Import job properties: Advanced
Contents
On the Advanced tab, which is present when you have opened a new or existing import job in the Import job editor, you may specify some special properties of the current import job, that are usually best left to their default settings though.
Click here for information on how to edit job properties in general.
On the current tab you'll find the following settings:
The input buffer is memory space allocated by the software to read data from the exchange file. Each time the exchange file is accessed for this purpose this amount of data is being read in. The bigger you set this input buffer, the less disk I/O (reading and writing actions) need to take place, and the faster importing will be. Therefore, by default this size is set to 32000, for maximum performance. Only if your computer is very old and has very little memory, you may want to choose a lower setting.
Every time data from the exchange file is read, minimally one field is read. If the input buffer size is smaller than a particular field, the input buffer will be resized dynamically and automatically to read the entire field. On the other hand, a full input buffer may contain multiple records too.
With this option, you can indicate whether you want to import the contents of long fields into one occurrence of the destination field (set the default word-wrap value to zero), or whether you want to spread such contents over multiple occurrences of the destination field (set the default word-wrap value to the maximum length that any destination field occurrence may have.
In most cases you will want to leave this setting to zero, so that the contents of long word-wrapped fields is imported into similar word-wrapped destination fields. More specifically, this setting at zero sets the source field length to the destination field length as defined in the data dictionary, during importing.
If the destination field length in the data dictionary is shorter than the character string you import, information at the end of the string will be lost.
With this option you set whether with this import job you want to import records from the exchange file into the destination database table. Since most import jobs are about importing data, you'll probably want to leave this option selected.
At first it may seem odd to have an option for this: aren't all imports about importing data? Well, most are: the only exception is when you only want to reindex your indexes (see the option below) with the help of a (dummy) exchange file but do not want to add or change anything to your database. This is an alternative use for import jobs as reindex jobs, in which nothing is imported; then deselect this option, and deselect the Reindex automatically option too.
For small to normal import jobs you can let the import job simultaneously (per imported record) reindex all appropriate indexes, while the database is being filled: then deselect the Reindex automatically option, and of course the Import database file option must be marked.
For a lot of exchange files it is very important to deselect this option! For instance, for importing files that contain records with internal links you must definitely deselect the Reindex automatically option, because processing those links is impossible if not all indexes are updated continuously.
You can also use this deselected option to reindex all indexes without filling the database; of course then you need to deselect the Import database file option. And then the Clear database option on the Options tab, needs to be deselected too. (You do need a dummy exchange file for this, even though you don't fill the database.)
Mark this Reindex automatically option though to reindex only the priref index and other unique indexes during import, and reindex all other indexes only after the database table has been filled, to speed up very large import jobs. But make sure this database table has no internal links, otherwise deselect this option.
Note: the setting of this option appears illogical, but you really have to deselect this option to activate automatic reindexing of all indexes during import.
This option specifies the encoding in which this import job (the .imp file) must be saved. This is only indicative of the encoding of the import job, not of the exchange file nor of the target database table. You can choose between Oem, ANSI and UTF8. Usually, ANSI is all you need. Only if text, file names or source field names in this import job contain exotic characters, you need Unicode in UTF-8 encoding. However, a bug in Designer versions older than 7.1.13318.8 caused the UTF8 option to store the import job in ANSI encoding anyway, thereby corrupting any exotic characters in the import job. So if you need UTF-8 encoding for your import job, you require Designer 7.1.13318.8 or higher: you'll find that the encoding no longer has to be specified because it now always defaults to UTF-8 (which suits all needs).