Kent Fredric πŸ”₯🐢🍡πŸ”₯

NAME

PPIx::DocumentName - Utility to extract a name from a PPI Document

VERSION

version 0.001003

DESCRIPTION

This module contains a few utilities for extracting a "name" out of an arbitrary Perl file.

Typically, this is the module name, in the form:

  package Foo

However, it also supports extraction of an override statement in the form:

  # PODNAME: OverrideName::Goes::Here

Which may be more applicable for documents that lack a package statement, or the package statement may be "wrong", but they still need the document parsed under the guise of having a name ( for purposes such as POD )

USAGE

The recommended approach is simply:

  use PPIx::DocumentName;

  # Get a PPI Document Somehow
  return PPIx::DocumentName->extract( $ppi_document );

METHODS

extract

  my $docname = PPIx::DocumentName->extract( $ppi_document );

This will first attempt to extract a name via the PODNAME: comment notation, and then fall back to using a package Package::Name statement.

$ppi_document is ideally a PPI::Document, but will be auto-up-cast if it is any of the parameters PPI::Document->new() understands.

extract_via_statement

  my $docname = PPIx::DocumentName->extract_via_statement( $ppi_document );

This only extract package Package::Name statement based document names.

$ppi_document is ideally a PPI::Document, but will be auto-up-cast if it is any of the parameters PPI::Document->new() understands.

extract_via_comment

  my $docname = PPIx::DocumentName->extract_via_comment( $ppi_document );

This will only extract PODNAME: comment based document names.

$ppi_document is ideally a PPI::Document, but will be auto-up-cast if it is any of the parameters PPI::Document->new() understands.

ALTERNATIVE NAMES

Other things I could have called this

  • PPIx::PodName - But it isn't, because it doesn't extract from POD, only returns data that may be useful FOR POD

  • PPIx::ModuleName - But it kinda isn't either, because its more generic than that and is tailored to extracting "a name" out of any PPI Document, and they're NOT all modules.

SIMILAR MODULES

Modules that are perceptibly similar to this ones tasks ( but are subtly different in important ways ) are as follows:

  • Module::Metadata - Module::Metadata does a bunch of things this module explicitly doesn't want or need to do, and it lacks a bunch of features this module needs.

    Module::Metadata is predominantly concerned with extracting ALL name spaces and ALL versions from a module for the purposes of indexing and indexing related tasks. This also means it has a notion of "hideable" name spaces with the purpose of hiding them from CPAN.

    Due to being core as well, it is not able to use PPI for its features, so the above concerns mean it is also mostly based on careful regex parsing, which can easily be false tripped on miscellaneous in document content.

    Whereas PPIx::DocumentName only cares about the first name of a given class, and it cares much more about nested strings being ignored intentionally. It also has a motive to show names even for documents that won't be indexed ( And Module::Metadata has no short term plans on exposing hidden document names ).

    PPIx::DocumentName also has special logic for the PODNAME: declaration, and may eventually support other mechanisms for extracting a name from "a document", which will be not in Module::Metadata's collection of desired use-cases.

  • Module::Extract::Namespaces - This is probably closer to PPIx::DocumentName's requirements, using PPI to extract content.

    Most of Module::Extract::Namespaces's code seems to be glue for legacy versions of PPI and the remaining code is for loading modules from @INC ( Which we don't need ), or special casing IO ( Which is also not necessary, as this module assumes you're moderately acquainted with PPI and can do IO yourself )

    Module::Extract::Namespaces also obliterates document comments, which of course stands in the way of our auxiliary requirements re PODNAME: declarations.

    It will also not be flexible enough to support other name extraction features we may eventually add.

    And like Module::Metadata, it also focuses on extracting many package declarations where this module prefers to extract only the first.

ACKNOWLEDGEMENTS

The bulk of this logic was extrapolated from Pod::Weaver::Section::Name and a related role, Pod::Weaver::Role::StringFromComment.

Thanks to RJBS for the initial implementation and DROLSKY for some of the improvement patches.

AUTHOR

Kent Fredric <kentnl@cpan.org>

COPYRIGHT AND LICENSE

This software is copyright (c) 2017 by Kent Fredric <kentfredric@gmail.com>.

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.