Delimited Text Files

Select this option if the data source is a text file. The Source tab layout is then as shown in the figure below - .

Figure 1 - The Source tab with File selected as the Source Type

• Formatting – data is copied out as a text file. Select comma-separated or tab-separated as required.
• Locale – specify the area or country in which the data in the file is to be used. This will allow the system to format fields such as date fields according to local standards.
• Unicode – check this box if the file is saved as Unicode. The Code Page property is then disabled.
• Code Page – select the code used in the file.
• Truncate Open Ends - when exporting data to File, system performance is heavily affected by the number of “unlimited width” text fields in the exports (i.e. Open Text fields with no field width settings). Open Text fields can be truncated by using this setting (default is 200 characters). Forsta strongly recommends truncating Open Text field widths with this setting rather than choosing “Full length”, as this will lead to significant improvements in processing time.
• Uploaded File - import files can either be uploaded directly in the browser, or placed on an FTP server and retrieved from there (see below). Select the required option.
• Pick up from FTP Location - if your company has licensed the FTP add-on and it has been enabled, this option will be active. You can then fetch the file to be uploaded from Forsta's FTP location. Forsta's FTP address is specified as part of the enabling procedure and cannot be changed, so you must add only the file name to this field. If you select this option, then you can also decide what to do with the original file on the FTP site; leave it there or delete it (see below). Note that fetching from FTP can allow automation of data flows, since you can set up the system to pick up files automatically
• File Handling - if you have selected to use the FTP option (see above), then you can decide what to do with the file on the FTP site after it is uploaded; leave it there or delete it.
• Encrypted File - if your company has licensed the Encryption add-on and it has been enabled, the data transfer encryption functionality makes it possible to perform secure data transfers in Forsta Plus. If the source file has been encrypted by the sender using your PGP public key, you will need to decrypt the files using your private key. Note that an administrator setting makes it possible to force every Forsta Plus user within a company to use encryption and FTP server during data transfer.
• Map Fields - Use this to map fields/columns with different names in source and target, from the source to the Survey Database (go to Mapping Fields for more information).

1. Make the required settings and save them.
2. In the event you wish to upload a file click Upload File, if you wish to pick up the file from the FTP site type the file name into the field.

For uploaded files, a File Upload dialog opens.

1. Browse to and select the file to be uploaded, then click OK to load the file.

Field names in the uploaded file that are not the same as field names in the database must be mapped manually to the appropriate fields in the database. Any field names in the uploaded file that are identical to field names in the database will be mapped automatically. If any columns in the file need to be mapped manually, a message will be presented below the Map Fields button. If this is the case and you wish to map the fields, click Map Fields(go to Mapping Fields for more information).

1. Click Save to save the changes.

Note: If you have more than one BitStream data source in a panel and use rules to update them (go to BitStream Files for more information), then when you need to update one BitStream set it is very important that you update all of them. If this is not done, then the “not-updated” BitStream files may be inconsistent and will then produce incorrect results.

An example to illustrate the situation:

Assume that you have two separate BitStream sets, A and B, and you have a number of sampling jobs working on each. When you run a sampling job on a BitStream set, this will cause the selected panelists to be noted in a table of “dirty” panelists – these are essentially panelists who have been selected for one or more jobs without the BitStream data source being informed (the BitStream data sources cannot know that a panelist has been selected for a job until the files are updated). When a sampling job is run, dirty panelists are detected by looking in the dirty table, so any panelists currently in the table will be checked and excluded/included as appropriate. If you now update BitStream set A, the panelists involved here will no longer be dirty, so will be removed from the dirty table. However BitStream set B has not been updated yet so still does not know about the dirty panelists. As these have now been removed from the dirty table they will not be noted, and there is a very real chance that when running another job on BitStream set B, panelists who should be excluded will be selected.