gen_curried_sub

Usage:

 gen_curried_sub( [ \%optional_named_args ] , $base_name, $output_name) -> any

Generate curried subroutine (and its metadata).

This is a more convenient helper than "gen_modified_sub" if you want to create a new subroutine that has some of its arguments preset (so they no longer need to be present in the new metadata).

For more general needs of modifying a subroutine (e.g. add some arguments, modify some arguments, etc) use "gen_modified_sub".

This function is not exported by default, but exportable.

Arguments ('*' denotes required arguments):

• $base_name* => str

  Subroutine name (either qualified or not).

  If not qualified with package name, will be searched in the caller's package. Rinci metadata will be searched in %SPEC package variable.

• $output_name* => str

  Where to install the modified sub.

  Subroutine will be put in the specified name. If the name is not qualified with package name, will use caller's package.

• set_args => hash

  Arguments to set.

Return value: (any)

gen_modified_sub

Usage:

 gen_modified_sub(%args) -> [status, msg, payload, meta]

Generate modified metadata (and subroutine) based on another.

Often you'll want to create another sub (and its metadata) based on another, but with some modifications, e.g. add/remove/rename some arguments, change summary, add/remove some properties, and so on.

Instead of cloning the Rinci metadata and modify it manually yourself, this routine provides some shortcuts.

You can specify base sub/metadata using "base_name" (string, subroutine name, either qualified or not) or "base_code" (coderef) + "base_meta" (hash).

This function is not exported by default, but exportable.

Arguments ('*' denotes required arguments):

• add_args => hash

  Arguments to add.

• base_code => code

  Base subroutine code.

  If you specify this, you'll also need to specify "base_meta".

  Alternatively, you can specify "base_name" instead, to let this routine search the base subroutine from existing Perl package.

• base_meta => hash

  Base Rinci metadata.

• base_name => str

  Subroutine name (either qualified or not).

  If not qualified with package name, will be searched in the caller's package. Rinci metadata will be searched in %SPEC package variable.

  Alternatively, you can also specify "base_code" and "base_meta".

• description => str

  Description for the mod subroutine.

• install_sub => bool (default: 1)
• modify_args => hash

  Arguments to modify.

  For each argument you can specify a coderef. The coderef will receive the argument ($arg_spec) and is expected to modify the argument specification.

• modify_meta => code

  Specify code to modify metadata.

  Code will be called with arguments ($meta) where $meta is the cloned Rinci metadata.

• output_code => code

  Code for the modified sub.

  Alternatively you can use "wrap_code". If both are not specified, will use "base_code" (which will then be required) as the modified subroutine's code.

• output_name => str

  Where to install the modified sub.

  Subroutine will be put in the specified name. If the name is not qualified with package name, will use caller's package. If no "output_code" is specified, the base subroutine reference will be assigned here.

  Note that this argument is optional.

• remove_args => array

  List of arguments to remove.

• rename_args => hash

  Arguments to rename.

• replace_args => hash

  Arguments to add.

• summary => str

  Summary for the mod subroutine.

• wrap_code => code

  Wrapper to generate the modified sub.

  The modified sub will become:

   sub { wrap_code->(base_code, @_) }
      

  Alternatively you can use "output_code". If both are not specified, will use "base_code" (which will then be required) as the modified subroutine's code.

Returns an enveloped result (an array).

First element (status) is an integer containing HTTP status code (200 means OK, 4xx caller error, 5xx function error). Second element (msg) is a string containing error message, or 'OK' if status is 200. Third element (payload) is optional, the actual result. Fourth element (meta) is called result metadata and is optional, a hash that contains extra information.

Return value: (hash)

caller([ $n ])

Just like Perl's builtin caller(), except that this one will ignore wrapper code in the call stack. You should use this if your code is potentially wrapped. See Perinci::Sub::Wrapper for more details.

err(...) => ARRAY

Experimental.

Generate an enveloped error response (see Rinci::function). Can accept arguments in an unordered fashion, by utilizing the fact that status codes are always integers, messages are strings, result metadata are hashes, and previous error responses are arrays. Error responses also seldom contain actual result. Status code defaults to 500, status message will default to "FUNC failed". This function will also fill the information in the "logs" result metadata.

Examples:

 err();    # => [500, "FUNC failed", undef, {...}];
 err(404); # => [404, "FUNC failed", undef, {...}];
 err(404, "Not found"); # => [404, "Not found", ...]
 err("Not found", 404); # => [404, "Not found", ...]; # order doesn't matter
 err([404, "Prev error"]); # => [500, "FUNC failed", undef,
                           #     {logs=>[...], prev=>[404, "Prev error"]}]

Will put "stack_trace" in logs only if "Carp::Always" module is loaded.

warn_err(...)

This is a shortcut for:

 $res = err(...);
 warn "ERROR $res->[0]: $res->[1]";

die_err(...)

This is a shortcut for:

 $res = err(...);
 die "ERROR $res->[0]: $res->[1]";

What if I want to put result ($res->[2]) into my result with err()?

You can do something like this:

 my $err = err(...) if ERROR_CONDITION;
 $err->[2] = SOME_RESULT;
 return $err;