开源软件名称（OpenSource Name）：

gvvaughan/lyaml

开源软件地址(OpenSource Url)：

https://github.com/gvvaughan/lyaml

开源编程语言(OpenSource Language)：

Lua 64.7%

开源软件介绍(OpenSource Introduction)：

LYAML

LibYAML binding for Lua, with a fast C implementation for converting between %YAML 1.1 and Lua tables, and a low-level YAML event parser for implementing more intricate YAML document loading.

Usage

High Level API

These functions quickly convert back and forth between Lua tables and %YAML 1.1 format strings.

local lyaml   = require "lyaml"
local t       = lyaml.load (YAML-STRING, [OPTS-TABLE])
local yamlstr = lyaml.dump (LUA-TABLE, [OPTS-TABLE])
local null    = lyaml.null ()

`lyaml.load`

lyaml.load accepts a YAML string for parsing. If the YAML string contains multiple documents, only the first document will be returned by default. To return multiple documents as a table, set all = true in the second argument OPTS-TABLE.

lyaml.load("foo: bar")
--> { foo = "bar" }

lyaml.load("foo: bar", { all = true })
--> { { foo = "bar" } }

multi_doc_yaml = [[
---
one
...
---
two
...
]]

lyaml.load(multi_doc_yaml)
--> "one"

lyaml.load(multi_doc_yaml, { all = true })
--> { "one", "two" }

You can supply an alternative function for converting implicit plain scalar values in the implicit_scalar field of the OPTS-TABLE argument; otherwise a default is composed from the functions in the lyaml.implicit module.

You can also supply an alternative table for coverting explicitly tagged scalar values in the explicit_scalar field of the OPTS-TABLE argument; otherwise all supported tags are parsed by default using the functions from the lyaml.explicit module.

`lyaml.dump`

lyaml.dump accepts a table of values to dump. Each value in the table represents a single YAML document. To dump a table of lua values this means the table must be wrapped in another table (the outer table represents the YAML documents, the inner table is the single document table to dump).

lyaml.dump({ { foo = "bar" } })
--> ---
--> foo: bar
--> ...

lyaml.dump({ "one", "two" })
--> --- one
--> ...
--> --- two
--> ...

If you need to round-trip load a dumped document, and you used a custom function for converting implicit scalars, then you should pass that same function in the implicit_scalar field of the OPTS-TABLE argument to lyaml.dump so that it can quote strings that might otherwise be implicitly converted on reload.

Nil Values

Lua tables treat nil valued keys as if they were not there, where YAML explicitly supports null values (and keys!). Lyaml will retain YAML null values as lyaml.null () by default, though it is straight forward to wrap the low level APIs to use nil, subject to the usual caveats of how nil values work in Lua tables.

Low Level APIs

local emitter = require ("yaml").emitter ()

emitter.emit {type = "STREAM_START"}
for _, event in ipairs (event_list) do
  emitter.emit (event)
end
str = emitter.emit {type = "STREAM_END"}

The yaml.emitter function returns an emitter object that has a single emit function, which you call with event tables, the last STREAM_END event returns a string formatted as a YAML 1.1 document.

local iter = require ("yaml").scanner (YAML-STRING)

for token_table in iter () do
  -- process token table
end

Each time the iterator returned by scanner is called, it returns a table describing the next token of YAML-STRING. See LibYAML's yaml.h for details of the contents and semantics of the various tokens produced by yaml_parser_scan, the underlying call made by the iterator.

LibYAML implements a fast parser in C using yaml_parser_scan, which is also bound to lyaml, and easier to use than the token API above:

local iter = require ("yaml").parser (YAML-STRING)

for event_table in iter () do
  -- process event table
end

Each time the iterator returned by parser is called, it returns a table describing the next event from the "Parse" process of the "Parse, Compose, Construct" processing model described in the YAML 1.1 specification using LibYAML.

Implementing the remaining "Compose" and "Construct" processes in Lua is left as an exercise for the reader -- though, unlike the high-level API, lyaml.parser exposes all details of the input stream events, such as line and column numbers.

Installation

There's no need to download an lyaml release, or clone the git repo, unless you want to modify the code. If you use LuaRocks, you can use it to install the latest release from its repository:

luarocks --server=http://rocks.moonscript.org install lyaml

Or from the rockspec in a release tarball:

luarocks make lyaml-?-1.rockspec

To install current git master from GitHub (for testing):

luarocks install http://raw.github.com/gvvaughan/lyaml/master/lyaml-git-1.rockspec

To install without LuaRocks, clone the sources from the repository, and then run the following commands:

cd lyaml
build-aux/luke LYAML_DIR=LIBYAML-INSTALL-PREFIX
sudo build-aux/luke PREFIX=LYAML-INSTALL-PREFIX install
specl -v1freport spec/*_spec.yaml

The dependencies are listed in the dependencies entry of the file rockspec.

Bug reports and code contributions

This library is maintained by its users.

Please make bug reports and suggestions as GitHub Issues. Pull requests are especially appreciated.

But first, please check that your issue has not already been reported by someone else, and that it is not already fixed by master in preparation for the next release (see Installation section above for how to temporarily install master with LuaRocks).

There is no strict coding style, but please bear in mind the following points when proposing changes:

Follow existing code. There are a lot of useful patterns and avoided traps there.
3-character indentation using SPACES in Lua sources: It makes rogue TABs easier to see, and lines up nicely with 'if' and 'end' keywords.
Simple strings are easiest to type using single-quote delimiters, saving double-quotes for where a string contains apostrophes.
Save horizontal space by only using SPACEs where the parser requires them.
Use vertical space to separate out compound statements to help the coverage reports discover untested lines.
Prefer explicit string function calls over object methods, to mitigate issues with monkey-patching in caller environment.