Sync from GitHub via hub-sync

filter_engine.py

@@ -1,6 +1,6 @@
 import marimo
 
-__generated_with = "0.14.16"
+__generated_with = "0.21.0"
 app = marimo.App(
     width="medium",
     app_title="Open Syndrome Definition - Data Browser",
@@ -100,16 +100,64 @@ def _(go, pl):
 
 @app.cell
 def _(mo):
-    mo.md(r"""# Open Syndrome Definition 👩🏽‍🔬""")
+    mo.md(r"""
+    # Open Syndrome Definition 👩🏽‍🔬
+    """)
     return
 
 
 @app.cell
-def _(mo):
+def _():
+    prompt = """
+    Role: Act as an expert in healthcare data engineering and the Open Syndrome Definition (OSD) framework.
+
+    Task: Generate two text files for testing data filtering and syndromic surveillance pipelines.
+
+    File 1: Synthetic Dataset (CSV Format)
+
+    Create a synthetic dataset of roughly 20 ambulatory care records.
+
+    The data must be in [Insert Language, e.g., Brazilian Portuguese, English, German].
+
+    Include the following columns: recording_ts (timestamp), icd_code (ICD-10 code), sex (encoded as [Insert Encoding, e.g., M/F/D]), age (integer), and chief_complaint (string of the symptoms).
+
+    Ensure the clinical presentation logically matches the ICD-10 code and age.
+
+    File 2: OSD Mapping File (YAML Format)
+
+    Create a YAML configuration file that maps the CSV columns to Open Syndrome Definition concepts.
+
+    Define a profile named ambulatory_care.
+
+    Include a value_encodings section that defines the mapping for the sex column.
+
+    Under columns, map each CSV column to its respective OSD concept (e.g., demographic_criteria, diagnosis), attribute (e.g., age, sex), and dtype (integer, string).
+
+    Please output the exact CSV and YAML code in clearly separated code blocks so I can copy them directly into my environment.
+    """
+    return (prompt,)
+
+
+@app.cell
+def _(mo, prompt):
     mo.callout(
         mo.md(
-            "This is a prototype of how to filter your data using definitions from the Open Syndrome Initiative.\n\n"
-            "We do not store any data."
+    f"""
+    This is a prototype for filtering your CSV data using definitions from the [Open Syndrome Initiative](https://opensyndrome.org/).
+
+    You can either provide a sample of your own data, up to 10 MB, or generate a toy dataset using your preferred GenAI tool.
+
+    <details>
+    <summary>Prompt</summary>
+    ```
+    {prompt}
+    ```
+    </details>
+
+    Next, you will need to create a map of your data and the Open Syndrome Definition concepts you want to filter on. Don't worry! We have an example ready for you.
+
+    **Please note that we do not store any data**.
+    """
         ),
         kind="neutral",
     )
@@ -140,8 +188,7 @@ def _(EXAMPLE_DATASETS, mo):
         value=_datasets[0] if _datasets else None,
         label="Example dataset",
     )
-    sample_file = mo.ui.file(kind="area", filetypes=[".csv"])
-
+    sample_file = mo.ui.file(kind="area", filetypes=[".csv"], max_size=10_000_000)
     return data_source, example_picker, sample_file
 
 
@@ -172,7 +219,7 @@ def _(EXAMPLE_DATASETS, data_source, example_picker, pl, sample_file):
 
 @app.cell
 def _(EXAMPLE_DATASETS, data_source, example_picker):
-    _default_yaml = """\
+    _default_yaml = """
     profiles:
     - name: my_dataset
       # value_encodings:  # optional — map OSD canonical values to dataset-specific ones
@@ -218,9 +265,14 @@ def _(df_selected, initial_date_column, initial_yaml, mo):
 
     date_format_input = mo.ui.text(
         value="%Y-%m-%d %H:%M:%S",
-        label="Date format",
+        label="Date format<sup>1</sup>",
     )
 
+    date_block = mo.vstack([
+        date_format_input,
+        mo.md("[^1]: A Python date format code compatible with your data. See other date formats [here](https://strftime.org/).")
+    ])
+
     _cols_hint = "`, `".join(df_selected.columns)
 
     mo.vstack(
@@ -233,7 +285,7 @@ def _(df_selected, initial_date_column, initial_yaml, mo):
                 f"Your dataset columns: `{_cols_hint}`"
             ),
             mo.hstack(
-                [yaml_editor, mo.vstack([date_column_picker, date_format_input])],
+                [yaml_editor, mo.vstack([date_column_picker, date_block])],
                 widths=[3, 1],
                 align="start",
             ),
@@ -307,7 +359,9 @@ def _(definition_options, mo):
 
 @app.cell
 def _(mo):
-    mo.md(r"""### Data sample""")
+    mo.md(r"""
+    ### Data sample
+    """)
     return
 
 
@@ -320,13 +374,17 @@ def _(df_selected, mo):
 
 @app.cell
 def _(mo):
-    mo.md(r"""---""")
+    mo.md(r"""
+    ---
+    """)
     return
 
 
 @app.cell
 def _(mo):
-    mo.md(r"""## Data & Definitions""")
+    mo.md(r"""
+    ## Data & Definitions
+    """)
     return
 
 

pyproject.toml

@@ -5,8 +5,8 @@ description = "Add your description here"
 readme = "README.md"
 requires-python = ">=3.12"
 dependencies = [
-    "marimo>=0.8.0",
-    "opensyndrome @ git+https://github.com/OpenSyndrome/open-syndrome-python.git@filter-engine",
+    "marimo>=0.21.0",
+    "opensyndrome @ git+https://github.com/OpenSyndrome/open-syndrome-python.git@main",
     "plotly>=6.2.0",
     "polars>=1.38.1",
 ]