The table_of_contents template was already too bulky and redundant and recognizing public and private components was only going to make it worse. This refactors the toc template and the toc class to use a generic outline for all components.
If there is a class, type, or function with no information documented but is pulled in by some fluke or tag we don't support, do not give it its own section
`puppet generate` behavior remains unchanged
`puppet generate --format markdown` now defaults to writing to
REFERENCE.md and will accept an override from --out
`puppet generate --format json` defaults to printing to stdout but will
accept the --out param
This implements the two remaining tags and removes the author tag. Had to do some deeper plumbing because options per function signature were not being passed along from YARD
This implements the author tag. We also claim to support the 'raise' and 'option' tags, but I don't know that we need to. This also consolidates some code in base.rb and adds docs
This makes sure the markdown doesn't include section headers that have no content. e.g. '## Classes' does not get put in if there are no classes. Also, we should deprecate --emit-json and --emit-json-stdout instead of just killing them.
This change does a few things:
1. Fixes up new api handler to return the stuff we want
2. Adds all the logic to parse YARD registries into markdown
3. Adds templates for markdown
4. Changes Face cli to use a --format option that can be used for either
markdown or json
In the rake task strings::generate, the arguments json and yard_args
are not applied properly. The logic uses args.key?, however args is not
a Hash but a Rake::TaskArguments (and Enumerable). There is now key?
method.
Use Rake::TaskArguments.has_key? instead.
Prevent mass-changes to files by removing the date in the footer of each page.
Since the generation date isn't particularly useful and isn't tied to either
a git revision or a released version, this could be an easy way to address this
bug. The files will of-course be regenerated but if there are no changes then
the content will be identical and git will not notice.
Because Puppet 4 is typed, parameter type information can
be automatically determined without any explicit documentation
in @param tags. However, users may desire to do so anyway for
consistency. This commit allows @param tags to include a [type]
in Puppet 4 code. Strings will emit a warning if the documented
type does not match the actual type. In such an event, the incorrect
documented type will be ignored in favor of the real one.
This commit adds a new top-level key to the JSON schema for Ruby 3x/4x
puppet functions: `signatures`. This key is an array and is constructed entirely
with data from existing elements of the puppet function object. The data
structure itself remains unchanged, as this commit only affects the final
JSON output for functions.
In the case of a puppet 4x function with multiple dispatches (overloads),
the signatures key is populated with the equivalent of an overload tag
for each which includes the signature and any param or return tags for
the dispatch.
In the case of a puppet function with only one dispatch, the array is
simply populated with a single element containing a lone dispatch
and its param and return tags.
Note that this change does not affect the internal structure or
representation of puppet function objects, but rather only affects
its final display form when serialized into JSON. All prior data
has been left intact, and the only change here is the addition
of the `signatures` member to the JSON output.
Previously, overload objects were not displaying their tags when
they had no docstring text. This was due to an issue in the overload
`to_hash` method, which prevented the tags from being serialized when
the dispatch had no top-level text. This commit updates that logic
so that the tags will always be included in the hash if they exist.
Previously, Strings ignored calls to `return_type` in Puppet 4.x API
function dispatches, preventing the return types of overloads from
being automatically determined. This commit adds a check for a node
with a `return_type` call and handles it properly.
Previously, Strings ignored the return type in Puppet language functions
that used the following syntax:
function foo() >> String {}
This commit updates the FunctionStatement class to use the return
type from such a statement if it exists. In addition, Strings will
now emit a warning if the return type specified in the @return tag
doesn't match the type specified in the function definition.
This commit updates the gh_pages task to reference the SHA
of the branch being documented in the documentation commit.
In addition, the description of each task except `update`
has been removed, as they are not useful on their own and
should be considered private. This is necessary in order
to hide the tasks from the output of `rake -T`.
Previously, YARD would parse ruby strings which used %Q notation
and return the syntax literally. This commit adds a new Util
module with a `scrub_string` method to check for such a string
and remove the errant %Q and %q's.
Previously, due to Jekyll's default rule of not allowing files beginning
with an '_' to be published, the gh_pages rake task led to broken links.
This commit adds a step to the task which adds a _config.yml file to the
`doc` directory of the gh-pages branch which explicitly allows the file
to be published.
Prior to this commit, strings did not properly handle providers
which had multiple related `defaultfor`s. In code, these are written
as comma-separated constraints. This commit updates strings' puppet
provider handler, as well as the template which generates HTML for
`defaultfor` statements.
Note that it was necessary to make a breaking change to the JSON
schema to accomodate multiple AND'ed defaults. Previously, provider
defaults were contained in a single key-value map. Now, they are
contained in an array of key-value lists, which allows multiple
constraints to be associated with each other.
When `newfunction` is separated from the Puppet::Parser::Functions module name by a
newline, YARD ignores the namespace and uses `newfunction` as the source of the
first statement.
Prior to this commit, strings didn't recognize this case, and 3.x functions written
in this way were not parsed as functions. This commit updates the ruby function handler
to identify and properly parse 3.x functions that include a newline between the
Puppet::Parser::Function namespace and the newfunction method call.
This commit implements the `strings:generate` and `strings:gh_pages:update`
rake tasks.
The `strings:generate` task is responsible for generating Puppet Strings
documentation.
The `strings:gh_pages:update` task is responsible for updating the `gh_pages`
branch of a GitHub repository with the latest Puppet Strings documentation.
This commit implements the Puppet language parser that future handlers will use
to generate YARD code objects for classes, defined types, and functions written
in Puppet.
This commit adds the initial YARD configuration. Subsequent commits will
integrate the code objects, handlers, parsers, tags, and templates to support
Puppet code.
This commit deletes the old implementation to assist in cleaner code reviews of
the upcoming reimplementation.
This commit also moves YARD to version 0.9.5 and lays down a bare bones
implementation of Puppet Strings that currently does nothing.
Ewoud Kohl van Wijngaarden