breakout_parser 0.0.12 → 0.0.13

data/ext/breakout_parser/parser.tab.h

@@ -44,32 +44,32 @@
      ITALIC_START = 260,
      BOLD_ITALIC_START = 261,
      ITALIC_BOLD_START = 262,
-     T_WORD = 263,
-     TICKET_LINK = 264,
-     LINK = 265,
-     SVN_REVISION_LINK = 266,
-     GIT_REVISION_LINK = 267,
-     WIKI_LINK = 268,
-     ANCHOR_LINK = 269,
-     SVN_N_REVISION_LINK = 270,
-     GIT_N_REVISION_LINK = 271,
-     URL_WITH_PROTO_LINK = 272,
-     URL_WITHOUT_PROTO_LINK = 273,
-     FILE_LINK = 274,
-     IMAGE_LINK = 275,
-     URL = 276,
-     EMAIL = 277,
-     UL = 278,
-     H1 = 279,
-     H2 = 280,
-     H3 = 281,
-     H4 = 282,
-     H5 = 283,
-     INLINE_CODE = 284,
-     SPACE = 285,
-     BR = 286,
-     OLI = 287,
-     ULI = 288,
+     ULI = 263,
+     OLI = 264,
+     T_WORD = 265,
+     TICKET_LINK = 266,
+     LINK = 267,
+     SVN_REVISION_LINK = 268,
+     GIT_REVISION_LINK = 269,
+     WIKI_LINK = 270,
+     ANCHOR_LINK = 271,
+     SVN_N_REVISION_LINK = 272,
+     GIT_N_REVISION_LINK = 273,
+     URL_WITH_PROTO_LINK = 274,
+     URL_WITHOUT_PROTO_LINK = 275,
+     FILE_LINK = 276,
+     IMAGE_LINK = 277,
+     URL = 278,
+     EMAIL = 279,
+     UL = 280,
+     H1 = 281,
+     H2 = 282,
+     H3 = 283,
+     H4 = 284,
+     H5 = 285,
+     INLINE_CODE = 286,
+     SPACE = 287,
+     BR = 288,
      PRE_CODE_START = 289,
      PRE_CODE_END = 290,
      PRE_START = 291,
@@ -92,7 +92,7 @@ typedef union YYSTYPE
 {
 
 /* Line 1676 of yacc.c  */
-#line 66 "parser.y"
+#line 68 "parser.y"
 
         double dvalue;
         int ivalue;

data/ext/breakout_parser/parser.y

@@ -25,6 +25,8 @@ size_t site_url_len = 0;
 
 extern VALUE git_url;
 
+int list_level = 1;
+
 #define CHECK_BUF_SIZE(len) \
     if( (bufptr - buf + len + 1) >= bufsize ){ \
         /*printf("[.] REALLOC oldsz=%d, newsz=%d\n",bufsize, (bufsize+((len > 0x1000) ? (len+0x1000) : 0x1000)));*/ \
@@ -72,6 +74,7 @@ void yyerror(const char *msg)
 
 %token <ivalue> T_CHAR BOLD_START ITALIC_START
 %token <ivalue> BOLD_ITALIC_START ITALIC_BOLD_START
+%token <ivalue> ULI OLI
 %token <svalue> T_WORD TICKET_LINK LINK SVN_REVISION_LINK GIT_REVISION_LINK WIKI_LINK ANCHOR_LINK
 %token <svalue> SVN_N_REVISION_LINK GIT_N_REVISION_LINK
 %token <svalue> URL_WITH_PROTO_LINK URL_WITHOUT_PROTO_LINK
@@ -80,7 +83,7 @@ void yyerror(const char *msg)
 %token <svalue> UL
 %token <svalue> H1 H2 H3 H4 H5
 %token <svalue> INLINE_CODE
-%token SPACE BR /*BRBR*/ OLI ULI
+%token SPACE BR /*BRBR*/
 %token PRE_CODE_START PRE_CODE_END PRE_START PRE_END CODE_START CODE_END
 %token NOTEXTILE_START NOTEXTILE_END
 %token BOLD_END ITALIC_END
@@ -101,8 +104,20 @@ textitem: br
      | h3 {concat("</h3>",5)}
      | h4 {concat("</h4>",5)}
      | h5 {concat("</h5>",5)}
-     | {concat("<ul>",4)} ulist {concat("</ul>",5)} textitem
-     | {concat("<ol>",4)} olist {concat("</ol>",5)} textitem
+     | {
+            list_level=1; 
+            concat("<ul>",4)
+       } ulist { 
+            concat("</ul>",5); 
+            for(; list_level>1 && list_level<4; list_level--) concat("</li></ul>",10);
+       } textitem
+     | {
+            list_level=1; 
+            concat("<ol>",4)
+       } olist { 
+            concat("</ol>",5); 
+            for(; list_level>1 && list_level<4; list_level--) concat("</li></ol>",10);
+       } textitem
      | code
 
 ulist: ulitem        {concat("</li>",5)}
@@ -157,14 +172,14 @@ char : T_CHAR {concat_escaped_char($1)}
 
 //raw_char : T_CHAR {concat_raw_char($1)}
 
-h1   : H1          {concat("<h1 id=\"",8); process_header($1)}
-h2   : H2          {concat("<h2 id=\"",8); process_header($1)}
-h3   : H3          {concat("<h3 id=\"",8); process_header($1)}
-h4   : H4          {concat("<h4 id=\"",8); process_header($1)}
-h5   : H5          {concat("<h5 id=\"",8); process_header($1)}
+h1   : H1          {concat("<h1 id=\"h-",10); process_header($1)}
+h2   : H2          {concat("<h2 id=\"h-",10); process_header($1)}
+h3   : H3          {concat("<h3 id=\"h-",10); process_header($1)}
+h4   : H4          {concat("<h4 id=\"h-",10); process_header($1)}
+h5   : H5          {concat("<h5 id=\"h-",10); process_header($1)}
 //ul   : UL          {concat("<ul>",4)}
-oli  : OLI         {concat("<li>",4)}
-uli  : ULI         {concat("<li>",4)}
+oli  : OLI         {process_oli($1)}
+uli  : ULI         {process_uli($1)}
 br   : BR          {concat("<br />",6)}
 //     | BRBR        {concat("<br /><br />",12)}
 
@@ -177,6 +192,36 @@ code : PRE_CODE_START {concat("<pre><code>",11)} chars PRE_CODE_END {concat("</c
 
 %%
 
+process_uli(int level){
+    if( level == list_level ){
+        concat("<li>",4);
+    } else if( level < list_level ){
+        list_level = level;
+        //unconcat("</li>");
+        concat("</ul></li><li>",14);
+    } else {
+        // if(level > list_level)
+        list_level = level;
+        unconcat("</li>");
+        concat("<ul><li>",8);
+    }
+}
+
+process_oli(int level){
+    if( level == list_level ){
+        concat("<li>",4);
+    } else if( level < list_level ){
+        list_level = level;
+        //unconcat("</li>");
+        concat("</ol></li><li>",14);
+    } else {
+        // if(level > list_level)
+        list_level = level;
+        unconcat("</li>");
+        concat("<ol><li>",8);
+    }
+}
+
 concat_hex_char(char c){
     unsigned char d;
     d = ((unsigned char)c)>>4;
@@ -274,7 +319,7 @@ process_anchor_link(const char*target){
 
     if((p = strchr(target,'|')) && (p<pend) && (p>target)) pend = p-1;
 
-    concat("<a href=\"#",10);
+    concat("<a href=\"#h-",12);
     if( need_hex_convert(target,pend) ){
         for(p = target; *p && p<=pend; p++) concat_hex_char( *p );
     } else {
@@ -362,7 +407,16 @@ process_wiki_link(const char*target){
     concat("/wiki/show/",11);
     concat(space_name,space_name_len);
     concat_raw_char('/');
-    for(c=target; *c && *c!=']' && *c!='|'; c++) concat_raw_char(*c);
+    for(c=target; *c && *c!=']' && *c!='|' && *c!='#'; c++) concat_raw_char(*c);
+    if( *c == '#' ){
+        concat_raw_char('#');
+        //if(memcmp(c, "#h-", 3) != 0){
+            // anchor w/o "h-" prefix, we need to add it
+            concat_raw_char('h');
+            concat_raw_char('-');
+        //}
+        for(c++; *c && *c!=']' && *c!='|'; c++) concat_raw_char(*c);
+    }
     process_link_tail(target,NULL,NULL);
 }
 
@@ -457,6 +511,11 @@ process_email(const char*url){
     process_link_tail(url,NULL,NULL);
 }
 
+unconcat(const char*what){
+    int l = strlen(what);
+    if( bufptr-buf > l && strncmp(bufptr-l,what,l) == 0 ) bufptr -= l;
+}
+
 revert_bold(){
     char *p;
     for( p=bufptr-1; p >= (buf+7) ; p--){

data/spec/parser_spec.rb

@@ -11,6 +11,19 @@ describe 'BreakoutParser' do
   end
   def hex_string s; self.class.hex_string(s); end
 
+  def unformat s
+    s.strip.split("\n").map(&:strip).join
+  end
+
+  def get_data fname
+    r = File.read(File.dirname(__FILE__) + '/data/' + fname)
+    if self.class.description[' numbered list multilevel']
+      r.gsub!('*','#')
+      r.gsub!('ul>','ol>')
+    end
+    r
+  end
+
   it 'accepts from 2 to 4 arguments' do
     [0,1,5,6,7,8,9,10].each do |argc|
       lambda{
@@ -319,6 +332,31 @@ describe 'BreakoutParser' 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
+
+    describe "multilevel" do
+      it "at start" do
+        # NOTE: not sure that list1.html is correct enough, but it should render fine
+        parse(get_data('list1')).should == unformat(get_data('list1.html'))
+      end
+
+      it "in middle" do
+        parse(get_data('list2')).should == unformat(get_data('list2.html'))
+      end
+
+      it "three levels" do
+        parse(get_data('list3')).should == unformat(get_data('list3.html'))
+      end
+
+      it "mess - should have matching count of opening and closing tags" do
+        r = parse(get_data('list4'))
+        r.scan('<ul>').count.should > 0
+        r.scan('<ul>').count.should <= r.scan('</ul>').count
+      end
+
+      it "at end" do
+        parse(get_data('list5')).should == unformat(get_data('list5.html'))
+      end
+    end
   end
 
 ###############################################################################
@@ -347,58 +385,84 @@ describe 'BreakoutParser' 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
+
+    describe "multilevel" do
+      it "at start" do
+        # NOTE: not sure that list1.html is correct enough, but it should render fine
+        parse(get_data('list1')).should == unformat(get_data('list1.html'))
+      end
+
+      it "in middle" do
+        parse(get_data('list2')).should == unformat(get_data('list2.html'))
+      end
+
+      it "three levels" do
+        parse(get_data('list3')).should == unformat(get_data('list3.html'))
+      end
+
+      it "mess - should have matching count of opening and closing tags" do
+        r = parse(unformat(get_data('list4')))
+        r.scan('<ol>').count.should > 0
+        r.scan('<ol>').count.should <= r.scan('</ol>').count
+      end
+
+      it "at end" do
+        parse(get_data('list5')).should == unformat(get_data('list5.html'))
+      end
+    end
   end
 
 ###############################################################################
+# headers
 
   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}>"
+        parse("h#{lvl}. xxx").should == "<h#{lvl} id=\"h-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}>"
+        parse("abcd\nh#{lvl}. xxx").should == "abcd<br /><h#{lvl} id=\"h-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}>"
+        parse("abcd\ndefgh\nh#{lvl}. xxx").should == "abcd<br />defgh<br /><h#{lvl} id=\"h-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"
+        parse("abcd defgh\nh#{lvl}. xxx\nyyy").should == "abcd defgh<br /><h#{lvl} id=\"h-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}>"
+        parse("h#{lvl}. xxx   yyy z").should == "<h#{lvl} id=\"h-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}>"
+        parse("h#{lvl}. xxx___yyy_z").should == "<h#{lvl} id=\"h-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}>"
+        parse("h#{lvl}. xxx---yyy-z").should == "<h#{lvl} id=\"h-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}>"
+        parse("h#{lvl}. xxx...yyy.z").should == "<h#{lvl} id=\"h-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}>"
+          parse("h#{lvl}. xxx#{c}yyy").should == "<h#{lvl} id=\"h-#{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}>"
+        parse("h#{lvl}.  \t  xxx   \t ").should == "<h#{lvl} id=\"h-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\ryyy").should == "<h#{lvl} id=\"h-xxx\">xxx</h#{lvl}><br />yyy"
+        parse("h#{lvl}. xxx\r").should == "<h#{lvl} id=\"h-xxx\">xxx</h#{lvl}>"
 
         parse("h#{lvl}. xxx\r yyy").sub(' yyy','yyy').should ==
-          "<h#{lvl} id=\"xxx\">xxx</h#{lvl}><br />yyy"
+          "<h#{lvl} id=\"h-xxx\">xxx</h#{lvl}><br />yyy"
       end
     end
   end
@@ -740,14 +804,15 @@ describe 'BreakoutParser' 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["Name#Ref"]   = '<a class="wiki_link" title="Name#Ref" href="/wiki/show/test_space/Name#h-Ref">Name#Ref</a>'
+    a["Name#h-Ref"]   = '<a class="wiki_link" title="Name#h-Ref" href="/wiki/show/test_space/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"] =

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: breakout_parser
 version: !ruby/object:Gem::Version 
-  version: 0.0.12
+  version: 0.0.13
 platform: ruby
 authors: 
 - Andrey "Zed" Zaikin
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-04-01 00:00:00 +06:00
+date: 2010-04-27 00:00:00 +06:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency