Staged (Multi-Stage)

Multiple stages with expressions:

transforms:
  - name: 'trim_space(value)'
    email: 'lower(value)'

  - full_name: 'record.first_name + " " + record.last_name'

  - tier: 'record.total_spent >= 1000 ? "premium" : "standard"'

String Functions

Function	Description	Example
`upper(s)`	Uppercase	`upper("hello")` -> "HELLO"
`lower(s)`	Lowercase	`lower("HELLO")` -> "hello"
`trim(s)`	Remove whitespace	`trim(" hi ")` -> "hi"
`replace(s, old, new)`	Replace text	`replace("a-b", "-", "_")` -> "a_b"
`substring(s, start, end)`	Extract substring	`substring("hello", 0, 2)` -> "he"
`split(s, sep)`	Split to array	`split("a,b", ",")` -> ["a","b"]
`split_part(s, sep, i)`	Get split part	`split_part("a,b,c", ",", 1)` -> "b"
`join(arr, sep)`	Join array	`join(["a","b"], "-")` -> "a-b"
`length(s)`	String length	`length("hello")` -> 5
`contains(s, sub)`	Check contains	`contains("hello", "ell")` -> true
`snake(s)`	Snake case	`snake("Hello World")` -> "hello_world"
`slugify(s)`	URL slug	`slugify("Hello World!")` -> "hello-world"
`remove_diacritics(s)`	Remove accents	`remove_diacritics("cafe")` -> "cafe"
`replace_non_printable(s)`	Clean string	Removes control chars
`replace_0x00(s)`	Remove null bytes	Cleans null chars

Numeric Functions

Function	Description	Example
`int_parse(v)`	Parse integer	`int_parse("42")` -> 42
`float_parse(v)`	Parse float	`float_parse("3.14")` -> 3.14
`bool_parse(v)`	Parse boolean	`bool_parse("true")` -> true
`int_format(n, fmt)`	Format integer	`int_format(42, "%05d")` -> "00042"
`float_format(n, fmt)`	Format float	`float_format(3.14, "%.1f")` -> "3.1"
`greatest(...)`	Maximum value	`greatest(1, 5, 3)` -> 5
`least(...)`	Minimum value	`least(1, 5, 3)` -> 1
`is_greater(a, b)`	Compare	`is_greater(5, 3)` -> true
`is_less(a, b)`	Compare	`is_less(3, 5)` -> true
`int_range(start, end)`	Generate range	`int_range(1, 3)` -> [1,2,3]

Date Functions

Function	Description	Example
`now()`	Current time	Returns datetime
`date_parse(s, fmt)`	Parse date	`date_parse("2024-01-15", "auto")`
`date_format(d, fmt)`	Format date	`date_format(now(), "%Y-%m-%d")`
`date_add(d, n, unit)`	Add duration	`date_add(now(), -7, "day")`
`date_diff(d1, d2, unit)`	Difference	`date_diff(d1, d2, "hour")`
`date_trunc(d, unit)`	Truncate	`date_trunc(now(), "month")`
`date_extract(d, part)`	Extract part	`date_extract(now(), "year")`
`date_first(d, period)`	First of period	`date_first(now(), "month")`
`date_last(d, period)`	Last of period	`date_last(now(), "month")`
`date_range(s, e, step)`	Date range	`date_range("2024-01-01", "2024-01-03", "1d")`
`date_timezone(d, tz)`	Convert TZ	`date_timezone(now(), "America/New_York")`

Date units: year, month, week, day, hour, minute, second

Format syntax: Uses strftime (e.g., %Y-%m-%d %H:%M:%S)

Value Functions

Function	Description	Example
`coalesce(...)`	First non-null	`coalesce(null, "default")`
`first_valid(...)`	First non-empty	`first_valid("", "value")`
`is_null(v)`	Check null	`is_null(value)`
`is_empty(v)`	Check empty	`is_empty("")` -> true
`cast(v, type)`	Convert type	`cast("42", "int")`
`try_cast(v, type)`	Safe convert	`try_cast("abc", "int")` -> null
`if(cond, t, f)`	Conditional	`if(x > 0, "pos", "neg")`
`equals(a, b)`	Deep equality	`equals(obj1, obj2)`
`require(v, msg)`	Assert non-null	`require(v, "Required!")`

Collection Functions

Function	Description	Example
`array(...)`	Create array	`array(1, 2, 3)`
`object(k, v, ...)`	Create object	`object("a", 1, "b", 2)`
`keys(map)`	Get keys	`keys({"a": 1})` -> ["a"]
`values(map)`	Get values	`values({"a": 1})` -> [1]
`exists(col, item)`	Check exists	`exists([1,2], 2)` -> true
`element(arr, i)`	Get element	`element([1,2,3], 1)` -> 2
`filter(arr, expr)`	Filter array	`filter([1,2,3], "element > 1")`
`sort(arr, desc?)`	Sort array	`sort([3,1,2])` -> [1,2,3]
`pluck(arr, key)`	Extract field	`pluck([{a:1},{a:2}], "a")` -> [1,2]
`jmespath(obj, expr)`	JMESPath query	`jmespath(data, "items[].name")`
`jq(obj, expr)`	jq filter (returns array)	`jq(data, ".items[].name")`
`get_path(obj, path)`	Get by path	`get_path(obj, "a.b[0]")`
`object_merge(...)`	Merge objects	`object_merge({a:1}, {b:2})`
`chunk(arr, size)`	Split chunks	`chunk([1,2,3,4], 2)`

Encoding Functions

Function	Description	Example
`encode_url(s)`	URL encode	`encode_url("a b")` -> "a%20b"
`decode_url(s)`	URL decode	`decode_url("a%20b")` -> "a b"
`encode_base64(s)`	Base64 encode	`encode_base64("hi")`
`decode_base64(s)`	Base64 decode	`decode_base64("aGk=")`
`hash(s, algo)`	Hash string	`hash("hello", "sha256")`
`json_parse(s)`	Parse JSON	`json_parse('{"a":1}')`

Utility Functions

Function	Description	Example
`uuid()`	Generate UUID	Returns UUID v4
`log(msg)`	Log message	Prints and returns msg
`regex_match(s, pat)`	Test pattern	`regex_match("a1", "\\d")`
`regex_extract(s, pat, i)`	Extract match	`regex_extract("id=123", "id=(\\d+)", 1)`
`regex_replace(s, pat, r)`	Replace pattern	`regex_replace("a1b2", "\\d", "X")`
`type_of(v)`	Get type	`type_of(42)` -> "integer"

Transform Examples

Clean and Normalize

transforms:
  - name: 'trim_space(upper(value))'
    email: 'lower(trim_space(value))'
    phone: 'replace(value, "[^0-9]", "")'

Computed Fields

transforms:
  - full_name: 'record.first_name + " " + record.last_name'
    domain: 'split_part(record.email, "@", 1)'
    age: 'date_diff(now(), record.birth_date, "year")'

Conditional Logic

transforms:
  - status: 'record.active ? "active" : "inactive"'
    tier: |
      record.total >= 1000 ? "gold" : (
        record.total >= 500 ? "silver" : "bronze"
      )

Hash/Anonymize

transforms:
  - email_hash: 'hash(record.email, "sha256")'
    masked_ssn: '"XXX-XX-" + substring(record.ssn, -4)'

Full Documentation

See https://docs.slingdata.io/concepts/functions.md for complete reference.

sling-transforms

Data Transformation Functions

Transform Syntax

Array (Global)

Map (Column-Specific)