Coverage data limitations

License:

The Coverage Toolbox is installed with ArcInfo Workstation and works with an ArcInfo license. You can work with coverages in ArcCatalog with an ArcEditor or ArcInfo license.

This provides you with a major overview of coverage data restrictions when working with the coverage tools in ArcGIS.

Size limits for a single coverage

Number of	Possible in ArcInfo Workstation
INFO data files	10,000 per workspace.
Nodes	MAXLIMIT.
Coordinates per arc	500. ArcInfo Workstation automatically creates a node at the 500th point and begins a new arc. A line feature can be represented by MAXLIMIT arcs.
Arcs per polygon	MAXLIMIT.
Arcs per node	360 arcs can connect at each node.
Arcs	MAXLIMIT.
Points	MAXLIMIT.
Polygons	MAXLIMIT.
Route systems	INFOLIMIT/2 (each system uses two INFO data files).
Routes	MAXLIMIT per route system.
Sections	MAXLIMIT per route.
Region subclasses	INFOLIMIT.
Regions	MAXLIMIT per region subclass.
Polygons per region	MAXLIMIT (constrained by number of polygons).
Tics	MAXLIMIT.
Annotation subclasses	MAXLIMIT.
Levels of annotation	MAXLIMIT.
Annotation strings	MAXLIMIT per annotation level.
User-ID values	MAXLIMIT, except for User-ID values for arcs. The valid range for User-ID in an AAT table is -1,073,741,823 to +1,073,741,823 (that is, 2**30 - 1). For the purpose of EXPORT, the valid User-ID range for all features is -999,999,999 to +2,147,483,647.

Table of size limits for a single coverage

MAXLIMIT is defined as 2,147,483,647 (that is, 2**31 - 1), which for most applications can be considered virtually unlimited. You are more likely to encounter limitations in disk space and increases in the time required to process large sets of data.

INFOLIMIT is defined as the number of INFO data files per workspace. The limit is 10,000, except for ARC/INFO releases prior to version 7, for which the limit is 1,000.

Feature attribute table limits

These constraints apply only to feature attribute tables stored in INFO and to INFO data files. They do not apply to external relational database management system (RDBMS) tables stored in Oracle, INGRES, Sybase, Informix, or others.

Maximum record size: The total width of a record in a feature attribute table cannot exceed 4,096 bytes or characters.

Maximum number of records per file: In theory, the number of records in a feature attribute table is unlimited, but in practice, it is constrained by disk space and processing time. The greater the number of bytes per record and the greater the number of records, the slower processing will be. There is one corresponding feature attribute record for each coverage feature.

Coverage processing limitations

There are two potential errors that may occur during the topology creation step of the Clean tool, the Build tool, and the Overlay tools. They are normally caused by use of extremely complex coverages that result in software limitations. Each condition is described below.

Too many arcs encountered in a scan line (SEGPUT): This error message tells you that the available memory used for storing active arcs has been filled. An active arc is one whose first node has been encountered but not its last node, since arc coordinates are scanned horizontally from top to bottom of a coverage. This happens when more than 2,666 arcs intersect any given scan line across a coverage.

This condition may be encountered in coverages that are wide or contain many long, vertical arcs that tend to run parallel instead of intersecting one another (for example, as in a large number of contours running vertically down a wide map). To reduce the storage requirements for active arcs, the size of the coverage can be reduced. For example, you can use the Clip tool to create a number of smaller coverages.

Segment table is full (PUTSEG) or Segment table full (CRESEG): This occurs in the intersection stage when there are too many band-related segments. This limitation results from a maximum of 8,000 entries in the active segment table. It could happen, for instance, when generating a buffer with a fuzzy tolerance that is relatively large compared to the buffer distance. Since the bandwidth is proportional to the fuzzy tolerance value, reducing the fuzzy tolerance value may help.