}
}
type boostFeatureWriter struct {
featureWriter
}
var (
goodQualClass = util.NewRegexFromWords(
"article",
"catchline",
"chapter",
"content",
"head",
"intro",
"introduction",
"leadin",
"main",
"post",
"story",
"summary",
"title",
)
poorQualClass = util.NewRegexFromWords(
"author",
"blog",
"byline",
"caption",
"col",
"comment",
"description",
}
}
}
}
var (
ignoreNames = util.NewRegexFromWords(
"breadcrumb",
"byline",
"caption",
"comment",
"community",
"credit",
"description",
"email",
"footer",
"gallery",
"hide",
"infotext",
"photo",
"related",
"shares",
"social",
"story[-_]?bar",
"story[-_]?feature",
)
ignoreStyle = util.NewRegex(`(?i)display:\s*none`)
)
// parseBody parses the <body>...</body> part of the HTML page. It creates
// Chunks for every html.TextNode found in the body.
func (doc *Document) parseBody(n *html.Node) {
