简体   繁体   English

Ruby on Rails - 如何使用以点开头的键生成 YAML 文件?

[英]Ruby on Rails - How to generate a YAML file with a key that starts with a dot?

I would like generate on Ruby on Rails a YAML file with a key that start with a dot ( . before_script_template ), like我想在 Ruby on Rails 上生成一个 YAML 文件,其中的键以点( . before_script_template )开头,例如

.before_script_template:
  before_script:
    - gem install bundler

but I am generating a string like ".before_script_template":但我正在生成一个类似".before_script_template":

".before_script_template":
  before_script:
  - gem install bundler

This is my code in Ruby to generate .yml file:这是我在 Ruby 中生成.yml文件的代码:

  pipeline_file = { 
                    ".before_script_template" => {"before_script" => ["gem install bundler"]}
                  }

  File.open("./my_project/folder/.gitlab-ci-template.yml", "r+") do |f|
    f.write(pipeline_file.to_yaml.gsub(/^---$/, ""))
  end

How can I generate a key like .before_script_template: instead of ".before_script_template": ?我怎样才能生成像.before_script_template:而不是".before_script_template":这样的密钥?

Both YAML docs are equivalent.两个 YAML 文档是等效的。 One is simply using quotes to clarify the key.一种是简单地使用引号来阐明关键。

require 'yaml'

# {".before_script_template"=>{"before_script"=>["gem install bundler"]}}
p YAML.load(<<-END)
".before_script_template":
  before_script:
  - gem install bundler
END

# {".before_script_template"=>{"before_script"=>["gem install bundler"]}}
p YAML.load(<<-END)
.before_script_template:
  before_script:
  - gem install bundler
END

This has nothing to do with YAML specifically as both .before_script_template and ".before_script_template" are valid tags in YAML.这与 YAML 无关,因为.before_script_template".before_script_template"都是 YAML 中的有效标签。

This has to do with the implementation determined by the parser/emitter ( Psych in this case).这与解析器/发射器(在本例中为Psych )确定的实现有关。

Specifically this is shown in Psych::Visitors::YAMLTree class in the method visit_String具体来说,这显示在方法visit_StringPsych::Visitors::YAMLTree类中

As far as the quoting style goes this code boils down to:就引用风格而言,这段代码归结为:

 PLAIN         = 1
 SINGLE_QUOTED = 2
 DOUBLE_QUOTED = 3
 LITERAL       = 4
 FOLDED        = 5

def quoting_style(o)
  style = PLAIN
  if o.encoding == Encoding::ASCII_8BIT && !o.ascii_only?
    style = LITERAL
  elsif o =~ /\n(?!\Z)/  
    style = LITERAL
  elsif o == '<<'
    style = SINGLE_QUOTED
  elsif o == 'y' || o == 'n'
    style = DOUBLE_QUOTED
  elsif @line_width && o.length > @line_width
    style = FOLDED
  elsif o =~ /^[^[:word:]][^"]*$/
    style = DOUBLE_QUOTED
 # elsif not String === @ss.tokenize(o) or /\A0[0-7]*[89]/ =~ o
 # commented out so you can see it but this question is not about the ScalarScanner
 # (https://github.com/ruby/psych/blob/v5.0.0/lib/psych/scalar_scanner.rb)
 #   style = SINGLE_QUOTED
  end
  style
end 

quoting_style("before_script") #=> 1
quoting_style(".before_script_template") #=> 3

In this case DOUBLE_QUOTED is utilized because:在这种情况下,使用DOUBLE_QUOTED是因为:

".before_script_template" =~ /^[^[:word:]][^"]*$/

Example例子

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM