From 84957a4260170c3b81a690853ba5b865f59e1217 Mon Sep 17 00:00:00 2001
From: Ralph Amissah <ralph@amissah.com>
Date: Sun, 19 Oct 2014 21:05:55 -0400
Subject: v5 v6: html, remove trailing backslash for empty linebreak &
 paragraph, <br> <p>

---
 lib/sisu/v5/db_sqltxt.rb | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

(limited to 'lib/sisu/v5/db_sqltxt.rb')

diff --git a/lib/sisu/v5/db_sqltxt.rb b/lib/sisu/v5/db_sqltxt.rb
index ffb5966b..6585fd66 100644
--- a/lib/sisu/v5/db_sqltxt.rb
+++ b/lib/sisu/v5/db_sqltxt.rb
@@ -62,11 +62,11 @@ module SiSU_DbText
     def special_character_escape(str)
       str=str.gsub(/'/,"''"). #string.gsub!(/'/,"\047") #string.gsub!(/'/,"\\'")
         gsub(/(\\)/m,'\1\1'). #ok but with warnings, double backslash on sqlite #str.gsub!(/[\\]/m,'\\x5C') #ok but with warnings, but not for sqlite #str.gsub!(/(\\)/m,'\1') #ok for sqlite not for pgsql
-        gsub(/#{Mx[:br_line]}|#{Mx[:br_nl]}/,"<br />\n").
-        gsub(/#{Mx[:tag_o]}\S+?#{Mx[:tag_c]}/,''). #check
-        gsub(/#{Mx[:lnk_o]}\s*(\S+?\.(?:png|jpg))(?:\s+\d+x\d+)?(.+?)#{Mx[:lnk_c]}\S+/,'[image: \1] \2').
-        gsub(/#{Mx[:lnk_o]}\s*(.+?)\s*#{Mx[:lnk_c]}(?:file|ftp):\/\/\S+?([.,!?]?(?:\s|$))/,'\1\2').
-        gsub(/#{Mx[:lnk_o]}\s*(.+?)\s*#{Mx[:lnk_c]}#{Mx[:url_o]}\S+?#{Mx[:url_c]}/,'\1')
+        gsub(/#{Mx[:br_line]}|#{Mx[:br_nl]}/m,"<br>\n").
+        gsub(/#{Mx[:tag_o]}\S+?#{Mx[:tag_c]}/m,''). #check
+        gsub(/#{Mx[:lnk_o]}\s*(\S+?\.(?:png|jpg))(?:\s+\d+x\d+)?(.+?)#{Mx[:lnk_c]}\S+/m,'[image: \1] \2').
+        gsub(/#{Mx[:lnk_o]}\s*(.+?)\s*#{Mx[:lnk_c]}(?:file|ftp):\/\/\S+?([.,!?]?(?:\s|$))/m,'\1\2').
+        gsub(/#{Mx[:lnk_o]}\s*(.+?)\s*#{Mx[:lnk_c]}#{Mx[:url_o]}\S+?#{Mx[:url_c]}/m,'\1')
     end
     def clean_searchable_text_from_document_objects(arr)
       txt_arr,en=[],[]
-- 
cgit v1.2.3


From 2c73f3060f9678f751c236fe17863d443f6a650f Mon Sep 17 00:00:00 2001
From: Ralph Amissah <ralph@amissah.com>
Date: Sun, 19 Oct 2014 21:13:52 -0400
Subject: v5 v6: db, text search & display field, footnotes moved to end of
 text object

* cleaner, more useful search results
* cleaner text search field
* separate footnote fields redundant for search purposes
---
 lib/sisu/v5/db_sqltxt.rb | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

(limited to 'lib/sisu/v5/db_sqltxt.rb')

diff --git a/lib/sisu/v5/db_sqltxt.rb b/lib/sisu/v5/db_sqltxt.rb
index 6585fd66..3f6cf951 100644
--- a/lib/sisu/v5/db_sqltxt.rb
+++ b/lib/sisu/v5/db_sqltxt.rb
@@ -60,7 +60,7 @@
 module SiSU_DbText
   class Prepare
     def special_character_escape(str)
-      str=str.gsub(/'/,"''"). #string.gsub!(/'/,"\047") #string.gsub!(/'/,"\\'")
+      str=str.gsub(/'/m,"''"). #string.gsub!(/'/,"\047") #string.gsub!(/'/,"\\'")
         gsub(/(\\)/m,'\1\1'). #ok but with warnings, double backslash on sqlite #str.gsub!(/[\\]/m,'\\x5C') #ok but with warnings, but not for sqlite #str.gsub!(/(\\)/m,'\1') #ok for sqlite not for pgsql
         gsub(/#{Mx[:br_line]}|#{Mx[:br_nl]}/m,"<br>\n").
         gsub(/#{Mx[:tag_o]}\S+?#{Mx[:tag_c]}/m,''). #check
@@ -80,13 +80,29 @@ module SiSU_DbText
           gsub(/#{Mx[:en_b_o]}.+?#{Mx[:en_b_c]}/m,'').
           gsub(/ \s+/m,' ')
         #p s if s =~/[^ \nA-Za-z0-9'"`?!#@$%^&*=+,.;:\[\]()<>{}‹›|\\\/~_-]/
-        s
+        txt_arr << s
       end
-      txt_arr << arr << en
-      #txt_arr=txt_arr.flatten
+      txt_arr=txt_arr << en
       txt=txt_arr.flatten.join("\n")
-      txt=special_character_escape(txt)
-      txt
+      special_character_escape(txt)
+    end
+    def clean_document_objects_body(arr)
+      txt_arr,en,en_arr=[],[],[]
+      arr=(arr.is_a?(String)) ? [ arr ] : arr
+      arr.each do |s|
+        en << s.scan(/#{Mx[:en_a_o]}\s*(.+?)\s*#{Mx[:en_a_c]}/m)
+        s=s.gsub(/#{Mx[:en_a_o]}\s*(\d+).+?#{Mx[:en_a_c]}/m,'<sup>\1</sup>').
+          gsub(/#{Mx[:en_b_o]}.+?#{Mx[:en_b_c]}/m,'').
+          gsub(/ \s+/m,' ')
+        txt_arr << s
+      end
+      en.flatten.each do |e|
+        e=e.sub(/^(\d+)\s*/,'<sup>\1</sup> ')
+        en_arr << e
+      end
+      txt_arr=txt_arr << en_arr
+      txt=txt_arr.flatten.join("\n<br>")
+      special_character_escape(txt)
     end
     def clean_searchable_text_from_document_source(arr)
       txt_arr,en=[],[]
-- 
cgit v1.2.3