Merger
All available mergers are defined in configuration file ./etc/admin/datawizard/ci_merger.properties . We provide a programming interface (API) that also allows you to develop your own classes in Java. For this, we offer in-depth training. If you are interested, please contact our support or sales staff.
Merge Edifact
The merger Merge Edifact can merge several EDIFACT documents. The merged document uses the UNA, UNB and UNZ segments of the first received EDIFACT document . Note: This means that the interchange control references of the individual documents get lost . The system variable MSG_CALL_REF_IDS (if defined) contains the interchange control references of all merged EDIFACT documents and can be read by subsequent profiles. Example: 339;338;337
Merge JSON
The merger Merge JSON merges several JSON files into an array. Example:
{ "fruit": "Apple", "color": "Red"}{ "fruit": "Apple", "color": "Green"}[{ "fruit": "Apple", "color": "Red"},{ "fruit": "Apple", "color": "Green"}]Merge XML
The merger Merge XML allows you to syntactically correct merge XML files below an artificial parent tag.
Assume the two following XML files.
<?xml version="1.0" encoding="ISO-8859-1"?><ORDER nr="9879871234" cust="John Doe"> <POS index="1" no="7687876" amount="1" artName="Article1"/> <POS index="2" no="5477969" amount="1" artName="Article2"/> <POS index="3" no="3687652" amount="2" artName="Article3"/></ORDER><?xml version="1.0" encoding="ISO-8859-1"?><ORDER nr="9527638478" cust="Test customer 1"> <POS index="1" no="6537896" amount="1" artName="Article4"/> <POS index="2" no="2971087" amount="1" artName="Article5"/></ORDER>The result would be the following XML file.
<?xml version="1.0" encoding="UTF-8"?><merge> <ORDER nr="9879871234" cust="John Doe"> <POS index="1" no="7687876" amount="1" artName="Article1"/> <POS index="2" no="5477969" amount="1" artName="Article2"/> <POS index="3" no="3687652" amount="2" artName="Article3"/> </ORDER> <ORDER nr="9527638478" cust="Test customer 1"> <POS index="1" no="6537896" amount="1" artName="Article4"/> <POS index="2" no="2971087" amount="1" artName="Article5"/> </ORDER></merge>Merge text
The first version of Merge text is only available for compatibility reasons and is not described here. Please always use version 2.
Merge text V2
The merger "Merge Text V2" allows you to merge files that match the defined file name patterns and belong to a context, which is defined by the first entry of the list (described in more detail below).
(1) Append line break: This allows you to insert a line break between the merged files.
(2) Hold back single files: If this option is selected, the File Pool must contain at least 2 files for the merger to start.
(3) List below: The list of entries. Note: At least two entries have to be defined.
General approach
For all files in the File Pool, the first entry generates a key value (if the pattern matches). Generation of the key value (with one of the two possible methods):
(4) Start/End: A substring is formed from the file name with a start index and an end index (inclusive). For example, Start=3 and End=3 for file h_1.txt generate the key value 1 (Prefix and Suffix must stay empty).
(5) Prefix/Suffix: The key value is generated from the text between Prefix and Suffix. For example, Prefix=h_ and Suffix=.txt for file h_1.txt generate the key value 1 (Start and End must stay 0).
For each different key value, a context is created.
The merger now checks (based on the other entries) the context to which any file in the File Pool belongs belong (when the File Pool is being processed). If the file name matches the pattern of the entry, the key value is also generated. If the key value is identical with that of a context, then the file is associated with that context. When all entries in the list have been processed for all files in the File Pool, the files of each context are merged in the order found if condition (6) "Min. (V2)" is met. This is the minimum number of files for the merging to start (you could, for example, ask for at least 2 item files or at least 1 header file).
Example
The files h_*.txt contain an order header and the associated items are in files d_*.txt.
So we want to merge files h_1.txt, d_11.txt and d_12.txt, ... and the files h_2.txt, d_21.txt, d_22.txt, ...
Therefore, we use to the third character of the file name to construct the key value that determines the context of the files to be merged.
For that, we create the first entry with pattern h_*.txt and Start=3 and End=3.
For the second entry, we use the pattern d_*.txt and also Start=3 and End=3.
The screenshot above shows the configuration for this example.
File h_1.txt matches the pattern h_*.txt of the first entry. The third character of the file name h_1.txt creates the key 1 for the first context. And we get key 2 from file h_2.txt for the second context.
File d_11.txt matches the pattern d_*.txt of the second entry. The third character of the file name d_1.txt creates the key 1, therefore the file is associated with the context with key 1. Analogously for d_12.txt, d_21.txt and d_22.txt.
Important note
Of course, the Content Inspector may also receive files that do not match the defined patterns. These files also end up in the File Pool of the Content Inspector.
If the File Pool is processed (due to a timeout or a reached maximum number of files), the merger Merge Text V2 merges the files in the File Pool according to the rules described above. However, files that do not match the defined pattern are not touched and remain in the File Pool, and the timer is reset to 0. But you can manually remove these remaining files from the File Pool (if you want to prevent an influence on the maximum number of files, for example).
Files can therefore only automatically leave a Content Inspector with configured merger Merge Text V2 via merging when the processing of the File Pool is triggered.
But a simultaneously configured splitter works in the usual way (and thus before the processing of the File Pool).