rfeedparser 0.9.8 → 0.9.9
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/rfeedparser.rb +170 -3345
- data/lib/rfeedparser/aliases.rb +432 -0
- data/lib/rfeedparser/better_attributelist.rb +41 -0
- data/lib/rfeedparser/better_sgmlparser.rb +264 -0
- data/lib/rfeedparser/encoding_helpers.rb +257 -0
- data/lib/rfeedparser/feedparserdict.rb +93 -0
- data/lib/rfeedparser/forgiving_uri.rb +93 -0
- data/lib/rfeedparser/markup_helpers.rb +73 -0
- data/lib/rfeedparser/parser_mixin.rb +1235 -0
- data/lib/rfeedparser/parsers.rb +177 -0
- data/lib/rfeedparser/scrub.rb +207 -0
- data/lib/rfeedparser/time_helpers.rb +408 -0
- data/tests/rfeedparsertest.rb +3 -1
- metadata +3271 -3250
@@ -0,0 +1,93 @@
|
|
1
|
+
#!/usr/bin/ruby
|
2
|
+
module FeedParserUtilities
|
3
|
+
class FeedParserDict < Hash
|
4
|
+
=begin
|
5
|
+
The naming of a certain common attribute (such as, "When was the last
|
6
|
+
time this feed was updated?") can have many different names depending
|
7
|
+
on the type of feed we are handling. This class allows us to satisfy
|
8
|
+
the expectations of both the developer who has prior knowledge of the
|
9
|
+
feed type as well as the developer who wants a consistent application
|
10
|
+
interface.
|
11
|
+
|
12
|
+
@@keymap is a Hash that contains information on what a certain
|
13
|
+
attribute names "really are" in each kind of feed. It does this by
|
14
|
+
providing a common name that will map to any feed type in the keys,
|
15
|
+
with possible "correct" attributes in the its values. the #[] and #[]=
|
16
|
+
methods check with keymaps to see what attribute the developer "really
|
17
|
+
means" if they've asked for one which happens to be in @@keymap's keys.
|
18
|
+
=end
|
19
|
+
@@keymap = {'channel' => 'feed',
|
20
|
+
'items' => 'entries',
|
21
|
+
'guid' => 'id',
|
22
|
+
'date' => 'updated',
|
23
|
+
'date_parsed' => 'updated_parsed',
|
24
|
+
'description' => ['subtitle', 'summary'],
|
25
|
+
'url' => ['href'],
|
26
|
+
'modified' => 'updated',
|
27
|
+
'modified_parsed' => 'updated_parsed',
|
28
|
+
'issued' => 'published',
|
29
|
+
'issued_parsed' => 'published_parsed',
|
30
|
+
'copyright' => 'rights',
|
31
|
+
'copyright_detail' => 'rights_detail',
|
32
|
+
'tagline' => 'subtitle',
|
33
|
+
'tagline_detail' => 'subtitle_detail'}
|
34
|
+
|
35
|
+
def entries # Apparently, Hash has an entries method! That blew a good 3 hours or more of my time
|
36
|
+
return self['entries']
|
37
|
+
end
|
38
|
+
|
39
|
+
# We could include the [] rewrite in new using Hash.new's fancy pants block thing
|
40
|
+
# but we'd still have to overwrite []= and such.
|
41
|
+
# I'm going to make it easy to turn lists of pairs into FeedParserDicts's though.
|
42
|
+
def initialize(pairs=nil)
|
43
|
+
if pairs.class == Array and pairs[0].class == Array and pairs[0].length == 2
|
44
|
+
pairs.each do |l|
|
45
|
+
k,v = l
|
46
|
+
self[k] = v
|
47
|
+
end
|
48
|
+
elsif pairs.class == Hash
|
49
|
+
self.merge!(pairs)
|
50
|
+
end
|
51
|
+
end
|
52
|
+
|
53
|
+
def [](key)
|
54
|
+
if key == 'category'
|
55
|
+
return self['tags'][0]['term']
|
56
|
+
end
|
57
|
+
if key == 'categories'
|
58
|
+
return self['tags'].collect{|tag| [tag['scheme'],tag['term']]}
|
59
|
+
end
|
60
|
+
realkey = @@keymap[key] || key
|
61
|
+
if realkey.class == Array
|
62
|
+
realkey.each{ |key| return self[key] if has_key?key }
|
63
|
+
end
|
64
|
+
# Note that the original key is preferred over the realkey we (might
|
65
|
+
# have) found in @@keymap
|
66
|
+
if has_key?(key)
|
67
|
+
return super(key)
|
68
|
+
end
|
69
|
+
return super(realkey)
|
70
|
+
end
|
71
|
+
|
72
|
+
def []=(key,value)
|
73
|
+
if @@keymap.key?key
|
74
|
+
key = @@keymap[key]
|
75
|
+
if key.class == Array
|
76
|
+
key = key[0]
|
77
|
+
end
|
78
|
+
end
|
79
|
+
super(key,value)
|
80
|
+
end
|
81
|
+
|
82
|
+
def method_missing(msym, *args)
|
83
|
+
methodname = msym.to_s
|
84
|
+
if methodname[-1] == '='
|
85
|
+
return self[methodname[0..-2]] = args[0]
|
86
|
+
elsif methodname[-1] != '!' and methodname[-1] != '?' and methodname[0] != "_" # FIXME implement with private?
|
87
|
+
return self[methodname]
|
88
|
+
else
|
89
|
+
raise NoMethodError, "whoops, we don't know about the attribute or method called `#{methodname}' for #{self}:#{self.class}"
|
90
|
+
end
|
91
|
+
end
|
92
|
+
end
|
93
|
+
end
|
@@ -0,0 +1,93 @@
|
|
1
|
+
#!/usr/bin/ruby
|
2
|
+
|
3
|
+
module URI
|
4
|
+
# NOTE I wish I didn't have to open this module up,but I cannot find a
|
5
|
+
# better way of accessing all of the instance methods of the URI module. I \
|
6
|
+
# may just be an idiot.
|
7
|
+
def self.split(uri)
|
8
|
+
case uri
|
9
|
+
when ''
|
10
|
+
# null uri
|
11
|
+
|
12
|
+
when ABS_URI
|
13
|
+
scheme, opaque, userinfo, host, port,
|
14
|
+
registry, path, query, fragment = $~[1..-1]
|
15
|
+
|
16
|
+
# URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
|
17
|
+
|
18
|
+
# absoluteURI = scheme ":" ( hier_part | opaque_part )
|
19
|
+
# hier_part = ( net_path | abs_path ) [ "?" query ]
|
20
|
+
# opaque_part = uric_no_slash *uric
|
21
|
+
|
22
|
+
# abs_path = "/" path_segments
|
23
|
+
# net_path = "//" authority [ abs_path ]
|
24
|
+
|
25
|
+
# authority = server | reg_name
|
26
|
+
# server = [ [ userinfo "@" ] hostport ]
|
27
|
+
|
28
|
+
if !scheme
|
29
|
+
raise InvalidURIError,
|
30
|
+
"bad URI(absolute but no scheme): #{uri}"
|
31
|
+
end
|
32
|
+
if !opaque && (!path && (!host && !registry))
|
33
|
+
raise InvalidURIError,
|
34
|
+
"bad URI(absolute but no path): #{uri}"
|
35
|
+
end
|
36
|
+
|
37
|
+
when REL_URI
|
38
|
+
scheme = nil
|
39
|
+
opaque = nil
|
40
|
+
|
41
|
+
userinfo, host, port, registry,
|
42
|
+
rel_segment, abs_path, query, fragment = $~[1..-1]
|
43
|
+
if rel_segment && abs_path
|
44
|
+
path = rel_segment + abs_path
|
45
|
+
elsif rel_segment
|
46
|
+
path = rel_segment
|
47
|
+
elsif abs_path
|
48
|
+
path = abs_path
|
49
|
+
end
|
50
|
+
|
51
|
+
# URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
|
52
|
+
|
53
|
+
# relativeURI = ( net_path | abs_path | rel_path ) [ "?" query ]
|
54
|
+
|
55
|
+
# net_path = "//" authority [ abs_path ]
|
56
|
+
# abs_path = "/" path_segments
|
57
|
+
# rel_path = rel_segment [ abs_path ]
|
58
|
+
|
59
|
+
# authority = server | reg_name
|
60
|
+
# server = [ [ userinfo "@" ] hostport ]
|
61
|
+
|
62
|
+
else
|
63
|
+
# NOTE this is the only part of the code that differs from the "clean"
|
64
|
+
# URI module.
|
65
|
+
return [nil,nil,uri,nil,nil,nil,nil,nil,nil]
|
66
|
+
end
|
67
|
+
|
68
|
+
path = '' if !path && !opaque # (see RFC2396 Section 5.2)
|
69
|
+
ret = [
|
70
|
+
scheme,
|
71
|
+
userinfo, host, port, # X
|
72
|
+
registry, # X
|
73
|
+
path, # Y
|
74
|
+
opaque, # Y
|
75
|
+
query,
|
76
|
+
fragment
|
77
|
+
]
|
78
|
+
return ret
|
79
|
+
end
|
80
|
+
end
|
81
|
+
|
82
|
+
def urljoin(base, uri)
|
83
|
+
urifixer = /^([A-Za-z][A-Za-z0-9+-.]*:\/\/)(\/*)(.*?)/u
|
84
|
+
uri = uri.sub(urifixer, '\1\3')
|
85
|
+
begin
|
86
|
+
return URI.join(base, uri).to_s
|
87
|
+
rescue URI::BadURIError => e
|
88
|
+
if URI.parse(base).relative?
|
89
|
+
return URI::parse(uri).to_s
|
90
|
+
end
|
91
|
+
end
|
92
|
+
end
|
93
|
+
|
@@ -0,0 +1,73 @@
|
|
1
|
+
#!/usr/bin/ruby
|
2
|
+
module FeedParserUtilities
|
3
|
+
#FIXME we need to find a better place for this method
|
4
|
+
def stripDoctype(data)
|
5
|
+
=begin
|
6
|
+
Strips DOCTYPE from XML document, returns (rss_version, stripped_data)
|
7
|
+
|
8
|
+
rss_version may be 'rss091n' or None
|
9
|
+
stripped_data is the same XML document, minus the DOCTYPE
|
10
|
+
=end
|
11
|
+
entity_pattern = /<!ENTITY(.*?)>/m # m is for Regexp::MULTILINE
|
12
|
+
data = data.gsub(entity_pattern,'')
|
13
|
+
|
14
|
+
doctype_pattern = /<!DOCTYPE(.*?)>/m
|
15
|
+
doctype_results = data.scan(doctype_pattern)
|
16
|
+
if doctype_results and doctype_results[0]
|
17
|
+
doctype = doctype_results[0][0]
|
18
|
+
else
|
19
|
+
doctype = ''
|
20
|
+
end
|
21
|
+
|
22
|
+
if /netscape/ =~ doctype.downcase
|
23
|
+
version = 'rss091n'
|
24
|
+
else
|
25
|
+
version = nil
|
26
|
+
end
|
27
|
+
data = data.sub(doctype_pattern, '')
|
28
|
+
return version, data
|
29
|
+
end
|
30
|
+
|
31
|
+
def resolveRelativeURIs(htmlSource, baseURI, encoding)
|
32
|
+
$stderr << "entering resolveRelativeURIs\n" if $debug # FIXME write a decent logger
|
33
|
+
relative_uris = [ ['a','href'],
|
34
|
+
['applet','codebase'],
|
35
|
+
['area','href'],
|
36
|
+
['blockquote','cite'],
|
37
|
+
['body','background'],
|
38
|
+
['del','cite'],
|
39
|
+
['form','action'],
|
40
|
+
['frame','longdesc'],
|
41
|
+
['frame','src'],
|
42
|
+
['iframe','longdesc'],
|
43
|
+
['iframe','src'],
|
44
|
+
['head','profile'],
|
45
|
+
['img','longdesc'],
|
46
|
+
['img','src'],
|
47
|
+
['img','usemap'],
|
48
|
+
['input','src'],
|
49
|
+
['input','usemap'],
|
50
|
+
['ins','cite'],
|
51
|
+
['link','href'],
|
52
|
+
['object','classid'],
|
53
|
+
['object','codebase'],
|
54
|
+
['object','data'],
|
55
|
+
['object','usemap'],
|
56
|
+
['q','cite'],
|
57
|
+
['script','src'],
|
58
|
+
]
|
59
|
+
h = Hpricot(htmlSource)
|
60
|
+
relative_uris.each do |l|
|
61
|
+
ename, eattr = l
|
62
|
+
h.search(ename).each do |elem|
|
63
|
+
euri = elem.attributes[eattr]
|
64
|
+
if euri and not euri.empty? and URI.parse(URI.encode(euri)).relative?
|
65
|
+
elem.attributes[eattr] = urljoin(baseURI, euri)
|
66
|
+
end
|
67
|
+
end
|
68
|
+
end
|
69
|
+
return h.to_html
|
70
|
+
end
|
71
|
+
end
|
72
|
+
|
73
|
+
|
@@ -0,0 +1,1235 @@
|
|
1
|
+
#!/usr/bin/ruby
|
2
|
+
module FeedParserMixin
|
3
|
+
attr_accessor :feeddata, :version, :namespacesInUse, :date_handlers
|
4
|
+
|
5
|
+
def startup(baseuri=nil, baselang=nil, encoding='utf-8')
|
6
|
+
$stderr << "initializing FeedParser\n" if $debug
|
7
|
+
|
8
|
+
@namespaces = {'' => '',
|
9
|
+
'http://backend.userland.com/rss' => '',
|
10
|
+
'http://blogs.law.harvard.edu/tech/rss' => '',
|
11
|
+
'http://purl.org/rss/1.0/' => '',
|
12
|
+
'http://my.netscape.com/rdf/simple/0.9/' => '',
|
13
|
+
'http://example.com/newformat#' => '',
|
14
|
+
'http://example.com/necho' => '',
|
15
|
+
'http://purl.org/echo/' => '',
|
16
|
+
'uri/of/echo/namespace#' => '',
|
17
|
+
'http://purl.org/pie/' => '',
|
18
|
+
'http://purl.org/atom/ns#' => '',
|
19
|
+
'http://www.w3.org/2005/Atom' => '',
|
20
|
+
'http://purl.org/rss/1.0/modules/rss091#' => '',
|
21
|
+
'http://webns.net/mvcb/' => 'admin',
|
22
|
+
'http://purl.org/rss/1.0/modules/aggregation/' => 'ag',
|
23
|
+
'http://purl.org/rss/1.0/modules/annotate/' => 'annotate',
|
24
|
+
'http://media.tangent.org/rss/1.0/' => 'audio',
|
25
|
+
'http://backend.userland.com/blogChannelModule' => 'blogChannel',
|
26
|
+
'http://web.resource.org/cc/' => 'cc',
|
27
|
+
'http://backend.userland.com/creativeCommonsRssModule' => 'creativeCommons',
|
28
|
+
'http://purl.org/rss/1.0/modules/company' => 'co',
|
29
|
+
'http://purl.org/rss/1.0/modules/content/' => 'content',
|
30
|
+
'http://my.theinfo.org/changed/1.0/rss/' => 'cp',
|
31
|
+
'http://purl.org/dc/elements/1.1/' => 'dc',
|
32
|
+
'http://purl.org/dc/terms/' => 'dcterms',
|
33
|
+
'http://purl.org/rss/1.0/modules/email/' => 'email',
|
34
|
+
'http://purl.org/rss/1.0/modules/event/' => 'ev',
|
35
|
+
'http://rssnamespace.org/feedburner/ext/1.0' => 'feedburner',
|
36
|
+
'http://freshmeat.net/rss/fm/' => 'fm',
|
37
|
+
'http://xmlns.com/foaf/0.1/' => 'foaf',
|
38
|
+
'http://www.w3.org/2003/01/geo/wgs84_pos#' => 'geo',
|
39
|
+
'http://postneo.com/icbm/' => 'icbm',
|
40
|
+
'http://purl.org/rss/1.0/modules/image/' => 'image',
|
41
|
+
'http://www.itunes.com/DTDs/PodCast-1.0.dtd' => 'itunes',
|
42
|
+
'http://example.com/DTDs/PodCast-1.0.dtd' => 'itunes',
|
43
|
+
'http://purl.org/rss/1.0/modules/link/' => 'l',
|
44
|
+
'http://search.yahoo.com/mrss' => 'media',
|
45
|
+
'http://madskills.com/public/xml/rss/module/pingback/' => 'pingback',
|
46
|
+
'http://prismstandard.org/namespaces/1.2/basic/' => 'prism',
|
47
|
+
'http://www.w3.org/1999/02/22-rdf-syntax-ns#' => 'rdf',
|
48
|
+
'http://www.w3.org/2000/01/rdf-schema#' => 'rdfs',
|
49
|
+
'http://purl.org/rss/1.0/modules/reference/' => 'ref',
|
50
|
+
'http://purl.org/rss/1.0/modules/richequiv/' => 'reqv',
|
51
|
+
'http://purl.org/rss/1.0/modules/search/' => 'search',
|
52
|
+
'http://purl.org/rss/1.0/modules/slash/' => 'slash',
|
53
|
+
'http://schemas.xmlsoap.org/soap/envelope/' => 'soap',
|
54
|
+
'http://purl.org/rss/1.0/modules/servicestatus/' => 'ss',
|
55
|
+
'http://hacks.benhammersley.com/rss/streaming/' => 'str',
|
56
|
+
'http://purl.org/rss/1.0/modules/subscription/' => 'sub',
|
57
|
+
'http://purl.org/rss/1.0/modules/syndication/' => 'sy',
|
58
|
+
'http://purl.org/rss/1.0/modules/taxonomy/' => 'taxo',
|
59
|
+
'http://purl.org/rss/1.0/modules/threading/' => 'thr',
|
60
|
+
'http://purl.org/rss/1.0/modules/textinput/' => 'ti',
|
61
|
+
'http://madskills.com/public/xml/rss/module/trackback/' =>'trackback',
|
62
|
+
'http://wellformedweb.org/commentAPI/' => 'wfw',
|
63
|
+
'http://purl.org/rss/1.0/modules/wiki/' => 'wiki',
|
64
|
+
'http://www.w3.org/1999/xhtml' => 'xhtml',
|
65
|
+
'http://www.w3.org/XML/1998/namespace' => 'xml',
|
66
|
+
'http://www.w3.org/1999/xlink' => 'xlink',
|
67
|
+
'http://schemas.pocketsoap.com/rss/myDescModule/' => 'szf'
|
68
|
+
}
|
69
|
+
@matchnamespaces = {}
|
70
|
+
@namespaces.each do |l|
|
71
|
+
@matchnamespaces[l[0].downcase] = l[1]
|
72
|
+
end
|
73
|
+
@can_be_relative_uri = ['link', 'id', 'wfw_comment', 'wfw_commentrss', 'docs', 'url', 'href', 'comments', 'license', 'icon', 'logo']
|
74
|
+
@can_contain_relative_uris = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description']
|
75
|
+
@can_contain_dangerous_markup = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description']
|
76
|
+
@html_types = ['text/html', 'application/xhtml+xml']
|
77
|
+
@feeddata = FeedParserDict.new # feed-level data
|
78
|
+
@encoding = encoding # character encoding
|
79
|
+
@entries = [] # list of entry-level data
|
80
|
+
@version = '' # feed type/version see SUPPORTED_VERSIOSN
|
81
|
+
@namespacesInUse = {} # hash of namespaces defined by the feed
|
82
|
+
|
83
|
+
# the following are used internall to track state;
|
84
|
+
# this is really out of control and should be refactored
|
85
|
+
@infeed = false
|
86
|
+
@inentry = false
|
87
|
+
@incontent = 0 # Yes, this needs to be zero until I work out popContent and pushContent
|
88
|
+
@intextinput = false
|
89
|
+
@inimage = false
|
90
|
+
@inauthor = false
|
91
|
+
@incontributor = false
|
92
|
+
@inpublisher = false
|
93
|
+
@insource = false
|
94
|
+
@sourcedata = FeedParserDict.new
|
95
|
+
@contentparams = FeedParserDict.new
|
96
|
+
@summaryKey = nil
|
97
|
+
@namespacemap = {}
|
98
|
+
@elementstack = []
|
99
|
+
@basestack = []
|
100
|
+
@langstack = []
|
101
|
+
@baseuri = baseuri || ''
|
102
|
+
@lang = baselang || nil
|
103
|
+
if baselang
|
104
|
+
@feeddata['language'] = baselang.gsub('_','-')
|
105
|
+
end
|
106
|
+
@date_handlers = [:_parse_date_rfc822,
|
107
|
+
:_parse_date_hungarian, :_parse_date_greek,:_parse_date_mssql,
|
108
|
+
:_parse_date_nate,:_parse_date_onblog,:_parse_date_w3dtf,:_parse_date_iso8601
|
109
|
+
]
|
110
|
+
$stderr << "Leaving startup\n" if $debug # My addition
|
111
|
+
end
|
112
|
+
|
113
|
+
def unknown_starttag(tag, attrsd)
|
114
|
+
$stderr << "start #{tag} with #{attrsd}\n" if $debug
|
115
|
+
# normalize attrs
|
116
|
+
attrsD = {}
|
117
|
+
attrsd = Hash[*attrsd.flatten] if attrsd.class == Array # Magic! Asterisk!
|
118
|
+
# LooseFeedParser needs the above because SGMLParser sends attrs as a
|
119
|
+
# list of lists (like [['type','text/html'],['mode','escaped']])
|
120
|
+
|
121
|
+
attrsd.each do |old_k,value|
|
122
|
+
# There has to be a better, non-ugly way of doing this
|
123
|
+
k = old_k.downcase # Downcase all keys
|
124
|
+
attrsD[k] = value
|
125
|
+
if ['rel','type'].include?value
|
126
|
+
attrsD[k].downcase! # Downcase the value if the key is 'rel' or 'type'
|
127
|
+
end
|
128
|
+
end
|
129
|
+
|
130
|
+
# track xml:base and xml:lang
|
131
|
+
baseuri = attrsD['xml:base'] || attrsD['base'] || @baseuri
|
132
|
+
@baseuri = urljoin(@baseuri, baseuri)
|
133
|
+
lang = attrsD['xml:lang'] || attrsD['lang']
|
134
|
+
if lang == '' # FIXME This next bit of code is right? Wtf?
|
135
|
+
# xml:lang could be explicitly set to '', we need to capture that
|
136
|
+
lang = nil
|
137
|
+
elsif lang.nil?
|
138
|
+
# if no xml:lang is specified, use parent lang
|
139
|
+
lang = @lang
|
140
|
+
end
|
141
|
+
if lang and not lang.empty? # Seriously, this cannot be correct
|
142
|
+
if ['feed', 'rss', 'rdf:RDF'].include?tag
|
143
|
+
@feeddata['language'] = lang.gsub('_','-')
|
144
|
+
end
|
145
|
+
end
|
146
|
+
@lang = lang
|
147
|
+
@basestack << @baseuri
|
148
|
+
@langstack << lang
|
149
|
+
|
150
|
+
# track namespaces
|
151
|
+
attrsd.each do |prefix, uri|
|
152
|
+
if /^xmlns:/ =~ prefix # prefix begins with xmlns:
|
153
|
+
trackNamespace(prefix[6..-1], uri)
|
154
|
+
elsif prefix == 'xmlns':
|
155
|
+
trackNamespace(nil, uri)
|
156
|
+
end
|
157
|
+
end
|
158
|
+
|
159
|
+
# track inline content
|
160
|
+
if @incontent != 0 and @contentparams.has_key?('type') and not ( /xml$/ =~ (@contentparams['type'] || 'xml') )
|
161
|
+
# element declared itself as escaped markup, but isn't really
|
162
|
+
|
163
|
+
@contentparams['type'] = 'application/xhtml+xml'
|
164
|
+
end
|
165
|
+
if @incontent != 0 and @contentparams['type'] == 'application/xhtml+xml'
|
166
|
+
# Note: probably shouldn't simply recreate localname here, but
|
167
|
+
# our namespace handling isn't actually 100% correct in cases where
|
168
|
+
# the feed redefines the default namespace (which is actually
|
169
|
+
# the usual case for inline content, thanks Sam), so here we
|
170
|
+
# cheat and just reconstruct the element based on localname
|
171
|
+
# because that compensates for the bugs in our namespace handling.
|
172
|
+
# This will horribly munge inline content with non-empty qnames,
|
173
|
+
# but nobody actually does that, so I'm not fixing it.
|
174
|
+
tag = tag.split(':')[-1]
|
175
|
+
attrsA = attrsd.to_a.collect{|l| "#{l[0]}=\"#{l[1]}\""}
|
176
|
+
attrsS = ' '+attrsA.join(' ')
|
177
|
+
return handle_data("<#{tag}#{attrsS}>", escape=false)
|
178
|
+
end
|
179
|
+
|
180
|
+
# match namespaces
|
181
|
+
if /:/ =~ tag
|
182
|
+
prefix, suffix = tag.split(':', 2)
|
183
|
+
else
|
184
|
+
prefix, suffix = '', tag
|
185
|
+
end
|
186
|
+
prefix = @namespacemap[prefix] || prefix
|
187
|
+
if prefix and not prefix.empty?
|
188
|
+
prefix = prefix + '_'
|
189
|
+
end
|
190
|
+
|
191
|
+
# special hack for better tracking of empty textinput/image elements in illformed feeds
|
192
|
+
if (not prefix and not prefix.empty?) and not (['title', 'link', 'description','name'].include?tag)
|
193
|
+
@intextinput = false
|
194
|
+
end
|
195
|
+
if (prefix.nil? or prefix.empty?) and not (['title', 'link', 'description', 'url', 'href', 'width', 'height'].include?tag)
|
196
|
+
@inimage = false
|
197
|
+
end
|
198
|
+
|
199
|
+
# call special handler (if defined) or default handler
|
200
|
+
begin
|
201
|
+
return send('_start_'+prefix+suffix, attrsD)
|
202
|
+
rescue NoMethodError
|
203
|
+
return push(prefix + suffix, true)
|
204
|
+
end
|
205
|
+
end # End unknown_starttag
|
206
|
+
|
207
|
+
def unknown_endtag(tag)
|
208
|
+
$stderr << "end #{tag}\n" if $debug
|
209
|
+
# match namespaces
|
210
|
+
if tag.index(':')
|
211
|
+
prefix, suffix = tag.split(':',2)
|
212
|
+
else
|
213
|
+
prefix, suffix = '', tag
|
214
|
+
end
|
215
|
+
prefix = @namespacemap[prefix] || prefix
|
216
|
+
if prefix and not prefix.empty?
|
217
|
+
prefix = prefix + '_'
|
218
|
+
end
|
219
|
+
|
220
|
+
# call special handler (if defined) or default handler
|
221
|
+
begin
|
222
|
+
send('_end_' + prefix + suffix) # NOTE no return here! do not add it!
|
223
|
+
rescue NoMethodError => details
|
224
|
+
pop(prefix + suffix)
|
225
|
+
end
|
226
|
+
|
227
|
+
# track inline content
|
228
|
+
if @incontent != 0 and @contentparams.has_key?'type' and /xml$/ =~ (@contentparams['type'] || 'xml')
|
229
|
+
# element declared itself as escaped markup, but it isn't really
|
230
|
+
@contentparams['type'] = 'application/xhtml+xml'
|
231
|
+
end
|
232
|
+
if @incontent != 0 and @contentparams['type'] == 'application/xhtml+xml'
|
233
|
+
tag = tag.split(':')[-1]
|
234
|
+
handle_data("</#{tag}>", escape=false)
|
235
|
+
end
|
236
|
+
|
237
|
+
# track xml:base and xml:lang going out of scope
|
238
|
+
if @basestack and not @basestack.empty?
|
239
|
+
@basestack.pop
|
240
|
+
if @basestack and @basestack[-1] and not (@basestack.empty? or @basestack[-1].empty?)
|
241
|
+
@baseuri = @basestack[-1]
|
242
|
+
end
|
243
|
+
end
|
244
|
+
if @langstack and not @langstack.empty?
|
245
|
+
@langstack.pop
|
246
|
+
if @langstack and not @langstack.empty? # and @langstack[-1] and not @langstack.empty?
|
247
|
+
@lang = @langstack[-1]
|
248
|
+
end
|
249
|
+
end
|
250
|
+
end
|
251
|
+
|
252
|
+
def handle_charref(ref)
|
253
|
+
# LooseParserOnly
|
254
|
+
# called for each character reference, e.g. for ' ', ref will be '160'
|
255
|
+
$stderr << "entering handle_charref with #{ref}\n" if $debug
|
256
|
+
return if @elementstack.nil? or @elementstack.empty?
|
257
|
+
ref.downcase!
|
258
|
+
chars = ['34', '38', '39', '60', '62', 'x22', 'x26', 'x27', 'x3c', 'x3e']
|
259
|
+
if chars.include?ref
|
260
|
+
text = "&##{ref};"
|
261
|
+
else
|
262
|
+
if ref[0..0] == 'x'
|
263
|
+
c = (ref[1..-1]).to_i(16)
|
264
|
+
else
|
265
|
+
c = ref.to_i
|
266
|
+
end
|
267
|
+
text = [c].pack('U*')
|
268
|
+
end
|
269
|
+
@elementstack[-1][2] << text
|
270
|
+
end
|
271
|
+
|
272
|
+
def handle_entityref(ref)
|
273
|
+
# LooseParserOnly
|
274
|
+
# called for each entity reference, e.g. for '©', ref will be 'copy'
|
275
|
+
|
276
|
+
return if @elementstack.nil? or @elementstack.empty?
|
277
|
+
$stderr << "entering handle_entityref with #{ref}\n" if $debug
|
278
|
+
ents = ['lt', 'gt', 'quot', 'amp', 'apos']
|
279
|
+
if ents.include?ref
|
280
|
+
text = "&#{ref};"
|
281
|
+
else
|
282
|
+
text = HTMLEntities::decode_entities("&#{ref};")
|
283
|
+
end
|
284
|
+
@elementstack[-1][2] << text
|
285
|
+
end
|
286
|
+
|
287
|
+
def handle_data(text, escape=true)
|
288
|
+
# called for each block of plain text, i.e. outside of any tag and
|
289
|
+
# not containing any character or entity references
|
290
|
+
return if @elementstack.nil? or @elementstack.empty?
|
291
|
+
if escape and @contentparams['type'] == 'application/xhtml+xml'
|
292
|
+
text = text.to_xs
|
293
|
+
end
|
294
|
+
@elementstack[-1][2] << text
|
295
|
+
end
|
296
|
+
|
297
|
+
def handle_comment(comment)
|
298
|
+
# called for each comment, e.g. <!-- insert message here -->
|
299
|
+
end
|
300
|
+
|
301
|
+
def handle_pi(text)
|
302
|
+
end
|
303
|
+
|
304
|
+
def handle_decl(text)
|
305
|
+
end
|
306
|
+
|
307
|
+
def parse_declaration(i)
|
308
|
+
# for LooseFeedParser
|
309
|
+
$stderr << "entering parse_declaration\n" if $debug
|
310
|
+
if @rawdata[i...i+9] == '<![CDATA['
|
311
|
+
k = @rawdata.index(/\]\]>/u,i+9)
|
312
|
+
k = @rawdata.length unless k
|
313
|
+
handle_data(@rawdata[i+9...k].to_xs,false)
|
314
|
+
return k+3
|
315
|
+
else
|
316
|
+
k = @rawdata.index(/>/,i).to_i
|
317
|
+
return k+1
|
318
|
+
end
|
319
|
+
end
|
320
|
+
|
321
|
+
def mapContentType(contentType)
|
322
|
+
contentType.downcase!
|
323
|
+
case contentType
|
324
|
+
when 'text'
|
325
|
+
contentType = 'text/plain'
|
326
|
+
when 'html'
|
327
|
+
contentType = 'text/html'
|
328
|
+
when 'xhtml'
|
329
|
+
contentType = 'application/xhtml+xml'
|
330
|
+
end
|
331
|
+
return contentType
|
332
|
+
end
|
333
|
+
|
334
|
+
def trackNamespace(prefix, uri)
|
335
|
+
|
336
|
+
loweruri = uri.downcase.strip
|
337
|
+
if [prefix, loweruri] == [nil, 'http://my.netscape.com/rdf/simple/0.9/'] and (@version.nil? or @version.empty?)
|
338
|
+
@version = 'rss090'
|
339
|
+
elsif loweruri == 'http://purl.org/rss/1.0/' and (@version.nil? or @version.empty?)
|
340
|
+
@version = 'rss10'
|
341
|
+
elsif loweruri == 'http://www.w3.org/2005/atom' and (@version.nil? or @version.empty?)
|
342
|
+
@version = 'atom10'
|
343
|
+
elsif /backend\.userland\.com\/rss/ =~ loweruri
|
344
|
+
# match any backend.userland.com namespace
|
345
|
+
uri = 'http://backend.userland.com/rss'
|
346
|
+
loweruri = uri
|
347
|
+
end
|
348
|
+
if @matchnamespaces.has_key? loweruri
|
349
|
+
@namespacemap[prefix] = @matchnamespaces[loweruri]
|
350
|
+
@namespacesInUse[@matchnamespaces[loweruri]] = uri
|
351
|
+
else
|
352
|
+
@namespacesInUse[prefix || ''] = uri
|
353
|
+
end
|
354
|
+
end
|
355
|
+
|
356
|
+
def resolveURI(uri)
|
357
|
+
return urljoin(@baseuri || '', uri)
|
358
|
+
end
|
359
|
+
|
360
|
+
def decodeEntities(element, data)
|
361
|
+
return data
|
362
|
+
end
|
363
|
+
|
364
|
+
def push(element, expectingText)
|
365
|
+
@elementstack << [element, expectingText, []]
|
366
|
+
end
|
367
|
+
|
368
|
+
def pop(element, stripWhitespace=true)
|
369
|
+
return if @elementstack.nil? or @elementstack.empty?
|
370
|
+
return if @elementstack[-1][0] != element
|
371
|
+
element, expectingText, pieces = @elementstack.pop
|
372
|
+
if pieces.class == Array
|
373
|
+
output = pieces.join('')
|
374
|
+
else
|
375
|
+
output = pieces
|
376
|
+
end
|
377
|
+
if stripWhitespace
|
378
|
+
output.strip!
|
379
|
+
end
|
380
|
+
return output if not expectingText
|
381
|
+
|
382
|
+
# decode base64 content
|
383
|
+
if @contentparams['base64']
|
384
|
+
out64 = Base64::decode64(output) # a.k.a. [output].unpack('m')[0]
|
385
|
+
if not output.empty? and not out64.empty?
|
386
|
+
output = out64
|
387
|
+
end
|
388
|
+
end
|
389
|
+
|
390
|
+
# resolve relative URIs
|
391
|
+
if @can_be_relative_uri.include?element and output and not output.empty?
|
392
|
+
output = resolveURI(output)
|
393
|
+
end
|
394
|
+
|
395
|
+
# decode entities within embedded markup
|
396
|
+
if not @contentparams['base64']
|
397
|
+
output = decodeEntities(element, output)
|
398
|
+
end
|
399
|
+
|
400
|
+
# remove temporary cruft from contentparams
|
401
|
+
@contentparams.delete('mode')
|
402
|
+
@contentparams.delete('base64')
|
403
|
+
|
404
|
+
# resolve relative URIs within embedded markup
|
405
|
+
if @html_types.include?mapContentType(@contentparams['type'] || 'text/html')
|
406
|
+
if @can_contain_relative_uris.include?element
|
407
|
+
output = FeedParser.resolveRelativeURIs(output, @baseuri, @encoding)
|
408
|
+
end
|
409
|
+
end
|
410
|
+
# sanitize embedded markup
|
411
|
+
if @html_types.include?mapContentType(@contentparams['type'] || 'text/html')
|
412
|
+
if @can_contain_dangerous_markup.include?element
|
413
|
+
output = FeedParser.sanitizeHTML(output, @encoding)
|
414
|
+
end
|
415
|
+
end
|
416
|
+
|
417
|
+
if @encoding and not @encoding.empty? and @encoding != 'utf-8'
|
418
|
+
output = uconvert(output, @encoding, 'utf-8')
|
419
|
+
# FIXME I turn everything into utf-8, not unicode, originally because REXML was being used but now beause I haven't tested it out yet.
|
420
|
+
end
|
421
|
+
|
422
|
+
# categories/tags/keywords/whatever are handled in _end_category
|
423
|
+
return output if element == 'category'
|
424
|
+
|
425
|
+
# store output in appropriate place(s)
|
426
|
+
if @inentry and not @insource
|
427
|
+
if element == 'content'
|
428
|
+
@entries[-1][element] ||= []
|
429
|
+
contentparams = Marshal.load(Marshal.dump(@contentparams)) # deepcopy
|
430
|
+
contentparams['value'] = output
|
431
|
+
@entries[-1][element] << contentparams
|
432
|
+
elsif element == 'link'
|
433
|
+
@entries[-1][element] = output
|
434
|
+
if output and not output.empty?
|
435
|
+
@entries[-1]['links'][-1]['href'] = output
|
436
|
+
end
|
437
|
+
else
|
438
|
+
element = 'summary' if element == 'description'
|
439
|
+
@entries[-1][element] = output
|
440
|
+
if @incontent != 0
|
441
|
+
contentparams = Marshal.load(Marshal.dump(@contentparams))
|
442
|
+
contentparams['value'] = output
|
443
|
+
@entries[-1][element + '_detail'] = contentparams
|
444
|
+
end
|
445
|
+
end
|
446
|
+
elsif (@infeed or @insource) and not @intextinput and not @inimage
|
447
|
+
context = getContext()
|
448
|
+
element = 'subtitle' if element == 'description'
|
449
|
+
context[element] = output
|
450
|
+
if element == 'link'
|
451
|
+
context['links'][-1]['href'] = output
|
452
|
+
elsif @incontent != 0
|
453
|
+
contentparams = Marshal.load(Marshal.dump(@contentparams))
|
454
|
+
contentparams['value'] = output
|
455
|
+
context[element + '_detail'] = contentparams
|
456
|
+
end
|
457
|
+
end
|
458
|
+
return output
|
459
|
+
end
|
460
|
+
|
461
|
+
def pushContent(tag, attrsD, defaultContentType, expectingText)
|
462
|
+
@incontent += 1 # Yes, I hate this.
|
463
|
+
type = mapContentType(attrsD['type'] || defaultContentType)
|
464
|
+
@contentparams = FeedParserDict.new({'type' => type,'language' => @lang,'base' => @baseuri})
|
465
|
+
@contentparams['base64'] = isBase64(attrsD, @contentparams)
|
466
|
+
push(tag, expectingText)
|
467
|
+
end
|
468
|
+
|
469
|
+
def popContent(tag)
|
470
|
+
value = pop(tag)
|
471
|
+
@incontent -= 1
|
472
|
+
@contentparams.clear
|
473
|
+
return value
|
474
|
+
end
|
475
|
+
|
476
|
+
def mapToStandardPrefix(name)
|
477
|
+
colonpos = name.index(':')
|
478
|
+
if colonpos
|
479
|
+
prefix = name[0..colonpos-1]
|
480
|
+
suffix = name[colonpos+1..-1]
|
481
|
+
prefix = @namespacemap[prefix] || prefix
|
482
|
+
name = prefix + ':' + suffix
|
483
|
+
end
|
484
|
+
return name
|
485
|
+
end
|
486
|
+
|
487
|
+
def getAttribute(attrsD, name)
|
488
|
+
return attrsD[mapToStandardPrefix(name)]
|
489
|
+
end
|
490
|
+
|
491
|
+
def isBase64(attrsD, contentparams)
|
492
|
+
return true if (attrsD['mode'] == 'base64')
|
493
|
+
if /(^text\/)|(\+xml$)|(\/xml$)/ =~ contentparams['type']
|
494
|
+
return false
|
495
|
+
end
|
496
|
+
return true
|
497
|
+
end
|
498
|
+
|
499
|
+
def itsAnHrefDamnIt(attrsD)
|
500
|
+
href= attrsD['url'] || attrsD['uri'] || attrsD['href']
|
501
|
+
if href
|
502
|
+
attrsD.delete('url')
|
503
|
+
attrsD.delete('uri')
|
504
|
+
attrsD['href'] = href
|
505
|
+
end
|
506
|
+
return attrsD
|
507
|
+
end
|
508
|
+
|
509
|
+
|
510
|
+
def _save(key, value)
|
511
|
+
context = getContext()
|
512
|
+
context[key] ||= value
|
513
|
+
end
|
514
|
+
|
515
|
+
def _start_rss(attrsD)
|
516
|
+
versionmap = {'0.91' => 'rss091u',
|
517
|
+
'0.92' => 'rss092',
|
518
|
+
'0.93' => 'rss093',
|
519
|
+
'0.94' => 'rss094'
|
520
|
+
}
|
521
|
+
|
522
|
+
if not @version or @version.empty?
|
523
|
+
attr_version = attrsD['version'] || ''
|
524
|
+
version = versionmap[attr_version]
|
525
|
+
if version and not version.empty?
|
526
|
+
@version = version
|
527
|
+
elsif /^2\./ =~ attr_version
|
528
|
+
@version = 'rss20'
|
529
|
+
else
|
530
|
+
@version = 'rss'
|
531
|
+
end
|
532
|
+
end
|
533
|
+
end
|
534
|
+
|
535
|
+
def _start_dlhottitles(attrsD)
|
536
|
+
@version = 'hotrss'
|
537
|
+
end
|
538
|
+
|
539
|
+
def _start_channel(attrsD)
|
540
|
+
@infeed = true
|
541
|
+
_cdf_common(attrsD)
|
542
|
+
end
|
543
|
+
alias :_start_feedinfo :_start_channel
|
544
|
+
|
545
|
+
def _cdf_common(attrsD)
|
546
|
+
if attrsD.has_key?'lastmod'
|
547
|
+
_start_modified({})
|
548
|
+
@elementstack[-1][-1] = attrsD['lastmod']
|
549
|
+
_end_modified
|
550
|
+
end
|
551
|
+
if attrsD.has_key?'href'
|
552
|
+
_start_link({})
|
553
|
+
@elementstack[-1][-1] = attrsD['href']
|
554
|
+
_end_link
|
555
|
+
end
|
556
|
+
end
|
557
|
+
|
558
|
+
def _start_feed(attrsD)
|
559
|
+
@infeed = true
|
560
|
+
versionmap = {'0.1' => 'atom01',
|
561
|
+
'0.2' => 'atom02',
|
562
|
+
'0.3' => 'atom03'
|
563
|
+
}
|
564
|
+
|
565
|
+
if not @version or @version.empty?
|
566
|
+
attr_version = attrsD['version']
|
567
|
+
version = versionmap[attr_version]
|
568
|
+
if @version and not @version.empty?
|
569
|
+
@version = version
|
570
|
+
else
|
571
|
+
@version = 'atom'
|
572
|
+
end
|
573
|
+
end
|
574
|
+
end
|
575
|
+
|
576
|
+
def _end_channel
|
577
|
+
@infeed = false
|
578
|
+
end
|
579
|
+
alias :_end_feed :_end_channel
|
580
|
+
|
581
|
+
def _start_image(attrsD)
|
582
|
+
@inimage = true
|
583
|
+
push('image', false)
|
584
|
+
context = getContext()
|
585
|
+
context['image'] ||= FeedParserDict.new
|
586
|
+
end
|
587
|
+
|
588
|
+
def _end_image
|
589
|
+
pop('image')
|
590
|
+
@inimage = false
|
591
|
+
end
|
592
|
+
|
593
|
+
def _start_textinput(attrsD)
|
594
|
+
@intextinput = true
|
595
|
+
push('textinput', false)
|
596
|
+
context = getContext()
|
597
|
+
context['textinput'] ||= FeedParserDict.new
|
598
|
+
end
|
599
|
+
alias :_start_textInput :_start_textinput
|
600
|
+
|
601
|
+
def _end_textinput
|
602
|
+
pop('textinput')
|
603
|
+
@intextinput = false
|
604
|
+
end
|
605
|
+
alias :_end_textInput :_end_textinput
|
606
|
+
|
607
|
+
def _start_author(attrsD)
|
608
|
+
@inauthor = true
|
609
|
+
push('author', true)
|
610
|
+
end
|
611
|
+
alias :_start_managingeditor :_start_author
|
612
|
+
alias :_start_dc_author :_start_author
|
613
|
+
alias :_start_dc_creator :_start_author
|
614
|
+
alias :_start_itunes_author :_start_author
|
615
|
+
|
616
|
+
def _end_author
|
617
|
+
pop('author')
|
618
|
+
@inauthor = false
|
619
|
+
_sync_author_detail()
|
620
|
+
end
|
621
|
+
alias :_end_managingeditor :_end_author
|
622
|
+
alias :_end_dc_author :_end_author
|
623
|
+
alias :_end_dc_creator :_end_author
|
624
|
+
alias :_end_itunes_author :_end_author
|
625
|
+
|
626
|
+
def _start_itunes_owner(attrsD)
|
627
|
+
@inpublisher = true
|
628
|
+
push('publisher', false)
|
629
|
+
end
|
630
|
+
|
631
|
+
def _end_itunes_owner
|
632
|
+
pop('publisher')
|
633
|
+
@inpublisher = false
|
634
|
+
_sync_author_detail('publisher')
|
635
|
+
end
|
636
|
+
|
637
|
+
def _start_contributor(attrsD)
|
638
|
+
@incontributor = true
|
639
|
+
context = getContext()
|
640
|
+
context['contributors'] ||= []
|
641
|
+
context['contributors'] << FeedParserDict.new
|
642
|
+
push('contributor', false)
|
643
|
+
end
|
644
|
+
|
645
|
+
def _end_contributor
|
646
|
+
pop('contributor')
|
647
|
+
@incontributor = false
|
648
|
+
end
|
649
|
+
|
650
|
+
def _start_dc_contributor(attrsD)
|
651
|
+
@incontributor = true
|
652
|
+
context = getContext()
|
653
|
+
context['contributors'] ||= []
|
654
|
+
context['contributors'] << FeedParserDict.new
|
655
|
+
push('name', false)
|
656
|
+
end
|
657
|
+
|
658
|
+
def _end_dc_contributor
|
659
|
+
_end_name
|
660
|
+
@incontributor = false
|
661
|
+
end
|
662
|
+
|
663
|
+
def _start_name(attrsD)
|
664
|
+
push('name', false)
|
665
|
+
end
|
666
|
+
alias :_start_itunes_name :_start_name
|
667
|
+
|
668
|
+
def _end_name
|
669
|
+
value = pop('name')
|
670
|
+
if @inpublisher
|
671
|
+
_save_author('name', value, 'publisher')
|
672
|
+
elsif @inauthor
|
673
|
+
_save_author('name', value)
|
674
|
+
elsif @incontributor
|
675
|
+
_save_contributor('name', value)
|
676
|
+
elsif @intextinput
|
677
|
+
context = getContext()
|
678
|
+
context['textinput']['name'] = value
|
679
|
+
end
|
680
|
+
end
|
681
|
+
alias :_end_itunes_name :_end_name
|
682
|
+
|
683
|
+
def _start_width(attrsD)
|
684
|
+
push('width', false)
|
685
|
+
end
|
686
|
+
|
687
|
+
def _end_width
|
688
|
+
value = pop('width').to_i
|
689
|
+
if @inimage
|
690
|
+
context = getContext
|
691
|
+
context['image']['width'] = value
|
692
|
+
end
|
693
|
+
end
|
694
|
+
|
695
|
+
def _start_height(attrsD)
|
696
|
+
push('height', false)
|
697
|
+
end
|
698
|
+
|
699
|
+
def _end_height
|
700
|
+
value = pop('height').to_i
|
701
|
+
if @inimage
|
702
|
+
context = getContext()
|
703
|
+
context['image']['height'] = value
|
704
|
+
end
|
705
|
+
end
|
706
|
+
|
707
|
+
def _start_url(attrsD)
|
708
|
+
push('href', true)
|
709
|
+
end
|
710
|
+
alias :_start_homepage :_start_url
|
711
|
+
alias :_start_uri :_start_url
|
712
|
+
|
713
|
+
def _end_url
|
714
|
+
value = pop('href')
|
715
|
+
if @inauthor
|
716
|
+
_save_author('href', value)
|
717
|
+
elsif @incontributor
|
718
|
+
_save_contributor('href', value)
|
719
|
+
elsif @inimage
|
720
|
+
context = getContext()
|
721
|
+
context['image']['href'] = value
|
722
|
+
elsif @intextinput
|
723
|
+
context = getContext()
|
724
|
+
context['textinput']['link'] = value
|
725
|
+
end
|
726
|
+
end
|
727
|
+
alias :_end_homepage :_end_url
|
728
|
+
alias :_end_uri :_end_url
|
729
|
+
|
730
|
+
def _start_email(attrsD)
|
731
|
+
push('email', false)
|
732
|
+
end
|
733
|
+
alias :_start_itunes_email :_start_email
|
734
|
+
|
735
|
+
def _end_email
|
736
|
+
value = pop('email')
|
737
|
+
if @inpublisher
|
738
|
+
_save_author('email', value, 'publisher')
|
739
|
+
elsif @inauthor
|
740
|
+
_save_author('email', value)
|
741
|
+
elsif @incontributor
|
742
|
+
_save_contributor('email', value)
|
743
|
+
end
|
744
|
+
end
|
745
|
+
alias :_end_itunes_email :_end_email
|
746
|
+
|
747
|
+
def getContext
|
748
|
+
if @insource
|
749
|
+
context = @sourcedata
|
750
|
+
elsif @inentry
|
751
|
+
context = @entries[-1]
|
752
|
+
else
|
753
|
+
context = @feeddata
|
754
|
+
end
|
755
|
+
return context
|
756
|
+
end
|
757
|
+
|
758
|
+
def _save_author(key, value, prefix='author')
|
759
|
+
context = getContext()
|
760
|
+
context[prefix + '_detail'] ||= FeedParserDict.new
|
761
|
+
context[prefix + '_detail'][key] = value
|
762
|
+
_sync_author_detail()
|
763
|
+
end
|
764
|
+
|
765
|
+
def _save_contributor(key, value)
|
766
|
+
context = getContext
|
767
|
+
context['contributors'] ||= [FeedParserDict.new]
|
768
|
+
context['contributors'][-1][key] = value
|
769
|
+
end
|
770
|
+
|
771
|
+
def _sync_author_detail(key='author')
|
772
|
+
context = getContext()
|
773
|
+
detail = context["#{key}_detail"]
|
774
|
+
if detail and not detail.empty?
|
775
|
+
name = detail['name']
|
776
|
+
email = detail['email']
|
777
|
+
|
778
|
+
if name and email and not (name.empty? or name.empty?)
|
779
|
+
context[key] = "#{name} (#{email})"
|
780
|
+
elsif name and not name.empty?
|
781
|
+
context[key] = name
|
782
|
+
elsif email and not email.empty?
|
783
|
+
context[key] = email
|
784
|
+
end
|
785
|
+
else
|
786
|
+
author = context[key].dup unless context[key].nil?
|
787
|
+
return if not author or author.empty?
|
788
|
+
emailmatch = author.match(/(([a-zA-Z0-9\_\-\.\+]+)@((\[[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.)|(([a-zA-Z0-9\-]+\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(\]?))/)
|
789
|
+
email = emailmatch[1]
|
790
|
+
author.gsub!(email, '')
|
791
|
+
author.gsub!("\(\)", '')
|
792
|
+
author.strip!
|
793
|
+
author.gsub!(/^\(/,'')
|
794
|
+
author.gsub!(/\)$/,'')
|
795
|
+
author.strip!
|
796
|
+
context["#{key}_detail"] ||= FeedParserDict.new
|
797
|
+
context["#{key}_detail"]['name'] = author
|
798
|
+
context["#{key}_detail"]['email'] = email
|
799
|
+
end
|
800
|
+
end
|
801
|
+
|
802
|
+
def _start_subtitle(attrsD)
|
803
|
+
pushContent('subtitle', attrsD, 'text/plain', true)
|
804
|
+
end
|
805
|
+
alias :_start_tagline :_start_subtitle
|
806
|
+
alias :_start_itunes_subtitle :_start_subtitle
|
807
|
+
|
808
|
+
def _end_subtitle
|
809
|
+
popContent('subtitle')
|
810
|
+
end
|
811
|
+
alias :_end_tagline :_end_subtitle
|
812
|
+
alias :_end_itunes_subtitle :_end_subtitle
|
813
|
+
|
814
|
+
def _start_rights(attrsD)
|
815
|
+
pushContent('rights', attrsD, 'text/plain', true)
|
816
|
+
end
|
817
|
+
alias :_start_dc_rights :_start_rights
|
818
|
+
alias :_start_copyright :_start_rights
|
819
|
+
|
820
|
+
def _end_rights
|
821
|
+
popContent('rights')
|
822
|
+
end
|
823
|
+
alias :_end_dc_rights :_end_rights
|
824
|
+
alias :_end_copyright :_end_rights
|
825
|
+
|
826
|
+
def _start_item(attrsD)
|
827
|
+
@entries << FeedParserDict.new
|
828
|
+
push('item', false)
|
829
|
+
@inentry = true
|
830
|
+
@guidislink = false
|
831
|
+
id = getAttribute(attrsD, 'rdf:about')
|
832
|
+
if id and not id.empty?
|
833
|
+
context = getContext()
|
834
|
+
context['id'] = id
|
835
|
+
end
|
836
|
+
_cdf_common(attrsD)
|
837
|
+
end
|
838
|
+
alias :_start_entry :_start_item
|
839
|
+
alias :_start_product :_start_item
|
840
|
+
|
841
|
+
def _end_item
|
842
|
+
pop('item')
|
843
|
+
@inentry = false
|
844
|
+
end
|
845
|
+
alias :_end_entry :_end_item
|
846
|
+
|
847
|
+
def _start_dc_language(attrsD)
|
848
|
+
push('language', true)
|
849
|
+
end
|
850
|
+
alias :_start_language :_start_dc_language
|
851
|
+
|
852
|
+
def _end_dc_language
|
853
|
+
@lang = pop('language')
|
854
|
+
end
|
855
|
+
alias :_end_language :_end_dc_language
|
856
|
+
|
857
|
+
def _start_dc_publisher(attrsD)
|
858
|
+
push('publisher', true)
|
859
|
+
end
|
860
|
+
alias :_start_webmaster :_start_dc_publisher
|
861
|
+
|
862
|
+
def _end_dc_publisher
|
863
|
+
pop('publisher')
|
864
|
+
_sync_author_detail('publisher')
|
865
|
+
end
|
866
|
+
alias :_end_webmaster :_end_dc_publisher
|
867
|
+
|
868
|
+
def _start_published(attrsD)
|
869
|
+
push('published', true)
|
870
|
+
end
|
871
|
+
alias :_start_dcterms_issued :_start_published
|
872
|
+
alias :_start_issued :_start_published
|
873
|
+
|
874
|
+
def _end_published
|
875
|
+
value = pop('published')
|
876
|
+
_save('published_parsed', parse_date(value))
|
877
|
+
end
|
878
|
+
alias :_end_dcterms_issued :_end_published
|
879
|
+
alias :_end_issued :_end_published
|
880
|
+
|
881
|
+
def _start_updated(attrsD)
|
882
|
+
push('updated', true)
|
883
|
+
end
|
884
|
+
alias :_start_modified :_start_updated
|
885
|
+
alias :_start_dcterms_modified :_start_updated
|
886
|
+
alias :_start_pubdate :_start_updated
|
887
|
+
alias :_start_dc_date :_start_updated
|
888
|
+
|
889
|
+
def _end_updated
|
890
|
+
value = pop('updated')
|
891
|
+
_save('updated_parsed', parse_date(value))
|
892
|
+
end
|
893
|
+
alias :_end_modified :_end_updated
|
894
|
+
alias :_end_dcterms_modified :_end_updated
|
895
|
+
alias :_end_pubdate :_end_updated
|
896
|
+
alias :_end_dc_date :_end_updated
|
897
|
+
|
898
|
+
def _start_created(attrsD)
|
899
|
+
push('created', true)
|
900
|
+
end
|
901
|
+
alias :_start_dcterms_created :_start_created
|
902
|
+
|
903
|
+
def _end_created
|
904
|
+
value = pop('created')
|
905
|
+
_save('created_parsed', parse_date(value))
|
906
|
+
end
|
907
|
+
alias :_end_dcterms_created :_end_created
|
908
|
+
|
909
|
+
def _start_expirationdate(attrsD)
|
910
|
+
push('expired', true)
|
911
|
+
end
|
912
|
+
def _end_expirationdate
|
913
|
+
_save('expired_parsed', parse_date(pop('expired')))
|
914
|
+
end
|
915
|
+
|
916
|
+
def _start_cc_license(attrsD)
|
917
|
+
push('license', true)
|
918
|
+
value = getAttribute(attrsD, 'rdf:resource')
|
919
|
+
if value and not value.empty?
|
920
|
+
@elementstack[-1][2] << value
|
921
|
+
pop('license')
|
922
|
+
end
|
923
|
+
end
|
924
|
+
|
925
|
+
def _start_creativecommons_license(attrsD)
|
926
|
+
push('license', true)
|
927
|
+
end
|
928
|
+
|
929
|
+
def _end_creativecommons_license
|
930
|
+
pop('license')
|
931
|
+
end
|
932
|
+
|
933
|
+
def addTag(term, scheme, label)
|
934
|
+
context = getContext()
|
935
|
+
context['tags'] ||= []
|
936
|
+
tags = context['tags']
|
937
|
+
if (term.nil? or term.empty?) and (scheme.nil? or scheme.empty?) and (label.nil? or label.empty?)
|
938
|
+
return
|
939
|
+
end
|
940
|
+
value = FeedParserDict.new({'term' => term, 'scheme' => scheme, 'label' => label})
|
941
|
+
if not tags.include?value
|
942
|
+
context['tags'] << FeedParserDict.new({'term' => term, 'scheme' => scheme, 'label' => label})
|
943
|
+
end
|
944
|
+
end
|
945
|
+
|
946
|
+
def _start_category(attrsD)
|
947
|
+
$stderr << "entering _start_category with #{attrsD}\n" if $debug
|
948
|
+
|
949
|
+
term = attrsD['term']
|
950
|
+
scheme = attrsD['scheme'] || attrsD['domain']
|
951
|
+
label = attrsD['label']
|
952
|
+
addTag(term, scheme, label)
|
953
|
+
push('category', true)
|
954
|
+
end
|
955
|
+
alias :_start_dc_subject :_start_category
|
956
|
+
alias :_start_keywords :_start_category
|
957
|
+
|
958
|
+
def _end_itunes_keywords
|
959
|
+
pop('itunes_keywords').split.each do |term|
|
960
|
+
addTag(term, 'http://www.itunes.com/', nil)
|
961
|
+
end
|
962
|
+
end
|
963
|
+
|
964
|
+
def _start_itunes_category(attrsD)
|
965
|
+
addTag(attrsD['text'], 'http://www.itunes.com/', nil)
|
966
|
+
push('category', true)
|
967
|
+
end
|
968
|
+
|
969
|
+
def _end_category
|
970
|
+
value = pop('category')
|
971
|
+
return if value.nil? or value.empty?
|
972
|
+
context = getContext()
|
973
|
+
tags = context['tags']
|
974
|
+
if value and not value.empty? and not tags.empty? and not tags[-1]['term']:
|
975
|
+
tags[-1]['term'] = value
|
976
|
+
else
|
977
|
+
addTag(value, nil, nil)
|
978
|
+
end
|
979
|
+
end
|
980
|
+
alias :_end_dc_subject :_end_category
|
981
|
+
alias :_end_keywords :_end_category
|
982
|
+
alias :_end_itunes_category :_end_category
|
983
|
+
|
984
|
+
def _start_cloud(attrsD)
|
985
|
+
getContext()['cloud'] = FeedParserDict.new(attrsD)
|
986
|
+
end
|
987
|
+
|
988
|
+
def _start_link(attrsD)
|
989
|
+
attrsD['rel'] ||= 'alternate'
|
990
|
+
attrsD['type'] ||= 'text/html'
|
991
|
+
attrsD = itsAnHrefDamnIt(attrsD)
|
992
|
+
if attrsD.has_key? 'href'
|
993
|
+
attrsD['href'] = resolveURI(attrsD['href'])
|
994
|
+
end
|
995
|
+
expectingText = @infeed || @inentry || @insource
|
996
|
+
context = getContext()
|
997
|
+
context['links'] ||= []
|
998
|
+
context['links'] << FeedParserDict.new(attrsD)
|
999
|
+
if attrsD['rel'] == 'enclosure'
|
1000
|
+
_start_enclosure(attrsD)
|
1001
|
+
end
|
1002
|
+
if attrsD.has_key? 'href'
|
1003
|
+
expectingText = false
|
1004
|
+
if (attrsD['rel'] == 'alternate') and @html_types.include?mapContentType(attrsD['type'])
|
1005
|
+
context['link'] = attrsD['href']
|
1006
|
+
end
|
1007
|
+
else
|
1008
|
+
push('link', expectingText)
|
1009
|
+
end
|
1010
|
+
end
|
1011
|
+
alias :_start_producturl :_start_link
|
1012
|
+
|
1013
|
+
def _end_link
|
1014
|
+
value = pop('link')
|
1015
|
+
context = getContext()
|
1016
|
+
if @intextinput
|
1017
|
+
context['textinput']['link'] = value
|
1018
|
+
end
|
1019
|
+
if @inimage
|
1020
|
+
context['image']['link'] = value
|
1021
|
+
end
|
1022
|
+
end
|
1023
|
+
alias :_end_producturl :_end_link
|
1024
|
+
|
1025
|
+
def _start_guid(attrsD)
|
1026
|
+
@guidislink = ((attrsD['ispermalink'] || 'true') == 'true')
|
1027
|
+
push('id', true)
|
1028
|
+
end
|
1029
|
+
|
1030
|
+
def _end_guid
|
1031
|
+
value = pop('id')
|
1032
|
+
_save('guidislink', (@guidislink and not getContext().has_key?('link')))
|
1033
|
+
if @guidislink:
|
1034
|
+
# guid acts as link, but only if 'ispermalink' is not present or is 'true',
|
1035
|
+
# and only if the item doesn't already have a link element
|
1036
|
+
_save('link', value)
|
1037
|
+
end
|
1038
|
+
end
|
1039
|
+
|
1040
|
+
|
1041
|
+
def _start_title(attrsD)
|
1042
|
+
pushContent('title', attrsD, 'text/plain', @infeed || @inentry || @insource)
|
1043
|
+
end
|
1044
|
+
alias :_start_dc_title :_start_title
|
1045
|
+
alias :_start_media_title :_start_title
|
1046
|
+
|
1047
|
+
def _end_title
|
1048
|
+
value = popContent('title')
|
1049
|
+
context = getContext()
|
1050
|
+
if @intextinput
|
1051
|
+
context['textinput']['title'] = value
|
1052
|
+
elsif @inimage
|
1053
|
+
context['image']['title'] = value
|
1054
|
+
end
|
1055
|
+
end
|
1056
|
+
alias :_end_dc_title :_end_title
|
1057
|
+
alias :_end_media_title :_end_title
|
1058
|
+
|
1059
|
+
def _start_description(attrsD)
|
1060
|
+
context = getContext()
|
1061
|
+
if context.has_key?('summary')
|
1062
|
+
@summaryKey = 'content'
|
1063
|
+
_start_content(attrsD)
|
1064
|
+
else
|
1065
|
+
pushContent('description', attrsD, 'text/html', @infeed || @inentry || @insource)
|
1066
|
+
end
|
1067
|
+
end
|
1068
|
+
|
1069
|
+
def _start_abstract(attrsD)
|
1070
|
+
pushContent('description', attrsD, 'text/plain', @infeed || @inentry || @insource)
|
1071
|
+
end
|
1072
|
+
|
1073
|
+
def _end_description
|
1074
|
+
if @summaryKey == 'content'
|
1075
|
+
_end_content()
|
1076
|
+
else
|
1077
|
+
value = popContent('description')
|
1078
|
+
context = getContext()
|
1079
|
+
if @intextinput
|
1080
|
+
context['textinput']['description'] = value
|
1081
|
+
elsif @inimage:
|
1082
|
+
context['image']['description'] = value
|
1083
|
+
end
|
1084
|
+
end
|
1085
|
+
@summaryKey = nil
|
1086
|
+
end
|
1087
|
+
alias :_end_abstract :_end_description
|
1088
|
+
|
1089
|
+
def _start_info(attrsD)
|
1090
|
+
pushContent('info', attrsD, 'text/plain', true)
|
1091
|
+
end
|
1092
|
+
alias :_start_feedburner_browserfriendly :_start_info
|
1093
|
+
|
1094
|
+
def _end_info
|
1095
|
+
popContent('info')
|
1096
|
+
end
|
1097
|
+
alias :_end_feedburner_browserfriendly :_end_info
|
1098
|
+
|
1099
|
+
def _start_generator(attrsD)
|
1100
|
+
if attrsD and not attrsD.empty?
|
1101
|
+
attrsD = itsAnHrefDamnIt(attrsD)
|
1102
|
+
if attrsD.has_key?('href')
|
1103
|
+
attrsD['href'] = resolveURI(attrsD['href'])
|
1104
|
+
end
|
1105
|
+
end
|
1106
|
+
getContext()['generator_detail'] = FeedParserDict.new(attrsD)
|
1107
|
+
push('generator', true)
|
1108
|
+
end
|
1109
|
+
|
1110
|
+
def _end_generator
|
1111
|
+
value = pop('generator')
|
1112
|
+
context = getContext()
|
1113
|
+
if context.has_key?('generator_detail')
|
1114
|
+
context['generator_detail']['name'] = value
|
1115
|
+
end
|
1116
|
+
end
|
1117
|
+
|
1118
|
+
def _start_admin_generatoragent(attrsD)
|
1119
|
+
push('generator', true)
|
1120
|
+
value = getAttribute(attrsD, 'rdf:resource')
|
1121
|
+
if value and not value.empty?
|
1122
|
+
@elementstack[-1][2] << value
|
1123
|
+
end
|
1124
|
+
pop('generator')
|
1125
|
+
getContext()['generator_detail'] = FeedParserDict.new({'href' => value})
|
1126
|
+
end
|
1127
|
+
|
1128
|
+
def _start_admin_errorreportsto(attrsD)
|
1129
|
+
push('errorreportsto', true)
|
1130
|
+
value = getAttribute(attrsD, 'rdf:resource')
|
1131
|
+
if value and not value.empty?
|
1132
|
+
@elementstack[-1][2] << value
|
1133
|
+
end
|
1134
|
+
pop('errorreportsto')
|
1135
|
+
end
|
1136
|
+
|
1137
|
+
def _start_summary(attrsD)
|
1138
|
+
context = getContext()
|
1139
|
+
if context.has_key?'summary'
|
1140
|
+
@summaryKey = 'content'
|
1141
|
+
_start_content(attrsD)
|
1142
|
+
else
|
1143
|
+
@summaryKey = 'summary'
|
1144
|
+
pushContent(@summaryKey, attrsD, 'text/plain', true)
|
1145
|
+
end
|
1146
|
+
end
|
1147
|
+
alias :_start_itunes_summary :_start_summary
|
1148
|
+
|
1149
|
+
def _end_summary
|
1150
|
+
if @summaryKey == 'content':
|
1151
|
+
_end_content()
|
1152
|
+
else
|
1153
|
+
popContent(@summaryKey || 'summary')
|
1154
|
+
end
|
1155
|
+
@summaryKey = nil
|
1156
|
+
end
|
1157
|
+
alias :_end_itunes_summary :_end_summary
|
1158
|
+
|
1159
|
+
def _start_enclosure(attrsD)
|
1160
|
+
attrsD = itsAnHrefDamnIt(attrsD)
|
1161
|
+
getContext()['enclosures'] ||= []
|
1162
|
+
getContext()['enclosures'] << FeedParserDict.new(attrsD)
|
1163
|
+
href = attrsD['href']
|
1164
|
+
if href and not href.empty?
|
1165
|
+
context = getContext()
|
1166
|
+
if not context['id']
|
1167
|
+
context['id'] = href
|
1168
|
+
end
|
1169
|
+
end
|
1170
|
+
end
|
1171
|
+
|
1172
|
+
def _start_source(attrsD)
|
1173
|
+
@insource = true
|
1174
|
+
end
|
1175
|
+
|
1176
|
+
def _end_source
|
1177
|
+
@insource = false
|
1178
|
+
getContext()['source'] = Marshal.load(Marshal.dump(@sourcedata))
|
1179
|
+
@sourcedata.clear()
|
1180
|
+
end
|
1181
|
+
|
1182
|
+
def _start_content(attrsD)
|
1183
|
+
pushContent('content', attrsD, 'text/plain', true)
|
1184
|
+
src = attrsD['src']
|
1185
|
+
if src and not src.empty?:
|
1186
|
+
@contentparams['src'] = src
|
1187
|
+
end
|
1188
|
+
push('content', true)
|
1189
|
+
end
|
1190
|
+
|
1191
|
+
def _start_prodlink(attrsD)
|
1192
|
+
pushContent('content', attrsD, 'text/html', true)
|
1193
|
+
end
|
1194
|
+
|
1195
|
+
def _start_body(attrsD)
|
1196
|
+
pushContent('content', attrsD, 'application/xhtml+xml', true)
|
1197
|
+
end
|
1198
|
+
alias :_start_xhtml_body :_start_body
|
1199
|
+
|
1200
|
+
def _start_content_encoded(attrsD)
|
1201
|
+
pushContent('content', attrsD, 'text/html', true)
|
1202
|
+
end
|
1203
|
+
alias :_start_fullitem :_start_content_encoded
|
1204
|
+
|
1205
|
+
def _end_content
|
1206
|
+
copyToDescription = (['text/plain'] + @html_types).include? mapContentType(@contentparams['type'])
|
1207
|
+
value = popContent('content')
|
1208
|
+
if copyToDescription
|
1209
|
+
_save('description', value)
|
1210
|
+
end
|
1211
|
+
alias :_end_body :_end_content
|
1212
|
+
alias :_end_xhtml_body :_end_content
|
1213
|
+
alias :_end_content_encoded :_end_content
|
1214
|
+
alias :_end_fullitem :_end_content
|
1215
|
+
alias :_end_prodlink :_end_content
|
1216
|
+
end
|
1217
|
+
|
1218
|
+
def _start_itunes_image(attrsD)
|
1219
|
+
push('itunes_image', false)
|
1220
|
+
getContext()['image'] = FeedParserDict.new({'href' => attrsD['href']})
|
1221
|
+
end
|
1222
|
+
alias :_start_itunes_link :_start_itunes_image
|
1223
|
+
|
1224
|
+
def _end_itunes_block
|
1225
|
+
value = pop('itunes_block', false)
|
1226
|
+
getContext()['itunes_block'] = (value == 'yes') and true or false
|
1227
|
+
end
|
1228
|
+
|
1229
|
+
def _end_itunes_explicit
|
1230
|
+
value = pop('itunes_explicit', false)
|
1231
|
+
getContext()['itunes_explicit'] = (value == 'yes') and true or false
|
1232
|
+
end
|
1233
|
+
end # End FeedParserMixin
|
1234
|
+
|
1235
|
+
|