Integration Toolkit Solutions Wiki

3rd Party Modules
,
22

Storing IPv4 addresses as integers (rather than strings)

String IP addresses displayed in tables don't sort naturally. e.g., x.x.x.1, x.x.x.10, x.x.x.11, x.x.x.2.
Storing the IP addresses in the database as a 32-bit integer (which is what they are supposed to represent), and returning those to a table column but rendering them with a view to convert to four-decimal octet format solves the sorting problem.

From IP address to integer

With the built-in expression language the job can be done in two stages (to simplify the expression):

Initial expression

split({path.to.IP.as.string}, '\\.')

Expression transform 
// Doing this in two steps allows the use of {value} rather than multiple splits.
  ({value}[0, 0] << 24)
+ ({value}[1, 0] << 16)
+ ({value}[2, 0] << 8)
+ ({value}[3, 0] << 0)

With the Integration Toolkit it can be done in one go.

Integration Toolkit solution 
transform(
	split({path.to.IP.as.string}, '\\.'),
	(value()[0, 0] << 24) +
	(value()[1, 0] << 16) +
	(value()[2, 0] << 8) +
	(value()[3, 0] << 0)
)

These expressions assume that the incoming IP ≥ 1.x.x.x.

From integer to IP address

With the built-in expression language the job can be done in two stages (to simplify the expression):

Initial expression

toInt({../lblIp2IntExpression.props.text})

Expression transform 
// Doing this in two steps allows the use of {value} rather than multiple splits.
stringFormat(
	"%d.%d.%d.%d",
	({value} >> 24) & 0xff,
	({value} >> 16) & 0xff,
	({value} >> 8) & 0xff,
	{value} & 0xff
)

With the Integration Toolkit it can be done in one go.

Integration Toolkit solution 
// This was my original attempt.
// See Phil's improved solution below.
transform(
	right('0' + toHex({path.to.IP.as.integer}), 8),
	fromHex(subString(value(), 0, 2)) + '.' +
	fromHex(subString(value(), 2, 4)) + '.' +
	fromHex(subString(value(), 4, 6)) + '.' +
	fromHex(subString(value(), 6, 8))
)

These expressions assume that the incoming integer represents an IP ≥ 1.x.x.x.

Edit by Phil: I would do the conversion to display without using hex strings.

Better Integration Toolkit Solution 
transform(
	toInt({path.to.IP.as.integer}),
	stringFormat(
		"%d.%d.%d.%d",
		(value() >> 24) & 0xff,
		(value() >> 16) & 0xff,
		(value() >> 8) & 0xff,
		value() & 0xff
	)
)

Does not require the leading octet to be ≥1.

IPv6 is not addressed by these functions.

1 Like
23

Iterating through an array-tag to determine if value is present

Casus: A camera is selected on a video-wall if it’s ID (UDT-parameter) is present in a tag of the type Integer array (external system). So I want to use an expression-tag to determine if the camera is selected (boolean).

Approach: Iterate through the array with forEach (or where?) and compare each value to the integer ID-value. If found -> true, else false

Initial expression

forEach returns a list if expression is met?
Then check if this list has one or more entries.

if(
	len(
		forEach( 
			{[~]MSBS/SLKRS1_I1/CCTV/Pnl/STS/CameraNr.value}, 
			it() = {ArrayID}
		)
	) > 0,
	true,
	false
)

Status: Error in the tag (Error Config)? What do I do wrong?
Tips:
forEach() always returns a list. But per my previous reply, use where() instead of forEach(). Do not wrap an if() around the whole thing, as a comparison with len() yields a boolean directly.

Improved expression

Note: I've moved the source (array) tag to be in the same folder as the expression-tag in order to supply an export-file (attached) if anybody is interested in reproducing/solving the issue.

len(
	where( 
		{[.]CameraNr.value}, 
		it() = {ArrayID}
	)
) > 0

Previous improvements applied, still an Error_Configuration (Error writing to Geselecteerd.value: Bad_Unsupported)

ExpressionTagAndSource.json (784 Bytes)

24

Creating a list of Months in format YYYY-MM between a start and end date

Note: forEach(10, it()) is equivalent to range(10) in Python.

Expression Structure Binding:

start: 2025-06-01 0:00:00
end: 2025-10-01 00:00:00

forEach(
	forEach(monthsBetween({value}['start'], {value}['end'])+1, it()),
	dateFormat(addMonths({value}['start'], it()), 'YYYY-MM')
)

Returns:

[
  "2025-06",
  "2025-07",
  "2025-08",
  "2025-09",
   "2025-10"
]

One of the forEach() loops is unneeded. forEach() returns a list, so your outer loop is actually forEach(asList(0,1,2,3,4),it())

forEach(
    monthsBetween({value}['start'],{value['end']) + 1,
    dateFormat(addMonths({value}['start'],it()),'YYYY-MM')
)

Will yeild the same result.

Features employed
  • Iterables General behavior of all iterators.
  • forEach() Loops through the dataset, calling the nested expression with one row at a time.
  • it() Delivers the one row in dataset format (same column names and types as the source).
  • monthsBetween() Calculates the number of whole months between two dates.
  • addMonths() Add or subtract an amount of months to a given date and time.
  • dateFormat() Returns the given date as a string, formatted according to a pattern.

Pssst! Use yyyy-MM to avoid year-end confusion over week-years.

4 Likes