requipso 0.2.2__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.
Files changed (80) hide show
  1. requipso-0.2.2/PKG-INFO +102 -0
  2. requipso-0.2.2/README.md +54 -0
  3. requipso-0.2.2/pyproject.toml +111 -0
  4. requipso-0.2.2/requipso/__init__.py +27 -0
  5. requipso-0.2.2/requipso/cli.py +11 -0
  6. requipso-0.2.2/requipso/completer.py +103 -0
  7. requipso-0.2.2/requipso/data.py +72 -0
  8. requipso-0.2.2/requipso/descriptive_pipeline.py +271 -0
  9. requipso-0.2.2/requipso/docs/macros.py +226 -0
  10. requipso-0.2.2/requipso/docs/mkdocs.yml +140 -0
  11. requipso-0.2.2/requipso/docs/pages/assets/consort_flowchart.png +0 -0
  12. requipso-0.2.2/requipso/docs/pages/assets/demo.gif +0 -0
  13. requipso-0.2.2/requipso/docs/pages/blocs/age.md +61 -0
  14. requipso-0.2.2/requipso/docs/pages/blocs/ald.md +65 -0
  15. requipso-0.2.2/requipso/docs/pages/blocs/biometry.md +98 -0
  16. requipso-0.2.2/requipso/docs/pages/blocs/cim10.md +80 -0
  17. requipso-0.2.2/requipso/docs/pages/blocs/cisp2.md +59 -0
  18. requipso-0.2.2/requipso/docs/pages/blocs/code-postal.md +47 -0
  19. requipso-0.2.2/requipso/docs/pages/blocs/consultations.md +65 -0
  20. requipso-0.2.2/requipso/docs/pages/blocs/departement.md +50 -0
  21. requipso-0.2.2/requipso/docs/pages/blocs/est-vivant.md +44 -0
  22. requipso-0.2.2/requipso/docs/pages/blocs/genre.md +46 -0
  23. requipso-0.2.2/requipso/docs/pages/blocs/index.md +40 -0
  24. requipso-0.2.2/requipso/docs/pages/blocs/insurance.md +55 -0
  25. requipso-0.2.2/requipso/docs/pages/blocs/medecin-traitant.md +56 -0
  26. requipso-0.2.2/requipso/docs/pages/blocs/paquet-annees.md +70 -0
  27. requipso-0.2.2/requipso/docs/pages/blocs/prescription-atc.md +69 -0
  28. requipso-0.2.2/requipso/docs/pages/blocs/regex-consultation.md +64 -0
  29. requipso-0.2.2/requipso/docs/pages/blocs/regex-nlp.md +107 -0
  30. requipso-0.2.2/requipso/docs/pages/blocs/regex-patient-observation.md +81 -0
  31. requipso-0.2.2/requipso/docs/pages/blocs/regex-prescription.md +64 -0
  32. requipso-0.2.2/requipso/docs/pages/guide/application.md +81 -0
  33. requipso-0.2.2/requipso/docs/pages/guide/combiner-blocs.md +61 -0
  34. requipso-0.2.2/requipso/docs/pages/guide/demarrage.md +74 -0
  35. requipso-0.2.2/requipso/docs/pages/guide/erreurs.md +83 -0
  36. requipso-0.2.2/requipso/docs/pages/guide/exemple-complet.md +67 -0
  37. requipso-0.2.2/requipso/docs/pages/guide/google-sheet.md +50 -0
  38. requipso-0.2.2/requipso/docs/pages/guide/onglet-inclusion.md +91 -0
  39. requipso-0.2.2/requipso/docs/pages/guide/onglet-projet.md +48 -0
  40. requipso-0.2.2/requipso/docs/pages/guide/onglet-tableau-descriptif.md +124 -0
  41. requipso-0.2.2/requipso/docs/pages/guide/parametres.md +162 -0
  42. requipso-0.2.2/requipso/docs/pages/index.md +73 -0
  43. requipso-0.2.2/requipso/docs/pages/stylesheets/extra.css +3 -0
  44. requipso-0.2.2/requipso/docs/pages/technique/ajouter-un-bloc.md +147 -0
  45. requipso-0.2.2/requipso/docs/pages/technique/architecture.md +80 -0
  46. requipso-0.2.2/requipso/docs/pages/technique/blocs-internes.md +176 -0
  47. requipso-0.2.2/requipso/docs/pages/technique/chargement-sheet.md +107 -0
  48. requipso-0.2.2/requipso/docs/pages/technique/developpement.md +124 -0
  49. requipso-0.2.2/requipso/docs/pages/technique/donnees.md +106 -0
  50. requipso-0.2.2/requipso/docs/pages/technique/export-signatures.md +131 -0
  51. requipso-0.2.2/requipso/docs/pages/technique/nlp.md +82 -0
  52. requipso-0.2.2/requipso/docs/pages/technique/parametres-internes.md +140 -0
  53. requipso-0.2.2/requipso/docs/pages/technique/pipeline.md +66 -0
  54. requipso-0.2.2/requipso/docs/pages/technique/tableau-descriptif.md +74 -0
  55. requipso-0.2.2/requipso/google.py +168 -0
  56. requipso-0.2.2/requipso/io.py +95 -0
  57. requipso-0.2.2/requipso/notebooks/marimo/run.py +352 -0
  58. requipso-0.2.2/requipso/pipeline.py +364 -0
  59. requipso-0.2.2/requipso/pipeline_loader.py +572 -0
  60. requipso-0.2.2/requipso/pipes/__init__.py +43 -0
  61. requipso-0.2.2/requipso/pipes/atcd.py +365 -0
  62. requipso-0.2.2/requipso/pipes/base.py +431 -0
  63. requipso-0.2.2/requipso/pipes/biometry.py +170 -0
  64. requipso-0.2.2/requipso/pipes/consultations.py +117 -0
  65. requipso-0.2.2/requipso/pipes/insurance.py +90 -0
  66. requipso-0.2.2/requipso/pipes/params.py +283 -0
  67. requipso-0.2.2/requipso/pipes/patient.py +377 -0
  68. requipso-0.2.2/requipso/pipes/prescription.py +134 -0
  69. requipso-0.2.2/requipso/pipes/processors.py +250 -0
  70. requipso-0.2.2/requipso/polars.py +24 -0
  71. requipso-0.2.2/requipso/resources/pathology/__init__.py +7 -0
  72. requipso-0.2.2/requipso/signature_exporter.py +473 -0
  73. requipso-0.2.2/requipso/stats.py +130 -0
  74. requipso-0.2.2/requipso.egg-info/PKG-INFO +102 -0
  75. requipso-0.2.2/requipso.egg-info/SOURCES.txt +78 -0
  76. requipso-0.2.2/requipso.egg-info/dependency_links.txt +1 -0
  77. requipso-0.2.2/requipso.egg-info/entry_points.txt +2 -0
  78. requipso-0.2.2/requipso.egg-info/requires.txt +41 -0
  79. requipso-0.2.2/requipso.egg-info/top_level.txt +1 -0
  80. requipso-0.2.2/setup.cfg +4 -0
@@ -0,0 +1,102 @@
1
+ Metadata-Version: 2.4
2
+ Name: requipso
3
+ Version: 0.2.2
4
+ Summary: Add your description here
5
+ Requires-Python: <3.13,>=3.12
6
+ Description-Content-Type: text/markdown
7
+ Requires-Dist: beautifulsoup4>=4.14.2
8
+ Requires-Dist: boto3>=1.41.2
9
+ Requires-Dist: boto3-stubs>=1.41.5
10
+ Requires-Dist: edsnlp[ml]>=0.23
11
+ Requires-Dist: google-api-core>=2.30.3
12
+ Requires-Dist: google-api-python-client>=2.194.0
13
+ Requires-Dist: google-auth>=2.49.2
14
+ Requires-Dist: google-auth-oauthlib>=1.2
15
+ Requires-Dist: graphviz>=0.21
16
+ Requires-Dist: gspread<6
17
+ Requires-Dist: gspread-formatting>=1.2.1
18
+ Requires-Dist: gspread-pandas>=3.2.3
19
+ Requires-Dist: ipykernel>=6.30.1
20
+ Requires-Dist: jupyterlab>=4.4.9
21
+ Requires-Dist: kaleido>=1.1.0
22
+ Requires-Dist: marimo[lsp,mcp,recommended]>=0.17.0
23
+ Requires-Dist: matplotlib>=3.10.6
24
+ Requires-Dist: metanno==1.0.0
25
+ Requires-Dist: mypy>=1.18.2
26
+ Requires-Dist: networkx>=3.5
27
+ Requires-Dist: openpyxl>=3.1.5
28
+ Requires-Dist: pandas>=2.3.2
29
+ Requires-Dist: pandas-stubs>=2.3.2.250926
30
+ Requires-Dist: pillow>=11.3.0
31
+ Requires-Dist: plotly>=6.3.0
32
+ Requires-Dist: polars>=1.33.1
33
+ Requires-Dist: polars-permutation>=0.1.0
34
+ Requires-Dist: pret>=0.4.1
35
+ Requires-Dist: pret-joy>=5.0.0b52
36
+ Requires-Dist: pyarrow>=21.0.0
37
+ Requires-Dist: pyzabbix>=1.3.1
38
+ Requires-Dist: rich>=14.1.0
39
+ Requires-Dist: ruff>=0.14.1
40
+ Requires-Dist: schemdraw>=0.21
41
+ Requires-Dist: sqlglot>=28.0.0
42
+ Requires-Dist: streamlit>=1.50.0
43
+ Requires-Dist: styleframe>=4.2
44
+ Requires-Dist: tabulate>=0.9.0
45
+ Requires-Dist: types-networkx>=3.6.0.20251127
46
+ Requires-Dist: xlsxwriter>=3.2.9
47
+ Requires-Dist: zabbix-utils>=2.0.3
48
+
49
+ # Requipso
50
+
51
+ ## Installation
52
+
53
+ Tout d'abord les dépendances:
54
+
55
+ ```
56
+ sudo apt update
57
+ sudo apt install build-essential graphviz
58
+ ```
59
+
60
+ Ensuite clonez le repo puis:
61
+
62
+ ```
63
+ cd requipso
64
+ uv sync
65
+ ```
66
+
67
+ ## Données
68
+
69
+ Demander à l"équipe RED d'ipso de vous fournir les données finales de l'ETL (ensemble de fichiers parquets)
70
+
71
+
72
+ ## Démo
73
+
74
+ Pour lancer la démo: `uv run requipso`.
75
+ Cela devrait ouvrir:
76
+
77
+ * un onglet sur navigateur
78
+ * un autre onglet contenant la documentation
79
+
80
+ [![demo.mp4 - Google Drive](https://res.cloudinary.com/marcomontalbano/image/upload/v1777483912/video_to_markdown/images/google-drive--1OTeSeuiUC5FhiQMEqwwUt_1l2KcusFtH-c05b58ac6eb4c4700831b2b3070cd403.jpg)](https://drive.google.com/file/d/1OTeSeuiUC5FhiQMEqwwUt_1l2KcusFtH/view "demo.mp4 - Google Drive")
81
+
82
+ ## Mode d'emploi
83
+
84
+ * Il faut créer une feuille de calcul dans [ce dossier Drive](https://drive.google.com/drive/folders/1vTA1zVLbyU6mVMkiTL1J1bLolOlsut7A). Vous pouvez dupliquer la feuille de calcul nommée **template**
85
+ * Ensuite:
86
+ * Dans l'onglet 'Projet'
87
+ * définir la date de référence
88
+ * les populations d'intérêt seront à définir une fois la pipeline d'inclusion définie
89
+ * Bien lire l'onglet "Explications" pour comprendre
90
+ * L'onglet 'Inclusion' permet de définir son flowchart
91
+ * Ensuite l'onglet 'Descriptive_table' permet de définir le tableau descriptif
92
+ * A ce moment là, vous pouvez renseigner les populations d'intérêt (qui formeront les colonnes du tableau desciptif) en faisant référence à certains blocs de votre diagramme d'inclusion
93
+ * Une fois que tout est rempli, cf plus haut (démo) pour lancer le requêteur.
94
+
95
+ ## Documentation
96
+
97
+ Pour build/serve la documentation seule:
98
+
99
+ ```bash
100
+ uv run mkdocs serve -f requipso/docs/mkdocs.yml
101
+ uv run mkdocs build -f requipso/docs/mkdocs.yml --strict
102
+ ```
@@ -0,0 +1,54 @@
1
+ # Requipso
2
+
3
+ ## Installation
4
+
5
+ Tout d'abord les dépendances:
6
+
7
+ ```
8
+ sudo apt update
9
+ sudo apt install build-essential graphviz
10
+ ```
11
+
12
+ Ensuite clonez le repo puis:
13
+
14
+ ```
15
+ cd requipso
16
+ uv sync
17
+ ```
18
+
19
+ ## Données
20
+
21
+ Demander à l"équipe RED d'ipso de vous fournir les données finales de l'ETL (ensemble de fichiers parquets)
22
+
23
+
24
+ ## Démo
25
+
26
+ Pour lancer la démo: `uv run requipso`.
27
+ Cela devrait ouvrir:
28
+
29
+ * un onglet sur navigateur
30
+ * un autre onglet contenant la documentation
31
+
32
+ [![demo.mp4 - Google Drive](https://res.cloudinary.com/marcomontalbano/image/upload/v1777483912/video_to_markdown/images/google-drive--1OTeSeuiUC5FhiQMEqwwUt_1l2KcusFtH-c05b58ac6eb4c4700831b2b3070cd403.jpg)](https://drive.google.com/file/d/1OTeSeuiUC5FhiQMEqwwUt_1l2KcusFtH/view "demo.mp4 - Google Drive")
33
+
34
+ ## Mode d'emploi
35
+
36
+ * Il faut créer une feuille de calcul dans [ce dossier Drive](https://drive.google.com/drive/folders/1vTA1zVLbyU6mVMkiTL1J1bLolOlsut7A). Vous pouvez dupliquer la feuille de calcul nommée **template**
37
+ * Ensuite:
38
+ * Dans l'onglet 'Projet'
39
+ * définir la date de référence
40
+ * les populations d'intérêt seront à définir une fois la pipeline d'inclusion définie
41
+ * Bien lire l'onglet "Explications" pour comprendre
42
+ * L'onglet 'Inclusion' permet de définir son flowchart
43
+ * Ensuite l'onglet 'Descriptive_table' permet de définir le tableau descriptif
44
+ * A ce moment là, vous pouvez renseigner les populations d'intérêt (qui formeront les colonnes du tableau desciptif) en faisant référence à certains blocs de votre diagramme d'inclusion
45
+ * Une fois que tout est rempli, cf plus haut (démo) pour lancer le requêteur.
46
+
47
+ ## Documentation
48
+
49
+ Pour build/serve la documentation seule:
50
+
51
+ ```bash
52
+ uv run mkdocs serve -f requipso/docs/mkdocs.yml
53
+ uv run mkdocs build -f requipso/docs/mkdocs.yml --strict
54
+ ```
@@ -0,0 +1,111 @@
1
+ [project]
2
+ name = "requipso"
3
+ version = "0.2.2"
4
+ description = "Add your description here"
5
+ readme = "README.md"
6
+ requires-python = ">=3.12,<3.13"
7
+ dependencies = [
8
+ "beautifulsoup4>=4.14.2",
9
+ "boto3>=1.41.2",
10
+ "boto3-stubs>=1.41.5",
11
+ "edsnlp[ml]>=0.23",
12
+ "google-api-core>=2.30.3",
13
+ "google-api-python-client>=2.194.0",
14
+ "google-auth>=2.49.2",
15
+ "google-auth-oauthlib>=1.2",
16
+ "graphviz>=0.21",
17
+ "gspread<6",
18
+ "gspread-formatting>=1.2.1",
19
+ "gspread-pandas>=3.2.3",
20
+ "ipykernel>=6.30.1",
21
+ "jupyterlab>=4.4.9",
22
+ "kaleido>=1.1.0",
23
+ "marimo[lsp,mcp,recommended]>=0.17.0",
24
+ "matplotlib>=3.10.6",
25
+ "metanno==1.0.0",
26
+ "mypy>=1.18.2",
27
+ "networkx>=3.5",
28
+ "openpyxl>=3.1.5",
29
+ "pandas>=2.3.2",
30
+ "pandas-stubs>=2.3.2.250926",
31
+ "pillow>=11.3.0",
32
+ "plotly>=6.3.0",
33
+ "polars>=1.33.1",
34
+ "polars-permutation>=0.1.0",
35
+ "pret>=0.4.1",
36
+ "pret-joy>=5.0.0b52",
37
+ "pyarrow>=21.0.0",
38
+ "pyzabbix>=1.3.1",
39
+ "rich>=14.1.0",
40
+ "ruff>=0.14.1",
41
+ "schemdraw>=0.21",
42
+ "sqlglot>=28.0.0",
43
+ "streamlit>=1.50.0",
44
+ "styleframe>=4.2",
45
+ "tabulate>=0.9.0",
46
+ "types-networkx>=3.6.0.20251127",
47
+ "xlsxwriter>=3.2.9",
48
+ "zabbix-utils>=2.0.3",
49
+ ]
50
+
51
+ [project.scripts]
52
+ requipso = "requipso.cli:main"
53
+
54
+ [build-system]
55
+ requires = ["setuptools>=61.0"]
56
+ build-backend = "setuptools.build_meta"
57
+
58
+ [tool.setuptools]
59
+ [tool.setuptools.packages.find]
60
+ include = ["requipso*"]
61
+
62
+ [tool.setuptools.package-data]
63
+ requipso = ["docs/mkdocs.yml", "docs/pages/**/*"]
64
+
65
+ [tool.uv]
66
+ default-groups = ["docs"]
67
+
68
+ [tool.marimo.runtime]
69
+ on_cell_change = "autorun"
70
+
71
+ [tool.mypy]
72
+ # --- Core ---
73
+ ignore_missing_imports = false
74
+ follow_imports = "normal"
75
+ namespace_packages = true
76
+ explicit_package_bases = true
77
+
78
+ # --- Warning settings ---
79
+ warn_unused_configs = true
80
+ warn_unused_ignores = true
81
+ warn_return_any = true
82
+ warn_redundant_casts = true
83
+ warn_unreachable = true
84
+
85
+ # --- Optional strictness ---
86
+ strict_equality = true
87
+ disallow_subclassing_any = true
88
+
89
+ # --- Level of strictness (balanced, not extreme) ---
90
+ check_untyped_defs = true
91
+ disallow_incomplete_defs = true
92
+ disallow_untyped_defs = false # Set to true if you want strict mode
93
+ disallow_untyped_calls = false # Set to true if stricter type checking is desired
94
+ disallow_untyped_decorators = false
95
+
96
+ # --- Error formatting ---
97
+ show_error_codes = true
98
+ pretty = true
99
+
100
+ [dependency-groups]
101
+ docs = [
102
+ "mike>=2.1",
103
+ "mkdocs-macros-plugin~=1.5.0",
104
+ "mkdocs-material~=9.7.7",
105
+ ]
106
+
107
+ # --- Plugins (enable if needed) ---
108
+ # plugins = [
109
+ # "pydantic.mypy"
110
+ # ]
111
+
@@ -0,0 +1,27 @@
1
+ from pathlib import Path
2
+
3
+ # Define paths FIRST before any other imports that might need them
4
+ BASE_DIR = Path(__file__).parent.resolve()
5
+ RESOURCES_DIR = BASE_DIR / "resources"
6
+
7
+ from .data import Data
8
+
9
+ # Registry and schema are imported lazily to avoid circular imports
10
+ # Use: from requipso.registry import registry
11
+ # Use: from requipso.schema import schema_provider
12
+
13
+ # Auto-install Polars column completer in IPython contexts
14
+ try:
15
+ from IPython import get_ipython
16
+
17
+ if get_ipython() is not None:
18
+ from .completer import install as _install_completer
19
+
20
+ _install_completer()
21
+ except ImportError:
22
+ pass
23
+
24
+ __all__ = [
25
+ "BASE_DIR",
26
+ "RESOURCES_DIR",
27
+ ]
@@ -0,0 +1,11 @@
1
+ import subprocess
2
+ import sys
3
+ from pathlib import Path
4
+
5
+
6
+ def main() -> None:
7
+ notebook = Path(__file__).parent / "notebooks" / "marimo" / "run.py"
8
+ code = subprocess.call(
9
+ ["marimo", "run", str(notebook), "--show-tracebacks", *sys.argv[1:]]
10
+ )
11
+ sys.exit(code)
@@ -0,0 +1,103 @@
1
+ """Custom IPython completer for Polars column names.
2
+
3
+ Provides tab-completion for column names inside ``pl.col("...")`` expressions.
4
+ Columns are sourced from all Polars DataFrames/LazyFrames referenced in the
5
+ current cell being edited.
6
+
7
+ Usage::
8
+
9
+ from requipso.completer import install
10
+ install()
11
+ """
12
+
13
+ from __future__ import annotations
14
+ from typing import Iterator
15
+
16
+ import ast
17
+ import re
18
+
19
+ import polars as pl
20
+ from IPython.core.getipython import get_ipython
21
+ from IPython.core.completer import Completion
22
+
23
+ # Matches cursor inside pl.col("...") / pl.cols("...") / pl.col(["..."])
24
+ _COL_RE = re.compile(r"pl\.cols?\(\[?[\"']([^\"']*?)$")
25
+
26
+
27
+ def _extract_names(code: str) -> set[str]:
28
+ """Extract variable names referenced in a code string.
29
+
30
+ Uses AST when possible; falls back to regex for incomplete code
31
+ (the usual state when the user is still typing).
32
+ """
33
+ try:
34
+ tree = ast.parse(code)
35
+ return {node.id for node in ast.walk(tree) if isinstance(node, ast.Name)}
36
+ except SyntaxError:
37
+ return set(re.findall(r"\b([a-zA-Z_]\w*)\b", code))
38
+
39
+
40
+ def _polars_col_completions(text: str, offset: int) -> Iterator[Completion]:
41
+ """Yield column-name ``Completion`` objects when the cursor is inside ``pl.col("…")``."""
42
+ text_to_cursor = text[:offset]
43
+ match = _COL_RE.search(text_to_cursor)
44
+ if not match:
45
+ return
46
+
47
+ prefix = match.group(1)
48
+ start = match.start(1)
49
+
50
+ ip = get_ipython()
51
+ if ip is None:
52
+ # IPython environment not available, can't provide completions
53
+ return
54
+ # Collect names referenced in the *current cell* (= full `text`)
55
+ cell_names = _extract_names(text)
56
+
57
+ columns: set[str] = set()
58
+ for name in cell_names:
59
+ obj = ip.user_ns.get(name)
60
+ if isinstance(obj, (pl.DataFrame, pl.LazyFrame)):
61
+ columns.update(obj.columns)
62
+
63
+ for col in sorted(columns):
64
+ if col.startswith(prefix):
65
+ yield Completion(start=start, end=offset, text=col, type="property")
66
+
67
+
68
+ # ---------------------------------------------------------------------------
69
+ # Public API
70
+ # ---------------------------------------------------------------------------
71
+
72
+
73
+ def install() -> None:
74
+ """Patch IPython's completer to suggest Polars column names."""
75
+ ip = get_ipython()
76
+ if ip is None:
77
+ return
78
+
79
+ completer = ip.Completer
80
+
81
+ # Store the original only once so repeated calls don't stack wrappers.
82
+ if not hasattr(completer, "_original_completions"):
83
+ completer._original_completions = completer.completions
84
+
85
+ original = completer._original_completions
86
+
87
+ def _patched(text, offset):
88
+ yield from _polars_col_completions(text, offset)
89
+ yield from original(text, offset)
90
+
91
+ completer.completions = _patched
92
+
93
+
94
+ def uninstall() -> None:
95
+ """Restore the original completer."""
96
+ ip = get_ipython()
97
+ if ip is None:
98
+ return
99
+ completer = ip.Completer
100
+ if hasattr(completer, "_original_completions"):
101
+ completer.completions = completer._original_completions
102
+ del completer._original_completions
103
+ print("✓ Polars column completer removed")
@@ -0,0 +1,72 @@
1
+ import os
2
+ from collections import defaultdict
3
+ from pathlib import Path
4
+
5
+ import polars as pl
6
+
7
+ from requipso.io import S3IO, BaseIO, LocalIO
8
+
9
+
10
+ class Data:
11
+ def __init__(
12
+ self,
13
+ data_path: Path | str,
14
+ empty: bool = False,
15
+ patient_ids: pl.Series | None = None,
16
+ ):
17
+ self.data_path = data_path
18
+ self.io: BaseIO
19
+ self.patient_ids = patient_ids
20
+
21
+ if isinstance(data_path, str) and data_path.startswith("s3://"):
22
+ self.io = S3IO(data_path)
23
+ else:
24
+ self.io = LocalIO(data_path)
25
+
26
+ self.data = self.load_data(empty=empty)
27
+
28
+ self.results = self.data["patient"].select("patient_id")
29
+ self.all_results = defaultdict(dict)
30
+
31
+ def load_data(self, empty: bool = False) -> dict[str, pl.DataFrame]:
32
+ n_rows = 0 if empty else None
33
+ tables = {
34
+ table_name: self.io.read_table(table_name, n_rows=n_rows)
35
+ for table_name in self.io.list_tables()
36
+ }
37
+
38
+ if self.patient_ids is not None:
39
+ tables = {
40
+ name: df.filter(pl.col("patient_id").is_in(self.patient_ids))
41
+ if "patient_id" in df.columns
42
+ else df
43
+ for name, df in tables.items()
44
+ }
45
+ return tables
46
+
47
+ def __getitem__(self, table_name: str) -> pl.DataFrame:
48
+ return self.data[table_name]
49
+
50
+ def __setitem__(self, table_name: str, table: pl.DataFrame) -> None:
51
+ self.data[table_name] = table
52
+
53
+ def __getattr__(self, name: str) -> pl.DataFrame:
54
+ if "data" in self.__dict__ and name in self.data:
55
+ return self.data[name]
56
+ raise AttributeError(f"'{type(self).__name__}' has no attribute '{name}'")
57
+
58
+ def __dir__(self) -> list[str]:
59
+ default = list(super().__dir__())
60
+ if "data" in self.__dict__:
61
+ return list(set(default) | set(self.data.keys()))
62
+ return default
63
+
64
+ def get_table(self, table_name: str) -> pl.DataFrame:
65
+ return self.data[table_name]
66
+
67
+ def set_table(self, table_name: str, table: pl.DataFrame) -> None:
68
+ self.data[table_name] = table
69
+
70
+ def reset(self) -> None:
71
+ self.all_results = defaultdict(dict)
72
+ self.results = self.data["patient"].select("patient_id")