nanoc3 3.0.9 → 3.1.0a1

data/LICENSE

@@ -1,4 +1,4 @@
-Copyright (c) 2007-2010 Denis Defreyne and contributors
+Copyright (c) 2007-2009 Denis Defreyne and contributors
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

data/NEWS.md

@@ -0,0 +1,360 @@
+# nanoc News
+
+## 3.1
+
+New:
+
+* An `Item#rep(name)` function for quickly getting a certain rep
+* An `Item#compiled_content` function for quickly getting compiled content
+* An `Item#path` function for quickly getting the path of an item rep
+* A new “+” wildcard in rule patterns that matches one or more characters
+* A `view` command that starts a web server in the output directory
+* A `debug` command that shows information about the items, reps and layouts
+* A `kramdown` filter ([kramdown site](http://kramdown.rubyforge.org/))
+* A plugin-loading system
+* A diff between the previously compiled content and the last compiled content
+  is now written to `output.diff`
+
+Changed:
+
+* New sites now come with a stylesheet item instead of a `style.css` file in
+  the output directory
+* The `deploy:rsync` task now use sensible default options
+* The `deploy:rsync` task now accepts a config environment variable
+* The `deploy:rsync` task now uses a lowercase `dry_run` environment variable
+* The `maruku` filter now accepts parameters
+* The `rainpress` filter now accepts parameters
+* The `filesystem` data source is now known as `filesystem_verbose`
+* Meta files and content files are now optional
+* The `filesystem_compact` and `filesystem_combined` data sources have been
+  merged into a new `filesystem_unified` data source
+* The metadata section in `filesystem_unified` is now optional [Christopher
+  Eppstein]
+* The `--server` autocompile option is now known as `--handler`
+* Assigns in filters are now available as instance variables and methods
+* The `#breadcrumbs_trail` function now allows missing parents
+
+Deprecated:
+
+* `Nanoc3::FileProxy`; use one of the filename attributes instead
+* `ItemRep#content_at_snapshot`; use `#compiled_content` instead
+* The `last_fm`, `delicious` and `twitter` data sources; fetch online content
+  into a cache by a rake task and load data from this cache instead
+
+## 3.0.7
+
+* Fixed a bug which could cause layout rules not be matched in order
+
+## 3.0.6
+
+* Error checking in `filesystem_combined` has been improved [Brian Candler]
+* Generated HTML files now have a default encoding of UTF-8
+* Periods in identifiers for layouts now behave correctly
+* The `relativize_paths` filter now correctly handles “/” [Eric Sunshine]
+
+## 3.0.5
+
+* Restored pre-3.0.3 behaviour of periods in identifiers. By default, a file
+  can have multiple extensions (e.g. `content/foo.html.erb` will have the
+  identifier `/foo/`), but if `allow_periods_in_identifiers` in the site
+  configuration is true, a file can have only one extension (e.g.
+  `content/blog/stuff.entry.html` will have the identifier
+  `/blog/stuff.entry/`).
+
+## 3.0.4
+
+* Fixed a bug which would cause the `filesystem_compact` data source to
+  incorrectly determine the content filename, leading to weird “Expected 1
+  content file but found 3” errors [Eric Sunshine]
+
+## 3.0.3
+
+* The `Blogging` helper now properly handles item reps without paths
+* The `relativize_paths` filter now only operates inside tags
+* The autocompiler now handles escaped paths
+* The `LinkTo` and `Tagging` helpers now output escaped HTML
+* Fixed `played_at` attribute assignment in the `LastFM` data source for
+  tracks playing now, and added a `now_playing` attribute [Nicky Peeters]
+* The `filesystem_*` data sources can now handle dots in identifiers
+* Required enumerator to make sure `#enum_with_index` always works
+* `Array#stringify_keys` now properly recurses
+
+## 3.0.2
+
+* Children-only identifier patterns no longer erroneously also match parent
+  (e.g.` /foo/*/` no longer matches `/foo/`)
+* The `create_site` command no longer uses those ugly HTML entities
+* Install message now mentions the IRC channel
+
+## 3.0.1
+
+* The proper exception is now raised when no matching compilation rules can
+  be found
+* The autocompile command no longer has a duplicate `--port` option
+* The `#url_for` and `#feed_url` methods now check the presence of the
+  `base_url` site configuration attribute
+* Several outdated URLs are now up-to-date
+* Error handling has been improved in general
+
+## 3.0
+
+New:
+
+* Multiple data sources
+* Dependency tracking between items
+* Filters can now optionally take arguments
+* `#create_page` and `#create_layout` methods in data sources
+* A new way to specify compilation/routing rules using a Rules file
+* A `coderay` filter ([CodeRay site](http://coderay.rubychan.de/))
+* A `filesystem_compact` data source which uses less directories
+
+Changed:
+
+* Pages and textual assets are now known as “items”
+
+Removed:
+
+* Support for drafts
+* Support for binary assets
+* Support for templates
+* Everything that was deprecated in nanoc 2.x
+* `save_*`, `move_*` and `delete_*` methods in data sources
+* Processing instructions in metadata
+
+## 2.2.2
+
+* Removed `relativize_paths` filter; use `relativize_paths_in_html` or
+  `relativize_paths_in_css` instead
+* Fixed bug which could cause nanoc to eat massive amounts of memory when an
+  exception occurs
+* Fixed bug which would cause nanoc to complain about the open file limit
+  being reached when using a large amount of assets
+
+## 2.2.1
+
+* Fixed bug which prevented `relative_path_to` from working
+* Split `relativize_paths` filter into two filter: `relativize_paths_in_html`
+  and `relativize_paths_in_css`
+* Removed bundled mime-types library
+
+## 2.2
+
+New:
+
+* `--pages` and `--assets` compiler options
+* `--no-color` commandline option
+* `Filtering` helper
+* `#relative_path_to` function in `LinkTo` helper
+* `rainpress` filter ([Rainpress site](http://code.google.com/p/rainpress/))
+* `relativize_paths` filter
+* The current layout is now accessible through the `@layout` variable
+* Much more informative stack traces when something goes wrong
+
+Changed:
+
+* The commandline option parser is now a lot more reliable
+* `#atom_feed` now takes optional `:content_proc`, `:excerpt_proc` and
+  `:articles` parameters
+* The compile command show non-written items (those with `skip_output: true`)
+* The compile command compiles everything by default
+* Added `--only-outdated` option to compile only outdated pages
+
+Removed:
+
+* deprecated extension-based code
+
+## 2.1.6
+
+* The `filesystem_combined` data source now supports empty metadata sections
+* The `rdoc` filter now works for both RDoc 1.x and 2.x
+* The autocompiler now serves a 500 when an exception occurs outside
+  compilation
+* The autocompiler no longer serves index files when the request path does not
+  end with a slash
+* The autocompiler now always serves asset content correctly
+
+## 2.1.5
+
+* Added Ruby 1.9 compatibility
+* The `filesystem` and `filesystem_combined` data sources now preserve custom
+  extensions
+
+## 2.1.4
+
+* Fixed an issue where the autocompiler in Windows would serve broken assets
+
+## 2.1.3
+
+* The `haml` and `sass` filters now correctly take their options from assets
+* The autocompiler now serves index files instead of 404s
+* Layouts named “index” are now handled correctly
+* The `filesystem_combined` data source now properly handles assets
+
+## 2.1.2
+
+* The utocompiler now compiles assets as well
+* The `sass` filter now takes options (just like the `haml` filter)
+* Haml/Sass options are now taken from the page *rep* instead of the page
+
+## 2.1.1
+
+* Fixed issue which would cause files not to be required in the right order
+
+## 2.1
+
+This is only a short summary of all changes in 2.1. For details, see the
+[nanoc web site](http://nanoc.stoneship.org/). Especially the blog and the
+updated manual will be useful.
+
+New:
+
+* New `rdiscount` filter ([RDiscount site](http://github.com/rtomayko/rdiscount))
+* New `maruku` filter ([Maruku site](http://maruku.rubyforge.org/))
+* New `erubis` filter ([Erubis site](http://www.kuwata-lab.com/erubis/))
+* A better commandline frontend
+* A new filesystem data source named `filesystem_combined`
+* Routers, which decide where compiled pages should be written to
+* Page/layout mtimes can now be retrieved through `page.mtime`/`layout.mtime`
+
+Changed:
+
+* Already compiled pages will no longer be re-compiled unless outdated
+* Layout processors and filters have been merged
+* Layouts no longer rely on file extensions to determine the layout processor
+* Greatly improved source code documentation
+* Greatly improved unit test suite
+
+Removed:
+
+* Several filters have been removed and replaced by newer filters:
+	* `eruby`: use `erb` or `erubis` instead
+	* `markdown`: use `bluecloth`, `rdiscount` or `maruku` instead
+	* `textile`: use `redcloth` instead
+
+## 2.0.4
+
+* Fixed `default.rb`’s `#html_escape`
+* Updated Haml filter and layout processor so that @page, @pages and @config
+  are now available as instance variables instead of local variables
+
+## 2.0.3
+
+* The autocompiler now honors custom paths
+* The autocompiler now attempts to serve pages with the most appropriate MIME
+  type, instead of always serving everything as `text/html`
+
+## 2.0.2
+
+* nanoc now requires Ruby 1.8.5 instead of 1.8.6
+
+## 2.0.1
+
+* Fixed a “too many open files” error that could appear during (auto)compiling
+
+## 2.0
+
+New:
+
+* Support for custom layout processors
+* Support for custom data sources
+* Database data source
+* An auto-compiler
+* Pages have `parent` and `children`
+
+Changed:
+
+* The source has been restructured and cleaned up a great deal
+* Filters are defined in a different way now
+* The `eruby` filter now uses ERB instead of Erubis
+
+Removed:
+
+* The `filters` property; use `filters_pre` instead
+* Support for Liquid
+
+## 1.6.2
+
+* Fixed an issue which prevented the content capturing plugin from working
+
+## 1.6.1
+
+* Removed a stray debug message
+
+## 1.6
+
+* Added support for post-layout filters
+* Added support for getting a File object for the page, so you can now e.g.
+  easily get the modification time for a given page (`@page.file.mtime`)
+* Cleaned up the source code a lot
+* Removed deprecated asset-copying functionality
+
+## 1.5
+
+* Added support for custom filters
+* Improved Liquid support -- Liquid is now a first-class nanoc citizen
+* Deprecated assets -- use something like rsync instead
+* Added `eruby_engine` option, which can be `erb` or `erubis`
+
+## 1.4
+
+* nanoc now supports ERB (as well as Erubis); Erubis no longer is a dependency
+* `meta.yaml` can now have `haml_options` property, which is passed to Haml
+* Pages can now have a `filename` property, which defaults to `index` [Dennis
+  Sutch]
+* Pages now know in what order they should be compiled, eliminating the need
+  for custom page ordering [Dennis Sutch]
+
+## 1.3.1
+
+* The contents of the `assets` directory are now copied into the output
+  directory specified in `config.yaml`
+
+## 1.3
+
+* The `@pages` array now also contains uncompiled pages
+* Pages with `skip_output` set to true will not be outputted
+* Added new filters
+	* Textile/RedCloth
+	* Sass
+* nanoc now warns before overwriting in `create_site`, `create_page` and
+  `create_template` (but not in compile)
+
+## 1.2
+
+* Sites now have an `assets` directory, whose contents are copied to the
+  `output` directory when compiling [Soryu]
+* Added support for non-eRuby layouts (Markaby, Haml, Liquid, ...)
+* Added more filters (Markaby, Haml, Liquid, RDoc [Dmitry Bilunov])
+* Improved error reporting
+* Accessing page attributes using instance variables, and not through `@page`,
+  is no longer possible
+* Page attributes can now be accessed using dot notation, i.e. `@page.title`
+  as well as `@page[:title]`
+
+## 1.1.3
+
+* Fixed bug which would cause layoutless pages to be outputted incorrectly
+
+## 1.1.2
+
+* Backup files (files ending with a “~”) are now ignored
+* Fixed bug which would cause subpages not to be generated correctly
+
+## 1.1
+
+* Added support for nested layouts
+* Added coloured logging
+* `@page` now hold the page that is currently being processed
+* Index files are now called “content” files and are now named after the
+  directory they are in [Colin Barrett]
+* It is now possible to access `@page` in the page’s content file
+
+## 1.0.1
+
+* Fixed a bug which would cause a “no such template” error to be displayed
+  when the template existed but compiling it would raise an exception
+* Fixed bug which would cause pages not to be sorted by order before compiling
+
+## 1.0
+
+* Initial release

data/README.md

@@ -0,0 +1,85 @@
+# nanoc 3
+
+nanoc is a simple but very flexible static site generator written in Ruby.
+It operates on local files, and therefore does not run on the server. nanoc
+"compiles" the local source files into HTML (usually), by evaluating eRuby,
+Markdown, etc.
+
+## Documentation
+
+The [nanoc3 web site](http://nanoc.stoneship.org) contains a few useful
+resources to help you get started with nanoc:
+
+* The [tutorial](http://nanoc.stoneship.org/tutorial)
+* The [manual](http://nanoc.stoneship.org/manual)
+* The [migration guide](http://nanoc.stoneship.org/migrating)
+
+It is probably also worth checking out and perhaps subscribing to the
+discussion groups:
+
+* The [discussion group in English](http://groups.google.com/group/nanoc)
+* The [discussion group in Spanish](http://groups.google.com/group/nanoc-es)
+
+### Source Code Documentation
+
+The source code is structured in a few directories:
+
+* `bin` contains the commandline tool aptly named `nanoc3`
+* `lib`
+  * `nanoc3`
+    * `base` contains the bare essentials necessary for nanoc to function
+    * `cli` contains the commandline interface
+    * `data_sources` contains the standard data sources (Nanoc3::DataSource
+      subclasses), such as the filesystem data source
+    * `helpers` contains helpers, which provide functionality some sites
+      may find useful, such as the blogging and tagging helpers
+    * `extra` contains stuff that is not needed by nanoc itself, but which may
+      be used by helpers, data sources, filters or VCSes.
+    * `filters` contains the standard filters (Nanoc3::Filter subclasses) such
+      as ERB, Markdown, Haml, ...
+* `test` contains testing code, structured in the same way as lib/nanoc
+
+The namespaces (modules) are organised like this:
+
+* `Nanoc3` is the namespace for everything nanoc-related (obviously). The
+  classes in `lib/nanoc3/base` are part of this module (not `Nanoc3::Base`)
+  * `CLI` containing everything related to the commandline tool.
+  * `DataSources` contains the data sources
+  * `Helpers` contains the helpers
+  * `Extra` contains useful stuff not needed by nanoc itself
+  * `Filters` contains the (textual) filters
+
+The central class in nanoc is `Nanoc3::Site`, so you should start there if
+you want to explore nanoc from a technical perspective.
+
+## Dependencies
+
+nanoc itself can be used without installing any dependencies. Some
+components, however, do have dependencies:
+
+* The **autocompiler** depends on `mime-types` and `rack`.
+* For **documentation generation** you’ll need `yard`.
+* For **packaging** you’ll need `rubygems` (1.3 or newer).
+* For **testing** you’ll need `mocha`.
+
+## Contributors
+
+(In alphabetical order)
+
+* Colin Barrett
+* Dmitry Bilunov
+* Brian Candler
+* Chris Eppstein
+* Starr Horne
+* Nicky Peeters
+* Christian Plessl
+* Šime Ramov
+* "Soryu"
+* Eric Sunshine
+* Dennis Sutch
+
+Special thanks to Ale Muñoz.
+
+## Contact
+
+You can reach me at <denis.defreyne@stoneship.org>.

data/Rakefile

@@ -7,5 +7,5 @@ require 'nanoc3'
 # Load tasks
 Dir.glob('tasks/**/*.rake').each { |r| Rake.application.add_import r }
 
-# Set default talk
-task :default => [ :fetch_dependencies, :test ]
+# Set default task
+task :default => :test

data/bin/nanoc3

@@ -8,9 +8,5 @@ $LOAD_PATH.unshift(File.expand_path(File.dirname(__FILE__) + '/../lib'))
 require 'nanoc3'
 require 'nanoc3/cli'
 
-# Load custom code that can't be load later
-Dir['lib/commands/*.rb'].sort.each     { |f| require f }
-Dir['lib/data_sources/*.rb'].sort.each { |f| require f }
-
 # Run base
 Nanoc3::CLI::Base.shared_base.run(ARGV)

data/lib/nanoc3/base/code_snippet.rb

@@ -7,25 +7,31 @@ module Nanoc3
   # speed up site compilation.
   class CodeSnippet
 
-    # The Nanoc3::Site this code snippet belongs to.
+    # The {Nanoc3::Site} this code snippet belongs to.
+    #
+    # @return [Nanoc3::Site]
     attr_accessor :site
 
     # A string containing the actual code in this code snippet.
+    #
+    # @return [String]
     attr_reader :data
 
     # The filename corresponding to this code snippet.
+    #
+    # @return [String]
     attr_reader :filename
 
     # The time where this code snippet was last modified.
+    #
+    # @return [Time]
     attr_reader :mtime
 
     # Creates a new code snippet.
     #
-    # +data+::     The raw source code which will be executed before compilation.
-    #
-    # +filename+:: The filename corresponding to this code snippet.
-    #
-    # +mtime+::    The time when the code was last modified (can be nil).
+    # @param [String] data     The raw source code which will be executed before compilation
+    # @param [String] filename The filename corresponding to this code snippet
+    # @param [Time]   mtime    The time when the code was last modified (can be nil)
     def initialize(data, filename, mtime=nil)
       @data     = data
       @filename = filename
@@ -33,6 +39,8 @@ module Nanoc3
     end
 
     # Loads the code by executing it.
+    #
+    # @return [void]
     def load
       eval(@data, TOPLEVEL_BINDING, @filename)
     end