# edgeNGram filter not keeping the whole words **URL:** \ **Category:** Elasticsearch\ **Created:** [July 26, 2013, 8:31am UTC](https://discuss.elastic.co/t/edgengram-filter-not-keeping-the-whole-words/12951 "2013-07-26T08:31:35Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![Patrick\_Norwood](https://avatars.discourse-cdn.com/v4/letter/p/ecd19e/32.png) [@Patrick\_Norwood](https://discuss.elastic.co/u/Patrick_Norwood)\ **Post date:** [July 26, 2013, 8:31am UTC](https://discuss.elastic.co/t/edgengram-filter-not-keeping-the-whole-words/12951/1 "2013-07-26T08:31:35Z")
Hi guys, I have a problem with my analyzer. Following is the setup in elasticsearch.yml: analyzer : default\_index : type : custom tokenizer : whitespace filter : [word\_delimiter, asciifolding, standard, lowercase, synonym, edgeNGram] filter : word\_delimiter : type : word\_delimiter preserve\_original : true edgeNGram : type : edgeNGram min\_gram : 2 max\_gram : 15 side : front The problem is that when I run curl -XGET 'localhost:9200/my\_index/\_analyze?field=email\_doc.body' -d 'WordWithOverFifteenChars', the result is 'WordWithOverFifteenChars' broken down into n-grams up to 15 chars, but the whole word does not get indexed. The same happens for words like 'something.company', that means the whole world (preserve\_orginal settings for word\_delimiter filter) doesn't get preserved, it only gets broken down into n-grams as a whole up to 'something.compa' (15 chars) as well as n-grams for 'company' and 'something' are created. Could you give me your insight on what I am missing? -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![jpountz](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/jpountz/32/45836_2.png) [@jpountz](https://discuss.elastic.co/u/jpountz)\ **Post date:** [July 26, 2013, 9:20am UTC](https://discuss.elastic.co/t/edgengram-filter-not-keeping-the-whole-words/12951/2 "2013-07-26T09:20:51Z")
Hi, Although preserve\_original is set to true on the word\_delimiter filter, the edgeNGram filter is applied afterwards, meaning that the analyzer will compute n-grams for both the original word and the sub-words created by the word\_delimiter filter. If you want to keep the original token, it might make sense to use a multi field[1] and analyze your field once with word\_delimiter and once with n-grams. Then you can use any of those fields depending on what you are trying to achieve : prefix search with the field analyzed with edgeNGram and standard search with the field analyzed with word\_delimiter. [1] [http://www.elasticsearch.org/guide/reference/mapping/multi-field-type/](http://www.elasticsearch.org/guide/reference/mapping/multi-field-type/) -- Adrien Grand -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 6, 2017, 2:24am UTC](https://discuss.elastic.co/t/edgengram-filter-not-keeping-the-whole-words/12951/3 "2017-07-06T02:24:23Z")