Skip to content

Dimensions

toolbox_pyspark.dimensions 🔗

Summary

The dimensions module is used for checking the dimensions of pyspark dataframe's.

get_dims 🔗

get_dims(
    dataframe: psDataFrame,
    use_names: bool = True,
    use_comma: bool = True,
) -> Union[
    dict[str, str],
    dict[str, int],
    tuple[str, str],
    tuple[int, int],
]

Summary

Extract the dimensions of a given dataframe.

Parameters:

Name Type Description Default
dataframe DataFrame

The table to check.

 required
use_names bool

Whether or not to add names to the returned object.
If True, then will return a dict with two keys only, for the number of rows and cols.
Defaults to True.

 True
use_comma bool

Whether or not to add a comma , to the returned object.
Defaults to True.

 True

Raises:

Type Description
TypeError

If any of the inputs parsed to the parameters of this function are not the correct type. Uses the @typeguard.typechecked decorator.

Returns:

Type Description
Union[Dict[str, Union[str, int]], tuple[str, ...], tTuple[int, ...]]

The dimensions of the given dataframe.
Examples

Set up
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
>>> # Imports
>>> import pandas as pd
>>> from pyspark.sql import SparkSession
>>> from toolbox_pyspark.dimensions import get_dims
>>>
>>> # Instantiate Spark
>>> spark = SparkSession.builder.getOrCreate()
>>>
>>> # Create data
>>> df = spark.createDataFrame(
...     pd.DataFrame({
...         'a': range(5000),
...         'b': range(5000),
...     })
... )
>>>
>>> # Check
>>> print(df.count())
>>> print(len(df.columns))
Terminal
5000
Terminal
2

Names and commas
1
>>> print(get_dims(dataframe=df, use_names=True, use_commas=True))
Terminal
{"rows": "5,000", "cols": "2"}

Names but no commas
1
>>> print(get_dims(dataframe=df, use_names=True, use_commas=False))
Terminal
{"rows": 5000, "cols": 2}

Commas but no names
1
>>> print(get_dims(dataframe=df, use_names=False, use_commas=True))
Terminal
("5,000", "2")

Neither names nor commas
1
>>> print(get_dims(dataframe=df, use_names=False, use_commas=False))
Terminal
(5000, 2)

Source code in src/toolbox_pyspark/dimensions.py 
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
@typechecked
def get_dims(
    dataframe: psDataFrame,
    use_names: bool = True,
    use_comma: bool = True,
) -> Union[dict[str, str], dict[str, int], tuple[str, str], tuple[int, int]]:
    """
    !!! note "Summary"
        Extract the dimensions of a given `dataframe`.

    Params:
        dataframe (psDataFrame):
            The table to check.
        use_names (bool, optional):
            Whether or not to add `names` to the returned object.<br>
            If `#!py True`, then will return a `#!py dict` with two keys only, for the number of `rows` and `cols`.<br>
            Defaults to `#!py True`.
        use_comma (bool, optional):
            Whether or not to add a comma `,` to the returned object.<br>
            Defaults to `#!py True`.

    Raises:
        TypeError:
            If any of the inputs parsed to the parameters of this function are not the correct type. Uses the [`@typeguard.typechecked`](https://typeguard.readthedocs.io/en/stable/api.html#typeguard.typechecked) decorator.

    Returns:
        (Union[Dict[str, Union[str, int]], tuple[str, ...], tTuple[int, ...]]):
            The dimensions of the given `dataframe`.

    ???+ example "Examples"

        ```{.py .python linenums="1" title="Set up"}
        >>> # Imports
        >>> import pandas as pd
        >>> from pyspark.sql import SparkSession
        >>> from toolbox_pyspark.dimensions import get_dims
        >>>
        >>> # Instantiate Spark
        >>> spark = SparkSession.builder.getOrCreate()
        >>>
        >>> # Create data
        >>> df = spark.createDataFrame(
        ...     pd.DataFrame({
        ...         'a': range(5000),
        ...         'b': range(5000),
        ...     })
        ... )
        >>>
        >>> # Check
        >>> print(df.count())
        >>> print(len(df.columns))
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        5000
        ```
        ```{.txt .text title="Terminal"}
        2
        ```
        </div>

        ```{.py .python linenums="1" title="Names and commas"}
        >>> print(get_dims(dataframe=df, use_names=True, use_commas=True))
        ```
        <div class="result" markdown>
        ```{.sh .shell  title="Terminal"}
        {"rows": "5,000", "cols": "2"}
        ```
        </div>

        ```{.py .python linenums="1" title="Names but no commas"}
        >>> print(get_dims(dataframe=df, use_names=True, use_commas=False))
        ```
        <div class="result" markdown>
        ```{.sh .shell  title="Terminal"}
        {"rows": 5000, "cols": 2}
        ```
        </div>

        ```{.py .python linenums="1" title="Commas but no names"}
        >>> print(get_dims(dataframe=df, use_names=False, use_commas=True))
        ```
        <div class="result" markdown>
        ```{.sh .shell  title="Terminal"}
        ("5,000", "2")
        ```
        </div>

        ```{.py .python linenums="1" title="Neither names nor commas"}
        >>> print(get_dims(dataframe=df, use_names=False, use_commas=False))
        ```
        <div class="result" markdown>
        ```{.sh .shell  title="Terminal"}
        (5000, 2)
        ```
        </div>
    """
    dims: tuple[int, int] = (dataframe.count(), len(dataframe.columns))
    if use_names and use_comma:
        return {"rows": f"{dims[0]:,}", "cols": f"{dims[1]:,}"}
    elif use_names and not use_comma:
        return {"rows": dims[0], "cols": dims[1]}
    elif not use_names and use_comma:
        return (f"{dims[0]:,}", f"{dims[1]:,}")
    else:
        return dims

get_dims_of_tables 🔗

get_dims_of_tables(
    tables: str_list,
    scope: Optional[dict] = None,
    use_comma: bool = True,
) -> pdDataFrame

Summary

Take in a list of the names of some tables, and for each of them, check their dimensions.

Details

This function will check against the global() scope. So you need to be careful if you're dealing with massive amounts of data in memory.

Parameters:

Name Type Description Default
tables str_list

The list of the tables that will be checked.

 required
scope dict

This is the scope against which the tables will be checked.
If None, then it will use the global() scope by default..
Defaults to None.

 None
use_comma bool

Whether or not the dimensions from the tables should be formatted as a string with a comma as the thousandths delimiter.
Defaults to True.

 True

Raises:

Type Description
TypeError

If any of the inputs parsed to the parameters of this function are not the correct type. Uses the @typeguard.typechecked decorator.

Returns:

Type Description
DataFrame

A pandas dataframe with four columns: ["table", "type", "rows", "cols"].
Examples

Set up
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
>>> # Imports
>>> import pandas as pd
>>> from pyspark.sql import SparkSession
>>> from toolbox_pyspark.dimensions import get_dims_of_tables, get_dims
>>>
>>> # Instantiate Spark
>>> spark = SparkSession.builder.getOrCreate()
>>>
>>> # Create data
>>> df1 = spark.createDataFrame(
...     pd.DataFrame({
...         'a': range(5000),
...         'b': range(5000),
...     })
... )
>>> df2 = spark.createDataFrame(
...     pd.DataFrame({
...         'a': range(10000),
...         'b': range(10000),
...         'c': range(10000),
...     })
... )
>>>
>>> # Check
>>> print(get_dims(df1))
>>> print(get_dims(df1))
Terminal
{"rows": "5000", "cols": "2"}
Terminal
{"rows": "10000", "cols": "3"}

Basic usage
1
>>> print(get_dims_of_tables(['df1', 'df2']))

  table type  rows cols
0   df1      5,000    2
1   df2      1,000    3

No commas
1
>>> print(get_dims_of_tables(['df1', 'df2'], use_commas=False))

  table type rows cols
0   df1      5000    2
1   df2      1000    3

Missing DF
1
>>> display(get_dims_of_tables(['df1', 'df2', 'df3'], use_comma=False))

  table type rows cols
0   df1      5000    2
1   df2      1000    3
1   df3       NaN  NaN

Notes
  • The first column of the returned table is the name of the table from the scope provided.
  • The second column of the returned table is the type of the table. That is, whether the table is one of ["prd", "arc", "acm"], which are for 'production', 'archive', accumulation' categories. This is designated by the table containing an underscore (_), and having a suffic of either one of: "prd", "arc", or "acm". If the table does not contain this info, then the value in this second column will just be blank.
  • If one of the tables given in the tables list does not exist in the scope, then the values given in the rows and cols columns will either be the values: np.nan or "Did not load".
Source code in src/toolbox_pyspark/dimensions.py 
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
@typechecked
def get_dims_of_tables(
    tables: str_list,
    scope: Optional[dict] = None,
    use_comma: bool = True,
) -> pdDataFrame:
    """
    !!! note "Summary"
        Take in a list of the names of some tables, and for each of them, check their dimensions.

    ???+ abstract "Details"
        This function will check against the `#!py global()` scope. So you need to be careful if you're dealing with massive amounts of data in memory.

    Params:
        tables (str_list):
            The list of the tables that will be checked.
        scope (dict, optional):
            This is the scope against which the tables will be checked.<br>
            If `#!py None`, then it will use the `#!py global()` scope by default..<br>
            Defaults to `#!py None`.
        use_comma (bool, optional):
            Whether or not the dimensions from the tables should be formatted as a string with a comma as the thousandths delimiter.<br>
            Defaults to `#!py True`.

    Raises:
        TypeError:
            If any of the inputs parsed to the parameters of this function are not the correct type. Uses the [`@typeguard.typechecked`](https://typeguard.readthedocs.io/en/stable/api.html#typeguard.typechecked) decorator.

    Returns:
        (pdDataFrame):
            A `pandas` `dataframe` with four columns: `#!py ["table", "type", "rows", "cols"]`.

    ???+ example "Examples"

        ```{.py .python linenums="1" title="Set up"}
        >>> # Imports
        >>> import pandas as pd
        >>> from pyspark.sql import SparkSession
        >>> from toolbox_pyspark.dimensions import get_dims_of_tables, get_dims
        >>>
        >>> # Instantiate Spark
        >>> spark = SparkSession.builder.getOrCreate()
        >>>
        >>> # Create data
        >>> df1 = spark.createDataFrame(
        ...     pd.DataFrame({
        ...         'a': range(5000),
        ...         'b': range(5000),
        ...     })
        ... )
        >>> df2 = spark.createDataFrame(
        ...     pd.DataFrame({
        ...         'a': range(10000),
        ...         'b': range(10000),
        ...         'c': range(10000),
        ...     })
        ... )
        >>>
        >>> # Check
        >>> print(get_dims(df1))
        >>> print(get_dims(df1))
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        {"rows": "5000", "cols": "2"}
        ```
        ```{.txt .text title="Terminal"}
        {"rows": "10000", "cols": "3"}
        ```
        </div>

        ```{.py .python linenums="1" title="Basic usage"}
        >>> print(get_dims_of_tables(['df1', 'df2']))
        ```
        <div class="result" markdown>
        ```{.txt .text}
          table type  rows cols
        0   df1      5,000    2
        1   df2      1,000    3
        ```
        </div>

        ```{.py .python linenums="1" title="No commas"}
        >>> print(get_dims_of_tables(['df1', 'df2'], use_commas=False))
        ```
        <div class="result" markdown>
        ```{.txt .text}
          table type rows cols
        0   df1      5000    2
        1   df2      1000    3
        ```
        </div>

        ```{.py .python linenums="1" title="Missing DF"}
        >>> display(get_dims_of_tables(['df1', 'df2', 'df3'], use_comma=False))
        ```
        <div class="result" markdown>
        ```{.txt .text}
          table type rows cols
        0   df1      5000    2
        1   df2      1000    3
        1   df3       NaN  NaN
        ```
        </div>

    ??? info "Notes"
        - The first column of the returned table is the name of the table from the `scope` provided.
        - The second column of the returned table is the `type` of the table. That is, whether the table is one of `#!py ["prd", "arc", "acm"]`, which are for 'production', 'archive', accumulation' categories. This is designated by the table containing an underscore (`_`), and having a suffic of either one of: `#!py "prd"`, `#!py "arc"`, or `#!py "acm"`. If the table does not contain this info, then the value in this second column will just be blank.
        - If one of the tables given in the `tables` list does not exist in the `scope`, then the values given in the `rows` and `cols` columns will either be the values: `#!py np.nan` or `#!py "Did not load"`.
    """
    sizes: Dict[str, list] = {
        "table": list(),
        "type": list(),
        "rows": list(),
        "cols": list(),
    }
    rows: Union[str, int, float]
    cols: Union[str, int, float]
    for tbl, typ in [
        (
            table.rsplit("_", 1)
            if "_" in table and table.endswith(("acm", "arc", "prd"))
            else (table, "")
        )
        for table in tables
    ]:
        try:
            tmp: psDataFrame = eval(
                f"{tbl}{f'_{typ}' if typ!='' else ''}",
                globals() if scope is None else scope,
            )
            rows, cols = get_dims(tmp, use_names=False, use_comma=use_comma)
        except Exception:
            if use_comma:
                rows = cols = "Did not load"
            else:
                rows = cols = np.nan
        sizes["table"].append(tbl)
        sizes["type"].append(typ)
        sizes["rows"].append(rows)
        sizes["cols"].append(cols)
    return pdDataFrame(sizes)

make_dimension_table 🔗

make_dimension_table(
    dataframe: psDataFrame,
    columns: Union[str, str_collection],
    index_prefix: str = "id",
) -> psDataFrame

Summary

Create a dimension table from the specified columns of a given pyspark dataframe.

Details

This function will create a dimension table from the specified columns of a given pyspark dataframe. The dimension table will contain the unique values of the specified columns, along with an index column that will be used to replace the original columns in the original dataframe.

index column will be named according to the index_prefix parameter. If only one column is specified, then the index column will be named according to the index_prefix parameter followed by the name of the column. If multiple columns are specified, then the index column will be named according to the index_prefix parameter only. The index column will be created by using the row_number() window function over the specified columns.

The dimension table will be created by selecting the specified columns from the original dataframe, then applying the distinct() function to get the unique values, and finally applying the row_number() window function to create the index column.

Parameters:

Name Type Description Default
dataframe DataFrame

The DataFrame to create the dimension table from.

 required
columns Union[str, str_collection]

The column(s) to include in the dimension table.

 required
index_prefix str

The prefix to use for the index column.
Defaults to "id".

 'id'

Raises:

Type Description
TypeError

If any of the inputs parsed to the parameters of this function are not the correct type. Uses the @typeguard.typechecked decorator.
ColumnDoesNotExistError

If any of the columns specified do not exist in the dataframe.

Returns:

Type Description
DataFrame

The dimension table.
Examples

Set up
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
>>> # Imports
>>> import pandas as pd
>>> from pyspark.sql import SparkSession
>>> from toolbox_pyspark.dimensions import make_dimension_table
>>>
>>> # Instantiate Spark
>>> spark = SparkSession.builder.getOrCreate()
>>>
>>> # Create data
>>> df = spark.createDataFrame(
...     pd.DataFrame(
...         {
...             "a": [1, 2, 3, 4],
...             "b": ["a", "b", "c", "d"],
...             "c": [1, 1, 2, 2],
...             "d": ["a", "b", "b", "b"],
...             "e": ["x", "x", "y", "z"],
...         }
...     )
... )
>>>
>>> # Check
>>> df.show()
Terminal
+---+---+---+---+---+
| a | b | c | d | e |
+---+---+---+---+---+
| 1 | a | 1 | a | x |
| 2 | b | 1 | b | x |
| 3 | c | 2 | b | y |
| 4 | d | 2 | b | z |
+---+---+---+---+---+

Example 1: Create dimension table with single column
1
2
>>> dim_table = make_dimension_table(df, "d")
>>> dim_table.show()
Terminal
+------+---+
| id_d | d |
+------+---+
|    1 | a |
|    2 | b |
+------+---+

Conclusion: Successfully created dimension table with single column.

Example 2: Create dimension table with multiple columns
1
2
>>> dim_table = make_dimension_table(df, ["c", "d"])
>>> dim_table.show()
Terminal
+----+---+---+
| id | c | d |
+----+---+---+
|  1 | 1 | a |
|  2 | 1 | b |
|  3 | 2 | b |
+----+---+---+

Conclusion: Successfully created dimension table with multiple columns.

Example 3: Use different prefix
1
2
>>> dim_table = make_dimension_table(df, "e", "index")
>>> dim_table.show()
Terminal
+---------+---+
| index_e | e |
+---------+---+
|       1 | x |
|       2 | y |
|       3 | z |
+---------+---+

Conclusion: Successfully created dimension table with different prefix.

Example 4: Invalid column
1
>>> dim_table = make_dimension_table(df, "123")
Terminal
ColumnDoesNotExistError: Column '123' does not exist in the DataFrame.

Conclusion: Failed to create dimension table due to invalid column name.

See Also
Source code in src/toolbox_pyspark/dimensions.py 
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
@typechecked
def make_dimension_table(
    dataframe: psDataFrame,
    columns: Union[str, str_collection],
    index_prefix: str = "id",
) -> psDataFrame:
    """
    !!! note "Summary"
        Create a dimension table from the specified columns of a given `pyspark` dataframe.

    ???+ abstract "Details"
        This function will create a dimension table from the specified columns of a given `pyspark` dataframe. The dimension table will contain the unique values of the specified columns, along with an index column that will be used to replace the original columns in the original dataframe.

        index column will be named according to the `index_prefix` parameter. If only one column is specified, then the index column will be named according to the `index_prefix` parameter followed by the name of the column. If multiple columns are specified, then the index column will be named according to the `index_prefix` parameter only. The index column will be created by using the `#!py row_number()` window function over the specified columns.

        The dimension table will be created by selecting the specified columns from the original dataframe, then applying the `#!py distinct()` function to get the unique values, and finally applying the `#!py row_number()` window function to create the index column.

    Params:
        dataframe (psDataFrame):
            The DataFrame to create the dimension table from.
        columns (Union[str, str_collection]):
            The column(s) to include in the dimension table.
        index_prefix (str, optional):
            The prefix to use for the index column.<br>
            Defaults to `#!py "id"`.

    Raises:
        TypeError:
            If any of the inputs parsed to the parameters of this function are not the correct type. Uses the [`@typeguard.typechecked`](https://typeguard.readthedocs.io/en/stable/api.html#typeguard.typechecked) decorator.
        ColumnDoesNotExistError:
            If any of the columns specified do not exist in the dataframe.

    Returns:
        (psDataFrame):
            The dimension table.

    ???+ example "Examples"

        ```{.py .python linenums="1" title="Set up"}
        >>> # Imports
        >>> import pandas as pd
        >>> from pyspark.sql import SparkSession
        >>> from toolbox_pyspark.dimensions import make_dimension_table
        >>>
        >>> # Instantiate Spark
        >>> spark = SparkSession.builder.getOrCreate()
        >>>
        >>> # Create data
        >>> df = spark.createDataFrame(
        ...     pd.DataFrame(
        ...         {
        ...             "a": [1, 2, 3, 4],
        ...             "b": ["a", "b", "c", "d"],
        ...             "c": [1, 1, 2, 2],
        ...             "d": ["a", "b", "b", "b"],
        ...             "e": ["x", "x", "y", "z"],
        ...         }
        ...     )
        ... )
        >>>
        >>> # Check
        >>> df.show()
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        +---+---+---+---+---+
        | a | b | c | d | e |
        +---+---+---+---+---+
        | 1 | a | 1 | a | x |
        | 2 | b | 1 | b | x |
        | 3 | c | 2 | b | y |
        | 4 | d | 2 | b | z |
        +---+---+---+---+---+
        ```
        </div>

        ```{.py .python linenums="1" title="Example 1: Create dimension table with single column"}
        >>> dim_table = make_dimension_table(df, "d")
        >>> dim_table.show()
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        +------+---+
        | id_d | d |
        +------+---+
        |    1 | a |
        |    2 | b |
        +------+---+
        ```
        !!! success "Conclusion: Successfully created dimension table with single column."
        </div>

        ```{.py .python linenums="1" title="Example 2: Create dimension table with multiple columns"}
        >>> dim_table = make_dimension_table(df, ["c", "d"])
        >>> dim_table.show()
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        +----+---+---+
        | id | c | d |
        +----+---+---+
        |  1 | 1 | a |
        |  2 | 1 | b |
        |  3 | 2 | b |
        +----+---+---+
        ```
        !!! success "Conclusion: Successfully created dimension table with multiple columns."
        </div>

        ```{.py .python linenums="1" title="Example 3: Use different prefix"}
        >>> dim_table = make_dimension_table(df, "e", "index")
        >>> dim_table.show()
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        +---------+---+
        | index_e | e |
        +---------+---+
        |       1 | x |
        |       2 | y |
        |       3 | z |
        +---------+---+
        ```
        !!! success "Conclusion: Successfully created dimension table with different prefix."
        </div>

        ```{.py .python linenums="1" title="Example 4: Invalid column"}
        >>> dim_table = make_dimension_table(df, "123")
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        ColumnDoesNotExistError: Column '123' does not exist in the DataFrame.
        ```
        !!! failure "Conclusion: Failed to create dimension table due to invalid column name."
        </div>

    ??? tip "See Also"
        - [`replace_columns_with_dimension_id`][toolbox_pyspark.dimensions.replace_columns_with_dimension_id]
    """
    columns = [columns] if is_type(columns, str) else columns
    assert_columns_exists(dataframe, columns)
    index_name: str = f"{index_prefix}_{columns[0]}" if len(columns) == 1 else index_prefix
    return (
        dataframe.select(*columns)
        .distinct()
        .withColumn(
            index_name,
            F.expr(f"row_number() over (order by {', '.join(columns)})").cast("int"),
        )
        .select(index_name, *columns)
    )

replace_columns_with_dimension_id 🔗

replace_columns_with_dimension_id(
    fct_dataframe: psDataFrame,
    dim_dataframe: psDataFrame,
    cols_to_replace: Union[str, str_collection],
    dim_id_col: Optional[str] = None,
) -> psDataFrame

Summary

Replace the specified columns in a given pyspark dataframe with the corresponding dimension table IDs.

Details

This function will replace the specified columns in a given pyspark dataframe with the corresponding dimension table IDs. The dimension table IDs will be obtained by joining the dimension table with the original dataframe on the specified columns. The original columns will then be dropped from the original dataframe.

The dimension table IDs will be added to the original dataframe to replace the columns specified in cols_to_replace. The dimension table IDs will be obtained by joining the dimension table with the original dataframe on the specified columns.

The join will be performed using a left join, so that any rows in the original dataframe that do not have a corresponding row in the dimension table will have a null value for the dimension table ID. The original columns will be dropped from the original dataframe after the join. The resulting dataframe will have the same number of rows as the original dataframe, but with the specified columns replaced by the dimension table IDs.

Parameters:

Name Type Description Default
fct_dataframe DataFrame

The DataFrame to replace the columns in.

 required
dim_dataframe DataFrame

The dimension table containing the IDs.

 required
cols_to_replace Union[str, str_collection]

The column(s) to replace with the dimension table IDs.

 required
dim_id_col str

The name of the column in the dimension table containing the IDs.
If None, then will use the first column of the dimension table.
Defaults to None.

 None

Raises:

Type Description
TypeError

If any of the inputs parsed to the parameters of this function are not the correct type. Uses the @typeguard.typechecked decorator.
ColumnDoesNotExistError

If any of the columns specified do not exist in the dataframes.

Returns:

Type Description
DataFrame

The DataFrame with the columns replaced by the dimension table IDs.
Examples

Set up
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
>>> # Imports
>>> import pandas as pd
>>> from pyspark.sql import SparkSession
>>>
>>> # Instantiate Spark
>>> from toolbox_pyspark.dimensions import make_dimension_table, replace_columns_with_dimension_id
>>> spark = SparkSession.builder.getOrCreate()
>>>
>>> # Create data
>>> df = spark.createDataFrame(
...     pd.DataFrame(
...         {
...             "a": [1, 2, 3, 4],
...             "b": ["a", "b", "c", "d"],
...             "c": [1, 1, 2, 2],
...             "d": ["a", "b", "b", "b"],
...             "e": ["x", "x", "y", "z"],
...         }
...     )
... )
>>> dim_table1 = make_dimension_table(df, "d")
>>> dim_table2 = make_dimension_table(df, "e")
>>> dim_table3 = make_dimension_table(df, ("c", "d"))
>>>
>>> # Check
>>> df.show()
>>> dim_table1.show()
>>> dim_table2.show()
>>> dim_table3.show()
Terminal
+---+---+---+---+---+
| a | b | c | d | e |
+---+---+---+---+---+
| 1 | a | 1 | a | x |
| 2 | b | 1 | b | x |
| 3 | c | 2 | b | y |
| 4 | d | 2 | b | z |
+---+---+---+---+---+
Terminal
+------+---+
| id_d | d |
+------+---+
|    1 | a |
|    2 | b |
+------+---+
Terminal
+------+---+
| id_e | e |
+------+---+
|    1 | x |
|    2 | y |
|    3 | z |
+------+---+
Terminal
+----+---+---+
| id | c | d |
+----+---+---+
|  1 | 1 | a |
|  2 | 1 | b |
|  3 | 2 | b |
+----+---+---+

Example 1: Replace single column with dimension ID
1
2
>>> df_replaced = replace_columns_with_dimension_id(df, dim_table1, "d")
>>> df_replaced.show()
Terminal
+---+---+---+------+---+
| a | b | c | id_d | e |
+---+---+---+------+---+
| 1 | a | 1 |    1 | x |
| 2 | b | 1 |    2 | x |
| 3 | c | 2 |    2 | y |
| 4 | d | 2 |    2 | z |
+---+---+---+------+---+

Conclusion: Successfully replaced single column with dimension ID.

Example 2: Replace single column with dimension ID
1
2
>>> df_replaced = replace_columns_with_dimension_id(df, dim_table2, "e")
>>> df_replaced.show()
Terminal
+---+---+---+---+------+
| a | b | c | d | id_e |
+---+---+---+---+------+
| 1 | a | 1 | a |    1 |
| 2 | b | 1 | b |    1 |
| 3 | c | 2 | b |    2 |
| 4 | d | 2 | b |    3 |
+---+---+---+---+------+

Conclusion: Successfully replaced single column with dimension ID.

Example 3: Replace multiple columns with dimension IDs
1
2
>>> df_replaced_multi = replace_columns_with_dimension_id(df, dim_table3, ["c", "d"])
>>> df_replaced_multi.show()
Terminal
+---+---+----+---+
| a | b | id | e |
+---+---+----+---+
| 1 | a |  1 | x |
| 2 | b |  2 | x |
| 3 | c |  3 | y |
| 4 | d |  3 | z |
+---+---+----+---+

Conclusion: Successfully replaced multiple columns with dimension IDs.

Example 4: Invalid column type
1
>>> df_replaced = replace_columns_with_dimension_id(df, dim_table, "123")
Terminal
ColumnDoesNotExistError: Column '123' does not exist in the DataFrame.

Conclusion: Failed to replace columns due to invalid column type.

See Also
Source code in src/toolbox_pyspark/dimensions.py 
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
@typechecked
def replace_columns_with_dimension_id(
    fct_dataframe: psDataFrame,
    dim_dataframe: psDataFrame,
    cols_to_replace: Union[str, str_collection],
    dim_id_col: Optional[str] = None,
) -> psDataFrame:
    """
    !!! note "Summary"
        Replace the specified columns in a given `pyspark` dataframe with the corresponding dimension table IDs.

    ???+ abstract "Details"
        This function will replace the specified columns in a given `pyspark` dataframe with the corresponding dimension table IDs. The dimension table IDs will be obtained by joining the dimension table with the original dataframe on the specified columns. The original columns will then be dropped from the original dataframe.

        The dimension table IDs will be added to the original dataframe to replace the columns specified in `cols_to_replace`. The dimension table IDs will be obtained by joining the dimension table with the original dataframe on the specified columns.

        The join will be performed using a left join, so that any rows in the original dataframe that do not have a corresponding row in the dimension table will have a `#!sql null` value for the dimension table ID. The original columns will be dropped from the original dataframe after the join.         The resulting dataframe will have the same number of rows as the original dataframe, but with the specified columns replaced by the dimension table IDs.

    Params:
        fct_dataframe (psDataFrame):
            The DataFrame to replace the columns in.
        dim_dataframe (psDataFrame):
            The dimension table containing the IDs.
        cols_to_replace (Union[str, str_collection]):
            The column(s) to replace with the dimension table IDs.
        dim_id_col (str, optional):
            The name of the column in the dimension table containing the IDs.<br>
            If `#!py None`, then will use the first column of the dimension table.<br>
            Defaults to `#!py None`.

    Raises:
        TypeError:
            If any of the inputs parsed to the parameters of this function are not the correct type. Uses the [`@typeguard.typechecked`](https://typeguard.readthedocs.io/en/stable/api.html#typeguard.typechecked) decorator.
        ColumnDoesNotExistError:
            If any of the columns specified do not exist in the dataframes.

    Returns:
        (psDataFrame):
            The DataFrame with the columns replaced by the dimension table IDs.

    ???+ example "Examples"

        ```{.py .python linenums="1" title="Set up"}
        >>> # Imports
        >>> import pandas as pd
        >>> from pyspark.sql import SparkSession
        >>>
        >>> # Instantiate Spark
        >>> from toolbox_pyspark.dimensions import make_dimension_table, replace_columns_with_dimension_id
        >>> spark = SparkSession.builder.getOrCreate()
        >>>
        >>> # Create data
        >>> df = spark.createDataFrame(
        ...     pd.DataFrame(
        ...         {
        ...             "a": [1, 2, 3, 4],
        ...             "b": ["a", "b", "c", "d"],
        ...             "c": [1, 1, 2, 2],
        ...             "d": ["a", "b", "b", "b"],
        ...             "e": ["x", "x", "y", "z"],
        ...         }
        ...     )
        ... )
        >>> dim_table1 = make_dimension_table(df, "d")
        >>> dim_table2 = make_dimension_table(df, "e")
        >>> dim_table3 = make_dimension_table(df, ("c", "d"))
        >>>
        >>> # Check
        >>> df.show()
        >>> dim_table1.show()
        >>> dim_table2.show()
        >>> dim_table3.show()
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        +---+---+---+---+---+
        | a | b | c | d | e |
        +---+---+---+---+---+
        | 1 | a | 1 | a | x |
        | 2 | b | 1 | b | x |
        | 3 | c | 2 | b | y |
        | 4 | d | 2 | b | z |
        +---+---+---+---+---+
        ```
        ```{.txt .text title="Terminal"}
        +------+---+
        | id_d | d |
        +------+---+
        |    1 | a |
        |    2 | b |
        +------+---+
        ```
        ```{.txt .text title="Terminal"}
        +------+---+
        | id_e | e |
        +------+---+
        |    1 | x |
        |    2 | y |
        |    3 | z |
        +------+---+
        ```
        ```{.txt .text title="Terminal"}
        +----+---+---+
        | id | c | d |
        +----+---+---+
        |  1 | 1 | a |
        |  2 | 1 | b |
        |  3 | 2 | b |
        +----+---+---+
        ```
        </div>

        ```{.py .python linenums="1" title="Example 1: Replace single column with dimension ID"}
        >>> df_replaced = replace_columns_with_dimension_id(df, dim_table1, "d")
        >>> df_replaced.show()
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        +---+---+---+------+---+
        | a | b | c | id_d | e |
        +---+---+---+------+---+
        | 1 | a | 1 |    1 | x |
        | 2 | b | 1 |    2 | x |
        | 3 | c | 2 |    2 | y |
        | 4 | d | 2 |    2 | z |
        +---+---+---+------+---+
        ```
        !!! success "Conclusion: Successfully replaced single column with dimension ID."
        </div>

        ```{.py .python linenums="1" title="Example 2: Replace single column with dimension ID"}
        >>> df_replaced = replace_columns_with_dimension_id(df, dim_table2, "e")
        >>> df_replaced.show()
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        +---+---+---+---+------+
        | a | b | c | d | id_e |
        +---+---+---+---+------+
        | 1 | a | 1 | a |    1 |
        | 2 | b | 1 | b |    1 |
        | 3 | c | 2 | b |    2 |
        | 4 | d | 2 | b |    3 |
        +---+---+---+---+------+
        ```
        !!! success "Conclusion: Successfully replaced single column with dimension ID."
        </div>

        ```{.py .python linenums="1" title="Example 3: Replace multiple columns with dimension IDs"}
        >>> df_replaced_multi = replace_columns_with_dimension_id(df, dim_table3, ["c", "d"])
        >>> df_replaced_multi.show()
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        +---+---+----+---+
        | a | b | id | e |
        +---+---+----+---+
        | 1 | a |  1 | x |
        | 2 | b |  2 | x |
        | 3 | c |  3 | y |
        | 4 | d |  3 | z |
        +---+---+----+---+
        ```
        !!! success "Conclusion: Successfully replaced multiple columns with dimension IDs."
        </div>

        ```{.py .python linenums="1" title="Example 4: Invalid column type"}
        >>> df_replaced = replace_columns_with_dimension_id(df, dim_table, "123")
        ```
        <div class="result" markdown>
        ```{.txt .text title="Terminal"}
        ColumnDoesNotExistError: Column '123' does not exist in the DataFrame.
        ```
        !!! failure "Conclusion: Failed to replace columns due to invalid column type."
        </div>

    ??? tip "See Also"
        - [`make_dimension_table`][toolbox_pyspark.dimensions.make_dimension_table]
    """

    # Generate variables ----
    cols_to_replace: str_list = (
        [cols_to_replace] if is_type(cols_to_replace, str) else list(cols_to_replace)
    )
    fct_cols: str_list = fct_dataframe.columns
    dim_cols: str_list = dim_dataframe.columns
    dim_id_col = dim_id_col or dim_cols[0]

    # Check variables ----
    assert_columns_exists(fct_dataframe, cols_to_replace)
    assert_columns_exists(dim_dataframe, cols_to_replace)
    assert_column_exists(dim_dataframe, dim_id_col)

    # Perform the replacement ----
    index_of_first_col: int = fct_cols.index(cols_to_replace[0])
    fct_new_cols: str_list = deepcopy(fct_cols)
    fct_new_cols = [
        *fct_new_cols[:index_of_first_col],
        dim_id_col,
        *fct_new_cols[index_of_first_col + 1 :],
    ]
    fct_removed_cols: str_list = [col for col in fct_new_cols if col not in cols_to_replace]

    # Return ----
    return (
        fct_dataframe.alias("a")
        .join(
            other=dim_dataframe.alias("b"),
            on=[F.col(f"a.{col}") == F.col(f"b.{col}") for col in cols_to_replace],
            how="left",
        )
        .select("a.*", f"b.{dim_id_col}")
        .select(*fct_removed_cols)
    )