scistackplotdb 0.2.5__tar.gz → 0.3.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.5
2
2
  Name: scistackplotdb
3
- Version: 0.2.5
3
+ Version: 0.3.0
4
4
  Summary: scidb-backed plotting: load variables into long tables and plot them with scistackplot
5
5
  Author: SciStack Contributors
6
6
  License-Expression: MIT
@@ -17,11 +17,11 @@ Classifier: Programming Language :: Python :: 3.13
17
17
  Classifier: Topic :: Scientific/Engineering :: Visualization
18
18
  Classifier: Typing :: Typed
19
19
  Requires-Python: >=3.10
20
- Requires-Dist: sciduckdb==0.2.5
21
- Requires-Dist: scifor==0.2.5
22
- Requires-Dist: scistack-db==0.2.5
23
- Requires-Dist: scistacklog==0.2.5
24
- Requires-Dist: scistackplot==0.2.5
20
+ Requires-Dist: sciduckdb==0.3.0
21
+ Requires-Dist: scifor==0.3.0
22
+ Requires-Dist: scistack-db==0.3.0
23
+ Requires-Dist: scistacklog==0.3.0
24
+ Requires-Dist: scistackplot==0.3.0
25
25
  Provides-Extra: dev
26
26
  Requires-Dist: matplotlib>=3.6; extra == 'dev'
27
27
  Requires-Dist: pytest-cov>=4.0; extra == 'dev'
@@ -24,7 +24,7 @@ import numpy as np
24
24
  import pandas as pd
25
25
  from sciduckdb import merge_tz_columns, stored_tz_companions, tz_column_name
26
26
  from scistacklog import Log
27
- from scistackplot import CODE_FACTOR_PREFIX, RUN_FACTOR_PREFIX, element_values
27
+ from scistackplot import CODE_FACTOR_PREFIX, RUN_FACTOR_PREFIX
28
28
  from scistackplot.text.framesize import format_extent, frame_extent
29
29
  from scistackplot.model.shape import Shape, classify_value
30
30
  from scistackplot.model.table import natural_sort_key
@@ -98,11 +98,6 @@ class VariableFrame:
98
98
  #: Name of the per-row "this is my location's newest code version" flag, or
99
99
  #: None when the variable holds only one version. See :data:`LATEST_COLUMN`.
100
100
  latest_column: str | None = None
101
- #: Data columns stored as ``row_list`` ("combine rows": one element per
102
- #: source row, sciduckdb.row_lists). Their cells arrive as arrays; the
103
- #: source splits them into rows with a ``Row`` factor
104
- #: (scistackplot.reduce.rows), never plots them as 1-D series.
105
- row_list_columns: list[str] = field(default_factory=list)
106
101
 
107
102
  @property
108
103
  def value_column(self) -> str:
@@ -273,13 +268,7 @@ def sample_value(db, variable: str) -> Any:
273
268
  columns = data_columns_for(db, variable)
274
269
  if not columns:
275
270
  return None
276
- value = db.record_store.sample_cell(variable, columns[0])
277
- # A row list's cell is its source rows, not a series: classify by one
278
- # ELEMENT, the shape the source turns it into (rows + a Row factor).
279
- if value is not None and _row_list_columns(db, variable, columns[:1]):
280
- elements = element_values(pd.Series([value], dtype=object)).dropna()
281
- value = elements.iloc[0] if len(elements) else None
282
- return value
271
+ return db.record_store.sample_cell(variable, columns[0])
283
272
 
284
273
 
285
274
  def sample_column_value(db, variable: str, column: str) -> Any:
@@ -308,18 +297,13 @@ def field_shapes(db, variable: str) -> dict[str, Shape]:
308
297
  a "(text)" entry that then says it has no text fields.
309
298
  """
310
299
  types = data_column_types_for(db, variable)
311
- row_lists = set(_row_list_columns(db, variable, list(types)))
312
300
  shapes: dict[str, Shape] = {}
313
301
  with Log.timer(f"field_shapes({variable})", layer=LAYER, extra=f"{len(types)} column(s)"):
314
302
  for column, data_type in types.items():
315
- if is_container_type(data_type) and column not in row_lists:
303
+ if is_container_type(data_type):
316
304
  shapes[column] = Shape.SERIES_1D
317
305
  continue
318
- value = sample_column_value(db, variable, column)
319
- if value is not None and column in row_lists:
320
- elements = element_values(pd.Series([value], dtype=object)).dropna()
321
- value = elements.iloc[0] if len(elements) else None
322
- shapes[column] = classify_value(value)
306
+ shapes[column] = classify_value(sample_column_value(db, variable, column))
323
307
  return shapes
324
308
 
325
309
 
@@ -441,28 +425,6 @@ def _key_text(value: Any) -> "str | None":
441
425
  return str(value)
442
426
 
443
427
 
444
- def _row_list_columns(db, variable: str, columns) -> list[str]:
445
- """Which columns are row lists (sciduckdb.row_lists: rows combined by
446
- ``on_duplicate="combine"``). This reader selects the stored cells as they
447
- are, so EVERY cell of such a column is an array (``[7]``, ``[5, 12]``).
448
- The source splits them into rows with a ``Row`` factor
449
- (scistackplot.reduce.rows) — they are observations, not a 1-D series. This
450
- replaced the "read as 1-D" decision of
451
- .claude/plan-combine-duplicate-locations.md §6 on 2026-10-30
452
- (.claude/plan-row-list-plots.md)."""
453
- from sciduckdb import is_row_list_meta
454
-
455
- stored = (db.get_dtype_meta(variable) or {}).get("columns", {})
456
- row_lists = [c for c in columns if is_row_list_meta(stored.get(c))]
457
- if row_lists:
458
- Log.debug(
459
- "variable %r: row-list column(s) %s (combined rows) — split into "
460
- "one row per element by the source",
461
- variable, row_lists, layer=LAYER,
462
- )
463
- return row_lists
464
-
465
-
466
428
  def load_variable(
467
429
  db, variable: str, *, with_variants: bool = True, include_data: bool = True
468
430
  ) -> VariableFrame:
@@ -487,7 +449,6 @@ def load_variable(
487
449
  keys = schema_keys(db)
488
450
  column_types = data_column_types_for(db, variable)
489
451
  columns = list(column_types)
490
- row_list_columns = _row_list_columns(db, variable, columns)
491
452
  if not columns:
492
453
  Log.warn("variable %r has no data columns", variable, layer=LAYER)
493
454
  return VariableFrame(name=variable, frame=pd.DataFrame())
@@ -581,7 +542,6 @@ def load_variable(
581
542
  variant_columns=variant_columns,
582
543
  variant_axes=variant_axes,
583
544
  latest_column=latest_column,
584
- row_list_columns=[c for c in row_list_columns if c in columns],
585
545
  )
586
546
 
587
547
 
@@ -21,11 +21,8 @@ from scistackplot import (
21
21
  Shape,
22
22
  classify_column,
23
23
  classify_value,
24
- element_values,
25
- explode_row_lists,
26
24
  is_plottable,
27
25
  natural_sort_key,
28
- row_origin,
29
26
  )
30
27
  from scistackplot.reduce.categorical import MAX_CATEGORY_LEVELS
31
28
  from scistackplot.sources.dedup import SingleFlight
@@ -656,8 +653,6 @@ class ScidbSource(BaseSource):
656
653
  primary = self._variable_frame(base_variable(measures[0]))
657
654
  own = self._own_column_groups(primary, factor_variables)
658
655
  field_columns: list[str] = []
659
- row_column: str | None = None
660
- row_info: dict | None = None
661
656
 
662
657
  if x_measure is None:
663
658
  if len(primary.data_columns) > 1:
@@ -668,20 +663,9 @@ class ScidbSource(BaseSource):
668
663
  text=text_only,
669
664
  )
670
665
  field_columns = [field_factor]
671
- melted = set(frame[field_factor].unique())
672
- row_fields = [c for c in primary.row_list_columns if c in melted]
673
666
  else:
674
667
  frame = self._named_frame(primary, measures[0])
675
668
  field_factor = None
676
- row_fields = (
677
- [measures[0]]
678
- if primary.value_column in primary.row_list_columns
679
- else []
680
- )
681
- if row_fields:
682
- frame, row_column, row_info = self._split_row_lists(
683
- primary.name, frame, measures[0], row_fields, field_factor
684
- )
685
669
  levels = primary.levels
686
670
  variant_columns = list(primary.variant_columns)
687
671
  variant_axes = list(primary.variant_axes)
@@ -692,22 +676,6 @@ class ScidbSource(BaseSource):
692
676
  else:
693
677
  secondary = self._variable_frame(x_measure)
694
678
  pair = (measures[0], x_measure)
695
- combined = [
696
- name
697
- for name, vf in zip(pair, (primary, secondary), strict=True)
698
- if vf.row_list_columns
699
- ]
700
- if combined:
701
- # Not split here: pairing one x per y row needs the two
702
- # variables' rows to line up, and nothing says row i of one
703
- # is row i of the other. Said, not silent.
704
- Log.warn(
705
- "%s hold combined rows (row lists); an x-y plot reads "
706
- "them as 1-D arrays — plot each on its own to see one "
707
- "point per source row",
708
- combined,
709
- layer=LAYER,
710
- )
711
679
  multi = [
712
680
  name
713
681
  for name, frame in zip(pair, (primary, secondary), strict=True)
@@ -766,13 +734,6 @@ class ScidbSource(BaseSource):
766
734
  factors.extend(c for c in field_columns if c in frame.columns)
767
735
  factor_origins = {axis["column"]: axis for axis in variant_axes}
768
736
  factor_depths = dict(group_depths)
769
- if row_column is not None and row_info is not None:
770
- factors.append(row_column)
771
- factor_origins[row_column] = row_info
772
- # Pinned by every schema key the record sits under plus itself:
773
- # one level below the variable's own, so the scalar default
774
- # groups it (one mark per source row) and iterates the rest.
775
- factor_depths[row_column] = len(levels) + 1
776
737
 
777
738
  level_order = {
778
739
  name: self._level_order(name, frame, is_field=name in field_columns)
@@ -1227,17 +1188,6 @@ class ScidbSource(BaseSource):
1227
1188
  ``spec.measures`` stop naming a real variable.
1228
1189
  """
1229
1190
  frames = [self._variable_frame(name) for name in measures]
1230
- combined = [f.name for f in frames if f.row_list_columns]
1231
- if combined:
1232
- # Splitting is per variable, and stacked variables share one
1233
- # value column; not done here yet. Said, not silent.
1234
- Log.warn(
1235
- "%s hold combined rows (row lists); stacked with other "
1236
- "variables they are read as 1-D arrays — plot each on its own "
1237
- "to see one point per source row",
1238
- combined,
1239
- layer=LAYER,
1240
- )
1241
1191
 
1242
1192
  # Dict/struct variables stack too, provided they carry the SAME fields:
1243
1193
  # RawEMG and FilteredEMG, both keyed by muscle, are the archetypal
@@ -1538,17 +1488,10 @@ class ScidbSource(BaseSource):
1538
1488
  carried = [c for c in (carry or []) if c in frame.columns]
1539
1489
  usable, skipped = [], []
1540
1490
  texts: list[str] = []
1541
- row_lists = set(variable_frame.row_list_columns)
1542
1491
  for column in variable_frame.data_columns:
1543
1492
  if column in carried:
1544
1493
  continue
1545
- # A row list is classified by its ELEMENTS (it is split into rows
1546
- # after the melt): a VARCHAR row list is text and joins the text
1547
- # fields, instead of melting in as a "1-D" column.
1548
- values = (
1549
- element_values(frame[column]) if column in row_lists else frame[column]
1550
- )
1551
- shape = classify_column(values)
1494
+ shape = classify_column(frame[column])
1552
1495
  if shape is Shape.CATEGORICAL:
1553
1496
  texts.append(column)
1554
1497
  elif is_plottable(shape):
@@ -1647,51 +1590,6 @@ class ScidbSource(BaseSource):
1647
1590
  )
1648
1591
  return melted, field_factor
1649
1592
 
1650
- def _split_row_lists(
1651
- self,
1652
- variable: str,
1653
- frame,
1654
- measure: str,
1655
- row_fields: list[str],
1656
- field_factor: str | None,
1657
- ):
1658
- """``(frame, row column or None, its origin or None)`` — every
1659
- row-list cell split into one row per element (scistackplot.reduce.rows,
1660
- the one owner of the reshape; codegen mirrors it from the origin).
1661
-
1662
- The Row factor is kept only when some cell held two or more rows: a
1663
- one-level factor is a constant (synthetic-factors.md invariant 6).
1664
- The cells are unwrapped either way, so ``[v]`` reads as ``v``."""
1665
- with Log.timer(
1666
- f"split_row_lists({variable})",
1667
- layer=LAYER,
1668
- extra=f"{len(row_fields)} field(s)",
1669
- ):
1670
- split = explode_row_lists(
1671
- frame, measure, row_fields, field_factor=field_factor
1672
- )
1673
- Log.info(
1674
- "split row-list field(s) %s of %r: %d cell(s), %d -> %d row(s), "
1675
- "at most %d per cell; %s",
1676
- row_fields,
1677
- variable,
1678
- split.cells,
1679
- split.rows_in,
1680
- split.rows_out,
1681
- split.max_rows,
1682
- f"factor {split.row_factor!r}"
1683
- if split.row_factor
1684
- else "no cell held two rows, so no Row factor",
1685
- layer=LAYER,
1686
- )
1687
- if split.row_factor is None:
1688
- return split.frame, None, None
1689
- return (
1690
- split.frame,
1691
- split.row_factor,
1692
- row_origin(row_fields, field_factor),
1693
- )
1694
-
1695
1593
  def _named_frame(self, variable_frame, measure: str):
1696
1594
  """The variable's frame with its data column renamed to the measure."""
1697
1595
  return variable_frame.frame.rename(
File without changes