giblish 0.3.1 → 0.4.0

data/lib/giblish/indexheadings.rb

@@ -0,0 +1,163 @@
+require "json"
+require "pathname"
+require "asciidoctor"
+require "asciidoctor/extensions"
+require_relative "./utils"
+
+module Giblish
+
+  # This hook is called by Asciidoctor once for each document _before_
+  # Asciidoctor processes the adoc content.
+  #
+  # It indexes all headings found in all documents in the tree.
+  # The resulting index can be serialized to a JSON file file
+  # with the following format:
+  #
+  # {
+  #   file_infos : [{
+  #     filepath : filepath_1,
+  #     title : Title,
+  #     sections : [{
+  #       id : section_id_1,
+  #       title : section_title_1,
+  #       line_no : line_no
+  #     },
+  #     {
+  #       id : section_id_1,
+  #       title : section_title_1,
+  #       line_no : line_no
+  #     },
+  #     ...
+  #     ]
+  #   },
+  #   {
+  #     filepath : filepath_1,
+  #     ...
+  #   }]
+  # }
+  class IndexHeadings < Asciidoctor::Extensions::Preprocessor
+
+    # Use a class-global heading_index dict since asciidoctor creates a new instance
+    # of this class for each processed file
+    @heading_index = {"file_infos" => []}
+
+    class << self
+      def heading_index
+        @heading_index
+      end
+
+      def clear_index
+        @heading_index = {"file_infos" => []}
+      end
+
+      # write the index to a file in dst_dir and remove the base_dir
+      # part of the path for each filename
+      def serialize(dst_dir, base_dir = "")
+        dst_dir = Pathname.new(dst_dir) unless dst_dir.respond_to?(:join)
+        base_dir = Pathname.new(base_dir) unless base_dir.respond_to?(:join)
+
+        if base_dir.to_s.empty?
+          heading_index
+        else
+          # remove the base_dir part of the file path
+          heading_index["file_infos"].each do |file_info|
+            file_info["filepath"] = Pathname.new(file_info["filepath"])
+                                        .relative_path_from(base_dir)
+          end
+        end
+
+        Giblog.logger.info { "writing json to #{dst_dir.join("heading_index.json").to_s}" }
+        File.open(dst_dir.join("heading_index.json").to_s,"w") do |f|
+          f.write(heading_index.to_json)
+        end
+      end
+    end
+
+    def process(document, reader)
+      # Add doc as a source dependency for doc ids
+      src_path = document.attributes["docfile"]
+
+      # Note: the nil check is there to prevent us adding generated
+      # asciidoc docs that does not exist in the file system (e.g. the
+      # generated index pages). This is a bit hackish and should maybe be
+      # done differently
+      return if src_path.nil?
+
+      # get the title from thw raw text (asciidoctor has not yet
+      # processed the text)
+      title = find_title reader.lines
+
+      # Index all headings in the doc
+      Giblog.logger.debug { "indexing headings in #{src_path}" }
+      sections = []
+      file_info_hash = {
+          "filepath" => src_path,
+          "title" => title,
+          "sections" => sections
+      }
+
+      # build the 'sections' array
+      line_no = 1
+      regex = Regexp.new(/^=+\s+(.*)$/)
+      reader.lines.each do |line|
+        m = regex.match(line)
+        if m
+          # We found a heading, index it
+          section = { "id" => get_unique_id(file_info_hash, m[1]) }
+          section["title"] = m[1].strip
+          section["line_no"] = line_no
+          sections << section
+        end
+        line_no += 1
+      end
+
+      heading_index["file_infos"] << file_info_hash
+      reader
+    end
+
+    private
+
+    def find_title(lines)
+      title = "No title Found!"
+      Giblish.process_header_lines(lines) do |line|
+        m = /^=+(.*)$/.match(line)
+        if m
+          # We found a decent title
+          title = m[1].strip
+        end
+      end
+      title
+    end
+
+    def get_unique_id(doc_heading_dict, heading_str)
+      id_base = Giblish.to_valid_id(heading_str)
+      return id_base if !doc_heading_dict.key? id_base
+
+      # handle the case with several sections with the same name
+      idx = 1
+      heading_id = ""
+      loop do
+        idx += 1
+        heading_id = "#{id_base}_#{idx}"
+        # some code here
+        break unless doc_heading_dict.key? heading_id
+      end
+      return heading_id
+    end
+
+    # Helper method to shorten calls to the heading_index from instance methods
+    def heading_index
+      self.class.heading_index
+    end
+  end
+
+
+  # Helper method to register the docid preprocessor extension with
+  # the asciidoctor engine.
+  def register_index_heading_extension
+    Asciidoctor::Extensions.register do
+      preprocessor IndexHeadings
+    end
+  end
+  module_function :register_index_heading_extension
+end

data/lib/giblish/utils.rb

@@ -38,12 +38,16 @@ module Giblish
     #            tree
     # resource_dir - a string or pathname with the directory containing
     #                resources
-    def initialize(src_root, dst_root, resource_dir = nil)
+    def initialize(src_root, dst_root, resource_dir = nil, web_root = false)
       # Make sure that the source root exists in the file system
       @src_root_abs = Pathname.new(src_root).realpath
       self.dst_root_abs = dst_root
+
       # Make sure that the resource dir exists if user gives a path to it
       resource_dir && (@resource_dir_abs = Pathname.new(resource_dir).realpath)
+
+      # Set web root if given by user
+      @web_root_abs = web_root ? Pathname.new(web_root) : nil
     end
 
     def dst_root_abs=(dst_root)
@@ -67,6 +71,13 @@ module Giblish
       src_abs.relative_path_from(@src_root_abs)
     end
 
+    def reldir_from_web_root(in_path)
+      p = in_path.is_a?(Pathname) ? in_path : Pathname.new(in_path)
+      return p if @web_root_abs.nil?
+
+      p.relative_path_from(@web_root_abs)
+    end
+
     def adoc_output_file(infile_path, extension)
       # Get absolute source dir path
       src_dir_abs = self.class.closest_dir infile_path
@@ -152,6 +163,50 @@ module Giblish
     end
   end
 
+  # Helper method that provides the user with a way of processing only the
+  # lines within the asciidoc header block.
+  # The user must return nil to get the next line.
+  #
+  # ex:
+  # process_header_lines(file_path) do |line|
+  #   if line == "Quack!"
+  #      puts "Donald!"
+  #      1
+  #   else
+  #      nil
+  #   end
+  # end
+  def process_header_lines(lines)
+    state = "before_header"
+    lines.each do |line|
+      case state
+      when "before_header" then (state = "in_header" if line =~ /^[=+]^.*$/ || yield(line))
+      when "in_header" then (state = "done" if line =~ /^\s*$/ || yield(line))
+      when "done" then break
+      end
+    end
+  end
+  module_function :process_header_lines
+
+  # Helper method that provides the user with a way of processing only the
+  # lines within the asciidoc header block.
+  # The user must return nil to get the next line.
+  #
+  # ex:
+  # process_header_lines_from_file(file_path) do |line|
+  #   if line == "Quack!"
+  #      puts "Donald!"
+  #      1
+  #   else
+  #      nil
+  #   end
+  # end
+  def process_header_lines_from_file(path)
+    lines = File.readlines(path)
+    process_header_lines(lines,&Proc.new)
+  end
+  module_function :process_header_lines_from_file
+
   # runs the supplied block but redirect stderr to a string
   # returns the string containing stderr contents
   def with_captured_stderr
@@ -166,8 +221,9 @@ module Giblish
 
   # transforms strings to valid asciidoctor id strings
   def to_valid_id(input_str)
-    id_str = "_#{input_str.downcase}"
-    id_str.gsub(%r{[^a-z0-9]+},"_")
+    id_str = "_#{input_str.strip.downcase}"
+    id_str.gsub!(%r{[^a-z0-9]+},"_")
+    id_str.chomp('_')
   end
   module_function :to_valid_id
 

data/lib/giblish/version.rb

@@ -1,3 +1,3 @@
 module Giblish
-  VERSION = "0.3.1".freeze
+  VERSION = "0.4.0".freeze
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: giblish
 version: !ruby/object:Gem::Version
-  version: 0.3.1
+  version: 0.4.0
 platform: ruby
 authors:
 - Anders Rillbert
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2019-02-11 00:00:00.000000000 Z
+date: 2019-03-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -61,7 +61,7 @@ dependencies:
         version: '1.5'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 1.5.7.1
+        version: 1.5.8
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -71,7 +71,7 @@ dependencies:
         version: '1.5'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 1.5.7.1
+        version: 1.5.8
 - !ruby/object:Gem::Dependency
   name: asciidoctor-diagram
   requirement: !ruby/object:Gem::Requirement
@@ -169,6 +169,7 @@ files:
 - data/testdocs/wellformed/source_highlighting/highlight_source.adoc
 - exe/giblish
 - giblish.gemspec
+- lib/giblish-search.rb
 - lib/giblish.rb
 - lib/giblish/application.rb
 - lib/giblish/buildgraph.rb
@@ -179,6 +180,7 @@ files:
 - lib/giblish/docid.rb
 - lib/giblish/docinfo.rb
 - lib/giblish/gititf.rb
+- lib/giblish/indexheadings.rb
 - lib/giblish/pathtree.rb
 - lib/giblish/utils.rb
 - lib/giblish/version.rb
@@ -210,8 +212,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.7.6
+rubygems_version: 3.0.1
 signing_key: 
 specification_version: 4
 summary: A tool for publishing asciidoc docs stored in git repos