C implementation of the Readability algorithm (plus some goodies)

Dependencies:

• libxml2
• libpcre or ICU for regular expressions support

Building:

• make

Building with ICU rather than pcre:

• ICU=1 make

By default, both the readable program and the Python extension will be built.

Building for OS X using Xcode

• Create a new directory named readable
• Copy readable.h and readable.c in the newly created directory
• Copy the directory named unicode from the ICU headers into your project (you can get it from the iPhoneSimulator SDK, under /usr/include/unicode)
• Add the readable parent directory, the unicode parent directory and /usr/include/libxml2 to Header Search Path under Build Settings
• Add libicucore.dylib and libxml2.xylib to the Link Binary with libraries Build Phase
• In your code, import readable.h

Building for iOS using Xcode

• Create a new directory named readable
• Copy readable.h and readable.c in the newly created directory
• Add the readable parent directory and /usr/include/libxml2 to Header Search Path under Build Settings
• Add libicucore.dylib and libxml2.xylib to the Link Binary with libraries Build Phase
• In your code, import readable.h

API:

char * readable(const char *html, const char *url, const char *encoding, int options)

Parses HTML to extract the interesting contents.

• html: HTML code to parse
• url: URL where this HTML was fetched from
• encoding: HTML encoding
• options: See readable.h for the avaialble options

char * next_page_url(const char *html, const char *url, const char *encoding);

Returns the url for the next page in a multipage article (pretty much in alpha):

• html: HTML code to parse
• url: URL where this HTML was fetched from
• encoding: HTML encoding

License

This code is licensed under the AGPLv3. If you'd like to use the code under a different license, drop me a line to alberto@garciahierro.com