Attribute value ignored when capturing another attribute value in the same tag

[ackalker]

First, thanks for this wonderful tool!

I have the following problem: when trying the following snippet:

import scrapemark

html = """
    <div>
    <a href="http://site.com/page1" title="Page 1">Page 1</a>
    <a href="http://site.com/page2" title="Page 2">Page 2</a>
    <a href="http://site.com/page3" title="Page 3">Page 3</a>
    <a href="http://site.com/page2" title="Next">&gt;</a>
    </div>
    """

res = scrapemark.scrape("""<a href="{{ nextpage }}" title="Next" />""", html)
print res

results in: {'nextpage': u'http://site.com/page1'} which is simply the first link, not the link to the next page as I would expect. Capturing to a list with:

res = scrapemark.scrape("""{* <a href="{{ [nextpage] }}" title="Next" /> *}""", html)

returns: {'nextpage': [u'http://site.com/page1', u'http://site.com/page2', u'http://site.com/page3', u'http://site.com/page2']} i.e. a list of all links, not a list with just the link to the next page.

It appears as if scrapemark is ignoring the title attribute's value when it can capture the href attribute's value. Am I doing something wrong here, is this simply a quirk we'll have to be aware of, or is this a bug?

[ackalker]

Pull request #15 from quink seems to resolve this issue.