Boost HTML Library - Release of the Boost HTML library, a cross-plat... | www.ibusy.com | HTML C++ Parser Library hash string_algo multi_index shared

Boost HTML Library

(2 votes)

Written by Andreas Haberstroh

Monday, 03 September 2007

Release of the Boost HTML library, a cross-platform C++ library for reading and writing HTML documents.

License

Boost-HTML is released under the Boost Software License.

Dependencies

Boost::HTML requires Boost version 1.33 or later.

Downloading

The Boost-HTML library is available here.

Using Boost-HTML

The Boost-HTML library is relatively simplistic in design. All read and write operations are based on streams.

#include <boost/html/document.hpp>
#include <iostream>

#include <fstream>
using namespace std;
int main(int argc, char* argv[])
{
fstream ifile("index.html", ios_base::in);
boost::html::document htmlDOM;

htmlDOM.read(ifile);

htmlDOM.write(cout);

return 0;
}

Why did I write Boost-HTML?

I was searching the Internet for a c++ HTML parser library that would create a tree of HTML elements and their attributes. I was looking for an c++ interface similar to JavaScript's DOM. I found plenty of XML parsers (Boost-Property Tree Library, Xerces) and a few HTML parsers that spat out information in a SAX like manner. I just simply couldn't find a C++ library that did what I wanted.

So, I did some reading of the W3C's Core DOM specification and a few other sources and decided to write my own HTML parser. The IDL interface in the Core DOM specification gave me the starting point for my c++ implementation, and that is where I started. Boost-HTML Library is simple, no frills, since it was a Labor Day weekend project. I can read in a document and write it back, almost the same as I read it. I say almost the same, because I decided to use the XHTML notation for elements that are singular in nature, for instance <br />.

With this library, you can also pull out a list of elements by searching for a tag name or a tag id attribute. That was really the main feature I was looking for, pull out all the <a> tags from a document.

Hopefully, you'll find this library useful and tell me about it.

Better yet, if you find a glaring bug, I hope you'll tell me about that too!

What Boost Libraries are Used ?

I've used a few of my favorite Boost libraries:

shared_ptr	Pretty much every element object is encapsulated as a shared_ptr. I like automagic clean up
multi_index	Attributes and element references are placed in multi_index_containers. Makes life easy to look up objects.
hash	Easier to search by a hash, and that is just what I do!
string_algo	Pretty cool library giving Perl like functionality for strings. I personally use the join function a few times.