NAME

docs/configuration.pod - Parrot Configuration System

DESCRIPTION

Parrot configuration is broken up into steps. Each step contains several related prompts, probes, or generations. Steps should be mostly of a single type, though some overlap is allowed (for example, allowing a probe to ask the user what to do in an exceptional situation).

The directory config contains subdirectories for each type of step. Each step should consist of exactly one .pm file and any number of supporting .c, .in, etc. files. Any supporting files should be in a folder whose name is the same as the basename of the step's .pm file; for example, if foo.pm uses bar_c.in, bar_c.in should be in a directory called foo; the full path might be config/auto/foo/bar_c.in.

Generally, when adding a new test you should add a new step unless a test clearly belongs in a current step. For example, if we added a new user-configurable type called FOOVAL, you should add the test for its size in config/auto/sizes.pm; however, if you were testing what dynaloading capabilities are available, you should create a new step.

Initialization Steps

Initialization steps are run before any other steps. They do tasks such as preparing the configuration system's data structures and checking the MANIFEST. These will rarely be added; when they are, it usually means that the configuration system is getting significant new capabilities. They're kept in the directory config/init.

Initialization steps usually do not output anything under normal circumstances.

Prompts

Prompts ask the user for some information. These should be used sparingly. A step containing prompts is an interactive step. Interactive steps should be in the config/inter folder.

Interactive steps often include simple probes to determine good guesses of what the user will answer. See "Prompt or Probe?" for more information.

Interactive steps virtually always output something.

Note that, by default, these prompts are turned off. To enable them run Configure.pl with the --ask option.

Probes

Probes are automated tests of some feature of the computer. These should be used wherever a value will not often need to be modified by the user. A step containing probes is an automatic step. Automatic steps should be in the config/auto folder.

Automatic steps usually do not output anything under normal circumstances.

Generations

Generations create files needed after configuration has completed, such as Makefiles and configuration headers. A step containing generations is a generation step. Generation steps should be in the config/gen folder.

Generation steps usually do not output anything under normal circumstances.

Templates for to be generated files usually have the extension '.in'. There is variable substitutes and funny macros like 'CONDITIONED_LINE' and INVERSE_CONDITIONED_LINE'.

Prompt or Probe?

It can sometimes be hard to decide whether a given step should be an automatic or an interactive step. The guiding question is Would a user ever want to change this?, or conversely, Is this something that can be completely determined without user intervention? A step figuring out what the compiler's command is would probably be an interactive step; conversely, a step figuring out if that command is connected to a specific compiler (like gcc) would be an automatic step.

Configuring Configuration

The configuration system gets it's own configuration data from and is invoked via the Configure.pl script. The system is invoked by instantiating a Parrot::Configure object, registers one or more steps with that object, and then call it's "runsteps" in Parrot::Configure method.

Adding New Steps

New steps should be added in one of the three folders mentioned above.

All steps are really classes; they each exist in a unique namespace. The namespace used depends on the tests relative path in the source tree minus sans the config prefix. For example, the step config/init/defaults.pm uses the init::defaults namespace.

They should define the following methods:

description: Returns a short descriptive message that should be printed before the step executes. Usually, interactive steps have long, friendly descriptions and other steps have terse descriptions ending in "...".; Some example descriptions:; Note that on non-interactive steps, the text done will be printed after the description when the step finishes executing; for example, the user will see:
args: Returns a list of the names of any command-line arguments the step cares about. Command-line arguments are standardized in Configure.pl; this will be described later in more detail.; XXX Note that this method is currently unused but will be used in the futher to register acceptable CLI parameters. New steps should continue to define this method.
runstep: This method is called to actually execute the step. The global Parrot::Configure is passed is as the first parameter. The return value is undefined.; XXX In the near future the return value of this method will be signifigant and there will be a means of passing additional parameters.

The configuration system won't execute your step by default unless it's specifically told to. To do this, edit the Configure.pl files. Steps are run in the order in which that are registered with the Parrot::Configure object.

Various utility functions for configuration steps are provided by the Parrot::Configure::Step module.

A template for a new step might look like this:

    package auto::newstep;

    use strict;
    use warnings;
    use vars qw($description $result @args);

    use base qw(Parrot::Configure::Step::Base);

    use Parrot::Configure::Step;

    $description = '<description>';
    @args = qw(<args>);

    sub runstep {
        my ($self, $conf) = @_
        ...
    }

Command-line Arguments

Command-line arguments look like /--[-\w]+(=.*)?/; the equals sign separates the name and the value. If the value is omitted, it's assumed to be 1. The options --help and --version are built in to Configure; any others are defined by steps.

If you add a new option, don't forget to add it to this documentation and the --help listing in Configure.pl.

Arguments passed to Configure.pl are held in a Parrot::Configure::Data object stored inside the Parrot::Configure object. The options data object may be accessed via the "options" in Parrot::Configure method.

Steps use the args method to list any options they're interested in. They should be listed without the dashes.

Building Up Configuration Data

The second step is config/init/defaults.pm, which sets up some defaults in a Parrot::Configure::Data object contained by the main Parrot::Configure object. It can be accessed via the "data" in Parrot::Configuremethod. You get and set configuration system's data by interacting with this object. Some of it's methods are summarized below.

get(keys): Returns the values for the given keys.
set(key, value, [key, value, ...]): Sets the given keys to the given values.
add(delim, key, value, [key, value, ...]): Sets the given keys to the given values or appends values delimited by delim to existing keys.
keys(): Returns a list of all keys.
dump(): Returns a string that can be evaled by Perl to create a hash representing the configuration system's data.; See the Parrot::Configure::Data documentation for further details.

Special Configuration Items

Some configuration items, by convention, have a special meaning (mainly prefixes) and are handled with some magic internally.

i_(\w+) include files: defines or undefs PARROT_HAS_HEADER_XXX in include/parrot/has_header.h
HAS_(\w+) features: defines PARROT_HAS_XXX in include/parrot/has_header.h
TEMP_(\w+) temporary settings: These settings are deleted before lib/Parrot/Config.pm is written. These entries are only used e.g. for Makefile creation.

HISTORY

The Parrot configuration system was created by Brent Dax and has been heavily mangled by Joshua Hoblitt.

parrotcode: Parrot Configuration System
Contents \| Configuration