breakout_parser 0.0.11 → 0.0.12

data/ext/breakout_parser/parser.tab.h

@@ -51,35 +51,37 @@
      GIT_REVISION_LINK = 267,
      WIKI_LINK = 268,
      ANCHOR_LINK = 269,
-     URL_WITH_PROTO_LINK = 270,
-     URL_WITHOUT_PROTO_LINK = 271,
-     FILE_LINK = 272,
-     IMAGE_LINK = 273,
-     URL = 274,
-     EMAIL = 275,
-     UL = 276,
-     H1 = 277,
-     H2 = 278,
-     H3 = 279,
-     H4 = 280,
-     H5 = 281,
-     INLINE_CODE = 282,
-     SPACE = 283,
-     BR = 284,
-     OLI = 285,
-     ULI = 286,
-     PRE_CODE_START = 287,
-     PRE_CODE_END = 288,
-     PRE_START = 289,
-     PRE_END = 290,
-     CODE_START = 291,
-     CODE_END = 292,
-     NOTEXTILE_START = 293,
-     NOTEXTILE_END = 294,
-     BOLD_END = 295,
-     ITALIC_END = 296,
-     REVERT_BOLD = 297,
-     REVERT_ITALIC = 298
+     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,
+     PRE_CODE_START = 289,
+     PRE_CODE_END = 290,
+     PRE_START = 291,
+     PRE_END = 292,
+     CODE_START = 293,
+     CODE_END = 294,
+     NOTEXTILE_START = 295,
+     NOTEXTILE_END = 296,
+     BOLD_END = 297,
+     ITALIC_END = 298,
+     REVERT_BOLD = 299,
+     REVERT_ITALIC = 300
    };
 #endif
 
@@ -90,7 +92,7 @@ typedef union YYSTYPE
 {
 
 /* Line 1676 of yacc.c  */
-#line 61 "parser.y"
+#line 66 "parser.y"
 
         double dvalue;
         int ivalue;
@@ -99,7 +101,7 @@ typedef union YYSTYPE
 
 
 /* Line 1676 of yacc.c  */
-#line 103 "parser.tab.h"
+#line 105 "parser.tab.h"
 } YYSTYPE;
 # define YYSTYPE_IS_TRIVIAL 1
 # define yystype YYSTYPE /* obsolescent; will be withdrawn */

data/ext/breakout_parser/parser.y

@@ -20,6 +20,11 @@ size_t bufsize = 0;
 const char *space_name = NULL;
 size_t space_name_len = 0;
 
+const char *site_url = NULL;
+size_t site_url_len = 0;
+
+extern VALUE git_url;
+
 #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)));*/ \
@@ -68,6 +73,7 @@ void yyerror(const char *msg)
 %token <ivalue> T_CHAR BOLD_START ITALIC_START
 %token <ivalue> BOLD_ITALIC_START ITALIC_BOLD_START
 %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
 %token <svalue> FILE_LINK IMAGE_LINK
 %token <svalue> URL EMAIL
@@ -130,8 +136,10 @@ word : chars
       | REVERT_ITALIC  {revert_italic()}
 
 link: TICKET_LINK            {process_ticket_link($1)}
-    | SVN_REVISION_LINK      {process_svn_link($1)}
-    | GIT_REVISION_LINK      {process_git_link($1)}
+    | SVN_REVISION_LINK      {process_svn_link($1,0)}
+    | GIT_REVISION_LINK      {process_git_link($1,0)}
+    | SVN_N_REVISION_LINK    {process_svn_link($1,1)}
+    | GIT_N_REVISION_LINK    {process_git_link($1,1)}
     | URL_WITH_PROTO_LINK    {process_url_link($1,NULL)}
     | URL_WITHOUT_PROTO_LINK {process_url_link($1,"http://")}
     | WIKI_LINK              {process_wiki_link($1)}
@@ -286,29 +294,72 @@ process_url_link(const char*target,const char* proto){
     process_link_tail(target,NULL,proto);
 }
 
-process_svn_link(const char*target){
+concat_site_url(){
+    if( site_url && site_url_len > 0 ){
+        concat(site_url, site_url_len);
+    }
+}
+
+process_svn_link(const char*target, int numbered_repo){
     const char *c;
-    concat("<a href=\"/code/",15);
+    // can use sprintf here.. but I think it's a way slower than raw concat
+    concat("<a href=\"",9);
+    concat_site_url();
+    concat("/code/",6);
     concat(space_name,space_name_len);
-    concat("/subversion/changesets/",23);
+    if( numbered_repo ){
+        concat("/subversion-",12);
+        while( *target >= '0' && *target <= '9' ) concat_raw_char(*target++);
+        while( *target == ':' ) target++;
+        concat("/changesets/",12);
+    } else {
+        concat("/subversion/changesets/",23);
+    }
     for(c=target; *c && *c!=']' && *c!='|'; c++) concat_raw_char(*c);
     process_link_tail(target,NULL,"revision:");
 }
 
-process_git_link(const char*target){
+int concat_custom_git_url(){
     const char *c;
-    concat("<a href=\"/code/",15);
-    concat(space_name,space_name_len);
-    concat("/git/changesets/",16);
+    if( git_url && (c = StringValuePtr(git_url)) ){
+        size_t l = strlen(c);
+        if( l>0 ){
+            concat2(c);
+            return 1;
+        }
+    }
+    return 0;
+}
+
+process_git_link(const char*target, int numbered_repo){
+    const char *c;
+    // can use sprintf here.. but I think it's a way slower than raw concat
+    concat("<a href=\"",9);
+    if( !concat_custom_git_url() ){
+        concat_site_url();
+        concat("/code/",6);
+        concat(space_name,space_name_len);
+        if( numbered_repo ){
+            concat("/git-",5);
+            while( *target >= '0' && *target <= '9' ) concat_raw_char(*target++);
+            while( *target == ':' ) target++;
+            concat("/changesets/",12);
+        } else {
+            concat("/git/changesets/",16);
+        }
+    }
     for(c=target; *c && *c!=']' && *c!='|'; c++) concat_raw_char(*c);
     process_link_tail(target,NULL,"revision:");
 }
 
 process_wiki_link(const char*target){
     const char *c;
+    // can use sprintf here.. but I think it's a way slower than raw concat
     concat("<a class=\"wiki_link\" title=\"",28);
     for(c=target; *c && *c!=']' && *c!='|'; c++) concat_raw_char(*c);
-    concat("\" href=\"/wiki/show/",19);
+    concat("\" href=\"",8);
+    concat_site_url();
+    concat("/wiki/show/",11);
     concat(space_name,space_name_len);
     concat_raw_char('/');
     for(c=target; *c && *c!=']' && *c!='|'; c++) concat_raw_char(*c);
@@ -317,7 +368,10 @@ process_wiki_link(const char*target){
 
 process_file_link(const char*target){
     const char *c;
-    concat("<a href=\"/spaces/",17);
+    // can use sprintf here.. but I think it's a way slower than raw concat
+    concat("<a href=\"",9);
+    concat_site_url();
+    concat("/spaces/",8);
     concat(space_name,space_name_len);
     concat("/documents/download/",20);
     for(c=target; *c && *c!=']' && *c!='|'; c++) concat_raw_char(*c);
@@ -326,7 +380,10 @@ process_file_link(const char*target){
 
 process_image_link(const char*target){
     const char *c, *p;
-    concat("<img src=\"/spaces/",18);
+    // can use sprintf here.. but I think it's a way slower than raw concat
+    concat("<img src=\"",10);
+    concat_site_url();
+    concat("/spaces/",8);
     concat(space_name,space_name_len);
     concat("/documents/download/",20);
     for(c=target; *c && *c!=']' && *c!='|'; c++) concat_raw_char(*c);
@@ -342,7 +399,10 @@ process_image_link(const char*target){
 process_ticket_link(const char*ticket_id){
     const char *c;
     while(*ticket_id && (*ticket_id < '0' || *ticket_id > '9') ) ticket_id++;
-    concat("<a href=\"/spaces/",17);
+    // can use sprintf here.. but I think it's a way slower than raw concat
+    concat("<a href=\"",9);
+    concat_site_url();
+    concat("/spaces/",8);
     concat(space_name,space_name_len);
     concat("/tickets/",9);
     for(c=ticket_id; *c && *c>='0' && *c<='9'; c++) concat_raw_char(*c);
@@ -375,36 +435,6 @@ concat_raw_char(int c){
     *bufptr++ = c;
 }
 
-/*
-concat_escaping_html(const char*what){
-    char *po = buf + strlen(buf);
-    const char *pi = what;
-
-    for(;*pi;pi++){
-        switch(*pi){
-            case '<':
-                strcpy(po,"&lt;");
-                po += 4;
-                break;
-            case '>':
-                strcpy(po,"&gt;");
-                po += 4;
-                break;
-            case '"':
-                strcpy(po,"&quot;");
-                po += 6;
-                break;
-            case '&':
-                strcpy(po,"&amp;");
-                po += 5;
-                break;
-            default:
-                *po++ = *pi;
-        }
-    }
-    *po = 0;
-}*/
-
 process_url(const char*url){
     const char *p;
 

data/ext/breakout_parser/ruby_ext.c

@@ -3,15 +3,15 @@
 #include "ruby.h"
 
 void Init_breakout_parser();
-VALUE method_parse(VALUE, VALUE, VALUE);
-VALUE method_parse_links_only(VALUE, VALUE, VALUE);
+VALUE method_parse(int, VALUE*, VALUE);
+VALUE method_parse_links_only(int, 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, 2);
-    rb_define_singleton_method(breakout_parser, "parse_links_only", method_parse_links_only, 2);
+    rb_define_singleton_method(breakout_parser, "parse", method_parse, -1);
+    rb_define_singleton_method(breakout_parser, "parse_links_only", method_parse_links_only, -1);
 }
 
 extern char *buf, *bufptr;
@@ -19,12 +19,40 @@ extern char *in_buf, *in_pos;
 extern const char *space_name;
 extern size_t in_buf_len, bufsize, space_name_len;
 
+extern const char *site_url;
+extern size_t site_url_len;
+
+VALUE git_url;
+
 extern int parse_links_only;
 
-VALUE do_parse(VALUE self, VALUE text, VALUE r_space_name) {
-    VALUE s;
+VALUE do_parse(int argc, VALUE *argv, VALUE self) {
+    VALUE s, text, r_space_name;
     char *p;
 
+    if( argc < 2 || argc > 4 ){
+        rb_raise(rb_eArgError, "wrong number of arguments (%d for 2..4)", argc);
+        return rb_str_new("",0); // unreachable code, but for double safety
+    }
+
+    text = argv[0];
+    r_space_name = argv[1];
+
+    site_url = NULL; site_url_len = 0;
+    if( argc > 2 && argv[2] != Qnil && argv[2] != Qfalse ){
+        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--; 
+        }
+    }
+
+    git_url = NULL;
+    if( argc > 3 && argv[3] != Qnil && argv[3] != Qfalse ){
+        git_url = argv[3];
+    }
+
     if(!text || text == Qnil || text == Qfalse){
         // NULL input string
         return rb_str_new("",0);
@@ -73,14 +101,14 @@ VALUE do_parse(VALUE self, VALUE text, VALUE r_space_name) {
     return s;
 }
 
-VALUE method_parse(VALUE self, VALUE text, VALUE r_space_name) {
+VALUE method_parse(int argc, VALUE *argv, VALUE self) {
     parse_links_only = 0;
-    return do_parse(self,text,r_space_name);
+    return do_parse(argc, argv, self);
 }
 
-VALUE method_parse_links_only(VALUE self, VALUE text, VALUE r_space_name) {
+VALUE method_parse_links_only(int argc, VALUE *argv, VALUE self) {
     parse_links_only = 1;
-    return do_parse(self,text,r_space_name);
+    return do_parse(argc, argv, self);
 }
 
 #endif // ifdef RUBY_VERSION

data/spec/parser_spec.rb

@@ -1,4 +1,5 @@
 require 'breakout_parser'
+require File.dirname(__FILE__) + '/obj_proxy'
 
 describe 'BreakoutParser' do
   def self.hex_string s
@@ -10,6 +11,19 @@ describe 'BreakoutParser' do
   end
   def hex_string s; self.class.hex_string(s); end
 
+  it 'accepts from 2 to 4 arguments' do
+    [0,1,5,6,7,8,9,10].each do |argc|
+      lambda{
+        BreakoutParser.parse(*(['a']*argc))
+      }.should raise_error(ArgumentError, "wrong number of arguments (#{argc} for 2..4)")
+    end
+    (2..4).each do |argc|
+      lambda{
+        BreakoutParser.parse(*(['a']*argc))
+      }.should_not raise_error
+    end
+  end
+
   it 'converts \n to <br />' do
     parse("aaa\nbbb").should match(%r"aaa ?<br /> ?bbb")
   end
@@ -31,22 +45,22 @@ describe 'BreakoutParser' do
 
   it "handles nil & false text well" do
     parse(false).should == ""
-    parse(false,false).should == ""
-    parse("",false).should == ""
+    parse(false, :space_name => false).should == ""
+    parse("",    :space_name => false).should == ""
     parse(nil).should == ""
-    parse(nil,nil).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",nil)
+      parse("#123", :space_name => nil)
     }.should raise_error(TypeError)
   end
 
   it "handles false space_name well" do
     lambda{
-      parse("#123",false)
+      parse("#123", :space_name => false)
     }.should raise_error(TypeError)
   end
 
@@ -740,8 +754,14 @@ describe 'BreakoutParser' do
       '<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>'
@@ -765,6 +785,69 @@ describe 'BreakoutParser' do
       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's no git links in text" do
+      git_url = Breakout::ObjProxy.new do
+        raise 'should not be raised'
+      end
+      lambda {
+        parse("[[r:2345]]", :git_url => git_url).should ==
+          '<a href="/code/test_space/subversion/changesets/2345">revision:2345</a>'
+      }.should_not raise_error
     end
 
     a = {}
@@ -823,7 +906,8 @@ describe 'BreakoutParser' do
     s.to_s.gsub(/[&"><]/) { |special| HTML_ESCAPE[special] }
   end
 
-  def parse s, space_name = "test_space"
-    BreakoutParser.parse(s, space_name).strip
+  def parse s, h = {}
+    h[:space_name] = "test_space" unless h.key?(:space_name)
+    BreakoutParser.parse(s, h[:space_name], h[:site_url], h[:git_url]).strip
   end
 end