Fix Unicode bug

Fix Diego Restrepo's bug (see Rev 184).
Update to latest HTML5lib.
This commit is contained in:
Jacques Distler 2007-12-17 03:17:43 -06:00
parent 18da1a1d71
commit 0f6889e09f
29 changed files with 380 additions and 498 deletions

View file

@ -9,6 +9,10 @@ require 'uri/common'
module Chunk
class Abstract
# Rails's default utf-8 support causes problems here. So, for Chunk::Abstract class, turn off
# multibyte character support.
$KCODE = 'iso-8859-1'
# automatically construct the array of derivatives of Chunk::Abstract
@derivatives = []

View file

@ -2,12 +2,18 @@ History.txt
Manifest.txt
README
Rakefile.rb
bin/html5
lib/core_ext/string.rb
lib/html5.rb
lib/html5/constants.rb
lib/html5/filters/base.rb
lib/html5/filters/inject_meta_charset.rb
lib/html5/filters/iso639codes.rb
lib/html5/filters/optionaltags.rb
lib/html5/filters/rfc2046.rb
lib/html5/filters/rfc3987.rb
lib/html5/filters/sanitizer.rb
lib/html5/filters/validator.rb
lib/html5/filters/whitespace.rb
lib/html5/html5parser.rb
lib/html5/html5parser/after_body_phase.rb
@ -34,6 +40,7 @@ lib/html5/sanitizer.rb
lib/html5/serializer.rb
lib/html5/serializer/htmlserializer.rb
lib/html5/serializer/xhtmlserializer.rb
lib/html5/sniffer.rb
lib/html5/tokenizer.rb
lib/html5/treebuilders.rb
lib/html5/treebuilders/base.rb
@ -46,14 +53,65 @@ lib/html5/treewalkers/hpricot.rb
lib/html5/treewalkers/rexml.rb
lib/html5/treewalkers/simpletree.rb
lib/html5/version.rb
parse.rb
testdata/encoding/chardet/test_big5.txt
testdata/encoding/test-yahoo-jp.dat
testdata/encoding/tests1.dat
testdata/encoding/tests2.dat
testdata/sanitizer/tests1.dat
testdata/serializer/core.test
testdata/serializer/injectmeta.test
testdata/serializer/optionaltags.test
testdata/serializer/options.test
testdata/serializer/whitespace.test
testdata/sites/google-results.htm
testdata/sites/python-ref-import.htm
testdata/sites/web-apps-old.htm
testdata/sites/web-apps.htm
testdata/sniffer/htmlOrFeed.json
testdata/tokenizer/contentModelFlags.test
testdata/tokenizer/entities.test
testdata/tokenizer/escapeFlag.test
testdata/tokenizer/test1.test
testdata/tokenizer/test2.test
testdata/tokenizer/test3.test
testdata/tokenizer/test4.test
testdata/tree-construction/tests1.dat
testdata/tree-construction/tests2.dat
testdata/tree-construction/tests3.dat
testdata/tree-construction/tests4.dat
testdata/tree-construction/tests5.dat
testdata/tree-construction/tests6.dat
testdata/validator/attributes.test
testdata/validator/base-href-attribute.test
testdata/validator/base-target-attribute.test
testdata/validator/blockquote-cite-attribute.test
testdata/validator/classattribute.test
testdata/validator/contenteditableattribute.test
testdata/validator/contextmenuattribute.test
testdata/validator/dirattribute.test
testdata/validator/draggableattribute.test
testdata/validator/html-xmlns-attribute.test
testdata/validator/idattribute.test
testdata/validator/inputattributes.test
testdata/validator/irrelevantattribute.test
testdata/validator/langattribute.test
testdata/validator/li-value-attribute.test
testdata/validator/link-href-attribute.test
testdata/validator/link-hreflang-attribute.test
testdata/validator/link-rel-attribute.test
testdata/validator/ol-start-attribute.test
testdata/validator/starttags.test
testdata/validator/style-scoped-attribute.test
testdata/validator/tabindexattribute.test
tests/preamble.rb
tests/test_encoding.rb
tests/test_lxp.rb
tests/test_parser.rb
tests/test_sanitizer.rb
tests/test_serializer.rb
tests/test_sniffer.rb
tests/test_stream.rb
tests/test_tokenizer.rb
tests/test_treewalkers.rb
tests/test_validator.rb
tests/tokenizer_test_parser.rb

View file

@ -18,16 +18,16 @@ end
require 'rcov/rcovtask'
namespace :test do
namespace :test do
namespace :coverage do
desc "Delete aggregate coverage data."
task(:clean) { rm_f "coverage.data" }
end
desc 'Aggregate code coverage for unit, functional and integration tests'
Rcov::RcovTask.new(:coverage => "test:coverage:clean") do |t|
t.libs << "tests"
t.test_files = FileList["tests/test_*.rb"]
t.output_dir = "tests/coverage/"
t.libs << "test"
t.test_files = FileList["test/test_*.rb"]
t.output_dir = "test/coverage/"
t.verbose = true
end
end

View file

@ -1,217 +1,5 @@
#!/usr/bin/env ruby
require 'core_ext/string'
$:.unshift File.dirname(__FILE__), 'lib'
require 'html5/cli'
def parse(opts, args)
encoding = nil
f = args[-1]
if f
begin
if f[0..6] == 'http://'
require 'open-uri'
f = URI.parse(f).open
encoding = f.charset
elsif f == '-'
f = $stdin
else
f = open(f)
end
rescue
end
else
$stderr.write("No filename provided. Use -h for help\n")
exit(1)
end
require 'html5/treebuilders'
treebuilder = HTML5::TreeBuilders[opts.treebuilder]
if opts.output == :xml
require 'html5/liberalxmlparser'
p = HTML5::XMLParser.new(:tree=>treebuilder)
else
require 'html5/html5parser'
p = HTML5::HTMLParser.new(:tree=>treebuilder)
end
if opts.parsemethod == :parse
args = [f, encoding]
else
args = [f, (opts.container || 'div'), encoding]
end
if opts.profile
require 'profiler'
Profiler__::start_profile
p.send(opts.parsemethod, *args)
Profiler__::stop_profile
Profiler__::print_profile($stderr)
elsif opts.time
require 'time' # TODO: switch to benchmark
t0 = Time.new
document = p.send(opts.parsemethod, *args)
t1 = Time.new
print_output(p, document, opts)
t2 = Time.new
puts "\n\nRun took: %fs (plus %fs to print the output)"%[t1-t0, t2-t1]
else
document = p.send(opts.parsemethod, *args)
print_output(p, document, opts)
end
end
def print_output(parser, document, opts)
puts "Encoding: #{parser.tokenizer.stream.char_encoding}" if opts.encoding
case opts.output
when :xml
print document
when :html
require 'html5/treewalkers'
tokens = HTML5::TreeWalkers[opts.treebuilder].new(document)
require 'html5/serializer'
puts HTML5::HTMLSerializer.serialize(tokens, opts.serializer)
when :hilite
print document.hilite
when :tree
document = [document] unless document.respond_to?(:each)
document.each {|fragment| puts parser.tree.testSerializer(fragment)}
end
if opts.error
errList=[]
for pos, errorcode, datavars in parser.errors
errList << "Line #{pos[0]} Col #{pos[1]} " + (HTML5::E[errorcode] || "Unknown error \"#{errorcode}\"") % datavars
end
$stdout.write("\nParse errors:\n" + errList.join("\n")+"\n")
end
end
require 'ostruct'
options = OpenStruct.new
options.profile = false
options.time = false
options.output = :html
options.treebuilder = 'simpletree'
options.error = false
options.encoding = false
options.parsemethod = :parse
options.serializer = {
:encoding => 'utf-8',
:omit_optional_tags => false,
:inject_meta_charset => false
}
require 'optparse'
opts = OptionParser.new do |opts|
opts.separator ""
opts.separator "Parse Options:"
opts.on("-b", "--treebuilder NAME") do |treebuilder|
options.treebuilder = treebuilder
end
opts.on("-f", "--fragment CONTAINER", "Parse as a fragment") do |container|
options.parsemethod = :parse_fragment
options.container = container if container
end
opts.separator ""
opts.separator "Filter Options:"
opts.on("--[no-]inject-meta-charset", "inject <meta charset>") do |inject|
options.serializer[:inject_meta_charset] = inject
end
opts.on("--[no-]strip-whitespace", "strip unnecessary whitespace") do |strip|
options.serializer[:strip_whitespace] = strip
end
opts.on("--[no-]sanitize", "escape unsafe tags") do |sanitize|
options.serializer[:sanitize] = sanitize
end
opts.separator ""
opts.separator "Output Options:"
opts.on("--tree", "output as debug tree") do |tree|
options.output = :tree
end
opts.on("-x", "--xml", "output as xml") do |xml|
options.output = :xml
options.treebuilder = "rexml"
end
opts.on("--[no-]html", "Output as html") do |html|
options.output = (html ? :html : nil)
end
opts.on("--hilite", "Output as formatted highlighted code.") do |hilite|
options.output = :hilite
end
opts.on("-e", "--error", "Print a list of parse errors") do |error|
options.error = error
end
opts.separator ""
opts.separator "Serialization Options:"
opts.on("--[no-]omit-optional-tags", "Omit optional tags") do |omit|
options.serializer[:omit_optional_tags] = omit
end
opts.on("--[no-]quote-attr-values", "Quote attribute values") do |quote|
options.serializer[:quote_attr_values] = quote
end
opts.on("--[no-]use-best-quote-char", "Use best quote character") do |best|
options.serializer[:use_best_quote_char] = best
end
opts.on("--quote-char C", "Use specified quote character") do |c|
options.serializer[:quote_char] = c
end
opts.on("--[no-]minimize-boolean-attributes", "Minimize boolean attributes") do |min|
options.serializer[:minimize_boolean_attributes] = min
end
opts.on("--[no-]use-trailing-solidus", "Use trailing solidus") do |slash|
options.serializer[:use_trailing_solidus] = slash
end
opts.on("--[no-]escape-lt-in-attrs", "Escape less than signs in attribute values") do |lt|
options.serializer[:escape_lt_in_attrs] = lt
end
opts.on("--[no-]escape-rcdata", "Escape rcdata element values") do |rcdata|
options.serializer[:escape_rcdata] = rcdata
end
opts.separator ""
opts.separator "Other Options:"
opts.on("-p", "--[no-]profile", "Profile the run") do |profile|
options.profile = profile
end
opts.on("-t", "--[no-]time", "Time the run") do |time|
options.time = time
end
opts.on("-c", "--[no-]encoding", "Print character encoding used") do |encoding|
options.encoding = encoding
end
opts.on_tail("-h", "--help", "Show this message") do
puts opts
exit
end
end
opts.parse!(ARGV)
parse options, ARGV
HTML5::CLI.run

View file

@ -8,6 +8,6 @@ module HTML5
end
def self.parse_fragment(stream, options={})
HTMLParser.parse(stream, options)
HTMLParser.parse_fragment(stream, options)
end
end

231
vendor/plugins/HTML5lib/lib/html5/cli.rb vendored Normal file
View file

@ -0,0 +1,231 @@
$:.unshift File.dirname(__FILE__), 'lib'
require 'html5'
require 'core_ext/string'
require 'ostruct'
require 'optparse'
module HTML5::CLI
def self.parse_opts argv
options = OpenStruct.new
options.profile = false
options.time = false
options.output = :html
options.treebuilder = 'simpletree'
options.error = false
options.encoding = false
options.parsemethod = :parse
options.serializer = {
:encoding => 'utf-8',
:omit_optional_tags => false,
:inject_meta_charset => false
}
opts = OptionParser.new do |opts|
opts.separator ""
opts.separator "Parse Options:"
opts.on("-b", "--treebuilder NAME") do |treebuilder|
options.treebuilder = treebuilder
end
opts.on("-f", "--fragment CONTAINER", "Parse as a fragment") do |container|
options.parsemethod = :parse_fragment
options.container = container if container
end
opts.separator ""
opts.separator "Filter Options:"
opts.on("--[no-]inject-meta-charset", "inject <meta charset>") do |inject|
options.serializer[:inject_meta_charset] = inject
end
opts.on("--[no-]strip-whitespace", "strip unnecessary whitespace") do |strip|
options.serializer[:strip_whitespace] = strip
end
opts.on("--[no-]sanitize", "escape unsafe tags") do |sanitize|
options.serializer[:sanitize] = sanitize
end
opts.separator ""
opts.separator "Output Options:"
opts.on("--tree", "output as debug tree") do |tree|
options.output = :tree
end
opts.on("-x", "--xml", "output as xml") do |xml|
options.output = :xml
options.treebuilder = "rexml"
end
opts.on("--[no-]html", "Output as html") do |html|
options.output = (html ? :html : nil)
end
opts.on("--hilite", "Output as formatted highlighted code.") do |hilite|
options.output = :hilite
end
opts.on("-e", "--error", "Print a list of parse errors") do |error|
options.error = error
end
opts.separator ""
opts.separator "Serialization Options:"
opts.on("--[no-]omit-optional-tags", "Omit optional tags") do |omit|
options.serializer[:omit_optional_tags] = omit
end
opts.on("--[no-]quote-attr-values", "Quote attribute values") do |quote|
options.serializer[:quote_attr_values] = quote
end
opts.on("--[no-]use-best-quote-char", "Use best quote character") do |best|
options.serializer[:use_best_quote_char] = best
end
opts.on("--quote-char C", "Use specified quote character") do |c|
options.serializer[:quote_char] = c
end
opts.on("--[no-]minimize-boolean-attributes", "Minimize boolean attributes") do |min|
options.serializer[:minimize_boolean_attributes] = min
end
opts.on("--[no-]use-trailing-solidus", "Use trailing solidus") do |slash|
options.serializer[:use_trailing_solidus] = slash
end
opts.on("--[no-]escape-lt-in-attrs", "Escape less than signs in attribute values") do |lt|
options.serializer[:escape_lt_in_attrs] = lt
end
opts.on("--[no-]escape-rcdata", "Escape rcdata element values") do |rcdata|
options.serializer[:escape_rcdata] = rcdata
end
opts.separator ""
opts.separator "Other Options:"
opts.on("-p", "--[no-]profile", "Profile the run") do |profile|
options.profile = profile
end
opts.on("-t", "--[no-]time", "Time the run") do |time|
options.time = time
end
opts.on("-c", "--[no-]encoding", "Print character encoding used") do |encoding|
options.encoding = encoding
end
opts.on_tail("-h", "--help", "Show this message") do
puts opts
exit
end
end
opts.parse!(argv)
options
end
def self.open_input f
if f
begin
if f[0..6] == 'http://'
require 'open-uri'
f = URI.parse(f).open
encoding = f.charset
elsif f == '-'
f = $stdin
else
f = open(f)
end
rescue
end
else
$stderr.write("No filename provided. Use -h for help\n")
exit(1)
end
f
end
def self.parse(opts, args)
encoding = nil
f = open_input args.last
require 'html5/treebuilders'
treebuilder = HTML5::TreeBuilders[opts.treebuilder]
if opts.output == :xml
require 'html5/liberalxmlparser'
p = HTML5::XMLParser.new(:tree=>treebuilder)
else
require 'html5/html5parser'
p = HTML5::HTMLParser.new(:tree=>treebuilder)
end
if opts.parsemethod == :parse
args = [f, encoding]
else
args = [f, (opts.container || 'div'), encoding]
end
if opts.profile
require 'profiler'
Profiler__::start_profile
p.send(opts.parsemethod, *args)
Profiler__::stop_profile
Profiler__::print_profile($stderr)
elsif opts.time
require 'time' # TODO: switch to benchmark
t0 = Time.new
document = p.send(opts.parsemethod, *args)
t1 = Time.new
print_output(p, document, opts)
t2 = Time.new
puts "\n\nRun took: %fs (plus %fs to print the output)"%[t1-t0, t2-t1]
else
document = p.send(opts.parsemethod, *args)
print_output(p, document, opts)
end
end
def self.print_output(parser, document, opts)
puts "Encoding: #{parser.tokenizer.stream.char_encoding}" if opts.encoding
case opts.output
when :xml
print document
when :html
require 'html5/treewalkers'
tokens = HTML5::TreeWalkers[opts.treebuilder].new(document)
require 'html5/serializer'
puts HTML5::HTMLSerializer.serialize(tokens, opts.serializer)
when :hilite
print document.hilite
when :tree
document = [document] unless document.respond_to?(:each)
document.each {|fragment| puts parser.tree.testSerializer(fragment)}
end
if opts.error
errList=[]
for pos, errorcode, datavars in parser.errors
errList << "Line #{pos[0]} Col #{pos[1]} " + (HTML5::E[errorcode] || "Unknown error \"#{errorcode}\"") % datavars
end
$stdout.write("\nParse errors:\n" + errList.join("\n")+"\n")
end
end
def self.run
options = parse_opts ARGV
parse options, ARGV
end
end

View file

@ -908,7 +908,7 @@ module HTML5
"eof-in-bogus-doctype" =>
_("Unexpected end of file in bogus doctype."),
"eof-in-innerhtml" =>
_("XXX innerHTML EOF"),
_("Unexpected EOF in inner html mode."),
"unexpected-doctype" =>
_("Unexpected DOCTYPE. Ignored."),
"non-html-root" =>
@ -1040,7 +1040,8 @@ module HTML5
_("Unexpected end tag (%(name))" +
". Expected end of file."),
"unexpected-end-table-in-caption" =>
_("Unexpected end table tag in caption. Generates implied end caption.")
_("Unexpected end table tag in caption. Generates implied end caption."),
"end-html-in-innerhtml" => _("Unexpected html end tag in inner html mode.")
}
end

View file

@ -1 +0,0 @@
require 'html5/filters/optionaltags'

View file

@ -25,7 +25,7 @@ module HTML5
def endTagHtml(name)
if @parser.inner_html
parse_error
parse_error "end-html-in-innerhtml"
else
# XXX: This may need to be done, not sure
# Don't set last_phase to the current phase but to the inBody phase

View file

@ -51,34 +51,22 @@ module HTML5
super(parser, tree)
# for special handling of whitespace in <pre>
if $-w
$-w = false
class << self; alias processSpaceCharactersNonPre processSpaceCharacters; end
$-w = true
else
class << self; alias processSpaceCharactersNonPre processSpaceCharacters; end
silence do
class << self
alias processSpaceCharactersNonPre processSpaceCharacters
end
end
end
def processSpaceCharactersDropNewline(data)
# #Sometimes (start of <pre> blocks) we want to drop leading newlines
if $-w
$-w = false
class << self
silence do
alias processSpaceCharacters processSpaceCharactersNonPre
end
end
$-w = true
else
class << self
silence do
alias processSpaceCharacters processSpaceCharactersNonPre
end
class << self
silence do
alias processSpaceCharacters processSpaceCharactersNonPre
end
end
if (data.length > 0 and data[0] == ?\n &&
%w[pre textarea].include?(@tree.open_elements.last.name) && !@tree.open_elements.last.hasContent)
data = data[1..-1]
@ -376,16 +364,6 @@ module HTML5
end
def endTagBlock(name)
#Put us back in the right whitespace handling mode
if name == 'pre'
class << self;
silence do
alias processSpaceCharacters processSpaceCharactersNonPre;
end
end
end
@tree.generateImpliedEndTags if in_scope?(name)
unless @tree.open_elements.last.name == name

View file

@ -144,7 +144,7 @@ module HTML5
def remove_open_elements_until(name=nil)
finished = false
until finished
until finished || @tree.open_elements.length == 0
element = @tree.open_elements.pop
finished = name.nil? ? yield(element) : element.name == name
end

View file

@ -1,217 +0,0 @@
#!/usr/bin/env ruby
#
# Parse a document to a simpletree tree, with optional profiling
$:.unshift File.dirname(__FILE__),'lib'
def parse(opts, args)
encoding = nil
f = args[-1]
if f
begin
if f[0..6] == 'http://'
require 'open-uri'
f = URI.parse(f).open
encoding = f.charset
elsif f == '-'
f = $stdin
else
f = open(f)
end
rescue
end
else
$stderr.write("No filename provided. Use -h for help\n")
exit(1)
end
require 'html5/treebuilders'
treebuilder = HTML5::TreeBuilders[opts.treebuilder]
if opts.output == :xml
require 'html5/liberalxmlparser'
p = HTML5::XHTMLParser.new(:tree=>treebuilder)
else
require 'html5/html5parser'
p = HTML5::HTMLParser.new(:tree=>treebuilder)
end
if opts.parsemethod == :parse
args = [f, encoding]
else
args = [f, 'div', encoding]
end
if opts.profile
require 'profiler'
Profiler__::start_profile
p.send(opts.parsemethod, *args)
Profiler__::stop_profile
Profiler__::print_profile($stderr)
elsif opts.time
require 'time'
t0 = Time.new
document = p.send(opts.parsemethod, *args)
t1 = Time.new
printOutput(p, document, opts)
t2 = Time.new
puts "\n\nRun took: %fs (plus %fs to print the output)"%[t1-t0, t2-t1]
else
document = p.send(opts.parsemethod, *args)
printOutput(p, document, opts)
end
end
def printOutput(parser, document, opts)
puts "Encoding: #{parser.tokenizer.stream.char_encoding}" if opts.encoding
case opts.output
when :xml
print document
when :html
require 'html5/treewalkers'
tokens = HTML5::TreeWalkers[opts.treebuilder].new(document)
require 'html5/serializer'
puts HTML5::HTMLSerializer.serialize(tokens, opts.serializer)
when :hilite
print document.hilite
when :tree
document = [document] unless document.respond_to?(:each)
document.each {|fragment| puts parser.tree.testSerializer(fragment)}
end
if opts.error
errList=[]
for pos, message in parser.errors
errList << ("Line %i Col %i"%pos + " " + message)
end
$stdout.write("\nParse errors:\n" + errList.join("\n")+"\n")
end
end
require 'ostruct'
options = OpenStruct.new
options.profile = false
options.time = false
options.output = :html
options.treebuilder = 'simpletree'
options.error = false
options.encoding = false
options.parsemethod = :parse
options.serializer = {
:encoding => 'utf-8',
:omit_optional_tags => false,
:inject_meta_charset => false
}
require 'optparse'
opts = OptionParser.new do |opts|
opts.separator ""
opts.separator "Parse Options:"
opts.on("-b", "--treebuilder NAME") do |treebuilder|
options.treebuilder = treebuilder
end
opts.on("-f", "--fragment", "Parse as a fragment") do |parse|
options.parsemethod = :parseFragment
end
opts.separator ""
opts.separator "Filter Options:"
opts.on("--[no-]inject-meta-charset", "inject <meta charset>") do |inject|
options.serializer[:inject_meta_charset] = inject
end
opts.on("--[no-]strip-whitespace", "strip unnecessary whitespace") do |strip|
options.serializer[:strip_whitespace] = strip
end
opts.on("--[no-]sanitize", "escape unsafe tags") do |sanitize|
options.serializer[:sanitize] = sanitize
end
opts.separator ""
opts.separator "Output Options:"
opts.on("--tree", "output as debug tree") do |tree|
options.output = :tree
end
opts.on("-x", "--xml", "output as xml") do |xml|
options.output = :xml
options.treebuilder = "rexml"
end
opts.on("--[no-]html", "Output as html") do |html|
options.output = (html ? :html : nil)
end
opts.on("--hilite", "Output as formatted highlighted code.") do |hilite|
options.output = :hilite
end
opts.on("-e", "--error", "Print a list of parse errors") do |error|
options.error = error
end
opts.separator ""
opts.separator "Serialization Options:"
opts.on("--[no-]omit-optional-tags", "Omit optional tags") do |omit|
options.serializer[:omit_optional_tags] = omit
end
opts.on("--[no-]quote-attr-values", "Quote attribute values") do |quote|
options.serializer[:quote_attr_values] = quote
end
opts.on("--[no-]use-best-quote-char", "Use best quote character") do |best|
options.serializer[:use_best_quote_char] = best
end
opts.on("--quote-char C", "Use specified quote character") do |c|
options.serializer[:quote_char] = c
end
opts.on("--[no-]minimize-boolean-attributes", "Minimize boolean attributes") do |min|
options.serializer[:minimize_boolean_attributes] = min
end
opts.on("--[no-]use-trailing-solidus", "Use trailing solidus") do |slash|
options.serializer[:use_trailing_solidus] = slash
end
opts.on("--[no-]escape-lt-in-attrs", "Escape less than signs in attribute values") do |lt|
options.serializer[:escape_lt_in_attrs] = lt
end
opts.on("--[no-]escape-rcdata", "Escape rcdata element values") do |rcdata|
options.serializer[:escape_rcdata] = rcdata
end
opts.separator ""
opts.separator "Other Options:"
opts.on("-p", "--[no-]profile", "Profile the run") do |profile|
options.profile = profile
end
opts.on("-t", "--[no-]time", "Time the run") do |time|
options.time = time
end
opts.on("-c", "--[no-]encoding", "Print character encoding used") do |encoding|
options.encoding = encoding
end
opts.on_tail("-h", "--help", "Show this message") do
puts opts
exit
end
end
opts.parse!(ARGV)
parse options, ARGV

View file

@ -2,15 +2,14 @@ require 'test/unit'
HTML5_BASE = File.dirname(File.dirname(File.dirname(File.expand_path(__FILE__))))
if File.exists?(File.join(HTML5_BASE, 'testdata'))
TESTDATA_DIR = File.join(HTML5_BASE, 'testdata')
if File.exists?(File.join(HTML5_BASE, 'ruby', 'testdata'))
TESTDATA_DIR = File.join(HTML5_BASE, 'ruby', 'testdata')
else
TESTDATA_DIR = File.join(File.dirname(File.dirname(File.expand_path(__FILE__))), 'testdata')
TESTDATA_DIR = File.join(HTML5_BASE, 'testdata')
end
# $:.unshift File.join(File.dirname(File.dirname(__FILE__)), 'lib')
# $:.unshift File.dirname(__FILE__)
$:.unshift File.join(File.dirname(File.dirname(__FILE__)), 'lib')
$:.unshift File.dirname(__FILE__)
require 'core_ext/string'

View file

@ -0,0 +1,16 @@
require File.join(File.dirname(__FILE__), 'preamble')
require "html5/cli"
class TestCli < Test::Unit::TestCase
def test_open_input
assert_equal $stdin, HTML5::CLI.open_input('-')
assert_kind_of StringIO, HTML5::CLI.open_input('http://whatwg.org/')
assert_kind_of File, HTML5::CLI.open_input('testdata/sites/google-results.htm')
end
def test_parse_opts
HTML5::CLI.parse_opts [] # TODO test defaults
assert_equal 'hpricot', HTML5::CLI.parse_opts(['-b', 'hpricot']).treebuilder
assert_equal 'hpricot', HTML5::CLI.parse_opts(['--treebuilder', 'hpricot']).treebuilder
end
end

View file

@ -14,4 +14,13 @@ class TestHtml5Inputstream < Test::Unit::TestCase
1022.times{stream.char}
assert_equal "i", stream.char
end
def test_chars_until
stream = HTML5::HTMLInputStream.new("aaaaaaab")
assert_equal "aaaaaaa", stream.chars_until("b")
stream = HTML5::HTMLInputStream.new("aaaaaaab")
assert_equal "aaaaaaab", stream.chars_until("c")
end
end

View file

@ -12,11 +12,6 @@ begin
rescue LoadError
end
$CHECK_PARSER_ERRORS = ARGV.delete('-p') # TODO
puts 'Testing tree builders: ' + $tree_types_to_test * ', '
class Html5ParserTestCase < Test::Unit::TestCase
include HTML5
include TestSupport
@ -25,8 +20,7 @@ class Html5ParserTestCase < Test::Unit::TestCase
test_name = File.basename(test_file).sub('.dat', '')
TestData.new(test_file, %w(data errors document-fragment document)).
each_with_index do |(input, errors, inner_html, expected), index|
TestData.new(test_file, %w(data errors document-fragment document)).each_with_index do |(input, errors, inner_html, expected), index|
errors = errors.split("\n")
expected = expected.gsub("\n| ","\n")[2..-1]
@ -35,13 +29,13 @@ class Html5ParserTestCase < Test::Unit::TestCase
define_method 'test_%s_%d_%s' % [ test_name, index + 1, tree_name ] do
parser = HTMLParser.new(:tree => TreeBuilders[tree_name])
if inner_html
parser.parse_fragment(input, inner_html)
else
parser.parse(input)
end
actual_output = convertTreeDump(parser.tree.testSerializer(parser.tree.document))
assert_equal sortattrs(expected), sortattrs(actual_output), [
@ -53,13 +47,13 @@ class Html5ParserTestCase < Test::Unit::TestCase
actual_errors = parser.errors.map do |(line, col), message, datavars|
'Line: %i Col: %i %s' % [line, col, E[message] % datavars]
end
assert_equal errors.length, parser.errors.length, [
assert_equal errors, actual_errors, [
'', 'Input', input,
'', "Expected errors (#{errors.length}):", errors.join("\n"),
'', "Actual errors (#{actual_errors.length}):",
actual_errors.join("\n")
actual_errors.join("\n") + "\n"
].join("\n")
end
end
end

View file

@ -42,9 +42,10 @@ class HTMLInputStreamTest < Test::Unit::TestCase
require 'iconv'
def test_utf_16
stream = HTMLInputStream.new("\xff\xfe" + " \x00"*1025)
assert(stream.char_encoding, 'utf-16-le')
assert_equal(1025, stream.chars_until(' ',true).length)
input = Iconv.new('utf-16', 'utf-8').iconv(' '*1025)
stream = HTMLInputStream.new(input)
assert('utf-16-le', stream.char_encoding)
assert_equal(1025, stream.chars_until(' ', true).length)
end
rescue LoadError
puts "iconv not found, skipping iconv tests"

View file

@ -433,9 +433,9 @@
{
"name": "uri_ref_with_space_in svg_attribute",
"input": "<rect fill='url(\n#foo)' />",
"rexml": "<rect fill=\'url(\n#foo)\'></rect>",
"xhtml": "<rect fill=\'url(\n#foo)\'></rect>",
"output": "<rect fill=\'url(\n#foo)\'/>"
"rexml": "<rect fill='url(\n#foo)'></rect>",
"xhtml": "<rect fill='url(\n#foo)'></rect>",
"output": "<rect fill='url(\n#foo)'/>"
},
{

View file

@ -41,7 +41,7 @@ plaintext
#data
setting html's innerHTML
#errors
Line: 1 Col: 24 XXX innerHTML EOF
Line: 1 Col: 24 Unexpected EOF in inner html mode.
#document-fragment
html
#document

View file

@ -608,4 +608,25 @@ Line: 1 Col: 6 Unexpected start tag (html). Expected DOCTYPE.
| <html>
| <head>
| <body>
| <!-- foo -->
| <!-- foo -->
#data
<body></body></html>
#document-fragment
html
#errors
Line: 1 Col: 20 Unexpected html end tag in inner html mode.
Line: 1 Col: 20 Unexpected EOF in inner html mode.
#document
| <head>
| <body>
#data
<html><frameset></frameset></html>
#errors
Line: 1 Col: 6 Unexpected start tag (html). Expected DOCTYPE.
#document
| <html>
| <head>
| <frameset>
| " "