Differences between revisions 5 and 6
Revision 5 as of 2009-02-04 08:45:08
Size: 4614
Editor: adsl-75-55-126-95
Comment:
Revision 6 as of 2009-02-05 22:18:26
Size: 5312
Editor: adsl-074-229-189-244
Comment:
Deletions are marked like this. Additions are marked like this.
Line 9: Line 9:
The Solr sort parameter uses the Lucene sorting tool. This creates an array containing an entry for every document in the index. Sorting is then done against this array. This array is cached across requests and so repeated sorts are fast. If the field type is 'integer' the array contains only that value and thus is 4 bytes * the number of documents. If the field type is anything else, this integer array is created and then a separate array is also created with that field's data per entry. Sorting is also slower if the type is not an 'integer'. The Solr sort parameter uses the Lucene sorting tool. This creates an array containing an entry for every document in the index using the !FieldCache. Sorting is then done against this array. This array is cached across requests using the !IndexReader and so repeated sorts are fast. If the field type is 'integer' the array contains only that value and thus is 4 bytes * the number of documents. If the field type is anything else, this integer array is created and then a separate array is also created with that field's data per entry. Sorting is also slower if the type is not an 'integer'.
Line 18: Line 18:

=== A Note on "sortable" FieldTypes ===
Sortable !FieldTypes like sint, sdouble are a bit of a misnomer. They are not needed for Sorting in the sense described above, but are needed when doing !RangeQuery queries. Sortables, in fact, refer to the notion of making the number sort correctly lexicographically as Strings. That is, if this is not done, the numbers 1..10 sort lexicographically as 1,10, 2, 3... Using an sint, however remedies this. If, however, you don't need to do !RangeQuery queries and only need to sort on the field, then just use an int or double or the equivalent appropriate class. You will save yourself time and memory.
Line 31: Line 35:
To create a phoneme-based field, you need a text filter stack that does not include stemming or stopwords, and add the solr.PhoneticFilterFactory (see [AnalyzersTokenizersTokenFilters]) with one of the available encoders. This must be in both the indexing and query stack. Of the several available the "Double Metaphone" filter is the most popular and does well with non-English text. There are as yet no language-specific phoneme encoders. To create a phoneme-based field, you need a text filter stack that does not include stemming or stopwords, and add the solr.!PhoneticFilterFactory (see AnalyzersTokenizersTokenFilters) with one of the available encoders. This must be in both the indexing and query stack. Of the several available the "Double Metaphone" filter is the most popular and does well with non-English text. There are as yet no language-specific phoneme encoders.
Line 33: Line 37:
For another take on assisting spelling, see [SpellCheckComponent]. For another take on assisting spelling, see SpellCheckComponent.

General tips & tricks in designing schemas.

Mapping databases to Solr

Solr provides one table. Storing a set database tables in an index generally requires denormalizing some of the tables. Attempts to avoid denormalizing usually fail.

Field contents

The more heterogeneous (different kinds of data) you have in one field or in one index, the less useful it is. For example, if you have text in different languages, it is more useful to store them in different fields: text_en, text_fr, etc. than all in one field. When you search against that one field English and French words and phrases will be searched with equal interest.

Sorting

There are two ways of sorting available in Solr 1.4: Lucene's sorting feature and function queries.

Lucene Sorting

The Solr sort parameter uses the Lucene sorting tool. This creates an array containing an entry for every document in the index using the FieldCache. Sorting is then done against this array. This array is cached across requests using the IndexReader and so repeated sorts are fast. If the field type is 'integer' the array contains only that value and thus is 4 bytes * the number of documents. If the field type is anything else, this integer array is created and then a separate array is also created with that field's data per entry. Sorting is also slower if the type is not an 'integer'. However, range checks do not work on an 'integer' field. If you want range checks and fast sorting, you can create a pair of fields, one of each type, with a copyField directive:

 <field name="popularity" type="sint" indexed="true" stored="true" multiValued="false"/>
 <field name="popularitySort" type="integer" indexed="true" stored="false" />
 ...
 <copyField source="popularity" dest="popularitySort"/>

Note that since multiValued=false is the default for these types, attempting to store a value to 'popularitySort' will cause an indexing error, since it also always receives a value from 'popularity'. Also there is no reason to store both fields, and so 'popularitySort' is index-only.

A Note on "sortable" FieldTypes

Sortable FieldTypes like sint, sdouble are a bit of a misnomer. They are not needed for Sorting in the sense described above, but are needed when doing RangeQuery queries. Sortables, in fact, refer to the notion of making the number sort correctly lexicographically as Strings. That is, if this is not done, the numbers 1..10 sort lexicographically as 1,10, 2, 3... Using an sint, however remedies this. If, however, you don't need to do RangeQuery queries and only need to sort on the field, then just use an int or double or the equivalent appropriate class. You will save yourself time and memory.

Function Query Sorting

Add this clause to your query string to sort the results using 'myIndexedField'. Do not use the 'sort=field+asc' parameter. See [FunctionQuery] for more.

_val_:"ord(myIndexedField)"

There may be performance differences with this technique v.s. the Lucene sorting algorithm.

Multiple Text Search Field types

The "text" field type in the example schema.xml provides basic text search for English text. But, it has a surprise: the actual text given to this field is not indexed as-is, and therefore searching for the raw text may not work. If you store "To Be Or Not To Be" in a "text" field, none of these words will found this document, nor will the phrase in quotes. The above words are all stopwords and are stripped from the input text. Another transform is stemming, which stores both 'change' and 'changing' as the word 'chang'.

If you want to have any phrase search work as well as individual words, you need to have two fields. Both should be processed similarly, but the phrase search field should not use stemming or stopwords.

Phonemes

Programmers are perfect spellers and expect the same of their users. A phoneme represents (roughly) the sound of one syllable. Phoneme-based searching can give users a better search experience. To support misspelled search words phoneme filters cause the index to store phoneme-base representations of the text instead of the input. This only finds misspellings which sound like the original word.

To create a phoneme-based field, you need a text filter stack that does not include stemming or stopwords, and add the solr.PhoneticFilterFactory (see AnalyzersTokenizersTokenFilters) with one of the available encoders. This must be in both the indexing and query stack. Of the several available the "Double Metaphone" filter is the most popular and does well with non-English text. There are as yet no language-specific phoneme encoders.

For another take on assisting spelling, see SpellCheckComponent.

Unicode processing

Searching text in different languages is very difficult. The Latin1Accent filters downgrade all European "special characters" down to their US Ascii equivalents: the French spelling protégé becomes the English spelling protege. In Solr-1.3, use this in the filter stack of your "text" field type:

<filter class="solr.ISOLatin1AccentFilterFactory" />

In Solr-1.4, use this:

<charFilter class="solr.MappingCharFilterFactory" mapping="mapping-ISOLatin1Accent.txt"/>

SchemaDesign (last edited 2012-05-21 19:13:42 by adsl-75-51-164-120)