url matching, refinement necessary, use of positive lookahead not avoided, correction and rebuilt

4 files changed, 10 insertions, 6 deletions

diff --git a/lib/sisu/v0/html_tune.rb b/lib/sisu/v0/html_tune.rb
index 66c45aed..ac8d6594 100644
--- a/lib/sisu/v0/html_tune.rb
+++ b/lib/sisu/v0/html_tune.rb
@@ -327,7 +327,7 @@ module SiSU_Tune
           end
           para.gsub!(/\b[_\\]((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,'<a href="\1" target="_top">\1</a>\2') #http ftp matches escaped, no decoration
           para.gsub!(/((?:^|\s)[}])((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,'\1<a href="\2" target="_top">\2</a>\3') #special case \{ e.g. \}http://url
-          para.gsub!(/\b((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,%{#{@url_brace.xml_open}<a href="\\1" target="_top">\\1</a>#{@url_brace.xml_close}\\2}) #http ftp matches with decoration
+          para.gsub!(/(^|\s)((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?=\s|$))/,%{\\1#{@url_brace.xml_open}<a href="\\2" target="_top">\\2</a>#{@url_brace.xml_close}\\3}) #http ftp matches with decoration
           if (para =~/..\/\S+/ and para !~/(\"..\/\S+?\"|>\s*..\/\S+<)/)
             para.gsub!(/(\.\.\/\S+)/,'<a href="\1">\1</a>')
           end
diff --git a/lib/sisu/v0/odf.rb b/lib/sisu/v0/odf.rb
index fbd4cc62..ff788116 100644
--- a/lib/sisu/v0/odf.rb
+++ b/lib/sisu/v0/odf.rb
@@ -299,8 +299,10 @@ module SiSU_ODF
           %{<text:a xlink:type="simple" xlink:href="\\1">\\1</text:a>\\2}) #http ftp matches escaped, no decoration
         para.gsub!(/((?:^|\s)[}])((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,
           %{\\1<text:a xlink:type="simple" xlink:href="\\2">\\2</text:a>\\3}) #special case \{ e.g. \}http://url
-        para.gsub!(/\b((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,
-          %{#{@url_brace.xml_open}<text:a xlink:type="simple" xlink:href="\\1">\\1</text:a>#{@url_brace.xml_close}\\2}) #http ftp matches with decoration
+        para.gsub!(/(^|\s)((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?=\s|$))/,
+          %{\\1#{@url_brace.xml_open}<text:a xlink:type="simple" xlink:href="\\2">\\2</text:a>#{@url_brace.xml_close}\\3}) #http ftp matches with decoration
+        #para.gsub!(/\b((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/, also works
+          #%{#{@url_brace.xml_open}<text:a xlink:type="simple" xlink:href="\\1">\\1</text:a>#{@url_brace.xml_close}\\2}) #http ftp matches with decoration
         para.gsub!(/([a-zA-Z0-9._-]+@[a-zA-Z0-9._-]+)/,
           %{#{@url_brace.xml_open}<text:a xlink:type="simple" xlink:href="mailto:\\1">\\1</text:a>#{@url_brace.xml_close}})
         para=case para
diff --git a/lib/sisu/v0/shared_html_lite.rb b/lib/sisu/v0/shared_html_lite.rb
index 16491ebf..43fb4446 100644
--- a/lib/sisu/v0/shared_html_lite.rb
+++ b/lib/sisu/v0/shared_html_lite.rb
@@ -133,7 +133,7 @@ module SiSU_Format_Shared
       end
       para.gsub!(/\b[_\\]((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,'<a href="\1" target="_top">\1</a>\2') #http ftp matches escaped, no decoration
       para.gsub!(/((?:^|\s)[}])((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,'\1<a href="\2" target="_top">\2</a>\3') #special case \{ e.g. \}http://url
-      para.gsub!(/\b((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,%{#{@url_brace.xml_open}<a href="\\1" target="_top">\\1</a>#{@url_brace.xml_close}\\2}) #http ftp matches with decoration
+      para.gsub!(/(^|\s)((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?=\s|$))/,%{\\1#{@url_brace.xml_open}<a href="\\2" target="_top">\\2</a>#{@url_brace.xml_close}\\3}) #http ftp matches with decoration
       para
     end
     def paragraph
diff --git a/lib/sisu/v0/shared_xml.rb b/lib/sisu/v0/shared_xml.rb
index 249085a1..914391b7 100644
--- a/lib/sisu/v0/shared_xml.rb
+++ b/lib/sisu/v0/shared_xml.rb
@@ -358,8 +358,10 @@ module SiSU_XML_munge
       para.gsub!(/(^|\s)\{\s*(\S+?\.(?:jpg|png|gif))(\s+[^}]+)?\}(https?:\/\/\S+)/,%{\\1<image xmlns:xlink="http://www.w3.org/1999/xlink" xlink:type="simple" xlink:actuate="onLoad" xlink:show="embed" xlink:href="#{@dir.url.images_local}/\\2"/>\\2})
       para.gsub!(/\B\{([^}]+)\}(https?:\/\/[^"><]+?)([,.:;"><]?(?:\s|$))/,
         '<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:type="simple" xlink:href="\2">\1</link>\3') #watch, compare html_tune
-      para.gsub!(/\b((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,
-        %{#{@url_brace.xml_open}<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:type="simple" xlink:href="\\1">\\1</link>#{@url_brace.xml_close}\\2})
+      para.gsub!(/(^|\s)((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?=\s|$))/,
+        %{\\1#{@url_brace.xml_open}<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:type="simple" xlink:href="\\2">\\2</link>#{@url_brace.xml_close}\\3})
+      #para.gsub!(/\b((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/, #also works
+        #%{#{@url_brace.xml_open}<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:type="simple" xlink:href="\\1">\\1</link>#{@url_brace.xml_close}\\2})
       para.gsub!(/\b[_\\]((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,'<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:type="simple" xlink:href="\1">\1</link>\2') #escaped urls not linked, deal with later
       #para.gsub!(/(^|\s)[_\\]((?:https?|ftp):\/\/\S+?\.[^'"><\s]+?)([.,]?(?:\s|$))/,'\1<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:type="simple" xlink:href="\2">\2</link>\3') #escaped urls not linked, deal with later
       para.gsub!(/&nbsp;/,' ') #clean