se parsea defensivamente; los pares cuya fecha no parsea se descartan junto con su valor."
- name: v
desc: "Lista de valores numericos (float/int) paralela a t. Puede contener None o valores no numericos: se ignoran en la agregacion pero la fila sigue contando en 'count' si su fecha es valida. bool, NaN e Inf se tratan como no numericos."
- name: freq
desc: "Granularidad del bucket: 'auto' (infiere del delta mediano en dias entre fechas: <=3 daily, <=16 weekly, <=75 monthly, <=200 quarterly, mayor yearly) o explicita en {daily, weekly, monthly, quarterly, yearly}. Una frecuencia desconocida cae a 'auto'."
- name: agg
desc: "Agregacion por bucket sobre los valores numericos validos: 'mean' | 'sum' | 'median' | 'last' (valor de la observacion cronologicamente mas reciente del bucket) | 'min' | 'max'. Una agregacion desconocida cae a 'mean'."
- name: max_points
desc: "Tope de buckets en la salida. Si n_buckets > max_points hace downsampling uniforme (1 de cada k buckets equiespaciados, conservando el primero y el ultimo) para no saturar el grafico del PDF/PPTX. max_points<=0 desactiva el limite."
output: "Dict siempre con las mismas claves: t (lista de etiquetas ISO 'YYYY-MM-DD' por bucket, orden cronologico), v (lista paralela del valor agregado por bucket segun agg; None si el bucket no tiene ningun valor numerico valido), count (lista paralela del nº de observaciones con fecha valida por bucket), freq (frecuencia efectivamente usada), agg (agregacion usada), n_in (nº de pares (t,v) con fecha valida que entraron), n_buckets (nº de buckets antes del downsample), downsampled (bool, True si se aplico downsampling), note ('' o 'datos insuficientes' cuando no hay pares validos / longitudes incompatibles / listas vacias). Numericos de v en float, count en int."
uses_functions: []
uses_types: []
returns: []
returns_optional: false
error_type: ""
imports: []
tested: true
tests: ["test_daily_a_mensual_mean", "test_agg_sum_y_last", "test_count_cuenta_observacion_con_valor_none", "test_downsampling_respeta_max_points_y_extremos", "test_freq_auto_infiere_mensual", "test_edge_listas_vacias_o_desiguales"]
test_file_path: "python/functions/datascience/resample_timeseries_test.py"
file_path: "python/functions/datascience/resample_timeseries.py"
---
## Ejemplo
```python
import sys, os
sys.path.insert(0, os.path.join("python", "functions"))
from datascience.resample_timeseries import resample_timeseries
# Serie diaria agregada a buckets mensuales: media del valor + conteo de filas.
t = ["2020-01-01", "2020-01-15", "2020-02-01", "2020-02-10", "2020-02-20"]
v = [10.0, 20.0, 30.0, 40.0, 50.0]
r = resample_timeseries(t, v, freq="monthly", agg="mean")
print(r["t"]) # ['2020-01-01', '2020-02-01']
print(r["v"]) # [15.0, 40.0]
print(r["count"]) # [2, 3] <- densidad: nº de observaciones por mes
print(r["freq"], r["downsampled"]) # monthly False
# freq='auto' infiere la granularidad del delta mediano entre fechas.
mensual = [f"2022-{m:02d}-01" for m in range(1, 13)]
print(resample_timeseries(mensual, list(range(1, 13)))["freq"]) # monthly
```
## Cuando usarla
- Usala en el capitulo TIMESERIES de `AutomaticEDA` para construir, a partir de una columna temporal (`detect_time_column`) y una columna numerica, la doble serie que el renderer dibuja: la EVOLUCION del valor agregado por periodo y el CONTEO de observaciones por periodo.
- Cuando ya tengas las fechas y los valores leidos en memoria (de DuckDB, polars, CSV, etc.) y solo necesites agregarlos por dia/semana/mes/trimestre/año sin volver a tocar la base de datos — esta funcion es pura y recibe los datos por parametro.
- Cuando quieras un downsampling controlado para que una serie muy larga (miles de fechas) quepa en un grafico de un PDF/PPTX sin saturarlo, conservando el primer y el ultimo punto.
- Cuando no sepas la cadencia de la serie: pasa `freq="auto"` y deja que la infiera del delta mediano.
## Gotchas
- Funcion pura, sin I/O y determinista. NUNCA lanza: ante entrada invalida (listas vacias, longitudes distintas o todas las fechas no parseables) devuelve listas vacias + `note="datos insuficientes"`.
- `count` cuenta OBSERVACIONES con fecha valida en el bucket (densidad temporal), aunque su valor numerico sea `None`/no numerico. `v` agrega SOLO los valores numericos validos del bucket; si no hay ninguno, `v` del bucket es `None` mientras `count` sigue reflejando las filas. No confundas `count` (filas) con el nº de valores agregados.
- `bool`, `NaN` e `Inf` se tratan como NO numericos (se ignoran en `v`). Un string que no parsea a numero tambien se ignora en `v` pero su fila cuenta si la fecha es valida.
- El truncado de bucket usa el inicio del periodo: semana = lunes ISO (`weekday()==0`), mes = dia 1, trimestre = primer dia del trimestre (ene/abr/jul/oct), año = 1 de enero. La etiqueta de cada bucket es esa fecha de inicio en ISO `YYYY-MM-DD`, no un rango.
- El downsampling (`n_buckets > max_points`) reduce la salida a `<= max_points` puntos equiespaciados conservando primero y ultimo, pero `n_buckets` SIEMPRE reporta el conteo real previo al recorte. Si necesitas todos los buckets, sube `max_points` o ponlo `<=0`.
- Las fechas con hora se truncan a su `date()` antes de agrupar: la granularidad minima es el dia (no hay buckets horarios).
- `freq` desconocida o no-string cae a `"auto"`; `agg` desconocida cae a `"mean"`. El campo devuelto refleja la opcion efectivamente usada.
@@ -0,0 +1,275 @@
"""Agrega una serie temporal por periodo para el capitulo TIMESERIES (grupo eda).
Funcion pura y determinista: recibe las fechas y los valores YA leidos (nunca
toca una base de datos ni hace I/O) y los agrega por bucket temporal para poder
graficar la evolucion de la serie y, en paralelo, el CONTEO de observaciones por
periodo (densidad temporal).
Estilo "dict-no-throw" del grupo eda: NUNCA lanza excepcion, siempre devuelve el
mismo conjunto de claves. Lectura y parseo de fechas 100% defensivos. Solo usa la
libreria estandar (``datetime``, ``statistics``, ``re``).
"""
from __future__ import annotations
import datetime
import re
import statistics
# Frecuencias soportadas, de mas fina a mas gruesa.
_FREQS = ("daily", "weekly", "monthly", "quarterly", "yearly")
# Agregaciones soportadas.
_AGGS = ("mean", "sum", "median", "last", "min", "max")
# Acepta el inicio de una fecha ISO con cualquier separador posterior
# (incluido un caracter raro entre la fecha y la hora).
_DATE_RE = re.compile(r"(\d{4})-(\d{2})-(\d{2})")
def _to_date(x) -> "datetime.date | None":
"""Parsea defensivamente un valor a ``datetime.date``; devuelve None si falla."""
if x is None:
return None
# datetime es subclase de date: comprobarlo primero.
if isinstance(x, datetime.datetime):
return x.date()
if isinstance(x, datetime.date):
return x
s = str(x).strip()
if not s:
return None
# Camino feliz: ISO completo (con o sin hora, con o sin 'Z' final).
try:
s2 = s[:-1] if s.endswith("Z") else s
return datetime.datetime.fromisoformat(s2).date()
except ValueError:
pass
# Fallback robusto: extrae el prefijo YYYY-MM-DD con cualquier separador.
m = _DATE_RE.match(s)
if m:
try:
return datetime.date(int(m.group(1)), int(m.group(2)), int(m.group(3)))
except ValueError:
return None
return None
def _to_number(x) -> "float | None":
"""Convierte a float si es numerico finito; devuelve None en otro caso."""
if x is None:
return None
if isinstance(x, bool):
# bool es subclase de int: lo tratamos como no-numerico para una serie.
return None
try:
f = float(x)
except (TypeError, ValueError):
return None
# Descarta NaN / Inf (no agregables de forma estable).
if f != f or f in (float("inf"), float("-inf")):
return None
return f
def _infer_freq(dates_sorted: list) -> str:
"""Infiere la frecuencia desde el delta mediano (en dias) entre fechas."""
if len(dates_sorted) < 2:
return "daily"
diffs = [
(dates_sorted[i + 1] - dates_sorted[i]).days
for i in range(len(dates_sorted) - 1)
]
diffs = [d for d in diffs if d > 0] # ignora duplicados del mismo dia
if not diffs:
return "daily"
med = statistics.median(diffs)
if med <= 3:
return "daily"
if med <= 16:
return "weekly"
if med <= 75:
return "monthly"
if med <= 200:
return "quarterly"
return "yearly"
def _bucket_start(d: "datetime.date", freq: str) -> "datetime.date":
"""Trunca una fecha al inicio de su bucket segun la frecuencia."""
if freq == "weekly":
return d - datetime.timedelta(days=d.weekday()) # lunes ISO
if freq == "monthly":
return datetime.date(d.year, d.month, 1)
if freq == "quarterly":
first_month = ((d.month - 1) // 3) * 3 + 1
return datetime.date(d.year, first_month, 1)
if freq == "yearly":
return datetime.date(d.year, 1, 1)
return d # daily (o cualquier otra cosa): la propia fecha
def _downsample_indices(n: int, max_points: int) -> list:
"""Indices equiespaciados conservando primero y ultimo (<= max_points)."""
if max_points <= 0 or max_points >= n:
return list(range(n))
if max_points == 1:
return [0]
idx = sorted({round(i * (n - 1) / (max_points - 1)) for i in range(max_points)})
return idx
def _empty(freq_req: str, agg: str) -> dict:
"""Resultado canonico cuando no hay datos suficientes."""
eff_freq = freq_req if freq_req in _FREQS else "auto"
return {
"t": [],
"v": [],
"count": [],
"freq": eff_freq,
"agg": agg if agg in _AGGS else "mean",
"n_in": 0,
"n_buckets": 0,
"downsampled": False,
"note": "datos insuficientes",
}
def resample_timeseries(
t: list,
v: list,
freq: str = "auto",
agg: str = "mean",
max_points: int = 400,
) -> dict:
"""Agrega una serie temporal por periodo (buckets) para graficarla.
Empareja ``t[i]`` con ``v[i]`` por indice, descarta los pares cuya fecha no
parsea, trunca cada fecha al inicio de su bucket segun ``freq`` y agrupa. Por
cada bucket devuelve el valor agregado (``agg`` sobre los valores numericos
validos) y el CONTEO de observaciones con fecha valida (densidad temporal),
independientemente de si su valor numerico es ``None``.
Funcion pura: no hace I/O, no muta los inputs, es determinista, NUNCA lanza.
Args:
t: lista de fechas paralela a ``v``. Acepta strings ISO
(``"YYYY-MM-DD"`` o ``"YYYY-MM-DDTHH:MM:SS"``, con ``Z`` opcional),
``datetime.date`` o ``datetime.datetime``. Se parsea defensivamente;
las fechas que no parsean se descartan junto con su valor.
v: lista de valores numericos (float/int). Puede contener ``None`` o
valores no numericos: estos se ignoran en la agregacion, pero la fila
sigue contando en ``count`` (siempre que su fecha sea valida).
freq: ``"auto"`` (infiere del delta mediano entre fechas) o uno de
``"daily"``, ``"weekly"``, ``"monthly"``, ``"quarterly"``,
``"yearly"``. Una frecuencia desconocida cae a ``"auto"``.
agg: agregacion por bucket: ``"mean"``, ``"sum"``, ``"median"``,
``"last"`` (valor de la observacion cronologicamente mas reciente),
``"min"`` o ``"max"``. Una agregacion desconocida cae a ``"mean"``.
max_points: si tras agregar hay mas buckets que este limite, se hace
downsampling uniforme (1 de cada k buckets equiespaciados,
conservando el primero y el ultimo) para no saturar el grafico.
Returns:
Siempre un dict con las mismas claves::
{
"t": [str, ...], # etiqueta ISO YYYY-MM-DD de cada bucket
"v": [float|None, ...], # valor agregado por bucket (None si vacio)
"count": [int, ...], # nº de observaciones con fecha valida
"freq": str, # frecuencia efectivamente usada
"agg": str, # agregacion usada
"n_in": int, # nº de pares (t,v) con fecha valida
"n_buckets": int, # nº de buckets antes del downsample
"downsampled": bool, # True si se aplico downsampling
"note": str, # "" o nota (p.ej. "datos insuficientes")
}
"""
agg = agg if agg in _AGGS else "mean"
freq_req = freq if isinstance(freq, str) else "auto"
# Validacion de entrada: deben ser listas de igual longitud y no vacias.
if (
not isinstance(t, list)
or not isinstance(v, list)
or len(t) == 0
or len(t) != len(v)
):
return _empty(freq_req, agg)
# Empareja por indice y descarta fechas no parseables.
parsed: list = [] # (date, original_index, number_or_None)
for i, (ti, vi) in enumerate(zip(t, v)):
d = _to_date(ti)
if d is None:
continue
parsed.append((d, i, _to_number(vi)))
n_in = len(parsed)
if n_in == 0:
return _empty(freq_req, agg)
# Resuelve la frecuencia efectiva.
if freq_req in _FREQS:
eff_freq = freq_req
else:
dates_sorted = sorted(d for d, _, _ in parsed)
eff_freq = _infer_freq(dates_sorted)
# Agrupa por bucket.
buckets: dict = {}
for d, idx, num in parsed:
b = _bucket_start(d, eff_freq)
slot = buckets.get(b)
if slot is None:
slot = {"count": 0, "vals": [], "last_key": None, "last_val": None}
buckets[b] = slot
slot["count"] += 1
if num is not None:
slot["vals"].append(num)
key = (d, idx)
if slot["last_key"] is None or key > slot["last_key"]:
slot["last_key"] = key
slot["last_val"] = num
ordered = sorted(buckets.items(), key=lambda kv: kv[0])
n_buckets = len(ordered)
def _aggregate(vals: list, last_val) -> "float | None":
if not vals:
return None
if agg == "sum":
return float(sum(vals))
if agg == "median":
return float(statistics.median(vals))
if agg == "last":
return float(last_val) if last_val is not None else None
if agg == "min":
return float(min(vals))
if agg == "max":
return float(max(vals))
return float(statistics.fmean(vals)) # mean (default)
t_out = [b.isoformat() for b, _ in ordered]
v_out = [_aggregate(s["vals"], s["last_val"]) for _, s in ordered]
c_out = [s["count"] for _, s in ordered]
downsampled = False
if n_buckets > max_points > 0:
keep = _downsample_indices(n_buckets, max_points)
t_out = [t_out[i] for i in keep]
v_out = [v_out[i] for i in keep]
c_out = [c_out[i] for i in keep]
downsampled = True
return {
"t": t_out,
"v": v_out,
"count": c_out,
"freq": eff_freq,
"agg": agg,
"n_in": n_in,
"n_buckets": n_buckets,
"downsampled": downsampled,
"note": "",
}
@@ -0,0 +1,118 @@
"""Tests para resample_timeseries (grupo eda)."""
import datetime
import os
import sys
sys.path.insert(0, os.path.dirname(__file__))
from resample_timeseries import resample_timeseries
def test_daily_a_mensual_mean():
# Serie diaria agregada a buckets mensuales con agg="mean".
t = [
"2020-01-01", "2020-01-15",
"2020-02-01", "2020-02-10", "2020-02-20",
]
v = [10.0, 20.0, 30.0, 40.0, 50.0]
r = resample_timeseries(t, v, freq="monthly", agg="mean")
assert r["t"] == ["2020-01-01", "2020-02-01"]
assert r["v"] == [15.0, 40.0] # (10+20)/2 ; (30+40+50)/3
assert r["count"] == [2, 3]
assert r["freq"] == "monthly"
assert r["agg"] == "mean"
assert r["n_in"] == 5
assert r["n_buckets"] == 2
assert r["downsampled"] is False
assert r["note"] == ""
def test_agg_sum_y_last():
t = [
"2020-01-01", "2020-01-15",
"2020-02-01", "2020-02-10", "2020-02-20",
]
v = [10.0, 20.0, 30.0, 40.0, 50.0]
r_sum = resample_timeseries(t, v, freq="monthly", agg="sum")
assert r_sum["v"] == [30.0, 120.0]
assert r_sum["agg"] == "sum"
# last = valor de la observacion cronologicamente mas reciente del bucket,
# aunque el orden de entrada este desordenado.
t2 = ["2020-02-20", "2020-02-01", "2020-02-10", "2020-01-15", "2020-01-01"]
v2 = [50.0, 30.0, 40.0, 20.0, 10.0]
r_last = resample_timeseries(t2, v2, freq="monthly", agg="last")
assert r_last["t"] == ["2020-01-01", "2020-02-01"]
assert r_last["v"] == [20.0, 50.0] # Jan->2020-01-15=20 ; Feb->2020-02-20=50
assert r_last["agg"] == "last"
def test_count_cuenta_observacion_con_valor_none():
# Un bucket con un valor None: count cuenta la fila, v ignora el None.
t = ["2020-03-05", "2020-03-06", "2020-03-20"]
v = [None, 7.0, 9.0]
r = resample_timeseries(t, v, freq="monthly", agg="mean")
assert r["t"] == ["2020-03-01"]
assert r["count"] == [3] # 3 filas con fecha valida
assert r["v"] == [8.0] # media de los validos: (7+9)/2
assert r["n_in"] == 3
# Bucket entero sin ningun valor numerico valido -> v = None, count sigue.
r2 = resample_timeseries(
["2020-04-01", "2020-04-02"], [None, "n/a"], freq="monthly"
)
assert r2["t"] == ["2020-04-01"]
assert r2["count"] == [2]
assert r2["v"] == [None]
def test_downsampling_respeta_max_points_y_extremos():
base = datetime.date(2021, 1, 1)
t = [(base + datetime.timedelta(days=i)).isoformat() for i in range(500)]
v = [float(i) for i in range(500)]
r = resample_timeseries(t, v, freq="daily", agg="mean", max_points=400)
assert r["n_buckets"] == 500
assert r["downsampled"] is True
assert len(r["t"]) <= 400
assert len(r["t"]) == len(r["v"]) == len(r["count"])
# Primero y ultimo bucket conservados.
assert r["t"][0] == "2021-01-01"
assert r["t"][-1] == (base + datetime.timedelta(days=499)).isoformat()
def test_freq_auto_infiere_mensual():
# Fechas separadas ~1 mes -> auto infiere "monthly".
t = [f"2022-{m:02d}-01" for m in range(1, 13)]
v = [float(m) for m in range(1, 13)]
r = resample_timeseries(t, v, freq="auto", agg="mean")
assert r["freq"] == "monthly"
assert r["n_buckets"] == 12
assert r["count"] == [1] * 12
# Fechas diarias consecutivas -> auto infiere "daily".
base = datetime.date(2023, 1, 1)
td = [(base + datetime.timedelta(days=i)).isoformat() for i in range(20)]
rd = resample_timeseries(td, [float(i) for i in range(20)], freq="auto")
assert rd["freq"] == "daily"
def test_edge_listas_vacias_o_desiguales():
vacio = resample_timeseries([], [])
assert vacio["t"] == [] and vacio["v"] == [] and vacio["count"] == []
assert vacio["note"] == "datos insuficientes"
assert vacio["n_in"] == 0 and vacio["n_buckets"] == 0
desigual = resample_timeseries(["2020-01-01", "2020-01-02"], [1.0])
assert desigual["note"] == "datos insuficientes"
assert desigual["t"] == []
# Todas las fechas invalidas -> tambien insuficiente.
invalidas = resample_timeseries(["no-fecha", "tampoco"], [1.0, 2.0])
assert invalidas["note"] == "datos insuficientes"
assert invalidas["n_in"] == 0
@@ -1,158 +0,0 @@
---
id: select_groupby_keys_py_datascience
name: select_groupby_keys
kind: function
lang: py
domain: datascience
version: "1.0.0"
purity: pure
signature: "def select_groupby_keys(profile: dict, max_keys: int = 3, max_card: int = 20, max_measures: int = 4) -> dict"
description: "Elige deterministicamente las columnas categoricas mas interesantes para GROUP BY, las numericas medida y pares pivote a partir de un TableProfile del grupo eda. Respaldo cuantitativo para el capitulo de agregacion/OLAP de un EDA. Funcion pura, no muta el input, nunca lanza."
tags: [eda, aggregation, groupby, olap, profiling, datascience]
uses_functions: []
uses_types: []
returns: []
returns_optional: false
error_type: ""
imports: []
example: |
from datascience import select_groupby_keys
profile = {
"n_rows": 891,
"key_candidates": ["passenger_id"],
"columns": [
{"name": "sex", "inferred_type": "categorical", "distinct_count": 2,
"unique_pct": 0.002, "null_pct": 0.0, "flags": [],
"categorical": {"imbalance": 1.8}, "numeric": None},
{"name": "pclass", "inferred_type": "categorical", "distinct_count": 3,
"unique_pct": 0.003, "null_pct": 0.0, "flags": [],
"categorical": {"imbalance": 2.5}, "numeric": None},
{"name": "fare", "inferred_type": "numeric", "distinct_count": 200,
"unique_pct": 0.2, "null_pct": 0.0, "flags": [],
"numeric": {"std": 49.7, "cv": 1.54}, "categorical": None},
],
}
select_groupby_keys(profile)
# {"group_keys": [{"col": "sex", ...}, {"col": "pclass", ...}],
# "measures": ["fare"],
# "pivots": [{"index": "sex", "columns": "pclass", "value": "fare"}],
# "note": "2 clave(s) de grupo: sex, pclass; 1 medida(s): fare; 1 pivot(s)."}
tested: true
tests:
- "test_titanic_picks_good_cats_excludes_id_and_constant"
- "test_titanic_measures_exclude_id_constant_and_keep_numerics"
- "test_titanic_generates_one_pivot"
- "test_empty_profile_returns_all_empty_and_does_not_crash"
- "test_none_profile_does_not_crash"
- "test_only_numerics_yields_empty_group_keys_and_no_pivots"
- "test_high_cardinality_and_max_card_are_excluded"
- "test_max_keys_limits_group_keys"
- "test_three_keys_cap_pivots_to_two"
- "test_does_not_mutate_input"
test_file_path: "python/functions/datascience/select_groupby_keys_test.py"
file_path: "python/functions/datascience/select_groupby_keys.py"
params:
- name: profile
desc: >
TableProfile dict del grupo eda (p.ej. salida de summarize_table_duckdb).
Se lee de forma defensiva (.get / or [] / isinstance). Claves usadas:
columns (list[ColumnProfile]), key_candidates (list de nombres de columna
o dicts {name}), n_rows. Cada ColumnProfile usa: name, inferred_type
("numeric"|"categorical"|"datetime"|"text"|"boolean"), distinct_count,
unique_pct (0..1), null_pct (0..1), flags (list[str], reconoce
"possible_id"/"high_cardinality"/"constant"), numeric ({std, cv, ...}|None)
y categorical ({imbalance, mode_pct, ...}|None).
- name: max_keys
desc: "Numero maximo de claves de grupo (group_keys) a devolver. Default 3."
- name: max_card
desc: >
Cardinalidad maxima (distinct_count) que una columna categorica puede
tener para seguir siendo candidata a clave de grupo. Default 20.
- name: max_measures
desc: "Numero maximo de columnas medida (nombres) a devolver. Default 4."
output: >
dict con group_keys (list de {col, cardinality, score} ordenada por score
desc), measures (list[str] de nombres de columnas numericas ordenadas por
dispersion), pivots (list de {index, columns, value}, hasta 2 pares
categorica x categorica con la primera measure como valor) y note (str,
resumen corto en espanol de lo elegido). Ante profile vacio/None devuelve
todas las listas vacias y una note descriptiva; nunca lanza.
---
## Ejemplo
```python
from datascience import select_groupby_keys
# TableProfile estilo titanic: 2 categoricas buenas, 1 numerica medida,
# 1 id secuencial (descartado) y un key_candidate declarado.
profile = {
"n_rows": 891,
"key_candidates": ["passenger_id"],
"columns": [
{"name": "sex", "inferred_type": "categorical", "distinct_count": 2,
"unique_pct": 0.002, "null_pct": 0.0, "flags": [],
"categorical": {"imbalance": 1.8}, "numeric": None},
{"name": "pclass", "inferred_type": "categorical", "distinct_count": 3,
"unique_pct": 0.003, "null_pct": 0.0, "flags": [],
"categorical": {"imbalance": 2.5}, "numeric": None},
{"name": "fare", "inferred_type": "numeric", "distinct_count": 200,
"unique_pct": 0.2, "null_pct": 0.0, "flags": [],
"numeric": {"std": 49.7, "cv": 1.54}, "categorical": None},
{"name": "passenger_id", "inferred_type": "numeric", "distinct_count": 891,
"unique_pct": 1.0, "null_pct": 0.0, "flags": ["possible_id"],
"numeric": {"std": 257.4, "cv": 0.58}, "categorical": None},
],
}
select_groupby_keys(profile)
# {
# "group_keys": [
# {"col": "sex", "cardinality": 2, "score": 0.5556},
# {"col": "pclass", "cardinality": 3, "score": 0.4},
# ],
# "measures": ["fare"], # passenger_id excluido (id secuencial)
# "pivots": [{"index": "sex", "columns": "pclass", "value": "fare"}],
# "note": "2 clave(s) de grupo: sex, pclass; 1 medida(s): fare; 1 pivot(s).",
# }
```
## Cuando usarla
Cuando hayas perfilado una tabla con el grupo `eda` (p.ej.
`summarize_table_duckdb`) y necesites decidir, sin mirar los datos, por qué
columnas merece la pena agrupar (GROUP BY) y qué métricas numéricas agregar:
el respaldo cuantitativo del capítulo de agregación/OLAP de un AutomaticEDA, o
para proponer pivotes en un dashboard. Es la capa de selección sobre el
TableProfile crudo: lee el perfil, ordena candidatos de forma determinista y
no toca los datos.
## Notas
Función pura, sin I/O ni dependencias externas (solo stdlib), no muta
`profile`. Lectura defensiva total (`.get`, `or []`, `isinstance`): un `{}` o
`None` produce `{"group_keys": [], "measures": [], "pivots": [], "note": ...}`
y nunca lanza.
Criterios de selección (deterministas):
- **group_keys** — candidatas con `inferred_type` en `("categorical","boolean")`.
Se descartan las que estén en `key_candidates`, con flag
`possible_id`/`high_cardinality`/`constant`, con `distinct_count` fuera de
`[2, max_card]`, o all-null (`null_pct >= 0.999`). `score = card_score *
balance_score`: `card_score` mantiene un plateau para cardinalidad moderada
(2..12) y decae hacia `max_card`; `balance_score = 1/imbalance` usando
`categorical.imbalance` si está, aproximando con `mode_pct` si no, o un valor
neutro (0.5) en último caso. Devuelve hasta `max_keys`, ordenadas por score
desc (empates por orden de columna).
- **measures** — candidatas con `inferred_type` en
`("numeric","integer","float")`. Se descartan id-like (flag `possible_id` y
`unique_pct >= 0.99`) y constantes (`numeric.std` == 0 o None). Se rankean por
dispersión informativa: `abs(cv)` si está, si no `abs(std)`. Devuelve hasta
`max_measures` **nombres** (strings).
- **pivots** — hasta 2 pares `(group_keys[i].col, group_keys[j].col)` con i<j y
la primera measure como valor. Vacío si hay menos de 2 group_keys.
Caveat de ranking de measures: mezclar `cv` (adimensional) con `std` (en
unidades de la columna) cuando una columna carece de `cv` puede dar órdenes
poco comparables entre columnas; se prefiere `cv` siempre que esté disponible.
@@ -1,310 +0,0 @@
"""Pure EDA helper: pick GROUP BY keys and measures from a TableProfile.
Given a ``TableProfile`` of the ``eda`` group (the dict produced by, e.g.,
``summarize_table_duckdb``), this function deterministically selects the most
interesting categorical columns to group by (GROUP BY), the numeric measure
columns to aggregate, and a couple of categorical x categorical pivot pairs.
It is the quantitative backbone for the aggregation / OLAP chapter of an
AutomaticEDA: a pure, deterministic ranking over the profile, with no I/O, no
mutation of the input and no external dependencies (stdlib only). It never
raises a missing or malformed profile yields an empty, well-formed result.
"""
def select_groupby_keys(
profile: dict,
max_keys: int = 3,
max_card: int = 20,
max_measures: int = 4,
) -> dict:
"""Select GROUP BY keys, measures and pivot pairs from a TableProfile.
Reads everything defensively (``.get(...)``, ``or []``, ``isinstance``) and
never raises. With an empty/None profile it returns every list empty.
Selection rules (deterministic):
- **group_keys** (categorical columns to group by): candidates have
``inferred_type`` in ``("categorical", "boolean")``. Discarded if they are
in ``profile['key_candidates']``, carry a ``possible_id`` /
``high_cardinality`` / ``constant`` flag, have ``distinct_count`` outside
``[2, max_card]``, or are all-null (``null_pct >= 0.999``). Each survivor
gets ``score = card_score * balance_score`` where ``card_score`` keeps a
plateau for moderate cardinality (2..12) and decays towards ``max_card``,
and ``balance_score = 1 / imbalance`` (``categorical.imbalance`` if
present, else approximated from ``mode_pct``, else a neutral default).
The top ``max_keys`` by score (desc, ties by column order) are returned.
- **measures** (numeric columns to aggregate): candidates have
``inferred_type`` in ``("numeric", "integer", "float")``. Discarded if
id-like (``possible_id`` flag *and* ``unique_pct >= 0.99``) or constant
(``numeric.std`` is ``0`` or ``None``). Ranked by informative dispersion:
``abs(cv)`` when available, else ``abs(std)``. The top ``max_measures``
**names** are returned.
- **pivots**: up to 2 ``(group_keys[i].col, group_keys[j].col)`` pairs with
``i < j``, using the first measure as the aggregated value. Empty when
fewer than 2 group keys were selected.
Args:
profile: TableProfile dict of the ``eda`` group. Relevant keys:
``columns`` (list[ColumnProfile]), ``key_candidates`` (list of
column names or ``{name}`` dicts), ``n_rows``. Each ColumnProfile
uses: ``name``, ``inferred_type``, ``distinct_count``,
``unique_pct`` (0..1), ``null_pct`` (0..1), ``flags`` (list[str]),
``numeric`` ({std, cv, ...}|None), ``categorical``
({imbalance, mode_pct, ...}|None).
max_keys: Maximum number of group-by keys to return. Default 3.
max_card: Maximum cardinality (``distinct_count``) a categorical column
may have to still qualify as a group key. Default 20.
max_measures: Maximum number of measure names to return. Default 4.
Returns:
dict with:
group_keys (list[{col, cardinality, score}], ordered by score desc),
measures (list[str], numeric column names ordered by dispersion),
pivots (list[{index, columns, value}], up to 2 pairs),
note (str, short summary of what was chosen).
"""
if not isinstance(profile, dict):
profile = {}
try:
max_keys = int(max_keys)
except (TypeError, ValueError):
max_keys = 3
try:
max_card = int(max_card)
except (TypeError, ValueError):
max_card = 20
try:
max_measures = int(max_measures)
except (TypeError, ValueError):
max_measures = 4
max_keys = max(max_keys, 0)
max_card = max(max_card, 2)
max_measures = max(max_measures, 0)
columns = profile.get("columns") or []
if not isinstance(columns, (list, tuple)):
columns = []
key_names = _key_candidate_names(profile.get("key_candidates"))
group_keys = _select_group_keys(columns, key_names, max_keys, max_card)
measures = _select_measures(columns, max_measures)
pivots = _select_pivots(group_keys, measures)
return {
"group_keys": group_keys,
"measures": measures,
"pivots": pivots,
"note": _build_note(group_keys, measures, pivots),
}
# ---------------------------------------------------------------------------
# group_keys
# ---------------------------------------------------------------------------
_GROUP_TYPES = ("categorical", "boolean")
_DISQUALIFYING_FLAGS = frozenset({"possible_id", "high_cardinality", "constant"})
_CARD_PLATEAU_HI = 12 # cardinalities 2..12 are all "moderate" (best).
def _select_group_keys(columns, key_names, max_keys, max_card) -> list:
"""Rank categorical/boolean columns suitable for GROUP BY."""
scored = []
for idx, col in enumerate(columns):
if not isinstance(col, dict):
continue
if (col.get("inferred_type") or "") not in _GROUP_TYPES:
continue
name = col.get("name")
if name is None:
continue
if name in key_names:
continue
flags = _as_set(col.get("flags"))
if flags & _DISQUALIFYING_FLAGS:
continue
if _num(col.get("null_pct"), 0.0) >= 0.999:
continue
card = _num(col.get("distinct_count"), 0.0)
if card < 2 or card > max_card:
continue
card_i = int(card)
score = _card_score(card_i, max_card) * _balance_score(col.get("categorical"))
scored.append((round(score, 6), idx, name, card_i))
# Deterministic: higher score first, ties broken by original column order.
scored.sort(key=lambda t: (-t[0], t[1]))
out = []
for score, _idx, name, card_i in scored[:max_keys]:
out.append({"col": name, "cardinality": card_i, "score": score})
return out
def _card_score(card: int, max_card: int) -> float:
"""Prefer moderate cardinality; plateau at 2..12, decay towards max_card."""
if card <= 1:
return 0.0
if card <= _CARD_PLATEAU_HI:
return 1.0
denom = max(max_card - _CARD_PLATEAU_HI, 1)
over = card - _CARD_PLATEAU_HI
return max(0.1, 1.0 - over / denom)
def _balance_score(categorical) -> float:
"""1.0 for a perfectly balanced category, decaying as imbalance grows.
Uses ``categorical.imbalance`` (max_count/min_count, >= 1) when available;
otherwise approximates from ``mode_pct`` (top-class dominance); otherwise a
neutral default so the column is still selectable.
"""
if isinstance(categorical, dict):
imbalance = categorical.get("imbalance")
if isinstance(imbalance, (int, float)) and imbalance >= 1.0:
return 1.0 / float(imbalance)
mode_pct = categorical.get("mode_pct")
if isinstance(mode_pct, (int, float)):
return _clamp(1.0 - float(mode_pct), 0.0, 1.0)
return 0.5
# ---------------------------------------------------------------------------
# measures
# ---------------------------------------------------------------------------
_NUMERIC_TYPES = ("numeric", "integer", "float")
def _select_measures(columns, max_measures) -> list:
"""Rank numeric columns by informative dispersion (cv, else std)."""
scored = []
for idx, col in enumerate(columns):
if not isinstance(col, dict):
continue
if (col.get("inferred_type") or "") not in _NUMERIC_TYPES:
continue
name = col.get("name")
if name is None:
continue
flags = _as_set(col.get("flags"))
unique_pct = _num(col.get("unique_pct"), 0.0)
if "possible_id" in flags and unique_pct >= 0.99:
continue # sequential id, not a measure.
numeric = col.get("numeric")
std = numeric.get("std") if isinstance(numeric, dict) else None
if not isinstance(std, (int, float)) or std == 0:
continue # constant or unknown spread -> not informative.
cv = numeric.get("cv") if isinstance(numeric, dict) else None
if isinstance(cv, (int, float)):
dispersion = abs(float(cv))
else:
dispersion = abs(float(std))
scored.append((dispersion, idx, name))
# Higher dispersion first, ties broken by original column order.
scored.sort(key=lambda t: (-t[0], t[1]))
return [name for _disp, _idx, name in scored[:max_measures]]
# ---------------------------------------------------------------------------
# pivots
# ---------------------------------------------------------------------------
def _select_pivots(group_keys, measures) -> list:
"""Up to 2 (cat_a, cat_b) pairs from the chosen group keys."""
if not isinstance(group_keys, list) or len(group_keys) < 2:
return []
value = measures[0] if measures else None
pairs = []
n = len(group_keys)
for i in range(n):
for j in range(i + 1, n):
pairs.append({
"index": group_keys[i].get("col"),
"columns": group_keys[j].get("col"),
"value": value,
})
if len(pairs) >= 2:
return pairs
return pairs
# ---------------------------------------------------------------------------
# helpers
# ---------------------------------------------------------------------------
def _build_note(group_keys, measures, pivots) -> str:
"""One-line Spanish summary of the selection."""
parts = []
if group_keys:
cols = ", ".join(str(g.get("col")) for g in group_keys)
parts.append(f"{len(group_keys)} clave(s) de grupo: {cols}")
else:
parts.append("sin categóricas agrupables")
if measures:
parts.append(f"{len(measures)} medida(s): " + ", ".join(str(m) for m in measures))
else:
parts.append("sin medidas numéricas")
if pivots:
parts.append(f"{len(pivots)} pivot(s)")
return "; ".join(parts) + "."
def _key_candidate_names(key_candidates) -> set:
"""Normalize ``key_candidates`` (strings or ``{name}`` dicts) to a name set."""
names = set()
if not isinstance(key_candidates, (list, tuple)):
return names
for entry in key_candidates:
if isinstance(entry, str):
names.add(entry)
elif isinstance(entry, dict):
nm = entry.get("name") or entry.get("col")
if nm is not None:
names.add(nm)
return names
def _as_set(flags) -> set:
"""Coerce a flags value into a set, tolerating None / non-iterables."""
if isinstance(flags, (list, tuple, set)):
return set(flags)
return set()
def _num(value, default: float) -> float:
"""Best-effort float conversion with a fallback default."""
if value is None:
return default
try:
return float(value)
except (TypeError, ValueError):
return default
def _clamp(x: float, lo: float, hi: float) -> float:
"""Recorta x al rango [lo, hi]."""
if x < lo:
return lo
if x > hi:
return hi
return x
@@ -1,213 +0,0 @@
"""Tests para select_groupby_keys (grupo eda, dominio datascience)."""
import os
import sys
sys.path.insert(0, os.path.dirname(__file__))
from select_groupby_keys import select_groupby_keys
def _cat_col(name, card, *, imbalance=2.0, flags=None, null_pct=0.0):
"""ColumnProfile categorico minimo con bloque categorical."""
return {
"name": name,
"inferred_type": "categorical",
"distinct_count": card,
"unique_pct": card / 1000.0,
"null_pct": null_pct,
"flags": flags or [],
"numeric": None,
"categorical": {"imbalance": imbalance, "mode_pct": 0.5, "n_distinct": card},
}
def _num_col(name, *, std, cv, flags=None, unique_pct=0.1):
"""ColumnProfile numerico minimo con bloque numeric."""
return {
"name": name,
"inferred_type": "numeric",
"distinct_count": 200,
"unique_pct": unique_pct,
"null_pct": 0.0,
"flags": flags or [],
"numeric": {"std": std, "cv": cv},
"categorical": None,
}
def _titanic_like_profile() -> dict:
"""Perfil estilo titanic: 2 categoricas buenas, 2 numericas, 1 id, 1 constante."""
return {
"n_rows": 891,
"key_candidates": ["passenger_id"],
"columns": [
_cat_col("sex", 2, imbalance=1.8),
_cat_col("pclass", 3, imbalance=2.5),
_num_col("age", std=14.5, cv=0.49),
_num_col("fare", std=49.7, cv=1.54),
# id secuencial: flag possible_id + unique_pct alto.
{
"name": "passenger_id",
"inferred_type": "numeric",
"distinct_count": 891,
"unique_pct": 1.0,
"null_pct": 0.0,
"flags": ["possible_id"],
"numeric": {"std": 257.4, "cv": 0.58},
"categorical": None,
},
# columna constante: flag constant + std 0.
{
"name": "embarked_const",
"inferred_type": "categorical",
"distinct_count": 1,
"unique_pct": 0.001,
"null_pct": 0.0,
"flags": ["constant"],
"numeric": None,
"categorical": {"imbalance": 1.0},
},
],
}
def test_titanic_picks_good_cats_excludes_id_and_constant():
out = select_groupby_keys(_titanic_like_profile())
# Elige las dos categoricas buenas.
chosen_cols = {g["col"] for g in out["group_keys"]}
assert chosen_cols == {"sex", "pclass"}
# Excluye la constante y el key_candidate.
assert "embarked_const" not in chosen_cols
assert "passenger_id" not in chosen_cols
# Cada group key trae col, cardinality y score.
for g in out["group_keys"]:
assert set(g.keys()) == {"col", "cardinality", "score"}
assert isinstance(g["score"], float)
by_col = {g["col"]: g for g in out["group_keys"]}
assert by_col["sex"]["cardinality"] == 2
assert by_col["pclass"]["cardinality"] == 3
# Ordenadas por score descendente.
scores = [g["score"] for g in out["group_keys"]]
assert scores == sorted(scores, reverse=True)
def test_titanic_measures_exclude_id_constant_and_keep_numerics():
out = select_groupby_keys(_titanic_like_profile())
# Solo nombres (strings) de numericas informativas, sin el id secuencial.
assert all(isinstance(m, str) for m in out["measures"])
assert "passenger_id" not in out["measures"]
assert set(out["measures"]) == {"age", "fare"}
# fare tiene mayor cv (1.54 > 0.49) -> primero.
assert out["measures"][0] == "fare"
def test_titanic_generates_one_pivot():
out = select_groupby_keys(_titanic_like_profile())
# Con 2 group keys -> exactamente 1 pivot.
assert len(out["pivots"]) == 1
pivot = out["pivots"][0]
assert set(pivot.keys()) == {"index", "columns", "value"}
assert {pivot["index"], pivot["columns"]} == {"sex", "pclass"}
# El valor es la primera measure (fare).
assert pivot["value"] == "fare"
def test_empty_profile_returns_all_empty_and_does_not_crash():
out = select_groupby_keys({})
assert out["group_keys"] == []
assert out["measures"] == []
assert out["pivots"] == []
assert isinstance(out["note"], str)
def test_none_profile_does_not_crash():
out = select_groupby_keys(None)
assert out == {
"group_keys": [],
"measures": [],
"pivots": [],
"note": out["note"],
}
assert isinstance(out["note"], str)
def test_only_numerics_yields_empty_group_keys_and_no_pivots():
profile = {
"n_rows": 500,
"key_candidates": [],
"columns": [
_num_col("price", std=12.0, cv=0.6),
_num_col("weight", std=3.0, cv=0.2),
],
}
out = select_groupby_keys(profile)
assert out["group_keys"] == []
assert out["pivots"] == []
# Las numericas si se eligen como measures.
assert set(out["measures"]) == {"price", "weight"}
assert out["measures"][0] == "price" # mayor cv.
def test_high_cardinality_and_max_card_are_excluded():
profile = {
"n_rows": 1000,
"key_candidates": [],
"columns": [
_cat_col("city", 50, flags=["high_cardinality"]), # flag -> fuera.
_cat_col("zone", 35), # card 35 > max_card 20 -> fuera.
_cat_col("region", 5), # valida.
],
}
out = select_groupby_keys(profile, max_card=20)
assert {g["col"] for g in out["group_keys"]} == {"region"}
def test_max_keys_limits_group_keys():
profile = {
"n_rows": 1000,
"key_candidates": [],
"columns": [
_cat_col("a", 4, imbalance=1.0),
_cat_col("b", 5, imbalance=1.2),
_cat_col("c", 6, imbalance=1.5),
_cat_col("d", 7, imbalance=2.0),
],
}
out = select_groupby_keys(profile, max_keys=2)
assert len(out["group_keys"]) == 2
# Hasta 2 pivots con >=2 keys (aqui exactamente 1 par posible entre 2 keys).
assert len(out["pivots"]) == 1
def test_three_keys_cap_pivots_to_two():
profile = {
"n_rows": 1000,
"key_candidates": [],
"columns": [
_cat_col("a", 4, imbalance=1.0),
_cat_col("b", 5, imbalance=1.1),
_cat_col("c", 6, imbalance=1.2),
_num_col("m", std=10.0, cv=0.5),
],
}
out = select_groupby_keys(profile, max_keys=3)
assert len(out["group_keys"]) == 3
# 3 keys -> 3 pares posibles, capado a 2.
assert len(out["pivots"]) == 2
for p in out["pivots"]:
assert p["value"] == "m"
def test_does_not_mutate_input():
profile = _titanic_like_profile()
before = repr(profile)
select_groupby_keys(profile)
assert repr(profile) == before
@@ -1,96 +0,0 @@
---
name: suggest_aggregations_llm
kind: function
lang: py
domain: datascience
version: "1.0.0"
purity: impure
signature: "def suggest_aggregations_llm(profile: dict, candidates: dict, max_aggs: int = 4, model: str = \"claude-haiku-4-5-20251001\") -> dict"
description: "MUST-11.1 del capitulo AGREGACION del AutomaticEDA (grupo eda). Dado el TableProfile de una tabla y los candidatos cuantitativos de select_groupby_keys ({group_keys:[{col,cardinality,score}], measures:[str], pivots:[{index,columns,value}]}), con UNA sola llamada al LLM elige y ordena las K agregaciones (GROUP BY categorica x medidas numericas) y los pivots MAS INFORMATIVOS para un analisis de grupos, con una razon corta cada uno, evitando la explosion combinatoria (no todo contra todo). Privacidad/coste: NO envia filas crudas, solo el resumen AGREGADO de los candidatos (tabla, columnas categoricas con cardinalidad/score, medidas, pivots). Reusa ask_llm del grupo claude-direct (API directa con token OAuth de Claude). Impura, dict-no-throw: NUNCA lanza y SIEMPRE devuelve algo usable; si el LLM falla, el JSON no parsea o no hay seleccion valida, cae a un fallback determinista construido desde los candidatos (source='fallback'). Toda columna que el LLM invente se descarta."
tags: [eda, claude-direct, llm, aggregation, groupby, pivot, datascience, automatic-eda]
params:
- name: profile
desc: "TableProfile del grupo eda. Solo se usa profile['table'] para nombrar la tabla en el prompt; puede ir vacio o sin esa clave (se usa '(tabla sin nombre)')."
- name: candidates
desc: "Salida de select_groupby_keys: {group_keys:[{col, cardinality, score}], measures:[str], pivots:[{index, columns, value}]}. group_keys = columnas categoricas candidatas para GROUP BY; measures = columnas numericas a agregar (sum/avg); pivots = cruces index x columns -> value sugeridos. Cualquier columna que el LLM elija debe existir aqui o se descarta. None o no-dict se trata como vacio."
- name: max_aggs
desc: "Tope de agregaciones a devolver. Default 4. Valores <1 o no-int se normalizan a 4. Limita tanto la seleccion del LLM como el fallback determinista, para evitar la explosion combinatoria."
- name: model
desc: "id del modelo Anthropic a usar en la unica llamada. Default 'claude-haiku-4-5-20251001' (haiku, coste bajo, ~2-3s). Para razones mas finas, pasar p.ej. 'claude-opus-4-8'."
output: "dict dict-no-throw: {status:'ok', source:'llm'|'fallback', aggregations:[{group_by:str, measures:[str], why:str}], pivots:[{index:str, columns:str, value:str|None, why:str}], note:str}. source=='llm' si el LLM produjo al menos una agregacion valida (columnas existentes en candidates); en cualquier otro caso (LLM caido, JSON invalido, seleccion vacia, sin candidatos) source=='fallback' y aggregations/pivots se derivan de candidates con why='selección cuantitativa (sin LLM)'. NUNCA lanza."
uses_functions: [ask_llm_py_core, select_groupby_keys_py_datascience]
uses_types: []
returns: []
returns_optional: false
error_type: "error_go_core"
imports: []
tested: true
tests: ["test_extract_json_object", "test_extract_json_wrapped_in_fences_and_junk", "test_extract_json_non_json_returns_none", "test_validate_aggregations_drops_invalid_columns", "test_llm_path_uses_selection", "test_llm_path_respects_max_aggs", "test_llm_invented_column_is_discarded", "test_fallback_on_empty_llm_response", "test_fallback_on_unparseable_response", "test_fallback_respects_max_aggs", "test_fallback_when_llm_raises", "test_no_candidates_returns_empty_fallback", "test_non_dict_candidates_does_not_raise"]
test_file_path: "python/functions/datascience/suggest_aggregations_llm_test.py"
file_path: "python/functions/datascience/suggest_aggregations_llm.py"
---
## Ejemplo
```python
import sys, os
sys.path.insert(0, os.path.join("python", "functions"))
from datascience.suggest_aggregations_llm import suggest_aggregations_llm
profile = {"table": "ventas"}
# candidates = salida de select_groupby_keys (aqui literal de ejemplo).
candidates = {
"group_keys": [
{"col": "categoria", "cardinality": 8, "score": 0.91},
{"col": "region", "cardinality": 5, "score": 0.74},
{"col": "canal", "cardinality": 3, "score": 0.60},
],
"measures": ["importe", "unidades"],
"pivots": [
{"index": "categoria", "columns": "region", "value": "importe"},
],
}
out = suggest_aggregations_llm(profile, candidates, max_aggs=4) # haiku por defecto
print("fuente:", out["source"]) # "llm" o "fallback" si no hay red
for agg in out["aggregations"]:
print(f"GROUP BY {agg['group_by']} -> {agg['measures']} ({agg['why']})")
for piv in out["pivots"]:
print(f"pivot {piv['index']} x {piv['columns']} = {piv['value']} ({piv['why']})")
```
## Cuando usarla
Justo despues de `select_groupby_keys` en el capitulo AGREGACION del AutomaticEDA:
cuando ya tienes los candidatos cuantitativos (columnas categoricas con cardinalidad,
medidas numericas y pivots posibles) y quieres que un LLM se quede con las K
agregaciones y pivots MAS INFORMATIVOS en vez de generar "todo contra todo". Usala para
priorizar el plan de analisis de grupos antes de materializar las tablas con
`aggregate_by_group` / pivots, manteniendo el coste y el ruido bajos. Si no hay red o
credenciales, sigue funcionando con un fallback determinista, asi que es seguro
ponerla en un pipeline.
## Gotchas
- **Impura: hace 1 llamada de red al LLM.** No es determinista ni gratis. Latencia
tipica ~2-3s con haiku. Una sola llamada cubre toda la seleccion.
- **Requiere token OAuth de Claude** en `~/.claude/.credentials.json` (via `ask_llm` /
grupo `claude-direct`). Sin token / sin red NO lanza: cae al **fallback
determinista** (`source="fallback"`) construido desde `candidates`
(group_keys x measures hasta `max_aggs`, pivots tal cual) con
`why="selección cuantitativa (sin LLM)"`. Comprueba `out["source"]` para saber si la
seleccion vino del LLM o del fallback.
- **NO envia filas crudas al LLM**, solo el resumen AGREGADO de los candidatos. Esto
exige que `candidates` venga ya calculado por `select_groupby_keys` (cardinalidades,
scores, medidas, pivots).
- **Valida columnas inventadas**: si el LLM propone un `group_by`/`measure`/`index`/
`columns` que no esta en `candidates`, esa entrada se descarta (las medidas se
recortan a las validas). Si tras validar no queda ninguna agregacion, cae al
fallback completo.
- **`max_aggs` acota la explosion combinatoria** tanto en el camino LLM como en el
fallback. Subirlo demasiado reintroduce el ruido que esta funcion evita.
- **Modelo `haiku` por defecto** para coste bajo; sube a `claude-opus-4-8` si necesitas
razones (`why`) mas finas (mas caro y lento).
@@ -1,405 +0,0 @@
"""suggest_aggregations_llm — el LLM elige las agregaciones mas informativas (grupo `eda`).
MUST-11.1 del capitulo AGREGACION del AutomaticEDA. Dado el `TableProfile` de una
tabla y los CANDIDATOS cuantitativos que produce `select_groupby_keys`
(`{group_keys:[{col,cardinality,score}], measures:[str], pivots:[{index,columns,value}]}`),
con UNA sola llamada al LLM elige y ordena las K agregaciones (GROUP BY categorica x
medidas numericas) y los pivots MAS INFORMATIVOS para un analisis de grupos, con una
razon corta cada uno. El objetivo es evitar la explosion combinatoria: en vez de
"todo contra todo", el LLM se queda con lo que mas informa.
Privacidad y coste: NO se envian filas crudas al LLM. El prompt solo lleva el resumen
AGREGADO de los candidatos (nombre de la tabla, columnas categoricas con su
cardinalidad/score, medidas y pivots posibles). Una sola llamada barata.
Reusa `ask_llm` del registry (grupo claude-direct, API directa con el token OAuth de
Claude en ~/.claude/.credentials.json, arranque 0). Impura: una llamada de red.
Estilo dict-no-throw con FALLBACK DETERMINISTA: la funcion NUNCA lanza y SIEMPRE
devuelve algo usable. Si `ask_llm` falla (devuelve ""), el JSON no parsea, o el LLM no
produce ninguna seleccion valida, se construye la respuesta directamente desde los
candidatos (group_keys x measures hasta max_aggs, pivots tal cual) con
`source="fallback"`. Ademas, toda columna que el LLM invente (no presente en los
candidatos) se descarta.
"""
import json
from core.ask_llm import ask_llm
_SYSTEM = (
"Eres un analista de datos conciso. Te dan los CANDIDATOS AGREGADOS de una tabla "
"(columnas categoricas para GROUP BY con su cardinalidad, medidas numericas y "
"pivots posibles) y eliges las agregaciones y pivots MAS INFORMATIVOS para "
"entender los grupos, evitando la explosion combinatoria (no todo contra todo). "
"No recibes filas crudas. Responde en espanol. Responde SIEMPRE y SOLO con un "
"unico objeto JSON valido, sin texto alrededor ni fences de markdown, con la forma "
'{"aggregations": [{"group_by": "<col categorica>", "measures": ["<medida>", ...], '
'"why": "<razon corta>"}], "pivots": [{"index": "<col>", "columns": "<col>", '
'"value": "<medida o null>", "why": "<razon corta>"}]}. Usa SOLO nombres de columna '
"que aparezcan en los candidatos; no inventes nombres."
)
def _fmt_num(value) -> str:
"""Formatea un numero de forma compacta para el prompt (None -> '?')."""
if value is None:
return "?"
if isinstance(value, bool):
return str(value)
if isinstance(value, float):
if value == int(value):
return str(int(value))
return f"{value:.4g}"
return str(value)
def _candidate_view(candidates: dict):
"""Extrae las vistas utiles de los candidatos. Funcion interna PURA.
Devuelve la tupla (group_cols, measures, measure_set, pivots, group_keys):
- group_cols: set de nombres de columna categorica validas (de group_keys[].col).
- measures: lista de medidas numericas (str) preservando orden.
- measure_set: set de las medidas para validar pertenencia rapido.
- pivots: lista de pivots candidatos (dicts) tal cual vienen.
- group_keys: lista de dicts {col, cardinality, score} ya filtrada a entradas validas.
Tolera estructuras incompletas o de tipo incorrecto sin lanzar.
"""
candidates = candidates if isinstance(candidates, dict) else {}
gk_raw = candidates.get("group_keys")
group_keys = []
if isinstance(gk_raw, list):
for gk in gk_raw:
if isinstance(gk, dict) and isinstance(gk.get("col"), str):
group_keys.append(gk)
group_cols = {gk["col"] for gk in group_keys}
m_raw = candidates.get("measures")
measures = [m for m in m_raw if isinstance(m, str)] if isinstance(m_raw, list) else []
measure_set = set(measures)
p_raw = candidates.get("pivots")
pivots = p_raw if isinstance(p_raw, list) else []
return group_cols, measures, measure_set, pivots, group_keys
def _sorted_group_cols(group_keys: list) -> list:
"""Nombres de columna categorica ordenados por score descendente. PURA."""
def _score(gk):
s = gk.get("score")
if isinstance(s, (int, float)) and not isinstance(s, bool):
return s
return 0.0
return [gk["col"] for gk in sorted(group_keys, key=_score, reverse=True)]
def _build_prompt(profile: dict, candidates: dict, max_aggs: int) -> str:
"""Construye el prompt compacto SOLO con agregados. Funcion interna PURA.
No toca red ni disco: testeable sin credenciales. Incluye el nombre de la tabla,
las columnas categoricas candidatas (con cardinalidad y score), las medidas
numericas y los pivots candidatos. Nunca filas crudas.
Args:
profile: TableProfile (se usa solo profile['table'] para nombrar la tabla).
candidates: salida de select_groupby_keys.
max_aggs: tope de agregaciones a pedir.
Returns:
El texto del prompt.
"""
profile = profile if isinstance(profile, dict) else {}
candidates = candidates if isinstance(candidates, dict) else {}
table = profile.get("table")
table = str(table) if table is not None else "(tabla sin nombre)"
lines = [
f"Tabla: {table}",
(
"Tarea: elegir las agregaciones (GROUP BY categorica x medidas numericas) y "
"los pivots MAS INFORMATIVOS para un analisis de grupos. Evita la explosion "
"combinatoria: NO combines todo contra todo, prioriza lo que mas informa."
),
f"Devuelve a lo sumo {max_aggs} agregaciones.",
"",
"Columnas categoricas candidatas para GROUP BY (col: cardinalidad, score):",
]
group_keys = candidates.get("group_keys") or []
for gk in group_keys:
if not isinstance(gk, dict) or not isinstance(gk.get("col"), str):
continue
lines.append(
f" - {gk['col']}: cardinalidad={_fmt_num(gk.get('cardinality'))}, "
f"score={_fmt_num(gk.get('score'))}"
)
measures = candidates.get("measures") or []
lines.append("")
lines.append("Medidas numericas disponibles (para sum/avg por grupo):")
lines.append(" " + ", ".join(str(m) for m in measures if isinstance(m, str)))
pivots = candidates.get("pivots") or []
if pivots:
lines.append("")
lines.append("Pivots candidatos (index x columns -> value):")
for p in pivots:
if not isinstance(p, dict):
continue
lines.append(
f" - index={p.get('index')}, columns={p.get('columns')}, "
f"value={p.get('value')}"
)
lines.append("")
lines.append(
"Usa SOLO columnas de las listas anteriores; no inventes nombres. Responde "
"SOLO con el JSON descrito en las instrucciones del sistema."
)
return "\n".join(lines)
def _extract_json(text: str):
"""Extrae el primer bloque JSON (objeto o array) de la respuesta. PURA.
Localiza el bloque que empieza antes (el primer '{' o el primer '[') y, para ese
delimitador, hace json.loads del rango hasta su ultimo cierre. Tolera texto basura
alrededor y fences ```json. NUNCA lanza: ante cualquier fallo devuelve None.
Args:
text: respuesta cruda del LLM.
Returns:
El objeto/lista deserializado, o None si no se pudo parsear.
"""
if not text or not isinstance(text, str):
return None
opens = []
i_obj = text.find("{")
if i_obj != -1:
opens.append((i_obj, "{", "}"))
i_arr = text.find("[")
if i_arr != -1:
opens.append((i_arr, "[", "]"))
opens.sort()
for _, open_c, close_c in opens:
start = text.find(open_c)
end = text.rfind(close_c)
if start != -1 and end != -1 and end > start:
try:
return json.loads(text[start : end + 1])
except (ValueError, TypeError):
continue
return None
def _validate_aggregations(raw_aggs, group_cols: set, measure_set: set, max_aggs: int) -> list:
"""Filtra las agregaciones del LLM a las que usan SOLO columnas candidatas. PURA.
Descarta cualquier agregacion cuyo group_by no este en group_cols o que no tenga
al menos una medida valida. Recorta las medidas a las presentes en measure_set.
Limita el resultado a max_aggs entradas.
"""
out = []
if not isinstance(raw_aggs, list):
return out
for item in raw_aggs:
if not isinstance(item, dict):
continue
gb = item.get("group_by")
if not isinstance(gb, str) or gb not in group_cols:
continue # columna inventada -> se descarta
raw_measures = item.get("measures")
if isinstance(raw_measures, str):
raw_measures = [raw_measures]
if not isinstance(raw_measures, list):
continue
measures = [m for m in raw_measures if isinstance(m, str) and m in measure_set]
if not measures:
continue # sin medidas validas -> agregacion inutil
why = item.get("why")
why = str(why) if why is not None else ""
out.append({"group_by": gb, "measures": measures, "why": why})
if len(out) >= max_aggs:
break
return out
def _validate_pivots(raw_pivots, group_cols: set, measure_set: set) -> list:
"""Filtra los pivots del LLM a los que usan SOLO columnas candidatas. PURA.
Descarta el pivot si index o columns no son columnas categoricas validas. Si el
value no es una medida valida, lo deja en None (un pivot de conteo sigue siendo util).
"""
out = []
if not isinstance(raw_pivots, list):
return out
for item in raw_pivots:
if not isinstance(item, dict):
continue
idx = item.get("index")
cols = item.get("columns")
if not (isinstance(idx, str) and idx in group_cols):
continue
if not (isinstance(cols, str) and cols in group_cols):
continue
val = item.get("value")
if not (isinstance(val, str) and val in measure_set):
val = None
why = item.get("why")
why = str(why) if why is not None else ""
out.append({"index": idx, "columns": cols, "value": val, "why": why})
return out
def _fallback_aggregations(group_cols_sorted: list, measures: list, max_aggs: int) -> list:
"""Agregaciones deterministas: cada columna categorica x todas las medidas. PURA."""
out = []
for col in group_cols_sorted:
out.append(
{
"group_by": col,
"measures": list(measures),
"why": "selección cuantitativa (sin LLM)",
}
)
if len(out) >= max_aggs:
break
return out
def _fallback_pivots(cand_pivots: list) -> list:
"""Normaliza los pivots candidatos a la forma de salida (tal cual + why). PURA."""
out = []
if not isinstance(cand_pivots, list):
return out
for p in cand_pivots:
if not isinstance(p, dict):
continue
idx = p.get("index")
cols = p.get("columns")
if not (isinstance(idx, str) and isinstance(cols, str)):
continue
val = p.get("value")
if not isinstance(val, str):
val = None
out.append(
{
"index": idx,
"columns": cols,
"value": val,
"why": "selección cuantitativa (sin LLM)",
}
)
return out
def suggest_aggregations_llm(
profile: dict,
candidates: dict,
max_aggs: int = 4,
model: str = "claude-haiku-4-5-20251001",
) -> dict:
"""Elige las agregaciones y pivots mas informativos con UNA llamada al LLM.
MUST-11.1 del capitulo AGREGACION del AutomaticEDA. Toma el perfil de la tabla y
los candidatos cuantitativos (salida de select_groupby_keys) y deja que el LLM
seleccione/ordene las K agregaciones (GROUP BY categorica x medidas) y los pivots
mas utiles, con una razon corta cada uno, evitando la explosion combinatoria.
Privacidad/coste: solo viaja al LLM el resumen AGREGADO de los candidatos, nunca
filas crudas. Una sola llamada barata.
dict-no-throw con fallback determinista: NUNCA lanza. Si el LLM falla, el JSON no
parsea, o no produce seleccion valida -> construye la respuesta desde los candidatos
(group_keys x measures hasta max_aggs, pivots tal cual) con source="fallback". Las
columnas que el LLM invente (no presentes en los candidatos) se descartan.
Args:
profile: TableProfile del grupo eda. Solo se usa profile['table'] para nombrar
la tabla en el prompt; puede ir vacio.
candidates: salida de select_groupby_keys, con la forma
{group_keys:[{col,cardinality,score}], measures:[str],
pivots:[{index,columns,value}]}.
max_aggs: tope de agregaciones a devolver. Default 4. Valores <1 o no-int se
normalizan a 4.
model: id del modelo Anthropic. Default 'claude-haiku-4-5-20251001' (haiku,
coste bajo, ~2-3s).
Returns:
dict {status:"ok", source:"llm"|"fallback",
aggregations:[{group_by:str, measures:[str], why:str}],
pivots:[{index:str, columns:str, value:str|None, why:str}], note:str}.
source=="llm" si el LLM produjo al menos una agregacion valida; en cualquier
otro caso "fallback". NUNCA lanza.
"""
if not isinstance(candidates, dict):
candidates = {}
if isinstance(max_aggs, bool) or not isinstance(max_aggs, int) or max_aggs < 1:
max_aggs = 4
group_cols, measures, measure_set, cand_pivots, group_keys = _candidate_view(candidates)
group_cols_sorted = _sorted_group_cols(group_keys)
# Sin material suficiente para agregar: no merece la pena llamar al LLM.
if not group_cols or not measures:
return {
"status": "ok",
"source": "fallback",
"aggregations": [],
"pivots": _fallback_pivots(cand_pivots),
"note": "sin candidatos suficientes para agregar",
}
prompt = _build_prompt(profile, candidates, max_aggs)
try:
text = ask_llm(prompt, model=model, system=_SYSTEM, echo=False)
except Exception: # noqa: BLE001 — degradacion: cualquier fallo de red/LLM.
text = ""
parsed = _extract_json(text)
if parsed is not None:
if isinstance(parsed, dict):
raw_aggs = parsed.get("aggregations")
raw_pivots = parsed.get("pivots")
elif isinstance(parsed, list):
raw_aggs = parsed
raw_pivots = None
else:
raw_aggs = None
raw_pivots = None
aggs = _validate_aggregations(raw_aggs, group_cols, measure_set, max_aggs)
if aggs:
pivots = _validate_pivots(raw_pivots, group_cols, measure_set)
if not pivots:
pivots = _fallback_pivots(cand_pivots)
return {
"status": "ok",
"source": "llm",
"aggregations": aggs,
"pivots": pivots,
"note": f"{len(aggs)} agregaciones y {len(pivots)} pivots seleccionados por el LLM",
}
# Fallback determinista.
note = (
"LLM no disponible; selección cuantitativa determinista"
if not text
else "LLM sin selección válida; selección cuantitativa determinista"
)
return {
"status": "ok",
"source": "fallback",
"aggregations": _fallback_aggregations(group_cols_sorted, measures, max_aggs),
"pivots": _fallback_pivots(cand_pivots),
"note": note,
}
@@ -1,198 +0,0 @@
"""Tests para suggest_aggregations_llm.
NO acceden a red ni a credenciales: las funciones internas (_build_prompt,
_extract_json, _validate_*, _fallback_*) son puras y testeables aisladas; la unica
via que llamaria al LLM (suggest_aggregations_llm) se prueba reemplazando el simbolo
`ask_llm` del modulo bajo prueba con una funcion simulada. Los candidatos van
literales en el test: NO se importa select_groupby_keys.
Cubre golden (LLM ok con columnas validas), edge (max_aggs respetado, sin candidatos)
y error (LLM caido -> fallback, JSON invalido -> fallback, columna inventada -> se
descarta). Todos sin tocar la red.
"""
import json
import datascience.suggest_aggregations_llm as M
from datascience.suggest_aggregations_llm import (
_extract_json,
_validate_aggregations,
suggest_aggregations_llm,
)
# Candidatos de ejemplo con la forma que produce select_groupby_keys (literales).
_CANDIDATES = {
"group_keys": [
{"col": "categoria", "cardinality": 8, "score": 0.91},
{"col": "region", "cardinality": 5, "score": 0.74},
{"col": "canal", "cardinality": 3, "score": 0.60},
],
"measures": ["importe", "unidades"],
"pivots": [
{"index": "categoria", "columns": "region", "value": "importe"},
],
}
_PROFILE = {"table": "ventas"}
def _fake_returner(text):
"""Devuelve un ask_llm simulado que ignora args y retorna `text`."""
def _fake(prompt, model="x", system="", echo=True, **kwargs):
return text
return _fake
# --- _extract_json (parser puro, sin red) ---
def test_extract_json_object():
obj = {"aggregations": [{"group_by": "categoria", "measures": ["importe"], "why": "x"}]}
assert _extract_json(json.dumps(obj)) == obj
def test_extract_json_wrapped_in_fences_and_junk():
obj = {"aggregations": [], "pivots": []}
text = "Claro, aqui tienes:\n```json\n" + json.dumps(obj) + "\n```\nFin."
assert _extract_json(text) == obj
def test_extract_json_non_json_returns_none():
assert _extract_json("no hay json aqui") is None
assert _extract_json("") is None
assert _extract_json(None) is None
# --- _validate_aggregations (puro) ---
def test_validate_aggregations_drops_invalid_columns():
group_cols = {"categoria", "region"}
measure_set = {"importe", "unidades"}
raw = [
{"group_by": "categoria", "measures": ["importe", "inventada"], "why": "ok"},
{"group_by": "no_existe", "measures": ["importe"], "why": "mala"},
{"group_by": "region", "measures": ["solo_inventada"], "why": "sin medidas"},
]
out = _validate_aggregations(raw, group_cols, measure_set, max_aggs=4)
# Solo sobrevive la primera, con las medidas recortadas a las validas.
assert out == [{"group_by": "categoria", "measures": ["importe"], "why": "ok"}]
# --- suggest_aggregations_llm: camino LLM (golden) ---
def test_llm_path_uses_selection(monkeypatch):
llm_obj = {
"aggregations": [
{"group_by": "categoria", "measures": ["importe"], "why": "ventas por familia"},
{"group_by": "region", "measures": ["importe", "unidades"], "why": "reparto geografico"},
],
"pivots": [
{"index": "categoria", "columns": "region", "value": "importe", "why": "cruce clave"},
],
}
monkeypatch.setattr(M, "ask_llm", _fake_returner(json.dumps(llm_obj)))
out = suggest_aggregations_llm(_PROFILE, _CANDIDATES)
assert out["status"] == "ok"
assert out["source"] == "llm"
assert out["aggregations"] == llm_obj["aggregations"]
assert out["pivots"][0]["index"] == "categoria"
assert out["pivots"][0]["why"] == "cruce clave"
def test_llm_path_respects_max_aggs(monkeypatch):
llm_obj = {
"aggregations": [
{"group_by": "categoria", "measures": ["importe"], "why": "a"},
{"group_by": "region", "measures": ["importe"], "why": "b"},
{"group_by": "canal", "measures": ["unidades"], "why": "c"},
],
"pivots": [],
}
monkeypatch.setattr(M, "ask_llm", _fake_returner(json.dumps(llm_obj)))
out = suggest_aggregations_llm(_PROFILE, _CANDIDATES, max_aggs=2)
assert out["source"] == "llm"
assert len(out["aggregations"]) == 2
def test_llm_invented_column_is_discarded(monkeypatch):
# El LLM mezcla una agregacion valida con otra de columna inexistente.
llm_obj = {
"aggregations": [
{"group_by": "categoria", "measures": ["importe"], "why": "valida"},
{"group_by": "columna_fantasma", "measures": ["importe"], "why": "inventada"},
],
"pivots": [
{"index": "fantasma", "columns": "region", "value": "importe", "why": "mala"},
],
}
monkeypatch.setattr(M, "ask_llm", _fake_returner(json.dumps(llm_obj)))
out = suggest_aggregations_llm(_PROFILE, _CANDIDATES)
assert out["source"] == "llm"
# La agregacion inventada se descarta; queda solo la valida.
assert [a["group_by"] for a in out["aggregations"]] == ["categoria"]
# El pivot con index fantasma se descarta -> cae a los pivots de candidates.
assert all(p["index"] in {"categoria", "region", "canal"} for p in out["pivots"])
# --- suggest_aggregations_llm: fallback determinista (error paths) ---
def test_fallback_on_empty_llm_response(monkeypatch):
monkeypatch.setattr(M, "ask_llm", _fake_returner(""))
out = suggest_aggregations_llm(_PROFILE, _CANDIDATES, max_aggs=4)
assert out["status"] == "ok"
assert out["source"] == "fallback"
# Las agregaciones se derivan de candidates (una por group_key, con todas las medidas).
assert out["aggregations"][0]["group_by"] in {"categoria", "region", "canal"}
assert out["aggregations"][0]["measures"] == ["importe", "unidades"]
assert out["aggregations"][0]["why"] == "selección cuantitativa (sin LLM)"
# Pivots tal cual de candidates.
assert out["pivots"][0]["index"] == "categoria"
def test_fallback_on_unparseable_response(monkeypatch):
monkeypatch.setattr(M, "ask_llm", _fake_returner("esto no es JSON {roto"))
out = suggest_aggregations_llm(_PROFILE, _CANDIDATES)
assert out["source"] == "fallback"
assert len(out["aggregations"]) >= 1
def test_fallback_respects_max_aggs(monkeypatch):
monkeypatch.setattr(M, "ask_llm", _fake_returner(""))
out = suggest_aggregations_llm(_PROFILE, _CANDIDATES, max_aggs=2)
assert out["source"] == "fallback"
assert len(out["aggregations"]) == 2
def test_fallback_when_llm_raises(monkeypatch):
def _boom(*args, **kwargs):
raise RuntimeError("sin red")
monkeypatch.setattr(M, "ask_llm", _boom)
out = suggest_aggregations_llm(_PROFILE, _CANDIDATES)
assert out["source"] == "fallback"
assert out["aggregations"] # no vacio, no lanza
def test_no_candidates_returns_empty_fallback():
# Sin red porque ni siquiera se llama al LLM (no hay material).
out = suggest_aggregations_llm(_PROFILE, {"group_keys": [], "measures": [], "pivots": []})
assert out["status"] == "ok"
assert out["source"] == "fallback"
assert out["aggregations"] == []
def test_non_dict_candidates_does_not_raise():
out = suggest_aggregations_llm(_PROFILE, None)
assert out["status"] == "ok"
assert out["aggregations"] == []