The Property Extraction configuration enables automatic extraction of metadata from interface raw data and maps it to structured properties for reporting and analysis. This feature allows you to capture technical details like web proxies, communication users and identifiers directly from interface configurations.
Purpose: Configure automated extraction rules that parse interface raw data and populate property fields, enabling consistent metadata capture across integration platforms without manual enrichment workflows.
Navigate to Settings → Properties → Extraction to manage your extraction rules using regex patterns and field mapping techniques.
For detailed guidance on property configuration, see 2.1.1 Properties in the Onboarding Guide.
Property Extraction Overview
Property Extraction interface with the two extraction method tabs, the context selector and the rule table
How Extraction Works
Extraction Methods
ID from Description/Name - finds an identifier inside a text field using two regular expressions. Use it when your teams write IDs into interface names or descriptions, for example an interface whose description reads “Order export - Interface ID: 4711-1234”.
Copy from Field - copies a value directly out of the interface raw data, without regular expressions. Use it for technical settings such as proxy hosts, communication users or endpoint addresses.
Interfaces or Systems
The context selector at the top right of the rule table decides where the value is written:
Interfaces - the value becomes a property of the individual interface. Use this when you report per interface, for example “which technical user does this interface use”.
Systems - the value becomes a property of the normalized system the interface belongs to. Use this to harmonize systems, for example when several IDoc interfaces should all point to the same logical system.
The same source field syntax works in both contexts - only the target changes. An IDoc partner number, for example, can either be stored on the individual interface or identify the system behind it.
The Systems context writes the extracted value to the normalized system instead of the interface
Which interface types are available
Not every platform delivers system-level data, so the selectable interface types depend on the tab and the context:
Three interface types need additional explanation:
RESOURCE covers SAP IS CI User Credentials and SAP BTP destinations
PLANNED covers interfaces from your planning tool (Jira, Azure DevOps, SAP Integration Assessment)
EAM covers systems from your EAM tool (e.g. LeanIX, Luy, Bee360).
Rule Management
Adding Extraction Rules
Pick the tab and the context first, then click + and fill the row. What goes into the middle columns differs per tab - see ID from Description/Name and Copy from Field below.
The Target Property dropdown shows the properties that match the selected tab and context
The dropdown shows Interface properties in the Interfaces context, System properties in the Systems context, and only ID-format properties on the ID from Description/Name tab.
Removing Extraction Rules
Select checkbox next to rules you want to remove
Click "-" button to delete selected rules
Click Save to confirm changes
Common Use Cases
One property is usually filled by several rules, because the same information sits in a different place on every platform.
Typical questions and the source fields that answer them
Which technical user does each interface run under? (e.g. into a Communication User property)
Run the job (from Settings → Properties → Overview or in your designated Data Source)
SAP_SOAP is read from data already stored in Integration Cockpit, so the INVENTORY job applies this rule (see Apply and Validate). Start it from your specific Data Source or click the Update Inventory button found in the Properties Overview.
Update Inventory button
Check the result
Go to Inventory → Interfaces → by Property, select Communication User and click Next. Every SAP_SOAP interface whose service binding uses basic authentication now shows its technical user.
Extend to the other platforms
The same property can be filled from every platform - add one rule per interface type, all pointing at Communication User:
Interface Type
Source Field
SAP_RFC
RFCDest:EXTERNAL:USERNAME
SAP_IDOC
RFCDest:EXTERNAL:USERNAME
RESOURCE
USER:Username
RESOURCE
BTPDestination:AuthenticationUser
After the next INVENTORY run you have one property that answers "which technical user does this interface run under" across your whole landscape - and you can filter or export by it.
Note
The two RESOURCE rules read different artifacts (Cloud Integration credentials and BTP destinations), so they never compete for the same interface. Do not add two rules for the same interface type and the same property - see the note under “What the extracted values look like”.
ID from Description/Name
ID rules using labels like ExternalID, InterfaceID and Jira # with matching extract patterns
What you see:
Context selector (Interfaces, Systems) to determine where the extracted value is stored
Interface Type - the platform the rule applies to
Source Field - which text field is searched (Name, Description, Title). The options depend on the Interface Type. Leave empty to search all available fields.
Search Pattern (Regex) - locates the label in the text
Extract Pattern (Regex) - reads the value that follows the label
Target Property - the property that receives the value
The available source fields depend on the selected interface type
How to use:
Click "+" to add a new rule
Select the Interface Type
Optionally narrow the Source Field to specific fields
Enter the Search Pattern - the label that precedes your ID
Enter the Extract Pattern - the shape of the ID itself
Select the Target Property
Click Save
The Search Pattern locates a label in the text. The Extract Pattern is then applied only to the text that follows that label, so keep the Search Pattern short and literal.
Leaving Source Field empty
An empty Source Field means all available fields are searched.
Target Property format
Only properties with the ID format are available in this tab. If a property is missing from the dropdown, check its format under Settings → Properties → Overview and make sure it is set to ID.
Common Extraction Patterns
Interface Identifiers
```
# Extract 8-digit numeric IDs (like Global Interface IDs)
# Search Pattern: GLIF
\d{8}
# Example: "SAP_ORDER_GLIF_12345678_PROCESS" → extracts "12345678"
# Extract a 6-digit ID in brackets - the digit count is up to you
# Search Pattern: \(
\d{6}
# Example: "Active Directory (260051)" → extracts "260051"
# Extract IDs that are separated by a space, colon or hyphen
# Search Pattern: GLIF
\b\d{8}\b
# Example: "Order export GLIF 12345678" → extracts "12345678"
# \b does not apply next to underscores - use the pattern above for SAP_GLIF_12345678
# Extract only part of a match using a capture group
# Search Pattern: ExternalID=
[A-Z]{2,4}-(\d+)
# Example: "ExternalID=SYS-4711" → extracts "4711", not "SYS-4711"
# Extract a fixed-length ID that follows a label
# Search Pattern: InterfaceID:
\b.{9}\b
# Example: "...InterfaceID: 4711-1234" → extracts "4711-1234"
# A different length yields no value
# Extract a planning key that follows a label
# Search Pattern: Interface ID:
[A-Z]{2,4}-\d+
# Example: "Order export - Interface ID: WIB-6" → extracts "WIB-6"
# Extract a ticket key that is not always labelled
# Search Pattern: ^
[A-Z]{2,5}-\d+
# Example: "Jira #WIC-1456", "WIC-1257" and "for WIC-638" all yield the key
# Extract alphanumeric codes with prefix
# Search Pattern: ^
[A-Z]{2,4}-\d{4,8}
# Example: "Interface CRM-20240115 export" → extracts "CRM-20240115"
# Extract an ID at the end of the name
# Search Pattern: ^
\d+$
# Example: "TRUSTING@ALM_ALM_0090110740" → extracts "0090110740"
# Extract GUID patterns
# Search Pattern: ^
[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}
# Example: "Interface_a1b2c3d4-e5f6-47a8-9b0c-d1e2f3a4b5c6" → extracts the GUID
```
# Find the label that precedes the ID
GLIF
# Example: Matches "HR_GLIF_12345678", extraction continues after "GLIF"
# Find a label regardless of upper or lower case
(?i)GLIF
# Example: Matches "HR_GLIF_12345678" and "hr_glif_12345678"
# Find a marker used in descriptions
##
# Example: Matches "This is demo interface ##47111234"
# Find a labelled ticket or interface ID
Jira #
# Example: Matches "Order import (Jira #WIC-1456)"
# Find a label written with an equals sign
Cockpit ID=
# Example: Matches "Cockpit ID=IF-0815: order export"
# Search the whole text, without a label
^
# Example: Matches every interface, the extract pattern then searches the entire text
# Leaving the Search Pattern empty extracts nothing - always use ^ instead
```
Copy from Field
Copy from Field rules reading proxy hosts, communication users and planning metadata
What you see:
Context selector (Interfaces, Systems) to determine where the extracted value is stored
Interface Type - the platform the rule applies to
Source Field - the field path in the raw data, entered as text
Target Property - the property that receives the value. All property formats are available on this tab.
How to use:
Click "+" to add a new rule
Select the Interface Type
Enter the Source Field using the syntax for that platform (see below)
RFCDest:EXTERNAL:WORKINGDIRECTORY → Extract the working directory
💡
Destination Type Rules
Each rule applies to one destination type. Add one rule per destination type you use. An interface has only one destination, so these rules never compete and can target the same property.
IDoc partner profiles (SAP_IDOC only)
Syntax: EDIPHONE:FieldName
Partner Identification Examples:
EDIPHONE:PartnerNumber → Extract the IDoc partner number, e.g. 0000004711
EDIPHONE:PartnerType → Extract the partner type, e.g. LI or LS
Contact Detail Examples:
EDIPHONE:NAME → Extract the contact name stored in the partner profile
EDIPHONE:TEL_NUMBER → Extract the contact telephone number
EDIPHONE:COMPANY → Extract the company / organization
EDIPHONE refers to the EDI phone/fax details of the IDoc partner profile in SAP (WE20). PartnerNumber and PartnerType identify the partner, typically the logical system, while NAME, TEL_NUMBER and COMPANY contain the associated contact details. Use PartnerNumber for an ID property. Field names are case-sensitive and must be written exactly as shown above.
In the Interfaces context, all five fields are available and both inbound and outbound partner profiles are evaluated. In the Systems context, only NAME, TEL_NUMBER and COMPANY are available. PartnerNumber and PartnerType identify the partner from which the system is created and are not available as system properties. Use one rule per field, as a second rule for the same field replaces the first.
Requirement: Rules in the Systems context are only applied when Create System from ALE Partner is enabled on the ABAP data source under Settings → Connectivity → Data Sources. The checkbox appears once IDocs is enabled. Without it, the rules are ignored entirely.
SAP_IS_CI Interfaces
For SAP Cloud Integration interfaces, values can be read from externalized iFlow parameters maintained in the Configure view:
Syntax: Configuration:ParameterName
Connection Configuration Examples:
Configuration:LocationID → Extract the Cloud Connector location ID
Configuration:Address → Extract the endpoint address
Configuration:ProxyType → Extract the proxy type
Technical Configuration Examples:
Configuration:Timeout → Extract the timeout configuration
Configuration:Credential Name → Extract the referenced credential
Parameter names depend on the iFlow and can be found in the Configuration table of the integration flow. They are case-insensitive, but spaces, hyphens, and underscores must match exactly.
For example a rule for Configuration:LocationID will not match a parameter named location-id or Location_ID.
Values can also be read from custom tags defined on the Integration Package:
Package:Tag:SystemID → Extract the value of the SystemID custom package tag and map it to a system property (e.g. System ID Harmonized)
This is useful for consistent system assignment across all interfaces in a package. The tag must exist on the package, and the tag name is case-sensitive.
Interface properties can be directly filled from Integration Package custom tags by enabling the flag on the designated Data Source “Copy Package Tag to Property”.
The name of the property must match the name of the custom tag.
PLANNED Interfaces
For planned interfaces, values are read from the connected planning tool. Available fields depend on the source tool.
Jira: Key, Title, Status, Due Date, Description, Assigned To, Label(s), Parent, Type, ID, Created, Changed, Confluence
Azure DevOps: Parent (other work item fields are currently not available)
SAP Integration Assessment: Interface Request, Interface Request Status, Interface Request Style, Business Solution Request, Domain, Selected Technology 1, Selected Technologies Coverage 1 (continues with 2, 3, etc. for additional technologies)
Syntax: Metadata:FieldName
Examples:
Metadata:Key → Extract the planning ID (e.g. into Global Interface ID)
Metadata:Due Date → Extract the planned date (e.g. into Go-Live Date)
Metadata:Status → Extract the project status (e.g. into Project Status)
Metadata:Domain → Extract the business domain (SAP Integration Assessment)
RESOURCE Interfaces
For resources, values can be read from the Security Material of the SAP Cloud Integration tenant:
Syntax: USER:FieldName
USER:Username → Extract the user name stored in the credential
USER:Name → Extract the name of the credential artifact
USER:Description → Extract the description
USER:DeployedBy → Extract who deployed the credential
Values can also be read from SAP BTP destinations. This requires a BTP data source with the Destination product enabled:
Syntax: BTPDestination:FieldName
Authentication Examples:
BTPDestination:AuthenticationUser → Extract the configured user
BTPDestination:AuthenticationType → Extract the authentication method (e.g. BasicAuthentication, CONFIGURED_USER, NoAuthentication)
Connection Configuration Examples:
BTPDestination:URLHTTP → Extract the destination URL
BTPDestination:HostRfc → Extract the RFC host
You can also write an authentication method instead of a field name to read the user only for destinations using it, for example BTPDestination:CONFIGURED_USER.
EAM Systems
In the Systems context, values can be read from the metadata delivered by your EAM tool (LeanIX, Luy, Bee360).
Syntax: Metadata:FieldName
Metadata:Name → Extract the system name (e.g. into System ID Harmonized)
Metadata:App owner → Extract the application owner (e.g. into System Owner)
Metadata:ID → Extract the external system ID (e.g. into Fact Sheet ID (LeanIX))
Metadata:Type → Extract the system type (e.g. into Fact Sheet Type (LeanIX))
Field names come from your EAM tool, so the available names depend on which fields you expose there.
Apply and Validate
Implementation Workflow:
Create Properties first in Settings → Properties → Overview
Configure extraction rules using the patterns and field paths described above
Run the job that applies your rules - INVENTORY (for Interface properties) or CONFIGURATION (for System properties)
Verify extraction in Inventory → Interfaces/Systems → by Property
Use extracted properties for reporting and filtering workflows
What happens on the next run
Extraction overwrites the value it manages, but it never removes values:
Value changed in the source system - the property is updated with the new value on the next run
Rule changed - the new rule takes effect on the next run and overwrites the value it wrote before
Rule deleted, or the value no longer exists in the source system - the property keeps the value it last had. Nothing cleans it up automatically
So if you remove a rule or reorganize your properties, delete the values you no longer want on the interface detail page, or through mass maintenance in Inventory → Interfaces. Otherwise old values stay in your reports.
If a rule stays empty
Rules and patterns are not validated when you save them, so a mistake shows up as an empty property rather than an error message. Check in this order:
Did you run the right job? See the table above - several syntaxes are only applied by the CONFIGURATION job, never by INVENTORY
Is the required data source option enabled? See Requirements and Setup
Does the source field exist and is it spelled exactly right? Look it up in SOAMANAGER, the iFlow Configure view or transaction WE20. RFCDest, the RFCType and package tag names are case sensitive, and a rule only reads the destination type it names
Is the Target Property a value list? Only values contained in that list are stored
Best Practices
Plan your properties before your rules. Use one property per piece of information and fill it from every platform, rather than one property per platform. That way a single filter or export column answers the question for your whole landscape
Start small. Create one rule for one interface type, run the job and check the result before rolling out broadly
Name the source field explicitly where an ID could appear in more than one field, instead of relying on the search order
Clean up when you change your mind. Removing a rule leaves its values behind - delete them through mass maintenance in Inventory → Interfaces → Mass Maintenance
Prefer fields the source cannot leave empty. An authentication method or a technical ID is maintained on every object, so a rule reading it covers your whole landscape. Optional settings that teams fill in only sometimes give you a sample, not a report - check how many objects actually carry a value before you build a property around it
Related Documentation
For comprehensive property extraction and configuration guidance:
Properties - Complete property reference and usage examples
Property Propagation - Carrying a property from one interface type to another, for example from a planned interface to the technical interface that implements it