\n" init_regexp_handlings - - init_tags end # :section: Regexp Handling @@ -72,6 +71,10 @@ def init_regexp_handlings # external links @markup.add_regexp_handling(/(?:link:|https?:|mailto:|ftp:|irc:|www\.)#{URL_CHARACTERS_REGEXP_STR}+\w/, :HYPERLINK) + + # suppress crossref: \#method \::method \ClassName \method_with_underscores + @markup.add_regexp_handling(/\\(?:[#:A-Z]|[a-z]+_[a-z0-9])/, :SUPPRESSED_CROSSREF) + init_link_notation_regexp_handlings end @@ -80,7 +83,6 @@ def init_regexp_handlings def init_link_notation_regexp_handlings add_regexp_handling_RDOCLINK - add_regexp_handling_TIDYLINK end def handle_RDOCLINK(url) # :nodoc: @@ -88,6 +90,7 @@ def handle_RDOCLINK(url) # :nodoc: when /^rdoc-ref:/ CGI.escapeHTML($') when /^rdoc-label:/ + return CGI.escapeHTML(url) if in_tidylink_label? text = $' text = case text @@ -113,11 +116,127 @@ def handle_RDOCLINK(url) # :nodoc: end end - ## - # +target+ is a
')
+ super
+ emit_inline('')
+ end
+
+ def handle_STRIKE(nodes)
+ emit_inline('') + end + + def emit_inline(text) + @inline_output << text + end - def handle_regexp_HARD_BREAK(target) - '
' + # Returns true if we are processing inside a tidy link label. + + def in_tidylink_label? + @in_tidylink_label + end + + # Special handling for tidy link labels. + # When a tidy link is {rdoc-image:path/to/image.jpg:alt text}[http://example.com], + # label part is normally considered RDOCLINK rdoc-image:path/to/image.jpg:alt and a text " text" + # but RDoc's test code expects the whole label part to be treated as RDOCLINK only in tidy link label. + # When a tidy link is {^1}[url] or {*1}[url], the label part needs to drop leading * or ^. + # TODO: reconsider this workaround. + + def apply_tidylink_label_special_handling(label, url) + # ^1 *1 will be converted to just 1 in tidy link label. + return label[1..] if label.match?(/\A[*^]\d+\z/) + + # rdoc-image in label specially allows spaces in alt text. + return handle_RDOCLINK(label) if label.start_with?('rdoc-image:') + end + + def handle_TIDYLINK(label_part, url) + # When url is an image, ignore label part (maybe bug?) and just generate img tag. + if url.match?(/\Ahttps?:\/\/.+\.(png|gif|jpg|jpeg|bmp)\z/) + emit_inline("
", ""
- add_tag :EM, "", ""
- add_tag :STRIKE, "def f(...); end` and `(c) in `` will be wrongly converted.
+ to_html_characters(handle_inline(item))
end
end
diff --git a/lib/rdoc/markup/to_html_crossref.rb b/lib/rdoc/markup/to_html_crossref.rb
index ad600663bf..fad87a5802 100644
--- a/lib/rdoc/markup/to_html_crossref.rb
+++ b/lib/rdoc/markup/to_html_crossref.rb
@@ -50,8 +50,6 @@ def init_link_notation_regexp_handlings
# will be processed as a tidylink first and will be broken.
crossref_re = @options.hyperlink_all ? ALL_CROSSREF_REGEXP : CROSSREF_REGEXP
@markup.add_regexp_handling crossref_re, :CROSSREF
-
- add_regexp_handling_TIDYLINK
end
##
@@ -80,9 +78,8 @@ def cross_reference(name, text = nil, code = true, rdoc_ref: false)
# example, ToHtml is found, even without the RDoc::Markup:: prefix,
# because we look for it in module Markup first.
- def handle_regexp_CROSSREF(target)
- name = target.text
-
+ def handle_regexp_CROSSREF(name)
+ return convert_string(name) if in_tidylink_label?
return name if @options.autolink_excluded_words&.include?(name)
return name if name =~ /@[\w-]+\.[\w-]/ # labels that look like emails
@@ -101,8 +98,8 @@ def handle_regexp_CROSSREF(target)
# Handles rdoc-ref: scheme links and allows RDoc::Markup::ToHtml to
# handle other schemes.
- def handle_regexp_HYPERLINK(target)
- url = target.text
+ def handle_regexp_HYPERLINK(url)
+ return convert_string(url) if in_tidylink_label?
case url
when /\Ardoc-ref:/
@@ -120,12 +117,14 @@ def handle_regexp_HYPERLINK(target)
# All other contents are handled by
# {the superclass}[rdoc-ref:RDoc::Markup::ToHtml#handle_regexp_RDOCLINK]
- def handle_regexp_RDOCLINK(target)
- url = target.text
-
+ def handle_regexp_RDOCLINK(url)
case url
when /\Ardoc-ref:/
- cross_reference $', rdoc_ref: true
+ if in_tidylink_label?
+ convert_string(url)
+ else
+ cross_reference $', rdoc_ref: true
+ end
else
super
end
@@ -203,73 +202,30 @@ def link(name, text, code = true, rdoc_ref: false)
end
end
- def convert_flow(flow_items, &block)
- res = []
-
- i = 0
- while i < flow_items.size
- item = flow_items[i]
-
- case item
- when RDoc::Markup::AttrChanger
- if !tidy_link_capturing? && (text = convert_tt_crossref(flow_items, i))
- text = block.call(text, res) if block
- append_flow_fragment res, text
- i += 3
- next
- end
+ def handle_TT(code)
+ emit_inline(tt_cross_reference(code) || "#{CGI.escapeHTML code}")
+ end
- off_tags res, item
- on_tags res, item
- i += 1
- when String
- text = convert_string(item)
- text = block.call(text, res) if block
- append_flow_fragment res, text
- i += 1
- when RDoc::Markup::RegexpHandling
- text = convert_regexp_handling(item)
- text = block.call(text, res) if block
- append_flow_fragment res, text
- i += 1
- else
- raise "Unknown flow element: #{item.inspect}"
- end
+ # Applies additional special handling on top of the one defined in ToHtml.
+ # When a tidy link is {Foo}[rdoc-ref:Foo], the label part is surrounded by ')
+ "#{convert_string(label)}"
+ else
+ super
end
-
- res.join('')
end
- private
-
- ##
- # Detects ... spans that contain a single cross-reference candidate.
- # When the candidate occupies the whole span (aside from trailing
- # punctuation), the tt markup is replaced by the resolved cross-reference.
-
- def convert_tt_crossref(flow_items, index)
- opener = flow_items[index]
- return unless tt_tag?(opener.turn_on)
-
- string = flow_items[index + 1]
- closer = flow_items[index + 2]
-
- return unless String === string
- return unless RDoc::Markup::AttrChanger === closer
- return unless tt_tag?(closer.turn_off, true)
+ def tt_cross_reference(code)
+ return if in_tidylink_label?
crossref_regexp = @options.hyperlink_all ? ALL_CROSSREF_REGEXP : CROSSREF_REGEXP
- match = crossref_regexp.match(string)
- return unless match
- return unless match.begin(1).zero?
-
- trailing = match.post_match
- # Only convert when the remainder is punctuation/whitespace so other tt text stays literal.
- return unless trailing.match?(/\A[[:punct:]\s]*\z/)
-
- text = cross_reference(string)
- return if text == string
+ match = crossref_regexp.match(code)
+ return unless match && match.begin(1).zero?
+ return unless match.post_match.match?(/\A[[:punct:]\s]*\z/)
- text
+ ref = cross_reference(code)
+ ref if ref != code
end
end
diff --git a/lib/rdoc/markup/to_html_snippet.rb b/lib/rdoc/markup/to_html_snippet.rb
index a671cf0f1d..52cc4543f3 100644
--- a/lib/rdoc/markup/to_html_snippet.rb
+++ b/lib/rdoc/markup/to_html_snippet.rb
@@ -107,8 +107,8 @@ def accept_list_start(list)
def accept_verbatim(verbatim)
throw :done if @characters >= @character_limit
input = verbatim.text.rstrip
-
- text = truncate input
+ text = truncate(input, @character_limit - @characters)
+ @characters += input.length
text << ' ...' unless text == input
super RDoc::Markup::Verbatim.new text
@@ -128,16 +128,8 @@ def start_accepting
##
# Removes escaping from the cross-references in +target+
- def handle_regexp_CROSSREF(target)
- target.text.sub(/\A\\/, '')
- end
-
- ##
- # +target+ is a
'
+ def handle_regexp_CROSSREF(text)
+ text.sub(/\A\\/, '')
end
##
@@ -212,74 +204,82 @@ def convert(content)
end_accepting
end
- ##
- # Converts flow items +flow+
+ def handle_PLAIN_TEXT(text) # :nodoc:
+ return if inline_limit_reached?
- def convert_flow(flow)
- throw :done if @characters >= @character_limit
+ truncated = truncate(text, @inline_character_limit)
+ @inline_character_limit -= text.size
+ emit_inline(convert_string(truncated))
+ end
- res = []
- @mask = 0
-
- flow.each do |item|
- case item
- when RDoc::Markup::AttrChanger then
- off_tags res, item
- on_tags res, item
- when String then
- text = convert_string item
- res << truncate(text)
- when RDoc::Markup::RegexpHandling then
- text = convert_regexp_handling item
- res << truncate(text)
- else
- raise "Unknown flow element: #{item.inspect}"
- end
-
- if @characters >= @character_limit then
- off_tags res, RDoc::Markup::AttrChanger.new(0, @mask)
- break
- end
- end
+ def handle_REGEXP_HANDLING_TEXT(text) # :nodoc:
+ return if inline_limit_reached?
- res << ' ...' if @characters >= @character_limit
+ # We can't truncate text including html tags.
+ # Just emit as is, and count all characters including html tag part.
+ emit_inline(text)
+ @inline_character_limit -= text.size
+ end
- res.join
+ def handle_BOLD(nodes)
+ super unless inline_limit_reached?
end
- ##
- # Maintains a bitmask to allow HTML elements to be closed properly. See
- # RDoc::Markup::Formatter.
+ def handle_BOLD_WORD(word)
+ super unless inline_limit_reached?
+ end
- def on_tags(res, item)
- @mask ^= item.turn_on
+ def handle_EM(nodes)
+ super unless inline_limit_reached?
+ end
- super
+ def handle_EM_WORD(word)
+ super unless inline_limit_reached?
end
- ##
- # Maintains a bitmask to allow HTML elements to be closed properly. See
- # RDoc::Markup::Formatter.
+ def handle_TT(code)
+ super unless inline_limit_reached?
+ end
- def off_tags(res, item)
- @mask ^= item.turn_off
+ def handle_STRIKE(nodes)
+ super unless inline_limit_reached?
+ end
- super
+ def handle_HARD_BREAK
+ super unless inline_limit_reached?
end
- ##
- # Truncates +text+ at the end of the first word after the character_limit.
+ def handle_TIDYLINK(label_part, url)
+ traverse_inline_nodes(label_part) unless inline_limit_reached?
+ end
- def truncate(text)
- length = text.length
- characters = @characters
- @characters += length
+ def inline_limit_reached?
+ @inline_character_limit <= 0
+ end
- return text if @characters < @character_limit
+ def handle_inline(text)
+ limit = @character_limit - @characters
+ return ['', 0] if limit <= 0
+ @inline_character_limit = limit
+ res = super
+ res << ' ...' if @inline_character_limit <= 0
+ @characters += limit - @inline_character_limit
+ res
+ end
+
+ def to_html(item)
+ throw :done if @characters >= @character_limit
+ to_html_characters(handle_inline(item))
+ end
+
+ ##
+ # Truncates +text+ at the end of the first word after the limit.
- remaining = @character_limit - characters
+ def truncate(text, limit)
+ return text if limit >= text.size
+ return '' if limit <= 0
- text =~ /\A(.{#{remaining},}?)(\s|$)/m # TODO word-break instead of \s?
+ text =~ /\A(.{#{limit},}?)(\s|$)/m # TODO word-break instead of \s?
$1
end
diff --git a/lib/rdoc/markup/to_label.rb b/lib/rdoc/markup/to_label.rb
index 22ec07f9c1..4ab030bb70 100644
--- a/lib/rdoc/markup/to_label.rb
+++ b/lib/rdoc/markup/to_label.rb
@@ -18,20 +18,33 @@ def initialize(markup = nil)
super nil, markup
@markup.add_regexp_handling RDoc::CrossReference::CROSSREF_REGEXP, :CROSSREF
- @markup.add_regexp_handling(/(((\{.*?\})|\b\S+?)\[\S+?\])/, :TIDYLINK)
- add_tag :BOLD, '', ''
- add_tag :TT, '', ''
- add_tag :EM, '', ''
+ @res = []
+ end
+
+ def handle_PLAIN_TEXT(text)
+ @res << text
+ end
+
+ def handle_REGEXP_HANDLING_TEXT(text)
+ @res << text
+ end
+
+ def handle_TT(text)
+ @res << text
+ end
+ def extract_plaintext(text)
@res = []
+ handle_inline(text)
+ @res.join
end
##
# Converts +text+ to an HTML-safe label using GitHub-style anchor formatting.
def convert(text)
- label = convert_flow @am.flow text
+ label = extract_plaintext(text)
RDoc::Text.to_anchor(label)
end
@@ -41,7 +54,7 @@ def convert(text)
# Used for generating backward-compatible anchor aliases.
def convert_legacy(text)
- label = convert_flow @am.flow text
+ label = extract_plaintext(text)
CGI.escape(label).gsub('%', '-').sub(/^-/, '')
end
@@ -50,23 +63,10 @@ def convert_legacy(text)
# Converts the CROSSREF +target+ to plain text, removing the suppression
# marker, if any
- def handle_regexp_CROSSREF(target)
- text = target.text
-
+ def handle_regexp_CROSSREF(text)
text.sub(/^\\/, '')
end
- ##
- # Converts the TIDYLINK +target+ to just the text part
-
- def handle_regexp_TIDYLINK(target)
- text = target.text
-
- return text unless text =~ /\{(.*?)\}\[(.*?)\]/ or text =~ /(\S+)\[(.*?)\]/
-
- $1
- end
-
alias accept_blank_line ignore
alias accept_block_quote ignore
alias accept_heading ignore
@@ -79,7 +79,6 @@ def handle_regexp_TIDYLINK(target)
alias accept_rule ignore
alias accept_verbatim ignore
alias end_accepting ignore
- alias handle_regexp_HARD_BREAK ignore
alias start_accepting ignore
end
diff --git a/lib/rdoc/markup/to_markdown.rb b/lib/rdoc/markup/to_markdown.rb
index ff3d1e6da8..4c19e460c4 100644
--- a/lib/rdoc/markup/to_markdown.rb
+++ b/lib/rdoc/markup/to_markdown.rb
@@ -20,27 +20,10 @@ def initialize(markup = nil)
@headings[6] = ['###### ', '']
add_regexp_handling_RDOCLINK
- add_regexp_handling_TIDYLINK
@hard_break = " \n"
end
- ##
- # Maps attributes to HTML sequences
-
- def init_tags
- add_tag :BOLD, '**', '**'
- add_tag :EM, '*', '*'
- add_tag :TT, '`', '`'
- end
-
- ##
- # Adds a newline to the output
-
- def handle_regexp_HARD_BREAK(target)
- " \n"
- end
-
##
# Finishes consumption of `list`
@@ -94,6 +77,65 @@ def accept_list_item_start(list_item)
end
end
+ def add_tag(tag, simple_tag, content)
+ if content.match?(/\A[\w\s]+\z/)
+ emit_inline("#{simple_tag}#{content}#{simple_tag}")
+ else
+ emit_inline("<#{tag}>#{content}")
+ end
+ end
+
+ def handle_tag(nodes, simple_tag, tag)
+ if nodes.size == 1 && String === nodes[0]
+ content = apply_regexp_handling(nodes[0]).map do |text, converted|
+ converted ? text : convert_string(text)
+ end.join
+ add_tag(tag, simple_tag, content)
+ else
+ emit_inline("<#{tag}>")
+ traverse_inline_nodes(nodes)
+ emit_inline("")
+ end
+ end
+
+ def handle_TIDYLINK(label_part, url)
+ if url =~ /^rdoc-label:foot/ then
+ emit_inline(handle_rdoc_link(url))
+ else
+ emit_inline('[')
+ traverse_inline_nodes(label_part)
+ emit_inline("](#{url})")
+ end
+ end
+
+ def handle_BOLD(nodes)
+ handle_tag(nodes, '**', 'strong')
+ end
+
+ def handle_EM(nodes)
+ handle_tag(nodes, '*', 'em')
+ end
+
+ def handle_BOLD_WORD(word)
+ add_tag('strong', '**', convert_string(word))
+ end
+
+ def handle_EM_WORD(word)
+ add_tag('em', '*', convert_string(word))
+ end
+
+ def handle_TT(text)
+ add_tag('code', '`', convert_string(text))
+ end
+
+ def handle_STRIKE(nodes)
+ handle_tag(nodes, '~~', 's')
+ end
+
+ def handle_HARD_BREAK
+ emit_inline(" \n")
+ end
+
##
# Prepares the visitor for consuming `list`
@@ -163,29 +205,11 @@ def handle_rdoc_link(url)
end
end
- ##
- # Converts the RDoc markup tidylink into a Markdown.style link.
-
- def handle_regexp_TIDYLINK(target)
- text = target.text
-
- return text unless text =~ /\{(.*?)\}\[(.*?)\]/ or text =~ /(\S+)\[(.*?)\]/
-
- label = $1
- url = $2
-
- if url =~ /^rdoc-label:foot/ then
- handle_rdoc_link url
- else
- gen_url url, label
- end
- end
-
##
# Converts the rdoc-...: links into a Markdown.style links.
- def handle_regexp_RDOCLINK(target)
- handle_rdoc_link target.text
+ def handle_regexp_RDOCLINK(text)
+ handle_rdoc_link text
end
end
diff --git a/lib/rdoc/markup/to_rdoc.rb b/lib/rdoc/markup/to_rdoc.rb
index 4185e8bea9..5eab835980 100644
--- a/lib/rdoc/markup/to_rdoc.rb
+++ b/lib/rdoc/markup/to_rdoc.rb
@@ -57,21 +57,11 @@ def initialize(markup = nil)
@markup.add_regexp_handling(/\\\S/, :SUPPRESSED_CROSSREF)
@width = 78
- init_tags
@headings = DEFAULT_HEADINGS.dup
@hard_break = "\n"
end
- ##
- # Maps attributes to HTML sequences
-
- def init_tags
- add_tag :BOLD, "", ""
- add_tag :TT, "", ""
- add_tag :EM, "", ""
- end
-
##
# Adds +blank_line+ to the output
@@ -282,12 +272,90 @@ def calculate_text_width(text)
text.size
end
+ def handle_PLAIN_TEXT(text)
+ add_text(text)
+ end
+
+ def handle_REGEXP_HANDLING_TEXT(text)
+ add_text(text)
+ end
+
+ def handle_BOLD(target)
+ on(:BOLD)
+ super
+ off(:BOLD)
+ end
+
+ def handle_EM(target)
+ on(:EM)
+ super
+ off(:EM)
+ end
+
+ def handle_BOLD_WORD(word)
+ on(:BOLD)
+ super
+ off(:BOLD)
+ end
+
+ def handle_EM_WORD(word)
+ on(:EM)
+ super
+ off(:EM)
+ end
+
+ def handle_TT(code)
+ on(:TT)
+ super
+ off(:TT)
+ end
+
+ def handle_STRIKE(target)
+ on(:STRIKE)
+ super
+ off(:STRIKE)
+ end
+
+ def handle_HARD_BREAK
+ add_text("\n")
+ end
+
+ def handle_TIDYLINK(label_part, url)
+ super
+ add_text("( #{url} )")
+ end
+
+ def handle_inline(text, initial_attributes = [])
+ @attributes = Hash.new(0)
+ initial_attributes.each { |attr| on(attr) }
+ out = @inline_output = +''
+ super(text)
+ @inline_output = nil
+ out
+ end
+
+ def on(attr)
+ @attributes[attr] += 1
+ end
+
+ def off(attr)
+ @attributes[attr] -= 1
+ @attributes.delete(attr) if @attributes[attr] == 0
+ end
+
+ def add_text(text)
+ emit_inline(text)
+ end
+
+ def emit_inline(text)
+ @inline_output << text
+ end
+
##
- # Applies attribute-specific markup to +text+ using RDoc::AttributeManager
+ # Applies attribute-specific markup to +text+ using RDoc::Markup::InlineParser
def attributes(text)
- flow = @am.flow text.dup
- convert_flow flow
+ handle_inline(text)
end
##
@@ -300,17 +368,8 @@ def end_accepting
##
# Removes preceding \\ from the suppressed crossref +target+
- def handle_regexp_SUPPRESSED_CROSSREF(target)
- text = target.text
- text = text.sub('\\', '') unless in_tt?
- text
- end
-
- ##
- # Adds a newline to the output
-
- def handle_regexp_HARD_BREAK(target)
- "\n"
+ def handle_regexp_SUPPRESSED_CROSSREF(text)
+ text.sub('\\', '')
end
##
diff --git a/lib/rdoc/markup/to_test.rb b/lib/rdoc/markup/to_test.rb
index 3cf20f384b..f2b7febe95 100644
--- a/lib/rdoc/markup/to_test.rb
+++ b/lib/rdoc/markup/to_test.rb
@@ -18,8 +18,16 @@ def end_accepting
@res
end
+ def handle_PLAIN_TEXT(text)
+ @res << text
+ end
+
+ def handle_REGEXP_HANDLING_TEXT(text)
+ @res << text
+ end
+
def accept_paragraph(paragraph)
- @res << convert_flow(@am.flow(paragraph.text))
+ handle_inline(paragraph.text)
end
def accept_raw(raw)
diff --git a/lib/rdoc/markup/to_tt_only.rb b/lib/rdoc/markup/to_tt_only.rb
index 7b59473aaf..19e3622f75 100644
--- a/lib/rdoc/markup/to_tt_only.rb
+++ b/lib/rdoc/markup/to_tt_only.rb
@@ -20,8 +20,6 @@ class RDoc::Markup::ToTtOnly < RDoc::Markup::Formatter
def initialize(markup = nil)
super nil, markup
-
- add_tag :TT, nil, nil
end
##
@@ -82,22 +80,18 @@ def do_nothing(markup_item)
# Extracts tt sections from +text+
def tt_sections(text)
- flow = @am.flow text.dup
-
- flow.each do |item|
- case item
- when String then
- @res << item if in_tt?
- when RDoc::Markup::AttrChanger then
- off_tags res, item
- on_tags res, item
- when RDoc::Markup::RegexpHandling then
- @res << convert_regexp_handling(item) if in_tt? # TODO can this happen?
+ parsed = RDoc::Markup::InlineParser.new(text).parse
+ traverse = -> node {
+ next if String === node
+ if node[:type] == :TT
+ res << nil
+ res << node[:children][0] || ''
+ res << nil
else
- raise "Unknown flow element: #{item.inspect}"
+ node[:children].each(&traverse)
end
- end
-
+ }
+ parsed.each(&traverse)
res
end
diff --git a/lib/rdoc/text.rb b/lib/rdoc/text.rb
index 7f715ae861..94c84037c8 100644
--- a/lib/rdoc/text.rb
+++ b/lib/rdoc/text.rb
@@ -193,11 +193,15 @@ def strip_stars(text)
text.gsub(/^\s+$/, empty)
end
+ def to_html(text)
+ to_html_characters(text)
+ end
+
##
# Converts ampersand, dashes, ellipsis, quotes, copyright and registered
# trademark symbols in +text+ to properly encoded characters.
- def to_html(text)
+ def to_html_characters(text)
html = (''.encode text.encoding).dup
encoded = RDoc::Text::TO_HTML_CHARACTERS[text.encoding]
@@ -210,15 +214,12 @@ def to_html(text)
until s.eos? do
case
when s.scan(/<(tt|code)>.*?<\/\1>/) then # skip contents of tt
- html << s.matched.gsub('\\\\', '\\')
+ html << s.matched
when s.scan(/<(tt|code)>.*?/) then
warn "mismatched <#{s[1]}> tag" # TODO signal file/line
html << s.matched
when s.scan(/<[^>]+\/?s*>/) then # skip HTML tags
html << s.matched
- when s.scan(/\\(\S)/) then # unhandled suppressed crossref
- html << s[1]
- after_word = nil
when s.scan(/\.\.\.(\.?)/) then
html << s[1] << encoded[:ellipsis]
after_word = nil
diff --git a/test/rdoc/markup/attribute_manager_test.rb b/test/rdoc/markup/attribute_manager_test.rb
deleted file mode 100644
index 903966b8e2..0000000000
--- a/test/rdoc/markup/attribute_manager_test.rb
+++ /dev/null
@@ -1,474 +0,0 @@
-# frozen_string_literal: true
-require_relative '../helper'
-
-class RDocMarkupAttributeManagerTest < RDoc::TestCase
-
- def setup
- super
-
- @am = RDoc::Markup::AttributeManager.new
-
- @bold_on = @am.changed_attribute_by_name([], [:BOLD])
- @bold_off = @am.changed_attribute_by_name([:BOLD], [])
-
- @tt_on = @am.changed_attribute_by_name([], [:TT])
- @tt_off = @am.changed_attribute_by_name([:TT], [])
-
- @em_on = @am.changed_attribute_by_name([], [:EM])
- @em_off = @am.changed_attribute_by_name([:EM], [])
-
- @strike_on = @am.changed_attribute_by_name([], [:STRIKE])
- @strike_off = @am.changed_attribute_by_name([:STRIKE], [])
-
- @bold_em_on = @am.changed_attribute_by_name([], [:BOLD] | [:EM])
- @bold_em_off = @am.changed_attribute_by_name([:BOLD] | [:EM], [])
-
- @em_then_bold = @am.changed_attribute_by_name([:EM], [:EM] | [:BOLD])
-
- @em_to_bold = @am.changed_attribute_by_name([:EM], [:BOLD])
-
- @am.add_word_pair("{", "}", :WOMBAT)
- @wombat_on = @am.changed_attribute_by_name([], [:WOMBAT])
- @wombat_off = @am.changed_attribute_by_name([:WOMBAT], [])
-
- @klass = RDoc::Markup::AttributeManager
- @formatter = RDoc::Markup::Formatter.new @rdoc.options
- @formatter.add_tag :BOLD, '', ''
- @formatter.add_tag :EM, '', ''
- @formatter.add_tag :TT, '', ''
- end
-
- def crossref(text)
- crossref_bitmap = @am.attributes.bitmap_for(:_REGEXP_HANDLING_) |
- @am.attributes.bitmap_for(:CROSSREF)
-
- [ @am.changed_attribute_by_name([], [:CROSSREF, :_REGEXP_HANDLING_]),
- RDoc::Markup::RegexpHandling.new(crossref_bitmap, text),
- @am.changed_attribute_by_name([:CROSSREF, :_REGEXP_HANDLING_], [])
- ]
- end
-
- def test_adding
- assert_equal(["cat ", @wombat_on, "and", @wombat_off, " dog" ],
- @am.flow("cat {and} dog"))
- #assert_equal(["cat {and} dog" ], @am.flow("cat \\{and} dog"))
- end
-
- def test_add_html_tag
- @am.add_html("Test", :TEST)
- tags = @am.html_tags
- assert_equal(8, tags.size)
- assert(tags.has_key?("test"))
- end
-
- def test_add_regexp_handling
- @am.add_regexp_handling "WikiWord", :WIKIWORD
- regexp_handlings = @am.regexp_handlings
-
- assert_equal 1, regexp_handlings.size
- assert regexp_handlings.assoc "WikiWord"
- end
-
- def test_add_word_pair
- @am.add_word_pair '%', '&', 'percent and'
-
- assert @am.word_pair_map.include?(/(%)(\S+)(&)/)
- assert @am.protectable.include?('%')
- assert !@am.protectable.include?('&')
- end
-
- def test_add_word_pair_angle
- e = assert_raise ArgumentError do
- @am.add_word_pair '<', '>', 'angles'
- end
-
- assert_equal "Word flags may not start with '<'", e.message
- end
-
- def test_add_word_pair_invalid
- assert_raise ArgumentError do
- @am.add_word_pair("<", "<", :TEST)
- end
- end
-
- def test_add_word_pair_map
- @am.add_word_pair("x", "y", :TEST)
-
- word_pair_map = @am.word_pair_map
-
- assert_includes word_pair_map.keys.map { |r| r.source }, "(x)(\\S+)(y)"
- end
-
- def test_add_word_pair_matching
- @am.add_word_pair '^', '^', 'caret'
-
- assert @am.matching_word_pairs.include?('^')
- assert @am.protectable.include?('^')
- end
-
- def test_basic
- assert_equal(["cat"], @am.flow("cat"))
-
- assert_equal(["cat ", @bold_on, "and", @bold_off, " dog"],
- @am.flow("cat *and* dog"))
-
- assert_equal(["cat ", @bold_on, "AND", @bold_off, " dog"],
- @am.flow("cat *AND* dog"))
-
- assert_equal(["cat ", @em_on, "And", @em_off, " dog"],
- @am.flow("cat _And_ dog"))
-
- assert_equal(["cat *and dog*"], @am.flow("cat *and dog*"))
-
- assert_equal(["*cat and* dog"], @am.flow("*cat and* dog"))
-
- assert_equal(["cat *and ", @bold_on, "dog", @bold_off],
- @am.flow("cat *and *dog*"))
-
- assert_equal(["cat ", @em_on, "and", @em_off, " dog"],
- @am.flow("cat _and_ dog"))
-
- assert_equal(["cat_and_dog"],
- @am.flow("cat_and_dog"))
-
- assert_equal(["cat ", @tt_on, "and", @tt_off, " dog"],
- @am.flow("cat +and+ dog"))
-
- assert_equal(["cat ", @tt_on, "X::Y", @tt_off, " dog"],
- @am.flow("cat +X::Y+ dog"))
-
- assert_equal(["cat ", @bold_on, "a_b_c", @bold_off, " dog"],
- @am.flow("cat *a_b_c* dog"))
-
- assert_equal(["cat __ dog"],
- @am.flow("cat __ dog"))
-
- assert_equal(["cat ", @em_on, "_", @em_off, " dog"],
- @am.flow("cat ___ dog"))
-
- assert_equal(["cat and ", @em_on, "5", @em_off, " dogs"],
- @am.flow("cat and _5_ dogs"))
-
- assert_equal([@tt_on, "__id__", @tt_off], @am.flow("+__id__+"))
- end
-
- def test_bold
- assert_equal [@bold_on, 'bold', @bold_off],
- @am.flow("*bold*")
-
- assert_equal [@bold_on, 'Bold:', @bold_off],
- @am.flow("*Bold:*")
-
- assert_equal [@bold_on, '\\bold', @bold_off],
- @am.flow("*\\bold*")
- end
-
- def test_bold_html_escaped
- assert_equal ['cat dog'], @am.flow('cat \dog')
- end
-
- def test_strike_html_escaped
- assert_equal ['cat dog'], @am.flow('cat \dog')
- assert_equal ['cat dog'], @am.flow('cat \dog')
- end
-
- def test_html_like_strike
- assert_equal ["cat ", @strike_on, "dog", @strike_off],
- @am.flow("cat dog")
- end
-
- def test_html_like_strike_del
- assert_equal ["cat ", @strike_on, "dog", @strike_off],
- @am.flow("cat dog")
- end
-
- def test_combined
- assert_equal(["cat ", @em_on, "and", @em_off, " ", @bold_on, "dog", @bold_off],
- @am.flow("cat _and_ *dog*"))
-
- assert_equal(["cat ", @em_on, "a__nd", @em_off, " ", @bold_on, "dog", @bold_off],
- @am.flow("cat _a__nd_ *dog*"))
- end
-
- def test_convert_attrs
- str = '+foo+'.dup
- attrs = RDoc::Markup::AttrSpan.new str.length, @am.exclusive_bitmap
-
- @am.convert_attrs str, attrs, true
- @am.convert_attrs str, attrs
-
- assert_equal "\000foo\000", str
-
- str = '+:foo:+'.dup
- attrs = RDoc::Markup::AttrSpan.new str.length, @am.exclusive_bitmap
-
- @am.convert_attrs str, attrs, true
- @am.convert_attrs str, attrs
-
- assert_equal "\000:foo:\000", str
-
- str = '+x-y+'.dup
- attrs = RDoc::Markup::AttrSpan.new str.length, @am.exclusive_bitmap
-
- @am.convert_attrs str, attrs, true
- @am.convert_attrs str, attrs
-
- assert_equal "\000x-y\000", str
- end
-
- def test_convert_attrs_ignores_code
- assert_equal 'foo __send__ bar', output('foo __send__ bar')
- end
-
- def test_convert_attrs_ignores_bold_inside_code
- assert_equal 'foo *bold* bar', output('foo *bold* bar')
- end
-
- def test_convert_attrs_ignores_em_inside_code
- assert_equal 'foo _em_ bar', output('foo _em_ bar')
- end
-
- def test_convert_attrs_ignores_tt_inside_code
- assert_equal 'foo +tt+ bar', output('foo +tt+ bar')
- end
-
- def test_convert_attrs_ignores_bold_inside_tt
- assert_equal 'foo *bold* bar', output('foo *bold* bar')
- end
-
- def test_convert_attrs_ignores_em_inside_tt
- assert_equal 'foo _em_ bar', output('foo _em_ bar')
- end
-
- def test_convert_attrs_ignores_tt_inside_tt
- assert_equal 'foo +tt+ bar', output('foo +tt+ bar')
- end
-
- def test_backtick_basic
- assert_equal(["cat ", @tt_on, "and", @tt_off, " dog"],
- @am.flow("cat `and` dog"))
-
- assert_equal(["cat ", @tt_on, "X::Y", @tt_off, " dog"],
- @am.flow("cat `X::Y` dog"))
- end
-
- def test_backtick_output
- assert_equal 'cat and dog', output('cat `and` dog')
- assert_equal 'cat X::Y dog', output('cat `X::Y` dog')
- end
-
- def test_convert_attrs_ignores_backtick_inside_code
- assert_equal 'foo `text` bar', output('foo `text` bar')
- end
-
- def test_convert_attrs_ignores_backtick_inside_tt
- assert_equal 'foo `text` bar', output('foo `text` bar')
- end
-
- def test_backtick_escaped
- assert_equal ['`text`'], @am.flow('\`text`')
- end
-
- def test_convert_attrs_ignores_del_inside_code
- assert_equal 'foo strikestrikestrikestrike bar')
- end
-
- def test_convert_attrs_ignores_s_inside_code
- assert_equal 'foo strikestrike__send__ bar', output('foo __send__ bar')
- end
-
- def test_convert_attrs_preserves_double
- assert_equal 'foo.__send__ :bar', output('foo.__send__ :bar')
- assert_equal 'use __FILE__ to', output('use __FILE__ to')
- end
-
- def test_convert_attrs_does_not_ignore_after_tt
- assert_equal 'the IF:key directive', output('the IF:_key_ directive')
- end
-
- def test_escapes
- assert_equal 'text', output('text')
- assert_equal 'text', output('\\text')
- assert_equal '', output('\\')
- assert_equal '\\', output('\\\\')
- assert_equal 'text', output('*text*')
- assert_equal '*text*', output('\\*text*')
- assert_equal '\\', output('\\')
- assert_equal '\\text', output('\\text')
- assert_equal '\\\\text', output('\\\\text')
- assert_equal 'text \\ text', output('text \\ text')
-
- assert_equal 'and \\s matches space',
- output('and \\s matches space')
- assert_equal 'use text for code',
- output('use \\text for code')
- assert_equal 'use text for code',
- output('use \\text\\ for code')
- assert_equal 'use text for code',
- output('use \\\\text for code')
- assert_equal 'use text for code',
- output('use \\text for code')
- assert_equal 'use +text+ for code',
- output('use \\+text+ for code')
- assert_equal 'use text for code',
- output('use \\+text+ for code')
- assert_equal 'illegal not changed',
- output('illegal not changed')
- assert_equal 'unhandled tag
unchanged',
- output('unhandled tag
unchanged')
- end
-
- def test_exclude_tag
- assert_equal 'aaa[:symbol]', output('+aaa+[:symbol]')
- assert_equal 'aaa[:symbol]', output('+aaa[:symbol]+')
- assert_equal 'aaa[:symbol]', output('aaa[:symbol]')
- assert_equal 'index', output('index')
- end
-
- def test_exclude_tag_flow
- assert_equal [@tt_on, "aaa", @tt_off, "[:symbol]"],
- @am.flow("+aaa+[:symbol]")
- assert_equal [@tt_on, "aaa[:symbol]", @tt_off],
- @am.flow("+aaa[:symbol]+")
- assert_equal ["aaa[:symbol]"],
- @am.flow("aaa[:symbol]")
- end
-
- def test_html_like_em_bold
- assert_equal ["cat ", @em_on, "and ", @em_to_bold, "dog", @bold_off],
- @am.flow("cat and dog")
- end
-
- def test_html_like_em_bold_SGML
- assert_equal ["cat ", @em_on, "and ", @em_to_bold, "dog", @bold_off],
- @am.flow("cat and dog")
- end
-
- def test_html_like_em_bold_nested_1
- assert_equal(["cat ", @bold_em_on, "and", @bold_em_off, " dog"],
- @am.flow("cat and dog"))
- end
-
- def test_html_like_em_bold_nested_2
- assert_equal ["cat ", @em_on, "and ", @em_then_bold, "dog", @bold_em_off],
- @am.flow("cat and dog")
- end
-
- def test_html_like_em_bold_nested_mixed_case
- assert_equal ["cat ", @em_on, "and ", @em_then_bold, "dog", @bold_em_off],
- @am.flow("cat and dog")
- end
-
- def test_html_like_em_bold_mixed_case
- assert_equal ["cat ", @em_on, "and", @em_off, " ", @bold_on, "dog", @bold_off],
- @am.flow("cat and dog")
- end
-
- def test_html_like_teletype
- assert_equal ["cat ", @tt_on, "dog", @tt_off],
- @am.flow("cat dog")
- end
-
- def test_html_like_teletype_em_bold_SGML
- assert_equal [@tt_on, "cat", @tt_off, " ", @em_on, "and ", @em_to_bold, "dog", @bold_off],
- @am.flow("cat and dog")
- end
-
- def test_initial_html
- html_tags = @am.html_tags
- assert html_tags.is_a?(Hash)
- assert_equal(7, html_tags.size)
- end
-
- def test_initial_word_pairs
- word_pairs = @am.matching_word_pairs
- assert word_pairs.is_a?(Hash)
- assert_equal(4, word_pairs.size)
- end
-
- def test_mask_protected_sequence
- def @am.str() @str end
- def @am.str=(str) @str = str end
-
- @am.str = 'foo'.dup
- @am.mask_protected_sequences
-
- assert_equal "foo", @am.str
-
- @am.str = 'foo\\'.dup
- @am.mask_protected_sequences
-
- assert_equal "foo<\x04/code>", @am.str, 'escaped close'
-
- @am.str = 'foo\\\\'.dup
- @am.mask_protected_sequences
-
- assert_equal "foo\\", @am.str, 'escaped backslash'
- end
-
- def test_protect
- assert_equal(['cat \\ dog'],
- @am.flow('cat \\ dog'))
-
- assert_equal(["cat dog"],
- @am.flow("cat \\dog"))
-
- assert_equal(["cat ", @em_on, "and", @em_off, " dog"],
- @am.flow("cat and \\dog"))
-
- assert_equal(["*word* or text"],
- @am.flow("\\*word* or \\text"))
-
- assert_equal(["_cat_", @em_on, "dog", @em_off],
- @am.flow("\\_cat_dog"))
- end
-
- def test_lost_tag_for_the_second_time
- str = "cat dog"
- assert_equal(["cat ", @tt_on, "dog", @tt_off],
- @am.flow(str))
- assert_equal(["cat ", @tt_on, "dog", @tt_off],
- @am.flow(str))
- end
-
- def test_regexp_handling
- @am.add_regexp_handling(RDoc::CrossReference::CROSSREF_REGEXP, :CROSSREF)
-
- #
- # The apostrophes in "cats'" and "dogs'" suppress the flagging of these
- # words as potential cross-references, which is necessary for the unit
- # tests. Unfortunately, the markup engine right now does not actually
- # check whether a cross-reference is valid before flagging it.
- #
- assert_equal(["cats'"], @am.flow("cats'"))
-
- assert_equal(["cats' ", crossref("#fred"), " dogs'"].flatten,
- @am.flow("cats' #fred dogs'"))
-
- assert_equal([crossref("#fred"), " dogs'"].flatten,
- @am.flow("#fred dogs'"))
-
- assert_equal(["cats' ", crossref("#fred")].flatten, @am.flow("cats' #fred"))
-
- assert_equal(["(", crossref("#fred"), ")"].flatten, @am.flow("(#fred)"))
- end
-
- def test_tt_html
- assert_equal [@tt_on, '"\n"', @tt_off],
- @am.flow('"\n"')
- end
-
- def output(str)
- @formatter.convert_flow @am.flow str
- end
-
-end
diff --git a/test/rdoc/markup/attributes_test.rb b/test/rdoc/markup/attributes_test.rb
deleted file mode 100644
index b46b7a0bf9..0000000000
--- a/test/rdoc/markup/attributes_test.rb
+++ /dev/null
@@ -1,39 +0,0 @@
-# frozen_string_literal: true
-require_relative '../helper'
-
-class RDocMarkupAttributesTest < RDoc::TestCase
-
- def setup
- super
-
- @as = RDoc::Markup::Attributes.new
- end
-
- def test_bitmap_for
- assert_equal 2, @as.bitmap_for('two')
- assert_equal 2, @as.bitmap_for('two')
- assert_equal 4, @as.bitmap_for('three')
- end
-
- def test_as_string
- @as.bitmap_for 'two'
- @as.bitmap_for 'three'
-
- assert_equal 'none', @as.as_string(0)
- assert_equal '_REGEXP_HANDLING_', @as.as_string(1)
- assert_equal 'two', @as.as_string(2)
- assert_equal '_REGEXP_HANDLING_,two', @as.as_string(3)
- end
-
- def test_each_name_of
- @as.bitmap_for 'two'
- @as.bitmap_for 'three'
-
- assert_equal %w[], @as.each_name_of(0).to_a
- assert_equal %w[], @as.each_name_of(1).to_a
- assert_equal %w[two], @as.each_name_of(2).to_a
- assert_equal %w[three], @as.each_name_of(4).to_a
- assert_equal %w[two three], @as.each_name_of(6).to_a
- end
-
-end
diff --git a/test/rdoc/markup/formatter_test.rb b/test/rdoc/markup/formatter_test.rb
index e17fa7725d..2ff7323622 100644
--- a/test/rdoc/markup/formatter_test.rb
+++ b/test/rdoc/markup/formatter_test.rb
@@ -7,20 +7,38 @@ class ToTest < RDoc::Markup::Formatter
def initialize(markup)
super nil, markup
-
- add_tag :TT, '', ''
end
def accept_paragraph(paragraph)
@res += attributes(paragraph.text)
end
+ def handle_PLAIN_TEXT(text)
+ @res << text
+ end
+
+ def handle_REGEXP_HANDLING_TEXT(text)
+ @res << text
+ end
+
+ def handle_TT(text)
+ @res << "#{text}"
+ end
+
+ def handle_TIDYLINK(label_part, url)
+ @res << '{'
+ super
+ @res << '}[' + url + ']'
+ end
+
def attributes(text)
- convert_flow @am.flow text.dup
+ @res = +""
+ handle_inline(text)
+ @res
end
- def handle_regexp_CAPS(target)
- "handled #{target.text}"
+ def handle_regexp_CAPS(text)
+ "handled #{text}"
end
def start_accepting
@@ -39,14 +57,7 @@ def setup
@markup = @RM.new
@markup.add_regexp_handling(/[A-Z]+/, :CAPS)
- @attribute_manager = @markup.attribute_manager
- @attributes = @attribute_manager.attributes
-
@to = ToTest.new @markup
-
- @caps = @attributes.bitmap_for :CAPS
- @regexp_handling = @attributes.bitmap_for :_REGEXP_HANDLING_
- @tt = @attributes.bitmap_for :TT
end
def test_class_gen_relative_url
@@ -63,9 +74,7 @@ def gen(from, to)
end
def regexp_handling_names
- @attribute_manager.regexp_handlings.map do |_, mask|
- @attributes.as_string mask
- end
+ @to.instance_variable_get(:@markup).regexp_handlings.map(&:last).map(&:to_s)
end
def test_add_regexp_handling_RDOCLINK
@@ -73,43 +82,15 @@ def test_add_regexp_handling_RDOCLINK
assert_includes regexp_handling_names, 'RDOCLINK'
- def @to.handle_regexp_RDOCLINK(target)
- "<#{target.text}>"
- end
-
- document = doc(para('{foo}[rdoc-label:bar].'))
-
- formatted = document.accept @to
-
- assert_equal '{foo}[].', formatted
- end
-
- def test_add_regexp_handling_TIDYLINK
- @to.add_regexp_handling_TIDYLINK
-
- assert_includes regexp_handling_names, 'TIDYLINK'
-
- def @to.handle_regexp_TIDYLINK(target)
- "<#{target.text}>"
+ def @to.handle_regexp_RDOCLINK(text)
+ "<#{text}>"
end
- document = doc(para('foo[rdoc-label:bar].'))
-
- formatted = document.accept @to
-
- assert_equal '.', formatted
-
- document = doc(para('{foo}[rdoc-label:bar].'))
-
- formatted = document.accept @to
-
- assert_equal '<{foo}[rdoc-label:bar]>.', formatted
-
- document = doc(para('{abc}: {foo}[rdoc-label:bar].'))
+ document = doc(para('{foo rdoc-label:bar baz}[url]'))
formatted = document.accept @to
- assert_equal '{abc}: <{foo}[rdoc-label:bar]>.', formatted
+ assert_equal '{foo baz}[url]', formatted
end
def test_parse_url
diff --git a/test/rdoc/markup/inline_parser_test.rb b/test/rdoc/markup/inline_parser_test.rb
new file mode 100644
index 0000000000..60f03bedd7
--- /dev/null
+++ b/test/rdoc/markup/inline_parser_test.rb
@@ -0,0 +1,269 @@
+# frozen_string_literal: true
+
+require_relative '../helper'
+require 'rdoc/markup/inline_parser'
+
+class RDocMarkupInlineParserTest < RDoc::TestCase
+ def parse(text)
+ RDoc::Markup::InlineParser.new(text).parse
+ end
+
+ def em_node(*children)
+ { type: :EM, children: children }
+ end
+
+ def bold_node(*children)
+ { type: :BOLD, children: children }
+ end
+
+ def bold_word(text)
+ { type: :BOLD_WORD, children: [text] }
+ end
+
+ def em_word(text)
+ { type: :EM_WORD, children: [text] }
+ end
+
+ def strike_node(*children)
+ { type: :STRIKE, children: children }
+ end
+
+ def tt_node(*children)
+ { type: :TT, children: children }
+ end
+
+ def tidylink_node(children, url)
+ { type: :TIDYLINK, children: children, url: url }
+ end
+
+ def hard_break_node
+ { type: :HARD_BREAK, children: [] }
+ end
+
+ def test_escape
+ # Escaping backslash are removed, other backslashes (suppressed crossref) remains
+ assert_equal(['\\', bold_node('\\Array'), bold_node('\\#to_s'), bold_node('\\::new')], parse('\\\\\\Array\\#to_s\\::new'))
+ assert_equal(['_a_ +a+ b \\n \\ABC'], parse('\\_a_ \\+a+ \\b\\ \\n \\ABC'))
+ assert_equal([bold_node('')], parse('\\'))
+ assert_equal([em_node('')], parse('\\'))
+ assert_equal(['a\\'], parse('a\\'))
+ assert_equal([tidylink_node([''], 'url')], parse('{\\\\}[url]'))
+ # Unescape \\ and \< in code blocks
+ assert_equal([tt_node('p(%(\\)+"\\a\\n")')], parse('p(%(\\\\\\\\)+"\\a\\n")'))
+ end
+
+ def test_bold
+ assert_equal([bold_node()], parse(''))
+ assert_equal(['*a b*'], parse('*a b*'))
+ assert_equal(['x*a* *b*x'], parse('x*a* *b*x'))
+ assert_equal([bold_word('bold')], parse('*bold*'))
+ assert_equal([bold_word('bold')], parse('**bold**'))
+ assert_equal([bold_node('bo ld')], parse('bo ld'))
+ assert_equal(
+ ['a ', bold_word('A'), ' b ', bold_word('B'), ' c ', bold_node('C C'), ' d'],
+ parse('a *A* b **B** c C C d')
+ )
+ assert_equal([bold_node('a', em_node('b'), bold_node('c'), 'd')], parse('abcd'))
+ end
+
+ def test_em
+ assert_equal([em_node()], parse(''))
+ assert_equal(['_a b_'], parse('_a b_'))
+ assert_equal(['x_a_ _b_x'], parse('x_a_ _b_x'))
+ assert_equal([em_word('em')], parse('_em_'))
+ assert_equal([em_word('F1LE')], parse('__F1LE__'))
+ assert_equal(['_foo_bar_baz'], parse('_foo_bar_baz'))
+
+ # _ inside _em_
+ assert_equal([em_word('foo_bar')], parse('_foo_bar_'))
+
+ # non-alphanumeric after _
+ assert_equal([em_word('host'), ':', em_word('port')], parse('_host_:_port_'))
+
+ # Special exception
+ assert_equal(['__send__'], parse('__send__'))
+ assert_equal(['__FILE__'], parse('__FILE__'))
+
+ assert_equal([em_node('e m')], parse('e m'))
+ assert_equal([em_node('e m')], parse('e m'))
+ assert_equal([em_node('a', bold_node('b'), em_node('c'), 'd')], parse('abcd'))
+ end
+
+ def test_method_like_words
+ assert_equal([bold_word('::Foo.bar-baz')], parse('*::Foo.bar-baz*'))
+ assert_equal([bold_word('#foo_bar=')], parse('*#foo_bar=*'))
+ assert_equal([bold_word('#foo_bar!')], parse('*#foo_bar!*'))
+ assert_equal([bold_word('#foo_bar?')], parse('*#foo_bar?*'))
+
+ assert_equal([em_word('::Foo.bar-baz')], parse('_::Foo.bar-baz_'))
+ assert_equal([em_word('#foo_bar=')], parse('_#foo_bar=_'))
+ assert_equal([em_word('#foo_bar!')], parse('_#foo_bar!_'))
+ assert_equal([em_word('#foo_bar?')], parse('_#foo_bar?_'))
+
+ assert_equal([em_word('::Foo.bar-baz')], parse('__::Foo.bar-baz__'))
+ assert_equal([em_word('#foo_bar=')], parse('__#foo_bar=__'))
+ assert_equal([em_word('#foo_bar!')], parse('__#foo_bar!__'))
+ assert_equal([em_word('#foo_bar?')], parse('__#foo_bar?__'))
+
+ assert_equal([tt_node('::Foo.bar-baz')], parse('+::Foo.bar-baz+'))
+ assert_equal([tt_node('#foo_bar=')], parse('+#foo_bar=+'))
+ assert_equal([tt_node('#foo_bar!')], parse('+#foo_bar!+'))
+ assert_equal([tt_node('#foo_bar?')], parse('+#foo_bar?+'))
+ end
+
+ def test_tt
+ assert_equal([tt_node()], parse(''))
+ assert_equal(['`a b`'], parse('`a b`'))
+ assert_equal(['x`a` `b`x'], parse('x`a` `b`x'))
+ assert_equal([tt_node('code')], parse('`code`'))
+ assert_equal([tt_node('code')], parse('+code+'))
+ assert_equal([tt_node('code')], parse('++code++'))
+ assert_equal([tt_node('code')], parse('``code``'))
+ assert_equal([tt_node('code(1 + 2)')], parse('code(1 + 2)'))
+ assert_equal([tt_node('code(1 + 2)')], parse('code(1 + 2)'))
+
+ # Detect closing tag with escaping
+ assert_equal([tt_node('ab\\')], parse('a\\b\\\\'))
+ assert_equal([tt_node('a b\\')], parse('a\\b\\\\'))
+
+ # Close with nearest non-escaped closing tag
+ assert_equal([tt_node('ab'), 'cd'], parse('a\\bcd'))
+ assert_equal([tt_node('ab'), 'cd'], parse('a\\bcd'))
+ end
+
+ def test_strike
+ assert_equal([strike_node()], parse('')) + assert_equal(['a', hard_break_node, 'b'], parse('a
b')) + assert_equal([hard_break_node, hard_break_node], parse('
')) + assert_equal([em_node('a', hard_break_node, 'b'), hard_break_node, 'c'], parse('a
b
c')) + end + + def test_simplified_tidylink + # Empty url is not allowed + assert_equal(['label[]'], parse('label[]')) + assert_equal([tidylink_node(['label'], 'url')], parse('label[url]')) + assert_equal([tidylink_node(['label'], 'http://example.com/?q=+1+')], parse('label[http://example.com/?q=+1+]')) + end + + def test_tidylink + # Empty label is allowed, empty url is not allowed + assert_equal([tidylink_node([], 'url')], parse('{}[url]')) + assert_equal(['{label}[]'], parse('{label}[]')) + + assert_equal( + [tidylink_node(['label'], 'http://example.com/')], + parse('{label}[http://example.com/]') + ) + assert_equal( + [tidylink_node(['label'], 'brac[ke]]t\\'), '_esc[]aped'], + parse('{label}[brac\[ke\]\]t\\\\]_esc\[\]aped') + ) + assert_equal( + ['See ', tidylink_node(['this link'], 'http://example.com/'), ' for more info.'], + parse('See {this link}[http://example.com/] for more info.') + ) + assert_equal( + [tidylink_node(['Label with ', bold_word('bold'), ' text'], 'http://example.com/')], + parse('{Label with *bold* text}[http://example.com/]') + ) + assert_equal( + [bold_node('bold', tidylink_node(['link'], 'http://example.com/'))], + parse('bold{link}[http://example.com/]') + ) + assert_equal( + [tidylink_node(['link'], 'http://example.com/?q=+1+')], + parse('{link}[http://example.com/?q=+1+]') + ) + assert_equal( + [tidylink_node([tt_node('}[]{')], 'url')], + parse('{
}[]{}[url]')
+ )
+ # Non-tidylink braces and brackets inside tidylink label are allowed
+ assert_equal(
+ [tidylink_node(['[a]{b}{c}d'], 'url')],
+ parse('{[a]{b}{c}d}[url]')
+ )
+ end
+
+ def test_invalid_nested_tidylink
+ # Simplified tidylink invalidates open tidylinks
+ assert_equal(
+ [bold_node('{a ', tidylink_node(['b'], 'url'), '}[', bold_word('c'), ']')],
+ parse('{a b[url]}[*c*]')
+ )
+ # Normal tidylink invalidates open tidylinks
+ assert_equal(
+ [bold_node('{a ', tidylink_node(['b'], 'url'), '}[', bold_word('c'), ']')],
+ parse('{a {b}[url]}[*c*]')
+ )
+ # Tidylink invalidates all open tidylinks
+ assert_equal(
+ [bold_node('{label', em_node('{label{label', tidylink_node(['label'], 'url'), '}[a]}[b]'), '}[c]')],
+ parse('{label{label{label{label}[url]}[a]}[b]}[c]')
+ )
+ # Valid tidylink inside invalidated tidylink
+ assert_equal(
+ [bold_node('{', tidylink_node(['label1'], 'url1'), ' ', tidylink_node(['label2'], 'url2'), '}[', bold_word('b'), ']')],
+ parse('{{label1}[url1] {label2}[url2]}[*b*]')
+ )
+ # Invalidated tidylink accepts tag break through brackets
+ assert_equal(
+ ['{', tidylink_node(['label'], 'url'), '}[', bold_node(']')],
+ parse('{{label}[url]}[]')
+ )
+ end
+
+ def test_unclosed_error_case
+ # Treat as normal text
+ assert_equal(['*unclosed bold'], parse('*unclosed bold'))
+ assert_equal(['_unclosed em'], parse('_unclosed em'))
+ assert_equal(['`unclosed tt'], parse('`unclosed tt'))
+ assert_equal(['unclosed tag'], parse('unclosed tag'))
+ assert_equal(['unclosed code'], parse('unclosed code'))
+ assert_equal(['{unclosed tidylink'], parse('{unclosed tidylink'))
+ assert_equal(['{label}[url'], parse('{label}[url'))
+ assert_equal(['label[url'], parse('label[url'))
+ end
+
+ def test_unknown_tag_as_normal_text
+ # Even if opening and closing tags are present, treat as normal text
+ assert_equal(['', strike_node(' ')], parse(''))
+ end
+
+ def test_invalid_closing_error_case
+ # No opening tag, then treat it as normal text
+ assert_equal([bold_node('')], parse(''))
+
+ # (strike closing) shouldn't close (also strike)
+ assert_equal([bold_node(strike_node(''))], parse(' '))
+
+ # Closing tag will close the last opened tag. Tag that has no corresponding open/close pair remains as normal text
+ assert_equal([em_node(''), ''], parse(''))
+
+ # Tag that has corresponding open/close pair will be parsed normally
+ assert_equal([em_node('a', bold_node('b'), 'c', bold_node('b'))], parse('b'))
+
+ # Tidylink closing brace will close the last opened tidylink
+ assert_equal([tidylink_node(['', bold_node('b')], 'url'), ''], parse('{b}[url]'))
+
+ # Tag closing will invalidate tidylink
+ assert_equal([em_node('{a', bold_node('b')), 'c}[url]'], parse('{abc}[url]'))
+
+ # Unclosed tidylink url will parsed as normal rdoc
+ assert_equal(['label[http://example.com/?q=', bold_node(), tt_node('1')], parse('label[http://example.com/?q=+1+'))
+ assert_equal(['{label}[http://example.com/?q=', bold_node(), tt_node('1')], parse('{label}[http://example.com/?q=+1+'))
+
+ # Closing brace invalidates unclosed tags
+ assert_equal(['{', bold_node('foo'), '}}[bar]'], parse('{foo}}[bar]'))
+ end
+end
diff --git a/test/rdoc/markup/markup_test.rb b/test/rdoc/markup/markup_test.rb
index 9945d83ef1..979d0a8c29 100644
--- a/test/rdoc/markup/markup_test.rb
+++ b/test/rdoc/markup/markup_test.rb
@@ -41,26 +41,6 @@ def test_convert
assert_equal expected, out
end
- def test_convert_custom_markup
- str = <<-STR
-{stricken}
- STR
-
- m = RDoc::Markup.new
- m.add_word_pair '{', '}', :STRIKE
-
- tt = RDoc::Markup::ToTest.new nil, m
- tt.add_tag :STRIKE, 'STRIKE ', ' STRIKE'
-
- out = m.convert str, tt
-
- expected = [
- "STRIKE stricken STRIKE",
- ]
-
- assert_equal expected, out
- end
-
def test_convert_document
doc = RDoc::Markup::Parser.parse <<-STR
now is
diff --git a/test/rdoc/markup/to_ansi_test.rb b/test/rdoc/markup/to_ansi_test.rb
index 19556a0f67..36a4a8bc4a 100644
--- a/test/rdoc/markup/to_ansi_test.rb
+++ b/test/rdoc/markup/to_ansi_test.rb
@@ -377,4 +377,10 @@ def test_convert_list_note
assert_equal expected, @to.convert(note_list)
end
+ def test_ansi_on_off
+ assert_equal "\e[1mab\e[m", @to.attributes('ab')
+ assert_equal "\e[1;4ma\e[0;9mb\e[m", @to.attributes('ab')
+ assert_equal "\e[1;4ma\e[24;9mb\e[m", @to.attributes('ab')
+ assert_equal "\e[1;4ma\e[22mb\e[m", @to.attributes('ab')
+ end
end
diff --git a/test/rdoc/markup/to_html_crossref_test.rb b/test/rdoc/markup/to_html_crossref_test.rb
index cf61460297..ce69aa50db 100644
--- a/test/rdoc/markup/to_html_crossref_test.rb
+++ b/test/rdoc/markup/to_html_crossref_test.rb
@@ -243,10 +243,27 @@ def test_handle_regexp_CROSSREF_show_hash_false
def test_handle_regexp_CROSSREF_with_arg_looks_like_TIDYLINK
result = @to.convert 'C1.m[:sym]'
- assert_equal para("C1.m[:sym]"), result,
+ assert_equal para("C1.m[:sym]"), result,
'C1.m[:sym]'
end
+ def test_suppress_link_inside_tidylink_label
+ result = @to.convert '{rdoc-ref:C1.m http://example.com C1}[url]'
+ assert_equal para('rdoc-ref:C1.m http://example.com C1'), result
+ end
+
+ def test_crossref_disabled_in_word_pair
+ result = @to.convert 'C1 *C1* _C1_ C1'
+ crossref = 'C1'
+ assert_equal para("#{crossref} C1 C1 #{crossref}"), result
+ end
+
+ def test_suppressed_crossref
+ result = @to.convert 'C1 \C1 \CC1'
+ crossref = 'C1'
+ assert_equal para("#{crossref} C1 CC1"), result
+ end
+
def test_handle_regexp_HYPERLINK_rdoc
readme = @store.add_file 'README.txt'
readme.parser = RDoc::Parser::Simple
@@ -266,31 +283,31 @@ def test_handle_regexp_HYPERLINK_rdoc
assert_equal 'README.txt', link
end
- def test_handle_regexp_TIDYLINK_rdoc
+ def test_handle_TIDYLINK_rdoc
readme = @store.add_file 'README.txt'
readme.parser = RDoc::Parser::Simple
@to = RDoc::Markup::ToHtmlCrossref.new @options, 'C2.html', @c2
- link = @to.handle_regexp_TIDYLINK tidy 'C2::C3'
+ link = @to.to_html tidy 'C2::C3'
assert_equal 'tidy', link
- link = @to.handle_regexp_TIDYLINK tidy 'C4'
+ link = @to.to_html tidy 'C4'
assert_equal 'tidy', link
- link = @to.handle_regexp_TIDYLINK tidy 'C1#m'
+ link = @to.to_html tidy 'C1#m'
assert_equal 'tidy', link
- link = @to.handle_regexp_TIDYLINK tidy 'README.txt'
+ link = @to.to_html tidy 'README.txt'
assert_equal 'tidy', link
end
def test_handle_regexp_TIDYLINK_label
- link = @to.handle_regexp_TIDYLINK tidy 'C1#m@foo'
+ link = @to.to_html tidy 'C1#m@foo'
assert_equal "tidy",
link, 'C1#m@foo'
@@ -341,19 +358,15 @@ def para(text)
end
def REGEXP_HANDLING(text)
- @to.handle_regexp_CROSSREF regexp_handling text
+ @to.handle_regexp_CROSSREF text
end
def hyper(reference)
- RDoc::Markup::RegexpHandling.new 0, "rdoc-ref:#{reference}"
- end
-
- def regexp_handling(text)
- RDoc::Markup::RegexpHandling.new 0, text
+ "rdoc-ref:#{reference}"
end
def tidy(reference)
- RDoc::Markup::RegexpHandling.new 0, "{tidy}[rdoc-ref:#{reference}]"
+ "{tidy}[rdoc-ref:#{reference}]"
end
end
diff --git a/test/rdoc/markup/to_html_snippet_test.rb b/test/rdoc/markup/to_html_snippet_test.rb
index a9bc00eef3..ce9118a18c 100644
--- a/test/rdoc/markup/to_html_snippet_test.rb
+++ b/test/rdoc/markup/to_html_snippet_test.rb
@@ -651,7 +651,7 @@ def test_convert_TIDYLINK_rdoc_label
end
def test_handle_regexp_HYPERLINK_link
- target = RDoc::Markup::RegexpHandling.new 0, 'link:README.txt'
+ target = 'link:README.txt'
link = @to.handle_regexp_HYPERLINK target
@@ -675,24 +675,6 @@ def test_list_verbatim_2
assert_equal 17, @to.characters
end
- def test_on_tags
- on = RDoc::Markup::AttrChanger.new 2, 0
-
- @to.on_tags [], on
-
- assert_equal 2, @to.mask
- end
-
- def test_off_tags
- on = RDoc::Markup::AttrChanger.new 2, 0
- off = RDoc::Markup::AttrChanger.new 0, 2
-
- @to.on_tags [], on
- @to.off_tags [], off
-
- assert_equal 0, @to.mask
- end
-
def test_to_html
assert_equal "--\n", util_format("--")
assert_equal 2, @to.characters
diff --git a/test/rdoc/markup/to_html_test.rb b/test/rdoc/markup/to_html_test.rb
index 3d2b010c4b..459bcb140e 100644
--- a/test/rdoc/markup/to_html_test.rb
+++ b/test/rdoc/markup/to_html_test.rb
@@ -907,7 +907,7 @@ def test_gen_url_rb_file
end
def test_handle_regexp_HYPERLINK_link
- target = RDoc::Markup::RegexpHandling.new 0, 'link:README.txt'
+ target = 'link:README.txt'
link = @to.handle_regexp_HYPERLINK target
@@ -915,7 +915,7 @@ def test_handle_regexp_HYPERLINK_link
end
def test_handle_regexp_HYPERLINK_irc
- target = RDoc::Markup::RegexpHandling.new 0, 'irc://irc.freenode.net/#ruby-lang'
+ target = 'irc://irc.freenode.net/#ruby-lang'
link = @to.handle_regexp_HYPERLINK target
@@ -998,7 +998,7 @@ def test_parseable_eh
end
def test_to_html
- assert_equal "\n
--
\n", util_format("--")
+ assert_equal "\n--\\\\
\n", util_format("--\\\\\\\\")
end
def util_format(text)
@@ -1040,6 +1040,11 @@ def test_accept_table
assert_include(res[%r<]*>C1>], 'C1')
end
+ def test_suppressed_crossref_and_backslashes
+ result = @to.convert('\\1 \\n \\Ruby \\::new \\foo_bar \\')
+ assert_equal "\n\\1 \\n Ruby ::new foo_bar \\
\n", result
+ end
+
def test_gen_url_markdown_anchor
assert_equal 'link', @to.gen_url('#hello-world', 'link')
end
@@ -1049,6 +1054,29 @@ def test_convert_tidy_link_markdown_anchor
assert_equal "\n\n", result
end
+ def test_convert_hyperlink_disabled_inside_tidylink
+ result = @to.convert '{See http://example.com}[README.txt] http://example.com'
+ assert_equal "\nSee http://example.com example.com
\n", result
+ end
+
+ def test_convert_rdoc_image_inside_tidylink
+ result = @to.convert '{See rdoc-image:image.png:text}[url] rdoc-image:image.jpg:text'
+ assert_equal "\n\n", result
+
+ # When `label =~ regexp_handling == 0`, label is handled specially in RDoc::Markup::ToHTML#apply_tidylink_label_special_regexp_handling
+ result = @to.convert '{rdoc-image:image.png:text}[url] rdoc-image:image.jpg:text'
+ assert_equal "\n![\"text\"](\"image.png\")
![\"text\"](\"image.jpg\")
\n", result
+ end
+
+ def test_convert_rdoc_label_disabled_inside_tidylink
+ result = @to.convert '{See rdoc-label:label}[url] rdoc-label:label'
+ assert_equal "\n\n", result
+
+ # When `label =~ regexp_handling == 0`, label is handled specially in RDoc::Markup::ToHTML#apply_tidylink_label_special_regexp_handling
+ result = @to.convert '{rdoc-label:label}[url] rdoc-label:label'
+ assert_equal "\n\n", result
+ end
+
def assert_escaped(unexpected, code)
result = @to.convert(code)
assert_not_include result, unexpected
diff --git a/test/rdoc/markup/to_rdoc_test.rb b/test/rdoc/markup/to_rdoc_test.rb
index 1ee6e0b26a..ea7326ec6f 100644
--- a/test/rdoc/markup/to_rdoc_test.rb
+++ b/test/rdoc/markup/to_rdoc_test.rb
@@ -2,6 +2,35 @@
require_relative '../helper'
class RDocMarkupToRDocTest < RDoc::Markup::TextFormatterTestCase
+ class ToBoldEmTT < RDoc::Markup::ToRdoc
+ def handle_BOLD(nodes)
+ emit_inline('')
+ super
+ emit_inline('')
+ end
+
+ def handle_EM(nodes)
+ emit_inline('')
+ super
+ emit_inline('')
+ end
+
+ def handle_BOLD_WORD(word)
+ emit_inline('')
+ super
+ emit_inline('')
+ end
+
+ def handle_EM_WORD(word)
+ emit_inline('')
+ super
+ emit_inline('')
+ end
+
+ def handle_TT(text)
+ emit_inline("#{text}")
+ end
+ end
add_visitor_tests
add_text_tests
@@ -9,7 +38,7 @@ class RDocMarkupToRDocTest < RDoc::Markup::TextFormatterTestCase
def setup
super
- @to = RDoc::Markup::ToRdoc.new
+ @to = ToBoldEmTT.new
end
def accept_blank_line
diff --git a/test/rdoc/rdoc_text_test.rb b/test/rdoc/rdoc_text_test.rb
index b691afa0ca..03c4167ac5 100644
--- a/test/rdoc/rdoc_text_test.rb
+++ b/test/rdoc/rdoc_text_test.rb
@@ -493,7 +493,8 @@ def test_to_html_apostrophe_entity
end
def test_to_html_backslash
- assert_equal 'S', to_html('\\S')
+ # Don't handle unescaped crossref. It should be handled in RDoc::Markup::ToHtml, not in RDoc::Text
+ assert_equal '\\S', to_html('\\S')
end
def test_to_html_br
@@ -562,8 +563,9 @@ def test_to_html_registered_trademark
end
def test_to_html_tt_tag
+ # tt tag content is already escaped
assert_equal 'hi\'s', to_html('hi\'s')
- assert_equal 'hi\\\'s', to_html('hi\\\\\'s')
+ assert_equal 'hi\\\\\'s', to_html('hi\\\\\'s')
end
def test_to_html_tt_tag_mismatch