gitoe 0.1.0

data/lib/gitoe/httpserver/repos.rb

@@ -0,0 +1,71 @@
+# rack app for repos
+require "gitoe"
+require "gitoe/repo/rugged"
+require "sinatra"
+require "active_support"
+
+module Gitoe::HTTPServer
+  class Repos < ::Sinatra::Base
+
+    Repo = ::Gitoe::Repo::RestfulRugged
+
+    set :environment, :production
+
+    def json reply
+      content_type 'application/json'
+      ::ActiveSupport::JSON.encode(reply)
+    end
+
+    error do
+      json \
+        :error_message => env['sinatra.error'],
+        :backtrace   => env['sinatra.error'].backtrace
+    end
+
+    # index
+    get "/" do
+      json Repo.instances[:by_arg]
+    end
+
+    # create
+    post "/new" do
+      # create instance if not existing
+      # and return id
+      path = params["path"] or raise "path not specified"
+      repo_id = Repo.id_for(path)
+      repo = Repo.find repo_id
+      json \
+        :id   => repo_id ,
+        :path => repo.path
+    end
+
+    # show
+    get "/:repo_id/?" do
+      repo = Repo.find params["repo_id"]
+      json repo.status
+    end
+
+    # namespace under /:repo_id/:resource
+    Resources = Set[ 'commits', 'commit' ].freeze
+    get "/:repo_id/**" do
+
+      repo = Repo.find( params["repo_id"] )
+
+      sub_str = params[:splat].last # "**" part
+
+      resource, url = sub_str.split('/',2)
+      raise "invalid resource '#{resource}'" unless Resources.include? resource
+      query_hash = env['rack.request.query_hash']
+      # /1/commits/aaaaaaa/b/c/d ? a=1 & b=1
+      # => {
+      #   repo_id: "1",
+      #   resource: "commits"
+      #   arg: "aaaaaaa/b/c/d"
+      #   query_hash:
+      # }
+      # json resource => repo.send(resource.to_sym, arg, env[])
+      json repo.send(resource, url, query_hash )
+    end
+
+  end
+end

data/lib/gitoe/httpserver/static.rb

@@ -0,0 +1,17 @@
+# rack app for static files
+require "gitoe"
+require "sinatra"
+
+module Gitoe::HTTPServer
+  class Static < ::Sinatra::Base
+
+    set :app_file, __FILE__
+    set :environment, :production
+    set :static_cache_control, [:public, max_age: 3600]
+
+    get "/" do
+      send_file File.join( settings.public_folder, "index.html" )
+    end
+
+  end
+end

data/lib/gitoe/repo/repo.rb

@@ -0,0 +1,174 @@
+require "gitoe"
+
+module Gitoe::Repo
+
+  # instance methods, enable querying of instances
+  module Query
+
+    def instances
+      @instances ||= {
+        :by_arg => {}, # arg => id
+        :by_id  => {}, # id  => instance
+      }
+    end
+
+    def create arg
+      id  = instances[:by_id].size
+      instances[:by_id][id] = self.new arg
+      instances[:by_arg][arg] = id
+    end
+
+    def destroy id
+      raise NotImplementedError # TODO
+    end
+
+    def find id
+      id = Integer(id) unless id.is_a? Integer
+      instances[:by_id][id] or raise "repo not found"
+    end
+
+    def id_for arg
+      existing = instances[:by_arg][arg]
+      if existing
+        existing
+      else
+        create arg
+      end
+    end
+  end
+
+  module RestfulRepo
+    # public methods( 'resources' ):
+    #   #status
+    #   #commits
+
+    def status
+      {
+        :refs           => refs         ,
+        :path           => path         ,
+        :cached_commits => size_of_cache,
+      }
+    end
+
+    def commits start, options={}
+      # start :
+      #   commits to start at,
+      #   possible multiple, ','
+      # options :
+      #   limit
+      limit = (options['limit'] || 1000).to_i
+
+      to_query = start.split ','
+      queried = {}
+
+      while to_query.size > 0 and queried.size < limit
+
+        another = to_query.shift
+        next if queried.has_key?(another)
+
+        parent = commit another
+        queried[ parent[:sha1] ] ||= parent
+        parent[:parents].each do |p|
+          next if queried.has_key? p
+          to_query.push p
+        end
+      end
+      $gitoe_log[ "gathered #{queried.size} from #{start}" ]
+
+      queried
+    end
+
+  end
+
+  module Cache
+    def initialize *args
+      init_cache
+      super
+    end
+
+    private
+
+    def ref name
+      follow_ref super
+    end
+
+    def follow_ref ref_hash
+      ref_hash
+    end
+
+    def commit sha1, options={}
+      # if in_cache?
+      #   just take from cache
+      # else
+      #   add sha1 and all its ancestors to cache, in topological order
+      if in_cache? sha1
+        return from_cache sha1
+      end
+      this = super(sha1).freeze
+      if not in_cache? this[:sha1]
+        add_to_cache this[:sha1], this
+      end
+      this
+    end
+
+    private
+
+    def add_to_cache sha1, content
+      if $gitoe_debug and in_cache?(sha1)
+        raise "#{sha1} already in cache"
+      end
+      $gitoe_log[ "add #{sha1} to cache" ]
+      @cached_commits[sha1] = content
+    end
+
+    def init_cache
+      @cached_commits = {} # { sha1:content, order:content }
+    end
+
+    def size_of_cache
+      @cached_commits.size
+    end
+
+    def from_cache sha1
+      @cached_commits[sha1] or raise "not in cache"
+    end
+
+    def in_cache? key
+      @cached_commits.has_key? key
+    end
+
+    def sort_topo queried
+
+      sorted = []
+
+      in_degree = Hash.new{|h,k| h[k] = 0 }
+      children = Hash.new{|h,k| h[k] = [] }
+
+      queried.each do |sha1,content|
+        in_degree[sha1]
+        content[:parents].each do |p|
+          in_degree[p]
+          in_degree[sha1] += 1
+          children[p] << sha1
+        end
+      end
+
+      to_remove = in_degree.keys.select{|k| in_degree[k] == 0 }
+      while to_remove.size > 0
+        # $gitoe_log[ "topo-sorting : remaining #{in_degree.size}" ]
+
+        n = to_remove.shift
+        sorted.push n
+        in_degree.delete n
+
+        children[n].each do |c|
+          new_in_degree = in_degree[c] -= 1
+          to_remove.push c if new_in_degree == 0
+        end
+      end
+
+      $gitoe_log[ "topo-sorting DONE" ]
+      return sorted.select{|s| queried.has_key? s }
+    end
+  end
+end

data/lib/gitoe/repo/rugged.rb

@@ -0,0 +1,121 @@
+# repo using rugged ( https://github.com/libgit2/rugged )
+require "gitoe/repo/repo"
+require "rugged"
+
+module Gitoe::Repo
+  class Rugged_backend
+
+    # implements
+    #   path       :: -> String
+    #   commit     :: sha1::String -> { prop:val }
+    #   refs       :: -> { ref_name : reflogs }
+    #   ref        :: String -> reflogs
+
+    # private
+    #   ref_names  :: -> [ String ]
+    #   reflog     :: String -> [ {} ]
+
+    include ::Rugged
+
+    def initialize path
+      @rugged = Repository.new path
+    end
+
+    def path
+      @rugged.path
+    end
+
+    def commit sha1
+      # $gitoe_log[ "query #{sha1}" ]
+      obj = @rugged.lookup(sha1)
+      case obj
+      when Commit
+        commit_to_hash obj
+      else
+        raise "#{obj} is not commit"
+      end
+    end
+
+    def refs
+      Hash[
+        ref_names.map do |ref_name|
+          [ ref_name, ref(ref_name) ]
+        end
+      ]
+    end
+
+    def ref(name)
+      resolved = Reference.lookup(@rugged, name)
+      basic = {
+          :name   => resolved.name     ,
+          :target => resolved.target   ,
+          :type   => resolved.type     ,
+          :log    => reflog( resolved ),
+        }
+      case name
+      when %r{^refs/remotes/}, %r{^refs/heads/}, 'HEAD'
+        extra = {}
+      when %r{^refs/tags/}
+        extra = deref_tag(resolved)
+      else
+        raise "error parsing ref <ref>"
+      end
+      basic.merge(extra).freeze
+    end
+
+    private
+
+    def ref_names
+      @rugged.refs.to_a.each do |ref_name|
+        ref_name.sub! %r{^/} , ""
+      end << "HEAD"
+    end
+
+    def deref_tag tag
+      target = @rugged.lookup(tag.target)
+      case target
+      when Tag
+        {
+          tag_type: 'annotated',
+          real_target: target.target.oid
+        }
+      when Commit
+        {
+          tag_type: 'lightweight'
+        }
+      else
+        raise "don't know #{tag}"
+      end
+    end
+
+    def reflog ref
+      raise "demand Reference" unless ref.is_a? Reference
+      log =
+        begin
+          ref.log
+        rescue
+          []
+        end
+      log.each do |change|
+        change[:committer][:time] = change[:committer][:time].to_i # seconds from epoch
+      end
+    end
+
+    def commit_to_hash commit_obj
+      {
+        :sha1    => commit_obj.oid,
+        :parents => commit_obj.parents.map(&:oid),
+        # :type    => :commit,
+        :author  => commit_obj.author,
+        :committer => commit_obj.committer
+      }.freeze
+    end
+  end
+
+  class RestfulRugged < Rugged_backend
+    extend Query
+    include Cache
+    include RestfulRepo
+    # ancestors: [RestfulRugged, RestfulRepo, Cache, Rugged_backend]
+  end
+end

data/lib/gitoe/version.rb

@@ -0,0 +1,3 @@
+module Gitoe
+  VERSION = "0.1.0"
+end

data/nanoc.yaml

@@ -0,0 +1,77 @@
+# A list of file extensions that nanoc will consider to be textual rather than
+# binary. If an item with an extension not in this list is found,  the file
+# will be considered as binary.
+text_extensions: [ 'coffee', 'haml', 'sass' ]
+
+# The path to the directory where all generated files will be written to. This
+# can be an absolute path starting with a slash, but it can also be path
+# relative to the site directory.
+output_dir: lib/gitoe/httpserver/public/
+
+# A list of index filenames, i.e. names of files that will be served by a web
+# server when a directory is requested. Usually, index files are named
+# “index.html”, but depending on the web server, this may be something else,
+# such as “default.htm”. This list is used by nanoc to generate pretty URLs.
+index_filenames: [ 'index.html' ]
+
+# Whether or not to generate a diff of the compiled content when compiling a
+# site. The diff will contain the differences between the compiled content
+# before and after the last site compilation.
+enable_output_diff: false
+
+prune:
+  # Whether to automatically remove files not managed by nanoc from the output
+  # directory. For safety reasons, this is turned off by default.
+  auto_prune: true
+
+  # Which files and directories you want to exclude from pruning. If you version
+  # your output directory, you should probably exclude VCS directories such as
+  # .git, .svn etc.
+  exclude: [ '.git', '.hg', '.svn', 'CVS' ]
+
+# The data sources where nanoc loads its data from. This is an array of
+# hashes; each array element represents a single data source. By default,
+# there is only a single data source that reads data from the “content/” and
+# “layout/” directories in the site directory.
+data_sources:
+  -
+    # The type is the identifier of the data source. By default, this will be
+    # `filesystem_unified`.
+    type: filesystem_unified
+
+    # The path where items should be mounted (comparable to mount points in
+    # Unix-like systems). This is “/” by default, meaning that items will have
+    # “/” prefixed to their identifiers. If the items root were “/en/”
+    # instead, an item at content/about.html would have an identifier of
+    # “/en/about/” instead of just “/about/”.
+    items_root: /
+
+    # The path where layouts should be mounted. The layouts root behaves the
+    # same as the items root, but applies to layouts rather than items.
+    layouts_root: /
+
+    # Whether to allow periods in identifiers. When turned off, everything
+    # past the first period is considered to be the extension, and when
+    # turned on, only the characters past the last period are considered to
+    # be the extension. For example,  a file named “content/about.html.erb”
+    # will have the identifier “/about/” when turned off, but when turned on
+    # it will become “/about.html/” instead.
+    allow_periods_in_identifiers: true
+
+# Configuration for the “watch” command, which watches a site for changes and
+# recompiles if necessary.
+watcher:
+  # A list of directories to watch for changes. When editing this, make sure
+  # that the “output/” and “tmp/” directories are _not_ included in this list,
+  # because recompiling the site will cause these directories to change, which
+  # will cause the site to be recompiled, which will cause these directories
+  # to change, which will cause the site to be recompiled again, and so on.
+  dirs_to_watch: [ 'content' ]
+
+  # A list of single files to watch for changes. As mentioned above, don’t put
+  # any files from the “output/” or “tmp/” directories in here.
+  files_to_watch: [ 'nanoc.yaml', 'Rules' ]
+
+  # When to send notifications (using Growl or notify-send).
+  notify_on_compilation_success: true
+  notify_on_compilation_failure: true