Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
W
webmagic
Project
Project
Details
Activity
Releases
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
沈俊林
webmagic
Commits
0633ea16
Commit
0633ea16
authored
Nov 10, 2013
by
yihua.huang
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
add javascript support
parent
df8ca8ad
Changes
5
Show whitespace changes
Inline
Side-by-side
Showing
5 changed files
with
116 additions
and
11 deletions
+116
-11
JsScriptProcessor.java
...va/us/codecraft/webmagic/processor/JsScriptProcessor.java
+61
-0
RubyScriptProcessor.java
.../us/codecraft/webmagic/processor/RubyScriptProcessor.java
+11
-11
defines.js
webmagic-scripts/src/main/resources/js/defines.js
+10
-0
oschina.js
webmagic-scripts/src/main/resources/js/oschina.js
+9
-0
TestJsCall.java
...ts/src/test/java/us/codecraft/webmagic/js/TestJsCall.java
+25
-0
No files found.
webmagic-scripts/src/main/java/us/codecraft/webmagic/processor/JsScriptProcessor.java
0 → 100644
View file @
0633ea16
package
us
.
codecraft
.
webmagic
.
processor
;
import
org.apache.commons.io.IOUtils
;
import
us.codecraft.webmagic.Page
;
import
us.codecraft.webmagic.Site
;
import
us.codecraft.webmagic.Spider
;
import
javax.script.ScriptContext
;
import
javax.script.ScriptEngine
;
import
javax.script.ScriptEngineManager
;
import
javax.script.ScriptException
;
import
java.io.IOException
;
import
java.io.InputStream
;
/**
* @author code4crafter@gmail.com
*/
public
class
JsScriptProcessor
implements
PageProcessor
{
private
ScriptEngine
rubyEngine
;
private
String
defines
;
private
String
script
;
public
JsScriptProcessor
(
String
filename
){
ScriptEngineManager
manager
=
new
ScriptEngineManager
();
rubyEngine
=
manager
.
getEngineByName
(
"javascript"
);
InputStream
resourceAsStream
=
this
.
getClass
().
getClassLoader
().
getResourceAsStream
(
"js/defines.js"
);
try
{
defines
=
IOUtils
.
toString
(
resourceAsStream
);
resourceAsStream
=
this
.
getClass
().
getClassLoader
().
getResourceAsStream
(
filename
);
script
=
IOUtils
.
toString
(
resourceAsStream
);
}
catch
(
IOException
e
)
{
e
.
printStackTrace
();
}
}
@Override
public
void
process
(
Page
page
)
{
ScriptContext
context
=
rubyEngine
.
getContext
();
context
.
setAttribute
(
"page"
,
page
,
ScriptContext
.
ENGINE_SCOPE
);
try
{
rubyEngine
.
eval
(
defines
+
script
,
context
);
}
catch
(
ScriptException
e
)
{
e
.
printStackTrace
();
}
}
@Override
public
Site
getSite
()
{
return
Site
.
me
();
}
public
static
void
main
(
String
[]
args
)
{
Spider
.
create
(
new
JsScriptProcessor
(
"js/oschina.js"
)).
addUrl
(
"http://my.oschina.net/flashsword/blog"
).
run
();
}
}
webmagic-scripts/src/main/java/us/codecraft/webmagic/processor/ScriptProcessor.java
→
webmagic-scripts/src/main/java/us/codecraft/webmagic/processor/
Ruby
ScriptProcessor.java
View file @
0633ea16
...
...
@@ -15,39 +15,39 @@ import java.io.InputStream;
/**
* @author code4crafter@gmail.com
*/
public
class
ScriptProcessor
implements
PageProcessor
{
public
class
Ruby
ScriptProcessor
implements
PageProcessor
{
private
ScriptEngine
rubyEngine
;
private
String
defines
;
ScriptProcessor
(){
private
String
script
;
public
RubyScriptProcessor
(
String
filename
){
ScriptEngineManager
manager
=
new
ScriptEngineManager
();
rubyEngine
=
manager
.
getEngineByName
(
"jruby"
);
InputStream
resourceAsStream
=
this
.
getClass
().
getClassLoader
().
getResourceAsStream
(
"ruby/defines.rb"
);
try
{
defines
=
IOUtils
.
toString
(
resourceAsStream
);
resourceAsStream
=
this
.
getClass
().
getClassLoader
().
getResourceAsStream
(
filename
);
script
=
IOUtils
.
toString
(
resourceAsStream
);
}
catch
(
IOException
e
)
{
e
.
printStackTrace
();
}
}
@Override
public
void
process
(
Page
page
)
{
ScriptContext
context
=
rubyEngine
.
getContext
();
context
.
setAttribute
(
"page"
,
page
,
ScriptContext
.
ENGINE_SCOPE
);
String
script
;
try
{
InputStream
resourceAsStream
=
this
.
getClass
().
getClassLoader
().
getResourceAsStream
(
"ruby/oschina.rb"
);
try
{
script
=
IOUtils
.
toString
(
resourceAsStream
);
rubyEngine
.
eval
(
defines
+
script
,
context
);
}
catch
(
IOException
e
)
{
e
.
printStackTrace
();
}
}
catch
(
ScriptException
e
)
{
e
.
printStackTrace
();
}
}
@Override
...
...
@@ -56,6 +56,6 @@ public class ScriptProcessor implements PageProcessor{
}
public
static
void
main
(
String
[]
args
)
{
Spider
.
create
(
new
ScriptProcessor
(
)).
addUrl
(
"http://my.oschina.net/flashsword/blog"
).
run
();
Spider
.
create
(
new
RubyScriptProcessor
(
"ruby/oschina.rb"
)).
addUrl
(
"http://my.oschina.net/flashsword/blog"
).
run
();
}
}
webmagic-scripts/src/main/resources/js/defines.js
0 → 100644
View file @
0633ea16
function
$
(
str
){
return
page
.
getHtml
().
$
(
str
).
toString
();
}
function
xpath
(
str
){
return
page
.
getHtml
().
xpath
(
str
).
toString
();
}
function
urls
(
str
){
links
=
page
.
getHtml
().
links
().
regex
(
str
).
all
();
page
.
addTargetRequests
(
links
);
}
\ No newline at end of file
webmagic-scripts/src/main/resources/js/oschina.js
0 → 100644
View file @
0633ea16
var
result
=
{
title
:
$
(
"div.BlogTitle h1"
),
content
:
$
(
"div.BlogContent"
)
}
var
config
=
{
ua
:
''
,
sleepTime
:
20
}
urls
(
"http://my
\\
.oschina
\\
.net/flashsword/blog/
\\
d+"
)
\ No newline at end of file
webmagic-scripts/src/test/java/us/codecraft/webmagic/js/TestJsCall.java
0 → 100644
View file @
0633ea16
package
us
.
codecraft
.
webmagic
.
js
;
import
org.junit.Test
;
import
javax.script.ScriptContext
;
import
javax.script.ScriptEngine
;
import
javax.script.ScriptEngineManager
;
import
javax.script.ScriptException
;
/**
* @author code4crafter@gmail.com
*/
public
class
TestJsCall
{
@Test
public
void
test
()
throws
ScriptException
{
ScriptEngineManager
manager
=
new
ScriptEngineManager
();
ScriptEngine
rubyEngine
=
manager
.
getEngineByName
(
"javascript"
);
ScriptContext
context
=
rubyEngine
.
getContext
();
context
.
setAttribute
(
"a"
,
"sad"
,
ScriptContext
.
ENGINE_SCOPE
);
// rubyEngine.eval("", context);
rubyEngine
.
eval
(
"print(a)"
,
context
);
}
}
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment