lm_docstache 3.0.0 → 3.0.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f59dcaa224f4e8e899df674bc5bd296432ebb24e8e4f929be6b414b8f3a8e0a9
-  data.tar.gz: b05e837bccad1978807e82bd7d13da786d24fc796213f4f2859f6ebf023522f1
+  metadata.gz: cc1a3839b3cfabfd78b144d3f2862aa4a8bf1650bc4037220b2e5af4feaa4a31
+  data.tar.gz: d573c864a49f7e2dcc07122fc242664b597a588253408f065ad94cb0f2823f0a
 SHA512:
-  metadata.gz: 900ff62d4ada49c3645abfd6fbb64f30b6d7f7e0bf0d619d378c1702890dc25d15685e0c950b152242ca3438d9c4e09658788cc36171d0dacad57c37046eb545
-  data.tar.gz: abd44d43c73c3012b1512e56c2c0fa985295876217d66d86a83670c73e23054362a29899608a86e90893cc1b82ce5306770a57e45fdbf71f0bec5cacf6518900
+  metadata.gz: 7b0fb9ff483de6b2e4315206d1961f3ff847612519ed7ff11203f8ca9e7aabd35fc8a5f8730ff552c171082a0d29d2a1a126f86e89a3a117e3a10ab0a5fb5222
+  data.tar.gz: a8c510d591b10ee2d66e6c3ac85995ee9eadb8d91f2178c127e70c12b4184d840cfa725929d4d02b40e62b2e2f9f2c7d629bf65866164833e421a5fdbb5e5c7b

data/CHANGELOG.md

@@ -1,5 +1,42 @@
 # Changelog
 
+## 3.0.5
+
+#### Bug fixes and improvements
+
+* Improve the way broken tags are detected, making the algorithm wider in terms
+  detecting broken tags, specially if the broken tag is the opening part of
+  conditional tag blocks (which was being detected before these improvements).
+* Improve the way the paragraphs with "unusable" tags are traversed and have
+  their same-style texts merged (hence the "unusable" tags becoming usable). So,
+  from now, `w:hyperlink` elements, for instance, are properly processed as
+  well.
+
+## 3.0.4
+* Allow replacement `data` argument to be an `Array`. This feature allow to replace blocks
+in a sequentially order following the sequence of matching blocks order.
+
+## 3.0.3
+
+### Bugfix
+
+* Hide custom tags arguments was pushing blocks tags to end of paragraph. There are cases this approach
+ doesn't work. I changed to be an ordered replacement when we match hide tags.
+* Avoid to merge Tab tags on fix errors methods. This was causing unexpected document changes.  
+
+## 3.0.2
+
+### Bugfix
+
+* Fix replacing tags related to hidden custom tags regexp formats. E.g. tab characters.
+
+## 3.0.1
+
+### Bugfix
+
+* Fix Hide Custom Tag feature when document there is no text inside a w|r we
+  can't split content.
+
 ## 3.0.0
 
 ## Breaking Changes

data/lib/lm_docstache.rb

@@ -1,9 +1,9 @@
 require 'nokogiri'
 require 'zip'
 require "lm_docstache/version"
+require "lm_docstache/parser"
 require "lm_docstache/document"
 require 'lm_docstache/hide_custom_tags'
-require "lm_docstache/parser"
 require "lm_docstache/condition"
 require "lm_docstache/conditional_block"
 require "lm_docstache/renderer"

data/lib/lm_docstache/document.rb

@@ -1,7 +1,10 @@
 module LMDocstache
   class Document
-    TAGS_REGEXP = /{{.+?}}/
+    WHOLE_BLOCK_START_REGEX = /^#{Parser::BLOCK_START_PATTERN}$/
+    GENERAL_TAG_REGEX = /\{\{[\/#^]?(.+?)(?:(\s((?:==|~=))\s?.+?))?\}\}/
     ROLES_REGEXP = /({{(sig|sigfirm|date|check|text|initial)\|(req|noreq)\|(.+?)}})/
+    BLOCK_CHILDREN_ELEMENTS = 'w|r,w|hyperlink,w|ins,w|del'
+    RUN_LIKE_ELEMENTS = 'w|r,w|ins'
 
     def initialize(*paths)
       raise ArgumentError if paths.empty?
@@ -34,38 +37,48 @@ module LMDocstache
 
     def tags
       @documents.values.flat_map do |document|
-        document.text.strip.scan(TAGS_REGEXP)
+        document_text = document.text
+        extract_tag_names(document_text) + extract_tag_names(document_text, true)
       end
     end
 
     def usable_tags
       @documents.values.reduce([]) do |tags, document|
         document.css('w|t').reduce(tags) do |document_tags, text_node|
-          document_tags.push(*text_node.text.scan(TAGS_REGEXP))
+          text = text_node.text
+          document_tags.push(*extract_tag_names(text))
+          document_tags.push(*extract_tag_names(text, true))
         end
       end
     end
 
     def usable_tag_names
-      usable_tags.reject { |tag| tag =~ ROLES_REGEXP }.map do |tag|
-        tag.scan(/\{\{[\/#^]?(.+?)(?:(\s((?:==|~=))\s?.+?))?\}\}/)
-        $1
+      usable_tags.reduce([]) do |memo, tag|
+        next memo if !tag.is_a?(Regexp) && tag =~ ROLES_REGEXP
+
+        tag = tag.source if tag.is_a?(Regexp)
+        memo << (tag.scan(GENERAL_TAG_REGEX) && $1)
       end.compact.uniq
     end
 
     def unusable_tags
-      unusable_tags = tags
+      conditional_start_tags = text_nodes_containing_only_starting_conditionals.map(&:text)
+
+      usable_tags.reduce(tags) do |broken_tags, usable_tag|
+        broken_tags.delete_at(broken_tags.index(usable_tag)) && broken_tags
+      end.reject do |broken_tag|
+        operator = broken_tag.is_a?(Regexp) ? :=~ : :==
+        start_tags_index = conditional_start_tags.find_index do |start_tag|
+          broken_tag.send(operator, start_tag)
+        end
 
-      usable_tags.each do |usable_tag|
-        index = unusable_tags.index(usable_tag)
-        unusable_tags.delete_at(index) if index
+        conditional_start_tags.delete_at(start_tags_index) if start_tags_index
+        !!start_tags_index
       end
-
-      unusable_tags
     end
 
     def fix_errors
-      problem_paragraphs.each { |pg| flatten_paragraph(pg) if pg }
+      problem_paragraphs.each { |pg| flatten_text_blocks(pg) if pg }
     end
 
     def errors?
@@ -99,6 +112,25 @@ module LMDocstache
 
     private
 
+    def text_nodes_containing_only_starting_conditionals
+      @documents.values.flat_map do |document|
+        document.css('w|t').select do |paragraph|
+          paragraph.text =~ WHOLE_BLOCK_START_REGEX
+        end
+      end
+    end
+
+    def extract_tag_names(text, conditional_tag = false)
+      if conditional_tag
+        text.scan(Parser::BLOCK_MATCHER).map do |match|
+          start_block_tag = "{{#{match[0]}#{match[1]} #{match[2]} #{match[3]}}}"
+          /#{Regexp.escape(start_block_tag)}/
+        end
+      else
+        text.scan(Parser::VARIABLE_MATCHER).map { |match| "{{#{match[0]}}}" }
+      end
+    end
+
     def render_documents(data, text = nil, render_options = {})
       Hash[
         @documents.map do |(path, document)|
@@ -115,37 +147,48 @@ module LMDocstache
     def problem_paragraphs
       unusable_tags.flat_map do |tag|
         @documents.values.inject([]) do |tags, document|
-          faulty_paragraphs = document
-            .css('w|p')
-            .select { |paragraph| paragraph.text =~ /#{Regexp.escape(tag)}/ }
+          faulty_paragraphs = document.css('w|p').select do |paragraph|
+            tag_regex = tag.is_a?(Regexp) ? tag : /#{Regexp.escape(tag)}/
+            paragraph.text =~ tag_regex
+          end
 
           tags + faulty_paragraphs
         end
       end
     end
 
-    def flatten_paragraph(paragraph)
-      return if (run_nodes = paragraph.css('w|r')).size < 2
+    def flatten_text_blocks(runs_wrapper)
+      return if (children = filtered_children(runs_wrapper)).size < 2
 
-      while run_node = run_nodes.pop
-        next if run_nodes.empty?
+      while node = children.pop
+        is_run_node = node.matches?(RUN_LIKE_ELEMENTS)
+        previous_node = children.last
 
-        style_node = run_node.at_css('w|rPr')
+        if !is_run_node && filtered_children(node, RUN_LIKE_ELEMENTS).any?
+          next flatten_text_blocks(node)
+        end
+        next if !is_run_node || children.empty? || !previous_node.matches?(RUN_LIKE_ELEMENTS)
+        next if node.at_css('w|tab') || previous_node.at_css('w|tab')
+
+        style_node = node.at_css('w|rPr')
         style_html = style_node ? style_node.inner_html : ''
-        previous_run_node = run_nodes.last
-        previous_style_node = previous_run_node.at_css('w|rPr')
+        previous_style_node = previous_node.at_css('w|rPr')
         previous_style_html = previous_style_node ? previous_style_node.inner_html : ''
-        previous_text_node = previous_run_node.at_css('w|t')
-        current_text_node = run_node.at_css('w|t')
+        previous_text_node = previous_node.at_css('w|t')
+        current_text_node = node.at_css('w|t')
 
         next if style_html != previous_style_html
         next if current_text_node.nil? || previous_text_node.nil?
 
-        previous_text_node.content = previous_text_node.text + run_node.text
-        run_node.unlink
+        previous_text_node.content = previous_text_node.text + current_text_node.text
+        node.unlink
       end
     end
 
+    def filtered_children(node, selector = BLOCK_CHILDREN_ELEMENTS)
+      Nokogiri::XML::NodeSet.new(node.document, node.children.filter(selector))
+    end
+
     def unzip_read(zip, zip_path)
       file = zip.find_entry(zip_path)
       contents = ""

data/lib/lm_docstache/hide_custom_tags.rb

@@ -28,16 +28,17 @@ module LMDocstache
           next unless paragraph.text =~ full_pattern
           run_nodes = paragraph.css('w|r')
           while run_node = run_nodes.shift
-            next if run_node.text.to_s.strip.size == 0
-            remainder_run_node = run_node.clone
-            run_node.unlink
-            tag_contents = split_tag_content(remainder_run_node.text, full_pattern)
+            next unless run_node.at_css('w|t')
+            next unless run_node.text =~ full_pattern
+            tag_contents = split_tag_content(run_node.text, full_pattern)
+            replacement_nodes = []
             tag_contents[:content_list].each_with_index do |content, idx|
+              remainder_run_node = run_node.clone
               replace_content(remainder_run_node, content)
-              run_node_with_match = remainder_run_node.dup
               matched_tag = tag_contents[:matched_tags][idx]
-              nodes_list = [remainder_run_node]
+              replacement_nodes << remainder_run_node
               if matched_tag
+                run_node_with_match = run_node.clone
                 replace_style(run_node_with_match)
                 matched_content = matched_tag
                 if value
@@ -46,11 +47,11 @@ module LMDocstache
                        value.to_s
                 end
                 replace_content(run_node_with_match, matched_content)
-                nodes_list << run_node_with_match
+                replacement_nodes << run_node_with_match
               end
-              paragraph << Nokogiri::XML::NodeSet.new(document, nodes_list)
-              remainder_run_node = remainder_run_node.clone
             end
+            run_node.add_next_sibling(Nokogiri::XML::NodeSet.new(document, replacement_nodes))
+            run_node.unlink
           end
         end
       end

data/lib/lm_docstache/parser.rb

@@ -18,7 +18,22 @@ module LMDocstache
     VARIABLE_MATCHER = /{{([^#\^\/].*?)}}/
 
     attr_reader :document, :data, :blocks, :special_variable_replacements, :hide_custom_tags
+    attr_reader :data_sequential_replacement
 
+    # Constructor +data+ argument is a +Hash+ where the key is
+    # expected to be a +String+ representing the replacement block value. +Hash+
+    # key must not contain the `{{}}` part, but only the pattern characters.
+    # As for the values of the +Hash+, we have options:
+    #
+    # * +String+  will be the value that will replace matching string.
+    # * +Array<String>+ will be an ordered sequence of values that will replace the matched string following
+    # document matching order.
+    #
+    # Example:
+    # { 'full_name' => 'John Doe', 'text|req|Client' => ['John', 'Matt', 'Paul'] }
+    #
+    # Constructor +options+ argument is a +Hash+ where keys can be:
+    #
     # The +special_variable_replacements+ option is a +Hash+ where the key is
     # expected to be either a +Regexp+ or a +String+ representing the pattern
     # of more specific type of variables that deserves a special treatment. The
@@ -47,7 +62,8 @@ module LMDocstache
     #   will be the value that will replace the matched string
     def initialize(document, data, options = {})
       @document = document
-      @data = data.transform_keys(&:to_s)
+      @data = data.transform_keys(&:to_s).select {|e, v| !v.is_a?(Array) }
+      @data_sequential_replacement = data.transform_keys(&:to_s).select {|e, v| v.is_a?(Array) }
       @special_variable_replacements = add_blocks_to_regexp(options.fetch(:special_variable_replacements, {}))
       @hide_custom_tags = add_blocks_to_regexp(options.fetch(:hide_custom_tags, {}))
     end
@@ -65,6 +81,7 @@ module LMDocstache
       hide_custom_tags!
       find_blocks
       replace_conditional_blocks_in_document!
+      replace_data_sequentially_in_document!
       replace_variables_in_document!
     end
 
@@ -140,8 +157,35 @@ module LMDocstache
       end
     end
 
+    def replace_data_sequentially_in_document!
+      data_sequential_replacement.each do |tag_key, values|
+
+        tag = Regexp.escape("{{#{tag_key}}}")
+        pattern_found = 0
+
+        document.css('w|t').each do |text_node|
+          text = text_node.text
+
+          if text.match(tag)
+
+            text.gsub!(/#{tag}/) do |_match|
+              value = values[pattern_found]
+              # if there is no more available value replace the content with empty string
+              return '' unless value
+
+              pattern_found +=1
+              value
+            end
+
+            text_node.content = text
+          end
+        end
+      end
+    end
+
     def has_skippable_variable?(text)
-      return true if hide_custom_tags.find { |(pattern, value)| text =~ pattern }
+      return true if hide_custom_tags.find { |(pattern, _)| text =~ pattern }
+
       !!special_variable_replacements.find do |(pattern, value)|
         text =~ pattern && value == false
       end

data/lib/lm_docstache/version.rb

@@ -1,3 +1,3 @@
 module LMDocstache
-  VERSION = "3.0.0"
+  VERSION = "3.0.5"
 end

data/spec/hide_custom_tags_spec.rb

@@ -85,5 +85,14 @@ describe LMDocstache::HideCustomTags do
         expect(total_replacement).to eq(2)
       end
     end
+
+    context 'giving a document with tabs spacing in the middle of replacement tags' do
+      let(:input_file) { "#{base_path}/sample-signature-with-tabs-spacing.docx" }
+      it 'expect to not replace tabs' do
+        hide_custom_tags.hide_custom_tags!
+        d = hide_custom_tags.document
+        expect(d.css('w|p w|tab').size).to eq(11)
+      end
+    end
   end
 end

data/spec/integration_spec.rb

@@ -62,7 +62,7 @@ describe 'integration test', integration: true do
     it 'fixes nested xml errors breaking tags' do
       expect { document.fix_errors }.to change {
         document.send(:problem_paragraphs).size
-      }.from(6).to(1)
+      }.from(7).to(1)
 
       expect(document.send(:problem_paragraphs).first.text).to eq(
         '{{TAG123-\\-//WITH WE👻IRD CHARS}}'
@@ -70,7 +70,7 @@ describe 'integration test', integration: true do
     end
 
     it 'has the expected amount of usable tags' do
-      expect(document.usable_tags.count).to eq(43)
+      expect(document.usable_tags.count).to eq(21)
     end
 
     it 'has the expected amount of usable roles tags' do
@@ -79,7 +79,7 @@ describe 'integration test', integration: true do
     end
 
     it 'has the expected amount of unique tag names' do
-      expect(document.usable_tag_names.count).to eq(19)
+      expect(document.usable_tag_names.count).to eq(14)
     end
 
     it 'renders file using data' do

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: lm_docstache
 version: !ruby/object:Gem::Version
-  version: 3.0.0
+  version: 3.0.5
 platform: ruby
 authors:
 - Roey Chasman
@@ -12,7 +12,7 @@ authors:
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2021-03-23 00:00:00.000000000 Z
+date: 2021-06-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
@@ -123,6 +123,7 @@ files:
 - spec/example_input/blank.docx
 - spec/example_input/docx-no-rpr.docx
 - spec/example_input/sample-signature-blue.docx
+- spec/example_input/sample-signature-with-tabs-spacing.docx
 - spec/example_input/sample-signature.docx
 - spec/hide_custom_tags_spec.rb
 - spec/integration_spec.rb
@@ -157,6 +158,7 @@ test_files:
 - spec/example_input/blank.docx
 - spec/example_input/docx-no-rpr.docx
 - spec/example_input/sample-signature-blue.docx
+- spec/example_input/sample-signature-with-tabs-spacing.docx
 - spec/example_input/sample-signature.docx
 - spec/hide_custom_tags_spec.rb
 - spec/integration_spec.rb