rbpdf 1.18.2 → 1.18.3

data/README.md

@@ -14,15 +14,15 @@ A template plugin allowing the inclusion of ERB-enabled RBPDF template files.
 * HTML tag support.
 * CSS minimum support.
 * Image
- - 8bit PNG image support without RMagick liblary.
- - PNG(with alpha channel)/JPEG/GIF image support. (use RMagick liblary)
+ - 8bit PNG image support without RMagick library.
+ - PNG(with alpha channel)/JPEG/GIF image support. (use RMagick library)
 
 
 ##
 ## Installing RBPDF
 ##
 
-RBPDF is distrifuted via RubyGems, and can be installed the usual way that you install gems: by simply typing `gem install rbpdf` on the command line. 
+RBPDF is distributed via RubyGems, and can be installed the usual way that you install gems: by simply typing `gem install rbpdf` on the command line. 
 
 ==
 

data/lib/rbpdf.rb

@@ -363,6 +363,7 @@ class RBPDF
     @radiobutton_groups ||= []
     @radio_groups ||= []
     @textindent ||= 0
+    @nested_table = false
 
     @start_transaction_y ||= 0
     @in_thead ||= false
@@ -1110,8 +1111,8 @@ class RBPDF
   # [@access public]
   # [@since 4.5.027 (2009-03-16)]
   #
-  def getPageDimensions(pagenum='')
-    if pagenum.empty?
+  def getPageDimensions(pagenum=0)
+    if !pagenum.is_a? Integer or pagenum.zero?
       pagenum = @page
     end
     return @pagedim[pagenum]
@@ -1161,15 +1162,15 @@ class RBPDF
   # [@since 1.5.2]
   # [@see] getPageDimensions()
   #
-  def getBreakMargin(pagenum='')
-    if pagenum.empty?
+  def getBreakMargin(pagenum=0)
+    if !pagenum.is_a? Integer or pagenum.zero?
       return @b_margin
     end
     return @pagedim[pagenum]['bm']
   end
   alias_method :get_break_margin, :getBreakMargin
 
-  def GetBreakMargin(pagenum='')
+  def GetBreakMargin(pagenum=0)
     warn "[DEPRECATION] 'GetBreakMargin' is deprecated. Please use 'get_break_margin' instead."
     getBreakMargin(pagenum)
   end
@@ -3287,6 +3288,7 @@ class RBPDF
   # [@since (4.5.019) 2009-02-28]
   #
   def removeSHY(txt='')
+    txt = txt.dup
     # Unicode Data
     # Name : SOFT HYPHEN, commonly abbreviated as SHY
     # HTML Entity (decimal): ­
@@ -4209,6 +4211,7 @@ class RBPDF
   # [@since 1.5]
   #
   def Write(h, txt, link=nil, fill=0, align='', ln=false, stretch=0, firstline=false, firstblock=false, maxh=0)
+    txt = txt.dup
     txt.force_encoding('ASCII-8BIT') if txt.respond_to?(:force_encoding)
     if txt.length == 0
       txt = ' '
@@ -4359,7 +4362,7 @@ 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-1], '', @tmprtl))
+          l = GetArrStringWidth(utf8Bidi(chars[j,i-j], '', @tmprtl))
         else
           l += GetCharWidth(c)
         end
@@ -4846,7 +4849,7 @@ class RBPDF
       if (type == '')
         type = getImageFileType(file, imsize)
       else
-        type.downcase!
+        type = type.downcase
         type = 'jpeg' if type == 'jpg'
       end
 
@@ -6366,7 +6369,7 @@ protected
     stream = stream.strip
     @apxo_obj_id += 1
     @offsets[@apxo_obj_id] = @bufferlen
-    out = @apxo_obj_id + ' 0 obj'
+    out = @apxo_obj_id.to_s + ' 0 obj'
     out << ' <<'
     out << ' /Type /XObject'
     out << ' /Subtype /Form'
@@ -7586,7 +7589,7 @@ protected
   def UTF8StringToArray(str)
     if @cache_utf8_string_to_array[str]
       # return cached value
-      return @cache_utf8_string_to_array[str]
+      return @cache_utf8_string_to_array[str].dup
     end
     # check cache size
     if @cache_size_utf8_string_to_array >= @cache_maxsize_utf8_string_to_array
@@ -7602,7 +7605,7 @@ protected
         strarr[pos] = char.unpack('C')[0]
       end
       # insert new value on cache
-      @cache_utf8_string_to_array[str] = strarr
+      @cache_utf8_string_to_array[str] = strarr.dup
       return strarr
     end
 
@@ -7661,7 +7664,7 @@ protected
       end
     end
     # insert new value on cache
-    @cache_utf8_string_to_array[str] = unicode
+    @cache_utf8_string_to_array[str] = unicode.dup
     return unicode;
   end
   
@@ -9011,6 +9014,7 @@ public
   # [@since 2.4.000 (2008-03-06)]
   #
   def utf8Bidi(ta, str='', forcertl=false)
+    str = str.dup
     # paragraph embedding level
     pel = 0
     # max level
@@ -9567,7 +9571,6 @@ public
     
     ordarray = []
     0.upto(numchars-1) do |i|
-      chardata[i][:char]
       ordarray.push chardata[i][:char]
     end
     
@@ -9685,8 +9688,8 @@ public
   # [@access public]
   # [@since 3.0.000 (2008-03-27)]
   #
-  def startPageGroup(page='')
-    if page.empty?
+  def startPageGroup(page=0)
+    if !page.is_a? Integer or page.zero?
       page = @page + 1
     end
     @newpagegroup[page] = true
@@ -10324,7 +10327,7 @@ protected
       return {}
     end
     # remove comments
-    cssdata.gsub!(/\/\*[^\*]*\*\//, '')
+    cssdata = cssdata.gsub(/\/\*[^\*]*\*\//, '')
     # remove newlines and multiple spaces
     cssdata.gsub!(/[\s]+/, ' ')
     # remove some spaces
@@ -10573,6 +10576,7 @@ protected
   # [@since 3.2.000 (2008-06-20)]
   #
   def getHtmlDomArray(html)
+    html = html.dup
     html.force_encoding('UTF-8') if @is_unicode and html.respond_to?(:force_encoding)
     # array of CSS styles ( selector => properties).
     css = {}
@@ -10642,7 +10646,6 @@ protected
     html = "%s" % sanitize(html, :tags=> %w(a b blockquote body br dd del div dl dt em font h1 h2 h3 h4 h5 h6 hr i img li ol p pre small span strong sub sup table tablehead td th thead tr tt u ins ul), :attributes => %w(cellspacing cellpadding bgcolor color value width height src size colspan rowspan style align border face href dir class id nobr stroke strokecolor fill nested))
 
     # replace some blank characters
-    html.gsub!(/<br>/, '<br/>')
     html.gsub!(/<pre/, '<xre') # preserve pre tag
     html.gsub!(/<(table|tr|td|th|blockquote|dd|div|dt|h1|h2|h3|h4|h5|h6|br|hr|li|ol|ul|p)([^\>]*)>[\n\r\t]+/, '<\\1\\2>')
     html.gsub!(/@(\r\n|\r)@/, "\n")
@@ -10777,7 +10780,11 @@ protected
           # store header rows on a new table
           if (dom[key]['value'] == 'tr') and (dom[(dom[key]['parent'])]['thead'] == true)
             if empty_string(dom[grandparent]['thead'])
-              dom[grandparent]['thead'] = a[dom[grandparent]['elkey']].dup
+              if dom[grandparent]['attribute'].nil? or dom[grandparent]['attribute']['style'].nil? 
+                dom[grandparent]['thead'] = a[dom[grandparent]['elkey']].dup
+              else
+                dom[grandparent]['thead'] = '<style>' + dom[grandparent]['value'] + ' {' + dom[grandparent]['attribute']['style'] + '}</style>' + a[dom[grandparent]['elkey']].dup
+              end
             end
             dom[key]['parent'].upto(key) do |i|
               dom[grandparent]['thead'] << a[dom[i]['elkey']]
@@ -10797,12 +10804,12 @@ protected
           # *** opening html tag
           dom[key]['opening'] = true
           dom[key]['parent'] = level[-1]
-          if element[-1, 1] != '/'
-            # not self-closing tag
+          if element[-1, 1] == '/' or (dom[key]['value'] =~ /(br|img|hr)/)
+            # self-closing tag
+            dom[key]['self'] = true
+          else
             level.push(key)
             dom[key]['self'] = false
-          else
-            dom[key]['self'] = true
           end
           # copy some values from parent
           parentkey = 0
@@ -11228,7 +11235,10 @@ public
   # [@see] Multicell(), writeHTML(), Cell()
   #
   def writeHTMLCell(w, h, x, y, html='', border=0, ln=0, fill=0, reseth=true, align='', autopadding=true)
-    return MultiCell(w, h, html, border, align, fill, ln, x, y, reseth, 0, true, autopadding, 0)
+    @nested_table = true
+    rtn = MultiCell(w, h, html, border, align, fill, ln, x, y, reseth, 0, true, autopadding, 0)
+    @nested_table = false
+    return rtn
   end
   alias_method :write_html_cell, :writeHTMLCell
 
@@ -11957,7 +11967,7 @@ public
             else
               wtmp = @w - @r_margin - @x
             end
-            if dom[key]['attribute']['nested'] and (dom[key]['attribute']['nested'] == 'true')
+            if (@nested_table == true) or (dom[key]['attribute']['nested'] and (dom[key]['attribute']['nested'] == 'true'))
               # add margin for nested tables
               wtmp -= @c_margin
             end
@@ -12541,6 +12551,12 @@ public
             xpos += 2 * GetStringWidth(32.chr)
           end
         end
+        if @rtl
+            xpos -= @line_width * 1.5
+        else
+            xpos += @line_width * 1.5
+        end
+
         imglink = ''
         if !@href['url'].nil? and !empty_string(@href['url'])
           imglink = @href['url']
@@ -12573,13 +12589,20 @@ public
           ih = getHTMLUnitToUnits(tag['attribute']['height'], 1, 'px', false)
         end
 
+        # store original margin values
+        l_margin = @l_margin
+        r_margin = @r_margin
+
+        SetLeftMargin(@l_margin + @line_width * 1.5)
+        SetRightMargin(@r_margin + @line_width * 1.5)
+
         begin
 #        if (type == 'eps') or (type == 'ai')
 #          ImageEps(tag['attribute']['src'], xpos, @y, iw, ih, imglink, true, align, '', border, true)
 #        elsif type == 'svg'
 #          ImageSVG(tag['attribute']['src'], xpos, @y, iw, ih, imglink, align, '', border, true)
 #        else
-          result_img = Image(tag['attribute']['src'], xpos, @y, iw, ih, '', imglink, align, false, 300, '', false, false, border, false, false, true)
+          result_img = Image(tag['attribute']['src'], xpos, @y + @line_width * 1.5, iw, ih, '', imglink, align, false, 300, '', false, false, border, false, false, true)
 #        end
         rescue => err
           logger.error "pdf: Image: error: #{err.message}"
@@ -12593,6 +12616,11 @@ public
         when 'B'
           @y = @img_rb_y - (tag['fontsize'] / @k)
         end
+
+        # restore original margin values
+        SetLeftMargin(l_margin)
+        SetRightMargin(r_margin)
+
         if result_img == false
           Write(@lasth, File::basename(img_name) + ' ', '', false, '', false, 0, false) unless img_name.nil?
         end
@@ -12961,7 +12989,7 @@ public
         }
         if !in_table_head
           # we are not inside a thead section
-          if !parent['cellpadding'].nil?
+          if dom[(parent['parent'])]['attribute']['cellpadding'] ### fix ###
             @c_margin = @old_c_margin
           end
           @lasth = @font_size * @cell_height_ratio
@@ -12980,6 +13008,9 @@ public
             @thead_margins = {}
           end
         end
+        if tag['block'] ### fix ###
+          addHTMLVertSpace(hbz / 2, 0, cell, (dom[key+1].nil? or (dom[key+1]['value'] != 'table')))
+        end
       when 'a'
         @href = {}
       when 'sup'

data/lib/rbpdf/version.rb

@@ -1,3 +1,3 @@
 module Rbpdf
-  VERSION = "1.18.2"
+  VERSION = "1.18.3"
 end

data/rbpdf.gemspec

@@ -20,7 +20,7 @@ Gem::Specification.new do |spec|
                        Dir.glob("lib/fonts/dejavu-fonts-ttf-*/{AUTHORS,BUGS,LICENSE,NEWS,README}") +
                        Dir.glob("test/*") +
                        ["Rakefile", "rbpdf.gemspec", "Gemfile",
-                        "CHANGELOG", "test_unicode.rbpdf", "README.md",
+                        "CHANGELOG", "test_unicode.rbpdf", "README.md", "LICENSE.TXT",
                         "utf8test.txt", "logo_example.png" ]
   spec.rdoc_options  += [ '--exclude', 'lib/fonts/',
                           '--exclude', 'lib/htmlcolors.rb',

data/test/rbpdf_bidi_test.rb

@@ -9,6 +9,9 @@ class RbpdfTest < ActiveSupport::TestCase
     def utf8Bidi(ta, str='', forcertl=false)
       super
     end
+    def cache_utf8_string_to_array(str)
+      @cache_utf8_string_to_array[str]
+    end
   end
 
   test "Bidi" do
@@ -62,4 +65,67 @@ class RbpdfTest < ActiveSupport::TestCase
     assert_equal [0x61, 0x62, 0x63, 0x5ea, 0x5d9, 0x5e8, 0x5d1, 0x5e2], ary_str
   end
 
+  test "Bidi arabic test" do
+    pdf = MYPDF.new
+
+    # Bidirectional Algorithm
+    ascii_str   = "role"
+    utf8_arabic_str_1  = "\xd8\xaf\xd9\x88\xd8\xb1"
+    utf8_arabic_char_1  = "\xd8\xaf"
+
+    # UCS4 charactor -> UTF-8 charactor
+    utf8_chr = pdf.unichr(0x62f)
+    assert_equal "\xd8\xaf", utf8_chr
+
+    # UTF-8 string -> array of UCS4 charactor
+    ary_ucs4 = pdf.UTF8StringToArray(ascii_str)
+    assert_equal [0x72, 0x6f, 0x6c, 0x65], ary_ucs4
+    ary_ucs4 = pdf.UTF8StringToArray(utf8_arabic_str_1)
+    assert_equal [0x62f, 0x648, 0x631], ary_ucs4
+    ary_ucs4 = pdf.UTF8StringToArray(utf8_arabic_char_1)
+    assert_equal [0x62f], ary_ucs4
+
+
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str))
+    assert_equal [0x72, 0x6f, 0x6c, 0x65], ary_ucs4
+
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(ascii_str), ascii_str, 'R')
+    assert_equal [0x72, 0x6f, 0x6c, 0x65], ary_ucs4
+
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_arabic_char_1))
+    assert_equal [0xfea9], ary_ucs4
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_arabic_char_1), utf8_arabic_char_1, 'R')
+    assert_equal [0xfea9], ary_ucs4
+
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_arabic_str_1))
+    assert_equal [0xfeae, 0xfeee, 0xfea9], ary_ucs4
+  end
+
+  test "Bidi date test" do
+    pdf = MYPDF.new
+
+    utf8_date_str_1  = '12/01/2014'
+
+    pdf.set_rtl(true)
+    pdf.set_temp_rtl('rtl')
+    # UTF-8 string -> array of UCS4 charactor
+    ary_ucs4 = pdf.UTF8StringToArray(utf8_date_str_1)
+    assert_equal [0x31, 0x32, 0x2f, 0x30, 0x31, 0x2f, 0x32, 0x30, 0x31, 0x34], ary_ucs4  # 12/01/2014
+
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_date_str_1))
+    assert_equal [0x31, 0x32, 0x2f, 0x30, 0x31, 0x2f, 0x32, 0x30, 0x31, 0x34], ary_ucs4  # 12/01/2014
+
+    ary_ucs4 = pdf.utf8Bidi(pdf.UTF8StringToArray(utf8_date_str_1), utf8_date_str_1, 'R')
+    assert_equal [0x31, 0x32, 0x2f, 0x30, 0x31, 0x2f, 0x32, 0x30, 0x31, 0x34], ary_ucs4  # 12/01/2014
+  end
+
+  test "UTF8StringToArray cache_utf8_string_to_array test" do
+    pdf = MYPDF.new
+
+    chars = pdf.UTF8StringToArray('1234')
+    chars.reverse!
+
+    rtn = pdf.cache_utf8_string_to_array('1234')
+    assert_equal rtn, [0x31, 0x32, 0x33, 0x34]
+  end
 end

data/test/rbpdf_cell_test.rb

@@ -250,4 +250,18 @@ class RbpdfTest < ActiveSupport::TestCase
     line = pdf.get_num_lines(txt, w)
     assert_equal line, 3
   end
+
+  test "removeSHY encoding test" do
+    return unless 'test'.respond_to?(:force_encoding)
+
+    pdf = RBPDF.new('P', 'mm', 'A4', true, "UTF-8", true)
+
+    str = 'test'.force_encoding('UTF-8')
+    txt = pdf.removeSHY(str)
+    assert_equal str.encoding.to_s, 'UTF-8'
+
+    str = 'test'.force_encoding('ASCII-8BIT')
+    txt = pdf.removeSHY(str)
+    assert_equal str.encoding.to_s, 'ASCII-8BIT'
+  end
 end

data/test/rbpdf_css_test.rb

@@ -181,12 +181,55 @@ class RbpdfCssTest < ActiveSupport::TestCase
     assert_equal dom[1]['value'], 'h2'
 
     # <table>
+    assert_equal dom[4]['elkey'], 3
     assert_equal dom[4]['value'], 'table'
     assert_equal dom[4]['attribute'], {'border'=>'2px #ff0000 solid', 'style'=>';border:2px #ff0000 solid;'}
     assert_equal dom[4]['style']['border'], '2px #ff0000 solid'
     assert_equal dom[4]['attribute']['border'], '2px #ff0000 solid'
   end
 
+  test "CSS Dom table thead test" do
+    pdf = MYPDF.new
+
+    html = '<style> table, td { border: 2px #ff0000 solid; } </style>
+            <h2>HTML TABLE THEAD:</h2>
+            <table><thead>
+            <tr> <th>abc</th> </tr>
+            </thead>
+            <tbody>
+            <tr> <td>def</td> </tr>
+            <tr> <td>ghi</td> </tr>
+            </tbody></table>'
+
+    dom = pdf.getHtmlDomArray(html)
+    ## remove style tag block (by getHtmlDomArray())       ##
+    ## remove thead/tbody tag block (by getHtmlDomArray()) ##
+    ## added marker tag (by getHtmlDomArray())             ##
+    # '<h2>HTML TABLE:</h2>
+    #  <table><tr><th>abc<marker style="font-size:0"/></th></tr>
+    #         <tr><td>def<marker style="font-size:0"/></td></tr></table>'
+    assert_equal dom.length, 24
+
+    assert_equal dom[0]['parent'], 0  # Root
+    assert_equal dom[0]['tag'], false
+    assert_equal dom[0]['attribute'], {}
+
+    # <h2>
+    assert_equal dom[1]['elkey'], 0
+    assert_equal dom[1]['parent'], 0   # parent -> parent tag key
+    assert_equal dom[1]['tag'], true
+    assert_equal dom[1]['opening'], true
+    assert_equal dom[1]['value'], 'h2'
+
+    # <table>
+    assert_equal dom[4]['elkey'], 3
+    assert_equal dom[4]['value'], 'table'
+    assert_equal dom[4]['attribute'], {'border'=>'2px #ff0000 solid', 'style'=>';border:2px #ff0000 solid;'}
+    assert_equal dom[4]['style']['border'], '2px #ff0000 solid'
+    assert_equal dom[4]['attribute']['border'], '2px #ff0000 solid'
+    assert_equal dom[4]['thead'], '<style>table {;border:2px #ff0000 solid;}</style><table><tr><th>abc<marker style="font-size:0"/></th></tr></tablehead>'
+  end
+
   test "CSS Dom line-height test normal" do
     pdf = MYPDF.new