breakout_parser 0.0.23 → 0.0.31

data/ext/breakout_parser/ruby_ext.c

@@ -1,6 +1,13 @@
 #ifdef RUBY_VERSION
 
 #include "ruby.h"
+#include "ruby_ext.h"
+
+#ifdef RUBY_19
+  #include "ruby/st.h"
+#else
+  #include "st.h"
+#endif
 
 void Init_breakout_parser();
 VALUE method_parse(int, VALUE*, VALUE);
@@ -20,55 +27,78 @@ extern const char *space_name;
 extern size_t in_buf_len, bufsize, space_name_len;
 
 extern const char *site_url;
+extern const char *large_files_url;
 extern size_t site_url_len;
+extern size_t large_files_url_len;
+
+extern VALUE meta_attributes;
 
 VALUE vcs_url;
 
+char **temp_names, **temp_values;
+
 extern int parse_links_only, absolute_urls;
 
 VALUE do_parse(int argc, VALUE *argv, VALUE self) {
     VALUE s, text, r_space_name;
     char *p;
 
-    if( argc < 2 || argc > 5 ){
-        rb_raise(rb_eArgError, "wrong number of arguments (%d for 2..5)", argc);
-        return rb_str_new("",0); // unreachable code, but for double safety
+    if (argc < 2 || argc > 7) {
+        rb_raise(rb_eArgError, "wrong number of arguments (%d for 2..7)", argc);
     }
 
     text = argv[0];
     r_space_name = argv[1];
 
     site_url = NULL; site_url_len = 0;
-    if( argc > 2 && argv[2] != Qnil && argv[2] != Qfalse ){
+    if (argc > 2 && RTEST(argv[2])){
         site_url     = StringValueCStr(argv[2]);
         site_url_len = site_url ? strlen(site_url) : 0;
         while( site_url && site_url_len > 0 && site_url[site_url_len-1] == '/' ) {
             // skip trailing slashes
-            site_url_len--; 
+            site_url_len--;
         }
     }
 
-    vcs_url = NULL;
-    if( argc > 3 && argv[3] != Qnil && argv[3] != Qfalse ){
+    vcs_url = rb_str_new2("");
+    if (argc > 3 && RTEST(argv[3])) {
         vcs_url = argv[3];
     }
 
     absolute_urls = 0;
-    if( argc > 4 && argv[4] != Qnil && argv[4] != Qfalse ){
+    if (argc > 4 && RTEST(argv[4])) {
         absolute_urls = 1;
     }
 
-    if(!text || text == Qnil || text == Qfalse){
-        // NULL input string
-        return rb_str_new("",0);
+    large_files_url = NULL; large_files_url_len = 0;
+    if (argc > 5 && RTEST(argv[5])) {
+        large_files_url = StringValueCStr(argv[5]);
+        large_files_url_len = large_files_url ? strlen(large_files_url) : 0;
     }
 
-    p          = RSTRING(text)->ptr;
-    in_buf_len = RSTRING(text)->len;
+    meta_attributes = rb_hash_new();
+    if (argc > 6 && RTEST(argv[6])) {
+        // More information about object types may be found in ruby.h
+        if (TYPE(argv[6]) != T_HASH) {
+            rb_raise(rb_eTypeError, "wrong type of meta attributes values; expected Hash, given %d", TYPE(argv[6]));
+        }
+
+        if (RHASH_SIZE(argv[6]) > 0) {
+            meta_attributes = argv[6];
+        }
+    }
+
+    if(!text || !(RTEST(text))){
+        // empty input string
+        return rb_str_new2("");
+    }
+
+    p          = RSTRING_PTR(text);
+    in_buf_len = RSTRING_LEN(text);
 
     if(!p || in_buf_len <= 0){
         // empty input string
-        return rb_str_new("",0);
+        return rb_str_new2("");
     }
 
     while( in_buf_len > 0 && (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t')){
@@ -84,25 +114,24 @@ VALUE do_parse(int argc, VALUE *argv, VALUE self) {
     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); 
+    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");
+    // printf("[.] yyparse() ended\n");
 
     // make ruby string from our char[] data
-    s = rb_str_new(buf,bufptr-buf);
+    s = rb_str_new(buf, bufptr-buf);
 
-    // cleanup
-    rb_iv_set(self,"@obj",Qnil);
     xfree(buf);
     buf = bufptr = NULL;
     bufsize = 0;
 
+    #ifdef RUBY_19
+      s = rb_funcall(s, rb_intern("force_encoding"), 1, rb_str_new2("UTF-8")); // encode string to UTF-8
+    #endif
+
     return s;
 }
 

data/ext/breakout_parser/ruby_ext.h

@@ -0,0 +1,2 @@
+void yyparse();
+void yylex_destroy();

data/spec/links_only_parser_spec.rb

@@ -0,0 +1,916 @@
+# -*- coding: utf-8 -*-
+require 'spec_helper'
+require File.dirname(__FILE__) + '/obj_proxy'
+
+describe 'BreakoutParser' do
+  def self.hex_string s
+    ## sexier, but not runs on ruby 1.8.6 patchlevel 383 i386-mingw32:
+    #s.each_byte.to_a.map{ |c| "%02x" % c }.join
+    r = ''
+    s.each_byte{ |c| r << "%02x" % c }
+    r
+  end
+  def hex_string s; self.class.hex_string(s); end
+
+  it 'accepts from 2 to 7 arguments' do
+    expect { BreakoutParser.parse }.to raise_error(ArgumentError, "wrong number of arguments (0 for 2..7)")
+    expect { BreakoutParser.parse('a') }.to raise_error(ArgumentError, "wrong number of arguments (1 for 2..7)")
+    [9, 10].each do |argc|
+      expect { BreakoutParser.parse(*(['a'] * (argc - 1) + [{}])) }.to raise_error(ArgumentError, "wrong number of arguments (#{argc} for 2..7)")
+    end
+    (2..6).each do |argc|
+      expect { BreakoutParser.parse(*(['a'] * argc)) }.to_not raise_error(ArgumentError)
+    end
+    expect { BreakoutParser.parse(*(['a'] * 6 + [{}])) }.to_not raise_error(ArgumentError)
+  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 "handles nil & false text well" do
+    parse(false).should == ""
+    parse(false, :space_name => false).should == ""
+    parse("",    :space_name => false).should == ""
+    parse(nil).should == ""
+    parse(nil,   :space_name => nil).should == ""
+    parse("",    :space_name => nil).should == ""
+  end
+
+  it "handles nil space_name well" do
+    lambda{
+      parse("#123", :space_name => nil)
+    }.should raise_error(TypeError)
+  end
+
+  it "handles false space_name well" do
+    lambda{
+      parse("#123", :space_name => false)
+    }.should raise_error(TypeError)
+  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 each newline to <br />" 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 /><br />bbb"
+    parse("aaa\n   \n    \nbbb").should == "aaa<br /><br /><br />bbb"
+    parse("aaa\r\n   \r\n    \r\nbbb").should == "aaa<br /><br /><br />bbb"
+    parse("aaa\n   \n\n    \nbbb").should == "aaa<br /><br /><br /><br />bbb"
+    parse("aaa\n   \n\n\n    \nbbb").should == "aaa" + "<br />"*5 + "bbb"
+    parse("aaa\n\n\n\n\n\n\nbbb").should == "aaa" + "<br />"*7 + "bbb"
+    parse("aaa\r\n\r\n\r\nbbb").should == "aaa" + "<br />"*3 + "bbb"
+  end
+
+###############################################################################
+
+  describe "@code@" do
+    it "only" do
+      parse("@smth@").should == '@smth@'
+    end
+    it "at beginning" do
+      parse("@smth@\nxxx").should == '@smth@<br />xxx'
+    end
+    it "in the middle of text" do
+      parse("xxx @smth@ yyy").should == 'xxx @smth@ yyy'
+    end
+    it "parses @multiline\\nsmth@" do
+      parse("@multiline\nsmth@").should == "@multiline<br />smth@"
+    end
+    it "not confuses" do
+      parse("look at @this code@ and mail me at xxx@yyy.com").should ==
+        'look at @this code@ and mail me at <a href="mailto:xxx@yyy.com">xxx@yyy.com</a>'
+    end
+    it "w/o closing tag" do
+      parse("@smth").should == '@smth'
+    end
+    it "nesting1 w/o closing tags" do
+      parse("@smth1 @smth2").should == '@smth1 @smth2'
+    end
+    it "nesting2 w/o closing tags" do
+      parse("@smth1  @smth2").should == '@smth1 @smth2'
+    end
+    it "two times" do
+      parse("@code1@ @code2@").should == "@code1@ @code2@"
+      parse("@code1@   @code2@").should == "@code1@ @code2@"
+      parse("  @code1@   @code2@  ").should == "@code1@ @code2@"
+      parse("  @code1@ xxx @code2@  ").should == "@code1@ xxx @code2@"
+    end
+  end
+
+###############################################################################
+
+  describe "*bold*" do
+    it "only" do
+      parse("*bold*").should == '*bold*'
+    end
+    it "at beginning" do
+      parse("*bold*\nxxx").should == "*bold*<br />xxx"
+    end
+    it "in the middle of text" do
+      parse("xxx *bold* yyy").should == "xxx *bold* yyy"
+    end
+    it "parses *multiline\\nbold*" do
+      parse("*multiline\nbold*").should == "*multiline<br />bold*"
+    end
+    it "skips lone star inside bold block" do
+      parse("*aaa * bbb*").should == "*aaa * bbb*"
+    end
+    it "skips lone star" do
+      parse("aaa * bbb").should == 'aaa * bbb'
+    end
+    it "w/o closing tag" do
+      parse("*bold").should == '*bold'
+    end
+    it "nesting1 w/o closing tags" do
+      parse("*bold1 *bold2").should == "*bold1 *bold2"
+    end
+    it "nesting2 w/o closing tags" do
+      parse("*bold1  *bold2").should == "*bold1 *bold2"
+    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
+      s = "_italic_"
+      parse(s).should == s
+    end
+    it "at beginning" do
+      s = "_italic_\nxxx"
+      parse(s).should == s.gsub("\n","<br />")
+    end
+    it "in the middle of text" do
+      s = "xxx _italic_ yyy"
+      parse(s).should == s
+    end
+    it "parses _multiline\\nitalic_" do
+      s = "_multiline\nitalic_"
+      parse(s).should == s.gsub("\n","<br />")
+    end
+    it "skips lone underscore inside italic block" do
+      s = "_aaa _ bbb_"
+      parse(s).should == s
+    end
+    it "skips lone underscore" do
+      s = "aaa _ bbb"
+      parse(s).should == s
+    end
+    it "w/o closing tag" do
+      s = "_italic"
+      parse(s).should == s
+    end
+    it "nesting1 w/o closing tags" do
+      s = "_italic1 _italic2"
+      parse(s).should == s
+    end
+    it "nesting2 w/o closing tags" do
+      s = "_italic1  _italic2"
+      parse(s).should == s.gsub(/ +/,' ')
+    end
+  end
+
+###############################################################################
+
+  describe "combinations" do
+    it "bold in italic" do
+      s = "_aaa *bbb* ccc_"
+      parse(s).should == s
+    end
+    it "bold in italic - no closing1" do
+      s = "_aaa *bbb* ccc"
+      parse(s).should == s
+    end
+    it "bold in italic - no closing2" do
+      s = "_aaa *bbb ccc"
+      parse(s).should == s
+    end
+
+    it "italic in bold" do
+      s = "*aaa _bbb_ ccc*"
+      parse(s).should == s
+    end
+    it "italic in bold - no closing1" do
+      s = "*aaa _bbb_ ccc"
+      parse(s).should == s
+    end
+    it "italic in bold - no closing2" do
+      s = "*aaa _bbb ccc"
+      parse(s).should == s
+    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 == "#{c} aaa <a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a>"
+      end
+      it "raw text link inside #{l.upcase}> #2" do
+        s = "#{c} aaa http://www.ru\n#{c} bbb"
+        parse(s).should == "#{c} aaa <a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a><br />#{c} bbb"
+      end
+      it "raw text link inside #{l.upcase}> #3" do
+        s = "#{c} http://www.ru"
+        parse(s).should == "#{c} <a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a>"
+      end
+      it "raw text link inside #{l.upcase}> #4" do
+        s = "#{c} aaa http://www.ru bbb"
+        parse(s).should == "#{c} aaa <a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a> bbb"
+      end
+      it "two links inside #{l.upcase}>" do
+        s = "#{c} aaa http://www.ru http://ya.ru bbb"
+        parse(s).should == "#{c} aaa <a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a> <a rel=\"nofollow\" href=\"http://ya.ru\">http://ya.ru</a> bbb"
+      end
+    end
+  end
+
+###############################################################################
+
+  describe "IGNORES unnumbered lists" do
+    it "should ignore" do
+      parse("* a\n* b\n* c").should == "* a<br />* b<br />* c"
+    end
+    it "two lists" do
+      s = "* a\n* b\n* c"
+      s = s + "\nxxx\n" + s
+      parse(s).should == "* a<br />* b<br />* c<br />xxx<br />* a<br />* b<br />* c"
+    end
+    it "in middle of text when begins with space" do
+      parse("hello\n * a\n  * b\n * c\nworld").should ==
+        "hello<br />* a<br />* b<br />* c<br />world"
+    end
+    it "in middle of text" do
+      parse("hello\n* a\n* b\n* c\nworld").should ==
+        "hello<br />* a<br />* b<br />* c<br />world"
+    end
+    it "after blank line" do
+      parse("hello\n\n * a\n * b\n * c\nworld").should ==
+        "hello<br /><br />* a<br />* b<br />* c<br />world"
+    end
+  end
+
+###############################################################################
+
+  describe "IGNORES numbered lists" do
+    it "should ignore" do
+      parse("# a\n# b\n# c").should == "# a<br /># b<br /># c"
+    end
+    it "two lists" do
+      s = "# a\n# b\n# c"
+      s = s + "\nxxx\n" + s
+      parse(s).should == h(s).gsub("\n","<br />")
+    end
+    it "in middle of text when begins with space" do
+      parse("hello\n # a\n  # b\n # c\nworld").should ==
+        "hello<br /># a<br /># b<br /># c<br />world"
+    end
+    it "in middle of text" do
+      parse("hello\n# a\n# b\n# c\nworld").should ==
+        "hello<br /># a<br /># b<br /># c<br />world"
+    end
+    it "after blank line" do
+      parse("hello\n\n # a\n # b\n # c\nworld").should ==
+        "hello<br /><br /># a<br /># b<br /># c<br />world"
+    end
+  end
+
+###############################################################################
+
+  1.upto(5) do |lvl|
+    describe "H#{lvl}" do
+      it "at the beginning" do
+        s = "h#{lvl}. xxx"
+        parse(s).should == h(s)
+      end
+      it "after 1 line of text" do
+        s = "abcd\nh#{lvl}. xxx"
+        parse(s).should == h(s).gsub("\n","<br />")
+      end
+      it "after 2 lines of text" do
+        s = "abcd\ndefgh\nh#{lvl}. xxx"
+        parse(s).should == h(s).gsub("\n","<br />")
+      end
+      it "in middle of other words" do
+        s = "abcd defgh h#{lvl}. xxx yyy"
+        parse(s).should == h(s).gsub("\n","<br />")
+      end
+      it "in middle of other lines" do
+        s = "abcd defgh\nh#{lvl}. xxx\nyyy"
+        parse(s).should == h(s).gsub("\n","<br />")
+      end
+
+      it "does nothing special on spaces in id" do
+        s = "h#{lvl}. xxx   yyy z"
+        parse(s).should == h(s).gsub(/ +/,' ')
+      end
+      it "does nothing special on underscores in id" do
+        s = "h#{lvl}. xxx___yyy_z"
+        parse(s).should == h(s)
+      end
+      it "does nothing special on dashes in id" do
+        s = "h#{lvl}. xxx---yyy-z"
+        parse(s).should == h(s)
+      end
+      it "does nothing special on dots in id" do
+        s = "h#{lvl}. xxx...yyy.z"
+        parse(s).should == h(s)
+      end
+
+      %w'Ъ ъ : ; , привет" \' ! < >'.each do |c|
+        it "does nothing special on \"#{c}\"" do
+          s = "h#{lvl}. xxx#{c}yyy"
+          parse(s).should == h(s)
+        end
+      end
+    end
+  end
+
+###############################################################################
+
+  describe "raw text links" do
+    describe "starting with 'http://'" do
+      it "at the beginning" do
+        parse("http://asd.ru").should == "<a rel=\"nofollow\" href=\"http://asd.ru\">http://asd.ru</a>"
+      end
+      it "in middle of other words" do
+        parse("aaa bbb ccc http://asd.ru ddd eee fff").should ==
+          "aaa bbb ccc <a rel=\"nofollow\" href=\"http://asd.ru\">http://asd.ru</a> ddd eee fff"
+      end
+      it "in new line" do
+        parse("aaa bbb ccc\nhttp://asd.ru\nddd eee fff").should match(
+          %r"aaa bbb ccc ?<br /> ?<a rel=\"nofollow\" href=\"http://asd.ru\">http://asd.ru</a> ?<br /> ?ddd eee fff"
+        )
+      end
+      it "escapes '&' in link _text_" do
+        parse("http://asd.ru/?a=1&b=2").should == "<a rel=\"nofollow\" href=\"http://asd.ru/?a=1&b=2\">http://asd.ru/?a=1&amp;b=2</a>"
+      end
+
+      it "parses https://" do
+        parse("https://asd.ru").should == "<a rel=\"nofollow\" href=\"https://asd.ru\">https://asd.ru</a>"
+      end
+
+      %w', .'.each do |c|
+        it "stops parsing on \"#{c} \"" do
+          parse("http://asd.ru#{c}").should == "<a rel=\"nofollow\" href=\"http://asd.ru\">http://asd.ru</a>#{c}"
+          parse(" http://asd.ru#{c} ").should == "<a rel=\"nofollow\" href=\"http://asd.ru\">http://asd.ru</a>#{c}"
+          parse(" http://asd.ru#{c} hello!").should == "<a rel=\"nofollow\" href=\"http://asd.ru\">http://asd.ru</a>#{c} hello!"
+          parse("xxx http://asd.ru#{c} hello!").should == "xxx <a rel=\"nofollow\" href=\"http://asd.ru\">http://asd.ru</a>#{c} hello!"
+          parse(" http://asd.ru/#{c} hello!").should == "<a rel=\"nofollow\" href=\"http://asd.ru/\">http://asd.ru/</a>#{c} hello!"
+          parse(" http://aaa.com#{c} http://bbb.com").should ==
+            "<a rel=\"nofollow\" href=\"http://aaa.com\">http://aaa.com</a>#{c} <a rel=\"nofollow\" href=\"http://bbb.com\">http://bbb.com</a>"
+        end
+      end
+    end
+
+    describe "starting with 'www.'" do
+      it "at the beginning" do
+        parse("www.ru").should == "<a rel=\"nofollow\" href=\"http://www.ru\">www.ru</a>"
+      end
+      it "in middle of other words" do
+        parse("aaa bbb ccc www.ru ddd eee fff").should ==
+          "aaa bbb ccc <a rel=\"nofollow\" href=\"http://www.ru\">www.ru</a> ddd eee fff"
+      end
+      it "in new line" do
+        parse("aaa bbb ccc\nwww.ru\nddd eee fff").should match(
+          %r"aaa bbb ccc ?<br /> ?<a rel=\"nofollow\" href=\"http://www.ru\">www.ru</a> ?<br /> ?ddd eee fff"
+        )
+      end
+      it "escapes '&' in link _text_" do
+        parse("www.ru/?a=1&b=2").should == "<a rel=\"nofollow\" href=\"http://www.ru/?a=1&b=2\">www.ru/?a=1&amp;b=2</a>"
+      end
+
+      %w', .'.each do |c|
+        it "stops parsing on \"#{c} \"" do
+          parse("www.ru#{c}").should == "<a rel=\"nofollow\" href=\"http://www.ru\">www.ru</a>#{c}"
+          parse(" www.ru#{c} ").should == "<a rel=\"nofollow\" href=\"http://www.ru\">www.ru</a>#{c}"
+          parse(" www.ru#{c} hello!").should == "<a rel=\"nofollow\" href=\"http://www.ru\">www.ru</a>#{c} hello!"
+          parse("xxx www.ru#{c} hello!").should == "xxx <a rel=\"nofollow\" href=\"http://www.ru\">www.ru</a>#{c} hello!"
+          parse(" www.ru/#{c} hello!").should == "<a rel=\"nofollow\" href=\"http://www.ru/\">www.ru/</a>#{c} hello!"
+          parse(" www.aaa.com#{c} www.bbb.com").should ==
+            "<a rel=\"nofollow\" href=\"http://www.aaa.com\">www.aaa.com</a>#{c} <a rel=\"nofollow\" href=\"http://www.bbb.com\">www.bbb.com</a>"
+        end
+      end
+    end
+
+    describe 'e-mails' do
+      it "at the beginning" do
+        parse("aaa@bbb.com").should == "<a href=\"mailto:aaa@bbb.com\">aaa@bbb.com</a>"
+      end
+      it "in middle of other words" do
+        parse("aaa bbb ccc xx@yy.cn ddd eee fff").should ==
+          "aaa bbb ccc <a href=\"mailto:xx@yy.cn\">xx@yy.cn</a> ddd eee fff"
+      end
+      it "in new line" do
+        parse("aaa bbb ccc\naa.bb@cc.dd.ee\nddd eee fff").should match(
+          %r"aaa bbb ccc ?<br /> ?<a href=\"mailto:aa.bb@cc.dd.ee\">aa.bb@cc.dd.ee</a> ?<br /> ?ddd eee fff"
+        )
+      end
+
+      %w', .'.each do |c|
+        it "stops parsing on \"#{c} \"" do
+          parse("a-b@c-d.efghjikl#{c}").should == "<a href=\"mailto:a-b@c-d.efghjikl\">a-b@c-d.efghjikl</a>#{c}"
+          parse(" a-b@c-d.efghjikl#{c} ").should == "<a href=\"mailto:a-b@c-d.efghjikl\">a-b@c-d.efghjikl</a>#{c}"
+          parse(" a-b@c-d.efghjikl#{c} hello!").should == "<a href=\"mailto:a-b@c-d.efghjikl\">a-b@c-d.efghjikl</a>#{c} hello!"
+          parse("xxx a-b@c-d.efghjikl#{c} hello!").should == "xxx <a href=\"mailto:a-b@c-d.efghjikl\">a-b@c-d.efghjikl</a>#{c} hello!"
+          parse(" a-b@c-d.efghjikl#{c} hello!").should == "<a href=\"mailto:a-b@c-d.efghjikl\">a-b@c-d.efghjikl</a>#{c} hello!"
+          parse(" www@aaa.com#{c} www@bbb.com").should ==
+            "<a href=\"mailto:www@aaa.com\">www@aaa.com</a>#{c} <a href=\"mailto:www@bbb.com\">www@bbb.com</a>"
+        end
+      end
+
+      it "not parses bad emails" do
+        s="a@b.c a@b a.b@c a.b@@c a@b@c.d a#b@c.d"
+        parse(s).should == s
+      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
+
+    ".,;!?:-".each_char do |c|
+      it "uses '#{c}' as separator" do
+        l = '<a href="/spaces/test_space/tickets/1234">#1234</a>'
+        t = "L#{c}L"
+        parse(t.gsub('L','#1234')).should == t.gsub('L',l)
+        t = "L#{c}#{c}L"
+        parse(t.gsub('L','#1234')).should == t.gsub('L',l)
+        t = "#{c}L#{c}L#{c}"
+        parse(t.gsub('L','#1234')).should == t.gsub('L',l)
+      end
+    end
+  end
+
+###############################################################################
+
+  [
+    %w'<pre><code> </code></pre>',
+    %w'<pre> </pre>',
+    %w'<notextile> </notextile>'
+  ].each do |ot,ct|
+    # ot - opening tag
+    # ct - closing tag
+
+#    if ot == '<notextile>'
+#      ote,cte = '',''
+#    else
+      ote,cte = ot,ct
+#    end
+
+    describe "#{ot}..#{ct}" 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("#{ot}#{s.strip}#{ct}").should ==
+          h("#{ot}#{s.strip}#{ct}").gsub(/[ \t]+/,' ').gsub("\n","<br />")
+
+        s = <<-EOF
+            while ( 1 < 2 ) do
+              puts "<b>12345\\t54321</b>"
+              // *bold* comment
+              // _italic_ comment
+            end
+            ---
+            * aaa
+            * bbb
+            * ccc
+
+        EOF
+        parse("#{ot}#{s.strip}#{ct}").should ==
+          h("#{ote}#{s.strip}#{cte}").gsub(/[ \t]+/,' ').gsub(/\n */,"<br />")
+      end
+      it "not parses *bold*" do
+        s = "#{ot} *bold*#{ct}"
+        parse(s).should == h(s)
+      end
+      it "not parses _italic_" do
+        s = "#{ot} _italic_#{ct}"
+        parse(s).should == h(s)
+      end
+      it "not parses UL lists" do
+        s = "#{ot}\n * l1\n * l2\n * l3#{ct}"
+        parse(s).should == h(s).gsub("\n ","<br />")
+      end
+      it "not parses OL lists" do
+        s = "#{ot}\n # l1\n # l2\n # l3#{ct}"
+        parse(s).should == h(s).gsub("\n ","<br />")
+      end
+      it "not parses H1..H5" do
+        1.upto(5) do |i|
+          s = "#{ot}\nh#{i}. zzzzzzz\n#{ct}"
+          parse(s).should == h(s).gsub("\n","<br />")
+        end
+      end
+      it "parses raw text links" do
+        s = "#{ot}xxx http://www.ru yyy#{ct}"
+        parse(s).should == "#{h(ote)}xxx <a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a> yyy#{h(cte)}"
+        s = "#{ot}http://www.ru#{ct}"
+        parse(s).should == "#{h(ote)}<a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a>#{h(cte)}"
+      end
+      it "keeps newlines" do
+        s = "#{ot}aaa\nbbb#{ct}"
+        parse(s).should == h(s).gsub("\n","<br />")
+        s = "#{ot}aaa\n\nbbb\nccc#{ct}"
+        parse(s).should == h(s).gsub("\n","<br />")
+      end
+
+      it "w/o closing tags" do
+        s = "#{ot}aaa"
+        parse(s).should == h(s)
+      end
+
+      it "in middle of text" do
+        s = "xxx #{ot}yyyy#{ct} jjj"
+        parse(s).should == h(s)
+      end
+
+      it "with 2 instances" do
+        s = "xxx #{ot}yyyy#{ct} <jjj> #{ot}asdkjaslkd#{ct} END"
+        parse(s).should == h(s)
+      end
+
+      it "works with unicode" do
+        s = "привет #{ot} жжж #{ct} пока!"
+        parse(s).should == h(s)
+
+        s = "#{ot}абвгдеёжзийклмнопрстуфхцчшщьыъэюя#{ct}"
+        parse(s).should == h(s)
+
+        s = "#{ot}АБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЬЫЪЭЮЯ#{ct}"
+        parse(s).should == h(s)
+
+        s = "#{ot}☸☹☺☻☼☽☾☿#{ct}"
+        parse(s).should == h(s)
+      end
+
+      it "should escape lone closing tags" do
+        s = "#{ct}"
+        parse(s).should == h(s)
+      end
+
+      it "should skip newlines and spaces at end" do
+        s = "#{ot} aaa bbb ccc \n\n\n  \t\n\n\n\r\n\r\n   \t  #{ct}"
+        parse(s).should == "#{h(ote)} aaa bbb ccc<br /><br /><br /><br /><br /><br /><br /><br />#{h(cte)}"
+      end
+
+      it "escapes html chars" do
+        HTML_ESCAPE.each do |k,v|
+          parse("#{ot}#{k}#{ct}").should == h("#{ote}#{k}#{cte}")
+        end
+      end
+    end
+  end
+
+  describe "NOT PARSES <pre><code>..</code></pre>" do
+    it "with no spaces between <pre> and <code>" do
+      s = "<pre><code>aaa</code></pre>"
+      parse(s).should == h(s).gsub(/ +/,' ')
+    end
+
+    it "with spaces between <pre> and <code>" do
+      s = "<pre>    <code>aaa</code>           </pre>"
+      parse(s).should == h(s).gsub(/ +/,' ')
+    end
+    it "with spaces between <pre> and <code> and inside" do
+      s = "<pre>    <code>  aaa   bbb   </code>           </pre>"
+      parse(s).should == h(s).gsub(/ +/,' ')
+    end
+  end
+
+###############################################################################
+
+  describe "<code>..</code>" do
+    it "keeps <code> tags" do
+      s = "<code>aaa</code>"
+      parse(s).should == h(s)
+    end
+    it "strips heading & tailing whitespace" do
+      s = "<code> \r\n \t  \t\r  aaa \r\n\t  \t\r  </code>"
+      parse(s).should == "&lt;code&gt;<br /><br /> aaa<br /><br /> &lt;/code&gt;"
+    end
+    it "not parses *bold*" do
+      s = "<code>aaa *bbb* ccc</code>"
+      parse(s).should == h(s)
+    end
+    it "not parses _italic_" do
+      s = "<code>aaa _bbb_ ccc</code>"
+      parse(s).should == h(s)
+    end
+    it "not parses headers" do
+      s = "<code>aaa\nh1. bbb\nccc</code>"
+      parse(s).should == h(s).gsub("\n","<br />")
+    end
+    it "not parses <pre>" do
+      s = "<code>aaa <pre>bbb</pre> ccc</code>"
+      parse(s).should == h(s)
+    end
+    it "NOT closes unclosed <code>" do
+      s = "aaa <code>bbb"
+      parse(s).should == h(s)
+    end
+    it "NOT escapes '&'" do
+      s = "<code>aaa & bbb</code>"
+      parse(s).should == h(s)
+    end
+    it "parses links" do
+      s = "<code>aaa #1 #2 #3 http://www.ru [[wiki:jjj]] [[url:http://www.ru]] bbb</code>"
+      parse(s).should == "&lt;code&gt;aaa <a href=\"/spaces/test_space/tickets/1\">#1</a> <a href=\"/spaces/test_space/tickets/2\">#2</a> <a href=\"/spaces/test_space/tickets/3\">#3</a> <a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a> <a class=\"wiki_link\" title=\"jjj\" href=\"/spaces/test_space/wiki/jjj\">jjj</a> <a rel=\"nofollow\" href=\"http://www.ru\">http://www.ru</a> bbb&lt;/code&gt;"
+    end
+    it "NOT works" do
+      s = "<code>aaa & bbb</code> xxx <code>jjj&hhh</code>"
+      parse(s).should == h(s)
+      s = "<code>   aaa      </code>xxx<code>      jjj      </code>"
+      parse(s).should == h(s).gsub(/ +/,' ')
+    end
+    it "NOT keeps code bold" do
+      s = "*aaa <code>bbb</code> ccc*"
+      parse(s).should == h(s)
+      s = "*<code>aaa</code>*"
+      parse(s).should == h(s)
+    end
+    it "NOT keeps code italic" do
+      s = "_aaa <code>bbb</code> ccc_"
+      parse(s).should == h(s)
+      s = "_<code>aaa</code>_"
+      parse(s).should == h(s)
+    end
+  end
+
+###############################################################################
+
+  describe "Assembla Links" do
+    a = {}
+    a["wiki:Name"]  = '<a class="wiki_link" title="Name" href="/spaces/test_space/wiki/Name">Name</a>'
+    a["Name"]       = '<a class="wiki_link" title="Name" href="/spaces/test_space/wiki/Name">Name</a>'
+    a["Name#Ref"]   = '<a class="wiki_link" title="Name#Ref" href="/spaces/test_space/wiki/Name#h-Ref">Name#Ref</a>'
+    a["Name#h-Ref"]   = '<a class="wiki_link" title="Name#h-Ref" href="/spaces/test_space/wiki/Name#h-h-Ref">Name#h-Ref</a>'
+    a["#Ref"]       = '<a href="#h-Ref" title="#Ref" class="wiki_link">#Ref</a>'
+    a["#привет"]    = %Q|<a href="#h-#{hex_string("привет")}" title="#привет" class="wiki_link">#привет</a>|
+    a["#with  spc"] = %Q|<a href="#h-with__spc" title="#with  spc" class="wiki_link">#with  spc</a>|
+    a["#with__usc"] = %Q|<a href="#h-with__usc" title="#with__usc" class="wiki_link">#with__usc</a>|
+    a["#with--dsh"] = %Q|<a href="#h-with--dsh" title="#with--dsh" class="wiki_link">#with--dsh</a>|
+    a["#with!xclm"] = %Q|<a href="#h-#{hex_string("with!xclm")}" title="#with!xclm" class="wiki_link">#with!xclm</a>|
+    a["#with&amp"]  = %Q|<a href="#h-#{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="/code/test_space/git/changesets/1f4bdab77be696efd">revision:1f4bdab77be696efd</a>'
+    a["revision:12345"] =
+      '<a href="/code/test_space/subversion/changesets/12345">revision:12345</a>'
+    a["revision:567:1f4bdab77be696efd"] =
+      '<a href="/code/test_space/git-567/changesets/1f4bdab77be696efd">revision:1f4bdab77be696efd</a>'
+    a["revision:3:12345"] =
+      '<a href="/code/test_space/subversion-3/changesets/12345">revision:12345</a>'
+    a["r:2345"] = '<a href="/code/test_space/subversion/changesets/2345">revision:2345</a>'
+    a["r:2345ef"] = '<a href="/code/test_space/git/changesets/2345ef">revision:2345ef</a>'
+    a["r:10:2345"] = '<a href="/code/test_space/subversion-10/changesets/2345">revision:2345</a>'
+    a["r:1:2345ef"] = '<a href="/code/test_space/git-1/changesets/2345ef">revision:2345ef</a>'
+
+    a["url:http://www.ru"]   = '<a rel="nofollow" href="http://www.ru">http://www.ru</a>'
+    a["url:https://www.ru"]  = '<a rel="nofollow" href="https://www.ru">https://www.ru</a>'
+    a["url:www.ru"]          = '<a rel="nofollow" href="http://www.ru">http://www.ru</a>'
+    a["url:www.ru/?a=1&b=2"] = '<a rel="nofollow" href="http://www.ru/?a=1&amp;b=2">http://www.ru/?a=1&amp;b=2</a>'
+    a["url:ftp://www.ru"]    = '<a rel="nofollow" href="ftp://www.ru">ftp://www.ru</a>'
+    a["url:/spaces/x2"]      = '<a rel="nofollow" href="/spaces/x2">/spaces/x2</a>'
+
+    a["file:ExistingFile.txt"] =
+      '<a href="/spaces/test_space/documents/download/ExistingFile.txt">file:ExistingFile.txt</a>'
+    a["file:cVJUz6ejWr35pEab_qKWB8"] =
+      '<a href="/spaces/test_space/documents/download/cVJUz6ejWr35pEab_qKWB8">file:cVJUz6ejWr35pEab_qKWB8</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
+      if v['href="/'] && !k['url:']
+        it "parses [[#{k}]] with a site url" do
+          site_url = "http://www.ru"
+          parse("[[#{k}]]", :site_url => site_url).should ==
+            v.gsub('href="/',"href=\"#{site_url}/")
+
+          # with extraordinary slash
+          parse("[[#{k}]]", :site_url => "#{site_url}/").should ==
+            v.gsub('href="/',"href=\"#{site_url}/")
+
+          site_url = "http://127.0.0.1:3000"
+          parse("[[#{k}]]", :site_url => site_url).should ==
+            v.gsub('href="/',"href=\"#{site_url}/")
+
+          # with extraordinary slash
+          parse("[[#{k}]]", :site_url => "#{site_url}/").should ==
+            v.gsub('href="/',"href=\"#{site_url}/")
+        end
+      end
+      if v['/git/']
+        it "parses [[#{k}]] with custom git_url (String)" do
+          git_url = "http://www.ru/"
+          rev = k.split(':').last.tr(']','')
+          parse("[[#{k}]]", :git_url => git_url).should ==
+            v.sub('/code/test_space/git/changesets/',git_url)
+        end
+
+        it "parses [[#{k}]] with custom git_url (ObjProxy)" do
+          rev = k.split(':').last.tr(']','')
+          @asdfg = 'http://mmm.us'
+          git_url = Breakout::ObjProxy.new do
+            @asdfg + '/'
+          end
+          parse("[[#{k}]]", :git_url => git_url).should ==
+            v.sub('/code/test_space/git/changesets/',git_url)
+        end
+
+        it "parses [[#{k}]] with NULL git_url (ObjProxy)" do
+          rev = k.split(':').last.tr(']','')
+          git_url = Breakout::ObjProxy.new do
+            nil
+          end
+          parse("[[#{k}]]", :git_url => git_url).should == v
+        end
+
+        it "parses [[#{k}]] with FALSE git_url (ObjProxy)" do
+          rev = k.split(':').last.tr(']','')
+          git_url = Breakout::ObjProxy.new do
+            false
+          end
+          parse("[[#{k}]]", :git_url => git_url).should == v
+        end
+      end
+    end
+
+    it "should not instantiate ObjProxy's internal object if there are no git or svn links in text" do
+      git_url = Breakout::ObjProxy.new do
+        raise 'should not be raised'
+      end
+      lambda {
+        parse("no revision tag", :git_url => git_url).should == 'no revision tag'
+      }.should_not raise_error
+    end
+
+    describe "absolute_urls" do
+      # 'true' values
+      [true, 1, 'x'].each do |v|
+        it "should NOT convert relative url to absolute when absolute_urls = #{v.inspect} AND site_url is NULL" do
+          parse("[[url:/rel]]", :absolute_urls => v).should ==
+            '<a rel="nofollow" href="/rel">/rel</a>'
+          parse("[[url:/rel|text]]", :absolute_urls => v).should ==
+            '<a rel="nofollow" href="/rel">text</a>'
+        end
+        it "should convert relative url to absolute when absolute_urls = #{v.inspect}" do
+          parse("[[url:/rel]]", :absolute_urls => v, :site_url => 'http://www.ru').should ==
+            '<a rel="nofollow" href="http://www.ru/rel">/rel</a>'
+          parse("[[url:/rel|text]]", :absolute_urls => v, :site_url => 'http://www.ru').should ==
+            '<a rel="nofollow" href="http://www.ru/rel">text</a>'
+        end
+      end
+      # 'false' values
+      [false, nil].each do |v|
+        it "should not convert relative url to absolute when absolute_urls = #{v.inspect}" do
+          parse("[[url:/rel]]", :absolute_urls => v).should ==
+            '<a rel="nofollow" href="/rel">/rel</a>'
+          parse("[[url:/rel|text]]", :absolute_urls => v).should ==
+            '<a rel="nofollow" href="/rel">text</a>'
+        end
+      end
+    end
+
+    a = {}
+    a["image:ExistingImage.png"] =
+      '<img src="/spaces/test_space/documents/download/ExistingImage.png" alt="ALT" />'
+    a["image:cVJUz6ejWr35pEab_qKWB8"] =
+      '<img src="/spaces/test_space/documents/download/cVJUz6ejWr35pEab_qKWB8" alt="ALT" />'
+
+    a.each do |k,v|
+      it "parses [[#{k}]]" do
+        parse("[[#{k}]]").should == v.sub('ALT',k.sub('image:',''))
+      end
+      it "parses [[#{k}|привет тест]]" do
+        parse("[[#{k}|привет тест]]").should == v.sub('ALT','привет тест')
+      end
+      it "parses [[#{k}|test & here]]" do
+        parse("[[#{k}|test & here]]").should == v.sub('ALT','test & here')
+      end
+    end
+
+    it "ignores unknown link types" do
+      s = "[[zzz:xxx]]"
+      parse(s).should == s
+      s = "[[abcd:1234]]"
+      parse(s).should == s
+      s = "[[abcd::1234]] [[abcd:1234]] [[uri:ww.ru]]"
+      parse(s).should == s
+    end
+
+    it "ignores file & image links with forbidden symbols" do
+      s = "[[file:aaa/bbb]]"
+      parse(s).should == s
+      s = "[[file:aaa\\bbb]]"
+      parse(s).should == s
+      s = "[[file:aaa bbb]]"
+      parse(s).should == s
+
+      s = "[[image:aaa/bbb]]"
+      parse(s).should == s
+      s = "[[image:aaa\\bbb]]"
+      parse(s).should == s
+      s = "[[image:aaa bbb]]"
+      parse(s).should == s
+    end
+  end
+
+###############################################################################
+###############################################################################
+###############################################################################
+
+  unless defined?(HTML_ESCAPE)
+    HTML_ESCAPE = { '&' => '&amp;', '>' => '&gt;', '<' => '&lt;', '"' => '&quot;' }
+  end
+
+  def h(s)
+    s.to_s.gsub(/[&"><]/) { |special| HTML_ESCAPE[special] }
+  end
+
+  def parse(s, h = {})
+    h[:space_name] = "test_space" unless h.key?(:space_name)
+    BreakoutParser.parse_links_only(s,
+      h[:space_name],
+      h[:site_url],
+      h[:git_url],
+      h[:absolute_urls],
+      h[:large_files_url],
+      h[:meta_attributes]
+    ).strip
+  end
+end