breakout_parser 0.0.0

data/ext/breakout_parser/ruby_ext.c

@@ -0,0 +1,55 @@
+#ifdef RUBY_VERSION
+
+#include "ruby.h"
+
+void Init_breakout_parser();
+VALUE method_parse(VALUE, VALUE);
+
+VALUE breakout_parser = Qnil;
+
+void Init_breakout_parser() {
+    breakout_parser = rb_define_class("BreakoutParser",rb_cObject);
+    rb_define_singleton_method(breakout_parser, "parse", method_parse, 1);
+}
+
+extern char *buf, *bufptr;
+extern char *in_buf, *in_pos;
+extern size_t in_buf_len, bufsize;
+
+VALUE method_parse(VALUE self, VALUE text) {
+    VALUE s;
+    char *p    = RSTRING(text)->ptr;
+    in_buf_len = RSTRING(text)->len;
+    while( in_buf_len > 0 && (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t')){
+        p++;
+        in_buf_len--;
+    }
+    in_buf = p;
+    in_pos = in_buf;
+
+    bufsize    = 1 + in_buf_len + in_buf_len/3; // reserve 30% of in_buf size
+    if(bufsize<0x100) bufsize = 0x100;
+
+    buf = ALLOC_N(char, bufsize); 
+    bufptr = buf;
+
+    // protect buf from GC (theoretically)
+    rb_iv_set(self,"@obj",Data_Wrap_Struct(rb_cData,NULL,NULL,buf));
+
+    yyparse();
+    yylex_destroy();
+//    printf("[.] yyparse() ended\n");
+
+    // make ruby string from our char[] data
+    s = rb_str_new(buf,bufptr-buf);
+
+    // cleanup
+    rb_iv_set(self,"@obj",Qnil);
+    xfree(buf);
+    buf = bufptr = NULL;
+    bufsize = 0;
+
+    return s;
+}
+
+#endif // ifdef RUBY_VERSION

data/ext/breakout_parser/test.rb

@@ -0,0 +1,3 @@
+#!/usr/bin/env ruby
+require 'breakout_parser'
+p BreakoutParser.parse("h1. aaa")

data/ext/breakout_parser/yywrap.c

@@ -0,0 +1,3 @@
+int yywrap(){
+    return 1;
+}

data/lib/breakout_parser.rb

@@ -0,0 +1,6 @@
+if RUBY_PLATFORM =~/(mswin|mingw)/i
+  # Fat binary gems, you make the Rockin' world go round
+  require "breakout_parser/win32-ruby#{RUBY_VERSION.sub(/\.\d+$/, '')}/breakout_parser"
+else
+  require 'breakout_parser/breakout_parser'
+end

data/spec/parser_examples_spec.rb

@@ -0,0 +1,101 @@
+describe 'BreakoutParser' do
+
+  describe "bad examples" do
+    Dir["examples/orig/*.bad"].sort.each do |fname|
+      it "should not die on #{fname} " do
+        data = File.read(fname)
+        parse_file(fname).size.should >= File.read(fname).strip.gsub(/\s+/,' ').size
+      end
+    end
+  end
+  describe "pending examples" do
+    Dir["examples/orig/*.pending"].sort.each do |fname|
+      it "should parse #{fname} "
+    end
+  end
+
+  describe "preparsed examples" do
+    Dir["examples/orig/*.txt"].sort.each do |fname|
+      bname = File.basename(fname)
+      it "should parse #{fname} " do
+        preparsed = File.read("examples/parsed/#{bname}")
+        preparsed = preparsed[3..-1] if preparsed[0..2] == '<p>'
+        preparsed = preparsed[0..-5] if preparsed[-4..-1] == '</p>'
+        preparsed.gsub!("&#8211;","-")
+        preparsed.gsub!("&#8212;","--")
+        preparsed.gsub!("&#8216;","'")
+        preparsed.gsub!("&#8217;","'")
+        preparsed.gsub!("&#8230;","...")
+        preparsed.gsub!("&#215;","x")
+        preparsed.gsub!("&#169;","(c)")
+        preparsed.gsub!("<br />\n","<br />")
+        preparsed.gsub!(/[ \t]+<br \/>/,"<br />")
+        preparsed.gsub!("\t"," ")
+        if preparsed['<hr />']
+          # find longest dash-line in source
+          dashline = File.read(fname).scan(/-+/).sort_by{ |x| -x.length }.first
+          preparsed.gsub!("</p>\n<hr />\n<p>","<br /><br />#{dashline}<br /><br />");
+        end
+
+#        preparsed.gsub!(/^<p>/,"");
+#        preparsed.gsub!(/<\/p>$/,"");
+        preparsed.gsub!("</pre>\n<ol>","</pre><br /><br /><ol>")
+        preparsed.gsub!(/<\/p>\s+<p>/,"<br /><br />")
+        preparsed.gsub!("</p>\n","<br /><br />")
+        preparsed.gsub!("<p>","<br /><br />")
+        preparsed.gsub!(/[\r\n]+ */," ")
+        preparsed.gsub!(/[ \t]{2,}/," ")
+
+        preparsed.gsub!("<del>","-")
+        preparsed.gsub!("</del>","-")
+        preparsed.gsub!(/<br \/>[ ]+/,"<br />")
+        preparsed.gsub!(/(<br \/>){2,}/,"<br /><br />")
+#        preparsed.gsub!("<br /><ol>","<ol>")
+#        preparsed.gsub!("<br /><ul>","<ul>")
+#        preparsed.gsub!("<br /><br /><ul>","<br /><ul>")
+
+        parsed = parse_file(fname)
+
+        # old parser not parses raw text urls
+        #parsed.gsub!(%r'<a href="([^<>"]+)">([^<>"]+)</a>',"\\1")
+
+        t1 = parsed
+        t2 = preparsed
+
+        [t1,t2].each do |t|
+          t.downcase!
+          t.gsub!(/(\s*<br \/>\s*)+/,' ')
+          t.gsub!(/\n\s*/,"\n")
+#          t.gsub!(/>[ \t]+</,"><")
+          t.gsub!(/>[ \t]+/,">")
+          t.gsub!(/[ \t]+</,"<")
+          t.gsub!(/[\r\n \t]+/," ")
+          t.strip!
+        end
+
+        if t1 != t2
+#          File.open("last-parsed.tmp","w"){ |f| f << parsed }
+#          File.open("last-preparsed.tmp","w"){ |f| f << preparsed }
+          pos = 0
+          pos += 1 while t1[0..pos] == t2[0..pos]
+          pos -= 5
+          pos = 0 if pos<0
+          t1[pos..-1].should == t2[pos..-1]
+        end
+        t1.should == t2
+      end
+      $n ||= 0
+      $n +=  1
+#      break if $n == 1900
+    end
+  end
+
+###############################################################################
+###############################################################################
+###############################################################################
+
+  def parse_file fname
+    r = `cat #{fname} | ./parser`
+    r.strip
+  end
+end

data/spec/parser_spec.rb

@@ -0,0 +1,549 @@
+require 'breakout_parser'
+
+describe 'BreakoutParser' do
+  def self.hex_string s
+    s.each_byte.to_a.map{ |c| "%02x" % c }.join
+  end
+  def hex_string s; self.class.hex_string(s); end
+
+  it 'converts \n to <br />' do
+    parse("aaa\nbbb").should match(%r"aaa ?<br /> ?bbb")
+  end
+
+  it "parses 1M file #1" do
+    s = 'a' * 1024 * 1024
+    parse(s).size.should == s.size
+  end
+
+  it "parses 1M file #2" do
+    s = 'a' + (' ' * 1024 * 1024) + 'b'
+    parse(s).should == 'a b'
+  end
+
+  it "parses 1M file #3" do
+    s = 'a ' * 1024 * 512
+    parse(s).size.should == s.strip.size
+  end
+
+  it "strips tailing spaces and newlines" do
+    parse("aaa         ").should == "aaa"
+    parse("aaa\t\t\t\t\t\t").should == "aaa"
+    parse("aaa\r\r\r\r\r").should == "aaa"
+    parse("aaa\n\n\n\n\n").should == "aaa"
+    parse("aaa\r\n\r\n\r\n\r\n").should == "aaa"
+    parse("aaa\r\n\t   \t  \n  \r   \n   \t  \t\n\r   ").should == "aaa"
+  end
+
+  it "strips leading spaces and newlines" do
+    parse("         aaa").should == "aaa"
+    parse("\t\t\t\t\t\taaa").should == "aaa"
+    parse("\r\r\r\r\raaa").should == "aaa"
+    parse("\n\n\n\n\naaa").should == "aaa"
+    parse("\r\n\r\n\r\n\r\naaa").should == "aaa"
+    parse("\r\n\t   \t  \n  \r   \n   \t  \t\n\r   aaa").should == "aaa"
+  end
+
+  it "converts two or more \\n to single empty line" do
+    parse("aaa\n\nbbb").should == "aaa<br /><br />bbb"
+    parse("aaa\n  \nbbb").should == "aaa<br /><br />bbb"
+    parse("aaa\n\n\nbbb").should == "aaa<br /><br />bbb"
+    parse("aaa\n   \n    \nbbb").should == "aaa<br /><br />bbb"
+    parse("aaa\r\n   \r\n    \r\nbbb").should == "aaa<br /><br />bbb"
+    parse("aaa\n   \n\n    \nbbb").should == "aaa<br /><br />bbb"
+    parse("aaa\n   \n\n\n    \nbbb").should == "aaa<br /><br />bbb"
+    parse("aaa\n\n\n\n\n\n\nbbb").should == "aaa<br /><br />bbb"
+    parse("aaa\r\n\r\n\r\nbbb").should == "aaa<br /><br />bbb"
+  end
+
+###############################################################################
+
+  describe "*bold*" do
+    it "only" do
+      parse("*bold*").should == '<strong>bold</strong>'
+    end
+    it "at beginning" do
+      parse("*bold*\nxxx").should == '<strong>bold</strong><br />xxx'
+    end
+    it "in the middle of text" do
+      parse("xxx *bold* yyy").should == 'xxx <strong>bold</strong> yyy'
+    end
+    it "parses *multiline\\nbold*" do
+      parse("*multiline\nbold*").should == "<strong>multiline<br />bold</strong>"
+    end
+    it "skips lone star inside bold block" do
+      parse("*aaa * bbb*").should == '<strong>aaa * bbb</strong>'
+    end
+    it "skips lone star" do
+      parse("aaa * bbb").should == 'aaa * bbb'
+    end
+    it "w/o closing tag" do
+      parse("*bold").should == '<strong>bold</strong>'
+    end
+    it "nesting1 w/o closing tags" do
+      parse("*bold1 *bold2").should == '<strong>bold1 <strong>bold2</strong></strong>'
+    end
+    it "nesting2 w/o closing tags" do
+      parse("*bold1  *bold2").should == '<strong>bold1 <strong>bold2</strong></strong>'
+    end
+
+    it "not parses '*.*'" do
+      parse("*.*").should == "*.*"
+      parse(" *.* ").should == "*.*"
+      parse("aaa *.* bbb").should == "aaa *.* bbb"
+    end
+
+    it "not parses '*.something'" do
+      parse("*.exe").should == "*.exe"
+      parse(" *.exe ").should == "*.exe"
+      parse("aaa *.exe bbb").should == "aaa *.exe bbb"
+    end
+
+  end
+
+###############################################################################
+
+  describe "_italic_" do
+    it "only" do
+      parse("_italic_").should == '<em>italic</em>'
+    end
+    it "at beginning" do
+      parse("_italic_\nxxx").should == '<em>italic</em><br />xxx'
+    end
+    it "in the middle of text" do
+      parse("xxx _italic_ yyy").should == 'xxx <em>italic</em> yyy'
+    end
+    it "parses _multiline\\nitalic_" do
+      parse("_multiline\nitalic_").should == "<em>multiline<br />italic</em>"
+    end
+    it "skips lone underscore inside italic block" do
+      parse("_aaa _ bbb_").should == '<em>aaa _ bbb</em>'
+    end
+    it "skips lone underscore" do
+      parse("aaa _ bbb").should == 'aaa _ bbb'
+    end
+    it "w/o closing tag" do
+      parse("_italic").should == '<em>italic</em>'
+    end
+    it "nesting1 w/o closing tags" do
+      parse("_italic1 _italic2").should == '<em>italic1 <em>italic2</em></em>'
+    end
+    it "nesting2 w/o closing tags" do
+      parse("_italic1  _italic2").should == '<em>italic1 <em>italic2</em></em>'
+    end
+  end
+
+###############################################################################
+
+  describe "combinations" do
+    it "bold in italic" do
+      s = "_aaa *bbb* ccc_"
+      parse(s).should == "<em>aaa <strong>bbb</strong> ccc</em>"
+    end
+    it "bold in italic - no closing1" do
+      s = "_aaa *bbb* ccc"
+      parse(s).should == "<em>aaa <strong>bbb</strong> ccc</em>"
+    end
+    it "bold in italic - no closing2" do
+      s = "_aaa *bbb ccc"
+      parse(s).should == "<em>aaa <strong>bbb ccc</strong></em>"
+    end
+
+    it "italic in bold" do
+      s = "*aaa _bbb_ ccc*"
+      parse(s).should == "<strong>aaa <em>bbb</em> ccc</strong>"
+    end
+    it "italic in bold - no closing1" do
+      s = "*aaa _bbb_ ccc"
+      parse(s).should == "<strong>aaa <em>bbb</em> ccc</strong>"
+    end
+    it "italic in bold - no closing2" do
+      s = "*aaa _bbb ccc"
+      parse(s).should == "<strong>aaa <em>bbb ccc</em></strong>"
+    end
+
+    {'ul' => '*', 'ol' => '#'}.each do |l,c|
+      it "raw text link inside #{l.upcase}> #1" do
+        s = "#{c} aaa http://www.ru"
+        parse(s).should == "<#{l}><li>aaa <a href=\"http://www.ru\">http://www.ru</a></li></#{l}>"
+      end
+      it "raw text link inside #{l.upcase}> #2" do
+        s = "#{c} aaa http://www.ru\n#{c} bbb"
+        parse(s).should == "<#{l}><li>aaa <a href=\"http://www.ru\">http://www.ru</a></li><li>bbb</li></#{l}>"
+      end
+      it "raw text link inside #{l.upcase}> #3" do
+        s = "#{c} http://www.ru"
+        parse(s).should == "<#{l}><li><a href=\"http://www.ru\">http://www.ru</a></li></#{l}>"
+      end
+      it "raw text link inside #{l.upcase}> #4" do
+        s = "#{c} aaa http://www.ru bbb"
+        parse(s).should == "<#{l}><li>aaa <a href=\"http://www.ru\">http://www.ru</a> bbb</li></#{l}>"
+      end
+      it "two links inside #{l.upcase}>" do
+        s = "#{c} aaa http://www.ru http://ya.ru bbb"
+        parse(s).should == "<#{l}><li>aaa <a href=\"http://www.ru\">http://www.ru</a> <a href=\"http://ya.ru\">http://ya.ru</a> bbb</li></#{l}>"
+      end
+    end
+  end
+
+###############################################################################
+
+  describe "unnumbered list" do
+    it "should work" do
+      parse("* a\n* b\n* c").should match(
+        %r"<ul><li>a</li><li>b</li><li>c</li></ul>"
+      )
+    end
+    it "two lists" do
+      s = "* a\n* b\n* c"
+      s = s + "\nxxx\n" + s
+      r = "<ul><li>a</li><li>b</li><li>c</li></ul>"
+      parse(s).should == "#{r}xxx<br />#{r}"
+    end
+    it "in middle of text when begins with space" do
+      parse("hello\n * a\n  * b\n * c\nworld").should ==
+        "hello<br /><ul><li>a</li><li>b</li><li>c</li></ul>world"
+    end
+    it "in middle of text" do
+      parse("hello\n* a\n* b\n* c\nworld").should ==
+        "hello<br /><ul><li>a</li><li>b</li><li>c</li></ul>world"
+    end
+    it "after blank line" do
+      parse("hello\n\n * a\n * b\n * c\nworld").should ==
+        "hello<br /><br /><ul><li>a</li><li>b</li><li>c</li></ul>world"
+    end
+  end
+
+###############################################################################
+
+  describe "numbered list" do
+    it "should work" do
+      parse("# a\n# b\n# c").should match(
+        %r"<ol><li>a</li><li>b</li><li>c</li></ol>"
+      )
+    end
+    it "two lists" do
+      s = "# a\n# b\n# c"
+      s = s + "\nxxx\n" + s
+      r = "<ol><li>a</li><li>b</li><li>c</li></ol>"
+      parse(s).should == "#{r}xxx<br />#{r}"
+    end
+    it "in middle of text when begins with space" do
+      parse("hello\n # a\n  # b\n # c\nworld").should ==
+        "hello<br /><ol><li>a</li><li>b</li><li>c</li></ol>world"
+    end
+    it "in middle of text" do
+      parse("hello\n# a\n# b\n# c\nworld").should ==
+        "hello<br /><ol><li>a</li><li>b</li><li>c</li></ol>world"
+    end
+    it "after blank line" do
+      parse("hello\n\n # a\n # b\n # c\nworld").should ==
+        "hello<br /><br /><ol><li>a</li><li>b</li><li>c</li></ol>world"
+    end
+  end
+
+###############################################################################
+
+  1.upto(5) do |lvl|
+    describe "H#{lvl}" do
+      it "at the beginning" do
+        parse("h#{lvl}. xxx").should == "<h#{lvl} id=\"xxx\">xxx</h#{lvl}>"
+      end
+      it "after 1 line of text" do
+        parse("abcd\nh#{lvl}. xxx").should == "abcd<br /><h#{lvl} id=\"xxx\">xxx</h#{lvl}>"
+      end
+      it "after 2 lines of text" do
+        parse("abcd\ndefgh\nh#{lvl}. xxx").should == "abcd<br />defgh<br /><h#{lvl} id=\"xxx\">xxx</h#{lvl}>"
+      end
+      it "in middle of other words" do
+        parse("abcd defgh h#{lvl}. xxx yyy").should == "abcd defgh h#{lvl}. xxx yyy"
+      end
+      it "in middle of other lines" do
+        parse("abcd defgh\nh#{lvl}. xxx\nyyy").should == "abcd defgh<br /><h#{lvl} id=\"xxx\">xxx</h#{lvl}><br />yyy"
+      end
+
+      it "converts spaces to underscores in id" do
+        parse("h#{lvl}. xxx   yyy z").should == "<h#{lvl} id=\"xxx___yyy_z\">xxx   yyy z</h#{lvl}>"
+      end
+      it "keeps underscores in id" do
+        parse("h#{lvl}. xxx___yyy_z").should == "<h#{lvl} id=\"xxx___yyy_z\">xxx___yyy_z</h#{lvl}>"
+      end
+      it "keeps dashes in id" do
+        parse("h#{lvl}. xxx---yyy-z").should == "<h#{lvl} id=\"xxx---yyy-z\">xxx---yyy-z</h#{lvl}>"
+      end
+      it "keeps dots in id" do
+        parse("h#{lvl}. xxx...yyy.z").should == "<h#{lvl} id=\"xxx...yyy.z\">xxx...yyy.z</h#{lvl}>"
+      end
+
+      %w'Ъ ъ : ; , привет" \' ! < >'.each do |c|
+        it "converts id to hex if it contains \"#{c}\"" do
+          idhex = hex_string("xxx#{c}yyy")
+          parse("h#{lvl}. xxx#{c}yyy").should == "<h#{lvl} id=\"#{idhex}\">xxx#{h(c)}yyy</h#{lvl}>"
+        end
+      end
+
+      it "skips excess spaces" do
+        parse("h#{lvl}.  \t  xxx   \t ").should == "<h#{lvl} id=\"xxx\">xxx</h#{lvl}>"
+      end
+
+      it "thinks that \\r is EOL" do
+        parse("h#{lvl}. xxx\ryyy").should == "<h#{lvl} id=\"xxx\">xxx</h#{lvl}><br />yyy"
+        parse("h#{lvl}. xxx\r").should == "<h#{lvl} id=\"xxx\">xxx</h#{lvl}>"
+
+        parse("h#{lvl}. xxx\r yyy").sub(' yyy','yyy').should ==
+          "<h#{lvl} id=\"xxx\">xxx</h#{lvl}><br />yyy"
+      end
+    end
+  end
+
+###############################################################################
+
+  describe "raw text links" do
+    it "at the beginning" do
+      parse("http://www.ru").should == "<a href=\"http://www.ru\">http://www.ru</a>"
+    end
+    it "in middle of other words" do
+      parse("aaa bbb ccc http://www.ru ddd eee fff").should ==
+        "aaa bbb ccc <a href=\"http://www.ru\">http://www.ru</a> ddd eee fff"
+    end
+    it "in new line" do
+      parse("aaa bbb ccc\nhttp://www.ru\nddd eee fff").should match(
+        %r"aaa bbb ccc ?<br /> ?<a href=\"http://www.ru\">http://www.ru</a> ?<br /> ?ddd eee fff"
+      )
+    end
+    it "escapes '&' in link _text_" do
+      parse("http://www.ru/?a=1&b=2").should == "<a href=\"http://www.ru/?a=1&b=2\">http://www.ru/?a=1&amp;b=2</a>"
+    end
+
+    it "parses https://" do
+      parse("https://www.ru").should == "<a href=\"https://www.ru\">https://www.ru</a>"
+    end
+
+    %w', .'.each do |c|
+      it "stops parsing on \"#{c} \"" do
+        parse("http://www.ru#{c}").should == "<a href=\"http://www.ru\">http://www.ru</a>#{c}"
+        parse(" http://www.ru#{c} ").should == "<a href=\"http://www.ru\">http://www.ru</a>#{c}"
+        parse(" http://www.ru#{c} hello!").should == "<a href=\"http://www.ru\">http://www.ru</a>#{c} hello!"
+        parse("xxx http://www.ru#{c} hello!").should == "xxx <a href=\"http://www.ru\">http://www.ru</a>#{c} hello!"
+        parse(" http://www.ru/#{c} hello!").should == "<a href=\"http://www.ru/\">http://www.ru/</a>#{c} hello!"
+        parse(" http://aaa.com#{c} http://bbb.com").should ==
+          "<a href=\"http://aaa.com\">http://aaa.com</a>#{c} <a href=\"http://bbb.com\">http://bbb.com</a>"
+      end
+    end
+  end
+
+###############################################################################
+
+  describe "#ticketNum ticket links" do
+    it "at the beginning" do
+      parse("#1234").should == '<a href="/spaces/test_space/tickets/1234">#1234</a>'
+    end
+    it "in middle of other words" do
+      parse("aaa bbb ccc #3476 ddd eee fff").should ==
+        'aaa bbb ccc <a href="/spaces/test_space/tickets/3476">#3476</a> ddd eee fff'
+    end
+    it "in new line" do
+      parse("aaa bbb ccc\n#1234\nddd eee fff").should match(
+        %r|aaa bbb ccc ?<br /> ?<a href="/spaces/test_space/tickets/1234">#1234</a> ?<br /> ?ddd eee fff|
+      )
+    end
+    it "ignores non-digits" do
+      parse("#1234d").should == '#1234d'
+      parse("#xxx").should == '#xxx'
+    end
+  end
+
+###############################################################################
+
+  describe "<pre><code>..</code></pre>" do
+    it "works" do
+      s = <<-EOF
+    for ( n = 0; n < max_size && \
+           (c = getc( yyin )) != EOF && c != '\\n'; ++n ) \
+      buf[n] = (char) c; \
+
+      EOF
+
+      parse("<pre><code>#{s.strip}</code></pre>").should ==
+        "<pre><code>#{h(s.strip)}</code></pre>"
+
+      s = <<-EOF
+          while ( 1 < 2 ) do
+            puts "<b>12345\\t54321</b>"
+            // *bold* comment
+            // _italic_ comment
+          end
+          ---
+          * aaa
+          * bbb
+          * ccc
+
+      EOF
+      parse("<pre><code>#{s.strip}</code></pre>").should ==
+        "<pre><code>#{h(s.strip)}</code></pre>"
+    end
+    it "not parses *bold*" do
+      s = "<pre><code> *bold*</code></pre>"
+      parse(s).should == s
+    end
+    it "not parses _italic_" do
+      s = "<pre><code> _italic_</code></pre>"
+      parse(s).should == s
+    end
+    it "not parses UL lists" do
+      s = "<pre><code>\n * l1\n * l2\n * l3</code></pre>"
+      parse(s).should == s.sub("<code>\n","<code>")
+    end
+    it "not parses OL lists" do
+      s = "<pre><code>\n # l1\n # l2\n # l3</code></pre>"
+      parse(s).should == s.sub("<code>\n","<code>")
+    end
+    it "not parses H1..H5" do
+      1.upto(5) do |i|
+        s = "<pre><code>\nh#{i}. zzzzzzz\n</code></pre>"
+        parse(s).should == "<pre><code>h#{i}. zzzzzzz</code></pre>"
+      end
+    end
+    it "not parses raw text links" do
+      s = "<pre><code>xxx http://www.ru yyy</code></pre>"
+      parse(s).should == s
+      s = "<pre><code>http://www.ru</code></pre>"
+      parse(s).should == s
+    end
+    it "keeps newlines" do
+      s = "<pre><code>aaa\nbbb</code></pre>"
+      parse(s).should == s
+      s = "<pre><code>aaa\n\nbbb\nccc</code></pre>"
+      parse(s).should == s
+    end
+
+    it "with no spaces between <pre> and <code>" do
+      s = "<pre><code>aaa</code></pre>"
+      parse(s).should == s
+    end
+
+    it "with spaces between <pre> and <code>" do
+      s = "<pre>    <code>aaa</code>           </pre>"
+      parse(s).should == s.tr(' ','')
+    end
+    it "with spaces between <pre> and <code> and inside" do
+      s = "<pre>    <code>  aaa   bbb   </code>           </pre>"
+      parse(s).should == "<pre><code>  aaa   bbb</code></pre>"
+    end
+
+    it "w/o closing tags" do
+      s = "<pre><code>aaa"
+      parse(s).should match(%r"<pre><code>aaa\n?</code></pre>")
+    end
+
+    it "in middle of text" do
+      s = "xxx <pre><code>yyyy</code></pre> jjj"
+      parse(s).should == s
+    end
+
+    it "with 2 instances" do
+      s = "xxx <pre><code>yyyy</code></pre> <jjj> <pre><code>asdkjaslkd</code></pre> END"
+      parse(s).should == s.sub('<jjj>','&lt;jjj&gt;')
+    end
+
+    it "works with unicode" do
+      s = "привет <pre><code> жжж </code></pre> пока!"
+      parse(s).should match(%r|привет ?<pre><code> жжж</code></pre> ?пока!|)
+
+      s = 'абвгдеёжзийклмнопрстуфхцчшщьыъэюя'
+      parse(s).should == s
+
+      s = 'АБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЬЫЪЭЮЯ'
+      parse(s).should == s
+
+      s = '☸☹☺☻☼☽☾☿'
+      parse(s).should == s
+    end
+
+    it "should escape lone closing tags" do
+      s = "</code></pre>"
+      parse(s).should == h(s)
+    end
+
+    it "should skip newlines and spaces at end" do
+      s = "<pre><code> aaa bbb ccc \n\n\n  \t\n\n\n\r\n\r\n   \t  </code></pre>"
+      parse(s).should == "<pre><code> aaa bbb ccc</code></pre>"
+    end
+
+    it "escapes html chars" do
+      HTML_ESCAPE.each do |k,v|
+        parse("<pre><code>#{k}</code></pre>").should == "<pre><code>#{v}</code></pre>"
+      end
+    end
+  end
+
+###############################################################################
+
+  describe "Assembla Links" do
+    a = {}
+    a["wiki:Name"]  = '<a class="wiki_link" title="Name" href="/wiki/show/test_space/Name">Name</a>'
+    a["Name"]       = '<a class="wiki_link" title="Name" href="/wiki/show/test_space/Name">Name</a>'
+    a["Name#Ref"]   = '<a class="wiki_link" title="Name#Ref" href="/wiki/show/test_space/Name#Ref">Name#Ref</a>'
+    a["#Ref"]       = '<a href="#Ref" title="#Ref" class="wiki_link">#Ref</a>'
+    a["#привет"]    = %Q|<a href="##{hex_string("привет")}" title="#привет" class="wiki_link">#привет</a>|
+    a["#with  spc"] = %Q|<a href="#with__spc" title="#with  spc" class="wiki_link">#with  spc</a>|
+    a["#with__usc"] = %Q|<a href="#with__usc" title="#with__usc" class="wiki_link">#with__usc</a>|
+    a["#with--dsh"] = %Q|<a href="#with--dsh" title="#with--dsh" class="wiki_link">#with--dsh</a>|
+    a["#with!xclm"] = %Q|<a href="##{hex_string("with!xclm")}" title="#with!xclm" class="wiki_link">#with!xclm</a>|
+    a["#with&amp"]  = %Q|<a href="##{hex_string("with&amp")}" title="#with&amp" class="wiki_link">#with&amp;amp</a>|
+
+    a["ticket:234"] = '<a href="/spaces/test_space/tickets/234">#234</a>'
+    a["revision:1f4bdab77be696efd"] =
+      '<a href="http://code.assembla.com/test_space/git/changesets/1f4bdab77be696efd">revision:1f4bdab77be696efd</a>'
+    a["revision:12345"] =
+      '<a href="http://code.assembla.com/test_space/svn/changesets/12345">revision:12345</a>'
+    a["r:2345"] = '<a href="http://code.assembla.com/test_space/svn/changesets/2345">revision:2345</a>'
+    a["r:2345ef"] = '<a href="http://code.assembla.com/test_space/git/changesets/2345ef">revision:2345ef</a>'
+
+    a["url:http://www.ru"]   = '<a href="http://www.ru">http://www.ru</a>'
+    a["url:https://www.ru"]  = '<a href="https://www.ru">https://www.ru</a>'
+    a["url:www.ru"]          = '<a href="http://www.ru">http://www.ru</a>'
+    a["url:www.ru/?a=1&b=2"] = '<a href="http://www.ru/?a=1&b=2">http://www.ru/?a=1&amp;b=2</a>'
+    a["url:ftp://www.ru"]    = '<a href="ftp://www.ru">ftp://www.ru</a>'
+    a["url:/spaces/x2"]      = '<a href="/spaces/x2">/spaces/x2</a>'
+
+    a.each do |k,v|
+      it "parses [[#{k}]]" do
+        parse("[[#{k}]]").should == v
+      end
+      it "parses [[#{k}|привет тест]]" do
+        parse("[[#{k}|привет тест]]").should == v.sub(/>.*</,">привет тест<")
+      end
+      it "parses [[#{k}|test & here]]" do
+        parse("[[#{k}|test & here]]").should == v.sub(/>.*</,">test &amp; here<")
+      end
+    end
+
+    it "keeps unknown link types" do
+      s = "[[zzz:xxx]]"
+      parse(s).should == s
+      s = "[[abcd:1234]]"
+      parse(s).should == s
+      s = "[[abcd::1234]] [[abcd:1234]] [[uri:www.ru]]"
+      parse(s).should == s
+    end
+
+    it "links to ExistingFile.txt"
+    it "links to NotExistingFile.txt"
+    it "links to ExistingImage.png"
+    it "links to NotExistingImage.png"
+  end
+
+###############################################################################
+###############################################################################
+###############################################################################
+
+  HTML_ESCAPE   =   { '&' => '&amp;', '>' => '&gt;', '<' => '&lt;', '"' => '&quot;' }
+
+  def h s
+    s.to_s.gsub(/[&"><]/) { |special| HTML_ESCAPE[special] }
+  end
+
+  def parse s
+    BreakoutParser.parse(s).strip
+  end
+end