1.4

record. So the boundaries for each record can be completely dependent on how you want to
use the data.
With no actual boundary markers inside the data, we have to figure out a way to identify specific
locations in the input stream and mark those locations as record boundaries. Fortunately for us,
it turns out that every single file format possesses intrinsic, natural delimiters that are used to
identify chunks of related data. These delimiters are key in helping us identify boundaries, so
it's important to understand what they are as well as when and why they occur in the Data
Source.
Let's start with a seemingly arbitrary assumption: a boundary can only occur on a natural
delimiter. That is to say, a record boundary never occurs between delimiters, it only occurs on a
delimiter. The actual information we need to determine whether a delimiter is a candidate for
being a boundary is very likely to be found between delimiters.
To do that, we'll take a closer look at the various input data types.
For a CSV or Database File
The natural delimiter for a CSV file is a data record, or to put it more visually, each line in a
spreadsheet or in a SQL data grid is a delimiter. Several such delimiters can be included in a
record, but you would never expect to find the end of one particular record right in the middle of
one of those lines on the grid. So here, the record occurs with a new line in the grid, but not on
each new line.
l
Record limit: Defines how many Source Records are displayed in the Data Viewer. This
does not affect output production, as generating output ignores this option. To disable the
limit, use the value "0".
l
Line limit: Defines the limit of detail lines in any detail table. This is useful for files with a
high number of detail lines, which in the DataMapper interface can slow down things.
This does not affect output production, as generating output ignores this option. To
disable the limit, use the value "0".
l
Trigger: Defines the type of rule that control when a boundary is created, establishing a
new record in the Data Sample (called a Source Record).
l
Record(s) per page: Defines a set number of lines in the file that go in each Source
Record.
l
Records: The number of records to show in each Source Records.
l
On change: Defines a new Source Record when a specific field (Field name) has a
new value.
Page 118