fixed MappingCharFilterFactory for en index

fruhstuck find tags frühstück also in english
but autocomplet still don't work well
This commit is contained in:
Bachir Soussi Chiadmi
2017-05-30 12:47:51 +02:00
parent 02d59eb06d
commit d10af95e24
2 changed files with 245 additions and 0 deletions
@@ -12,3 +12,225 @@
# "\u00c4" => "\u0041"
# "æ" => "ae"
# "\n" => " "
# À => A
"\u00C0" => "A"
# Á => A
"\u00C1" => "A"
# Â => A
"\u00C2" => "A"
# Ã => A
"\u00C3" => "A"
# Ä => A
"\u00C4" => "A"
# Å => A
"\u00C5" => "A"
# Æ => AE
"\u00C6" => "AE"
# Ç => C
"\u00C7" => "C"
# È => E
"\u00C8" => "E"
# É => E
"\u00C9" => "E"
# Ê => E
"\u00CA" => "E"
# Ë => E
"\u00CB" => "E"
# Ì => I
"\u00CC" => "I"
# Í => I
"\u00CD" => "I"
# Î => I
"\u00CE" => "I"
# Ï => I
"\u00CF" => "I"
# IJ => IJ
"\u0132" => "IJ"
# Ð => D
"\u00D0" => "D"
# Ñ => N
"\u00D1" => "N"
# Ò => O
"\u00D2" => "O"
# Ó => O
"\u00D3" => "O"
# Ô => O
"\u00D4" => "O"
# Õ => O
"\u00D5" => "O"
# Ö => O
"\u00D6" => "O"
# Ø => O
"\u00D8" => "O"
# Œ => OE
"\u0152" => "OE"
# Þ
"\u00DE" => "TH"
# Ù => U
"\u00D9" => "U"
# Ú => U
"\u00DA" => "U"
# Û => U
"\u00DB" => "U"
# Ü => U
"\u00DC" => "U"
# Ý => Y
"\u00DD" => "Y"
# Ÿ => Y
"\u0178" => "Y"
# à => a
"\u00E0" => "a"
# á => a
"\u00E1" => "a"
# â => a
"\u00E2" => "a"
# ã => a
"\u00E3" => "a"
# ä => a
"\u00E4" => "a"
# å => a
"\u00E5" => "a"
# æ => ae
"\u00E6" => "ae"
# ç => c
"\u00E7" => "c"
# è => e
"\u00E8" => "e"
# é => e
"\u00E9" => "e"
# ê => e
"\u00EA" => "e"
# ë => e
"\u00EB" => "e"
# ì => i
"\u00EC" => "i"
# í => i
"\u00ED" => "i"
# î => i
"\u00EE" => "i"
# ï => i
"\u00EF" => "i"
# ij => ij
"\u0133" => "ij"
# ð => d
"\u00F0" => "d"
# ñ => n
"\u00F1" => "n"
# ò => o
"\u00F2" => "o"
# ó => o
"\u00F3" => "o"
# ô => o
"\u00F4" => "o"
# õ => o
"\u00F5" => "o"
# ö => o
"\u00F6" => "o"
# ø => o
"\u00F8" => "o"
# œ => oe
"\u0153" => "oe"
# ß => ss
"\u00DF" => "ss"
# þ => th
"\u00FE" => "th"
# ù => u
"\u00F9" => "u"
# ú => u
"\u00FA" => "u"
# û => u
"\u00FB" => "u"
# ü => u
"\u00FC" => "u"
# ý => y
"\u00FD" => "y"
# ÿ => y
"\u00FF" => "y"
# ff => ff
"\uFB00" => "ff"
# fi => fi
"\uFB01" => "fi"
# fl => fl
"\uFB02" => "fl"
# ffi => ffi
"\uFB03" => "ffi"
# ffl => ffl
"\uFB04" => "ffl"
# ſt => ft
"\uFB05" => "ft"
# st => st
"\uFB06" => "st"
+23
View File
@@ -310,6 +310,27 @@
<filter class="solr.LowerCaseFilterFactory"/>
</analyzer>
</fieldType>
<!-- [MATERIO] Added custom Edge N gram Type with MappingCharFilterFactory to remove accent
see materio_search_api_search_api_data_type_info() in materio_search_api.module
purpose is to get the following feature :
"be" give béton and bettrave
BUT "bé" give only béton and NOT betterave
-->
<fieldType name="edge_n2_kw_mapped_text" class="solr.TextField" omitNorms="true" positionIncrementGap="100">
<analyzer type="index">
<charFilter class="solr.MappingCharFilterFactory" mapping="mapping-ISOLatin1Accent.txt"/>
<tokenizer class="solr.KeywordTokenizerFactory"/>
<filter class="solr.LowerCaseFilterFactory"/>
<filter class="solr.EdgeNGramFilterFactory" minGramSize="2" maxGramSize="25" />
</analyzer>
<analyzer type="query">
<tokenizer class="solr.KeywordTokenizerFactory"/>
<filter class="solr.LowerCaseFilterFactory"/>
</analyzer>
</fieldType>
<!-- Setup simple analysis for spell checking -->
<fieldType name="textSpell" class="solr.TextField" positionIncrementGap="100">
@@ -574,6 +595,8 @@
<!-- Special-purpose text fields -->
<dynamicField name="tes_*" type="edge_n2_kw_text" indexed="true" stored="true" multiValued="false" omitTermFreqAndPositions="true" />
<dynamicField name="tem_*" type="edge_n2_kw_text" indexed="true" stored="true" multiValued="true" omitTermFreqAndPositions="true" />
<!-- [materio] -->
<dynamicField name="temmap_*" type="edge_n2_kw_mapped_text" indexed="true" stored="false" multiValued="true" omitTermFreqAndPositions="true" />
<dynamicField name="tws_*" type="text_ws" indexed="true" stored="true" multiValued="false"/>
<dynamicField name="twm_*" type="text_ws" indexed="true" stored="true" multiValued="true"/>