public static AnalysisDescriptor SetAnalysis(AnalysisDescriptor analysis)
{
return(analysis
.CharFilters(c => c
.Mapping("mapping", f => f.Mappings(GetCharMapping()))
.PatternReplace("digits", f => f.Pattern("[^0-9]").Replacement(""))
)
.TokenFilters(f => f
.NGram("digits_ngram", t => t.MinGram(3).MaxGram(8))
.Length("length_limit", t => t.Min(1).Max(20))
.EdgeNGram("custom_edge_ngram", t => t.MinGram(1).MaxGram(50))
)
.Tokenizers(t => t
.NGram("ngram", d => d.MinGram(2).MaxGram(30)))
.Analyzers(a => a
.UserDefined(Replace, new CustomAnalyzer {
Tokenizer = "whitespace", Filter = new[] { "lowercase", "scandinavian_folding", "unique" }, CharFilter = new[] { "mapping" }
})
.UserDefined(ReplaceNgram, new CustomAnalyzer {
Tokenizer = "whitespace", Filter = new[] { "lowercase", "scandinavian_folding", "custom_edge_ngram", "unique" }, CharFilter = new[] { "mapping", "html_strip" }
})
.UserDefined(Key, new CustomAnalyzer {
Tokenizer = "keyword", Filter = new[] { "lowercase", "scandinavian_folding" }
})
.UserDefined(Digits, new CustomAnalyzer {
Tokenizer = "keyword", CharFilter = new[] { "digits" }, Filter = new[] { "digits_ngram", "unique", "length_limit" }
})
.UserDefined(Lowercase, new CustomAnalyzer {
Tokenizer = "keyword", Filter = new[] { "lowercase" }
})
));
}
private static AnalysisDescriptor CreateAnalysisDescriptor(string charHtmlFilter, TokenFiltersDescriptor tokenFilters, AnalyzersDescriptor analyzers)
{
var analysisDescriptor = new AnalysisDescriptor();
analysisDescriptor.CharFilters(c => c.HtmlStrip(charHtmlFilter));
analysisDescriptor.TokenFilters(t => tokenFilters);
analysisDescriptor.Analyzers(a => analyzers);
return(analysisDescriptor);
}
protected override AnalysisDescriptor Contribute(AnalysisDescriptor descriptor, IEnumerable <KeyValuePair <string, CharFilterBase> > build)
{
return(descriptor.CharFilters(a =>
{
foreach (var item in build.Where(x => CanContribute(x, a)))
{
a.Add(item.Key, item.Value);
}
return a;
}));
}
private static AnalysisDescriptor Analysis(AnalysisDescriptor analysis) => analysis
.CharFilters(c => c.Mapping("swedish_char_mapping", m => m.Mappings("w => v", "W => V")))
.TokenFilters(tf => tf.Hunspell("sv_SE", x => x.Dedup().Locale("sv_SE")))
.Analyzers(analyzers => analyzers
.Custom(LowercaseKeywordAnalyserName, c => c
.Tokenizer("keyword")
.Filters("lowercase")
)
.Custom(SwedishTextAnalyserName, c => c
.Tokenizer("standard")
.Filters("lowercase", "sv_SE")
.CharFilters("html_strip", "swedish_char_mapping")
)
);
protected override IAnalysis FluentAnalysis(AnalysisDescriptor an) => an.CharFilters(d => AssertionSetup.Fluent(AssertionSetup.Name, d));
public static AnalysisDescriptor CreateRemoveNumbersCharFilters(this AnalysisDescriptor analysisDescriptor)
{
return(analysisDescriptor.CharFilters(cFilter => cFilter
.Mapping(REMOVE_NUMBERS_CHAR_FILTER_NAME, mpf => mpf
.Mappings("1=>", "2=>", "3=>", "4=>", "5=>", "6=>", "7=>", "8=>", "9=>"))));
}
