pipe-config

The pipe-config applet generates a new pipeline file from a set of inputs. All the information from the config files and settings is merged into the specified pipe file and written as output. This can be used to flatten a complicated pipe file which may include other pipe and config files into a single dasily distributed file.

kwiver pipe-config [options] pipe-file

Options are:

-h, --help

Display applet usage information.

output options:

-o, --output arg

Name of output file or ‘-‘ for stdout. (default: -)

pipe options:

-c, --config <file>

Specifies a file containing supplemental configuration entries. Can occur multiple times on the command line.

The supplemental configuration files can contain additional configuration entires for processes or global config blocks. Configuration items for processes start with the process name. For example The input_path config element for the process reader would be specified as follows:

reader:input_path = input/file/path

These configuration values are applied to the pipeline configuration after the pipeline file has been processes and can overwrite values previously defined in the pipe file.

-s, --settings VAR=VALUE

Specifies additional configuration settings to be applied to the pipeline configuration.

The settings option can be used to specify a single configuration value. This command line option can appear multiple times on the command line. Using the same example above, that configuration item can be specified as follows:

--settings reader:input_path=input/file/path

Note that there can be no embedded spaces in the argument.

-I, --include <path>

Specifies a directory to be added to the configuration include path. This option can occur multiple times on the command line

pipe-file - name of pipeline file.