Easy Html Parser is an AST generator for html/xml documents. EHP is a nice tool to parse html content.
It has a short learning curve compared to other parsers. You don't need to lose time going through massive
documentation to do simple stuff. EHP handles broken html nicely.
EHP has a short learning curve, you can go through some examples, in a few minutes you can implement cool stuff.
python setup.py install
That is all.
from ehp import *
html = Html()
data = '''
<p>
This is a paragraph.
</p>
'''
dom = html.feed(data)
print domOutput:
<p >
This is a paragraph.
</p>
That above example shows how to generate a tree representation for html content. The Html class is the parser class it implements the Html.feed method which is used to build a tree representation for html content that is passed through a string. All classes that map a html entity have a str method.
from ehp import *
data = '''<html><body> <font size="+3" > <p> It is simple.</p>
</font> </body></html>'''
dom = Html().feed(data)
for ind, name, attr in dom.walk():
attr['size'] = '+2'
attr['color'] = 'red'
print domOutput.
<html color="red" size="+2" ><body color="red" size="+2" > <font color="red" size="+2" > <p color="red" size="+2" > It is simple.</p>
</font> </body></html>
The example shows how to walk through all html entities and manipulate their attributes. The variable ind in that example holds the tree representation a html entity that is being visited. The variable name holds the visited html entity name and the attr variable holds a dictionary whose keys are html entity attributes.
from ehp import *
data = '''<body> <em> </em> </body>'''
dom = Html().feed(data)
for ind in dom.find('em'):
x = Data('It is cool')
ind.append(x)
print dom*Output:
<body > <em > It is cool</em> </body>
This example shows how to visit html entities whose names match a given string. Notice it is possible to add new html entites to the html by instantiating the class Tag, XTag, Data, Amp etc. All objects used to represent html entities inherit from python list class so they all have a method list.append.
from ehp import *
html = Html()
data = '''
<body> <em> foo </em> </body>
'''
dom = html.feed(data)
for root, item in dom.find_with_root('em'):
root.remove(item)
print domOutput:
<body > </body>
That example shows how to use the Root.find_with_root method to match html entities with specific html types and how to remove them from the outmost html entity. The Root.find_with_root method returns an iterator holding the outmost html entity for a node being visited.
from ehp import *
data = ''' <body><em> foo </em></body>'''
dom = Html().feed(data)
for ind in dom.find('em'):
x = Tag('font', {'color':'red'})
ind.append(x)
print domOutput:
<body ><em > foo <font color="red" ></font></em></body>
The example shown above examplifies how to insert a new html entity in a tree representation.
from ehp import *
data = '''<html> <body> <em> Hello world. </em> </body> </html>'''
html = Html()
dom = html.feed(data)
for ind in dom.find('em'):
print ind.text()Output:
Hello world.
That one shows the usage of the Root.text method that is used to retrieve all raw data from html entities. The Root class abstracts all common methods of html entities, all classes mapping html entities inherit from Root.
from ehp import *
data = '''<html> <body>
<em style="background:blue"> It is a python. </em>
<p> cool </p></body> </html>'''
dom = Html().feed(data)
for ind in dom.match(('style', 'background:blue')):
print ind.text()Output:
It is a python.
The above example shows how to visit html entities that match a given attribute condition. It is very useful sometimes.
from ehp import *
html = Html().feed(data)
data = '''<tag> The & is a good & symbol. </tag>'''
dom = html.feed(data)
for root, ind in dom.find_with_root(AMP):
print indOutput:
&
&
There are special classes in EHP that represent special html elements. These classes are Amp, Meta, Code, Comment and Data.
The AMP, DATA, META, CODE, COMMENT values are the names of such entities in EHP. One would call the Root.find method with a string or with one of these values to find nodes.
The DATA name means it is raw data, CODE means hexadecimal numeric characters, COMMENT means html comments, Pi means processing instructions like ~~~ <?proc color='red'> ~~~ , META means html doctypes.
There are other handy methods to manipulate a tree html representation, for help on these methods see the EHP built-in docs.
If you have sugestions or need help you can find me at irc.freenode.org channel #vy.
You can contact me through email as well. [email protected]