|
| 1 | +.. Licensed to the Apache Software Foundation (ASF) under one |
| 2 | +.. or more contributor license agreements. See the NOTICE file |
| 3 | +.. distributed with this work for additional information |
| 4 | +.. regarding copyright ownership. The ASF licenses this file |
| 5 | +.. to you under the Apache License, Version 2.0 (the |
| 6 | +.. "License"); you may not use this file except in compliance |
| 7 | +.. with the License. You may obtain a copy of the License at |
| 8 | +
|
| 9 | +.. http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | +
|
| 11 | +.. Unless required by applicable law or agreed to in writing, |
| 12 | +.. software distributed under the License is distributed on an |
| 13 | +.. "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
| 14 | +.. KIND, either express or implied. See the License for the |
| 15 | +.. specific language governing permissions and limitations |
| 16 | +.. under the License. |
| 17 | +
|
| 18 | +DataFrames |
| 19 | +========== |
| 20 | + |
| 21 | +Overview |
| 22 | +-------- |
| 23 | + |
| 24 | +DataFusion's DataFrame API provides a powerful interface for building and executing queries against data sources. |
| 25 | +It offers a familiar API similar to pandas and other DataFrame libraries, but with the performance benefits of Rust |
| 26 | +and Arrow. |
| 27 | + |
| 28 | +A DataFrame represents a logical plan that can be composed through operations like filtering, projection, and aggregation. |
| 29 | +The actual execution happens when terminal operations like ``collect()`` or ``show()`` are called. |
| 30 | + |
| 31 | +Basic Usage |
| 32 | +----------- |
| 33 | + |
| 34 | +.. code-block:: python |
| 35 | +
|
| 36 | + import datafusion |
| 37 | + from datafusion import col, lit |
| 38 | +
|
| 39 | + # Create a context and register a data source |
| 40 | + ctx = datafusion.SessionContext() |
| 41 | + ctx.register_csv("my_table", "path/to/data.csv") |
| 42 | + |
| 43 | + # Create and manipulate a DataFrame |
| 44 | + df = ctx.sql("SELECT * FROM my_table") |
| 45 | + |
| 46 | + # Or use the DataFrame API directly |
| 47 | + df = (ctx.table("my_table") |
| 48 | + .filter(col("age") > lit(25)) |
| 49 | + .select([col("name"), col("age")])) |
| 50 | + |
| 51 | + # Execute and collect results |
| 52 | + result = df.collect() |
| 53 | + |
| 54 | + # Display the first few rows |
| 55 | + df.show() |
| 56 | +
|
| 57 | +HTML Rendering |
| 58 | +-------------- |
| 59 | + |
| 60 | +When working in Jupyter notebooks or other environments that support HTML rendering, DataFrames will |
| 61 | +automatically display as formatted HTML tables, making it easier to visualize your data. |
| 62 | + |
| 63 | +The ``_repr_html_`` method is called automatically by Jupyter to render a DataFrame. This method |
| 64 | +controls how DataFrames appear in notebook environments, providing a richer visualization than |
| 65 | +plain text output. |
| 66 | + |
| 67 | +Customizing HTML Rendering |
| 68 | +-------------------------- |
| 69 | + |
| 70 | +You can customize how DataFrames are rendered in HTML by configuring the formatter: |
| 71 | + |
| 72 | +.. code-block:: python |
| 73 | +
|
| 74 | + from datafusion.html_formatter import configure_formatter |
| 75 | + |
| 76 | + # Change the default styling |
| 77 | + configure_formatter( |
| 78 | + max_rows=50, # Maximum number of rows to display |
| 79 | + max_width=None, # Maximum width in pixels (None for auto) |
| 80 | + theme="light", # Theme: "light" or "dark" |
| 81 | + precision=2, # Floating point precision |
| 82 | + thousands_separator=",", # Separator for thousands |
| 83 | + date_format="%Y-%m-%d", # Date format |
| 84 | + truncate_width=20 # Max width for string columns before truncating |
| 85 | + ) |
| 86 | +
|
| 87 | +The formatter settings affect all DataFrames displayed after configuration. |
| 88 | + |
| 89 | +Custom Style Providers |
| 90 | +---------------------- |
| 91 | + |
| 92 | +For advanced styling needs, you can create a custom style provider: |
| 93 | + |
| 94 | +.. code-block:: python |
| 95 | +
|
| 96 | + from datafusion.html_formatter import StyleProvider, configure_formatter |
| 97 | + |
| 98 | + class MyStyleProvider(StyleProvider): |
| 99 | + def get_table_styles(self): |
| 100 | + return { |
| 101 | + "table": "border-collapse: collapse; width: 100%;", |
| 102 | + "th": "background-color: #007bff; color: white; padding: 8px; text-align: left;", |
| 103 | + "td": "border: 1px solid #ddd; padding: 8px;", |
| 104 | + "tr:nth-child(even)": "background-color: #f2f2f2;", |
| 105 | + } |
| 106 | + |
| 107 | + def get_value_styles(self, dtype, value): |
| 108 | + """Return custom styles for specific values""" |
| 109 | + if dtype == "float" and value < 0: |
| 110 | + return "color: red;" |
| 111 | + return None |
| 112 | + |
| 113 | + # Apply the custom style provider |
| 114 | + configure_formatter(style_provider=MyStyleProvider()) |
| 115 | +
|
| 116 | +Creating a Custom Formatter |
| 117 | +--------------------------- |
| 118 | + |
| 119 | +For complete control over rendering, you can implement a custom formatter: |
| 120 | + |
| 121 | +.. code-block:: python |
| 122 | +
|
| 123 | + from datafusion.html_formatter import Formatter, get_formatter |
| 124 | + |
| 125 | + class MyFormatter(Formatter): |
| 126 | + def format_html(self, batches, schema, has_more=False, table_uuid=None): |
| 127 | + # Create your custom HTML here |
| 128 | + html = "<div class='my-custom-table'>" |
| 129 | + # ... formatting logic ... |
| 130 | + html += "</div>" |
| 131 | + return html |
| 132 | + |
| 133 | + # Set as the global formatter |
| 134 | + configure_formatter(formatter_class=MyFormatter) |
| 135 | + |
| 136 | + # Or use the formatter just for specific operations |
| 137 | + formatter = get_formatter() |
| 138 | + custom_html = formatter.format_html(batches, schema) |
| 139 | +
|
| 140 | +Managing Formatters |
| 141 | +------------------- |
| 142 | + |
| 143 | +Reset to default formatting: |
| 144 | + |
| 145 | +.. code-block:: python |
| 146 | +
|
| 147 | + from datafusion.html_formatter import reset_formatter |
| 148 | + |
| 149 | + # Reset to default settings |
| 150 | + reset_formatter() |
| 151 | +
|
| 152 | +Get the current formatter settings: |
| 153 | + |
| 154 | +.. code-block:: python |
| 155 | +
|
| 156 | + from datafusion.html_formatter import get_formatter |
| 157 | + |
| 158 | + formatter = get_formatter() |
| 159 | + print(formatter.max_rows) |
| 160 | + print(formatter.theme) |
| 161 | +
|
| 162 | +Contextual Formatting |
| 163 | +--------------------- |
| 164 | + |
| 165 | +You can also use a context manager to temporarily change formatting settings: |
| 166 | + |
| 167 | +.. code-block:: python |
| 168 | +
|
| 169 | + from datafusion.html_formatter import formatting_context |
| 170 | + |
| 171 | + # Default formatting |
| 172 | + df.show() |
| 173 | + |
| 174 | + # Temporarily use different formatting |
| 175 | + with formatting_context(max_rows=100, theme="dark"): |
| 176 | + df.show() # Will use the temporary settings |
| 177 | + |
| 178 | + # Back to default formatting |
| 179 | + df.show() |
0 commit comments