Semantic link functions for holidays

0.14.0 · active · verified Wed Apr 15

The `semantic-link-functions-holidays` library provides semantic functions, such as `is_holiday`, that enable the enrichment of a FabricDataFrame with public holiday information. It is part of the broader Microsoft Fabric semantic link feature, designed to bridge Power BI datasets and Synapse Data Science notebooks. The current version is 0.14.0, released on March 17, 2026, with updates typically tied to the `semantic-link` project's release cadence.

Warnings

gotcha This library is primarily designed for use within Microsoft Fabric Synapse Data Science notebooks. While parts of `sempy` can be installed standalone, the full integration and benefits of semantic functions are realized in the Fabric environment.
Fix: Ensure you are operating within a Microsoft Fabric notebook for optimal experience, or be aware of the limitations outside of it.
gotcha Semantic functions like `is_holiday` are not directly imported but are dynamically exposed as methods on a `FabricDataFrame`. This occurs when the DataFrame contains columns with appropriate data types and metadata (e.g., 'date' and 'country' columns with 'Date' and 'Country' data categories).
Fix: Structure your `FabricDataFrame` with correct column names and metadata categories for functions to become available. Refer to the quickstart for an example.
gotcha The `semantic-link-functions-holidays` package wraps the `holidays` Python package. Any underlying issues, breaking changes, or specific usage patterns related to the `holidays` package might indirectly affect this semantic function.
Fix: Consult the documentation for the `holidays` package if encountering unexpected behavior with holiday calculations.
gotcha For a complete Semantic Link experience, it is often recommended to install the meta-package `semantic-link` (`pip install semantic-link`). This ensures all individual semantic link function packages, including holidays, are installed. Only installing `semantic-link-functions-holidays` will limit available functionality.
Fix: Consider `pip install semantic-link` if you need broader semantic link features beyond just holiday detection.
gotcha In Microsoft Fabric with Spark 3.4 or later, the core `semantic-link` package is preinstalled. However, specific function packages like `semantic-link-functions-holidays` may still need to be explicitly installed or updated to their latest version.
Fix: Always run `%pip install -U semantic-link` (or the specific function package) in your notebook cell to ensure you have the latest versions of all semantic link components.

Install

pip install semantic-link-functions-holidays Install only holidays functions
pip install semantic-link Install all semantic link packages (recommended)

Imports

FabricDataFrame
```
from sempy.fabric import FabricDataFrame
```
Semantic functions like `is_holiday` are dynamically exposed as methods on a FabricDataFrame, rather than imported directly. FabricDataFrame itself is imported from `sempy.fabric`.

Quickstart

This quickstart demonstrates how to create a `FabricDataFrame` and use the dynamically exposed `is_holiday` semantic function to determine if a given date is a public holiday in a specified country. The function relies on column metadata to identify relevant date and country columns.

from sempy.fabric import FabricDataFrame
import pandas as pd

# Create a FabricDataFrame with relevant columns and metadata
# In a real Microsoft Fabric environment, FabricDataFrame often comes from reading a Power BI semantic model.
df = FabricDataFrame(
    pd.DataFrame({
        "country": ["US", "AT", "DE", "US"],
        "date": ["2023-01-01", "2023-01-06", "2023-10-03", "2023-07-04"]
    }),
    column_metadata={
        "country": {"data_category": "Country"},
        "date": {"data_category": "Date"}
    }
)

# The is_holiday function is dynamically exposed if appropriate columns and metadata exist.
# It will return True if the date is a holiday in the given country.
df["is_public_holiday"] = df.is_holiday(col_country="country", col_date="date")

print(df)

view raw JSON →