hpricot 0.6-jruby
Sign up to get free protection for your applications and to get access to all the features.
- data/CHANGELOG +62 -0
- data/COPYING +18 -0
- data/README +284 -0
- data/Rakefile +211 -0
- data/ext/hpricot_scan/HpricotScanService.java +1340 -0
- data/ext/hpricot_scan/extconf.rb +6 -0
- data/ext/hpricot_scan/hpricot_common.rl +76 -0
- data/ext/hpricot_scan/hpricot_scan.c +5976 -0
- data/ext/hpricot_scan/hpricot_scan.h +79 -0
- data/ext/hpricot_scan/hpricot_scan.java.rl +363 -0
- data/ext/hpricot_scan/hpricot_scan.rl +273 -0
- data/extras/mingw-rbconfig.rb +176 -0
- data/lib/hpricot.rb +26 -0
- data/lib/hpricot/blankslate.rb +63 -0
- data/lib/hpricot/builder.rb +200 -0
- data/lib/hpricot/elements.rb +510 -0
- data/lib/hpricot/htmlinfo.rb +672 -0
- data/lib/hpricot/inspect.rb +107 -0
- data/lib/hpricot/modules.rb +37 -0
- data/lib/hpricot/parse.rb +297 -0
- data/lib/hpricot/tag.rb +228 -0
- data/lib/hpricot/tags.rb +164 -0
- data/lib/hpricot/traverse.rb +821 -0
- data/lib/hpricot/xchar.rb +94 -0
- data/lib/i686-linux/hpricot_scan.jar +0 -0
- data/test/files/basic.xhtml +17 -0
- data/test/files/boingboing.html +2266 -0
- data/test/files/cy0.html +3653 -0
- data/test/files/immob.html +400 -0
- data/test/files/pace_application.html +1320 -0
- data/test/files/tenderlove.html +16 -0
- data/test/files/uswebgen.html +220 -0
- data/test/files/utf8.html +1054 -0
- data/test/files/week9.html +1723 -0
- data/test/files/why.xml +19 -0
- data/test/load_files.rb +7 -0
- data/test/test_alter.rb +65 -0
- data/test/test_builder.rb +24 -0
- data/test/test_parser.rb +379 -0
- data/test/test_paths.rb +16 -0
- data/test/test_preserved.rb +66 -0
- data/test/test_xml.rb +28 -0
- metadata +98 -0
data/CHANGELOG
ADDED
@@ -0,0 +1,62 @@
|
|
1
|
+
= 0.6
|
2
|
+
=== 15th June, 2007
|
3
|
+
* Hpricot for JRuby -- nice work Ola Bini!
|
4
|
+
* Inline Markaby for Hpricot documents.
|
5
|
+
* XML tags and attributes are no longer downcased like HTML is.
|
6
|
+
* new syntax for grabbing everything between two elements using a Range in the search method: (doc/("font".."font/br")) or in nodes_at like so: (doc/"font").nodes_at("*".."br"). Only works with either a pair of siblings or a set of a parent and a sibling.
|
7
|
+
* Ignore self-closing endings on tags (such as form) which are containers. Treat them like open parent tags. Reported by Jonathan Nichols on the hpricot list.
|
8
|
+
* Escaping of attributes, yanked from Jim Weirich and Sam Ruby's work in Builder.
|
9
|
+
* Element#raw_attributes gives unescaped data. Element#attributes gives escaped.
|
10
|
+
* Added: Elements#attr, Elements#remove_attr, Elements#remove_class.
|
11
|
+
* Added: Traverse#preceding, Traverse#following, Traverse#previous, Traverse#next.
|
12
|
+
|
13
|
+
= 0.5
|
14
|
+
=== 31rd January, 2007
|
15
|
+
|
16
|
+
* support for a[text()="Click Me!"] and h3[text()*="space"] and the like.
|
17
|
+
* Hpricot.buffer_size accessor for increasing Hpricot's buffer if you're encountering huge ASP.NET viewstate attribs.
|
18
|
+
* some support for colons in tag names (not full namespace support yet.)
|
19
|
+
* Element.to_original_html will attempt to preserve the original HTML while merging your changes.
|
20
|
+
* Element.to_plain_text converts an element's contents to a simple text format.
|
21
|
+
* Element.inner_text removes all tags and returns text nodes concatenated into a single string.
|
22
|
+
* no @raw_string variable kept for comments, text, and cdata -- as it's redundant.
|
23
|
+
* xpath-style indices (//p/a[1]) but keep in mind that they aren't zero-based.
|
24
|
+
* node_position is the index among all sibling nodes, while position is the position among children of identical type.
|
25
|
+
* comment() and text() search criteria, like: //p/text(), which selects all text inside paragraph tags.
|
26
|
+
* every element has css_path and xpath methods which return respective absolute paths.
|
27
|
+
* more flexibility all around: in parsing attributes, tags, comments and cdata.
|
28
|
+
|
29
|
+
= 0.4
|
30
|
+
=== 11th August, 2006
|
31
|
+
|
32
|
+
* The :fixup_tags option will try to sort out the hierarchy so elements end up with the right parents.
|
33
|
+
* Elements such as *script* and *style* (identified as having CDATA contents) receive a single text node as their children now. Previously, Hpricot was parsing out tags found in scripts.
|
34
|
+
* Better scanning of partially quoted attributes (found by Brent Beardsly on http://uswebgen.com/)
|
35
|
+
* Better scanning of unquoted attributes -- thanks to Aaron Patterson for the test cases!
|
36
|
+
* Some tags were being output in the empty tag style, although browsers hated that. FIXED!
|
37
|
+
* Added Elements#at for finding single elements.
|
38
|
+
* Added Elem::Trav#[] and Elem::Trav#[]= for reading and writing attributes.
|
39
|
+
|
40
|
+
= 0.3
|
41
|
+
=== 7th July, 2006
|
42
|
+
|
43
|
+
* Fixed negative string size error on empty tokens. (news.bbc.co.uk)
|
44
|
+
* Allow the parser to accept just text nodes. (such as: <tt>Hpricot.parse('TEXT')</tt>)
|
45
|
+
* from JQuery to Hpricot::Elements: remove, empty, append, prepend, before, after, wrap, set,
|
46
|
+
html(...), to_html, to_s.
|
47
|
+
* on containers: to_html, replace_child, insert_before, insert_after, innerHTML=.
|
48
|
+
* Hpricot(...) is an alias for parse.
|
49
|
+
* open up all properties to setters, let people do as they may.
|
50
|
+
* use to_html for the full html of a node or set of elements.
|
51
|
+
* doctypes were messed.
|
52
|
+
|
53
|
+
= 0.2
|
54
|
+
=== 4th July, 2006
|
55
|
+
|
56
|
+
* Rewrote the HTree parser to be simpler, more adequate for the common man. Will add encoding back in later.
|
57
|
+
|
58
|
+
= 0.1
|
59
|
+
=== 3rd July, 2006
|
60
|
+
|
61
|
+
* For whatever reason, wrote this HTML parser in C.
|
62
|
+
I guess Ragel is addictive and I want to improve HTree.
|
data/COPYING
ADDED
@@ -0,0 +1,18 @@
|
|
1
|
+
Copyright (c) 2006 why the lucky stiff
|
2
|
+
|
3
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
4
|
+
of this software and associated documentation files (the "Software"), to
|
5
|
+
deal in the Software without restriction, including without limitation the
|
6
|
+
rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
|
7
|
+
sell copies of the Software, and to permit persons to whom the Software is
|
8
|
+
furnished to do so, subject to the following conditions:
|
9
|
+
|
10
|
+
The above copyright notice and this permission notice shall be included in
|
11
|
+
all copies or substantial portions of the Software.
|
12
|
+
|
13
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
14
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
15
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
16
|
+
THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
|
17
|
+
IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
|
18
|
+
CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README
ADDED
@@ -0,0 +1,284 @@
|
|
1
|
+
= Hpricot, Read Any HTML
|
2
|
+
|
3
|
+
Hpricot is a fast, flexible HTML parser written in C. It's designed to be very
|
4
|
+
accommodating (like Tanaka Akira's HTree) and to have a very helpful library
|
5
|
+
(like some JavaScript libs -- JQuery, Prototype -- give you.) The XPath and CSS
|
6
|
+
parser, in fact, is based on John Resig's JQuery.
|
7
|
+
|
8
|
+
Also, Hpricot can be handy for reading broken XML files, since many of the same
|
9
|
+
techniques can be used. If a quote is missing, Hpricot tries to figure it out.
|
10
|
+
If tags overlap, Hpricot works on sorting them out. You know, that sort of
|
11
|
+
thing.
|
12
|
+
|
13
|
+
*Please read this entire document* before making assumptions about how this
|
14
|
+
software works.
|
15
|
+
|
16
|
+
== An Overview
|
17
|
+
|
18
|
+
Let's clear up what Hpricot is.
|
19
|
+
|
20
|
+
# Hpricot is *a standalone library*. It requires no other libraries. Just Ruby!
|
21
|
+
# While priding itself on speed, Hpricot *works hard to sort out bad HTML* and
|
22
|
+
pays a small penalty in order to get that right. So that's slightly more important
|
23
|
+
to me than speed.
|
24
|
+
# *If you can see it in Firefox, then Hpricot should parse it.* That's
|
25
|
+
how it should be! Let me know the minute it's otherwise.
|
26
|
+
# Primarily, Hpricot is used for reading HTML and tries to sort out troubled
|
27
|
+
HTML by having some idea of what good HTML is. Some people still like to use
|
28
|
+
Hpricot for XML reading, but *remember to use the Hpricot::XML() method* for that!
|
29
|
+
|
30
|
+
== The Hpricot Kingdom
|
31
|
+
|
32
|
+
First, here are all the links you need to know:
|
33
|
+
|
34
|
+
* http://code.whytheluckystiff.net/hpricot is the Hpricot wiki and bug tracker.
|
35
|
+
Go there for news and recipes and patches. It's the center of activity.
|
36
|
+
* http://code.whytheluckystiff.net/svn/hpricot/trunk is the main Subversion
|
37
|
+
repository for Hpricot. You can get the latest code there.
|
38
|
+
* http://code.whytheluckystiff.net/doc/hpricot is the home for the latest copy of
|
39
|
+
this reference.
|
40
|
+
* See COPYING for the terms of this software. (Spoiler: it's absolutely free.)
|
41
|
+
|
42
|
+
If you have any trouble, don't hesitate to contact the author. As always, I'm
|
43
|
+
not going to say "Use at your own risk" because I don't want this library to be
|
44
|
+
risky. If you trip on something, I'll share the liability by repairing things
|
45
|
+
as quickly as I can. Your responsibility is to report the inadequacies.
|
46
|
+
|
47
|
+
== Installing Hpricot
|
48
|
+
|
49
|
+
You may get the latest stable version from Rubyforge. Win32 binaries and source
|
50
|
+
gems are available.
|
51
|
+
|
52
|
+
$ gem install hpricot
|
53
|
+
|
54
|
+
As Hpricot is still under active development, you can also try the most recent
|
55
|
+
candidate build here:
|
56
|
+
|
57
|
+
$ gem install hpricot --source http://code.whytheluckystiff.net
|
58
|
+
|
59
|
+
The development gem is usually in pretty good shape actually. You can also
|
60
|
+
get the bleeding edge code or plain Ruby tarballs on the wiki.
|
61
|
+
|
62
|
+
== An Hpricot Showcase
|
63
|
+
|
64
|
+
We're going to run through a big pile of examples to get you jump-started.
|
65
|
+
Many of these examples are also found at
|
66
|
+
http://code.whytheluckystiff.net/hpricot/wiki/HpricotBasics, in case you
|
67
|
+
want to add some of your own.
|
68
|
+
|
69
|
+
=== Loading Hpricot Itself
|
70
|
+
|
71
|
+
You have probably got the gem, right? To load Hpricot:
|
72
|
+
|
73
|
+
require 'rubygems'
|
74
|
+
require 'hpricot'
|
75
|
+
|
76
|
+
If you've installed the plain source distribution, go ahead and just:
|
77
|
+
|
78
|
+
require 'hpricot'
|
79
|
+
|
80
|
+
=== Load an HTML Page
|
81
|
+
|
82
|
+
The <tt>Hpricot()</tt> method takes a string or any IO object and loads the
|
83
|
+
contents into a document object.
|
84
|
+
|
85
|
+
doc = Hpricot("<p>A simple <b>test</b> string.</p>")
|
86
|
+
|
87
|
+
To load from a file, just get the stream open:
|
88
|
+
|
89
|
+
doc = open("index.html") { |f| Hpricot(f) }
|
90
|
+
|
91
|
+
To load from a web URL, use <tt>open-uri</tt>, which comes with Ruby:
|
92
|
+
|
93
|
+
require 'open-uri'
|
94
|
+
doc = open("http://qwantz.com/") { |f| Hpricot(f) }
|
95
|
+
|
96
|
+
Hpricot uses an internal buffer to parse the file, so the IO will stream
|
97
|
+
properly and large documents won't be loaded into memory all at once. However,
|
98
|
+
the parsed document object will be present in memory, in its entirety.
|
99
|
+
|
100
|
+
=== Search for Elements
|
101
|
+
|
102
|
+
Use <tt>Doc.search</tt>:
|
103
|
+
|
104
|
+
doc.search("//p[@class='posted']")
|
105
|
+
#=> #<Hpricot:Elements[{p ...}, {p ...}]>
|
106
|
+
|
107
|
+
<tt>Doc.search</tt> can take an XPath or CSS expression. In the above example,
|
108
|
+
all paragraph <tt><p></tt> elements are grabbed which have a <tt>class</tt>
|
109
|
+
attribute of <tt>"posted"</tt>.
|
110
|
+
|
111
|
+
A shortcut is to use the divisor:
|
112
|
+
|
113
|
+
(doc/"p.posted")
|
114
|
+
#=> #<Hpricot:Elements[{p ...}, {p ...}]>
|
115
|
+
|
116
|
+
=== Finding Just One Element
|
117
|
+
|
118
|
+
If you're looking for a single element, the <tt>at</tt> method will return the
|
119
|
+
first element matched by the expression. In this case, you'll get back the
|
120
|
+
element itself rather than the <tt>Hpricot::Elements</tt> array.
|
121
|
+
|
122
|
+
doc.at("body")['onload']
|
123
|
+
|
124
|
+
The above code will find the body tag and give you back the <tt>onload</tt>
|
125
|
+
attribute. This is the most common reason to use the element directly: when
|
126
|
+
reading and writing HTML attributes.
|
127
|
+
|
128
|
+
=== Fetching the Contents of an Element
|
129
|
+
|
130
|
+
Just as with browser scripting, the <tt>inner_html</tt> property can be used to
|
131
|
+
get the inner contents of an element.
|
132
|
+
|
133
|
+
(doc/"#elementID").inner_html
|
134
|
+
#=> "..<b>contents</b>.."
|
135
|
+
|
136
|
+
If your expression matches more than one element, you'll get back the contents
|
137
|
+
of ''all the matched elements''. So you may want to use <tt>first</tt> to be
|
138
|
+
sure you get back only one.
|
139
|
+
|
140
|
+
(doc/"#elementID").first.inner_html
|
141
|
+
#=> "..<b>contents</b>.."
|
142
|
+
|
143
|
+
=== Fetching the HTML for an Element
|
144
|
+
|
145
|
+
If you want the HTML for the whole element (not just the contents), use
|
146
|
+
<tt>to_html</tt>:
|
147
|
+
|
148
|
+
(doc/"#elementID").to_html
|
149
|
+
#=> "<div id='elementID'>...</div>"
|
150
|
+
|
151
|
+
=== Looping
|
152
|
+
|
153
|
+
All searches return a set of <tt>Hpricot::Elements</tt>. Go ahead and loop
|
154
|
+
through them like you would an array.
|
155
|
+
|
156
|
+
(doc/"p/a/img").each do |img|
|
157
|
+
puts img.attributes['class']
|
158
|
+
end
|
159
|
+
|
160
|
+
=== Continuing Searches
|
161
|
+
|
162
|
+
Searches can be continued from a collection of elements, in order to search deeper.
|
163
|
+
|
164
|
+
# find all paragraphs.
|
165
|
+
elements = doc.search("/html/body//p")
|
166
|
+
# continue the search by finding any images within those paragraphs.
|
167
|
+
(elements/"img")
|
168
|
+
#=> #<Hpricot::Elements[{img ...}, {img ...}]>
|
169
|
+
|
170
|
+
Searches can also be continued by searching within container elements.
|
171
|
+
|
172
|
+
# find all images within paragraphs.
|
173
|
+
doc.search("/html/body//p").each do |para|
|
174
|
+
puts "== Found a paragraph =="
|
175
|
+
pp para
|
176
|
+
|
177
|
+
imgs = para.search("img")
|
178
|
+
if imgs.any?
|
179
|
+
puts "== Found #{imgs.length} images inside =="
|
180
|
+
end
|
181
|
+
end
|
182
|
+
|
183
|
+
Of course, the most succinct ways to do the above are using CSS or XPath.
|
184
|
+
|
185
|
+
# the xpath version
|
186
|
+
(doc/"/html/body//p//img")
|
187
|
+
# the css version
|
188
|
+
(doc/"html > body > p img")
|
189
|
+
# ..or symbols work, too!
|
190
|
+
(doc/:html/:body/:p/:img)
|
191
|
+
|
192
|
+
=== Looping Edits
|
193
|
+
|
194
|
+
You may certainly edit objects from within your search loops. Then, when you
|
195
|
+
spit out the HTML, the altered elements will show.
|
196
|
+
|
197
|
+
(doc/"span.entryPermalink").each do |span|
|
198
|
+
span.attributes['class'] = 'newLinks'
|
199
|
+
end
|
200
|
+
puts doc
|
201
|
+
|
202
|
+
This changes all <tt>span.entryPermalink</tt> elements to
|
203
|
+
<tt>span.newLinks</tt>. Keep in mind that there are often more convenient ways
|
204
|
+
of doing this. Such as the <tt>set</tt> method:
|
205
|
+
|
206
|
+
(doc/"span.entryPermalink").set(:class => 'newLinks')
|
207
|
+
|
208
|
+
=== Figuring Out Paths
|
209
|
+
|
210
|
+
Every element can tell you its unique path (either XPath or CSS) to get to the
|
211
|
+
element from the root tag.
|
212
|
+
|
213
|
+
The <tt>css_path</tt> method:
|
214
|
+
|
215
|
+
doc.at("div > div:nth(1)").css_path
|
216
|
+
#=> "div > div:nth(1)"
|
217
|
+
doc.at("#header").css_path
|
218
|
+
#=> "#header"
|
219
|
+
|
220
|
+
Or, the <tt>xpath</tt> method:
|
221
|
+
|
222
|
+
doc.at("div > div:nth(1)").xpath
|
223
|
+
#=> "/div/div:eq(1)"
|
224
|
+
doc.at("#header").xpath
|
225
|
+
#=> "//div[@id='header']"
|
226
|
+
|
227
|
+
== Hpricot Fixups
|
228
|
+
|
229
|
+
When loading HTML documents, you have a few settings that can make Hpricot more
|
230
|
+
or less intense about how it gets involved.
|
231
|
+
|
232
|
+
== :fixup_tags
|
233
|
+
|
234
|
+
Really, there are so many ways to clean up HTML and your intentions may be to
|
235
|
+
keep the HTML as-is. So Hpricot's default behavior is to keep things flexible.
|
236
|
+
Making sure to open and close all the tags, but ignore any validation problems.
|
237
|
+
|
238
|
+
As of Hpricot 0.4, there's a new <tt>:fixup_tags</tt> option which will attempt
|
239
|
+
to shift the document's tags to meet XHTML 1.0 Strict.
|
240
|
+
|
241
|
+
doc = open("index.html") { |f| Hpricot f, :fixup_tags => true }
|
242
|
+
|
243
|
+
This doesn't quite meet the XHTML 1.0 Strict standard, it just tries to follow
|
244
|
+
the rules a bit better. Like: say Hpricot finds a paragraph in a link, it's
|
245
|
+
going to move the paragraph below the link. Or up and out of other elements
|
246
|
+
where paragraphs don't belong.
|
247
|
+
|
248
|
+
If an unknown element is found, it is ignored. Again, <tt>:fixup_tags</tt>.
|
249
|
+
|
250
|
+
== :xhtml_strict
|
251
|
+
|
252
|
+
So, let's go beyond just trying to fix the hierarchy. The
|
253
|
+
<tt>:xhtml_strict</tt> option really tries to force the document to be an XHTML
|
254
|
+
1.0 Strict document. Even at the cost of removing elements that get in the way.
|
255
|
+
|
256
|
+
doc = open("index.html") { |f| Hpricot f, :xhtml_strict => true }
|
257
|
+
|
258
|
+
What measures does <tt>:xhtml_strict</tt> take?
|
259
|
+
|
260
|
+
1. Shift elements into their proper containers just like :fixup_tags.
|
261
|
+
2. Remove unknown elements.
|
262
|
+
3. Remove unknown attributes.
|
263
|
+
4. Remove illegal content.
|
264
|
+
5. Alter the doctype to XHTML 1.0 Strict.
|
265
|
+
|
266
|
+
== Hpricot.XML()
|
267
|
+
|
268
|
+
The last option is the <tt>:xml</tt> option, which makes some slight variations
|
269
|
+
on the standard mode. The main difference is that :xml mode won't try to output
|
270
|
+
tags which are friendlier for browsers. For example, if an opening and closing
|
271
|
+
<tt>br</tt> tag is found, XML mode won't try to turn that into an empty element.
|
272
|
+
|
273
|
+
XML mode also doesn't downcase the tags and attributes for you. So pay attention
|
274
|
+
to case, friends.
|
275
|
+
|
276
|
+
The primary way to use Hpricot's XML mode is to call the Hpricot.XML method:
|
277
|
+
|
278
|
+
doc = open("http://redhanded.hobix.com/index.xml") do |f|
|
279
|
+
Hpricot.XML(f)
|
280
|
+
end
|
281
|
+
|
282
|
+
*Also, :fixup_tags is canceled out by the :xml option.* This is because
|
283
|
+
:fixup_tags makes assumptions based how HTML is structured. Specifically, how
|
284
|
+
tags are defined in the XHTML 1.0 DTD.
|
data/Rakefile
ADDED
@@ -0,0 +1,211 @@
|
|
1
|
+
require 'rake'
|
2
|
+
require 'rake/clean'
|
3
|
+
require 'rake/gempackagetask'
|
4
|
+
require 'rake/rdoctask'
|
5
|
+
require 'rake/testtask'
|
6
|
+
require 'fileutils'
|
7
|
+
include FileUtils
|
8
|
+
|
9
|
+
NAME = "hpricot"
|
10
|
+
REV = `svn info`[/Revision: (\d+)/, 1] rescue nil
|
11
|
+
VERS = ENV['VERSION'] || "0.6" + (REV ? ".#{REV}" : "")
|
12
|
+
PKG = "#{NAME}-#{VERS}"
|
13
|
+
BIN = "*.{bundle,jar,so,obj,pdb,lib,def,exp}"
|
14
|
+
ARCHLIB = "lib/#{::Config::CONFIG['arch']}"
|
15
|
+
CLEAN.include ["ext/hpricot_scan/#{BIN}", "lib/**/#{BIN}", 'ext/hpricot_scan/Makefile',
|
16
|
+
'**/.*.sw?', '*.gem', '.config']
|
17
|
+
RDOC_OPTS = ['--quiet', '--title', 'The Hpricot Reference', '--main', 'README', '--inline-source']
|
18
|
+
PKG_FILES = %w(CHANGELOG COPYING README Rakefile) +
|
19
|
+
Dir.glob("{bin,doc,test,lib,extras}/**/*") +
|
20
|
+
Dir.glob("ext/**/*.{h,java,c,rb,rl}") +
|
21
|
+
%w[ext/hpricot_scan/hpricot_scan.c] # needed because it's generated later
|
22
|
+
SPEC =
|
23
|
+
Gem::Specification.new do |s|
|
24
|
+
s.name = NAME
|
25
|
+
s.version = VERS
|
26
|
+
s.platform = Gem::Platform::RUBY
|
27
|
+
s.has_rdoc = true
|
28
|
+
s.rdoc_options += RDOC_OPTS
|
29
|
+
s.extra_rdoc_files = ["README", "CHANGELOG", "COPYING"]
|
30
|
+
s.summary = "a swift, liberal HTML parser with a fantastic library"
|
31
|
+
s.description = s.summary
|
32
|
+
s.author = "why the lucky stiff"
|
33
|
+
s.email = 'why@ruby-lang.org'
|
34
|
+
s.homepage = 'http://code.whytheluckystiff.net/hpricot/'
|
35
|
+
s.files = PKG_FILES
|
36
|
+
s.require_paths = [ARCHLIB, "lib"]
|
37
|
+
s.extensions = FileList["ext/**/extconf.rb"].to_a
|
38
|
+
s.bindir = "bin"
|
39
|
+
end
|
40
|
+
|
41
|
+
desc "Does a full compile, test run"
|
42
|
+
task :default => [:compile, :test]
|
43
|
+
|
44
|
+
desc "Packages up Hpricot."
|
45
|
+
task :package => [:clean, :ragel]
|
46
|
+
|
47
|
+
desc "Releases packages for all Hpricot packages and platforms."
|
48
|
+
task :release => [:package, :package_win32, :package_jruby]
|
49
|
+
|
50
|
+
desc "Run all the tests"
|
51
|
+
Rake::TestTask.new do |t|
|
52
|
+
t.libs << "test" << ARCHLIB
|
53
|
+
t.test_files = FileList['test/test_*.rb']
|
54
|
+
t.verbose = true
|
55
|
+
end
|
56
|
+
|
57
|
+
Rake::RDocTask.new do |rdoc|
|
58
|
+
rdoc.rdoc_dir = 'doc/rdoc'
|
59
|
+
rdoc.options += RDOC_OPTS
|
60
|
+
rdoc.main = "README"
|
61
|
+
rdoc.rdoc_files.add ['README', 'CHANGELOG', 'COPYING', 'lib/**/*.rb']
|
62
|
+
end
|
63
|
+
|
64
|
+
Rake::GemPackageTask.new(SPEC) do |p|
|
65
|
+
p.need_tar = true
|
66
|
+
p.gem_spec = SPEC
|
67
|
+
end
|
68
|
+
|
69
|
+
extension = "hpricot_scan"
|
70
|
+
ext = "ext/hpricot_scan"
|
71
|
+
ext_so = "#{ext}/#{extension}.#{Config::CONFIG['DLEXT']}"
|
72
|
+
ext_files = FileList[
|
73
|
+
"#{ext}/*.c",
|
74
|
+
"#{ext}/*.h",
|
75
|
+
"#{ext}/*.rl",
|
76
|
+
"#{ext}/extconf.rb",
|
77
|
+
"#{ext}/Makefile",
|
78
|
+
"lib"
|
79
|
+
]
|
80
|
+
|
81
|
+
task "lib" do
|
82
|
+
directory "lib"
|
83
|
+
end
|
84
|
+
|
85
|
+
desc "Compiles the Ruby extension"
|
86
|
+
task :compile => [:hpricot_scan] do
|
87
|
+
if Dir.glob(File.join(ARCHLIB,"hpricot_scan.*")).length == 0
|
88
|
+
STDERR.puts "!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"
|
89
|
+
STDERR.puts "Gem actually failed to build. Your system is"
|
90
|
+
STDERR.puts "NOT configured properly to build hpricot."
|
91
|
+
STDERR.puts "!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"
|
92
|
+
exit(1)
|
93
|
+
end
|
94
|
+
end
|
95
|
+
task :hpricot_scan => [:ragel]
|
96
|
+
|
97
|
+
desc "Builds just the #{extension} extension"
|
98
|
+
task extension.to_sym => ["#{ext}/Makefile", ext_so ]
|
99
|
+
|
100
|
+
file "#{ext}/Makefile" => ["#{ext}/extconf.rb"] do
|
101
|
+
Dir.chdir(ext) do ruby "extconf.rb" end
|
102
|
+
end
|
103
|
+
|
104
|
+
file ext_so => ext_files do
|
105
|
+
Dir.chdir(ext) do
|
106
|
+
sh(PLATFORM =~ /win32/ ? 'nmake' : 'make')
|
107
|
+
end
|
108
|
+
mkdir_p ARCHLIB
|
109
|
+
cp ext_so, ARCHLIB
|
110
|
+
end
|
111
|
+
|
112
|
+
desc "returns the ragel version"
|
113
|
+
task :ragel_version do
|
114
|
+
@ragel_v = `ragel -v`[/(version )(\S*)/,2].to_f
|
115
|
+
end
|
116
|
+
|
117
|
+
desc "Generates the C scanner code with Ragel."
|
118
|
+
task :ragel => [:ragel_version] do
|
119
|
+
sh %{ragel ext/hpricot_scan/hpricot_scan.rl | #{@ragel_v >= 5.18 ? 'rlgen-cd' : 'rlcodegen'} -G2 -o ext/hpricot_scan/hpricot_scan.c}
|
120
|
+
end
|
121
|
+
|
122
|
+
desc "Generates the Java scanner code with Ragel."
|
123
|
+
task :ragel_java => [:ragel_version] do
|
124
|
+
sh %{ragel -J ext/hpricot_scan/hpricot_scan.java.rl | #{@ragel_v >= 5.18 ? 'rlgen-java' : 'rlcodegen'} -o ext/hpricot_scan/HpricotScanService.java}
|
125
|
+
end
|
126
|
+
|
127
|
+
### Win32 Packages ###
|
128
|
+
|
129
|
+
Win32Spec = SPEC.dup
|
130
|
+
Win32Spec.platform = Gem::Platform::WIN32
|
131
|
+
Win32Spec.files = PKG_FILES + ["#{ARCHLIB}/hpricot_scan.so"]
|
132
|
+
Win32Spec.extensions = []
|
133
|
+
|
134
|
+
WIN32_PKG_DIR = "#{PKG}-mswin32"
|
135
|
+
|
136
|
+
desc "Package up the Win32 distribution."
|
137
|
+
file WIN32_PKG_DIR => [:package] do
|
138
|
+
sh "tar zxf pkg/#{PKG}.tgz"
|
139
|
+
mv PKG, WIN32_PKG_DIR
|
140
|
+
end
|
141
|
+
|
142
|
+
desc "Cross-compile the hpricot_scan extension for win32"
|
143
|
+
file "hpricot_scan_win32" => [WIN32_PKG_DIR] do
|
144
|
+
cp "extras/mingw-rbconfig.rb", "#{WIN32_PKG_DIR}/ext/hpricot_scan/rbconfig.rb"
|
145
|
+
sh "cd #{WIN32_PKG_DIR}/ext/hpricot_scan/ && ruby -I. extconf.rb && make"
|
146
|
+
mv "#{WIN32_PKG_DIR}/ext/hpricot_scan/hpricot_scan.so", "#{WIN32_PKG_DIR}/#{ARCHLIB}"
|
147
|
+
end
|
148
|
+
|
149
|
+
desc "Build the binary RubyGems package for win32"
|
150
|
+
task :package_win32 => ["hpricot_scan_win32"] do
|
151
|
+
Dir.chdir("#{WIN32_PKG_DIR}") do
|
152
|
+
Gem::Builder.new(Win32Spec).build
|
153
|
+
verbose(true) {
|
154
|
+
mv Dir["*.gem"].first, "../pkg/#{WIN32_PKG_DIR}.gem"
|
155
|
+
}
|
156
|
+
end
|
157
|
+
end
|
158
|
+
|
159
|
+
CLEAN.include WIN32_PKG_DIR
|
160
|
+
|
161
|
+
### JRuby Packages ###
|
162
|
+
|
163
|
+
compile_java = proc do
|
164
|
+
sh %{javac -source 1.4 -target 1.4 -classpath $JRUBY_HOME/lib/jruby.jar HpricotScanService.java}
|
165
|
+
sh %{jar cf hpricot_scan.jar HpricotScanService.class}
|
166
|
+
end
|
167
|
+
|
168
|
+
desc "Compiles the JRuby extension"
|
169
|
+
task :hpricot_scan_java => [:ragel_java] do
|
170
|
+
Dir.chdir("ext/hpricot_scan", &compile_java)
|
171
|
+
end
|
172
|
+
|
173
|
+
JRubySpec = SPEC.dup
|
174
|
+
JRubySpec.platform = 'jruby'
|
175
|
+
JRubySpec.files = PKG_FILES + ["#{ARCHLIB}/hpricot_scan.jar"]
|
176
|
+
JRubySpec.extensions = []
|
177
|
+
|
178
|
+
JRUBY_PKG_DIR = "#{PKG}-jruby"
|
179
|
+
|
180
|
+
desc "Package up the JRuby distribution."
|
181
|
+
file JRUBY_PKG_DIR => [:ragel_java, :package] do
|
182
|
+
sh "tar zxf pkg/#{PKG}.tgz"
|
183
|
+
mv PKG, JRUBY_PKG_DIR
|
184
|
+
end
|
185
|
+
|
186
|
+
desc "Cross-compile the hpricot_scan extension for JRuby"
|
187
|
+
file "hpricot_scan_jruby" => [JRUBY_PKG_DIR] do
|
188
|
+
Dir.chdir("#{JRUBY_PKG_DIR}/ext/hpricot_scan", &compile_java)
|
189
|
+
mv "#{JRUBY_PKG_DIR}/ext/hpricot_scan/hpricot_scan.jar", "#{JRUBY_PKG_DIR}/#{ARCHLIB}"
|
190
|
+
end
|
191
|
+
|
192
|
+
desc "Build the RubyGems package for JRuby"
|
193
|
+
task :package_jruby => ["hpricot_scan_jruby"] do
|
194
|
+
Dir.chdir("#{JRUBY_PKG_DIR}") do
|
195
|
+
Gem::Builder.new(JRubySpec).build
|
196
|
+
verbose(true) {
|
197
|
+
mv Dir["*.gem"].first, "../pkg/#{JRUBY_PKG_DIR}.gem"
|
198
|
+
}
|
199
|
+
end
|
200
|
+
end
|
201
|
+
|
202
|
+
CLEAN.include JRUBY_PKG_DIR
|
203
|
+
|
204
|
+
task :install do
|
205
|
+
sh %{rake package}
|
206
|
+
sh %{sudo gem install pkg/#{NAME}-#{VERS}}
|
207
|
+
end
|
208
|
+
|
209
|
+
task :uninstall => [:clean] do
|
210
|
+
sh %{sudo gem uninstall #{NAME}}
|
211
|
+
end
|