rbpdf 1.18.3 → 1.18.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: ca9832fc7040403705b87fe67df785e4450cef00
-  data.tar.gz: 31c6bee90f97b4ddc284f670b74eeb09c36d5a1d
+  metadata.gz: 5329616012db52c1743a298afb11aa835126a284
+  data.tar.gz: 6d92283c3fa55edaa59b68edab0b2db7f1311ae5
 SHA512:
-  metadata.gz: aa5d54a75bba745946e90af5e31444efb9ce8a36e12cf6028d92508987dbd8e1d30bdfffe894677f98712921b90e7520d6009e36bb1a2085f9b632c0ec7c37eb
-  data.tar.gz: c0b469c4c2544ea86297485535cda927c762019ed0b94d1661566b722a5febcd91d48bb5cea827c75337d0b6c8244415bdbc6a2baea7c6e2e686af4b4d80dd9f
+  metadata.gz: 74738ca0557dbe52a46177277d58b92ce1dc4e590fea0cfb6c4711e94f657db4359fe455aab918aeb67d55353fd4044d6bea4de05eaa7c1189b22e7daa05d918
+  data.tar.gz: f29d02d42ac883a7781314deed127c67216fc780b9df7d742cefeb8b839fc7e896fc2638a9459f6eb1010eb3a4b5e478450d523e6a55a9b55f045914efda2000

data/CHANGELOG

@@ -1,3 +1,12 @@
+1.18.4 2014-12-21
+       - utf8Bidi() Persion 0x200C(8204) and endedletter bug fixed.
+       - utf8Bidi() Bidirectional Algorithm X9/I1 rule BN(ZERO WIDTH NON-JOINER) bug fixed.
+       - utf8Bidi() Bidirectional Algorithm N1/N2 rule B/S/WS/ON bug fixed.
+       - utf8Bidi() Bidirectional Algorithm W2/W7 rule bug fixed.
+       - utf8Bidi() Bidirectional Algorithm N1/N2 rule multiple NIs bug fixed.
+       - RTL direction problem fixed. isRTLTextDir direction bug fixed.
+       - utf8Bidi() speed was improved.
+
 1.18.3 2014-12-06
        - fixed img and hr self html tag DOM problem.
        - fixed HTML table problem in writeHTMLCell function.

data/lib/rbpdf.rb

@@ -989,15 +989,33 @@ class RBPDF
 
   #
   # Return the current temporary RTL status
-  # [@return boolean]
+  # [@return boolean] true: RTL, false: LTR
   # [@access public]
   # [@since 4.8.014 (2009-11-04)]
   #
   def isRTLTextDir()
-    return (@rtl or (@tmprtl == 'R'))
+    if @tmprtl != false
+      return @tmprtl == 'R'
+    else
+      return @rtl
+    end
   end
   alias_method :is_rtl_text_dir, :isRTLTextDir
 
+  #
+  # Return the current text RTL status
+  # [@return direction] 'R' : RTL, 'L' LTR
+  # [@access protected]
+  #
+  def rtl_text_dir()
+    if @tmprtl != false
+      return @tmprtl # 'R' or 'L'
+    else
+      return @rtl ? 'R' : 'L'
+    end
+  end
+  protected :rtl_text_dir
+
   #
   # Set the last cell height.
   # [@param float :h] cell height.
@@ -2432,7 +2450,7 @@ class RBPDF
   # [@since 1.2]
   #
   def GetStringWidth(s, fontname='', fontstyle='', fontsize=0, getarray=false)
-    return GetArrStringWidth(utf8Bidi(UTF8StringToArray(s), s, @tmprtl), fontname, fontstyle, fontsize, getarray)
+    return GetArrStringWidth(utf8Bidi(UTF8StringToArray(s), s, rtl_text_dir), fontname, fontstyle, fontsize, getarray)
   end
   alias_method :get_string_width, :GetStringWidth
 
@@ -3509,7 +3527,7 @@ class RBPDF
           txt2 = UTF8ToLatin1(txt2)
         else
           unicode = UTF8StringToArray(txt) # array of UTF-8 unicode values
-          unicode = utf8Bidi(unicode, '', @tmprtl)
+          unicode = utf8Bidi(unicode, '', rtl_text_dir)
           if @@k_thai_topchars and @@k_thai_topchars == true
             # ---- Fix for bug #2977340 "Incorrect Thai characters position arrangement" ----
             # NOTE: this doesn't work with HTML justification
@@ -4078,7 +4096,7 @@ class RBPDF
     lines = 1
     sum = 0
     chars = UTF8StringToArray(txt)
-    chars = utf8Bidi(chars, txt, @tmprtl)
+    chars = utf8Bidi(chars, txt, rtl_text_dir)
     charsWidth = GetArrStringWidth(chars, '', '', 0, true)
     if @rtl
       charsWidth.reverse!
@@ -4228,7 +4246,7 @@ class RBPDF
     end
 
     # check if string contains RTL text
-    if arabic or (@tmprtl == 'R') or (txt =~ @@k_re_pattern_rtl)
+    if arabic or isRTLTextDir or (txt =~ @@k_re_pattern_rtl)
       rtlmode = true
     else
       rtlmode = false
@@ -4303,7 +4321,7 @@ class RBPDF
           startx = @x
           tmparr = chars[j, i - j]
           if rtlmode
-            tmparr = utf8Bidi(tmparr, tmpstr, @tmprtl)
+            tmparr = utf8Bidi(tmparr, tmpstr, rtl_text_dir)
           end
           linew = GetArrStringWidth(tmparr)
           tmparr = ''
@@ -4362,7 +4380,7 @@ class RBPDF
         if ((@current_font['type'] == 'TrueTypeUnicode') or (@current_font['type'] == 'cidfont0')) and arabic
           # with bidirectional algorithm some chars may be changed affecting the line length
           # *** very slow ***
-          l = GetArrStringWidth(utf8Bidi(chars[j,i-j], '', @tmprtl))
+          l = GetArrStringWidth(utf8Bidi(chars[j,i-j], '', rtl_text_dir))
         else
           l += GetCharWidth(c)
         end
@@ -4385,7 +4403,7 @@ class RBPDF
                 startx = @x
                 tmparr = chars[j, i - j]
                 if rtlmode
-                  tmparr = utf8Bidi(tmparr, tmpstr, @tmprtl)
+                  tmparr = utf8Bidi(tmparr, tmpstr, rtl_text_dir)
                 end
                 linew = GetArrStringWidth(tmparr)
                 tmparr = ''
@@ -4439,7 +4457,7 @@ class RBPDF
               startx = @x
               tmparr = chars[j, sep + endspace - j]
               if rtlmode
-                tmparr = utf8Bidi(tmparr, tmpstr, @tmprtl)
+                tmparr = utf8Bidi(tmparr, tmpstr, rtl_text_dir)
               end
               linew = GetArrStringWidth(tmparr)
               tmparr = ''
@@ -4514,7 +4532,7 @@ class RBPDF
         startx = @x
         tmparr = chars[j, nb - j]
         if rtlmode
-          tmparr = utf8Bidi(tmparr, tmpstr, @tmprtl)
+          tmparr = utf8Bidi(tmparr, tmpstr, rtl_text_dir)
         end
         linew = GetArrStringWidth(tmparr)
         tmparr = ''
@@ -5592,8 +5610,8 @@ protected
       if @is_unicode
         alias_b = escape(UTF8ToLatin1(@alias_nb_pages))
         alias_bu = escape(UTF8ToLatin1('{' + @alias_nb_pages + '}'))
-        alias_c = escape(utf8StrRev(@alias_nb_pages, false, @tmprtl))
-        alias_cu = escape(utf8StrRev('{' + @alias_nb_pages + '}', false, @tmprtl))
+        alias_c = escape(utf8StrRev(@alias_nb_pages, false, rtl_text_dir))
+        alias_cu = escape(utf8StrRev('{' + @alias_nb_pages + '}', false, rtl_text_dir))
       end
     end
     if @alias_num_page
@@ -5602,8 +5620,8 @@ protected
       if @is_unicode
         alias_pb = escape(UTF8ToLatin1(@alias_num_page))
         alias_pbu = escape(UTF8ToLatin1('{' + @alias_num_page + '}'))
-        alias_pc = escape(utf8StrRev(@alias_num_page, false, @tmprtl))
-        alias_pcu = escape(utf8StrRev('{' + @alias_num_page + '}', false, @tmprtl))
+        alias_pc = escape(utf8StrRev(@alias_num_page, false, rtl_text_dir))
+        alias_pcu = escape(utf8StrRev('{' + @alias_num_page + '}', false, rtl_text_dir))
       end
     end
     pagegroupnum = 0
@@ -5624,8 +5642,8 @@ protected
           if @is_unicode
             alias_gb = escape(UTF8ToLatin1(k))
             alias_gbu = escape(UTF8ToLatin1('{' + k + '}'))
-            alias_gc = escape(utf8StrRev(k.dup, false, @tmprtl))
-            alias_gcu = escape(utf8StrRev('{' + k + '}', false, @tmprtl))
+            alias_gc = escape(utf8StrRev(k.dup, false, rtl_text_dir))
+            alias_gcu = escape(utf8StrRev('{' + k + '}', false, rtl_text_dir))
           end
           temppage = temppage.gsub(alias_gau, vu)
           if @is_unicode
@@ -5644,8 +5662,8 @@ protected
           if @is_unicode
             alias_pgb = escape(UTF8ToLatin1(pk))
             alias_pgbu = escape(UTF8ToLatin1('{' + pk + '}'))
-            alias_pgc = escape(utf8StrRev(pk, false, @tmprtl))
-            alias_pgcu = escape(utf8StrRev('{' + pk + '}', false, @tmprtl))
+            alias_pgc = escape(utf8StrRev(pk, false, rtl_text_dir))
+            alias_pgcu = escape(utf8StrRev('{' + pk + '}', false, rtl_text_dir))
           end
           temppage = temppage.gsub(alias_pgau, pvu)
           if @is_unicode
@@ -7497,7 +7515,7 @@ protected
         s = UTF8ToLatin1(s)
       else
         # Convert string to UTF-16BE and reverse RTL language
-        s = utf8StrRev(s, false, @tmprtl)
+        s = utf8StrRev(s, false, rtl_text_dir)
       end
     end
     return escape(s);
@@ -9048,7 +9066,7 @@ public
     else
       # P2. In each paragraph, find the first character of type L, AL, or R.
       # P3. If a character is found in P2 and it is of type AL or R, then set the paragraph embedding level to one; otherwise, set it to zero.
-      0.upto(numchars -1) do |i|
+      numchars.times do |i|
         type = @@unicode[ta[i]]
         if type == 'L'
           pel = 0
@@ -9074,63 +9092,78 @@ public
       
     # X1. Begin by setting the current embedding level to the paragraph embedding level. Set the directional override status to neutral. Process each character iteratively, applying rules X2 through X9. Only embedding levels from 0 to 61 are valid in this phase.
     #   In the resolution of levels in rules I1 and I2, the maximum embedding level of 62 can be reached.
-    0.upto(numchars-1) do |i|
-      if ta[i] == @@k_rle
+    reg_KRP = /^(@@k_rle|@@k_lre|@@k_rlo|@@k_lro|@@k_pdf)$/
+    reg_KR = /^(@@k_rle|@@k_lre|@@k_rlo|@@k_lro)$/
+    numchars.times do |i|
+      if ta[i] !~ reg_KRP
+        # X6. For all types besides RLE, LRE, RLO, LRO, and PDF:
+        #  a. Set the level of the current character to the current embedding level.
+        #  b. Whenever the directional override status is not neutral, reset the current character type to the directional override status.
+        if dos != 'N'
+          chardir = dos
+        else
+          chardir = @@unicode[ta[i]]
+          chardir = 'L' if chardir.nil?
+        end
+        # stores string characters and other information
+        chardata << {:char => ta[i], :level => cel, :type => chardir, :sor => sor, :eor => eor}
+        next
+      end
+
+      case ta[i]
+      when @@k_rle
         # X2. With each RLE, compute the least greater odd embedding level.
         #  a. If this new level would be valid, then this embedding code is valid. Remember (push) the current embedding level and override status. Reset the current level to this new level, and reset the override status to neutral.
         #  b. If the new level would not be valid, then this code is invalid. Do not change the current level or override status.
         next_level = cel + (cel % 2) + 1
         if next_level < 62
-          remember.push :num => @@k_rle, :cel => cel, :dos => dos
+          remember << {:num => @@k_rle, :cel => cel, :dos => dos}
           cel = next_level
           dos = 'N'
           sor = eor
           eor = (cel % 2 == 1) ? 'R' : 'L'
         end
-      elsif ta[i] == @@k_lre
+      when @@k_lre
         # X3. With each LRE, compute the least greater even embedding level.
         #  a. If this new level would be valid, then this embedding code is valid. Remember (push) the current embedding level and override status. Reset the current level to this new level, and reset the override status to neutral.
         #  b. If the new level would not be valid, then this code is invalid. Do not change the current level or override status.
         next_level = cel + 2 - (cel % 2)
         if next_level < 62
-          remember.push :num => @@k_lre, :cel => cel, :dos => dos
+          remember << {:num => @@k_lre, :cel => cel, :dos => dos}
           cel = next_level
           dos = 'N'
           sor = eor
           eor = (cel % 2 == 1) ? 'R' : 'L'
         end
-      elsif ta[i] == @@k_rlo
+      when @@k_rlo
         # X4. With each RLO, compute the least greater odd embedding level.
         #  a. If this new level would be valid, then this embedding code is valid. Remember (push) the current embedding level and override status. Reset the current level to this new level, and reset the override status to right-to-left.
         #  b. If the new level would not be valid, then this code is invalid. Do not change the current level or override status.
         next_level = cel + (cel % 2) + 1
         if next_level < 62
-          remember.push :num => @@k_rlo, :cel => cel, :dos => dos
+          remember << {:num => @@k_rlo, :cel => cel, :dos => dos}
           cel = next_level
           dos = 'R'
           sor = eor
           eor = (cel % 2 == 1) ? 'R' : 'L'
         end
-      elsif ta[i] == @@k_lro
+      when @@k_lro
         # X5. With each LRO, compute the least greater even embedding level.
         #  a. If this new level would be valid, then this embedding code is valid. Remember (push) the current embedding level and override status. Reset the current level to this new level, and reset the override status to left-to-right.
         #  b. If the new level would not be valid, then this code is invalid. Do not change the current level or override status.
         next_level = cel + 2 - (cel % 2)
         if next_level < 62
-          remember.push :num => @@k_lro, :cel => cel, :dos => dos
+          remember << {:num => @@k_lro, :cel => cel, :dos => dos}
           cel = next_level
           dos = 'L'
           sor = eor
           eor = (cel % 2 == 1) ? 'R' : 'L'
         end
-      elsif ta[i] == @@k_pdf
+      when @@k_pdf
         # X7. With each PDF, determine the matching embedding or override code. If there was a valid matching code, restore (pop) the last remembered (pushed) embedding level and directional override.
         if remember.length
           last = remember.length - 1
-          if (remember[last][:num] == @@k_rle) or 
-              (remember[last][:num] == @@k_lre) or 
-              (remember[last][:num] == @@k_rlo) or 
-              (remember[last][:num] == @@k_lro)
+          if remember[last][:num] =~ reg_KR
             match = remember.pop
             cel = match[:cel]
             dos = match[:dos]
@@ -9138,25 +9171,6 @@ public
             eor = ((cel > match[:cel] ? cel : match[:cel]) % 2 == 1) ? 'R' : 'L'
           end
         end
-      elsif (ta[i] != @@k_rle) and
-           (ta[i] != @@k_lre) and
-           (ta[i] != @@k_rlo) and
-           (ta[i] != @@k_lro) and
-           (ta[i] != @@k_pdf)
-        # X6. For all types besides RLE, LRE, RLO, LRO, and PDF:
-        #  a. Set the level of the current character to the current embedding level.
-        #  b. Whenever the directional override status is not neutral, reset the current character type to the directional override status.
-        if dos != 'N'
-          chardir = dos
-        else
-          if !@@unicode[ta[i]].nil?
-            chardir = @@unicode[ta[i]]
-          else
-            chardir = 'L'
-          end
-        end
-        # stores string characters and other information
-        chardata.push :char => ta[i], :level => cel, :type => chardir, :sor => sor, :eor => eor
       end
     end # end for each char
       
@@ -9172,7 +9186,7 @@ public
     # W1. Examine each nonspacing mark (NSM) in the level run, and change the type of the NSM to the type of the previous character. If the NSM is at the start of the level run, it will get the type of sor.
     prevlevel = -1 # track level changes
     levcount = 0 # counts consecutive chars at the same level
-    0.upto(numchars-1) do |i|
+    numchars.times do |i|
       if chardata[i][:type] == 'NSM'
         if levcount
           chardata[i][:type] = chardata[i][:sor]
@@ -9191,8 +9205,8 @@ public
     # W2. Search backward from each instance of a European number until the first strong type (R, L, AL, or sor) is found. If an AL is found, change the type of the European number to Arabic number.
     prevlevel = -1
     levcount = 0
-    0.upto(numchars-1) do |i|
-      if chardata[i][:char] == 'EN'
+    numchars.times do |i|
+      if chardata[i][:type] == 'EN'
         levcount.downto(0) do |j|
           if chardata[j][:type] == 'AL'
             chardata[i][:type] = 'AN'
@@ -9210,7 +9224,7 @@ public
     end
     
     # W3. Change all ALs to R.
-    0.upto(numchars-1) do |i|
+    numchars.times do |i|
       if chardata[i][:type] == 'AL'
         chardata[i][:type] = 'R'
       end
@@ -9219,7 +9233,7 @@ public
     # W4. A single European separator between two European numbers changes to a European number. A single common separator between two numbers of the same type changes to that type.
     prevlevel = -1
     levcount = 0
-    0.upto(numchars-1) do |i|
+    numchars.times do |i|
       if (levcount > 0) and (i+1 < numchars) and (chardata[i+1][:level] == prevlevel)
         if (chardata[i][:type] == 'ES') and (chardata[i-1][:type] == 'EN') and (chardata[i+1][:type] == 'EN')
           chardata[i][:type] = 'EN'
@@ -9240,7 +9254,7 @@ public
     # W5. A sequence of European terminators adjacent to European numbers changes to all European numbers.
     prevlevel = -1
     levcount = 0
-    0.upto(numchars-1) do |i|
+    numchars.times do |i|
       if chardata[i][:type] == 'ET'
         if (levcount > 0) and (chardata[i-1][:type] == 'EN')
           chardata[i][:type] = 'EN'
@@ -9268,8 +9282,9 @@ public
     # W6. Otherwise, separators and terminators change to Other Neutral.
     prevlevel = -1
     levcount = 0
-    0.upto(numchars-1) do |i|
-      if (chardata[i][:type] == 'ET') or (chardata[i][:type] == 'ES') or (chardata[i][:type] == 'CS')
+    reg_ET_ES_CS = /^(ET|ES|CS)$/
+    numchars.times do |i|
+      if chardata[i][:type] =~ reg_ET_ES_CS
         chardata[i][:type] = 'ON'
       end
       if chardata[i][:level] != prevlevel
@@ -9283,8 +9298,8 @@ public
     # W7. Search backward from each instance of a European number until the first strong type (R, L, or sor) is found. If an L is found, then change the type of the European number to L.
     prevlevel = -1
     levcount = 0
-    0.upto(numchars-1) do |i|
-      if chardata[i][:char] == 'EN'
+    numchars.times do |i|
+      if chardata[i][:type] == 'EN'
         levcount.downto(0) do |j|
           if chardata[j][:type] == 'L'
             chardata[i][:type] = 'L'
@@ -9304,45 +9319,62 @@ public
     # N1. A sequence of neutrals takes the direction of the surrounding strong text if the text on both sides has the same direction. European and Arabic numbers act as if they were R in terms of their influence on neutrals. Start-of-level-run (sor) and end-of-level-run (eor) are used at level run boundaries.
     prevlevel = -1
     levcount = 0
-    0.upto(numchars-1) do |i|
-      if (levcount > 0) and (i+1 < numchars) and (chardata[i+1][:level] == prevlevel)
-        if (chardata[i][:type] == 'N') and (chardata[i-1][:type] == 'L') and (chardata[i+1][:type] == 'L')
-          chardata[i][:type] = 'L'
-        elsif (chardata[i][:type] == 'N') and
-         ((chardata[i-1][:type] == 'R') or (chardata[i-1][:type] == 'EN') or (chardata[i-1][:type] == 'AN')) and
-         ((chardata[i+1][:type] == 'R') or (chardata[i+1][:type] == 'EN') or (chardata[i+1][:type] == 'AN'))
-          chardata[i][:type] = 'R'
-        elsif chardata[i][:type] == 'N'
-          # N2. Any remaining neutrals take the embedding direction
-          chardata[i][:type] = chardata[i][:sor]
-        end
-      elsif (levcount == 0) and (i+1 < numchars) and (chardata[i+1][:level] == prevlevel)
-        # first char
-        if (chardata[i][:type] == 'N') and (chardata[i][:sor] == 'L') and (chardata[i+1][:type] == 'L')
-          chardata[i][:type] = 'L'
-        elsif (chardata[i][:type] == 'N') and
-         ((chardata[i][:sor] == 'R') or (chardata[i][:sor] == 'EN') or (chardata[i][:sor] == 'AN')) and
-         ((chardata[i+1][:type] == 'R') or (chardata[i+1][:type] == 'EN') or (chardata[i+1][:type] == 'AN'))
-          chardata[i][:type] = 'R'
-        elsif chardata[i][:type] == 'N'
-          # N2. Any remaining neutrals take the embedding direction
-          chardata[i][:type] = chardata[i][:sor]
-        end
-      elsif (levcount > 0) and ((i+1 == numchars) or ((i+1 < numchars) and (chardata[i+1][:level] != prevlevel)))
-        # last char
-        if (chardata[i][:type] == 'N') and (chardata[i-1][:type] == 'L') and (chardata[i][:eor] == 'L')
-          chardata[i][:type] = 'L'
-        elsif (chardata[i][:type] == 'N') and
-         ((chardata[i-1][:type] == 'R') or (chardata[i-1][:type] == 'EN') or (chardata[i-1][:type] == 'AN')) and
-         ((chardata[i][:eor] == 'R') or (chardata[i][:eor] == 'EN') or (chardata[i][:eor] == 'AN'))
-          chardata[i][:type] = 'R'
-        elsif chardata[i][:type] == 'N'
+    reg_NI = /^(B|S|WS|ON)$/
+    reg_R_EN_AN = /^(R|EN|AN)$/
+    reg_EN_AN = /^(EN|AN)$/
+    ni = nil
+    numchars.times do |i|
+      if (chardata[i][:type] =~ reg_NI)
+        if (levcount > 0) and (i+1 < numchars) and (chardata[i+1][:level] == prevlevel)
+          if !ni.nil? and ni > i
+            next_non_space_char = chardata[ni][:type]
+          else
+            ni = chardata[i+1..-1].index {|item| item[:type] !~ reg_NI}
+            unless ni.nil?
+              ni += i+1
+              next_non_space_char = chardata[ni][:type]
+            end
+          end
+          if (chardata[i-1][:type] == 'L') and (next_non_space_char == 'L')
+            chardata[i][:type] = 'L'
+          elsif ((chardata[i-1][:type] == 'R') and (next_non_space_char =~ reg_R_EN_AN) or
+                 (chardata[i-1][:type] =~ reg_EN_AN) and (next_non_space_char == 'R'))
+            chardata[i][:type] = 'R'
+          else
+            # N2. Any remaining neutrals take the embedding direction
+            chardata[i][:type] = chardata[i][:sor]
+          end
+        elsif (levcount == 0) and (i+1 < numchars) and (chardata[i+1][:level] == prevlevel)
+          ni = chardata[i+1..-1].index {|item| item[:type] !~ reg_NI}
+          unless ni.nil?
+            ni += i+1
+            next_non_space_char = chardata[ni][:type]
+          end
+          # first char
+          if (chardata[i][:sor] == 'L') and (next_non_space_char == 'L')
+            chardata[i][:type] = 'L'
+          elsif ((chardata[i][:sor] == 'R') and (next_non_space_char =~ reg_R_EN_AN) or
+                 (chardata[i][:sor] =~ reg_EN_AN) and (next_non_space_char == 'R'))
+            chardata[i][:type] = 'R'
+          else
+            # N2. Any remaining neutrals take the embedding direction
+            chardata[i][:type] = chardata[i][:sor]
+          end
+        elsif (levcount > 0) and ((i+1 == numchars) or ((i+1 < numchars) and (chardata[i+1][:level] != prevlevel)))
+          # last char
+          if (chardata[i-1][:type] == 'L') and (chardata[i][:eor] == 'L')
+            chardata[i][:type] = 'L'
+          elsif ((chardata[i-1][:type] == 'R') and (chardata[i][:eor] =~ reg_R_EN_AN) or
+                 (chardata[i-1][:type] =~ reg_EN_AN) and (chardata[i][:eor] == 'R'))
+            chardata[i][:type] = 'R'
+          else
+            # N2. Any remaining neutrals take the embedding direction
+            chardata[i][:type] = chardata[i][:sor]
+          end
+        else
           # N2. Any remaining neutrals take the embedding direction
           chardata[i][:type] = chardata[i][:sor]
         end
-      elsif chardata[i][:type] == 'N'
-        # N2. Any remaining neutrals take the embedding direction
-        chardata[i][:type] = chardata[i][:sor]
       end
       if chardata[i][:level] != prevlevel
         levcount = 0
@@ -9354,19 +9386,26 @@ public
     
     # I1. For all characters with an even (left-to-right) embedding direction, those of type R go up one level and those of type AN or EN go up two levels.
     # I2. For all characters with an odd (right-to-left) embedding direction, those of type L, EN or AN go up one level.
-    0.upto(numchars-1) do |i|
+    prevlevel = -1
+    reg_L_AN_EN = /^(L|AN|EN)$/
+    reg_AN_EN = /^(AN|EN)$/
+
+    numchars.times do |i|
       odd = chardata[i][:level] % 2
-      if odd == 1
-        if (chardata[i][:type] == 'L') or (chardata[i][:type] == 'AN') or (chardata[i][:type] == 'EN')
+      if odd == 1 # I2.
+        if chardata[i][:type] =~ reg_L_AN_EN
           chardata[i][:level] += 1
         end
-      else
+      else # I1.
         if chardata[i][:type] == 'R'
           chardata[i][:level] += 1
-        elsif (chardata[i][:type] == 'AN') or (chardata[i][:type] == 'EN')
+        elsif chardata[i][:type] == 'BN' and prevlevel != -1
+          chardata[i][:level] = prevlevel
+        elsif chardata[i][:type] =~ reg_AN_EN
           chardata[i][:level] += 2
         end
       end
+      prevlevel = chardata[i][:level]
       maxlevel = [chardata[i][:level],maxlevel].max
     end
     
@@ -9375,13 +9414,14 @@ public
     #  2. Paragraph separators,
     #  3. Any sequence of whitespace characters preceding a segment separator or paragraph separator, and
     #  4. Any sequence of white space characters at the end of the line.
-    0.upto(numchars-1) do |i|
-      if (chardata[i][:type] == 'B') or (chardata[i][:type] == 'S')
+    reg_B_S = /^(B|S)$/
+    numchars.times do |i|
+      if chardata[i][:type] =~ reg_B_S
         chardata[i][:level] = pel
       elsif chardata[i][:type] == 'WS'
         j = i+1
         while j < numchars
-          if ((chardata[j][:type] == 'B') or (chardata[j][:type] == 'S')) or
+          if (chardata[j][:type] =~ reg_B_S) or
             ((j == numchars-1) and (chardata[j][:type] == 'WS'))
             chardata[i][:level] = pel
             break
@@ -9402,16 +9442,18 @@ public
       laaletter = false
       charAL = []
       x = 0
-      0.upto(numchars-1) do |i|
-        if (@@unicode[chardata[i][:char]] == 'AL') or (chardata[i][:char] == 32) or (chardata[i]['char'] == 8204) # 4.0.008 - Arabic shaping for "Zero-Width Non-Joiner" character (U+200C) was fixed.
-          charAL[x] = chardata[i]
+      numchars.times do |i|
+        c = chardata[i][:char]
+        if (@@unicode[c] == 'AL') or (c == 32) or (c == 8204) # Unicode Character 'ZERO WIDTH NON-JOINER' (U+200C)
+          charAL[x] = chardata[i].dup
           charAL[x][:i] = i
           chardata[i][:x] = x
           x += 1
         end
       end
       numAL = x
-      0.upto(numchars-1) do |i|
+      reg_AL_NSM = /^(AL|NSM)$/
+      numchars.times do |i|
         thischar = chardata[i]
         if i > 0
           prevchar = chardata[i-1]
@@ -9449,8 +9491,8 @@ public
             laaletter = false
           end
           if (prevchar != false) and (nextchar != false) and
-            ((@@unicode[prevchar[:char]] == 'AL') or (@@unicode[prevchar[:char]] == 'NSM')) and
-            ((@@unicode[nextchar[:char]] == 'AL') or (@@unicode[nextchar[:char]] == 'NSM')) and
+            (@@unicode[prevchar[:char]] =~ reg_AL_NSM) and
+            (@@unicode[nextchar[:char]] =~ reg_AL_NSM) and
             (nextchar[:type] == thischar[:type]) and
             (nextchar[:char] != 1567)
             # medial
@@ -9466,7 +9508,7 @@ public
               end
             end
           elsif (nextchar != false) and
-            ((@@unicode[nextchar[:char]] == 'AL') or (@@unicode[nextchar[:char]] == 'NSM')) and
+            (@@unicode[nextchar[:char]] =~ reg_AL_NSM) and
             (nextchar[:type] == thischar[:type]) and
             (nextchar[:char] != 1567)
             if !arabicarr[thischar[:char]].nil? and !arabicarr[thischar[:char]][2].nil?
@@ -9474,7 +9516,7 @@ public
               chardata2[i][:char] = arabicarr[thischar[:char]][2]
             end
           elsif ((prevchar != false) and
-            ((@@unicode[prevchar[:char]] == 'AL') or (@@unicode[prevchar[:char]] == 'NSM')) and
+            (@@unicode[prevchar[:char]] =~ reg_AL_NSM) and
             (prevchar[:type] == thischar[:type])) or
             ((nextchar != false) and (nextchar[:char] == 1567))
             # final
@@ -9516,7 +9558,7 @@ public
       # Putting the combining mark and shadda in the same glyph allows us to avoid the two marks overlapping each other in an illegible manner.
       #
       cw = @current_font['cw']
-                        0.upto(numchars-2) do |i|
+      (numchars-1).times do |i|
         if (chardata2[i][:char] == 1617) and !@@diacritics[chardata2[i+1][:char]].nil?
           # check if the subtitution font is defined on current font
           unless cw[@@diacritics[chardata2[i+1][:char]]].nil?
@@ -9544,14 +9586,15 @@ public
       ordarray = []
       revarr = []
       onlevel = false
-      0.upto(numchars-1) do |i|
+      numchars.times do |i|
         if chardata[i][:level] >= j
           onlevel = true
-          unless @@unicode_mirror[chardata[i][:char]].nil?
+          um = @@unicode_mirror[chardata[i][:char]]
+          if um
             # L4. A character is depicted by a mirrored glyph if and only if (a) the resolved directionality of that character is R, and (b) the Bidi_Mirrored property value of that character is true.
-            chardata[i][:char] = @@unicode_mirror[chardata[i][:char]]
+            chardata[i][:char] = um
           end
-          revarr.push chardata[i]
+          revarr << chardata[i]
         else
           if onlevel
             revarr.reverse!
@@ -9559,7 +9602,7 @@ public
             revarr = []
             onlevel = false
           end
-          ordarray.push chardata[i]
+          ordarray << chardata[i]
         end
       end
       if onlevel
@@ -9570,7 +9613,7 @@ public
     end
     
     ordarray = []
-    0.upto(numchars-1) do |i|
+    numchars.times do |i|
       ordarray.push chardata[i][:char]
     end
     
@@ -14124,8 +14167,8 @@ public
           if @is_unicode
             alias_b = escape(UTF8ToLatin1(k))
             alias_bu = escape(UTF8ToLatin1(ku))
-            alias_c = escape(utf8StrRev(k, false, @tmprtl))
-            alias_cu = escape(utf8StrRev(ku, false, @tmprtl))
+            alias_c = escape(utf8StrRev(k, false, rtl_text_dir))
+            alias_cu = escape(utf8StrRev(ku, false, rtl_text_dir))
           end
           if n >= page
             np = n + numpages
@@ -14226,8 +14269,8 @@ public
           if @is_unicode
             alias_b = escape(UTF8ToLatin1(k))
             alias_bu = escape(UTF8ToLatin1(ku))
-            alias_c = escape(utf8StrRev(k, false, @tmprtl))
-            alias_cu = escape(utf8StrRev(ku, false, @tmprtl))
+            alias_c = escape(utf8StrRev(k, false, rtl_text_dir))
+            alias_cu = escape(utf8StrRev(ku, false, rtl_text_dir))
           end
           if n >= page
             np = n + numpages

data/lib/rbpdf/version.rb

@@ -1,3 +1,3 @@
 module Rbpdf
-  VERSION = "1.18.3"
+  VERSION = "1.18.4"
 end

data/test/rbpdf_bidi_test.rb

@@ -12,6 +12,42 @@ class RbpdfTest < ActiveSupport::TestCase
     def cache_utf8_string_to_array(str)
       @cache_utf8_string_to_array[str]
     end
+    def rtl_text_dir
+      super
+    end
+  end
+
+  test "RTL test" do
+    pdf = MYPDF.new
+
+    # LTR
+    rtl = pdf.get_rtl
+    assert_equal rtl, false
+    rtl = pdf.is_rtl_text_dir
+    assert_equal rtl, false
+    rtl = pdf.rtl_text_dir
+    assert_equal rtl, 'L'
+
+    pdf.set_temp_rtl('rtl')
+    rtl = pdf.is_rtl_text_dir
+    assert_equal rtl, true
+    rtl = pdf.rtl_text_dir
+    assert_equal rtl, 'R'
+
+    # RTL
+    pdf.set_rtl(true)
+    rtl = pdf.get_rtl
+    assert_equal rtl, true
+    rtl = pdf.is_rtl_text_dir
+    assert_equal rtl, true
+    rtl = pdf.rtl_text_dir
+    assert_equal rtl, 'R'
+
+    pdf.set_temp_rtl('ltr')
+    rtl = pdf.is_rtl_text_dir
+    assert_equal rtl, false
+    rtl = pdf.rtl_text_dir
+    assert_equal rtl, 'L'
   end
 
   test "Bidi" do
@@ -65,6 +101,99 @@ class RbpdfTest < ActiveSupport::TestCase
     assert_equal [0x61, 0x62, 0x63, 0x5ea, 0x5d9, 0x5e8, 0x5d1, 0x5e2], ary_str
   end
 
+  test "Bidi ascii space test" do
+    pdf = MYPDF.new
+
+    ascii_str   = "abc def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x20, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+
+    ascii_str   = "abc  def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x20, 0x20, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+
+    ascii_str   = "abc  "
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x20, 0x20]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, [0x20, 0x20, 0x61, 0x62, 0x63]
+
+    ascii_str   = "abc_def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x5f, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+  end
+
+  test "Bidi ascii numeric space test" do
+    pdf = MYPDF.new
+
+    ascii_str   = "abc 123 def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x20, 0x31, 0x32, 0x33, 0x20, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+
+    ascii_str   = "abc_123_def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x5f, 0x31, 0x32, 0x33, 0x5f, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+  end
+
+  test "Bidi ascii colon test" do
+    pdf = MYPDF.new
+
+    ascii_str   = "abc:def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x3a, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+
+    ascii_str   = "abc: def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x3a, 0x20, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+
+    ascii_str   = "abc : def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x20, 0x3a, 0x20, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+
+    ascii_str   = "abc  ::  def"
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal ary_ucs4_1, [0x61, 0x62, 0x63, 0x20, 0x20, 0x3a, 0x3a, 0x20, 0x20, 0x64, 0x65, 0x66]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'R')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+  end
+
+
   test "Bidi arabic test" do
     pdf = MYPDF.new
 
@@ -98,7 +227,108 @@ class RbpdfTest < ActiveSupport::TestCase
     assert_equal [0xfea9], ary_ucs4
 
     ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_arabic_str_1))
-    assert_equal [0xfeae, 0xfeee, 0xfea9], ary_ucs4
+    assert_equal [0xfead, 0xfeed, 0xfea9], ary_ucs4
+  end
+
+  test "Bidi Persian Sunday test" do
+    pdf = MYPDF.new
+
+    utf8_persian_str_1  = "\xdb\x8c"
+    utf8_persian_str_2  = "\xdb\x8c\xda\xa9"
+    utf8_persian_str_3  = "\xdb\x8c\xda\xa9\xe2\x80\x8c"
+    utf8_persian_str_4  = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4"
+    utf8_persian_str_5  = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4\xd9\x86"
+    utf8_persian_str_6  = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4\xd9\x86\xd8\xa8"
+    utf8_persian_str_7  = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4\xd9\x86\xd8\xa8\xd9\x87" # Sunday
+
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_1))
+    assert_equal ary_ucs4, [0xfbfc]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_2))
+    assert_equal ary_ucs4, [0xfb8f, 0xfbfe]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_3))
+    assert_equal ary_ucs4, [0x200C, 0xfb8f, 0xfbfe]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_4))
+    assert_equal ary_ucs4, [0xfeb5, 0x200C, 0xfb8f, 0xfbfe]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_5))
+    assert_equal ary_ucs4, [0xfee6, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_6))
+    assert_equal ary_ucs4, [0xfe90, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_7))
+    assert_equal ary_ucs4, [0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe]
+  end
+
+  test "Bidi Persian Sunday forcertl test" do
+    pdf = MYPDF.new
+    utf8_persian_str_sunday = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4\xd9\x86\xd8\xa8\xd9\x87"
+
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_sunday), '', 'R')
+    assert_equal ary_ucs4_1, [0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_sunday), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+  end
+
+  test "Bidi Persian Monday test" do
+    pdf = MYPDF.new
+
+    utf8_persian_str_1  = "\xd8\xaf"
+    utf8_persian_str_2  = "\xd8\xaf\xd9\x88"
+    utf8_persian_str_3  = "\xd8\xaf\xd9\x88\xd8\xb4"
+    utf8_persian_str_4  = "\xd8\xaf\xd9\x88\xd8\xb4\xd9\x86"
+    utf8_persian_str_5  = "\xd8\xaf\xd9\x88\xd8\xb4\xd9\x86\xd8\xa8"
+    utf8_persian_str_6  = "\xd8\xaf\xd9\x88\xd8\xb4\xd9\x86\xd8\xa8\xd9\x87" # Monday
+
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_1))
+    assert_equal ary_ucs4, [0xfea9]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_2))
+    assert_equal ary_ucs4, [0xfeed, 0xfea9]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_3))
+    assert_equal ary_ucs4, [0xfeb5, 0xfeed, 0xfea9]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_4))
+    assert_equal ary_ucs4, [0xfee6, 0xfeb7, 0xfeed, 0xfea9]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_5))
+    assert_equal ary_ucs4, [0xfe90, 0xfee8, 0xfeb7, 0xfeed, 0xfea9]
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_6))
+    assert_equal ary_ucs4, [0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0xfeed, 0xfea9]
+  end
+
+  test "Bidi Persian Monday forcertl test" do
+    pdf = MYPDF.new
+    utf8_persian_str_monday = "\xd8\xaf\xd9\x88\xd8\xb4\xd9\x86\xd8\xa8\xd9\x87"
+
+    ary_ucs4_1 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_monday), '', 'R')
+    assert_equal ary_ucs4_1, [0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0xfeed, 0xfea9]
+    ary_ucs4_2 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_monday), '', 'L')
+    assert_equal ary_ucs4_2, ary_ucs4_1
+  end
+
+  test "Bidi Persian and English test" do
+    pdf = MYPDF.new
+
+    utf8_persian_str_sunday = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4\xd9\x86\xd8\xa8\xd9\x87"
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_sunday + ' abc'))
+    assert_equal ary_ucs4, [0x61, 0x62, 0x63, 0x20, # 'abc '
+                            0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe] # Sunday
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_sunday + ' abc'), '', 'R')
+    assert_equal ary_ucs4, [0x61, 0x62, 0x63, 0x20, # 'abc '
+                            0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe] # Sunday
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_persian_str_sunday + ' abc'), '', 'L')
+    assert_equal ary_ucs4, [0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe, # Sunday
+                            0x20, 0x61, 0x62, 0x63] # 'abc '
+  end
+
+  test "Bidi English and Persian test" do
+    pdf = MYPDF.new
+
+    utf8_persian_str_sunday = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4\xd9\x86\xd8\xa8\xd9\x87"
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray('abc ' + utf8_persian_str_sunday))
+    assert_equal ary_ucs4, [0x61, 0x62, 0x63, 0x20, # 'abc '
+                            0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe] # Sunday
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray('abc ' + utf8_persian_str_sunday), '', 'L')
+    assert_equal ary_ucs4, [0x61, 0x62, 0x63, 0x20, # 'abc '
+                            0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe] # Sunday
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray('abc ' + utf8_persian_str_sunday), '', 'R')
+    assert_equal ary_ucs4, [0xfeea, 0xfe92, 0xfee8, 0xfeb7, 0x200C, 0xfb8f, 0xfbfe, # Sunday
+                            0x20, 0x61, 0x62, 0x63] # 'abc '
   end
 
   test "Bidi date test" do
@@ -128,4 +358,10 @@ class RbpdfTest < ActiveSupport::TestCase
     rtn = pdf.cache_utf8_string_to_array('1234')
     assert_equal rtn, [0x31, 0x32, 0x33, 0x34]
   end
+
+  test "UniArrSubString test" do
+    pdf = RBPDF.new
+    str = pdf.uni_arr_sub_string(['a', 'b', 'c', ' ', 'd', 'e', 'f'])
+    assert_equal str, 'abc def'
+  end
 end

data/test/rbpdf_content_test.rb

@@ -1,3 +1,4 @@
+# coding: ASCII-8BIT
 require 'test_helper'
 
 class RbpdfPageTest < ActiveSupport::TestCase
@@ -112,6 +113,84 @@ class RbpdfPageTest < ActiveSupport::TestCase
     assert_equal content[12], '340.88 141.17 370.62 158.34 392.04 183.86 c'  # 8/9 circle
     assert_equal content[13], '413.45 209.38 425.20 241.65 425.20 274.96 c'  # 9/9 circle
     assert_equal content[14], 'S'
+  end
+
+  test "write content test" do
+    pdf = MYPDF.new
+    pdf.add_page()
+    page = pdf.get_page
+    assert_equal 1, page
+
+    content = []
+    line = pdf.write(0, "abc def")
+    contents = pdf.getPageBuffer(page)
+    contents.each_line {|line| content.push line.chomp }
+    assert_equal content.length, 22
+    assert_equal content[21], "BT 31.19 801.84 Td 0 Tr 0.00 w [(abc def)] TJ ET"
+  end
+
+  test "write content RTL test" do
+    pdf = MYPDF.new
+    pdf.set_rtl(true)
+    pdf.add_page()
+    page = pdf.get_page
+    assert_equal 1, page
+
+    content = []
+    line = pdf.write(0, "abc def")
+    contents = pdf.getPageBuffer(page)
+    contents.each_line {|line| content.push line.chomp }
+    assert_equal content.length,  22
+    assert_equal content[21], "BT 524.73 801.84 Td 0 Tr 0.00 w [(abc def)] TJ ET"
+  end
+
+  test "write Persian Sunday content test" do
+    pdf = MYPDF.new
+    pdf.set_font('dejavusans', '', 18)
+    pdf.add_page()
+    page = pdf.get_page
+    assert_equal 1, page
 
+    utf8_persian_str_sunday = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4\xd9\x86\xd8\xa8\xd9\x87"
+    content = []
+    line = pdf.write(0, utf8_persian_str_sunday)
+    contents = pdf.getPageBuffer(page)
+
+    contents.each_line {|line| content.push line.chomp }
+    assert_equal content.length, 22
+    assert_equal content[21], "BT 31.19 796.06 Td 0 Tr 0.00 w [(\xFE\xEA\xFE\x92\xFE\xE8\xFE\xB7 \f\xFB\x8F\xFB\xFE)] TJ ET"
+
+    pdf.set_rtl(true)
+    line = pdf.write(0, utf8_persian_str_sunday)
+    contents = pdf.getPageBuffer(page)
+
+    contents.each_line {|line| content.push line.chomp }
+    assert_equal content.length, 46
+    assert_equal content[45], "BT 507.38 796.06 Td 0 Tr 0.00 w [(\xFE\xEA\xFE\x92\xFE\xE8\xFE\xB7 \f\xFB\x8F\xFB\xFE)] TJ ET"
+  end
+
+  test "write English and Persian Sunday content test" do
+    pdf = MYPDF.new
+    pdf.set_font('dejavusans', '', 18)
+    pdf.add_page()
+    page = pdf.get_page
+    assert_equal 1, page
+
+    utf8_persian_str_sunday = "\xdb\x8c\xda\xa9\xe2\x80\x8c\xd8\xb4\xd9\x86\xd8\xa8\xd9\x87"
+    content = []
+    line = pdf.write(0, 'abc def ' + utf8_persian_str_sunday)
+    contents = pdf.getPageBuffer(page)
+
+    contents.each_line {|line| content.push line.chomp }
+    assert_equal content.length, 22
+    assert_equal content[21], "BT 31.19 796.06 Td 0 Tr 0.00 w [(\x00a\x00b\x00c\x00 \x00d\x00e\x00f\x00 \xFE\xEA\xFE\x92\xFE\xE8\xFE\xB7 \f\xFB\x8F\xFB\xFE)] TJ ET"
+
+    pdf.set_rtl(true)
+    line = pdf.write(0, 'abc def ' + utf8_persian_str_sunday)
+    contents = pdf.getPageBuffer(page)
+
+    contents.each_line {|line| content.push line.chomp }
+    assert_equal content.length, 46
+    assert_equal content[45], "BT 434.73 796.06 Td 0 Tr 0.00 w [(\xFE\xEA\xFE\x92\xFE\xE8\xFE\xB7 \f\xFB\x8F\xFB\xFE\x00 \x00a\x00b\x00c\x00 \x00d\x00e\x00f)] TJ ET"
   end
 end

data/test/rbpdf_dom_test.rb

@@ -8,6 +8,9 @@ class RbpdfTest < ActiveSupport::TestCase
     def openHTMLTagHandler(dom, key, cell)
       super
     end
+    def get_temp_rtl
+      @tmprtl
+    end
   end
 
   test "Dom Basic" do
@@ -179,6 +182,89 @@ class RbpdfTest < ActiveSupport::TestCase
     assert_equal dom1, dom2
   end
 
+  test "Dom HTMLTagHandler DIR RTL test" do
+    pdf = MYPDF.new
+    pdf.add_page
+    temprtl = pdf.get_temp_rtl
+    assert_equal temprtl, false
+
+    # LTR, ltr
+    htmlcontent = '<p dir="ltr">HTML Example</p>'
+    dom = pdf.getHtmlDomArray(htmlcontent)
+    dom = pdf.openHTMLTagHandler(dom, 1, false)
+
+    assert_equal dom[1]['tag'], true
+    assert_equal dom[1]['opening'], true
+    assert_equal dom[1]['value'], 'p'
+    assert_equal dom[1]['attribute']['dir'], 'ltr'
+
+    temprtl = pdf.get_temp_rtl
+    assert_equal temprtl, false
+
+    # LTR, rtl
+    htmlcontent = '<p dir="rtl">HTML Example</p>'
+    dom = pdf.getHtmlDomArray(htmlcontent)
+    dom = pdf.openHTMLTagHandler(dom, 1, false)
+    assert_equal dom.length, 4
+
+    assert_equal dom[1]['tag'], true
+    assert_equal dom[1]['opening'], true
+    assert_equal dom[1]['value'], 'p'
+    assert_equal dom[1]['attribute']['dir'], 'rtl'
+
+    temprtl = pdf.get_temp_rtl
+    assert_equal temprtl, 'R'
+
+    # LTR, ltr
+    htmlcontent = '<p dir="ltr">HTML Example</p>'
+    dom = pdf.getHtmlDomArray(htmlcontent)
+    dom = pdf.openHTMLTagHandler(dom, 1, false)
+
+    assert_equal dom[1]['tag'], true
+    assert_equal dom[1]['opening'], true
+    assert_equal dom[1]['value'], 'p'
+    assert_equal dom[1]['attribute']['dir'], 'ltr'
+
+    temprtl = pdf.get_temp_rtl
+    assert_equal temprtl, false
+  end
+
+  test "Dom HTMLTagHandler DIR LTR test" do
+    pdf = MYPDF.new
+    pdf.add_page
+    temprtl = pdf.get_temp_rtl
+    assert_equal temprtl, false
+    pdf.set_rtl(true)
+
+    # RTL, ltr
+    htmlcontent = '<p dir="ltr">HTML Example</p>'
+    dom = pdf.getHtmlDomArray(htmlcontent)
+    dom = pdf.openHTMLTagHandler(dom, 1, false)
+    assert_equal dom.length, 4
+
+    assert_equal dom[1]['tag'], true
+    assert_equal dom[1]['opening'], true
+    assert_equal dom[1]['value'], 'p'
+    assert_equal dom[1]['attribute']['dir'], 'ltr'
+
+    temprtl = pdf.get_temp_rtl
+    assert_equal temprtl, 'L'
+
+    # RTL, rtl
+    htmlcontent = '<p dir="rtl">HTML Example</p>'
+    dom = pdf.getHtmlDomArray(htmlcontent)
+    dom = pdf.openHTMLTagHandler(dom, 1, false)
+    assert_equal dom.length, 4
+
+    assert_equal dom[1]['tag'], true
+    assert_equal dom[1]['opening'], true
+    assert_equal dom[1]['value'], 'p'
+    assert_equal dom[1]['attribute']['dir'], 'rtl'
+
+    temprtl = pdf.get_temp_rtl
+    assert_equal temprtl, false
+  end
+
   test "Dom HTMLTagHandler img test" do
     pdf = MYPDF.new
     pdf.add_page

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rbpdf
 version: !ruby/object:Gem::Version
-  version: 1.18.3
+  version: 1.18.4
 platform: ruby
 authors:
 - NAITOH Jun
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-12-06 00:00:00.000000000 Z
+date: 2014-12-21 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler