augusts_fancy_blog_post_parser 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 99d7da55d26f476768aa953faba2cfbd46df0db9
+  data.tar.gz: e240d0f31fc833d20c9711294fe66291d989eebd
+SHA512:
+  metadata.gz: 389ec05498d9c781fc24dfeaa4afced9a8a20d2df0eb0c46f13bca4f5cbd6df6b669193d6047bdad20bc927b9e3c3a7a66b9aba3ca0d92ef66730359a1f331c1
+  data.tar.gz: a5a18957a04d413e1481a00f89bdcd1572833be1630e40d98e643cd3ca3b1caf6f8dc1529fc6ab3ed73334e208f0c031cd412098b98f69becd8432178dbf1a77

data/augusts_fancy_blog_post_parser.gemspec

@@ -0,0 +1,13 @@
+Gem::Specification.new do |s|
+  s.name = "augusts_fancy_blog_post_parser"
+  s.version = "0.1.0"
+  s.authors = ["August Lilleaas"]
+  s.email = "august@augustl.com"
+  s.summary = "August's fancy blog post parser"
+  s.add_runtime_dependency "nokogiri", "~> 1.5.0"
+  s.add_runtime_dependency "parallel", "~> 0.7.0"
+  s.files = [
+    "lib/augusts_fancy_blog_post_parser.rb",
+    "augusts_fancy_blog_post_parser.gemspec"
+  ]
+end

data/lib/augusts_fancy_blog_post_parser.rb

@@ -0,0 +1,124 @@
+require "nokogiri"
+require "parallel"
+require "open3"
+
+class AugustsFancyBlogPostParser
+  attr_reader :url
+  attr_reader :body
+
+  def self.escape_html(html)
+    html
+      .gsub(/&(?!\w+;)/, '&')
+      .gsub(/</, '&lt;')
+      .gsub(/>/, '&gt;')
+      .gsub(/"/, '&quot;')
+  end
+
+  def self.parse_html_attrs(attrs)
+    if attrs.blank?
+      return {}
+    else
+      Nokogiri::HTML("<div#{attrs}></div>").css("div")[0].attributes.inject({}) do |prev, curr|
+        key = curr[0]
+        value = curr[1]
+        prev[key] = value.value
+        prev
+      end
+    end
+  end
+
+  def self.format_code(code, lang)
+    if lang
+      formatted = nil
+      Open3.popen3("pygmentize", "-l", lang, "-f", "html", "-P", "nowrap=true") do |stdin, stdout, stderr, wait_thr|
+        stdin.write(code)
+        stdin.close_write
+
+        lines = []
+        until (line = stdout.gets).nil?
+          lines.push(line)
+        end
+
+        formatted = lines.join("").rstrip
+      end
+
+      "<code class=\"highlight\">#{formatted}</code>"
+    else
+      "<code>" + escape_html(code) + "</code>"
+    end
+  end
+
+  def self.separate_on(str, re, result = [])
+    match_data = str.match(re)
+    if match_data
+      result.push(match_data.pre_match)
+      result.push(match_data)
+      separate_on(match_data.post_match, re, result)
+    else
+      result.push(str)
+      result
+    end
+  end
+
+  def initialize(cwd, path)
+    @path = path
+    @cwd = cwd
+    @url = "/" + @path[0...-(File.extname(@path).length)]
+    reload!
+  end
+
+  def self.parse_body(body)
+    chunks = separate_on(body, /\<code(.*?)\>(.*?)\<\/code\>/m)
+    Parallel.map(chunks, :in_threads => chunks.length) do |chunk|
+      case chunk
+      when String
+        chunk
+      when MatchData
+        attrs = parse_html_attrs(chunk[1])
+        code = chunk[2]
+        format_code(code, attrs["data-lang"])
+      else
+        raise ArgumentError.new("Unexpected chunk #{chunk.class}.")
+      end
+    end.join("")
+  end
+
+  def reload!
+    html = File.read(@cwd + "/" + @path)
+
+    @headers = {}
+    num_header_chars = 0
+
+    html.each_line.each do |line|
+      if line =~ /[a-z]+\:./
+        key, _, value = *line.partition(":")
+        @headers[key] = value.chomp
+        num_header_chars += line.length
+      else
+        break
+      end
+    end
+
+    @body = self.class.parse_body(html[num_header_chars..-1])
+  end
+
+  def id
+    url.gsub("/", ":")
+  end
+
+  def title
+    @headers["title"].strip
+  end
+
+  def timestamp
+    Time.parse(@headers["date"])
+  end
+
+  def html_date
+    timestamp.strftime("%Y-%m-%d")
+  end
+
+  def display_date
+    timestamp.strftime("%B %d, %Y")
+  end
+end

metadata

@@ -0,0 +1,72 @@
+--- !ruby/object:Gem::Specification
+name: augusts_fancy_blog_post_parser
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- August Lilleaas
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2013-12-30 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.5.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.5.0
+- !ruby/object:Gem::Dependency
+  name: parallel
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 0.7.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 0.7.0
+description: 
+email: august@augustl.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/augusts_fancy_blog_post_parser.rb
+- augusts_fancy_blog_post_parser.gemspec
+homepage: 
+licenses: []
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.0.14
+signing_key: 
+specification_version: 4
+summary: August's fancy blog post parser
+test_files: []