feedtools 0.2.26 → 0.2.27
Sign up to get free protection for your applications and to get access to all the features.
- data/CHANGELOG +232 -216
- data/db/migration.rb +2 -0
- data/db/schema.mysql.sql +2 -0
- data/db/schema.postgresql.sql +3 -1
- data/db/schema.sqlite.sql +3 -1
- data/lib/feed_tools.rb +37 -14
- data/lib/feed_tools/database_feed_cache.rb +13 -2
- data/lib/feed_tools/feed.rb +430 -104
- data/lib/feed_tools/feed_item.rb +533 -268
- data/lib/feed_tools/helpers/generic_helper.rb +1 -1
- data/lib/feed_tools/helpers/html_helper.rb +78 -116
- data/lib/feed_tools/helpers/retrieval_helper.rb +33 -3
- data/lib/feed_tools/helpers/uri_helper.rb +46 -54
- data/lib/feed_tools/monkey_patch.rb +27 -1
- data/lib/feed_tools/vendor/html5/History.txt +10 -0
- data/lib/feed_tools/vendor/html5/Manifest.txt +117 -0
- data/lib/feed_tools/vendor/html5/README +45 -0
- data/lib/feed_tools/vendor/html5/Rakefile.rb +33 -0
- data/lib/feed_tools/vendor/html5/bin/html5 +217 -0
- data/lib/feed_tools/vendor/html5/lib/core_ext/string.rb +17 -0
- data/lib/feed_tools/vendor/html5/lib/html5.rb +13 -0
- data/lib/feed_tools/vendor/html5/lib/html5/constants.rb +1046 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/base.rb +10 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/inject_meta_charset.rb +82 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/iso639codes.rb +752 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/optionaltags.rb +198 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/rfc2046.rb +30 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/rfc3987.rb +89 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/sanitizer.rb +15 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/validator.rb +830 -0
- data/lib/feed_tools/vendor/html5/lib/html5/filters/whitespace.rb +36 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser.rb +248 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/after_body_phase.rb +46 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/after_frameset_phase.rb +33 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/after_head_phase.rb +50 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/before_head_phase.rb +41 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_body_phase.rb +613 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_caption_phase.rb +69 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_cell_phase.rb +78 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_column_group_phase.rb +55 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_frameset_phase.rb +57 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_head_phase.rb +138 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_row_phase.rb +89 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_select_phase.rb +85 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_table_body_phase.rb +86 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/in_table_phase.rb +115 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/initial_phase.rb +133 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/phase.rb +154 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/root_element_phase.rb +41 -0
- data/lib/feed_tools/vendor/html5/lib/html5/html5parser/trailing_end_phase.rb +35 -0
- data/lib/feed_tools/vendor/html5/lib/html5/inputstream.rb +648 -0
- data/lib/feed_tools/vendor/html5/lib/html5/liberalxmlparser.rb +158 -0
- data/lib/feed_tools/vendor/html5/lib/html5/sanitizer.rb +188 -0
- data/lib/feed_tools/vendor/html5/lib/html5/serializer.rb +2 -0
- data/lib/feed_tools/vendor/html5/lib/html5/serializer/htmlserializer.rb +179 -0
- data/lib/feed_tools/vendor/html5/lib/html5/serializer/xhtmlserializer.rb +20 -0
- data/lib/feed_tools/vendor/html5/lib/html5/sniffer.rb +45 -0
- data/lib/feed_tools/vendor/html5/lib/html5/tokenizer.rb +966 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treebuilders.rb +24 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treebuilders/base.rb +334 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treebuilders/hpricot.rb +231 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treebuilders/rexml.rb +209 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treebuilders/simpletree.rb +185 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treewalkers.rb +26 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treewalkers/base.rb +162 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treewalkers/hpricot.rb +48 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treewalkers/rexml.rb +48 -0
- data/lib/feed_tools/vendor/html5/lib/html5/treewalkers/simpletree.rb +48 -0
- data/lib/feed_tools/vendor/html5/lib/html5/version.rb +3 -0
- data/lib/feed_tools/vendor/html5/testdata/encoding/chardet/test_big5.txt +51 -0
- data/lib/feed_tools/vendor/html5/testdata/encoding/test-yahoo-jp.dat +10 -0
- data/lib/feed_tools/vendor/html5/testdata/encoding/tests1.dat +394 -0
- data/lib/feed_tools/vendor/html5/testdata/encoding/tests2.dat +81 -0
- data/lib/feed_tools/vendor/html5/testdata/sanitizer/tests1.dat +416 -0
- data/lib/feed_tools/vendor/html5/testdata/serializer/core.test +104 -0
- data/lib/feed_tools/vendor/html5/testdata/serializer/injectmeta.test +65 -0
- data/lib/feed_tools/vendor/html5/testdata/serializer/optionaltags.test +900 -0
- data/lib/feed_tools/vendor/html5/testdata/serializer/options.test +60 -0
- data/lib/feed_tools/vendor/html5/testdata/serializer/whitespace.test +51 -0
- data/lib/feed_tools/vendor/html5/testdata/sites/google-results.htm +1 -0
- data/lib/feed_tools/vendor/html5/testdata/sites/python-ref-import.htm +1 -0
- data/lib/feed_tools/vendor/html5/testdata/sites/web-apps-old.htm +1 -0
- data/lib/feed_tools/vendor/html5/testdata/sites/web-apps.htm +34275 -0
- data/lib/feed_tools/vendor/html5/testdata/sniffer/htmlOrFeed.json +43 -0
- data/lib/feed_tools/vendor/html5/testdata/tokenizer/contentModelFlags.test +48 -0
- data/lib/feed_tools/vendor/html5/testdata/tokenizer/entities.test +2339 -0
- data/lib/feed_tools/vendor/html5/testdata/tokenizer/escapeFlag.test +21 -0
- data/lib/feed_tools/vendor/html5/testdata/tokenizer/test1.test +172 -0
- data/lib/feed_tools/vendor/html5/testdata/tokenizer/test2.test +129 -0
- data/lib/feed_tools/vendor/html5/testdata/tokenizer/test3.test +367 -0
- data/lib/feed_tools/vendor/html5/testdata/tokenizer/test4.test +198 -0
- data/lib/feed_tools/vendor/html5/testdata/tree-construction/tests1.dat +1950 -0
- data/lib/feed_tools/vendor/html5/testdata/tree-construction/tests2.dat +773 -0
- data/lib/feed_tools/vendor/html5/testdata/tree-construction/tests3.dat +270 -0
- data/lib/feed_tools/vendor/html5/testdata/tree-construction/tests4.dat +60 -0
- data/lib/feed_tools/vendor/html5/testdata/tree-construction/tests5.dat +175 -0
- data/lib/feed_tools/vendor/html5/testdata/tree-construction/tests6.dat +196 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/attributes.test +1035 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/base-href-attribute.test +787 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/base-target-attribute.test +35 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/blockquote-cite-attribute.test +7 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/classattribute.test +152 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/contenteditableattribute.test +59 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/contextmenuattribute.test +115 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/dirattribute.test +59 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/draggableattribute.test +63 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/html-xmlns-attribute.test +23 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/idattribute.test +115 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/inputattributes.test +2795 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/irrelevantattribute.test +63 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/langattribute.test +5579 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/li-value-attribute.test +7 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/link-href-attribute.test +7 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/link-hreflang-attribute.test +7 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/link-rel-attribute.test +271 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/ol-start-attribute.test +7 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/starttags.test +375 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/style-scoped-attribute.test +7 -0
- data/lib/feed_tools/vendor/html5/testdata/validator/tabindexattribute.test +79 -0
- data/lib/feed_tools/vendor/html5/tests/preamble.rb +72 -0
- data/lib/feed_tools/vendor/html5/tests/test_encoding.rb +35 -0
- data/lib/feed_tools/vendor/html5/tests/test_lxp.rb +279 -0
- data/lib/feed_tools/vendor/html5/tests/test_parser.rb +68 -0
- data/lib/feed_tools/vendor/html5/tests/test_sanitizer.rb +142 -0
- data/lib/feed_tools/vendor/html5/tests/test_serializer.rb +68 -0
- data/lib/feed_tools/vendor/html5/tests/test_sniffer.rb +27 -0
- data/lib/feed_tools/vendor/html5/tests/test_stream.rb +62 -0
- data/lib/feed_tools/vendor/html5/tests/test_tokenizer.rb +94 -0
- data/lib/feed_tools/vendor/html5/tests/test_treewalkers.rb +135 -0
- data/lib/feed_tools/vendor/html5/tests/test_validator.rb +31 -0
- data/lib/feed_tools/vendor/html5/tests/tokenizer_test_parser.rb +63 -0
- data/lib/feed_tools/vendor/uri.rb +781 -0
- data/lib/feed_tools/version.rb +1 -1
- data/rakefile +27 -6
- data/test/unit/atom_test.rb +298 -210
- data/test/unit/helper_test.rb +7 -12
- data/test/unit/rdf_test.rb +51 -1
- data/test/unit/rss_test.rb +13 -3
- metadata +239 -116
- data/lib/feed_tools/vendor/htree.rb +0 -97
- data/lib/feed_tools/vendor/htree/container.rb +0 -10
- data/lib/feed_tools/vendor/htree/context.rb +0 -67
- data/lib/feed_tools/vendor/htree/display.rb +0 -27
- data/lib/feed_tools/vendor/htree/doc.rb +0 -149
- data/lib/feed_tools/vendor/htree/elem.rb +0 -262
- data/lib/feed_tools/vendor/htree/encoder.rb +0 -163
- data/lib/feed_tools/vendor/htree/equality.rb +0 -218
- data/lib/feed_tools/vendor/htree/extract_text.rb +0 -37
- data/lib/feed_tools/vendor/htree/fstr.rb +0 -33
- data/lib/feed_tools/vendor/htree/gencode.rb +0 -97
- data/lib/feed_tools/vendor/htree/htmlinfo.rb +0 -672
- data/lib/feed_tools/vendor/htree/inspect.rb +0 -108
- data/lib/feed_tools/vendor/htree/leaf.rb +0 -94
- data/lib/feed_tools/vendor/htree/loc.rb +0 -367
- data/lib/feed_tools/vendor/htree/modules.rb +0 -48
- data/lib/feed_tools/vendor/htree/name.rb +0 -124
- data/lib/feed_tools/vendor/htree/output.rb +0 -207
- data/lib/feed_tools/vendor/htree/parse.rb +0 -409
- data/lib/feed_tools/vendor/htree/raw_string.rb +0 -124
- data/lib/feed_tools/vendor/htree/regexp-util.rb +0 -15
- data/lib/feed_tools/vendor/htree/rexml.rb +0 -130
- data/lib/feed_tools/vendor/htree/scan.rb +0 -166
- data/lib/feed_tools/vendor/htree/tag.rb +0 -111
- data/lib/feed_tools/vendor/htree/template.rb +0 -909
- data/lib/feed_tools/vendor/htree/text.rb +0 -115
- data/lib/feed_tools/vendor/htree/traverse.rb +0 -465
@@ -0,0 +1,135 @@
|
|
1
|
+
require File.join(File.dirname(__FILE__), 'preamble')
|
2
|
+
|
3
|
+
require 'html5/html5parser'
|
4
|
+
require 'html5/treewalkers'
|
5
|
+
require 'html5/treebuilders'
|
6
|
+
|
7
|
+
$tree_types_to_test = {
|
8
|
+
'simpletree' =>
|
9
|
+
{:builder => HTML5::TreeBuilders['simpletree'],
|
10
|
+
:walker => HTML5::TreeWalkers['simpletree']},
|
11
|
+
'rexml' =>
|
12
|
+
{:builder => HTML5::TreeBuilders['rexml'],
|
13
|
+
:walker => HTML5::TreeWalkers['rexml']},
|
14
|
+
'hpricot' =>
|
15
|
+
{:builder => HTML5::TreeBuilders['hpricot'],
|
16
|
+
:walker => HTML5::TreeWalkers['hpricot']},
|
17
|
+
}
|
18
|
+
|
19
|
+
puts 'Testing tree walkers: ' + $tree_types_to_test.keys * ', '
|
20
|
+
|
21
|
+
class TestTreeWalkers < Test::Unit::TestCase
|
22
|
+
include HTML5::TestSupport
|
23
|
+
|
24
|
+
def concatenateCharacterTokens(tokens)
|
25
|
+
charactersToken = nil
|
26
|
+
for token in tokens
|
27
|
+
type = token[:type]
|
28
|
+
if [:Characters, :SpaceCharacters].include?(type)
|
29
|
+
if charactersToken == nil
|
30
|
+
charactersToken = {:type => :Characters, :data => token[:data]}
|
31
|
+
else
|
32
|
+
charactersToken[:data] += token[:data]
|
33
|
+
end
|
34
|
+
else
|
35
|
+
if charactersToken != nil
|
36
|
+
yield charactersToken
|
37
|
+
charactersToken = nil
|
38
|
+
end
|
39
|
+
yield token
|
40
|
+
end
|
41
|
+
end
|
42
|
+
yield charactersToken if charactersToken != nil
|
43
|
+
end
|
44
|
+
|
45
|
+
def convertTokens(tokens)
|
46
|
+
output = []
|
47
|
+
indent = 0
|
48
|
+
concatenateCharacterTokens(tokens) do |token|
|
49
|
+
case token[:type]
|
50
|
+
when :StartTag, :EmptyTag
|
51
|
+
output << "#{' '*indent}<#{token[:name]}>"
|
52
|
+
indent += 2
|
53
|
+
for name, value in token[:data].to_a.sort
|
54
|
+
next if name=='xmlns'
|
55
|
+
output << "#{' '*indent}#{name}=\"#{value}\""
|
56
|
+
end
|
57
|
+
indent -= 2 if token[:type] == :EmptyTag
|
58
|
+
when :EndTag
|
59
|
+
indent -= 2
|
60
|
+
when :Comment
|
61
|
+
output << "#{' '*indent}<!-- #{token[:data]} -->"
|
62
|
+
when :Doctype
|
63
|
+
if token[:name] and token[:name].any?
|
64
|
+
output << "#{' '*indent}<!DOCTYPE #{token[:name]}>"
|
65
|
+
else
|
66
|
+
output << "#{' '*indent}<!DOCTYPE >"
|
67
|
+
end
|
68
|
+
when :Characters, :SpaceCharacters
|
69
|
+
output << "#{' '*indent}\"#{token[:data]}\""
|
70
|
+
end
|
71
|
+
end
|
72
|
+
output.join("\n")
|
73
|
+
end
|
74
|
+
|
75
|
+
html5_test_files('tree-construction').each do |test_file|
|
76
|
+
|
77
|
+
test_name = File.basename(test_file).sub('.dat', '')
|
78
|
+
next if test_name == 'tests5' # TODO
|
79
|
+
|
80
|
+
TestData.new(test_file, %w(data errors document-fragment document)).
|
81
|
+
each_with_index do |(input, errors, inner_html, expected), index|
|
82
|
+
|
83
|
+
expected = expected.gsub("\n| ","\n")[2..-1]
|
84
|
+
|
85
|
+
$tree_types_to_test.each do |tree_name, tree_class|
|
86
|
+
|
87
|
+
define_method "test_#{test_name}_#{index}_#{tree_name}" do
|
88
|
+
|
89
|
+
parser = HTML5::HTMLParser.new(:tree => tree_class[:builder])
|
90
|
+
|
91
|
+
if inner_html
|
92
|
+
parser.parse_fragment(input, inner_html)
|
93
|
+
else
|
94
|
+
parser.parse(input)
|
95
|
+
end
|
96
|
+
|
97
|
+
document = parser.tree.get_document
|
98
|
+
|
99
|
+
begin
|
100
|
+
output = sortattrs(convertTokens(tree_class[:walker].new(document)))
|
101
|
+
expected = sortattrs(expected)
|
102
|
+
assert_equal expected, output, [
|
103
|
+
'', 'Input:', input,
|
104
|
+
'', 'Expected:', expected,
|
105
|
+
'', 'Recieved:', output
|
106
|
+
].join("\n")
|
107
|
+
rescue NotImplementedError
|
108
|
+
# Amnesty for those that confess...
|
109
|
+
end
|
110
|
+
end
|
111
|
+
end
|
112
|
+
end
|
113
|
+
end
|
114
|
+
|
115
|
+
def test_all_tokens
|
116
|
+
expected = [
|
117
|
+
{:data => [], :type => :StartTag, :name => 'html'},
|
118
|
+
{:data => [], :type => :StartTag, :name => 'head'},
|
119
|
+
{:data => [], :type => :EndTag, :name => 'head'},
|
120
|
+
{:data => [], :type => :StartTag, :name => 'body'},
|
121
|
+
{:data => [], :type => :EndTag, :name => 'body'},
|
122
|
+
{:data => [], :type => :EndTag, :name => 'html'}]
|
123
|
+
for treeName, tree_class in $tree_types_to_test
|
124
|
+
p = HTML5::HTMLParser.new(:tree => tree_class[:builder])
|
125
|
+
document = p.parse("<html></html>")
|
126
|
+
# document = tree_class.get(:adapter)(document)
|
127
|
+
output = tree_class[:walker].new(document)
|
128
|
+
expected.zip(output) do |expected_token, output_token|
|
129
|
+
assert_equal(expected_token, output_token)
|
130
|
+
end
|
131
|
+
end
|
132
|
+
end
|
133
|
+
|
134
|
+
|
135
|
+
end
|
@@ -0,0 +1,31 @@
|
|
1
|
+
#!/usr/bin/env ruby -wKU
|
2
|
+
|
3
|
+
require File.join(File.dirname(__FILE__), 'preamble')
|
4
|
+
|
5
|
+
require 'html5'
|
6
|
+
require 'html5/filters/validator'
|
7
|
+
|
8
|
+
class TestValidator < Test::Unit::TestCase
|
9
|
+
def run_validator_test(test)
|
10
|
+
p = HTML5::HTMLParser.new(:tokenizer => HTMLConformanceChecker)
|
11
|
+
p.parse(test['input'])
|
12
|
+
errorCodes = p.errors.collect{|e| e[1]}
|
13
|
+
if test.has_key?('fail-if')
|
14
|
+
assert !errorCodes.include?(test['fail-if'])
|
15
|
+
end
|
16
|
+
if test.has_key?('fail-unless')
|
17
|
+
assert errorCodes.include?(test['fail-unless'])
|
18
|
+
end
|
19
|
+
end
|
20
|
+
|
21
|
+
for filename in html5_test_files('validator')
|
22
|
+
tests = JSON.load(open(filename))
|
23
|
+
testName = File.basename(filename).sub(".test", "")
|
24
|
+
tests['tests'].each_with_index do |test, index|
|
25
|
+
define_method "test_#{testName}_#{index}" do
|
26
|
+
run_validator_test(test)
|
27
|
+
end
|
28
|
+
end
|
29
|
+
end
|
30
|
+
end
|
31
|
+
|
@@ -0,0 +1,63 @@
|
|
1
|
+
require 'html5/constants'
|
2
|
+
|
3
|
+
class TokenizerTestParser
|
4
|
+
def initialize(tokenizer)
|
5
|
+
@tokenizer = tokenizer
|
6
|
+
end
|
7
|
+
|
8
|
+
def parse
|
9
|
+
@outputTokens = []
|
10
|
+
|
11
|
+
debug = nil
|
12
|
+
for token in @tokenizer
|
13
|
+
debug = token.inspect if token[:type] == :ParseError
|
14
|
+
send(('process' + token[:type].to_s), token)
|
15
|
+
end
|
16
|
+
|
17
|
+
return @outputTokens
|
18
|
+
end
|
19
|
+
|
20
|
+
def processDoctype(token)
|
21
|
+
@outputTokens.push(["DOCTYPE", token[:name], token[:publicId],
|
22
|
+
token[:systemId], token[:correct]])
|
23
|
+
end
|
24
|
+
|
25
|
+
def processStartTag(token)
|
26
|
+
@outputTokens.push(["StartTag", token[:name], token[:data]])
|
27
|
+
end
|
28
|
+
|
29
|
+
def processEmptyTag(token)
|
30
|
+
if not HTML5::VOID_ELEMENTS.include? token[:name]
|
31
|
+
@outputTokens.push("ParseError")
|
32
|
+
end
|
33
|
+
@outputTokens.push(["StartTag", token[:name], token[:data]])
|
34
|
+
end
|
35
|
+
|
36
|
+
def processEndTag(token)
|
37
|
+
if token[:data].length > 0
|
38
|
+
self.processParseError(token)
|
39
|
+
end
|
40
|
+
@outputTokens.push(["EndTag", token[:name]])
|
41
|
+
end
|
42
|
+
|
43
|
+
def processComment(token)
|
44
|
+
@outputTokens.push(["Comment", token[:data]])
|
45
|
+
end
|
46
|
+
|
47
|
+
def processCharacters(token)
|
48
|
+
@outputTokens.push(["Character", token[:data]])
|
49
|
+
end
|
50
|
+
|
51
|
+
alias processSpaceCharacters processCharacters
|
52
|
+
|
53
|
+
def processCharacters(token)
|
54
|
+
@outputTokens.push(["Character", token[:data]])
|
55
|
+
end
|
56
|
+
|
57
|
+
def process_eof(token)
|
58
|
+
end
|
59
|
+
|
60
|
+
def processParseError(token)
|
61
|
+
@outputTokens.push("ParseError")
|
62
|
+
end
|
63
|
+
end
|
@@ -0,0 +1,781 @@
|
|
1
|
+
module FeedTools
|
2
|
+
# This is an implementation of a URI parser based on RFC 3986.
|
3
|
+
class URI
|
4
|
+
# Raised if something other than a uri is supplied.
|
5
|
+
class InvalidURIError < StandardError
|
6
|
+
end
|
7
|
+
# Raised if an invalid method option is supplied.
|
8
|
+
class InvalidOptionError < StandardError
|
9
|
+
end
|
10
|
+
|
11
|
+
# Returns a URI object based on the parsed string.
|
12
|
+
def self.parse(uri_string)
|
13
|
+
return nil if uri_string.nil?
|
14
|
+
|
15
|
+
# If a URI object is passed, just return itself.
|
16
|
+
return uri_string if uri_string.kind_of?(self)
|
17
|
+
|
18
|
+
uri_regex =
|
19
|
+
/^(([^:\/?#]+):)?(\/\/([^\/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?/
|
20
|
+
scan = uri_string.scan(uri_regex)
|
21
|
+
fragments = scan[0]
|
22
|
+
return nil if fragments.nil?
|
23
|
+
scheme = fragments[1]
|
24
|
+
authority = fragments[3]
|
25
|
+
path = fragments[4]
|
26
|
+
query = fragments[6]
|
27
|
+
fragment = fragments[8]
|
28
|
+
userinfo = nil
|
29
|
+
host = nil
|
30
|
+
port = nil
|
31
|
+
if authority != nil
|
32
|
+
userinfo = authority.scan(/^([^\[\]]*)@/).flatten[0]
|
33
|
+
host = authority.gsub(/^([^\[\]]*)@/, "").gsub(/:([^:@\[\]]*?)$/, "")
|
34
|
+
port = authority.scan(/:([^:@\[\]]*?)$/).flatten[0]
|
35
|
+
end
|
36
|
+
if port.nil? || port == ""
|
37
|
+
port = nil
|
38
|
+
end
|
39
|
+
|
40
|
+
# WARNING: Not standards-compliant, but follows the theme
|
41
|
+
# of Postel's law:
|
42
|
+
#
|
43
|
+
# Special exception for dealing with the retarded idea of the
|
44
|
+
# feed pseudo-protocol. Without this exception, the parser will read
|
45
|
+
# the URI as having a blank port number, instead of as having a second
|
46
|
+
# URI embedded within. This exception translates these broken URIs
|
47
|
+
# and instead treats the inner URI as opaque.
|
48
|
+
if scheme == "feed" && host == "http"
|
49
|
+
userinfo = nil
|
50
|
+
host = nil
|
51
|
+
port = nil
|
52
|
+
path = authority + path
|
53
|
+
end
|
54
|
+
|
55
|
+
return URI.new(scheme, userinfo, host, port, path, query, fragment)
|
56
|
+
end
|
57
|
+
|
58
|
+
# Converts a path to a file protocol URI. If the path supplied is
|
59
|
+
# relative, it will be returned as a relative URI. If the path supplied
|
60
|
+
# is actually a URI, it will return the parsed URI.
|
61
|
+
def self.convert_path(path)
|
62
|
+
return nil if path.nil?
|
63
|
+
|
64
|
+
converted_uri = path.strip
|
65
|
+
if converted_uri.length > 0 && converted_uri[0..0] == "/"
|
66
|
+
converted_uri = "file://" + converted_uri
|
67
|
+
end
|
68
|
+
if converted_uri.length > 0 &&
|
69
|
+
converted_uri.scan(/^[a-zA-Z]:[\\\/]/).size > 0
|
70
|
+
converted_uri = "file:///" + converted_uri
|
71
|
+
end
|
72
|
+
converted_uri.gsub!(/^file:\/*/i, "file:///")
|
73
|
+
if converted_uri =~ /^file:/i
|
74
|
+
# Adjust windows-style uris
|
75
|
+
converted_uri.gsub!(/^file:\/\/\/([a-zA-Z])\|/i, 'file:///\1:')
|
76
|
+
converted_uri.gsub!(/\\/, '/')
|
77
|
+
converted_uri = self.parse(converted_uri).normalize
|
78
|
+
else
|
79
|
+
converted_uri = self.parse(converted_uri)
|
80
|
+
end
|
81
|
+
|
82
|
+
return converted_uri
|
83
|
+
end
|
84
|
+
|
85
|
+
# Joins several uris together.
|
86
|
+
def self.join(*uris)
|
87
|
+
uri_objects = uris.collect do |uri|
|
88
|
+
uri.kind_of?(self) ? uri : self.parse(uri.to_s)
|
89
|
+
end
|
90
|
+
result = uri_objects.shift.dup
|
91
|
+
for uri in uri_objects
|
92
|
+
result.merge!(uri)
|
93
|
+
end
|
94
|
+
return result
|
95
|
+
end
|
96
|
+
|
97
|
+
# Correctly escapes a uri.
|
98
|
+
def self.escape(uri)
|
99
|
+
uri_object = uri.kind_of?(self) ? uri : self.parse(uri.to_s)
|
100
|
+
return URI.new(
|
101
|
+
uri_object.scheme,
|
102
|
+
uri_object.userinfo,
|
103
|
+
uri_object.host,
|
104
|
+
uri_object.specified_port,
|
105
|
+
self.normalize_escaping(uri_object.path),
|
106
|
+
self.normalize_escaping(uri_object.query),
|
107
|
+
self.normalize_escaping(uri_object.fragment)
|
108
|
+
).to_s
|
109
|
+
end
|
110
|
+
|
111
|
+
# Extracts uris from an arbitrary body of text.
|
112
|
+
def self.extract(text, options={})
|
113
|
+
defaults = {:base => nil, :parse => false}
|
114
|
+
options = defaults.merge(options)
|
115
|
+
raise InvalidOptionError unless (options.keys - defaults.keys).empty?
|
116
|
+
# This regular expression needs to be less forgiving or else it would
|
117
|
+
# match virtually all text. Which isn't exactly what we're going for.
|
118
|
+
extract_regex = /((([a-z\+]+):)[^ \n\<\>\"\\]+[\w\/])/
|
119
|
+
extracted_uris =
|
120
|
+
text.scan(extract_regex).collect { |match| match[0] }
|
121
|
+
sgml_extract_regex = /<[^>]+href=\"([^\"]+?)\"[^>]*>/
|
122
|
+
sgml_extracted_uris =
|
123
|
+
text.scan(sgml_extract_regex).collect { |match| match[0] }
|
124
|
+
extracted_uris.concat(sgml_extracted_uris - extracted_uris)
|
125
|
+
textile_extract_regex = /\".+?\":([^ ]+\/[^ ]+)[ \,\.\;\:\?\!\<\>\"]/i
|
126
|
+
textile_extracted_uris =
|
127
|
+
text.scan(textile_extract_regex).collect { |match| match[0] }
|
128
|
+
extracted_uris.concat(textile_extracted_uris - extracted_uris)
|
129
|
+
parsed_uris = []
|
130
|
+
base_uri = nil
|
131
|
+
if options[:base] != nil
|
132
|
+
base_uri = options[:base] if options[:base].kind_of?(self)
|
133
|
+
base_uri = self.parse(options[:base].to_s) if base_uri == nil
|
134
|
+
end
|
135
|
+
for uri_string in extracted_uris
|
136
|
+
begin
|
137
|
+
if base_uri == nil
|
138
|
+
parsed_uris << self.parse(uri_string)
|
139
|
+
else
|
140
|
+
parsed_uris << (base_uri + self.parse(uri_string))
|
141
|
+
end
|
142
|
+
rescue Exception
|
143
|
+
nil
|
144
|
+
end
|
145
|
+
end
|
146
|
+
parsed_uris.reject! do |uri|
|
147
|
+
(uri.scheme =~ /T\d+/ ||
|
148
|
+
uri.scheme == "xmlns" ||
|
149
|
+
uri.scheme == "xml" ||
|
150
|
+
uri.scheme == "thr" ||
|
151
|
+
uri.scheme == "this" ||
|
152
|
+
uri.scheme == "float" ||
|
153
|
+
uri.scheme == "user" ||
|
154
|
+
uri.scheme == "username" ||
|
155
|
+
uri.scheme == "out")
|
156
|
+
end
|
157
|
+
if options[:parse]
|
158
|
+
return parsed_uris
|
159
|
+
else
|
160
|
+
return parsed_uris.collect { |uri| uri.to_s }
|
161
|
+
end
|
162
|
+
end
|
163
|
+
|
164
|
+
# Creates a new uri object from component parts. Passing nil for
|
165
|
+
# any of these parameters is acceptable.
|
166
|
+
def initialize(scheme, userinfo, host, port, path, query, fragment)
|
167
|
+
assign_components(scheme, userinfo, host, port, path, query, fragment)
|
168
|
+
end
|
169
|
+
|
170
|
+
# Returns the scheme (protocol) for this URI.
|
171
|
+
def scheme
|
172
|
+
return nil if @scheme.nil? || @scheme.strip == ""
|
173
|
+
return @scheme
|
174
|
+
end
|
175
|
+
|
176
|
+
# Returns the username and password segment of this URI.
|
177
|
+
def userinfo
|
178
|
+
return @userinfo
|
179
|
+
end
|
180
|
+
|
181
|
+
# Returns the host for this URI.
|
182
|
+
def host
|
183
|
+
return @host
|
184
|
+
end
|
185
|
+
|
186
|
+
# Returns the authority segment of this URI.
|
187
|
+
def authority
|
188
|
+
if !defined?(@authority) || @authority.nil?
|
189
|
+
return nil if self.host.nil?
|
190
|
+
@authority = ""
|
191
|
+
if self.userinfo != nil
|
192
|
+
@authority << "#{self.userinfo}@"
|
193
|
+
end
|
194
|
+
@authority << self.host
|
195
|
+
if self.specified_port != nil
|
196
|
+
@authority << ":#{self.specified_port}"
|
197
|
+
end
|
198
|
+
end
|
199
|
+
return @authority
|
200
|
+
end
|
201
|
+
|
202
|
+
# Returns the user for this URI.
|
203
|
+
def user
|
204
|
+
if !defined?(@user) || @user.nil?
|
205
|
+
@user = nil
|
206
|
+
return @user if @userinfo.nil?
|
207
|
+
@user = @userinfo.strip.scan(/^(.*):/).flatten[0].strip
|
208
|
+
end
|
209
|
+
return @user
|
210
|
+
end
|
211
|
+
|
212
|
+
# Returns the password for this URI.
|
213
|
+
def password
|
214
|
+
if !defined?(@password) || @password.nil?
|
215
|
+
@password = nil
|
216
|
+
return @password if @userinfo.nil?
|
217
|
+
@password = @userinfo.strip.scan(/:(.*)$/).flatten[0].strip
|
218
|
+
end
|
219
|
+
return @password
|
220
|
+
end
|
221
|
+
|
222
|
+
# Returns an array of known ip-based schemes. These schemes typically
|
223
|
+
# use a similar URI form:
|
224
|
+
# //<user>:<password>@<host>:<port>/<url-path>
|
225
|
+
def self.ip_based_schemes
|
226
|
+
return self.scheme_mapping.keys
|
227
|
+
end
|
228
|
+
|
229
|
+
# Returns a hash of common IP-based schemes and their default port
|
230
|
+
# numbers. Adding new schemes to this hash, as necessary, will allow
|
231
|
+
# for better URI normalization.
|
232
|
+
def self.scheme_mapping
|
233
|
+
if !defined?(@protocol_mapping) || @protocol_mapping.nil?
|
234
|
+
@protocol_mapping = {
|
235
|
+
"http" => 80,
|
236
|
+
"https" => 443,
|
237
|
+
"ftp" => 21,
|
238
|
+
"tftp" => 69,
|
239
|
+
"ssh" => 22,
|
240
|
+
"svn+ssh" => 22,
|
241
|
+
"telnet" => 23,
|
242
|
+
"nntp" => 119,
|
243
|
+
"gopher" => 70,
|
244
|
+
"wais" => 210,
|
245
|
+
"prospero" => 1525
|
246
|
+
}
|
247
|
+
end
|
248
|
+
return @protocol_mapping
|
249
|
+
end
|
250
|
+
|
251
|
+
# Returns the port number for this URI. This method will normalize to the
|
252
|
+
# default port for the URI's scheme if the port isn't explicitly specified
|
253
|
+
# in the URI.
|
254
|
+
def port
|
255
|
+
if @port.to_i == 0
|
256
|
+
if self.scheme.nil?
|
257
|
+
@port = nil
|
258
|
+
else
|
259
|
+
@port = self.class.scheme_mapping[self.scheme.strip.downcase]
|
260
|
+
end
|
261
|
+
return @port
|
262
|
+
else
|
263
|
+
@port = @port.to_i
|
264
|
+
return @port
|
265
|
+
end
|
266
|
+
end
|
267
|
+
|
268
|
+
# Returns the port number that was actually specified in the URI string.
|
269
|
+
def specified_port
|
270
|
+
@specified_port = nil if !defined?(@specified_port)
|
271
|
+
return nil if @specified_port.nil?
|
272
|
+
port = @specified_port.to_s.to_i
|
273
|
+
if port == 0
|
274
|
+
return nil
|
275
|
+
else
|
276
|
+
return port
|
277
|
+
end
|
278
|
+
end
|
279
|
+
|
280
|
+
# Returns the path for this URI.
|
281
|
+
def path
|
282
|
+
return @path
|
283
|
+
end
|
284
|
+
|
285
|
+
# Returns the query string for this URI.
|
286
|
+
def query
|
287
|
+
return @query
|
288
|
+
end
|
289
|
+
|
290
|
+
# Returns the fragment for this URI.
|
291
|
+
def fragment
|
292
|
+
return @fragment
|
293
|
+
end
|
294
|
+
|
295
|
+
# Returns true if the URI uses an IP-based protocol.
|
296
|
+
def ip_based?
|
297
|
+
return false if self.scheme.nil?
|
298
|
+
return self.class.ip_based_schemes.include?(self.scheme.strip.downcase)
|
299
|
+
end
|
300
|
+
|
301
|
+
# Returns true if this URI is known to be relative.
|
302
|
+
def relative?
|
303
|
+
return self.scheme.nil?
|
304
|
+
end
|
305
|
+
|
306
|
+
# Returns true if this URI is known to be absolute.
|
307
|
+
def absolute?
|
308
|
+
return !relative?
|
309
|
+
end
|
310
|
+
|
311
|
+
# Joins two URIs together.
|
312
|
+
def +(uri)
|
313
|
+
if !uri.kind_of?(self.class)
|
314
|
+
uri = URI.parse(uri.to_s)
|
315
|
+
end
|
316
|
+
if uri.to_s == ""
|
317
|
+
return self.dup
|
318
|
+
end
|
319
|
+
|
320
|
+
joined_scheme = nil
|
321
|
+
joined_userinfo = nil
|
322
|
+
joined_host = nil
|
323
|
+
joined_port = nil
|
324
|
+
joined_path = nil
|
325
|
+
joined_query = nil
|
326
|
+
joined_fragment = nil
|
327
|
+
|
328
|
+
# Section 5.2.2 of RFC 3986
|
329
|
+
if uri.scheme != nil
|
330
|
+
joined_scheme = uri.scheme
|
331
|
+
joined_userinfo = uri.userinfo
|
332
|
+
joined_host = uri.host
|
333
|
+
joined_port = uri.specified_port
|
334
|
+
joined_path = self.class.normalize_path(uri.path)
|
335
|
+
joined_query = uri.query
|
336
|
+
else
|
337
|
+
if uri.authority != nil
|
338
|
+
joined_userinfo = uri.userinfo
|
339
|
+
joined_host = uri.host
|
340
|
+
joined_port = uri.specified_port
|
341
|
+
joined_path = self.class.normalize_path(uri.path)
|
342
|
+
joined_query = uri.query
|
343
|
+
else
|
344
|
+
if uri.path == nil || uri.path == ""
|
345
|
+
joined_path = self.path
|
346
|
+
if uri.query != nil
|
347
|
+
joined_query = uri.query
|
348
|
+
else
|
349
|
+
joined_query = self.query
|
350
|
+
end
|
351
|
+
else
|
352
|
+
if uri.path[0..0] == "/"
|
353
|
+
joined_path = self.class.normalize_path(uri.path)
|
354
|
+
else
|
355
|
+
base_path = self.path.nil? ? "" : self.path.dup
|
356
|
+
base_path = self.class.normalize_path(base_path)
|
357
|
+
base_path.gsub!(/\/[^\/]+$/, "/")
|
358
|
+
joined_path = self.class.normalize_path(base_path + uri.path)
|
359
|
+
end
|
360
|
+
joined_query = uri.query
|
361
|
+
end
|
362
|
+
joined_userinfo = self.userinfo
|
363
|
+
joined_host = self.host
|
364
|
+
joined_port = self.specified_port
|
365
|
+
end
|
366
|
+
joined_scheme = self.scheme
|
367
|
+
end
|
368
|
+
joined_fragment = uri.fragment
|
369
|
+
|
370
|
+
return URI.new(
|
371
|
+
joined_scheme,
|
372
|
+
joined_userinfo,
|
373
|
+
joined_host,
|
374
|
+
joined_port,
|
375
|
+
joined_path,
|
376
|
+
joined_query,
|
377
|
+
joined_fragment
|
378
|
+
)
|
379
|
+
end
|
380
|
+
|
381
|
+
# Merges two URIs together.
|
382
|
+
def merge(uri)
|
383
|
+
return self + uri
|
384
|
+
end
|
385
|
+
|
386
|
+
# Destructive form of merge.
|
387
|
+
def merge!(uri)
|
388
|
+
replace_self(self.merge(uri))
|
389
|
+
end
|
390
|
+
|
391
|
+
# Returns a normalized URI object.
|
392
|
+
#
|
393
|
+
# NOTE: This method does not attempt to conform to specifications. It
|
394
|
+
# exists largely to correct other people's failures to read the
|
395
|
+
# specifications, and also to deal with caching issues since several
|
396
|
+
# different URIs may represent the same resource and should not be
|
397
|
+
# cached multiple times.
|
398
|
+
def normalize
|
399
|
+
normalized_scheme = nil
|
400
|
+
normalized_scheme = self.scheme.strip.downcase if self.scheme != nil
|
401
|
+
normalized_scheme = "svn+ssh" if normalized_scheme == "ssh+svn"
|
402
|
+
if normalized_scheme == "feed"
|
403
|
+
if self.to_s =~ /^feed:\/*http:\/*/
|
404
|
+
return self.class.parse(
|
405
|
+
self.to_s.scan(/^feed:\/*(http:\/*.*)/).flatten[0]).normalize
|
406
|
+
end
|
407
|
+
end
|
408
|
+
normalized_userinfo = nil
|
409
|
+
normalized_userinfo = self.userinfo.strip if self.userinfo != nil
|
410
|
+
normalized_host = nil
|
411
|
+
normalized_host = self.host.strip.downcase if self.host != nil
|
412
|
+
if normalized_host != nil
|
413
|
+
begin
|
414
|
+
normalized_host = URI::IDNA.to_ascii(normalized_host)
|
415
|
+
rescue Exception
|
416
|
+
end
|
417
|
+
end
|
418
|
+
|
419
|
+
# Normalize IPv4 addresses that were generated with the stupid
|
420
|
+
# assumption that inet_addr() would be used to parse the IP address.
|
421
|
+
if normalized_host != nil && normalized_host.strip =~ /^\d+$/
|
422
|
+
# Decimal IPv4 address.
|
423
|
+
decimal = normalized_host.to_i
|
424
|
+
if decimal < (256 ** 4)
|
425
|
+
octets = [0,0,0,0]
|
426
|
+
octets[0] = decimal >> 24
|
427
|
+
decimal -= (octets[0] * (256 ** 3))
|
428
|
+
octets[1] = decimal >> 16
|
429
|
+
decimal -= (octets[1] * (256 ** 2))
|
430
|
+
octets[2] = decimal >> 8
|
431
|
+
decimal -= (octets[2] * (256 ** 1))
|
432
|
+
octets[3] = decimal
|
433
|
+
normalized_host = octets.join(".")
|
434
|
+
end
|
435
|
+
elsif (normalized_host != nil && normalized_host.strip =~
|
436
|
+
/^0+[0-7]{3}.0+[0-7]{3}.0+[0-7]{3}.0+[0-7]{3}$/)
|
437
|
+
# Octal IPv4 address.
|
438
|
+
octet_strings = normalized_host.split('.')
|
439
|
+
octets = []
|
440
|
+
octet_strings.each do |octet_string|
|
441
|
+
decimal = octet_string.to_i(8)
|
442
|
+
octets << decimal
|
443
|
+
end
|
444
|
+
normalized_host = octets.join(".")
|
445
|
+
elsif (normalized_host != nil && normalized_host.strip =~
|
446
|
+
/^0x[0-9a-f]{2}.0x[0-9a-f]{2}.0x[0-9a-f]{2}.0x[0-9a-f]{2}$/i)
|
447
|
+
# Hexidecimal IPv4 address.
|
448
|
+
octet_strings = normalized_host.split('.')
|
449
|
+
octets = []
|
450
|
+
octet_strings.each do |octet_string|
|
451
|
+
decimal = octet_string[2...4].to_i(16)
|
452
|
+
octets << decimal
|
453
|
+
end
|
454
|
+
normalized_host = octets.join(".")
|
455
|
+
end
|
456
|
+
normalized_port = self.port
|
457
|
+
if self.class.scheme_mapping[normalized_scheme] == normalized_port
|
458
|
+
normalized_port = nil
|
459
|
+
end
|
460
|
+
normalized_path = nil
|
461
|
+
normalized_path = self.path.strip if self.path != nil
|
462
|
+
if normalized_scheme != nil && normalized_host == nil
|
463
|
+
if self.class.ip_based_schemes.include?(normalized_scheme) &&
|
464
|
+
normalized_path =~ /[\w\.]+/
|
465
|
+
normalized_host = normalized_path
|
466
|
+
normalized_path = nil
|
467
|
+
unless normalized_host =~ /\./
|
468
|
+
normalized_host = normalized_host + ".com"
|
469
|
+
end
|
470
|
+
end
|
471
|
+
end
|
472
|
+
if normalized_path == nil &&
|
473
|
+
normalized_scheme != nil &&
|
474
|
+
normalized_host != nil
|
475
|
+
normalized_path = "/"
|
476
|
+
end
|
477
|
+
if normalized_path != nil
|
478
|
+
normalized_path = self.class.normalize_path(normalized_path)
|
479
|
+
normalized_path = self.class.normalize_escaping(normalized_path)
|
480
|
+
end
|
481
|
+
if normalized_path == ""
|
482
|
+
if ["http", "https", "ftp", "tftp"].include?(normalized_scheme)
|
483
|
+
normalized_path = "/"
|
484
|
+
end
|
485
|
+
end
|
486
|
+
normalized_path.gsub!(/%3B/, ";") if normalized_path != nil
|
487
|
+
normalized_path.gsub!(/%3A/, ":") if normalized_path != nil
|
488
|
+
normalized_path.gsub!(/%40/, "@") if normalized_path != nil
|
489
|
+
normalized_path.gsub!(/%2B/, "+") if normalized_path != nil
|
490
|
+
|
491
|
+
normalized_query = nil
|
492
|
+
normalized_query = self.query.strip if self.query != nil
|
493
|
+
normalized_query = self.class.normalize_escaping(normalized_query)
|
494
|
+
normalized_query.gsub!(/%3D/, "=") if normalized_query != nil
|
495
|
+
normalized_query.gsub!(/%26/, "&") if normalized_query != nil
|
496
|
+
normalized_query.gsub!(/%2B/, "+") if normalized_query != nil
|
497
|
+
|
498
|
+
normalized_fragment = nil
|
499
|
+
normalized_fragment = self.fragment.strip if self.fragment != nil
|
500
|
+
normalized_fragment = self.class.normalize_escaping(normalized_fragment)
|
501
|
+
return URI.new(
|
502
|
+
normalized_scheme,
|
503
|
+
normalized_userinfo,
|
504
|
+
normalized_host,
|
505
|
+
normalized_port,
|
506
|
+
normalized_path,
|
507
|
+
normalized_query,
|
508
|
+
normalized_fragment
|
509
|
+
)
|
510
|
+
end
|
511
|
+
|
512
|
+
# Destructively normalizes this URI object.
|
513
|
+
def normalize!
|
514
|
+
replace_self(self.normalize)
|
515
|
+
end
|
516
|
+
|
517
|
+
# Creates a URI suitable for display to users. If semantic attacks are
|
518
|
+
# likely, the application should try to detect these and warn the user.
|
519
|
+
# See RFC 3986 section 7.6 for more information.
|
520
|
+
def display_uri
|
521
|
+
display_uri = self.normalize
|
522
|
+
begin
|
523
|
+
display_uri.instance_variable_set("@host",
|
524
|
+
URI::IDNA.to_unicode(display_uri.host))
|
525
|
+
rescue Exception
|
526
|
+
end
|
527
|
+
return display_uri
|
528
|
+
end
|
529
|
+
|
530
|
+
# Returns true if the URI objects are equal. This method normalizes
|
531
|
+
# both URIs before doing the comparison, and allows comparison against
|
532
|
+
# strings.
|
533
|
+
def ===(uri)
|
534
|
+
uri_string = nil
|
535
|
+
if uri.respond_to?(:normalize)
|
536
|
+
uri_string = uri.normalize.to_s
|
537
|
+
else
|
538
|
+
begin
|
539
|
+
uri_string = URI.parse(uri.to_s).normalize.to_s
|
540
|
+
rescue Exception
|
541
|
+
return false
|
542
|
+
end
|
543
|
+
end
|
544
|
+
return self.normalize.to_s == uri_string
|
545
|
+
end
|
546
|
+
|
547
|
+
# Returns true if the URI objects are equal. This method normalizes
|
548
|
+
# both URIs before doing the comparison.
|
549
|
+
def ==(uri)
|
550
|
+
return false unless uri.kind_of?(self.class)
|
551
|
+
return self.normalize.to_s == uri.normalize.to_s
|
552
|
+
end
|
553
|
+
|
554
|
+
# Returns true if the URI objects are equal. This method does NOT
|
555
|
+
# normalize either URI before doing the comparison.
|
556
|
+
def eql?(uri)
|
557
|
+
return false unless uri.kind_of?(self.class)
|
558
|
+
return self.to_s == uri.to_s
|
559
|
+
end
|
560
|
+
|
561
|
+
# Clones the URI object.
|
562
|
+
def dup
|
563
|
+
duplicated_scheme = nil
|
564
|
+
duplicated_scheme = self.scheme.dup if self.scheme != nil
|
565
|
+
duplicated_userinfo = nil
|
566
|
+
duplicated_userinfo = self.userinfo.dup if self.userinfo != nil
|
567
|
+
duplicated_host = nil
|
568
|
+
duplicated_host = self.host.dup if self.host != nil
|
569
|
+
duplicated_port = self.port
|
570
|
+
duplicated_path = nil
|
571
|
+
duplicated_path = self.path.dup if self.path != nil
|
572
|
+
duplicated_query = nil
|
573
|
+
duplicated_query = self.query.dup if self.query != nil
|
574
|
+
duplicated_fragment = nil
|
575
|
+
duplicated_fragment = self.fragment.dup if self.fragment != nil
|
576
|
+
duplicated_uri = URI.new(
|
577
|
+
duplicated_scheme,
|
578
|
+
duplicated_userinfo,
|
579
|
+
duplicated_host,
|
580
|
+
duplicated_port,
|
581
|
+
duplicated_path,
|
582
|
+
duplicated_query,
|
583
|
+
duplicated_fragment
|
584
|
+
)
|
585
|
+
@specified_port = nil if !defined?(@specified_port)
|
586
|
+
duplicated_uri.instance_variable_set("@specified_port", @specified_port)
|
587
|
+
return duplicated_uri
|
588
|
+
end
|
589
|
+
|
590
|
+
# Returns the assembled URI as a string.
|
591
|
+
def to_s
|
592
|
+
uri_string = ""
|
593
|
+
if self.scheme != nil
|
594
|
+
uri_string << "#{self.scheme}:"
|
595
|
+
end
|
596
|
+
if self.authority != nil
|
597
|
+
uri_string << "//#{self.authority}"
|
598
|
+
end
|
599
|
+
if self.path != nil
|
600
|
+
uri_string << self.path
|
601
|
+
end
|
602
|
+
if self.query != nil
|
603
|
+
uri_string << "?#{self.query}"
|
604
|
+
end
|
605
|
+
if self.fragment != nil
|
606
|
+
uri_string << "##{self.fragment}"
|
607
|
+
end
|
608
|
+
return uri_string
|
609
|
+
end
|
610
|
+
|
611
|
+
# Returns a string representation of the URI object's state.
|
612
|
+
def inspect
|
613
|
+
sprintf("#<%s:%#0x URL:%s>", self.class.to_s, self.object_id, self.to_s)
|
614
|
+
end
|
615
|
+
|
616
|
+
# This module handles internationalized domain names. When Ruby has an
|
617
|
+
# implementation of nameprep, stringprep, punycode, etc, this
|
618
|
+
# module should contain an actual implementation of IDNA instead of
|
619
|
+
# returning nil if libidn can't be used.
|
620
|
+
module IDNA
|
621
|
+
# Returns the ascii representation of the label.
|
622
|
+
def self.to_ascii(label)
|
623
|
+
return nil if label.nil?
|
624
|
+
if self.use_libidn?
|
625
|
+
return IDN::Idna.toASCII(label)
|
626
|
+
else
|
627
|
+
raise NotImplementedError,
|
628
|
+
"There is no available pure-ruby implementation. " +
|
629
|
+
"Install libidn bindings."
|
630
|
+
end
|
631
|
+
end
|
632
|
+
|
633
|
+
# Returns the unicode representation of the label.
|
634
|
+
def self.to_unicode(label)
|
635
|
+
return nil if label.nil?
|
636
|
+
if self.use_libidn?
|
637
|
+
return IDN::Idna.toUnicode(label)
|
638
|
+
else
|
639
|
+
raise NotImplementedError,
|
640
|
+
"There is no available pure-ruby implementation. " +
|
641
|
+
"Install libidn bindings."
|
642
|
+
end
|
643
|
+
end
|
644
|
+
|
645
|
+
private
|
646
|
+
# Determines if the libidn bindings are available and able to be used.
|
647
|
+
def self.use_libidn?
|
648
|
+
if !defined?(@use_libidn) || @use_libidn.nil?
|
649
|
+
begin
|
650
|
+
require 'rubygems'
|
651
|
+
rescue LoadError
|
652
|
+
end
|
653
|
+
begin
|
654
|
+
require 'idn'
|
655
|
+
rescue LoadError
|
656
|
+
end
|
657
|
+
@use_libidn = !!(defined?(IDN::Idna))
|
658
|
+
end
|
659
|
+
return @use_libidn
|
660
|
+
end
|
661
|
+
end
|
662
|
+
|
663
|
+
private
|
664
|
+
# Resolves paths to their simplest form.
|
665
|
+
def self.normalize_path(path)
|
666
|
+
return nil if path.nil?
|
667
|
+
normalized_path = path.dup
|
668
|
+
previous_state = normalized_path.dup
|
669
|
+
begin
|
670
|
+
previous_state = normalized_path.dup
|
671
|
+
normalized_path.gsub!(/\/\.\//, "/")
|
672
|
+
normalized_path.gsub!(/\/\.$/, "/")
|
673
|
+
parent = normalized_path.scan(/\/([^\/]+)\/\.\.\//).flatten[0]
|
674
|
+
if parent != "." && parent != ".."
|
675
|
+
normalized_path.gsub!(/\/#{parent}\/\.\.\//, "/")
|
676
|
+
end
|
677
|
+
parent = normalized_path.scan(/\/([^\/]+)\/\.\.$/).flatten[0]
|
678
|
+
if parent != "." && parent != ".."
|
679
|
+
normalized_path.gsub!(/\/#{parent}\/\.\.$/, "/")
|
680
|
+
end
|
681
|
+
normalized_path.gsub!(/^\.\.?\/?/, "")
|
682
|
+
normalized_path.gsub!(/^\/\.\.?\//, "/")
|
683
|
+
end until previous_state == normalized_path
|
684
|
+
return normalized_path
|
685
|
+
end
|
686
|
+
|
687
|
+
# Normalizes percent escaping of characters
|
688
|
+
def self.normalize_escaping(escaped_section)
|
689
|
+
return nil if escaped_section.nil?
|
690
|
+
normalized_section = escaped_section.dup
|
691
|
+
normalized_section.gsub!(/%[0-9a-f]{2}/i) do |sequence|
|
692
|
+
sequence[1..3].to_i(16).chr
|
693
|
+
end
|
694
|
+
if URI::IDNA.send(:use_libidn?)
|
695
|
+
normalized_section =
|
696
|
+
IDN::Stringprep.nfkc_normalize(normalized_section)
|
697
|
+
end
|
698
|
+
new_section = ""
|
699
|
+
for index in 0...normalized_section.size
|
700
|
+
if self.unreserved?(normalized_section[index]) ||
|
701
|
+
normalized_section[index] == '/'[0]
|
702
|
+
new_section << normalized_section[index..index]
|
703
|
+
else
|
704
|
+
new_section << ("%" + normalized_section[index].to_s(16).upcase)
|
705
|
+
end
|
706
|
+
end
|
707
|
+
normalized_section = new_section
|
708
|
+
return normalized_section
|
709
|
+
end
|
710
|
+
|
711
|
+
# Returns true if the specified character is unreserved.
|
712
|
+
def self.unreserved?(character)
|
713
|
+
character_string = nil
|
714
|
+
character_string = character.chr if character.respond_to?(:chr)
|
715
|
+
character_string = character[0..0] if character.kind_of?(String)
|
716
|
+
return self.unreserved.include?(character_string)
|
717
|
+
end
|
718
|
+
|
719
|
+
# Returns a list of unreserved characters.
|
720
|
+
def self.unreserved
|
721
|
+
if !defined?(@unreserved) || @unreserved.nil?
|
722
|
+
@unreserved = ["-", ".", "_", "~"]
|
723
|
+
for c in "a".."z"
|
724
|
+
@unreserved << c
|
725
|
+
@unreserved << c.upcase
|
726
|
+
end
|
727
|
+
for c in "0".."9"
|
728
|
+
@unreserved << c
|
729
|
+
end
|
730
|
+
@unreserved.sort!
|
731
|
+
end
|
732
|
+
return @unreserved
|
733
|
+
end
|
734
|
+
|
735
|
+
# Assigns the specified components to the appropriate instance variables.
|
736
|
+
# Used in destructive operations to avoid code repetition.
|
737
|
+
def assign_components(scheme, userinfo, host, port, path, query, fragment)
|
738
|
+
if scheme == nil && userinfo == nil && host == nil && port == nil &&
|
739
|
+
path == nil && query == nil && fragment == nil
|
740
|
+
raise InvalidURIError, "All parameters were nil."
|
741
|
+
end
|
742
|
+
@scheme = scheme
|
743
|
+
@userinfo = userinfo
|
744
|
+
@host = host
|
745
|
+
@specified_port = port.to_s
|
746
|
+
@port = port
|
747
|
+
@port = @port.to_s if @port.kind_of?(Fixnum)
|
748
|
+
if @port != nil && !(@port =~ /^\d+$/)
|
749
|
+
raise InvalidURIError,
|
750
|
+
"Invalid port number: #{@port.inspect}"
|
751
|
+
end
|
752
|
+
@port = @port.to_i
|
753
|
+
@port = nil if @port == 0
|
754
|
+
@path = path
|
755
|
+
@query = query
|
756
|
+
@fragment = fragment
|
757
|
+
if @scheme != nil && @host == "" && @path == ""
|
758
|
+
raise InvalidURIError,
|
759
|
+
"Absolute URI missing hierarchical segment."
|
760
|
+
end
|
761
|
+
end
|
762
|
+
|
763
|
+
# Replaces the internal state of self with the specified URI's state.
|
764
|
+
# Used in destructive operations to avoid code repetition.
|
765
|
+
def replace_self(uri)
|
766
|
+
@authority = nil
|
767
|
+
@user = nil
|
768
|
+
@password = nil
|
769
|
+
|
770
|
+
@scheme = uri.scheme
|
771
|
+
@userinfo = uri.userinfo
|
772
|
+
@host = uri.host
|
773
|
+
@specified_port = uri.instance_variable_get("@specified_port")
|
774
|
+
@port = @specified_port.to_s.to_i
|
775
|
+
@path = uri.path
|
776
|
+
@query = uri.query
|
777
|
+
@fragment = uri.fragment
|
778
|
+
return self
|
779
|
+
end
|
780
|
+
end
|
781
|
+
end
|