superset/viz.py
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
# pylint: disable=too-many-lines
"""This module contains the 'Viz' objects
These objects represent the backend of all the visualizations that
Superset can render.
"""
from __future__ import annotations
import copy
import dataclasses
import logging
import math
import re
from collections import defaultdict, OrderedDict
from datetime import datetime, timedelta
from itertools import product
from typing import Any, cast, Optional, TYPE_CHECKING
import geohash
import numpy as np
import pandas as pd
import polyline
from dateutil import relativedelta as rdelta
from deprecation import deprecated
from flask import request
from flask_babel import lazy_gettext as _
from geopy.point import Point
from pandas.tseries.frequencies import to_offset
from superset import app
from superset.common.db_query_status import QueryStatus
from superset.constants import NULL_STRING
from superset.errors import ErrorLevel, SupersetError, SupersetErrorType
from superset.exceptions import (
CacheLoadError,
NullValueException,
QueryObjectValidationError,
SpatialException,
SupersetSecurityException,
)
from superset.extensions import cache_manager, security_manager
from superset.models.helpers import QueryResult
from superset.sql_parse import sanitize_clause
from superset.superset_typing import (
Column,
Metric,
QueryObjectDict,
VizData,
VizPayload,
)
from superset.utils import core as utils, csv, json
from superset.utils.cache import set_and_log_cache
from superset.utils.core import (
apply_max_row_limit,
DateColumn,
DTTM_ALIAS,
ExtraFiltersReasonType,
get_column_name,
get_column_names,
get_column_names_from_columns,
JS_MAX_INTEGER,
merge_extra_filters,
simple_filter_to_adhoc,
)
from superset.utils.date_parser import get_since_until, parse_past_timedelta
from superset.utils.hashing import md5_sha_from_str
if TYPE_CHECKING:
from superset.connectors.sqla.models import BaseDatasource
config = app.config
stats_logger = config["STATS_LOGGER"]
relative_start = config["DEFAULT_RELATIVE_START_TIME"]
relative_end = config["DEFAULT_RELATIVE_END_TIME"]
logger = logging.getLogger(__name__)
METRIC_KEYS = [
"metric",
"metrics",
"percent_metrics",
"metric_2",
"secondary_metric",
"x",
"y",
"size",
]
class BaseViz: # pylint: disable=too-many-public-methods
"""All visualizations derive this base class"""
viz_type: str | None = None
verbose_name = "Base Viz"
credits = ""
is_timeseries = False
cache_type = "df"
enforce_numerical_metrics = True
@deprecated(deprecated_in="3.0")
def __init__(
self,
datasource: BaseDatasource,
form_data: dict[str, Any],
force: bool = False,
force_cached: bool = False,
) -> None:
if not datasource:
raise QueryObjectValidationError(_("Viz is missing a datasource"))
self.datasource = datasource
self.request = request
self.viz_type = form_data.get("viz_type")
self.form_data = form_data
self.query = ""
self.token = utils.get_form_data_token(form_data)
self.groupby: list[Column] = self.form_data.get("groupby") or []
self.time_shift = timedelta()
self.status: str | None = None
self.error_msg = ""
self.results: QueryResult | None = None
self.applied_filter_columns: list[Column] = []
self.rejected_filter_columns: list[Column] = []
self.errors: list[dict[str, Any]] = []
self.force = force
self._force_cached = force_cached
self.from_dttm: datetime | None = None
self.to_dttm: datetime | None = None
self._extra_chart_data: list[tuple[str, pd.DataFrame]] = []
self.process_metrics()
self.applied_filters: list[dict[str, str]] = []
self.rejected_filters: list[dict[str, str]] = []
@property
@deprecated(deprecated_in="3.0")
def force_cached(self) -> bool:
return self._force_cached
@deprecated(deprecated_in="3.0")
def process_metrics(self) -> None:
# metrics in Viz is order sensitive, so metric_dict should be
# OrderedDict
self.metric_dict = OrderedDict()
for mkey in METRIC_KEYS:
val = self.form_data.get(mkey)
if val:
if not isinstance(val, list):
val = [val]
for o in val:
label = utils.get_metric_name(o)
self.metric_dict[label] = o
# Cast to list needed to return serializable object in py3
self.all_metrics = list(self.metric_dict.values())
self.metric_labels = list(self.metric_dict.keys())
@staticmethod
@deprecated(deprecated_in="3.0")
def handle_js_int_overflow(
data: dict[str, list[dict[str, Any]]],
) -> dict[str, list[dict[str, Any]]]:
for record in data.get("records", {}):
for k, v in list(record.items()):
if isinstance(v, int):
# if an int is too big for Java Script to handle
# convert it to a string
if abs(v) > JS_MAX_INTEGER:
record[k] = str(v)
return data
@deprecated(deprecated_in="3.0")
def run_extra_queries(self) -> None:
"""Lifecycle method to use when more than one query is needed
In rare-ish cases, a visualization may need to execute multiple
queries. That is the case for FilterBox or for time comparison
in Line chart for instance.
In those cases, we need to make sure these queries run before the
main `get_payload` method gets called, so that the overall caching
metadata can be right. The way it works here is that if any of
the previous `get_df_payload` calls hit the cache, the main
payload's metadata will reflect that.
The multi-query support may need more work to become a first class
use case in the framework, and for the UI to reflect the subtleties
(show that only some of the queries were served from cache for
instance). In the meantime, since multi-query is rare, we treat
it with a bit of a hack. Note that the hack became necessary
when moving from caching the visualization's data itself, to caching
the underlying query(ies).
"""
@deprecated(deprecated_in="3.0")
def apply_rolling(self, df: pd.DataFrame) -> pd.DataFrame:
rolling_type = self.form_data.get("rolling_type")
rolling_periods = int(self.form_data.get("rolling_periods") or 0)
min_periods = int(self.form_data.get("min_periods") or 0)
if rolling_type in ("mean", "std", "sum") and rolling_periods:
kwargs = {"window": rolling_periods, "min_periods": min_periods}
if rolling_type == "mean":
df = df.rolling(**kwargs).mean()
elif rolling_type == "std":
df = df.rolling(**kwargs).std()
elif rolling_type == "sum":
df = df.rolling(**kwargs).sum()
elif rolling_type == "cumsum":
df = df.cumsum()
if min_periods:
df = df[min_periods:]
if df.empty:
raise QueryObjectValidationError(
_(
"Applied rolling window did not return any data. Please make sure "
"the source query satisfies the minimum periods defined in the "
"rolling window."
)
)
return df
@deprecated(deprecated_in="3.0")
def get_samples(self) -> dict[str, Any]:
query_obj = self.query_obj()
query_obj.update(
{
"is_timeseries": False,
"groupby": [],
"metrics": [],
"orderby": [],
"row_limit": config["SAMPLES_ROW_LIMIT"],
"columns": [o.column_name for o in self.datasource.columns],
"from_dttm": None,
"to_dttm": None,
}
)
payload = self.get_df_payload(query_obj) # leverage caching logic
return {
"data": payload["df"].to_dict(orient="records"),
"colnames": payload.get("colnames"),
"coltypes": payload.get("coltypes"),
"rowcount": payload.get("rowcount"),
"sql_rowcount": payload.get("sql_rowcount"),
}
@deprecated(deprecated_in="3.0")
def get_df(self, query_obj: QueryObjectDict | None = None) -> pd.DataFrame:
"""Returns a pandas dataframe based on the query object"""
if not query_obj:
query_obj = self.query_obj()
if not query_obj:
return pd.DataFrame()
self.error_msg = ""
timestamp_format = None
if self.datasource.type == "table":
granularity_col = self.datasource.get_column(query_obj["granularity"])
if granularity_col:
timestamp_format = granularity_col.python_date_format
# The datasource here can be different backend but the interface is common
self.results = self.datasource.query(query_obj)
self.applied_filter_columns = self.results.applied_filter_columns or []
self.rejected_filter_columns = self.results.rejected_filter_columns or []
self.query = self.results.query
self.status = self.results.status
self.errors = self.results.errors
df = self.results.df
# Transform the timestamp we received from database to pandas supported
# datetime format. If no python_date_format is specified, the pattern will
# be considered as the default ISO date format
# If the datetime format is unix, the parse will use the corresponding
# parsing logic.
if not df.empty:
utils.normalize_dttm_col(
df=df,
dttm_cols=tuple(
[
DateColumn.get_legacy_time_column(
timestamp_format=timestamp_format,
offset=self.datasource.offset,
time_shift=self.form_data.get("time_shift"),
)
]
),
)
if self.enforce_numerical_metrics:
self.df_metrics_to_num(df)
df.replace([np.inf, -np.inf], np.nan, inplace=True)
return df
@deprecated(deprecated_in="3.0")
def df_metrics_to_num(self, df: pd.DataFrame) -> None:
"""Converting metrics to numeric when pandas.read_sql cannot"""
metrics = self.metric_labels
for col, dtype in df.dtypes.items():
if dtype.type == np.object_ and col in metrics:
df[col] = pd.to_numeric(df[col], errors="coerce")
@deprecated(deprecated_in="3.0")
def process_query_filters(self) -> None:
utils.convert_legacy_filters_into_adhoc(self.form_data)
merge_extra_filters(self.form_data)
utils.split_adhoc_filters_into_base_filters(self.form_data)
@staticmethod
@deprecated(deprecated_in="3.0")
def dedup_columns(*columns_args: list[Column] | None) -> list[Column]:
# dedup groupby and columns while preserving order
labels: list[str] = []
deduped_columns: list[Column] = []
for columns in columns_args:
for column in columns or []:
label = get_column_name(column)
if label not in labels:
deduped_columns.append(column)
return deduped_columns
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict: # pylint: disable=too-many-locals
"""Building a query object"""
self.process_query_filters()
metrics = self.all_metrics or []
groupby = self.dedup_columns(self.groupby, self.form_data.get("columns"))
is_timeseries = self.is_timeseries
if DTTM_ALIAS in (groupby_labels := get_column_names(groupby)):
del groupby[groupby_labels.index(DTTM_ALIAS)]
is_timeseries = True
granularity = self.form_data.get("granularity_sqla")
limit = int(self.form_data.get("limit") or 0)
timeseries_limit_metric = self.form_data.get("timeseries_limit_metric")
# apply row limit to query
row_limit = int(self.form_data.get("row_limit") or config["ROW_LIMIT"])
row_limit = apply_max_row_limit(row_limit)
# default order direction
order_desc = self.form_data.get("order_desc", True)
try:
since, until = get_since_until(
relative_start=relative_start,
relative_end=relative_end,
time_range=self.form_data.get("time_range"),
since=self.form_data.get("since"),
until=self.form_data.get("until"),
)
except ValueError as ex:
raise QueryObjectValidationError(str(ex)) from ex
time_shift = self.form_data.get("time_shift", "")
self.time_shift = parse_past_timedelta(time_shift)
from_dttm = None if since is None else (since - self.time_shift)
to_dttm = None if until is None else (until - self.time_shift)
if from_dttm and to_dttm and from_dttm > to_dttm:
raise QueryObjectValidationError(
_("From date cannot be larger than to date")
)
self.from_dttm = from_dttm
self.to_dttm = to_dttm
# validate sql filters
for param in ("where", "having"):
clause = self.form_data.get(param)
if clause:
sanitized_clause = sanitize_clause(clause)
if sanitized_clause != clause:
self.form_data[param] = sanitized_clause
# extras are used to query elements specific to a datasource type
# for instance the extra where clause that applies only to Tables
extras = {
"having": self.form_data.get("having", ""),
"time_grain_sqla": self.form_data.get("time_grain_sqla"),
"where": self.form_data.get("where", ""),
}
return {
"granularity": granularity,
"from_dttm": from_dttm,
"to_dttm": to_dttm,
"is_timeseries": is_timeseries,
"groupby": groupby,
"metrics": metrics,
"row_limit": row_limit,
"filter": self.form_data.get("filters", []),
"timeseries_limit": limit,
"extras": extras,
"timeseries_limit_metric": timeseries_limit_metric,
"order_desc": order_desc,
}
@property
@deprecated(deprecated_in="3.0")
def cache_timeout(self) -> int:
if self.form_data.get("cache_timeout") is not None:
return int(self.form_data["cache_timeout"])
if self.datasource.cache_timeout is not None:
return self.datasource.cache_timeout
if (
hasattr(self.datasource, "database")
and self.datasource.database.cache_timeout
) is not None:
return self.datasource.database.cache_timeout
if config["DATA_CACHE_CONFIG"].get("CACHE_DEFAULT_TIMEOUT") is not None:
return config["DATA_CACHE_CONFIG"]["CACHE_DEFAULT_TIMEOUT"]
return config["CACHE_DEFAULT_TIMEOUT"]
@deprecated(deprecated_in="3.0")
def get_json(self) -> str:
return json.dumps(
self.get_payload(), default=json.json_int_dttm_ser, ignore_nan=True
)
@deprecated(deprecated_in="3.0")
def cache_key(self, query_obj: QueryObjectDict, **extra: Any) -> str:
"""
The cache key is made out of the key/values in `query_obj`, plus any
other key/values in `extra`.
We remove datetime bounds that are hard values, and replace them with
the use-provided inputs to bounds, which may be time-relative (as in
"5 days ago" or "now").
The `extra` arguments are currently used by time shift queries, since
different time shifts will differ only in the `from_dttm`, `to_dttm`,
`inner_from_dttm`, and `inner_to_dttm` values which are stripped.
"""
cache_dict = copy.copy(query_obj)
cache_dict.update(extra)
for k in ["from_dttm", "to_dttm", "inner_from_dttm", "inner_to_dttm"]:
if k in cache_dict:
del cache_dict[k]
cache_dict["time_range"] = self.form_data.get("time_range")
cache_dict["datasource"] = self.datasource.uid
cache_dict["extra_cache_keys"] = self.datasource.get_extra_cache_keys(query_obj)
cache_dict["rls"] = security_manager.get_rls_cache_key(self.datasource)
cache_dict["changed_on"] = self.datasource.changed_on
json_data = self.json_dumps(cache_dict, sort_keys=True)
return md5_sha_from_str(json_data)
@deprecated(deprecated_in="3.0")
def get_payload(self, query_obj: QueryObjectDict | None = None) -> VizPayload:
"""Returns a payload of metadata and data"""
try:
self.run_extra_queries()
except SupersetSecurityException as ex:
error = dataclasses.asdict(ex.error)
self.errors.append(error)
self.status = QueryStatus.FAILED
payload = self.get_df_payload(query_obj)
# if payload does not have a df, we are raising an error here.
df = cast(Optional[pd.DataFrame], payload["df"])
if self.status != QueryStatus.FAILED:
payload["data"] = self.get_data(df)
if "df" in payload:
del payload["df"]
applied_filter_columns = self.applied_filter_columns or []
rejected_filter_columns = self.rejected_filter_columns or []
applied_time_extras = self.form_data.get("applied_time_extras", {})
applied_time_columns, rejected_time_columns = utils.get_time_filter_status(
self.datasource, applied_time_extras
)
payload["applied_filters"] = [
{"column": get_column_name(col)} for col in applied_filter_columns
] + applied_time_columns
payload["rejected_filters"] = [
{
"reason": ExtraFiltersReasonType.COL_NOT_IN_DATASOURCE,
"column": get_column_name(col),
}
for col in rejected_filter_columns
] + rejected_time_columns
if df is not None:
payload["colnames"] = list(df.columns)
return payload
@deprecated(deprecated_in="3.0")
def get_df_payload( # pylint: disable=too-many-statements
self, query_obj: QueryObjectDict | None = None, **kwargs: Any
) -> dict[str, Any]:
"""Handles caching around the df payload retrieval"""
if not query_obj:
query_obj = self.query_obj()
cache_key = self.cache_key(query_obj, **kwargs) if query_obj else None
cache_value = None
logger.info("Cache key: %s", cache_key)
is_loaded = False
stacktrace = None
df = None
cache_timeout = self.cache_timeout
force = self.force or cache_timeout == -1
if cache_key and cache_manager.data_cache and not force:
cache_value = cache_manager.data_cache.get(cache_key)
if cache_value:
stats_logger.incr("loading_from_cache")
try:
df = cache_value["df"]
self.query = cache_value["query"]
self.applied_filter_columns = cache_value.get(
"applied_filter_columns", []
)
self.rejected_filter_columns = cache_value.get(
"rejected_filter_columns", []
)
self.status = QueryStatus.SUCCESS
is_loaded = True
stats_logger.incr("loaded_from_cache")
except Exception as ex: # pylint: disable=broad-except
logger.exception(ex)
logger.error(
"Error reading cache: %s",
utils.error_msg_from_exception(ex),
exc_info=True,
)
logger.info("Serving from cache")
if query_obj and not is_loaded:
if self.force_cached:
logger.warning(
"force_cached (viz.py): value not found for cache key %s",
cache_key,
)
raise CacheLoadError(_("Cached value not found"))
try:
invalid_columns = [
col
for col in get_column_names_from_columns(
query_obj.get("columns") or []
)
+ get_column_names_from_columns(query_obj.get("groupby") or [])
+ utils.get_column_names_from_metrics(
cast(list[Metric], query_obj.get("metrics") or [])
)
if col not in self.datasource.column_names
]
if invalid_columns:
raise QueryObjectValidationError(
_(
"Columns missing in datasource: %(invalid_columns)s",
invalid_columns=invalid_columns,
)
)
df = self.get_df(query_obj)
if self.status != QueryStatus.FAILED:
stats_logger.incr("loaded_from_source")
if not self.force:
stats_logger.incr("loaded_from_source_without_force")
is_loaded = True
except QueryObjectValidationError as ex:
error = dataclasses.asdict(
SupersetError(
message=str(ex),
level=ErrorLevel.ERROR,
error_type=SupersetErrorType.VIZ_GET_DF_ERROR,
)
)
self.errors.append(error)
self.status = QueryStatus.FAILED
except Exception as ex: # pylint: disable=broad-except
logger.exception(ex)
error = dataclasses.asdict(
SupersetError(
message=str(ex),
level=ErrorLevel.ERROR,
error_type=SupersetErrorType.VIZ_GET_DF_ERROR,
)
)
self.errors.append(error)
self.status = QueryStatus.FAILED
stacktrace = utils.get_stacktrace()
if is_loaded and cache_key and self.status != QueryStatus.FAILED:
set_and_log_cache(
cache_instance=cache_manager.data_cache,
cache_key=cache_key,
cache_value={"df": df, "query": self.query},
cache_timeout=cache_timeout,
datasource_uid=self.datasource.uid,
)
return {
"cache_key": cache_key,
"cached_dttm": cache_value["dttm"] if cache_value is not None else None,
"cache_timeout": cache_timeout,
"df": df,
"errors": self.errors,
"form_data": self.form_data,
"is_cached": cache_value is not None,
"query": self.query,
"from_dttm": self.from_dttm,
"to_dttm": self.to_dttm,
"status": self.status,
"stacktrace": stacktrace,
"rowcount": len(df.index) if df is not None else 0,
"colnames": list(df.columns) if df is not None else None,
"coltypes": utils.extract_dataframe_dtypes(df, self.datasource)
if df is not None
else None,
}
@staticmethod
@deprecated(deprecated_in="3.0")
def json_dumps(query_obj: Any, sort_keys: bool = False) -> str:
return json.dumps(
query_obj,
default=json.json_int_dttm_ser,
ignore_nan=True,
sort_keys=sort_keys,
)
@staticmethod
@deprecated(deprecated_in="3.0")
def has_error(payload: VizPayload) -> bool:
return (
payload.get("status") == QueryStatus.FAILED
or payload.get("error") is not None
or bool(payload.get("errors"))
)
@deprecated(deprecated_in="3.0")
def payload_json_and_has_error(self, payload: VizPayload) -> tuple[str, bool]:
return self.json_dumps(payload), self.has_error(payload)
@property
@deprecated(deprecated_in="3.0")
def data(self) -> dict[str, Any]:
"""This is the data object serialized to the js layer"""
content = {
"form_data": self.form_data,
"token": self.token,
"viz_name": self.viz_type,
"filter_select_enabled": self.datasource.filter_select_enabled,
}
return content
@deprecated(deprecated_in="3.0")
def get_csv(self) -> str | None:
df = self.get_df_payload()["df"] # leverage caching logic
include_index = not isinstance(df.index, pd.RangeIndex)
return csv.df_to_escaped_csv(df, index=include_index, **config["CSV_EXPORT"])
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
return df.to_dict(orient="records")
@property
@deprecated(deprecated_in="3.0")
def json_data(self) -> str:
return json.dumps(self.data)
@deprecated(deprecated_in="3.0")
def raise_for_access(self) -> None:
"""
Raise an exception if the user cannot access the resource.
:raises SupersetSecurityException: If the user cannot access the resource
"""
security_manager.raise_for_access(viz=self)
class TimeTableViz(BaseViz):
"""A data table with rich time-series related columns"""
viz_type = "time_table"
verbose_name = _("Time Table View")
credits = 'a <a href="https://github.com/airbnb/superset">Superset</a> original'
is_timeseries = True
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
if not self.form_data.get("metrics"):
raise QueryObjectValidationError(_("Pick at least one metric"))
if self.form_data.get("groupby") and len(self.form_data["metrics"]) > 1:
raise QueryObjectValidationError(
_("When using 'Group By' you are limited to use a single metric")
)
sort_by = utils.get_first_metric_name(query_obj["metrics"])
is_asc = not query_obj.get("order_desc")
query_obj["orderby"] = [(sort_by, is_asc)]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
columns = None
values: list[str] | str = self.metric_labels
if self.form_data.get("groupby"):
values = self.metric_labels[0]
columns = get_column_names(self.form_data.get("groupby"))
pt = df.pivot_table(index=DTTM_ALIAS, columns=columns, values=values)
pt.index = pt.index.map(str)
pt = pt.sort_index()
return {
"records": pt.to_dict(orient="index"),
"columns": list(pt.columns),
"is_group_by": bool(self.form_data.get("groupby")),
}
class CalHeatmapViz(BaseViz):
"""Calendar heatmap."""
viz_type = "cal_heatmap"
verbose_name = _("Calendar Heatmap")
credits = "<a href=https://github.com/wa0x6e/cal-heatmap>cal-heatmap</a>"
is_timeseries = True
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData: # pylint: disable=too-many-locals
if df.empty:
return None
form_data = self.form_data
data = {}
records = df.to_dict("records")
for metric in self.metric_labels:
values = {}
for query_obj in records:
v = query_obj[DTTM_ALIAS]
if hasattr(v, "value"):
v = v.value
values[str(v / 10**9)] = query_obj.get(metric)
data[metric] = values
try:
start, end = get_since_until(
relative_start=relative_start,
relative_end=relative_end,
time_range=form_data.get("time_range"),
since=form_data.get("since"),
until=form_data.get("until"),
)
except ValueError as ex:
raise QueryObjectValidationError(str(ex)) from ex
if not start or not end:
raise QueryObjectValidationError(
"Please provide both time bounds (Since and Until)"
)
domain = form_data.get("domain_granularity")
diff_delta = rdelta.relativedelta(end, start)
diff_secs = (end - start).total_seconds()
if domain == "year":
range_ = end.year - start.year + 1
elif domain == "month":
range_ = diff_delta.years * 12 + diff_delta.months + 1
elif domain == "week":
range_ = diff_delta.years * 53 + diff_delta.weeks + 1
elif domain == "day":
range_ = diff_secs // (24 * 60 * 60) + 1 # type: ignore
else:
range_ = diff_secs // (60 * 60) + 1 # type: ignore
return {
"data": data,
"start": start,
"domain": domain,
"subdomain": form_data.get("subdomain_granularity"),
"range": range_,
}
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
query_obj["metrics"] = self.form_data.get("metrics")
mapping = {
"min": "PT1M",
"hour": "PT1H",
"day": "P1D",
"week": "P1W",
"month": "P1M",
"year": "P1Y",
}
query_obj["extras"]["time_grain_sqla"] = mapping[
self.form_data.get("subdomain_granularity", "min")
]
return query_obj
class NVD3Viz(BaseViz):
"""Base class for all nvd3 vizs"""
credits = '<a href="http://nvd3.org/">NVD3.org</a>'
viz_type: str | None = None
verbose_name = "Base NVD3 Viz"
is_timeseries = False
class BubbleViz(NVD3Viz):
"""Based on the NVD3 bubble chart"""
viz_type = "bubble"
verbose_name = _("Bubble Chart")
is_timeseries = False
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
query_obj["groupby"] = [self.form_data.get("entity")]
if self.form_data.get("series"):
query_obj["groupby"].append(self.form_data.get("series"))
# dedup groupby if it happens to be the same
query_obj["groupby"] = self.dedup_columns(query_obj["groupby"])
# pylint: disable=attribute-defined-outside-init
self.x_metric = self.form_data["x"]
self.y_metric = self.form_data["y"]
self.z_metric = self.form_data["size"]
self.entity = self.form_data.get("entity")
self.series = self.form_data.get("series") or self.entity
query_obj["row_limit"] = self.form_data.get("limit")
query_obj["metrics"] = [self.z_metric, self.x_metric, self.y_metric]
if len(set(self.metric_labels)) < 3:
raise QueryObjectValidationError(_("Please use 3 different metric labels"))
if not all(query_obj["metrics"] + [self.entity]):
raise QueryObjectValidationError(_("Pick a metric for x, y and size"))
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
df["x"] = df[[utils.get_metric_name(self.x_metric)]]
df["y"] = df[[utils.get_metric_name(self.y_metric)]]
df["size"] = df[[utils.get_metric_name(self.z_metric)]]
df["shape"] = "circle"
df["group"] = df[[get_column_name(self.series)]] # type: ignore
series: dict[Any, list[Any]] = defaultdict(list)
for row in df.to_dict(orient="records"):
series[row["group"]].append(row)
chart_data = []
for k, v in series.items():
chart_data.append({"key": k, "values": v})
return chart_data
class BulletViz(NVD3Viz):
"""Based on the NVD3 bullet chart"""
viz_type = "bullet"
verbose_name = _("Bullet Chart")
is_timeseries = False
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
form_data = self.form_data
query_obj = super().query_obj()
self.metric = form_data[ # pylint: disable=attribute-defined-outside-init
"metric"
]
query_obj["metrics"] = [self.metric]
if not self.metric:
raise QueryObjectValidationError(_("Pick a metric to display"))
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
df["metric"] = df[[utils.get_metric_name(self.metric)]]
values = df["metric"].values
return {
"measures": values.tolist(),
}
class NVD3TimeSeriesViz(NVD3Viz):
"""A rich line chart component with tons of options"""
viz_type = "line"
verbose_name = _("Time Series - Line Chart")
sort_series = False
is_timeseries = True
pivot_fill_value: int | None = None
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
sort_by = self.form_data.get(
"timeseries_limit_metric"
) or utils.get_first_metric_name(query_obj.get("metrics") or [])
is_asc = not self.form_data.get("order_desc")
if sort_by:
sort_by_label = utils.get_metric_name(sort_by)
if sort_by_label not in utils.get_metric_names(query_obj["metrics"]):
query_obj["metrics"].append(sort_by)
query_obj["orderby"] = [(sort_by, is_asc)]
return query_obj
@deprecated(deprecated_in="3.0")
def to_series( # pylint: disable=too-many-branches
self, df: pd.DataFrame, classed: str = "", title_suffix: str = ""
) -> list[dict[str, Any]]:
cols = []
for col in df.columns:
if col == "":
cols.append("N/A")
elif col is None:
cols.append("NULL")
else:
cols.append(col)
df.columns = cols
series = df.to_dict("series")
chart_data = []
for name in df.T.index.tolist():
ys = series[name]
if df[name].dtype.kind not in "biufc":
continue
series_title: list[str] | str | tuple[str, ...]
if isinstance(name, list):
series_title = [str(title) for title in name]
elif isinstance(name, tuple):
series_title = tuple(str(title) for title in name)
else:
series_title = str(name)
if (
isinstance(series_title, (list, tuple))
and len(series_title) > 1
and len(self.metric_labels) == 1
):
# Removing metric from series name if only one metric
series_title = series_title[1:]
if title_suffix:
if isinstance(series_title, str):
series_title = (series_title, title_suffix)
elif isinstance(series_title, list):
series_title = series_title + [title_suffix]
elif isinstance(series_title, tuple):
series_title = series_title + (title_suffix,)
values = []
non_nan_cnt = 0
for ds in df.index:
if ds in ys:
data = {"x": ds, "y": ys[ds]}
if not np.isnan(ys[ds]):
non_nan_cnt += 1
else:
data = {}
values.append(data)
if non_nan_cnt == 0:
continue
data = {"key": series_title, "values": values}
if classed:
data["classed"] = classed
chart_data.append(data)
return chart_data
@deprecated(deprecated_in="3.0")
def process_data(self, df: pd.DataFrame, aggregate: bool = False) -> VizData:
if df.empty:
return df
if aggregate:
df = df.pivot_table(
index=DTTM_ALIAS,
columns=get_column_names(self.form_data.get("groupby")),
values=self.metric_labels,
fill_value=0,
aggfunc=sum,
)
else:
df = df.pivot_table(
index=DTTM_ALIAS,
columns=get_column_names(self.form_data.get("groupby")),
values=self.metric_labels,
fill_value=self.pivot_fill_value,
)
rule = self.form_data.get("resample_rule")
method = self.form_data.get("resample_method")
if rule and method:
df = getattr(df.resample(rule), method)()
if self.sort_series:
dfs = df.sum()
dfs.sort_values(ascending=False, inplace=True)
df = df[dfs.index]
df = self.apply_rolling(df)
if self.form_data.get("contribution"):
dft = df.T
df = (dft / dft.sum()).T
return df
@deprecated(deprecated_in="3.0")
def run_extra_queries(self) -> None:
time_compare = self.form_data.get("time_compare") or []
# backwards compatibility
if not isinstance(time_compare, list):
time_compare = [time_compare]
for option in time_compare:
query_object = self.query_obj()
try:
delta = parse_past_timedelta(option)
except ValueError as ex:
raise QueryObjectValidationError(str(ex)) from ex
query_object["inner_from_dttm"] = query_object["from_dttm"]
query_object["inner_to_dttm"] = query_object["to_dttm"]
if not query_object["from_dttm"] or not query_object["to_dttm"]:
raise QueryObjectValidationError(
_(
"An enclosed time range (both start and end) must be specified "
"when using a Time Comparison."
)
)
query_object["from_dttm"] -= delta
query_object["to_dttm"] -= delta
df2 = self.get_df_payload(query_object, time_compare=option).get("df")
if df2 is not None and DTTM_ALIAS in df2:
dttm_series = df2[DTTM_ALIAS] + delta
df2 = df2.drop(DTTM_ALIAS, axis=1)
df2 = pd.concat([dttm_series, df2], axis=1)
label = f"{option} offset"
df2 = self.process_data(df2)
self._extra_chart_data.append((label, df2))
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
comparison_type = self.form_data.get("comparison_type") or "values"
df = self.process_data(df)
if comparison_type == "values":
# Filter out series with all NaN
chart_data = self.to_series(df.dropna(axis=1, how="all"))
for i, (label, df2) in enumerate(self._extra_chart_data):
chart_data.extend(
self.to_series(df2, classed=f"time-shift-{i}", title_suffix=label)
)
else:
chart_data = []
for i, (label, df2) in enumerate(self._extra_chart_data):
# reindex df2 into the df2 index
combined_index = df.index.union(df2.index)
df2 = (
df2.reindex(combined_index)
.interpolate(method="time")
.reindex(df.index)
)
if comparison_type == "absolute":
diff = df - df2
elif comparison_type == "percentage":
diff = (df - df2) / df2
elif comparison_type == "ratio":
diff = df / df2
else:
raise QueryObjectValidationError(
f"Invalid `comparison_type`: {comparison_type}"
)
# remove leading/trailing NaNs from the time shift difference
diff = diff[diff.first_valid_index() : diff.last_valid_index()]
chart_data.extend(
self.to_series(diff, classed=f"time-shift-{i}", title_suffix=label)
)
if not self.sort_series:
chart_data = sorted(chart_data, key=lambda x: tuple(x["key"]))
return chart_data
class NVD3TimeSeriesBarViz(NVD3TimeSeriesViz):
"""A bar chart where the x axis is time"""
viz_type = "bar"
sort_series = True
verbose_name = _("Time Series - Bar Chart")
class NVD3TimePivotViz(NVD3TimeSeriesViz):
"""Time Series - Periodicity Pivot"""
viz_type = "time_pivot"
sort_series = True
verbose_name = _("Time Series - Period Pivot")
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
query_obj["metrics"] = [self.form_data.get("metric")]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
df = self.process_data(df)
freq = to_offset(self.form_data.get("freq"))
try:
freq = type(freq)(freq.n, normalize=True, **freq.kwds)
except ValueError:
freq = type(freq)(freq.n, **freq.kwds)
df.index.name = None
df[DTTM_ALIAS] = df.index.map(freq.rollback)
df["ranked"] = df[DTTM_ALIAS].rank(method="dense", ascending=False) - 1
df.ranked = df.ranked.map(int)
df["series"] = "-" + df.ranked.map(str)
df["series"] = df["series"].str.replace("-0", "current")
rank_lookup = {
row["series"]: row["ranked"] for row in df.to_dict(orient="records")
}
max_ts = df[DTTM_ALIAS].max()
max_rank = df["ranked"].max()
df[DTTM_ALIAS] = df.index + (max_ts - df[DTTM_ALIAS])
df = df.pivot_table(
index=DTTM_ALIAS,
columns="series",
values=utils.get_metric_name(self.form_data["metric"]),
)
chart_data = self.to_series(df)
for series in chart_data:
series["rank"] = rank_lookup[series["key"]]
series["perc"] = 1 - (series["rank"] / (max_rank + 1))
return chart_data
class NVD3CompareTimeSeriesViz(NVD3TimeSeriesViz):
"""A line chart component where you can compare the % change over time"""
viz_type = "compare"
verbose_name = _("Time Series - Percent Change")
class NVD3TimeSeriesStackedViz(NVD3TimeSeriesViz):
"""A rich stack area chart"""
viz_type = "area"
verbose_name = _("Time Series - Stacked")
sort_series = True
pivot_fill_value = 0
class HistogramViz(BaseViz):
"""Histogram"""
viz_type = "histogram"
verbose_name = _("Histogram")
is_timeseries = False
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
"""Returns the query object for this visualization"""
query_obj = super().query_obj()
numeric_columns = self.form_data.get("all_columns_x")
if numeric_columns is None:
raise QueryObjectValidationError(
_("Must have at least one numeric column specified")
)
self.columns = ( # pylint: disable=attribute-defined-outside-init
numeric_columns
)
query_obj["columns"] = numeric_columns + self.groupby
# override groupby entry to avoid aggregation
query_obj["groupby"] = None
query_obj["metrics"] = None
return query_obj
@deprecated(deprecated_in="3.0")
def labelify(self, keys: list[str] | str, column: str) -> str:
if isinstance(keys, str):
keys = [keys]
# removing undesirable characters
labels = [re.sub(r"\W+", r"_", k) for k in keys]
if len(self.columns) > 1 or not self.groupby:
# Only show numeric column in label if there are many
labels = [column] + labels
return "__".join(labels)
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
"""Returns the chart data"""
if df.empty:
return None
chart_data = []
if len(self.groupby) > 0:
groups = df.groupby(get_column_names(self.groupby))
else:
groups = [((), df)]
for keys, data in groups:
chart_data.extend(
[
{
"key": self.labelify(keys, get_column_name(column)),
"values": data[get_column_name(column)].tolist(),
}
for column in self.columns
]
)
return chart_data
class DistributionBarViz(BaseViz):
"""A good old bar chart"""
viz_type = "dist_bar"
verbose_name = _("Distribution - Bar Chart")
is_timeseries = False
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
if len(query_obj["groupby"]) < len(self.form_data.get("groupby") or []) + len(
self.form_data.get("columns") or []
):
raise QueryObjectValidationError(
_("Can't have overlap between Series and Breakdowns")
)
if not self.form_data.get("metrics"):
raise QueryObjectValidationError(_("Pick at least one metric"))
if not self.form_data.get("groupby"):
raise QueryObjectValidationError(_("Pick at least one field for [Series]"))
if sort_by := self.form_data.get("timeseries_limit_metric"):
sort_by_label = utils.get_metric_name(sort_by)
if sort_by_label not in utils.get_metric_names(query_obj["metrics"]):
query_obj["metrics"].append(sort_by)
query_obj["orderby"] = [
(sort_by, not self.form_data.get("order_desc", True))
]
elif query_obj["metrics"]:
# Legacy behavior of sorting by first metric by default
first_metric = query_obj["metrics"][0]
query_obj["orderby"] = [
(first_metric, not self.form_data.get("order_desc", True))
]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData: # pylint: disable=too-many-locals
if df.empty:
return None
metrics = self.metric_labels
columns = get_column_names(self.form_data.get("columns"))
groupby = get_column_names(self.groupby)
# pandas will throw away nulls when grouping/pivoting,
# so we substitute NULL_STRING for any nulls in the necessary columns
filled_cols = groupby + columns
df = df.copy()
df[filled_cols] = df[filled_cols].fillna(value=NULL_STRING)
sortby = utils.get_metric_name(
self.form_data.get("timeseries_limit_metric") or metrics[0]
)
row = df.groupby(groupby)[sortby].sum().copy()
is_asc = not self.form_data.get("order_desc")
row.sort_values(ascending=is_asc, inplace=True)
pt = df.pivot_table(index=groupby, columns=columns, values=metrics)
if self.form_data.get("contribution"):
pt = pt.T
pt = (pt / pt.sum()).T
pt = pt.reindex(row.index)
# Re-order the columns adhering to the metric ordering.
pt = pt[metrics]
chart_data = []
for name, ys in pt.items():
if pt[name].dtype.kind not in "biufc" or name in groupby:
continue
if isinstance(name, str):
series_title = name
else:
offset = 0 if len(metrics) > 1 else 1
series_title = ", ".join([str(s) for s in name[offset:]])
values = []
for i, v in ys.items():
x = i
if isinstance(x, (tuple, list)):
x = ", ".join([str(s) for s in x])
else:
x = str(x)
values.append({"x": x, "y": v})
chart_data.append({"key": series_title, "values": values})
return chart_data
class SankeyViz(BaseViz):
"""A Sankey diagram that requires a parent-child dataset"""
viz_type = "sankey"
verbose_name = _("Sankey")
is_timeseries = False
credits = '<a href="https://www.npmjs.com/package/d3-sankey">d3-sankey on npm</a>'
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
if len(query_obj["groupby"]) != 2:
raise QueryObjectValidationError(
_("Pick exactly 2 columns as [Source / Target]")
)
query_obj["metrics"] = [self.form_data["metric"]]
if self.form_data.get("sort_by_metric", False):
query_obj["orderby"] = [(query_obj["metrics"][0], False)]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
source, target = get_column_names(self.groupby)
(value,) = self.metric_labels
df.rename(
columns={
source: "source",
target: "target",
value: "value",
},
inplace=True,
)
df["source"] = df["source"].astype(str)
df["target"] = df["target"].astype(str)
recs = df.to_dict(orient="records")
hierarchy: dict[str, set[str]] = defaultdict(set)
for row in recs:
hierarchy[row["source"]].add(row["target"])
@deprecated(deprecated_in="3.0")
def find_cycle(graph: dict[str, set[str]]) -> tuple[str, str] | None:
"""Whether there's a cycle in a directed graph"""
path = set()
@deprecated(deprecated_in="3.0")
def visit(vertex: str) -> tuple[str, str] | None:
path.add(vertex)
for neighbour in graph.get(vertex, ()):
if neighbour in path or visit(neighbour):
return (vertex, neighbour)
path.remove(vertex)
return None
for vertex in graph:
cycle = visit(vertex)
if cycle:
return cycle
return None
cycle = find_cycle(hierarchy)
if cycle:
raise QueryObjectValidationError(
_(
"There's a loop in your Sankey, please provide a tree. "
"Here's a faulty link: {}"
).format(cycle)
)
return recs
class ChordViz(BaseViz):
"""A Chord diagram"""
viz_type = "chord"
verbose_name = _("Directed Force Layout")
credits = '<a href="https://github.com/d3/d3-chord">Bostock</a>'
is_timeseries = False
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
query_obj["groupby"] = [
self.form_data.get("groupby"),
self.form_data.get("columns"),
]
query_obj["metrics"] = [self.form_data.get("metric")]
if self.form_data.get("sort_by_metric", False):
query_obj["orderby"] = [(query_obj["metrics"][0], False)]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
df.columns = ["source", "target", "value"]
# Preparing a symmetrical matrix like d3.chords calls for
nodes = list(set(df["source"]) | set(df["target"]))
matrix = {}
for source, target in product(nodes, nodes):
matrix[(source, target)] = 0
for source, target, value in df.to_records(index=False):
matrix[(source, target)] = value
return {
"nodes": list(nodes),
"matrix": [[matrix[(n1, n2)] for n1 in nodes] for n2 in nodes],
}
class CountryMapViz(BaseViz):
"""A country centric"""
viz_type = "country_map"
verbose_name = _("Country Map")
is_timeseries = False
credits = "From bl.ocks.org By john-guerra"
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
metric = self.form_data.get("metric")
entity = self.form_data.get("entity")
if not self.form_data.get("select_country"):
raise QueryObjectValidationError("Must specify a country")
if not metric:
raise QueryObjectValidationError("Must specify a metric")
if not entity:
raise QueryObjectValidationError("Must provide ISO codes")
query_obj["metrics"] = [metric]
query_obj["groupby"] = [entity]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
cols = get_column_names([self.form_data.get("entity")]) # type: ignore
metric = self.metric_labels[0]
cols += [metric]
ndf = df[cols]
df = ndf
df.columns = ["country_id", "metric"]
return df.to_dict(orient="records")
class WorldMapViz(BaseViz):
"""A country centric world map"""
viz_type = "world_map"
verbose_name = _("World Map")
is_timeseries = False
credits = 'datamaps on <a href="https://www.npmjs.com/package/datamaps">npm</a>'
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
query_obj["groupby"] = [self.form_data["entity"]]
if self.form_data.get("sort_by_metric", False):
query_obj["orderby"] = [(query_obj["metrics"][0], False)]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
# pylint: disable=import-outside-toplevel
from superset.examples import countries
cols = get_column_names([self.form_data.get("entity")]) # type: ignore
metric = utils.get_metric_name(self.form_data["metric"])
secondary_metric = (
utils.get_metric_name(self.form_data["secondary_metric"])
if self.form_data.get("secondary_metric")
else None
)
columns = ["country", "m1", "m2"]
if metric == secondary_metric:
ndf = df[cols]
ndf["m1"] = df[metric]
ndf["m2"] = ndf["m1"]
else:
if secondary_metric:
cols += [metric, secondary_metric]
else:
cols += [metric]
columns = ["country", "m1"]
ndf = df[cols]
df = ndf
df.columns = columns
data = df.to_dict(orient="records")
for row in data:
country = None
if isinstance(row["country"], str):
if "country_fieldtype" in self.form_data:
country = countries.get(
self.form_data["country_fieldtype"], row["country"]
)
if country:
row["country"] = country["cca3"]
row["latitude"] = country["lat"]
row["longitude"] = country["lng"]
row["name"] = country["name"]
else:
row["country"] = "XXX"
return data
class ParallelCoordinatesViz(BaseViz):
"""Interactive parallel coordinate implementation
Uses this amazing javascript library
https://github.com/syntagmatic/parallel-coordinates
"""
viz_type = "para"
verbose_name = _("Parallel Coordinates")
credits = (
'<a href="https://syntagmatic.github.io/parallel-coordinates/">'
"Syntagmatic's library</a>"
)
is_timeseries = False
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
query_obj["groupby"] = [self.form_data.get("series")]
if sort_by := self.form_data.get("timeseries_limit_metric"):
sort_by_label = utils.get_metric_name(sort_by)
if sort_by_label not in utils.get_metric_names(query_obj["metrics"]):
query_obj["metrics"].append(sort_by)
if self.form_data.get("order_desc"):
query_obj["orderby"] = [
(sort_by, not self.form_data.get("order_desc", True))
]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
return df.to_dict(orient="records")
class HeatmapViz(BaseViz):
"""A nice heatmap visualization that support high density through canvas"""
viz_type = "heatmap"
verbose_name = _("Heatmap")
is_timeseries = False
credits = (
'inspired from mbostock @<a href="http://bl.ocks.org/mbostock/3074470">'
"bl.ocks.org</a>"
)
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
query_obj["metrics"] = [self.form_data.get("metric")]
query_obj["groupby"] = [
self.form_data.get("all_columns_x"),
self.form_data.get("all_columns_y"),
]
if self.form_data.get("sort_by_metric", False):
query_obj["orderby"] = [(query_obj["metrics"][0], False)]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
x = get_column_name(self.form_data.get("all_columns_x")) # type: ignore
y = get_column_name(self.form_data.get("all_columns_y")) # type: ignore
v = self.metric_labels[0]
if x == y:
df.columns = ["x", "y", "v"]
else:
df = df[[x, y, v]]
df.columns = ["x", "y", "v"]
norm = self.form_data.get("normalize_across")
overall = False
max_ = df.v.max()
min_ = df.v.min()
if norm == "heatmap":
overall = True
else:
gb = df.groupby(norm, group_keys=False)
if len(gb) <= 1:
overall = True
else:
df["perc"] = gb.apply(
lambda x: (x.v - x.v.min()) / (x.v.max() - x.v.min())
)
df["rank"] = gb.apply(lambda x: x.v.rank(pct=True))
if overall:
df["perc"] = (df.v - min_) / (max_ - min_)
df["rank"] = df.v.rank(pct=True)
return {"records": df.to_dict(orient="records"), "extents": [min_, max_]}
class HorizonViz(NVD3TimeSeriesViz):
"""Horizon chart
https://www.npmjs.com/package/d3-horizon-chart
"""
viz_type = "horizon"
verbose_name = _("Horizon Charts")
credits = (
'<a href="https://www.npmjs.com/package/d3-horizon-chart">'
"d3-horizon-chart</a>"
)
class MapboxViz(BaseViz):
"""Rich maps made with Mapbox"""
viz_type = "mapbox"
verbose_name = _("Mapbox")
is_timeseries = False
credits = "<a href=https://www.mapbox.com/mapbox-gl-js/api/>Mapbox GL JS</a>"
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
label_col = self.form_data.get("mapbox_label")
if not self.form_data.get("groupby"):
if (
self.form_data.get("all_columns_x") is None
or self.form_data.get("all_columns_y") is None
):
raise QueryObjectValidationError(
_("[Longitude] and [Latitude] must be set")
)
query_obj["columns"] = [
self.form_data.get("all_columns_x"),
self.form_data.get("all_columns_y"),
]
if label_col and len(label_col) >= 1:
if label_col[0] == "count":
raise QueryObjectValidationError(
_(
"Must have a [Group By] column to have 'count' as the "
+ "[Label]"
)
)
query_obj["columns"].append(label_col[0])
if self.form_data.get("point_radius") != "Auto":
query_obj["columns"].append(self.form_data.get("point_radius"))
# Ensure this value is sorted so that it does not
# cause the cache key generation (which hashes the
# query object) to generate different keys for values
# that should be considered the same.
query_obj["columns"] = sorted(set(query_obj["columns"]))
else:
# Ensuring columns chosen are all in group by
if (
label_col
and len(label_col) >= 1
and label_col[0] != "count"
and label_col[0] not in self.form_data["groupby"]
):
raise QueryObjectValidationError(
_("Choice of [Label] must be present in [Group By]")
)
if (
self.form_data.get("point_radius") != "Auto"
and self.form_data.get("point_radius") not in self.form_data["groupby"]
):
raise QueryObjectValidationError(
_("Choice of [Point Radius] must be present in [Group By]")
)
if (
self.form_data.get("all_columns_x") not in self.form_data["groupby"]
or self.form_data.get("all_columns_y") not in self.form_data["groupby"]
):
raise QueryObjectValidationError(
_(
"[Longitude] and [Latitude] columns must be present in "
+ "[Group By]"
)
)
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
label_col = self.form_data.get("mapbox_label")
has_custom_metric = label_col is not None and len(label_col) > 0
metric_col = [None] * len(df.index)
if has_custom_metric:
if label_col[0] == self.form_data.get("all_columns_x"): # type: ignore
metric_col = df[self.form_data.get("all_columns_x")]
elif label_col[0] == self.form_data.get("all_columns_y"): # type: ignore
metric_col = df[self.form_data.get("all_columns_y")]
else:
metric_col = df[label_col[0]] # type: ignore
point_radius_col = (
[None] * len(df.index)
if self.form_data.get("point_radius") == "Auto"
else df[self.form_data.get("point_radius")]
)
# limiting geo precision as long decimal values trigger issues
# around json-bignumber in Mapbox
geo_precision = 10
# using geoJSON formatting
geo_json = {
"type": "FeatureCollection",
"features": [
{
"type": "Feature",
"properties": {"metric": metric, "radius": point_radius},
"geometry": {
"type": "Point",
"coordinates": [
round(lon, geo_precision),
round(lat, geo_precision),
],
},
}
for lon, lat, metric, point_radius in zip(
df[self.form_data.get("all_columns_x")],
df[self.form_data.get("all_columns_y")],
metric_col,
point_radius_col,
)
],
}
x_series, y_series = (
df[self.form_data.get("all_columns_x")],
df[self.form_data.get("all_columns_y")],
)
south_west = [x_series.min(), y_series.min()]
north_east = [x_series.max(), y_series.max()]
return {
"geoJSON": geo_json,
"hasCustomMetric": has_custom_metric,
"mapboxApiKey": config["MAPBOX_API_KEY"],
"mapStyle": self.form_data.get("mapbox_style"),
"aggregatorName": self.form_data.get("pandas_aggfunc"),
"clusteringRadius": self.form_data.get("clustering_radius"),
"pointRadiusUnit": self.form_data.get("point_radius_unit"),
"globalOpacity": self.form_data.get("global_opacity"),
"bounds": [south_west, north_east],
"renderWhileDragging": self.form_data.get("render_while_dragging"),
"tooltip": self.form_data.get("rich_tooltip"),
"color": self.form_data.get("mapbox_color"),
}
class DeckGLMultiLayer(BaseViz):
"""Pile on multiple DeckGL layers"""
viz_type = "deck_multi"
verbose_name = _("Deck.gl - Multiple Layers")
is_timeseries = False
credits = '<a href="https://uber.github.io/deck.gl/">deck.gl</a>'
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
return {}
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
# Late imports to avoid circular import issues
# pylint: disable=import-outside-toplevel
from superset import db
from superset.models.slice import Slice
slice_ids = self.form_data.get("deck_slices")
slices = db.session.query(Slice).filter(Slice.id.in_(slice_ids)).all()
return {
"mapboxApiKey": config["MAPBOX_API_KEY"],
"slices": [slc.data for slc in slices],
}
class BaseDeckGLViz(BaseViz):
"""Base class for deck.gl visualizations"""
is_timeseries = False
credits = '<a href="https://uber.github.io/deck.gl/">deck.gl</a>'
spatial_control_keys: list[str] = []
@deprecated(deprecated_in="3.0")
def get_metrics(self) -> list[str]:
# pylint: disable=attribute-defined-outside-init
self.metric = self.form_data.get("size")
return [self.metric] if self.metric else []
@deprecated(deprecated_in="3.0")
def process_spatial_query_obj(self, key: str, group_by: list[str]) -> None:
group_by.extend(self.get_spatial_columns(key))
@deprecated(deprecated_in="3.0")
def get_spatial_columns(self, key: str) -> list[str]:
spatial = self.form_data.get(key)
if spatial is None:
raise ValueError(_("Bad spatial key"))
if spatial.get("type") == "latlong":
return [spatial.get("lonCol"), spatial.get("latCol")]
if spatial.get("type") == "delimited":
return [spatial.get("lonlatCol")]
if spatial.get("type") == "geohash":
return [spatial.get("geohashCol")]
return []
@staticmethod
@deprecated(deprecated_in="3.0")
def parse_coordinates(latlong: Any) -> tuple[float, float] | None:
if not latlong:
return None
try:
point = Point(latlong)
return (point.latitude, point.longitude)
except Exception as ex:
raise SpatialException(
_("Invalid spatial point encountered: %(latlong)s", latlong=latlong)
) from ex
@staticmethod
@deprecated(deprecated_in="3.0")
def reverse_geohash_decode(geohash_code: str) -> tuple[str, str]:
lat, lng = geohash.decode(geohash_code)
return (lng, lat)
@staticmethod
@deprecated(deprecated_in="3.0")
def reverse_latlong(df: pd.DataFrame, key: str) -> None:
df[key] = [tuple(reversed(o)) for o in df[key] if isinstance(o, (list, tuple))]
@deprecated(deprecated_in="3.0")
def process_spatial_data_obj(self, key: str, df: pd.DataFrame) -> pd.DataFrame:
spatial = self.form_data.get(key)
if spatial is None:
raise ValueError(_("Bad spatial key"))
if spatial.get("type") == "latlong":
df[key] = list(
zip(
pd.to_numeric(df[spatial.get("lonCol")], errors="coerce"),
pd.to_numeric(df[spatial.get("latCol")], errors="coerce"),
)
)
elif spatial.get("type") == "delimited":
lon_lat_col = spatial.get("lonlatCol")
df[key] = df[lon_lat_col].apply(self.parse_coordinates)
del df[lon_lat_col]
elif spatial.get("type") == "geohash":
df[key] = df[spatial.get("geohashCol")].map(self.reverse_geohash_decode)
del df[spatial.get("geohashCol")]
if spatial.get("reverseCheckbox"):
self.reverse_latlong(df, key)
if df.get(key) is None:
raise NullValueException(
_(
"Encountered invalid NULL spatial entry, \
please consider filtering those out"
)
)
return df
@deprecated(deprecated_in="3.0")
def add_null_filters(self) -> None:
spatial_columns = set()
for key in self.spatial_control_keys:
for column in self.get_spatial_columns(key):
spatial_columns.add(column)
if self.form_data.get("adhoc_filters") is None:
self.form_data["adhoc_filters"] = []
if line_column := self.form_data.get("line_column"):
spatial_columns.add(line_column)
for column in sorted(spatial_columns):
filter_ = simple_filter_to_adhoc(
{"col": column, "op": "IS NOT NULL", "val": ""}
)
self.form_data["adhoc_filters"].append(filter_)
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
# add NULL filters
if self.form_data.get("filter_nulls", True):
self.add_null_filters()
query_obj = super().query_obj()
group_by: list[str] = []
for key in self.spatial_control_keys:
self.process_spatial_query_obj(key, group_by)
if self.form_data.get("dimension"):
group_by += [self.form_data["dimension"]]
if self.form_data.get("js_columns"):
group_by += self.form_data.get("js_columns") or []
# Ensure this value is sorted so that it does not
# cause the cache key generation (which hashes the
# query object) to generate different keys for values
# that should be considered the same.
group_by = sorted(set(group_by))
if metrics := self.get_metrics():
query_obj["groupby"] = group_by
query_obj["metrics"] = metrics
query_obj["columns"] = []
first_metric = query_obj["metrics"][0]
query_obj["orderby"] = [
(first_metric, not self.form_data.get("order_desc", True))
]
else:
query_obj["columns"] = group_by
return query_obj
@deprecated(deprecated_in="3.0")
def get_js_columns(self, data: dict[str, Any]) -> dict[str, Any]:
cols = self.form_data.get("js_columns") or []
return {col: data.get(col) for col in cols}
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
# Processing spatial info
for key in self.spatial_control_keys:
df = self.process_spatial_data_obj(key, df)
features = []
for data in df.to_dict(orient="records"):
feature = self.get_properties(data)
extra_props = self.get_js_columns(data)
if extra_props:
feature["extraProps"] = extra_props
features.append(feature)
return {
"features": features,
"mapboxApiKey": config["MAPBOX_API_KEY"],
"metricLabels": self.metric_labels,
}
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
raise NotImplementedError()
class DeckScatterViz(BaseDeckGLViz):
"""deck.gl's ScatterLayer"""
viz_type = "deck_scatter"
verbose_name = _("Deck.gl - Scatter plot")
spatial_control_keys = ["spatial"]
is_timeseries = True
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
# pylint: disable=attribute-defined-outside-init
self.is_timeseries = bool(self.form_data.get("time_grain_sqla"))
self.point_radius_fixed = self.form_data.get("point_radius_fixed") or {
"type": "fix",
"value": 500,
}
return super().query_obj()
@deprecated(deprecated_in="3.0")
def get_metrics(self) -> list[str]:
# pylint: disable=attribute-defined-outside-init
self.metric = None
if self.point_radius_fixed.get("type") == "metric":
self.metric = self.point_radius_fixed["value"]
return [self.metric]
return []
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
return {
"metric": data.get(self.metric_label) if self.metric_label else None,
"radius": self.fixed_value
if self.fixed_value
else data.get(self.metric_label)
if self.metric_label
else None,
"cat_color": data.get(self.dim) if self.dim else None,
"position": data.get("spatial"),
DTTM_ALIAS: data.get(DTTM_ALIAS),
}
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
# pylint: disable=attribute-defined-outside-init
self.metric_label = utils.get_metric_name(self.metric) if self.metric else None
self.point_radius_fixed = self.form_data.get("point_radius_fixed")
self.fixed_value = None
self.dim = self.form_data.get("dimension")
if self.point_radius_fixed and self.point_radius_fixed.get("type") != "metric":
self.fixed_value = self.point_radius_fixed.get("value")
return super().get_data(df)
class DeckScreengrid(BaseDeckGLViz):
"""deck.gl's ScreenGridLayer"""
viz_type = "deck_screengrid"
verbose_name = _("Deck.gl - Screen Grid")
spatial_control_keys = ["spatial"]
is_timeseries = True
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
self.is_timeseries = bool(self.form_data.get("time_grain_sqla"))
return super().query_obj()
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
return {
"position": data.get("spatial"),
"weight": (data.get(self.metric_label) if self.metric_label else None) or 1,
"__timestamp": data.get(DTTM_ALIAS) or data.get("__time"),
}
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
self.metric_label = ( # pylint: disable=attribute-defined-outside-init
utils.get_metric_name(self.metric) if self.metric else None
)
return super().get_data(df)
class DeckGrid(BaseDeckGLViz):
"""deck.gl's DeckLayer"""
viz_type = "deck_grid"
verbose_name = _("Deck.gl - 3D Grid")
spatial_control_keys = ["spatial"]
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
return {
"position": data.get("spatial"),
"weight": (data.get(self.metric_label) if self.metric_label else None) or 1,
}
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
self.metric_label = ( # pylint: disable=attribute-defined-outside-init
utils.get_metric_name(self.metric) if self.metric else None
)
return super().get_data(df)
@deprecated(deprecated_in="3.0")
def geohash_to_json(geohash_code: str) -> list[list[float]]:
bbox = geohash.bbox(geohash_code)
return [
[bbox.get("w"), bbox.get("n")],
[bbox.get("e"), bbox.get("n")],
[bbox.get("e"), bbox.get("s")],
[bbox.get("w"), bbox.get("s")],
[bbox.get("w"), bbox.get("n")],
]
class DeckPathViz(BaseDeckGLViz):
"""deck.gl's PathLayer"""
viz_type = "deck_path"
verbose_name = _("Deck.gl - Paths")
deck_viz_key = "path"
is_timeseries = True
deser_map = {
"json": json.loads,
"polyline": polyline.decode,
"geohash": geohash_to_json,
}
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
# pylint: disable=attribute-defined-outside-init
self.is_timeseries = bool(self.form_data.get("time_grain_sqla"))
query_obj = super().query_obj()
self.metric = self.form_data.get("metric")
line_col = self.form_data.get("line_column")
if query_obj["metrics"]:
self.has_metrics = True
query_obj["groupby"].append(line_col)
else:
self.has_metrics = False
query_obj["columns"].append(line_col)
return query_obj
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
line_type = self.form_data["line_type"]
deser = self.deser_map[line_type]
line_column = self.form_data["line_column"]
path = deser(data[line_column])
if self.form_data.get("reverse_long_lat"):
path = [(o[1], o[0]) for o in path]
data[self.deck_viz_key] = path
if line_type != "geohash":
del data[line_column]
data["__timestamp"] = data.get(DTTM_ALIAS) or data.get("__time")
return data
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
self.metric_label = ( # pylint: disable=attribute-defined-outside-init
utils.get_metric_name(self.metric) if self.metric else None
)
return super().get_data(df)
class DeckPolygon(DeckPathViz):
"""deck.gl's Polygon Layer"""
viz_type = "deck_polygon"
deck_viz_key = "polygon"
verbose_name = _("Deck.gl - Polygon")
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
# pylint: disable=attribute-defined-outside-init
self.elevation = self.form_data.get("point_radius_fixed") or {
"type": "fix",
"value": 500,
}
return super().query_obj()
@deprecated(deprecated_in="3.0")
def get_metrics(self) -> list[str]:
metrics = [self.form_data.get("metric")]
if self.elevation.get("type") == "metric":
metrics.append(self.elevation.get("value"))
return [metric for metric in metrics if metric]
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
super().get_properties(data)
elevation = self.form_data["point_radius_fixed"]["value"]
type_ = self.form_data["point_radius_fixed"]["type"]
data["elevation"] = (
data.get(utils.get_metric_name(elevation))
if type_ == "metric"
else elevation
)
return data
class DeckHex(BaseDeckGLViz):
"""deck.gl's DeckLayer"""
viz_type = "deck_hex"
verbose_name = _("Deck.gl - 3D HEX")
spatial_control_keys = ["spatial"]
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
return {
"position": data.get("spatial"),
"weight": (data.get(self.metric_label) if self.metric_label else None) or 1,
}
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
self.metric_label = ( # pylint: disable=attribute-defined-outside-init
utils.get_metric_name(self.metric) if self.metric else None
)
return super().get_data(df)
class DeckHeatmap(BaseDeckGLViz):
"""deck.gl's HeatmapLayer"""
viz_type = "deck_heatmap"
verbose_name = _("Deck.gl - Heatmap")
spatial_control_keys = ["spatial"]
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
return {
"position": data.get("spatial"),
"weight": (data.get(self.metric_label) if self.metric_label else None) or 1,
}
def get_data(self, df: pd.DataFrame) -> VizData:
self.metric_label = ( # pylint: disable=attribute-defined-outside-init
utils.get_metric_name(self.metric) if self.metric else None
)
return super().get_data(df)
class DeckContour(BaseDeckGLViz):
"""deck.gl's ContourLayer"""
viz_type = "deck_contour"
verbose_name = _("Deck.gl - Contour")
spatial_control_keys = ["spatial"]
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
return {
"position": data.get("spatial"),
"weight": (data.get(self.metric_label) if self.metric_label else None) or 1,
}
def get_data(self, df: pd.DataFrame) -> VizData:
self.metric_label = ( # pylint: disable=attribute-defined-outside-init
utils.get_metric_name(self.metric) if self.metric else None
)
return super().get_data(df)
class DeckGeoJson(BaseDeckGLViz):
"""deck.gl's GeoJSONLayer"""
viz_type = "deck_geojson"
verbose_name = _("Deck.gl - GeoJSON")
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
query_obj["columns"] += [self.form_data.get("geojson")]
query_obj["metrics"] = []
query_obj["groupby"] = []
return query_obj
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
geojson = data[get_column_name(self.form_data["geojson"])]
return json.loads(geojson)
class DeckArc(BaseDeckGLViz):
"""deck.gl's Arc Layer"""
viz_type = "deck_arc"
verbose_name = _("Deck.gl - Arc")
spatial_control_keys = ["start_spatial", "end_spatial"]
is_timeseries = True
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
self.is_timeseries = bool(self.form_data.get("time_grain_sqla"))
return super().query_obj()
@deprecated(deprecated_in="3.0")
def get_properties(self, data: dict[str, Any]) -> dict[str, Any]:
dim = self.form_data.get("dimension")
return {
"sourcePosition": data.get("start_spatial"),
"targetPosition": data.get("end_spatial"),
"cat_color": data.get(dim) if dim else None,
DTTM_ALIAS: data.get(DTTM_ALIAS),
}
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
return {
"features": super().get_data(df)["features"],
"mapboxApiKey": config["MAPBOX_API_KEY"],
}
class EventFlowViz(BaseViz):
"""A visualization to explore patterns in event sequences"""
viz_type = "event_flow"
verbose_name = _("Event flow")
credits = 'from <a href="https://github.com/williaster/data-ui">@data-ui</a>'
is_timeseries = True
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query = super().query_obj()
form_data = self.form_data
event_key = form_data["all_columns_x"]
entity_key = form_data["entity"]
meta_keys = [
col
for col in form_data["all_columns"] or []
if col not in (event_key, entity_key)
]
query["columns"] = [event_key, entity_key] + meta_keys
if form_data["order_by_entity"]:
query["orderby"] = [(entity_key, True)]
return query
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
return df.to_dict(orient="records")
class PairedTTestViz(BaseViz):
"""A table displaying paired t-test values"""
viz_type = "paired_ttest"
verbose_name = _("Time Series - Paired t-test")
sort_series = False
is_timeseries = True
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
if sort_by := self.form_data.get("timeseries_limit_metric"):
sort_by_label = utils.get_metric_name(sort_by)
if sort_by_label not in utils.get_metric_names(query_obj["metrics"]):
query_obj["metrics"].append(sort_by)
if self.form_data.get("order_desc"):
query_obj["orderby"] = [
(sort_by, not self.form_data.get("order_desc", True))
]
return query_obj
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
"""
Transform received data frame into an object of the form:
{
'metric1': [
{
groups: ('groupA', ... ),
values: [ {x, y}, ... ],
}, ...
], ...
}
"""
if df.empty:
return None
groups = get_column_names(self.form_data.get("groupby"))
metrics = self.metric_labels
df = df.pivot_table(index=DTTM_ALIAS, columns=groups, values=metrics)
cols = []
# Be rid of falsy keys
for col in df.columns:
if col == "":
cols.append("N/A")
elif col is None:
cols.append("NULL")
else:
cols.append(col)
df.columns = cols
data: dict[str, list[dict[str, Any]]] = {}
series = df.to_dict("series")
for name_set in df.columns:
# If no groups are defined, nameSet will be the metric name
has_group = not isinstance(name_set, str)
data_ = {
"group": name_set[1:] if has_group else "All",
"values": [
{
"x": t,
"y": series[name_set][t] if t in series[name_set] else None,
}
for t in df.index
],
}
key = name_set[0] if has_group else name_set
if key in data:
data[key].append(data_)
else:
data[key] = [data_]
return data
class RoseViz(NVD3TimeSeriesViz):
viz_type = "rose"
verbose_name = _("Time Series - Nightingale Rose Chart")
sort_series = False
is_timeseries = True
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
data = super().get_data(df)
result: dict[str, list[dict[str, str]]] = {}
for datum in data:
key = datum["key"]
for val in datum["values"]:
timestamp = val["x"].value
if not result.get(timestamp):
result[timestamp] = []
value = 0 if math.isnan(val["y"]) else val["y"]
result[timestamp].append(
{
"key": key,
"value": value,
"name": ", ".join(key) if isinstance(key, list) else key,
"time": val["x"],
}
)
return result
class PartitionViz(NVD3TimeSeriesViz):
"""
A hierarchical data visualization with support for time series.
"""
viz_type = "partition"
verbose_name = _("Partition Diagram")
@deprecated(deprecated_in="3.0")
def query_obj(self) -> QueryObjectDict:
query_obj = super().query_obj()
time_op = self.form_data.get("time_series_option", "not_time")
# Return time series data if the user specifies so
query_obj["is_timeseries"] = time_op != "not_time"
return query_obj
@staticmethod
@deprecated(deprecated_in="3.0")
def levels_for(
time_op: str, groups: list[str], df: pd.DataFrame
) -> dict[int, pd.Series]:
"""
Compute the partition at each `level` from the dataframe.
"""
levels = {}
for i in range(0, len(groups) + 1):
agg_df = df.groupby(groups[:i]) if i else df
levels[i] = (
agg_df.mean(numeric_only=True)
if time_op == "agg_mean"
else agg_df.sum(numeric_only=True)
)
return levels
@staticmethod
@deprecated(deprecated_in="3.0")
def levels_for_diff(
time_op: str, groups: list[str], df: pd.DataFrame
) -> dict[int, pd.DataFrame]:
# Obtain a unique list of the time grains
times = list(set(df[DTTM_ALIAS]))
times.sort()
until = times[len(times) - 1]
since = times[0]
# Function describing how to calculate the difference
func = {
"point_diff": [pd.Series.sub, lambda a, b, fill_value: a - b],
"point_factor": [pd.Series.div, lambda a, b, fill_value: a / float(b)],
"point_percent": [
lambda a, b, fill_value=0: a.div(b, fill_value=fill_value) - 1,
lambda a, b, fill_value: a / float(b) - 1,
],
}[time_op]
agg_df = df.groupby(DTTM_ALIAS).sum(numeric_only=True)
levels = {
0: pd.Series(
{
m: func[1](agg_df[m][until], agg_df[m][since], 0)
for m in agg_df.columns
}
)
}
for i in range(1, len(groups) + 1):
agg_df = df.groupby([DTTM_ALIAS] + groups[:i]).sum(numeric_only=True)
levels[i] = pd.DataFrame(
{
m: func[0](agg_df[m][until], agg_df[m][since], fill_value=0)
for m in agg_df.columns
}
)
return levels
@deprecated(deprecated_in="3.0")
def levels_for_time(
self, groups: list[str], df: pd.DataFrame
) -> dict[int, VizData]:
procs = {}
for i in range(0, len(groups) + 1):
self.form_data["groupby"] = groups[:i]
df_drop = df.drop(groups[i:], axis=1)
procs[i] = self.process_data(df_drop, aggregate=True)
self.form_data["groupby"] = groups
return procs
@deprecated(deprecated_in="3.0")
def nest_values(
self,
levels: dict[int, pd.DataFrame],
level: int = 0,
metric: str | None = None,
dims: list[str] | None = None,
) -> list[dict[str, Any]]:
"""
Nest values at each level on the back-end with
access and setting, instead of summing from the bottom.
"""
if dims is None:
dims = []
if not level:
return [
{
"name": m,
"val": levels[0][m],
"children": self.nest_values(levels, 1, m),
}
for m in levels[0].index
]
if level == 1:
metric_level = levels[1][metric]
return [
{
"name": i,
"val": metric_level[i],
"children": self.nest_values(levels, 2, metric, [i]),
}
for i in metric_level.index
]
if level >= len(levels):
return []
dim_level = levels[level][metric][[dims[0]]]
return [
{
"name": i,
"val": dim_level[i],
"children": self.nest_values(levels, level + 1, metric, dims + [i]),
}
for i in dim_level.index
]
@deprecated(deprecated_in="3.0")
def nest_procs(
self,
procs: dict[int, pd.DataFrame],
level: int = -1,
dims: tuple[str, ...] | None = None,
time: Any = None,
) -> list[dict[str, Any]]:
if dims is None:
dims = ()
if level == -1:
return [
{"name": m, "children": self.nest_procs(procs, 0, (m,))}
for m in procs[0].columns
]
if not level:
return [
{
"name": t,
"val": procs[0][dims[0]][t],
"children": self.nest_procs(procs, 1, dims, t),
}
for t in procs[0].index
]
if level >= len(procs):
return []
return [
{
"name": i,
"val": procs[level][dims][i][time],
"children": self.nest_procs(procs, level + 1, dims + (i,), time),
}
for i in procs[level][dims].columns
]
@deprecated(deprecated_in="3.0")
def get_data(self, df: pd.DataFrame) -> VizData:
if df.empty:
return None
groups = get_column_names(self.form_data.get("groupby"))
time_op = self.form_data.get("time_series_option", "not_time")
if not groups:
raise ValueError(_("Please choose at least one groupby"))
if time_op == "not_time":
levels = self.levels_for("agg_sum", groups, df)
elif time_op in ["agg_sum", "agg_mean"]:
levels = self.levels_for(time_op, groups, df)
elif time_op in ["point_diff", "point_factor", "point_percent"]:
levels = self.levels_for_diff(time_op, groups, df)
elif time_op == "adv_anal":
procs = self.levels_for_time(groups, df)
return self.nest_procs(procs)
else:
levels = self.levels_for("agg_sum", [DTTM_ALIAS] + groups, df)
return self.nest_values(levels)
@deprecated(deprecated_in="3.0")
def get_subclasses(cls: type[BaseViz]) -> set[type[BaseViz]]:
return set(cls.__subclasses__()).union(
[sc for c in cls.__subclasses__() for sc in get_subclasses(c)]
)
viz_types = {
o.viz_type: o
for o in get_subclasses(BaseViz)
if o.viz_type not in config["VIZ_TYPE_DENYLIST"]
}