Linefeed filtering issue where: (a) Atom feed and (b) title element type != HTML or XHTML

[Echelon9]

As reported on the forums: http://forums.cocoaforge.com/viewtopic.php?f=18&t=26298&sid=8f90ce829f0abb3d7314e3d71e2c8a3f

A corner case remains where linefeed type characters can make their way into a title element of an Atom feed.

Following circumstances must exist:

• Atom feed
• Title element has a type that is not "HTML" or "XHTML"
• Linefeed characters exist in the text of the title

Kwik was able to helpfully provide an example Atom feed which reproduces the issue. The feed is of comments to a Google Blogspot dummy feed: http://rssfeedtesting.blogspot.com/feeds/comments/default (available at time of this posting)

The issue is that in src/RichXMLParser.m the Atom branch of the parsing code applies summaryTextFromHTML only to titles explicitly marked with HTML or XHTML types.

Given this is data from the network, we should enforce regardless.

Resolving this issue will have the side effect of bringing the RSS and Atom parsing into line with each other for the title element.

// Parse item title
if ([itemNodeName isEqualToString:@"title"])
{
        NSString * newTitle = [[subItemTree valueOfElement] stringByUnescapingExtendedCharacters];
        NSString * titleType = [subItemTree valueOfAttribute:@"type"];

        if ([titleType isEqualToString:@"html"] || [titleType isEqualToString:@"xhtml"])
                newTitle = [newTitle summaryTextFromHTML];

        [newItem setTitle:newTitle];
        continue;
}