Directory input connector

This connector retrieves files from a named directory. The source application sends files to this directory, and the StreamServer retrieves the files.

Comments

Several connectors can scan the same directory for input. The connectors can belong to different StreamServer instances running on the same machine or on different machines. The Directory connector that first finds a match moves the file to a temporary directory. The reason is to prevent the file from being processed more than once. After the file is moved it is opened and processed. A temporary directory streamserve_proc is automatically created under the scanned directory. The default behavior is to move the files to this directory. If the StreamServer scans a directory located on a remote machine, it is safer to move the files to the tmp directory of the StreamServer (normally export/tmp). The reason is that if the network goes down, the StreamServer may also go down if the file is not stored locally.

The connector settings are described below.

Folder

The folder to scan for input. Use an absolute path, or a path relative to the working directory of the StreamServer. Two sub directories (streamserve_lock and streamserve_proc) are created when you run the Project. These directories ensure that a file is processed only once. These directories should not be tampered with.

File name pattern

The match criterion for the files to retrieve.

Example 4

Pattern for retrieving all *.txt files

*.txt

 

Interval for file size check

This interval specifies how frequently to check the file size. The file is not moved to the temporary directory immediately after it is found. First the file size is checked, and the file is moved only if the file size remains the same after this time interval. If the file size has changed, the connector waits for yet another interval to pass before it tries to move the file. This prevents the connector from moving the file before it is completed.

Sort by

StreamServer generates a list of all the files found in the scanned directory, and uses this list to find files that match File name pattern above. You can sort this list by:

•

Name

•	Date and time

•	File extension

•

File size

The files are by default sorted in descending order.

Sort ascending

Select to change the order in the sort list to ascending.

Save path

When a file is retrieved by the StreamServer it is removed from the scanned directory. If you want to save copies of the retrieved files you must specify a Save path. Use an absolute path, or a path relative to the working directory of the StreamServer.

Create Folder

If the input folder (path specified in Folder above) does not already exist, it is created by StreamServer if Create Folder is enabled.

Delete scanned files

If enabled, files are deleted from the scanned folder after they are retrieved by StreamServer. This ensures that an input file is not processed more than once.

If disabled, StreamServer will retrieve the files over and over again, using the polling interval specified (see Schedule below). This can be used, for example, to scan a database for changes once per polling interval.

Use spool directory as temporary storage

Select to use streamserve_proc as the temporary directory for the retrieved files. If the scanned directory is located on a remote machine, it is safer to use the tmp directory of the StreamServer (normally export/tmp). In this case you should not select this option.

Synchronize scanning

Select to enable synchronized scanning for the connector. The connector runs single-threaded and retrieves files one by one from the input folder.

If you do not select this option, the connector scans the input folder asynchronously. The connector runs multi-threaded and retrieves several files in parallel.

Schedule

Opens the Scheduler Configuration dialog where you specify when and how often to scan the directory. See Scheduling actions in the Design Center documentation for more information about scheduling.