What is a geoprocessing package?
In ArcGIS, a package is a compressed file containing GIS data. You share a package like any other file—via e-mail, FTP, the cloud, thumb drives, and so on. You share them between colleagues in a work group, between departments in an organization, or with any other ArcGIS users via ArcGIS Online. The recipient of your package unpacks it—typically by dragging and dropping it onto ArcMap—and immediately begins using its contents.
Geoprocessing packages were introduced in ArcGIS 10.1 for Desktop. You cannot create or use geoprocessing packages with versions earlier than 10.1.
Currently, there are five kinds of packages you can make with ArcGIS:
- A layer package (.lpk) includes both the layer properties and the dataset referenced by the layer. With a layer package, you can save and share everything about the layer—its symbolization, labeling, table properties, and the data.
- A map package (.mpk) contains a map document (.mxd); all the data referenced by the layers it contains; and other map items such as graphics, layouts, and so on. Essentially, it's a collection of layer packages along with the map document.
- A locator package (.gcpk) contains one address locator or a composite locator along with its participating locators.
- A tile package (.tpk) contains a tile cache of data that you can display as a basemap with ArcGIS Runtime applications.
- A geoprocessing package (.gpk) is how you share your geoprocessing workflows.
Geoprocessing packages are created from one or more results in the Results window. All the data and tools used to create the result are included in the package. You can add additional files to the package, such as text documents, slide shows, and compressed ZIP files. Your colleague unpacks the package to immediately begin using its contents.
In general, there are two methods for creating packages: using a menu in ArcMap (for example, right-click a result and choose Share As > Geoprocessing Package) or using geoprocessing tools in the Package toolset. In general, the geoprocessing tools give you a finer degree of control over how data is packaged.
For more information on how to create packages and unpack them, see A quick tour of geoprocessing packages.
Ideas for using geoprocessing packages
Below are some ideas for how you can use geoprocessing packages. Although these descriptions are for geoprocessing packages, they can be generalized to apply to map, layer, and locator packages as well.
You may have an expertise in a particular subject, such as surface water runoff modeling, or expertise in a technique, such as linear referencing. You can share your expertise using a geoprocessing package, delivering sample data and models you've developed, along with ancillary files such as Word documents that further explain your methods. The recipients of your package can study your methodologies and apply them to their own data.
Suppose you have a project where your colleagues are in different locations and are working on different aspects of the project. For example, in a site suitability study, one of your colleagues is responsible for evaluating soils, while another is dealing with transportation issues. The results of their individual investigations can be packaged and shared with everyone working on the project. At key points in the project, these individual packages can be consolidated into one package for redistribution to all team members.
This may have happened to you: you're working in ArcMap, you've got 15 layers in your table of contents, and their data is from different places—some is on your local disks, some consists of UNC paths to different machines on your network, and some may be data from enterprise databases. You're deep into developing some models that use some but not all the data. And now you've got to unplug your laptop from the network and go on the road, and you want to continue working on your models, but you're going to be off the network. All you need to do is run the models you're working on to create results, then package the results into a geoprocessing package. All the data and models are included in the package. (When packaging with the Package Result tool, you can specify that you only want data in the current map extent and to extract enterprise database data into a file geodatabase that's included in the package.) You then unpack the package to a folder on your local disk, and you have everything you need to continue working consolidated into one folder.
Consolidating data into a single folder is such a common task that there are a suite of tools in the Package toolset that consolidate data. These tools are a shortcut to packing and the unpacking to a specific folder.
Your project may involve creating many different scenarios. For example, you might be developing land-use plans for a region, and you need alternative plans for different assumptions, such as high growth in some areas and negative growth in others, planned improvements in transportation and other infrastructure, and funds available for open space purchases. Each plan forms a scenario: a unique collection of data, edits to the data, models, and model parameters. You can create a geoprocessing package to quickly make a snapshot of a scenario, freeing you to create new scenarios within the same map document. At some future point, you can unpack each scenario into its own folder and map document.
If one of your jobs is to provide training courses for your staff or others outside your organization, you can use packages to deliver your course materials.
We've all experienced this—you just can't seem to get your model or script to work. You need help from a colleague. You can package your result and ship the package to your colleague.
You cannot create a package from a failed result . You can, however, use the Consolidate Result tool to create a folder containing all data and tools used by the failed result, then compress this consolidated folder (using the ZIP utility, for example) to give to your colleague. They uncompress the file into a local folder and use its contents to troubleshoot your tools.