Utility Functions

Built-in functions for string processing, math, date operations, regex, and data types.

Source: module/script/helper/script_processor.py:get_allowed_functions_list()

---

String Functions

Type Conversion

str(value)       # Convert to string
int(value)       # Convert to integer
float(value)     # Convert to float
str_to_bool(s)   # "true"/"1"/"yes" → True, everything else → False

String Methods

lower(s)              # str.lower — "ABC" → "abc"
upper(s)              # str.upper — "abc" → "ABC"
split(s, sep)         # str.split — "a,b,c".split(",") → ["a","b","c"]
strip(s)              # str.strip — " abc " → "abc"
startswith(s, prefix) # str.startswith
endswith(s, suffix)   # str.endswith

These are available as standalone functions but can also be called as methods on strings:

# Both work:
result = upper("hello")      # → "HELLO"
result = "hello".upper()     # → "HELLO"

---

Fuzzy String Matching

levenshtein_distance()

Calculates the edit distance between two strings (number of changes needed).

levenshtein_distance(s1, s2)

Example:

dist = levenshtein_distance("ACME Corp", "ACME Corporation")
# dist = 7
if dist < 5:
    # Strings are similar enough
    pass

jaro_winkler_similarity()

Calculates a similarity score between 0.0 and 1.0.

jaro_winkler_similarity(s1, s2)

Example:

score = jaro_winkler_similarity("Invoice", "Invocie")
# score ≈ 0.96 (very similar, typo)

score = jaro_winkler_similarity("Invoice", "Receipt")
# score ≈ 0.45 (dissimilar)

When to use which?

• Levenshtein: Good for short strings, exact error count

• Jaro-Winkler: Better for names/addresses, weights matches at the beginning more heavily

---

Regex Functions

Based on Python's re module, but available as standalone functions.

re_search()

Searches for the first occurrence of a pattern.

re_search(pattern, string)

Returns: Match object or None

Example — Extract order number from fulltext:

content = get_document_content(document_data)
match = re_search(r"Order number:\s*(\d+)", content)
if match:
    po_number = match.group(1)
    set_field_value(document_data, "purchase_order", po_number)

re_sub()

Replaces pattern matches with a replacement string.

re_sub(pattern, replacement, string)

Example — Remove special characters from invoice ID:

inv_id = get_field_value(document_data, "invoice_id", "")
cleaned = re_sub(r"[^A-Za-z0-9\-]", "", inv_id)
set_field_value(document_data, "invoice_id", cleaned)

re_findall()

Finds all occurrences of a pattern.

re_findall(pattern, string)

Example — Find all PO numbers in document:

content = get_document_content(document_data)
po_numbers = re_findall(r"PO[- ]?\d{6,}", content)
if po_numbers:
    set_field_value(document_data, "purchase_order", po_numbers[0])

---

Date/Time Functions

datetime_today()

Returns today's date as a datetime object.

today = datetime_today()

datetime_date

The date class for date creation.

d = datetime_date(2026, 3, 25)  # date(2026, 3, 25)

strptime()

Parses a date string into a datetime object.

strptime(date_string, format)

Example — Parse and use invoice date:

inv_date_str = get_field_value(document_data, "invoice_date", "")
try:
    date_obj = strptime(inv_date_str, "%Y-%m-%d")
    year = date_obj.year
    month = date_obj.month
except ValueError:
    pass

strftime()

Formats a datetime object as a string.

strftime(datetime_obj, format)

Example — Set processing date:

today = datetime_today()
formatted = strftime(today, "%d.%m.%Y")  # "25.03.2026"
set_field_value(document_data, "processing_date", formatted)

fromisocalendar()

Creates a date from ISO calendar week.

fromisocalendar(year, week, day)

Example — Convert calendar week to date:

# CW 12/2026, Monday
d = fromisocalendar(2026, 12, 1)  # date(2026, 3, 16)
set_date_value(document_data, "delivery_date", str(d))

Common with German customers: Delivery dates are often specified as "KW 12". Pattern:

content = get_document_content(document_data)
match = re_search(r"KW\s*(\d{1,2})[/\s]*(\d{4})", content)
if match:
    week = int(match.group(1))
    year = int(match.group(2))
    d = fromisocalendar(year, week, 1)  # Monday of the CW
    set_date_value(document_data, "delivery_date", str(d))

calendar_monthrange()

Returns the weekday of the 1st and number of days in a month.

weekday_of_first, num_days = calendar_monthrange(year, month)

Example:

_, days_in_month = calendar_monthrange(2026, 2)
# days_in_month = 28

---

Decimal/Locale Functions

parse_decimal()

Parses a string to a decimal number with locale detection.

parse_decimal(value, locale="en_US")

Example:

amount = parse_decimal("1.234,56", "de_DE")  # → Decimal('1234.56')
amount = parse_decimal("1,234.56", "en_US")  # → Decimal('1234.56')

format_decimal_to_locale()

Formats a decimal number according to locale.

format_decimal_to_locale(value, locale)

Example:

formatted = format_decimal_to_locale(1234.56, "de_DE")  # → "1.234,56"
formatted = format_decimal_to_locale(1234.56, "en_US")  # → "1,234.56"

---

Math Functions

The full math module is available:

Function	Description
abs(x)	Absolute value
round(x, n)	Round to n decimal places
floor(x)	Floor (round down)
ceil(x)	Ceiling (round up)
sqrt(x)	Square root
pow(x, y)	Power
log(x) / log10(x)	Logarithm
pi	π (3.14159...)
e	Euler's number (2.71828...)
sin, cos, tan	Trigonometry
acos, asin, atan	Inverse trigonometry
degrees, radians	Degrees ↔ Radians
exp, fabs, fmod	Additional math functions
hypot, ldexp, frexp	Special calculations
modf	Separate integer/decimal parts

---

Data Structure Functions

dict()            # New dictionary
list()            # New list
set()             # New set
tuple()           # New tuple
defaultdict(type) # collections.defaultdict

len(x)            # Length
isinstance(x, t)  # Type check
type(x)           # Get type
deepcopy(x)       # Deep copy (copy.deepcopy)

print(x)          # Debug output (for testing only)

print() is only available for debugging purposes. Output goes to the server log, not to the user.