Compute Wrangles

Compute wrangles, with recipe examples, parameters, and behavior.

Case When

Assign values to a column based on conditional logic.

Parameters

Name	Description	Accepted Values	Default	Required
I/O
output	Name of the output column.	string	—	Yes
Options
cases	List of conditions and corresponding values.	array	—	Yes
default	Value to assign if no conditions are met. Default None.	string, number, integer, boolean, null	null	No
Conditions
if	Condition that determines whether the wrangle runs as a whole. Recipe variables may be referenced with ${variable}.	string	—	No
where	Filter rows before applying the wrangle using SQL-like criteria, such as column1 = 123 OR column2 = 'abc'.	string	—	No
where_params	Values used with where for parameterized criteria. Uses SQLite placeholder syntax such as ? or :name.	array, object	—	No

Examples

wrangles:
  - compute.case_when:
      output: Letter Grade
      cases:
        - condition: Grade > 89
          value: 'A'
        - condition: 90 > Grade > 79
          value: 'B'
        - condition: 80 > Grade > 69
          value: 'C'
        - condition: 70 > Grade
          value: 'F'

Student	Grade
Billy	62
Sarah	91
Timmy	88
Tammy	74

Letter Grade
F
A
B
C

Conditions can be combined so multiple criteria must be met before assigning a value.

wrangles:
  - compute.case_when:
      output: Letter Grade
      cases:
        - condition: (Grade > .89) & (Attendance == 'Good')
          value: 'A'
        - condition: (.90 > Grade > .79) & (Attendance == 'Good')
          value: 'B'
        - condition: (.80 > Grade > .69) & (Attendance == 'Good')
          value: 'C'
        - condition: (.70 > Grade) or (Attendance == 'Poor')
          value: 'F'

Student	Grade	Attendance
Billy	62	Poor
Sarah	91	Poor
Timmy	88	Good
Tammy	74	Good

Letter Grade
F
F
B
C

Access

Requirement	Value
AI-powered	No
Requires WrangleWorks account	No
Requires subscription	No
Requires external API key	No

Technical details

Field	Value
Catalog ID	3
Catalog key	compute.case_when
Recipe key	compute.case_when
Catalog status	active
Lifecycle status	active
Recipe Writer eligible	Yes
Namespace	compute
Documentation group	compute
Aliases	None
Runtime symbol	wrangles.recipe_wrangles.compute.case_when
Legacy UUID	9a9662e4-53d1-4932-8adf-bc3e7aa364ad

Sources

• WranglesPY compute.case_when implementation
• Archived compute.case_when Markdown

---

Score Search Results

Scores and filters search results based on progressive partial/exact matching. Can return dictionaries or a parallel list of formatted strings.

Parameters

Name	Description	Accepted Values	Default	Required
I/O
input	List of 3 to 5 columns -> [results, suppliers, part_codes, mpns (optional), descriptions (optional)].	array	—	Yes
output	Output column for the dictionaries. If a list of 2 is provided, outputs [dicts_column, pretty_strings_column].	string, array	—	Yes
Options
must_match_part_code	If true, filters out results that don't satisfy the allowed match types.	boolean	true	No
allow_mpn_exact	Treat exact MPN matches as valid part code matches.	boolean	true	No
allow_mpn_partial	Treat partial MPN matches as valid part code matches.	boolean	true	No
allow_other_exact	Treat exact other part code matches as valid part code matches.	boolean	true	No
allow_other_partial	Treat partial other part code matches as valid part code matches.	boolean	true	No
blacklist_keywords	Comma-separated list or array of keywords to filter out URLs containing them.	string	""	No
mpn_exact_score	Mpn Exact Score value accepted by the runtime.	number	8	No
mpn_partial_base	Mpn Partial Base value accepted by the runtime.	number	4	No
part_code_exact_score	Part Code Exact Score value accepted by the runtime.	number	6	No
part_code_partial_base	Part Code Partial Base value accepted by the runtime.	number	2	No
supplier_exact_score	Supplier Exact Score value accepted by the runtime.	number	3	No
supplier_partial_base	Supplier Partial Base value accepted by the runtime.	number	1	No
context_match_base	Context Match Base value accepted by the runtime.	number	2	No
fuzzy_match_threshold	Fuzzy Match Threshold value accepted by the runtime.	number	0.8	No
Conditions
if	Condition that determines whether the wrangle runs as a whole. Recipe variables may be referenced with ${variable}.	string	—	No
where	Filter rows before applying the wrangle using SQL-like criteria, such as column1 = 123 OR column2 = 'abc'.	string	—	No
where_params	Values used with where for parameterized criteria. Uses SQLite placeholder syntax such as ? or :name.	array, object	—	No

Examples

This template scores search results using supplier and part-code context. The fields added to each result dictionary depend on the scoring configuration.

wrangles:
  - compute.score_search_results:
      input:
        - Search Results
        - Suppliers
        - Part Codes
      output:
        - Scored Results
      allow_other_exact: true
      must_match_part_code: true

Search Results	Suppliers	Part Codes
[{"title": "SKF 6202 bearing", "link": "https://example.com/6202"}]	["SKF"]	["6202"]

Scored Results
Ranked result dictionaries

Template based on the documented parameters; no published source example is currently available.

Access

Requirement	Value
AI-powered	No
Requires WrangleWorks account	No
Requires subscription	No
Requires external API key	No

Technical details

Field	Value
Catalog ID	4
Catalog key	compute.score_search_results
Recipe key	compute.score_search_results
Catalog status	active
Lifecycle status	active
Recipe Writer eligible	Yes
Namespace	compute
Documentation group	compute
Aliases	None
Runtime symbol	wrangles.recipe_wrangles.compute.score_search_results
Legacy UUID	313a8ec0-cf13-4956-8d3b-5362b8641d0f

Sources

• WranglesPY compute.score_search_results implementation
• Archived compute.score_search_results Markdown

---

Math

Apply a mathematical calculation.

Apply mathematical calculations to columns. Also called as maths.

info

Spaces within column headers are replaced with underscores automatically. Account for this when writing expressions in input.

Parameters

Name	Description	Accepted Values	Default	Required
I/O
input	The mathematical expression using column names. e.g. column1 * column2 + column3. Note: spaces within column names are replaced by underscores (_).	string	—	Yes
output	The column to output the results to.	string	—	Yes
Conditions
if	Condition that determines whether the wrangle runs as a whole. Recipe variables may be referenced with ${variable}.	string	—	No
where	Filter rows before applying the wrangle using SQL-like criteria, such as column1 = 123 OR column2 = 'abc'.	string	—	No
where_params	Values used with where for parameterized criteria. Uses SQLite placeholder syntax such as ? or :name.	array, object	—	No

Examples

wrangles:
  - math:
      input: sqrt(Values)
      output: Square Root

Values
4
9
16

Square Root
2
3
4

Access

Requirement	Value
AI-powered	No
Requires WrangleWorks account	No
Requires subscription	No
Requires external API key	No

Technical details

Field	Value
Catalog ID	82
Catalog key	math
Recipe key	math
Catalog status	active
Lifecycle status	active
Recipe Writer eligible	Yes
Namespace	Root-level
Documentation group	compute
Aliases	None
Runtime symbol	wrangles.recipe_wrangles.main.math
Legacy UUID	494b11ad-00c1-4748-8b93-6bec982f4fec

Sources

• WranglesPY math implementation
• Archived math Markdown

---

Python

Apply a simple single-line python command. For more complex python use a custom function. Note, this evaluates the python command - be especially cautious including variables from untrusted sources within the command string. The python command will be evaluated once for each row and the result returned. Reference column values by using their name. Non-alphanumeric characters within column names are replaced by underscores (_) Additionally, all columns are available as a dict named kwargs. Additional parameters set for the wrangle will also be available to the command.

The Python wrangle executes simple Python commands inline within a recipe. Row values are referenced by column name and commands are evaluated once per row. Spaces within column names are replaced by underscores (_). All columns are also available as a dictionary named kwargs. For more complex Python, use custom functions.

caution

This wrangle evaluates the Python command. Be cautious when including variables from untrusted sources in the command string.

Parameters

Name	Description	Accepted Values	Default	Required
I/O
output	Name or list of output column(s). To output multiple columns, return a list of the corresponding length.	string, array	—	Yes
input	Name or list of input column(s) to filter the data available to the command. Useful in conjunction with kwargs to target a variable range of columns. If not specified, all columns will be available.	string, integer, array, null	null	No
Options
command	Python command. This must return a value. Note: any non-alphanumeric characters in variable names are replaced by underscores (_).	string	—	Yes
Conditions
if	Condition that determines whether the wrangle runs as a whole. Recipe variables may be referenced with ${variable}.	string	—	No
where	Filter rows before applying the wrangle using SQL-like criteria, such as column1 = 123 OR column2 = 'abc'.	string	—	No
where_params	Values used with where for parameterized criteria. Uses SQLite placeholder syntax such as ? or :name.	array, object	—	No
Errors
except	Value to return for the row if an exception occurs during the evaluation. If not provided, an exception will be raised as normal. If multiple output columns are specified, this must match the length.	string, array, number, integer, boolean, object	—	No

Examples

wrangles:
  - python:
      output: result
      command: My_Column.upper()

My Column
example text

result
EXAMPLE TEXT

The Python wrangle supports parameters so values of unknown origin can be injected safely.

wrangles:
  - python:
      output: sliced
      command: input_column[:i]
      i: ${var}

No sample available.

No sample available.

Access

Requirement	Value
AI-powered	No
Requires WrangleWorks account	No
Requires subscription	No
Requires external API key	No

Technical details

Field	Value
Catalog ID	85
Catalog key	python
Recipe key	python
Catalog status	active
Lifecycle status	active
Recipe Writer eligible	Yes
Namespace	Root-level
Documentation group	compute
Aliases	None
Runtime symbol	wrangles.recipe_wrangles.main.python
Legacy UUID	c0398a11-7731-4e47-8df0-b07eea0b1d6c

Sources

• WranglesPY python implementation
• Archived python Markdown

---

SQL

Apply a SQL command to the current dataframe. Only SELECT statements are supported - the result will be the output.

Apply a SQL command to the current dataframe. Only SELECT statements are supported; the result becomes the output. The current table is called df.

info

SQL does not currently work with objects. If your table contains objects, use convert.to_json before using SQL. SQL is not compatible with where filtering.

Parameters

Name	Description	Accepted Values	Default	Required
Options
command	SQL Command. The table is called df. For specific SQL syntax, this uses the SQLite dialect.	string	—	Yes
params	Variables to use in conjunctions with query. This allows the query to be parameterized. This uses sqlite syntax (? or :name).	array, object, null	null	No
preserve_data_types	Preserve Data Types value accepted by the runtime.	boolean	true	No
Formatting
preserve_index	Preserve Index value accepted by the runtime.	boolean	false	No
Conditions
if	Condition that determines whether the wrangle runs as a whole. Recipe variables may be referenced with ${variable}.	string	—	No
where	Filter rows before applying the wrangle using SQL-like criteria, such as column1 = 123 OR column2 = 'abc'.	string	—	No
where_params	Values used with where for parameterized criteria. Uses SQLite placeholder syntax such as ? or :name.	array, object	—	No

Examples

wrangles:
  - sql:
      command: |
        SELECT header1, header2
        FROM df
        WHERE header1 >= 2

header1	header2	header3
1	a	x
2	b	y
3	c	z

header1	header2
2	b
3	c

Access

Requirement	Value
AI-powered	No
Requires WrangleWorks account	No
Requires subscription	No
Requires external API key	No

Technical details

Field	Value
Catalog ID	94
Catalog key	sql
Recipe key	sql
Catalog status	active
Lifecycle status	active
Recipe Writer eligible	Yes
Namespace	Root-level
Documentation group	compute
Aliases	None
Runtime symbol	wrangles.recipe_wrangles.main.sql
Legacy UUID	467a06b1-a697-4d31-8061-7d83a719fd79

Sources

• WranglesPY sql implementation
• Archived sql Markdown

---

Maths (Deprecated)

Deprecated

This compatibility wrangle remains available for existing recipes. Use math for new recipes.

Deprecated alias for math; evaluate an expression and write its result to an output column.

Behavior

Deprecated alias for math; evaluate an expression and write its result to an output column.

This guidance was derived from the callable signature.

Parameters

Name	Description	Accepted Values	Default	Required
I/O
input	Name, index, or list of input columns.	string	—	Yes
output	Name or list of output columns.	string	—	Yes
Conditions
if	Condition that determines whether the wrangle runs as a whole. Recipe variables may be referenced with ${variable}.	string	—	No
where	Filter rows before applying the wrangle using SQL-like criteria, such as column1 = 123 OR column2 = 'abc'.	string	—	No
where_params	Values used with where for parameterized criteria. Uses SQLite placeholder syntax such as ? or :name.	array, object	—	No

Access

Requirement	Value
AI-powered	No
Requires WrangleWorks account	No
Requires subscription	No
Requires external API key	No

Technical details

Field	Value
Catalog ID	83
Catalog key	maths
Recipe key	maths
Catalog status	active
Lifecycle status	deprecated
Recipe Writer eligible	No
Recipe Writer exclusion	Deprecated compatibility key; use math.
Replaced by	math
Namespace	Root-level
Documentation group	compute
Aliases	None
Runtime symbol	wrangles.recipe_wrangles.main.maths
Legacy UUID	None

Sources

• WranglesPY maths implementation