class Ferret::Analysis::PerFieldAnalyzer

Summary¶ ↑

The PerFieldAnalyzer is for use when you want to analyze different fields with different analyzers. With the PerFieldAnalyzer you can specify how you want each field analyzed.

Example¶ ↑

# Create a new PerFieldAnalyzer which uses StandardAnalyzer by default
pfa = PerFieldAnalyzer.new(StandardAnalyzer.new())

# Use the WhiteSpaceAnalyzer with no lowercasing on the :title field
pfa[:title] = WhiteSpaceAnalyzer.new(false)

# Use a custom analyzer on the :created_at field
pfa[:created_at] = DateAnalyzer.new

Public Class Methods

new(default_analyzer) → analyzer click to toggle source

Create a new PerFieldAnalyzer specifying the default analyzer to use on all fields that are set specifically.

default_analyzer: analyzer to be used on fields that aren't otherwise specified

static VALUE
frb_per_field_analyzer_init(VALUE self, VALUE ranalyzer)
{
    Analyzer *def = frb_get_cwrapped_analyzer(ranalyzer);
    Analyzer *a = per_field_analyzer_new(def);
    Frt_Wrap_Struct(self, &frb_pfa_mark, &frb_analyzer_free, a);
    object_add(a, self);
    return self;
}

Public Instance Methods

per_field_analyzer[field_name] = default_analyzer → self click to toggle source

Set the analyzer to be used on field field_name. Note that field_name should be a symbol.

field_name: field we wish to set the analyzer for
analyzer: analyzer to be used on field_name

static VALUE
frb_per_field_analyzer_add_field(VALUE self, VALUE rfield, VALUE ranalyzer)
{
    Analyzer *pfa, *a;
    Data_Get_Struct(self, Analyzer, pfa);
    a = frb_get_cwrapped_analyzer(ranalyzer);

    pfa_add_field(pfa, frb_field(rfield), a);
    return self;
}

add_field(field_name, default_analyzer) → self click to toggle source

Set the analyzer to be used on field field_name. Note that field_name should be a symbol.

field_name: field we wish to set the analyzer for
analyzer: analyzer to be used on field_name

static VALUE
frb_per_field_analyzer_add_field(VALUE self, VALUE rfield, VALUE ranalyzer)
{
    Analyzer *pfa, *a;
    Data_Get_Struct(self, Analyzer, pfa);
    a = frb_get_cwrapped_analyzer(ranalyzer);

    pfa_add_field(pfa, frb_field(rfield), a);
    return self;
}

token_stream(field_name, input) → token_stream click to toggle source

Create a new TokenStream to tokenize input. The TokenStream created will also depend on the field_name in the case of the PerFieldAnalyzer.

field_name: name of the field to be tokenized
input: data from the field to be tokenized

static VALUE
frb_pfa_analyzer_token_stream(VALUE self, VALUE rfield, VALUE rstring)
{
    Analyzer *pfa, *a;
    Symbol field = frb_field(rfield);
    GET_A(pfa, self);

    StringValue(rstring);
    a = (Analyzer *)h_get(PFA(pfa)->dict, field);
    if (a == NULL) {
        a = PFA(pfa)->default_a;
    }
    if (a->get_ts == cwa_get_ts) {
        return rb_funcall(CWA(a)->ranalyzer, id_token_stream, 2,
                          FSYM2SYM(field), rb_str_new2(rs2s(rstring)));
    }
    else {
        return get_rb_ts_from_a(a, rfield, rstring);
    }
}