Merge pull request #33 from blocknotes/v0.6.2

v0.6.2
blocknotes · Sep 20, 2021 · 3bfb469 · 3bfb469
2 parents 9230efc + 87b66df
commit 3bfb469
Show file tree

Hide file tree

Showing 22 changed files with 165 additions and 130 deletions.
diff --git a/README.md b/README.md
@@ -35,34 +35,34 @@ To check some examples with the PDF output see [examples](examples/) folder.
 
 ## Supported tags & attributes
 
-HTML tags:
-
-- **a**: link
-- **b**: bold
-- **blockquote**: block quotation element
-- **br**: new line
-- **code**: inline code element
-- **del**: strike-through
-- **div**: block element
-- **em**: italic
-- **h1** - **h6**: headings
-- **hr**: horizontal line
-- **i**: italic
-- **ins**: underline
-- **img**: image
-- **li**: list item
-- **mark**: highlight
-- **ol**: ordered list
-- **p**: block element
-- **pre**: preformatted text element
-- **s**: strike-through
-- **small**: smaller text
-- **span**: inline element
-- **strong**: bold
-- **sub**: subscript element
-- **sup**: superscript element
-- **u**: underline
-- **ul**: unordered list
+HTML tags (using MDN definitions):
+
+- **a**: the Anchor element
+- **b**: the Bring Attention To element
+- **blockquote**: the Block Quotation element
+- **br**: the Line Break element
+- **code**: the Inline Code element
+- **del**: the Deleted Text element
+- **div**: the Content Division element
+- **em**: the Emphasis element
+- **h1** - **h6**: the HTML Section Heading elements
+- **hr**: the Thematic Break (Horizontal Rule) element
+- **i**: the Idiomatic Text element
+- **ins**: the added text element
+- **img**: the Image Embed element
+- **li**: the list item element
+- **mark**: the Mark Text element
+- **ol**: the Ordered List element
+- **p**: the Paragraph element
+- **pre**: the Preformatted Text element
+- **s**: the strike-through text element
+- **small**: the side comment element
+- **span**: the generic inline element
+- **strong**: the Strong Importance element
+- **sub**: the Subscript element
+- **sup**: the Superscript element
+- **u**: the Unarticulated Annotation (Underline) element
+- **ul**: the Unordered List element
 
 CSS attributes (dimensional units are ignored and considered in pixel):
 

diff --git a/examples/elements.html b/examples/elements.html
@@ -0,0 +1,44 @@
+<!DOCTYPE html>
+<html>
+<head>
+  <title>Supported elements</title>
+</head>
+<body style="font-family: 'Times-Roman';">
+  tag A: <a href="https://www.google.com">link</a> <br>
+  tag B: <b>bold</b> <br>
+  tag Blockquote: <blockquote>block quotation element</blockquote>
+  tag Br: new line <br>
+  tag Code: <code>inline code element</code> <br>
+  tag Del: <del>strike-through</del> <br>
+  tag Div: <div>block element</div>
+  tag Em: <em>italic</em> <br>
+  tag H1: <h1>heading h1</h1>
+  tag H2: <h2>heading h2</h2>
+  tag H3: <h3>heading h3</h3>
+  tag H4: <h4>heading h4</h4>
+  tag H5: <h5>heading h5</h5>
+  tag H6: <h6>heading h6</h6>
+  tag Hr: horizontal line <hr>
+  tag I: <i>italic</i> <br>
+  tag Ins: <ins>underline</ins> <br>
+  tag Img: <img src="./examples/image.jpg" style="width: 100px"> image <br>
+  tag Mark: <mark>highlight</mark> <br>
+  tag Ol: <ol>
+    <li>ordered list</li>
+    <li>tag Li: list item (ordered)</li>
+  </ol>
+  tag P: <p>block element</p>
+  tag Pre: <pre>preformatted text element</pre>
+  tag S: <s>strike-through</s> <br>
+  tag Small: <small>smaller text</small> <br>
+  tag Span: <span>inline element</span> <br>
+  tag Strong: <strong>bold</strong> <br>
+  tag Sub: <sub>subscript element</sub> <br>
+  tag Sup: <sup>superscript element</sup> <br>
+  tag U: <u>underline</u> <br>
+  tag Ul: <ul>
+    <li>unordered list</li>
+    <li>tag Li: list item (unordered)</li>
+  </ul>
+</body>
+</html>
diff --git a/examples/elements.pdf b/examples/elements.pdf
diff --git a/examples/headings.pdf b/examples/headings.pdf
diff --git a/examples/misc_elements.html b/examples/misc_elements.html
diff --git a/examples/misc_elements.pdf b/examples/misc_elements.pdf
diff --git a/examples/random_content.html b/examples/random_content.html
@@ -4,12 +4,22 @@
   <title>Random content</title>
 </head>
 <body style="font-family: 'Times-Roman';">
-  <h1>A test document <small>(with the supported elements)</small></h1>
-  This is <i style="color: #df2818">italic</i>, <u>underline</u> and <b>bold, <i>bold italic</i></b>
+  <h1>Random content <small>(with various elements and styles)</small></h1>
+
+  Test inline elements:
+  this is <i style="color: #df2818">italic</i>, <u>underline</u> and <b>bold, <i>bold italic</i></b>
+
   <br />
   <br />
   <br />
 
+  <div>div element</div>
+  <div>div (2) element</div>
+  <span>span element</span>
+  <span>span (2) element</span>
+
+  <blockquote>The Block Quotation element</blockquote>
+
   <hr style="color: #080" />
 
   <div style="text-align: left">text-align: left</div>
@@ -26,7 +36,7 @@ <h1>A test document <small>(with the supported elements)</small></h1>
   <hr/>
 
   Begin of list
-  <ul style="margin-bottom: 20px">
+  <ul>
     <li>Level 1a</li>
     <li>Level 1b
       <ul>
@@ -35,10 +45,13 @@ <h1>A test document <small>(with the supported elements)</small></h1>
           <ul>
             <li>Level 3a</li>
             <li>Level 3b</li>
+            <li>Level 3c</li>
           </ul>
         </li>
+        <li>Level 2c</li>
       </ul>
     </li>
+    <li>Level 1c</li>
   </ul>
   End of list
 
@@ -86,10 +99,21 @@ <h3>More text</h3>
   <hr />
   <br />
 
+  This is an inline <code>code</code> element.
+  <pre>  this is  pre element  </pre>
+  last line.
+
   <div style="text-align: center">
     An image:<br/>
     <img src="./examples/image.jpg" style="width: 50%" />
   </div>
+  <p>Some other paragraph</p>
+
+  <div style="color: red; position: absolute; top: 100px; left: 300px">
+    <em>Position: absolute</em>
+  </div>
+
+  <h2 style="break-before: auto">New page</h2>
 
   <b>End of document</b>
 </body>

diff --git a/examples/random_content.pdf b/examples/random_content.pdf
diff --git a/examples/styles.pdf b/examples/styles.pdf
diff --git a/lib/prawn-html.rb b/lib/prawn-html.rb
@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 
 module PrawnHtml
+  ADJUST_LEADING = { nil => 0.18, 'Courier' => -0.07, 'Helvetica' => -0.17, 'Times-Roman' => 0.03 }.freeze
   PX = 0.6 # conversion constant for pixel sixes
 
   COLORS = {

diff --git a/lib/prawn_html/attributes.rb b/lib/prawn_html/attributes.rb
@@ -154,7 +154,7 @@ def apply_rule!(merged_styles:, rule:, value:, options:)
       return (@initial << rule) if value == 'initial'
 
       if rule[:set] == :append_styles
-        val = Utils.normalize_style(value)
+        val = Utils.normalize_style(value, rule[:values])
         (merged_styles[rule[:key]] ||= []) << val if val
       else
         opts = rule[:options] ? options[rule[:options]] : nil

diff --git a/lib/prawn_html/context.rb b/lib/prawn_html/context.rb
@@ -67,10 +67,17 @@ def remove_last
       last.on_context_remove(self) if last.respond_to?(:on_context_remove)
       @merged_styles = nil
       @last_text_node = false
-      @previous_tag = last.tag
+      @previous_tag = last
       pop
     end
 
+    # White space is equal to 'pre'?
+    #
+    # @return [boolean] white space property of the last element is equal to 'pre'
+    def white_space_pre?
+      last && last.styles[:white_space] == :pre
+    end
+
     private
 
     def evaluate_element_styles(element, res)

diff --git a/lib/prawn_html/document_renderer.rb b/lib/prawn_html/document_renderer.rb
@@ -12,6 +12,8 @@ def initialize(pdf)
       @buffer = []
       @context = Context.new
       @last_margin = 0
+      @last_text = ''
+      @last_tag_open = false
       @pdf = pdf
     end
 
@@ -22,6 +24,8 @@ def on_tag_close(element)
       render_if_needed(element)
       apply_tag_close_styles(element)
       context.remove_last
+      @last_tag_open = false
+      @last_text = ''
     end
 
     # On tag open callback
@@ -38,6 +42,7 @@ def on_tag_open(tag_name, attributes:, element_styles: '')
       options = { width: pdf.page_width, height: pdf.page_height }
       tag_class.new(tag_name, attributes: attributes, options: options).tap do |element|
         setup_element(element, element_styles: element_styles)
+        @last_tag_open = true
       end
     end
 
@@ -47,9 +52,10 @@ def on_tag_open(tag_name, attributes:, element_styles: '')
     #
     # @return [NilClass] nil value (=> no element)
     def on_text_node(content)
-      return if content.match?(/\A\s*\Z/)
+      return if context.previous_tag&.block? && content.match?(/\A\s*\Z/)
 
-      buffer << context.merged_styles.merge(text: prepare_text(content))
+      text = prepare_text(content)
+      buffer << context.merged_styles.merge(text: text) unless text.empty?
       context.last_text_node = true
       nil
     end
@@ -70,17 +76,13 @@ def render
     attr_reader :buffer, :context, :last_margin, :pdf
 
     def setup_element(element, element_styles:)
-      add_space_if_needed unless render_if_needed(element)
+      render_if_needed(element)
       context.add(element)
       element.process_styles(element_styles: element_styles)
       apply_tag_open_styles(element)
       element.custom_render(pdf, context) if element.respond_to?(:custom_render)
     end
 
-    def add_space_if_needed
-      buffer << SPACE if buffer.any? && !context.last_text_node && ![NEW_LINE, SPACE].include?(buffer.last)
-    end
-
     def render_if_needed(element)
       render_needed = element&.block? && buffer.any? && buffer.last != NEW_LINE
       return false unless render_needed
@@ -104,10 +106,12 @@ def apply_tag_open_styles(element)
     end
 
     def prepare_text(content)
-      white_space_pre = context.last && context.last.styles[:white_space] == :pre
-      text = ::Oga::HTML::Entities.decode(context.before_content)
-      text += white_space_pre ? content : content.gsub(/\A\s*\n\s*|\s*\n\s*\Z/, '').delete("\n").squeeze(' ')
-      text
+      text = context.before_content ? ::Oga::HTML::Entities.decode(context.before_content) : ''
+      return (@last_text = text + content) if context.white_space_pre?
+
+      content = content.lstrip if @last_text[-1] == ' ' || @last_tag_open
+      text += content.tr("\n", ' ').squeeze(' ')
+      @last_text = text
     end
 
     def output_content(buffer, block_styles)
@@ -129,7 +133,10 @@ def apply_callbacks(buffer)
     def adjust_leading(buffer, leading)
       return leading if leading
 
-      (buffer.map { |item| item[:size] || Context::DEFAULT_STYLES[:size] }.max * 0.055).round(4)
+      leadings = buffer.map do |item|
+        (item[:size] || Context::DEFAULT_STYLES[:size]) * (ADJUST_LEADING[item[:font]] || ADJUST_LEADING[nil])
+      end
+      leadings.max.round(4)
     end
 
     def bounds(buffer, options, block_styles)

diff --git a/lib/prawn_html/tags/br.rb b/lib/prawn_html/tags/br.rb
@@ -12,7 +12,7 @@ def block?
       end
 
       def custom_render(pdf, context)
-        return if context.last_text_node || context.previous_tag != :br
+        return if context.last_text_node || !context.previous_tag.is_a?(Br)
 
         pdf.advance_cursor(BR_SPACING)
       end

diff --git a/lib/prawn_html/utils.rb b/lib/prawn_html/utils.rb
@@ -103,11 +103,13 @@ def copy_value(value, options: nil)
     # Normalize a style value
     #
     # @param value [String] string value
+    # @param accepted_values [Array] allowlist of valid values (symbols)
     #
     # @return [Symbol] style value or nil
-    def normalize_style(value)
+    def normalize_style(value, accepted_values)
       val = value&.strip&.downcase
-      NORMALIZE_STYLES[val]
+      ret = NORMALIZE_STYLES[val]
+      accepted_values.include?(ret) ? ret : nil
     end
 
     # Unquotes a string