#! /usr/bin/python
"""
urlresolver:  makes relative urls absolute

TODO:
    - check for malformed urls
      possibly a seperate function.
"""


from exceptions import Exception
from lib import textfilters 
filters = textfilters

class URLResolveError(Exception):
    pass

def urlresolver(abs_url, rel_url):
    abs_url = abs_url.strip()
    rel_url = rel_url.strip()
 
    orig_abs_url = abs_url
    orig_rel_url = rel_url   

    #print "ORIG_ABS:", orig_abs_url
    #print "ORIG_REL:", orig_rel_url

    # first check to see if we need to do anything.
    if not rel_url: 
        return abs_url
    if 'http' in rel_url: 
        #print "IS ABS"
        return rel_url
    
    # check for http:// and remove before processing
    if "http://" in abs_url:
        abs_url = abs_url[7:]

    # there may be a file at the end of the abs_url 
    if abs_url[-1] != "/":
        abs_url = '/'.join(abs_url.split('/')[:-1])
        abs_url += "/"

    # remove './' '../' '/'
    def remove_levels(rel_url, up_levels=0):
        url = ''
        if rel_url:
            if up_levels > 0 and rel_url[0] == '/':
                url = rel_url[1:]
            elif rel_url[:2] == './':
                url = rel_url[2:]
            elif rel_url[:3] == '../':
                url = rel_url[3:]
                up_levels += 1
                return remove_levels(url, up_levels)
            else: 
                url = rel_url
        else:
            return rel_url, up_levels
    
        return url, up_levels

    url, up_levels = remove_levels(rel_url)
    
    for i in range(up_levels):
        parts = abs_url.split('/')
        parts = filter(lambda s: s != '', parts)[:-1]
        if len(parts) == 0:
            error_msg = "URL Cannot Be Resolved Base: %s, Relative: %s" % (orig_abs_url, orig_rel_url)
            #print error_msg
            raise URLResolveError(error_msg)
                                  
        abs_url = '/'.join(parts)

    if abs_url[-1] != '/':
        abs_url += "/"

    url = abs_url + url
    url = url.replace("//", "/")   
    url = "http://" + url
    
    #print "--> REL:", url
    #print "-" * 30
    #print
    return filters.remove_quotes(url)


if __name__ == '__main__':
    # normal cases
    u1 = 'http://www.python.org/doc/2.3.4/lib/module-exceptions.html'
    u2 = 'http://www.python.org/doc/2.3.4/lib/'
    r0 = 'http://www.python.org/doc/2.3.4/lib/'
    r1 = "../tut/tut.html"
    r2 = 'module-weakref.html'
    r3 = './module-weakref.html'
    r4 = "../../tut/tut.html"

    NORMAL_TEST = 1
    VERBOSE = 0
    BREAK_TEST = 1

    if NORMAL_TEST:
        assert(urlresolver(u1, r4) == 'http://www.python.org/doc/tut/tut.html')
        assert(urlresolver(u1, r0) == r0)
        assert(urlresolver(u1, r1) == 'http://www.python.org/doc/2.3.4/tut/tut.html')
        assert(urlresolver(u1, r2) == 'http://www.python.org/doc/2.3.4/lib/module-weakref.html')
        assert(urlresolver(u1, r3) == 'http://www.python.org/doc/2.3.4/lib/module-weakref.html')

        assert(urlresolver(u2, r0) == r0)
        assert(urlresolver(u2, r1) == 'http://www.python.org/doc/2.3.4/tut/tut.html')
        assert(urlresolver(u2, r2) == 'http://www.python.org/doc/2.3.4/lib/module-weakref.html')
        assert(urlresolver(u2, r3) == 'http://www.python.org/doc/2.3.4/lib/module-weakref.html')

    if VERBOSE:
        absolute = [u1, u2]
        relative = [r0, r1, r2, r3]
        for u in absolute:
            for r in relative:
                print urlresolver(u, r)

    if BREAK_TEST:
        u0 = 'http://www.python.org/doc/2.3.4/lib/'
        u1 = ""
        u2 = 'http://www.python.org/'
        r1 = ""
        r2 = "../tut/tut.html"
        assert(urlresolver(u0, r1) == u0)
        assert(urlresolver(u1, r1) == '')

        exception_raised = False
        try:
            urlresolver(u2, r2)
        except URLResolveError:
            exception_raised = True

        assert(exception_raised == True)