Introduction:
This blog describes the configuration steps for converting the FlatFile to XML with ‘n’ nested levels using custom Adapter Module ContentConversionBean ( Direction -> Sender ).
Pre-Requisites:
For novice users, Please read the link provided in help.sap.com related to File Content Conversion
Java Version:
This adapter module is written using JDK 1.5 API’s, so before deploying the Adapter Module (.ear file) make sure the JDK version 1.5 and above.
Problem:
Many times, we get requirements to convert the Flat File into XML with more than 3 nested levels. Using Standard Adapter/Adapter Module we will be able to achieve the above requirement with limitation up to 4 levels only. If the requirement is to generate the XML with more than 4 nested levels we can achieve this using Java Mapping. The problem here is that the java mapping is confined only to that particular interface, and hence it cannot be re-used for other scenarios.
Solution:
I have created a Generic Adapter Module to address the above problem for nested/deep hierarchies of generating XML.
This module uses the most of the standard FCC parameters and the use of the those parameters are listed in the Table2.
Table 1:
The following properties should be configured only in the module configuration of the channel.
| Property Name | Required/Optional | Description |
|---|---|---|
| Direction | Required | If the module used in sender channel provide the value as sender and if it is used in receiver channel specify receiver |
| Configuration_Mode | Optional | Configuration_Mode contains either file or local as the value.
|
| File_Path | Optional | 1) If Configuration_Mode = local, then this property will not be applicable. 2) If Configuration_Mode = file, then the properties which is specified in the Table 2 should be specified in the properties file and provide the absolute path of the properties file. Note: Properties file should be stored in the NFS file system of SAP PI. |
| Validate_XSD | Optional | This property is similar to the Validation By Adapter, but the functionality is achieved through Adapter Module.
Specify the absolute path of the XSD file which is stored in the NFS file system of SAP PI. Note : This property will be used only in case of sender channel. |
Table 2:
The following properties can be specified either in the file or in the Module Configuration of the channel depends on the value specified in the Configuration_Mode property specified in the Table1.
| Property Name | Required/Optional | Description |
|---|---|---|
| DocumentName | Required | The value of the property specifies the root element of the XML document. |
| DocumentNamespace | Optional | If values is specified for this property, namespace is added to the root element i.e.,DocumentName is namespace qualified. |
| DocumentOffset | Optional | The value of this property, specifies the number of lines needs to be skipped at the beginning of the document. This enables you to skip comment lines or column names during processing |
| RecordSetName.recordSetStructure | Optional | Specify the recordset names and the cardinality of substructures as follows:<NameA,nA,NameB,nB,...> , where nA=1,2,3,... or* (for a variable, unlimited number, including 0) Here NameA, NameB,… is the recordset Names nA specifies the cardinality. |
| RecordSetSequence | Optional | This property either contains Ascendingor Variable. If not configured, then Ascending will be considered as default. Ascending : The sequence of the recordset structures is assumed to be unique. A new recordset is started as soon as an earlier structure occurs. Variable : The sequence of the recordset structures is assumed not to be fixed. A new recordset is not started until another structure occurs that is defined with a fixed number. If all structures are defined as variable, the system interprets the entire document as a single recordset. |
| KeyFieldName | Optional | If you have entered a variable number of substructures under Recordset Structure, that is, you have entered the value * for at least one structure, specify a Key Field Name. The parser identifies the substructures by their content. This happens using the key field with different constants for the substructures. In this case, you must specify a key field, and the field name must occur in all substructures. |
| RecordSetName.fieldNames | Required | Specify the fieldNames separater by comma (,) which belongs to the corresponding recordSet structure |
| RecordSetName.beginSeparator | Optional | If you want to define an additional character string as a separator before the first column in a row, make a specification here. The system skips this separator when it processes this column (otherwise the system would treat it as part of the first column). |
| RecordSetName.endSeparator | Optional | If you want to define an additional string as a separator after the last column in a row, specify it here. The system skips this separator when it processes the last column (otherwise the system would treat it as part of the last column). Note:if fieldNames property isdefined for a recordSet, then endSeparator is required for that corresponding recordSet. |
| RecordSetName.fieldSeparator | Optional | If you make an entry here, the Module expects that the structure contains the specified character string (one or more characters) as a separator between the individual columns. If you have not made an entry for fieldFixedLengths , this is the only specification to identify the individual columns in a row. |
| RecordSetName.fieldFixedLengths | Optional | If you make an entry here, the Module expects a character string that contains the lengths of the structure columns as arguments separated by commas. If you also specify a separator for the columns, you must not add its length to the length of the columns. This entry is mandatory if you have not made an entry for RecordSetName.fieldSeparator |
| RecordSetName.enclosureSign | Optional | Specify a string that acts as a text delimiter. Text enclosed by such delimiters is transferred to the target structure unchanged, although the default setting is to remove all text delimiters. Separators within such texts are ignored. This parameter is optional. The default setting is an empty value (no text delimiter). Note : This property will be used only whenRecordSetName.fieldSeparator is used. |
| RecordSetName.enclosureSignEnd | Optional | If the text delimiters for the beginning and end of the text are different, specify the text delimiter for the end of the text here. If you do not make an entry here, the entry from NameA.enclosureSign is used. Note : This property will be used only whenRecordSetName.fieldSeparator is used. |
| RecordSetName.enclosureSignEscape | Optional | Specify a string that replaces the text delimiter if it occurs within a text that it delimits. When the text is transferred the string is replaced by the value specified in NameA.enclosureSign Note : This property will be used only whenRecordSetName.fieldSeparator is used. |
| RecordSetName.enclosureSignEndEscape | Optional | Specify a string that replaces the text delimiter for the end of the text if it occurs within a text that it delimits. When the text is transferred the string is replaced by the value specified in NameA.enclosureSignEnd . Note : This property will be used only whenRecordSetName.fieldSeparator is used. |
| RecordSetName.enclosureConversion | Optional | Specify YES if the text delimiter is to be removed or if the escape character is to be replaced when the files are transferred. This is the default value. Enter NO if the character is to be transferred unchanged. If you specify xml.enclosureSign=" and xml.enclosureSignEsc="" , text enclosed in quotation marks is transferred unchanged and the quotation marks are removed. If the escape character for a quotation mark ("" ) occurs in the text itself, it is replaced by the quotation mark during the transfer. Note : This property will be used only whenRecordSetName.fieldSeparator is used. |
| RecordSetName.keyFieldInStructure | Optional | If the key field of the substructure is to be added to the XML document, enter add . This is the default. If the key field is to be ignored, enter ignore . |
| RecordSetName.missingLastFields | Optional |
|
| RecordSetName.additionalLastFields | Optional |
|
| RecordsetName.fieldContentFormatting | Optional |
|
| RecordSetName.fieldNamesMandatory | Optional | Acceptable values are yes or no. Note : This is a custom property and the use case is provided below. |
How to create the Properties file:
- Open any text editor
- Save it with extension .properties
- Each and every property occurs in combination of key and value.
Eg: propertyname=propertyvalue
Scenario 1:
The below snapshot depicts the xsd which is created with 5 levels of depth
Module Configuration:
In the below configuration, Configuration_Mode is specified as file, we have specified all the FCC properties in the file and stored in the NFS path and specified the path of the file to File_Path property.
The advantage here is, user no need to modify/change the channels in order to change the FCC property.
If the Configuration_Mode is either not specified or configured as local, then all the FCC properties should be configured in the Module Configuration.
The FCC for the above structure is provided in the properties file as given below.
Illustration of FCC configuration in the properties file:
- DocumentName = MessageType
- KeyFieldName=key,specifies the key field name which is available in the first column of each recordSet, Module will read each and every line based on the end separator and needs to identify to which recordset the line should be associated.
- root.fieldNames=key, Name, Value
- root.recordsetStructure=Records,*
- root.Records.fieldNames=key, R_Name, R_Value
- root.Records.recordsetStructure =RHeader,1,RItem,*,Details,2
- root.Records.RHeader.fieldNames=key,RH_Name,RH_Value specifies the field names which is created under the RHeader element. To identify any line belongs to RHeader use the keyFieldValue property as root.Records.RHeader.keyFieldValue=3
- root.Records.RItem.fieldNames=key,RI_Name,RI_Value specifies the field names which is created under the RItem element. To identify any line belongs to RItem use the keyFieldValue property as root.Records.RItem.keyFieldValue=4
- root.Records.Details.fieldNames=key,D_Name,D_Value specifies the field names which is created under the Details element. To identify any line belongs to Details use the keyFieldValue property as root.Records.Details.keyFieldValue=4
- root.Records.Details.recordsetStructure=DHeader,1 specifies the recordset name DHeader should appear exactly once.
root.Records.Details.DHeader.fieldName= key,DH_Name,DH_Value
specifies the field names which is created under the DHeader element. To identify any line belongs to DHeader use the keyFieldValue property as root.Records.Details.DHeader.keyFieldValue=6
Input Data:
Output XML :
Scenario : 2
Requirement: consider the below requirement, where keyFieldValue is not configured for one of the recordset
- if the line starts with value HDR, then the line is associated to Header recordset
- if the line starts with value TRL, then the line is associated to Trailer recordset
- if the line starts neither with HDR nor with TRL then it has to be associated to Details recordset
The above requirement cannot be achieved using the Standard FCC and module addressed the above problem, by not configuring the keyFieldValue property to that corresponding recordset where fieldNames property is configured.
Note: if keyFieldValue property is not configured for 2 recordsets, for which fieldNames property is configured, then error will be raised, because module will be unable to determine to which recordset the line should be associated.
All the FCC parameters are placed in the properties file:
FCC Configuration:
Input File:
Output XML:
Scenario : 3
The below scenario illustrates how to use the missingLastFields , additionalLastFields and enclosureSign properties.
FCC properties is provided in the properties file for ease of use, as channel is not required to be activated if any of the FCC properties gets changed.
FCC Configuration:
Input File:
Output XML:
Scenario :4
As per the Standard FCC documentation, KeyFieldName is required, if any of the Recordset cardinality is configured as *. If KeyFieldName is not configured then the cardinality should not be configured as * for any of the recordset.
Note : If the KeyFieldName is not configured and if the keyFieldValue is specified for each of the recordset, then keyFieldValue will not be considered while building the XML.
For the sake of simplicity, the FCC is created in the properties file.
In the below FCC configuration, KeyFieldName property is not configured.
FCC Configuration:
Input File:
Output XML:
Scenario 5:
Hope you all know that, in PI Data Type, Message Type can be created from scratch or it can be imported as an external definition.
Consider the scenario, where the Message Type XSD is provided by the third party.
Message Type:
FCC Configuration:
Input File:
Output XML:
Description : In the above XML, fields Rec_Type, Fld1,Fld2,Fld3 is missing in the first LineItem, as the corresponding line is not available in the FlatFile.txt. This is the default behaviour.
If the default behaviour is not desired, then use the custom property fieldNamesMandatory as root.LineItems.fieldNamesMandatory=yes . If this property is configured and if we run with above FlatFile.txt then an error will be thrown by the Adapter Module.
Note:
- In General, if the Recordset( in our eg: LineItems) configured with both the fieldNames property and recordsetStructure property, then fieldNames property considered as optional. i.e.,. if the line belongs to fieldNames ( eg: root.LineItems.fieldNames) is available in the flatfile, the fields configured in the fieldNames property will be generated, else it will not be generated.
- The custom property fieldNamesMandatory is effective, if the recordset is configured with both the fieldNames property and the recordsetStructure property.
In the above FCC, root.LineItems recordset is configured with both the root.LineItems.fieldNames property and
root.LineItems.fieldNames property and root.LineItems.recordsetStructure property and root.LineItems.recordsetStructure property and hence the root.LineItems.fieldNamesMandatory=yes is used.
- If either the fieldNames property or recordsetStructure propery is not configured then fieldNamesMandatory property will not be taken into consideration.
In the above FCC, root.LineItems.Details recordset is configured with root.LineItems.Details.fieldNames property but not
configured root.LineItems.Details.recordsetStructure property and hence the custom propery
root.LineItems.Details.fieldNamesMandatory=yes will not be effective.
Use the below link which explains conversion of XML with Nested Structures to FlatFile,
Content Conversion for 'n' Nested Structures( XML -> Flat File) Using Adapter Module - Receiver
Disclaimer:
- This Module is optimized to process messages with an approximate payload size <= 5 MB.
- This Module will not work with messages having attachments