qipowl 0.9.3 → 0.9.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 !binary "U0hBMQ==":
-  metadata.gz: b1285e55189abe3129f00711fb21fbc905440796
-  data.tar.gz: 2b46cc8dbd9b675f5293a8ebd3242aee06a2e06b
+  metadata.gz: 7f1c631668f2211baa4e0104c6d9c7d474679f8e
+  data.tar.gz: 4acc209241f865925f2ad27657943b910b611bb0
 !binary "U0hBNTEy":
-  metadata.gz: 84bca4876f808702f0c4d8ae4a28ea46a9319fb7319c855c231d3b27ffadebaafecc85f8c94821d9373e6c8d087a526355900ba0d61a1668ad82e8cfd57ed605
-  data.tar.gz: 045c623b7344d3c84e47beb8240742fb04cabb329007e7f9a14814fec415efe0289381eaa505d7150e144780caf8b518386356f73a38dff112f34d30407256f8
+  metadata.gz: 065e805c8794134debf98cb13f7362184902b05616eb2c6a17af4e7df2d77ec4fa153666a8191418fcf54e622549f819668d191a0be6489d2443f506224a317d
+  data.tar.gz: 9f821d61819a6c2b7d95004dc48d0a1240ec107be804f46a582aa0275ed386676bd0585ec34d249d8dad15e6b04482984470114f8a57204be4bbfb3ff8544e13

data/Gemfile

@@ -4,5 +4,6 @@ source 'https://rubygems.org'
 gemspec
 
 gem 'aquarium', '~> 0.6', :git => 'git://github.com/deanwampler/Aquarium.git'
+gem 'typogrowth', '~> 0.9', :git => 'git://github.com/mudasobwa/typogrowth.git', :branch => 'master'
 
 gem 'rake', :group => :test

data/README.md

@@ -16,17 +16,158 @@ _qipowl_ (pronounced as **keep all**)
 
 ## Intro
 
-_qipowl_ is the next generation parser environment. It’s not the
-library for parsing, rather it is the framework to build extensive
-parsers for virtually every markup anyone may imagine.
-
 The main idea of _qipowl_ is to yield the power of 
 [DSL in Ruby](http://jroller.com/rolsen/entry/building_a_dsl_in_ruby).
 The whole input text is treated neither more nor less than `DSL`. 
 That gives the user an ability to make virtually every term in input text
 the _operating entity_.
 
-## Examples
+## Principles
+
+**Qipowl** is a Ruby parsing library. The parsing is done via
+DSL exactly as [Ouroboros](http://en.wikipedia.org/wiki/Ouroboros)
+eats it’s own tail. 
+
+The whole input is treated as Ruby source code and executed respectively.
+To prevent collisions of input with built-in ruby methods, the ASCII symbols
+in the input are being translated into their
+[fullwidth equivalents](http://en.wikipedia.org/wiki/Halfwidth_and_fullwidth_forms#Chart)
+before execution (and back to ASCII after the parsing is done.)
+
+Let’s say we have a string “Hello world” as input. It became ‘encoded’ into:
+“Ｈｅｌｌｏ ｗｏｒｌｄ”, executed as Ruby code (exactly as e. g. `puts rand`
+would) and finally ‘decoded’ back to ASCII. Whether the parser knows anything
+about ‘Ｈｅｌｌｏ’ or ‘ｗｏｒｌｄ’ it would be executed. Say, we have
+
+    def ｗｏｒｌｄ *args
+      "ｂｒａｖｅ ｎｅｗ #{__callee__}"
+    end
+
+thus the output will be:
+
+    # ⇒ Hello brave new world
+
+More about may be found at [project page](http://rocket-science.ru/qipowl/).
+
+## Applications
+
+**Qipowl** has a wide list of applications. The “markright”, descendant
+of “markup” and “markdown” is presented [here](http://qipowl.herokuapp.com).
+
+**Qipowl HTML** uses extended unicode symbols
+to specify more clean and readable source files and (boom!) ruby DSL to
+interpret them. E.g. the data definitions look like:
+
+    ▶ Data term — definition goes here
+
+Headings:
+
+    §1 This is a second-level heading
+
+Bold and emphasis:
+
+    The following ≡text≡ goes strong and this one is ≈emphasized≈.
+
+Comments are possible as well:
+
+    ✍ FIXME! 
+    Not to forget add this to parsing!
+    ✍
+
+etc.
+
+## Why?
+
+Just because it’s 2013 all around. Unicode came already and those fancy 
+symbols are easily mapped to the keyboard layouts. The brackets, used
+in old good Markdown are ugly, look at how they might be introduced:
+
+    I like Markdown¹http://daringfireball.net/projects/markdown/syntax
+
+Markdown lacks a lot of modern features (properties of text).
+
+Markdown does not provide a blanket set of marks, fully covering 
+claims to markup language.
+
+## Parsing
+
+Parsing is the most sexy part of **Qipowl** bowels, since it’s done
+almost without any external parsing; input files are the ruby scripts
+themselves. WTF? Let me explain.
+
+Let we have an input file of the following structure:
+
+    §1 Qipowl
+
+    ✍ FIXME 
+    include language reference here
+    ✍
+
+    ≡Qipowl≡ is the most exciting ruby DSL application example. As it
+    is stated in markdown reference:
+
+    〉 Readability, however, is emphasized above all else. 
+    A Markdown-formatted document should be publishable as-is, 
+    as plain text, without looking like it’s been marked up with 
+    tags or formatting instructions.
+    — http://daringfireball.net/projects/markdown/syntax
+
+Now we simply give the source to ruby interpreter, which knowns, that
+`§1` is *in fact* ruby function, which transforms that to any other syntax
+we want. To HTML, for instance.
+
+## Parsing problems
+
+Not all the constructions may be passed to ruby script as is. There are
+four exceptions:
+
+- **blockquotes**, which are in fact kinda documents inside documents, because
+they might be nested and they may include any other markup;
+- **images**, **videos**, etc. which may be typed as the hyperlink only;
+- **anchors, abbrs etc.**, the elements which are not “symbol-text” formed.
+They rather are looking like “text-symbol-text” and unfortunately should
+be preparsed to supply correct ruby DSL;
+- **lists and data definitions**, are to be surrounded with `<ul>`/`<dd>` tags;
+- **tables**… Bah, I didn’t think most about tables yet. They are ugly.
+
+### Links
+
+Links might be:
+- **anchors** 
+  - Wiki says¹http://wikipedia.org
+  - Wiki clone¹/wiki
+  - — Wikipedia, http://wikipedia.org   
+the latter may be found in quotations only.
+- **images**
+  - http://localhost/a.png
+  - Best views of Hornsjø¹http://localhost/a.png
+- **videos**
+  - http://youtu.be/SAJ_TzLqy1U
+  - http://www.youtube.com/watch?v=SAJ_TzLqy1U
+
+Abbrs are looking (and processing) mostly like links, but now we may
+forget about them:
+- **abbrs**
+  - Wiki†Best online knowledge base ever†
+
+
+Links are being parsed in the following manner:
+
+- find the link in the input, according to simple pattern `URI.regexp`
+- determine whether it is an image, video or link to page by downloading
+and analyzing the headers
+- TODO copying the image to the host computer, providing the watermark
+with copyright and any other significant information
+- TODO instead of previous two actions we might simply analyze it by extension
+e.g. if there is no internet connection available
+- prepending the link with special character (understood by DSL)
+
+After all is done, we yield smth like `⚐ http://localhost.a.png` in place of
+`http://localhost.a.png` and `⚓ http://localhost/index.html` in place of
+`http://localhost/index.html`
+
+
+### Examples
 
 This chapter should be the last one, but who wants to read technical details
 without any clue of how they might be applied? So, here we go.
@@ -94,11 +235,9 @@ _qipowl_ understands six types of ‘operators’:
 * flush
 * block
 * magnet
-* inplace
-* linewide
-* handshake
-* kiss
-* custom
+* grip
+* regular
+* self
 
 #### :flush
 
@@ -155,14 +294,14 @@ for the markup:
 
     ☎ +1(987)5554321
 
-#### :inplace
+#### :grip
 
 Acts mostly like `:block` but inside one text block (text blocks are
 likely paragraphs, delimited with double carriage returns.) Requires
 closing element. Inplace operators are of highest priority and may
 overlap.
 
-    :inplace
+    :grip
       :≡ : :strong
 
 will convert
@@ -173,12 +312,12 @@ into
 
     That is <strong>bold</strong> text.
 
-#### :linewide
+#### :regular
 
 Those are not require closings, since they are operated on the _rest_ of
 the text. Support nesting by prepending tags with _non-breakable space_:
 
-    :linewide
+    :regular
       :• : li
 
 The following syntax 
@@ -195,53 +334,6 @@ will produce:
     <li>Nested li 2</li></ul>
     <li>Line item 2</li></ul>
 
-#### :handshake
-
-**TODO** rewrite examples for latex
-
-The group contains operators, acting on left and right operands between
-the delimiters given. By default it takes the whole line from `^` till `$`.
-
-    :handshake :
-      :∈ : :mathml
-      :⊂ :
-        :tag  : :mathml
-        :from : '\s'
-        :till : '.'
-        
-The following syntax 
-
-    Let we have A ⊂ ∅. Then the following formula is OK:
-    ∀ a ∈ ∅
-    which is evident, though.
-    
-will produce:
-
-    Let we have <mathml>A ⊂ ∅</mathml>. Then the following formula is OK:
-    <mathml>∀ a ∈ ∅</mathml>
-    which is evident, though.
-
-#### :kiss
-
-Almost the same as `:handshake` but operates on the preceeding/following pair of
-text piece without spaces. E.g.
-
-    :kiss
-      :÷ : :mathml
-
-The following syntax 
-
-    The formula 12 ÷ 5 is simple.
-    
-will produce:
-
-    The formula <mathml>12 ÷ 5</mathml> is simple.
-
-#### :custom
-
-Custom is not yet fully powerful mechanism to make substitutions inplace
-for generic words. Please use on your own risk.
-
 ### Extending
 
 Extending _qipowl_ is as easy as writing a couple of strings in YAML format.
@@ -307,7 +399,7 @@ link to video into embedded frame with video content as by YouTube.
 
 Add this line to your application's Gemfile:
 
-    gem 'typogrowl'
+    gem 'qipowl'
 
 And then execute:
 
@@ -315,25 +407,14 @@ And then execute:
 
 Or install it yourself as:
 
-    $ gem install typogrowl
+    $ gem install qipowl
 
 ## Usage
 
 ```ruby
-require 'typogrowl'
-…
-tg =  Qipowl::Html.new 
-puts tg.parse_and_roll(text)
-```
-
-or even simplier
-
-```ruby
-require 'typogrowl'
+require 'qipowl'
 …
-tg =  Qipowl.tg_md__html # typogrowl markup _and_ markdown
-
-puts tg.parse_and_roll(text)
+result =  Qipowl.parse text # qipowl markup _and_ markdown
 ```
 
 ## Contributing

data/bin/bowler

@@ -11,7 +11,7 @@ OptionParser.new do |opts|
   opts.banner = "Usage: #{$0} FILE|STRING"
 
   # Bowl result?
-  opts.on("-a", "--action ACTION", [:bowl, :ruby, :cmd, :yaml, :html],
+  opts.on("-a", "--action ACTION", [:bowl, :unbowl, :ruby, :cmd, :yaml, :html],
           "Action to apply on input (bowl, html); default: html") do |action|
     options[:action] = action || :bowl
   end
@@ -31,6 +31,8 @@ file_or_string = File.read(file_or_string) if File.exist?(file_or_string)
 case options[:action]
 when :bowl
   puts file_or_string.bowl
+when :unbowl
+  puts file_or_string.unbowl
 else
   puts Qipowl::Html.parse file_or_string
-end  
+end

data/config/bowlers/html.yaml

@@ -11,7 +11,16 @@
 
 :entities       :
   :self         :
-    :qipowl     : :strong
+    :qipowl     :
+      :tag      : :strong
+      :format   : "<a href='http://\\1.github.com'>\\1</a>"
+    :tag        :
+      :format   : :tagger_format
+    :tag2       :
+      :tag      : :b
+      :format   : :tagger_format
+      :synonyms :
+        - :tag3
   :block        :
     :✍             :
       :synonyms :
@@ -29,8 +38,6 @@
     :✉             :
       :tag      : :span
       :class    : :email
-      :synonyms :
-        - :mail
     :✎             : :lj
     :☇              : :a
   :grip         :
@@ -38,11 +45,17 @@
     :≈          : :em
     :↑             : :sup
     :↓             : :small
-    :λ          : :code
+    :λ          :
+      :tag      : :code
+      :synonyms :
+        - :✿_span_fixedfont
     :⚓             : :a
     :†          : :abbr
     :✁             : :del
     :✿_span_nobr : :nobr
+    :✓          :
+      :tag      : :span
+      :class    : :notypo
   :regular      :
     :•          :
       :tag      : :li
@@ -60,7 +73,7 @@
       :tag      : :dt
       :parent   :
         :tag    : :dl
-        :class  : :dl_horizontal
+        :class  : :'dl-horizontal'
     :℁            : :address
     :〉          :
       :tag      : :p
@@ -84,6 +97,8 @@
     :☛             :
       :tag      : :p
       :class    : :'text-success'
+      :synonyms :
+        - :✿_div_center
     :☞             :
       :tag      : :p
       :class    : :'text-info'
@@ -92,7 +107,7 @@
       :class    : :'text-warning'
     :☢             :
       :tag      : :p
-      :class    : :text_danger
+      :class    : :'text-danger'
     :✇             : :video
     :⚘             : :figure
     :✿_p_address :
@@ -117,8 +132,8 @@
 # http://www.youtube.com/watch?v=SAJ_TzLqy1U
   'http://www\.youtube\.com/(?:watch\?v=|v/)([A-Za-z0-9_]+)' : '✇ \1'
 # Standalone images and quotes
-  '^(https?://\S+)(\s*.*)$' : '⚘ \1 \2'
-  
+  '^(https?://\S+)(\s*.*?)$' : '⚘ \1 \2'
+
 :enclosures :
   :•       :
     :tag   : :ul
@@ -127,6 +142,6 @@
   :▶ :     :dl
   :▷         :
     :tag   : :dl
-    :class : :dl_horizontal
+    :class : :'dl-horizontal'
   :〉       : :blockquote
 

data/extras/drafts/parsing.md

@@ -1,9 +1,38 @@
 Principles
 ==========
 
-**Typegrowl** is kinda 
-[Markdown](http://daringfireball.net/projects/markdown/syntax) successor,
-or, if you like, a Markdown². **Typegrowl** uses extended unicode symbols
+**Qipowl** is a Ruby parsing library. The parsing is done via
+DSL exactly as [Ouroboros](http://en.wikipedia.org/wiki/Ouroboros)
+eats it’s own tail. 
+
+The whole input is treated as Ruby source code and executed respectively.
+To prevent collisions of input with built-in ruby methods, the ASCII symbols
+in the input are being translated into their
+[fullwidth equivalents](http://en.wikipedia.org/wiki/Halfwidth_and_fullwidth_forms#Chart)
+before execution (and back to ASCII after the parsing is done.)
+
+Let’s say we have a string “Hello world” as input. It became ‘encoded’ into:
+“Ｈｅｌｌｏ ｗｏｒｌｄ”, executed as Ruby code (exactly as e. g. `puts rand`
+would) and finally ‘decoded’ back to ASCII. Whether the parser knows anything
+about ‘Ｈｅｌｌｏ’ or ‘ｗｏｒｌｄ’ it would be executed. Say, we have
+
+    def ｗｏｒｌｄ *args
+      "ｂｒａｖｅ ｎｅｗ #{__callee__}"
+    end
+
+thus the output will be:
+
+    # ⇒ Hello brave new world
+
+More about may be found at [project page](http://rocket-science.ru/qipowl/).
+
+Applications
+============
+
+**Qipowl** has a wide list of applications. The “markright”, descendant
+of “markup” and “markdown” is presented [here](http://qipowl.herokuapp.com).
+
+**Qipowl HTML** uses extended unicode symbols
 to specify more clean and readable source files and (boom!) ruby DSL to
 interpret them. E.g. the data definitions look like:
 
@@ -11,7 +40,7 @@ interpret them. E.g. the data definitions look like:
 
 Headings:
 
-    §§ This is a second-level heading
+    §1 This is a second-level heading
 
 Bold and emphasis:
 
@@ -19,13 +48,12 @@ Bold and emphasis:
 
 Comments are possible as well:
 
-    ✎ FIXME! Not to forget add this to parsing!
+    ✍ FIXME! 
+    Not to forget add this to parsing!
+    ✍
 
 etc.
 
-The other goal is to generate more typographically correct output, with
-proper quotation marks (“” instead of "" etc.)
-
 Why?
 ====
 
@@ -35,37 +63,37 @@ in old good Markdown are ugly, look at how they might be introduced:
 
     I like Markdown¹http://daringfireball.net/projects/markdown/syntax
 
-Markdown lacks a lot of modern features (properties of text,) such as:
-- tags
-- ???
+Markdown lacks a lot of modern features (properties of text).
 
 Markdown does not provide a blanket set of marks, fully covering 
 claims to markup language.
 
-
 Parsing
 =======
 
-Parsing is the most sexy part of **Typegrowl** bowels, since it’s done
-almost without any external parsing; `.tg` files are the ruby scripts
+Parsing is the most sexy part of **Qipowl** bowels, since it’s done
+almost without any external parsing; input files are the ruby scripts
 themselves. WTF? Let me explain.
 
-Let we have a `.tg` file of the following structure:
+Let we have an input file of the following structure:
 
-    § Typegrowl
+    §1 Qipowl
 
-    ✎ include language reference here
-    ≡Typegrowl≡ is the most exciting ruby DSL application example. As it
+    ✍ FIXME 
+    include language reference here
+    ✍
+
+    ≡Qipowl≡ is the most exciting ruby DSL application example. As it
     is stated in markdown reference:
 
-    » Readability, however, is emphasized above all else. 
-    » A Markdown-formatted document should be publishable as-is, 
-    » as plain text, without looking like it’s been marked up with 
-    » tags or formatting instructions.
-    » — http://daringfireball.net/projects/markdown/syntax
+    〉 Readability, however, is emphasized above all else. 
+    A Markdown-formatted document should be publishable as-is, 
+    as plain text, without looking like it’s been marked up with 
+    tags or formatting instructions.
+    — http://daringfireball.net/projects/markdown/syntax
 
 Now we simply give the source to ruby interpreter, which knowns, that
-`§` is *in fact* ruby function, which transforms that to any other syntax
+`§1` is *in fact* ruby function, which transforms that to any other syntax
 we want. To HTML, for instance.
 
 ## Parsing problems
@@ -81,7 +109,6 @@ They rather are looking like “text-symbol-text” and unfortunately should
 be preparsed to supply correct ruby DSL;
 - **lists and data definitions**, are to be surrounded with `<ul>`/`<dd>` tags;
 - **tables**… Bah, I didn’t think most about tables yet. They are ugly.
-    
 
 ### Links
 
@@ -106,8 +133,7 @@ forget about them:
 
 Links are being parsed in the following manner:
 
-- find the link in the input, according to simple pattern like 
-`(?:^|\P{L})(?<proto>[hftps/:]*)(?<path>\S+?)(?:\s|$)`
+- find the link in the input, according to simple pattern `URI.regexp`
 - determine whether it is an image, video or link to page by downloading
 and analyzing the headers
 - TODO copying the image to the host computer, providing the watermark
@@ -120,18 +146,3 @@ After all is done, we yield smth like `⚐ http://localhost.a.png` in place of
 `http://localhost.a.png` and `⚓ http://localhost/index.html` in place of
 `http://localhost/index.html`
 
-We will store all the links in some array and substitute them in input
-with array index as shown below:
-
-    Wiki¹⚓0 says, that Saint-Petersburg is the most beatuful city in the world:
-    SPb in the night¹⚐1
-
-### Links and abbrs: final normalization
-
-Now there are no nasty “http://” links without prepending DSL. Nice.
-Let’s go further: we don’t like links and abbrs in format 
-
-- `title¹href`
-- and `term†explanation`
-
-