Fixed replacing reserved words. #722
Conversation
| @@ -63,7 +63,7 @@ public static function escapeTerm($term) | |||
| */ | |||
| public static function replaceBooleanWords($term) | |||
| { | |||
| $replacementMap = array('AND'=>'&&', 'OR'=>'||', 'NOT'=>'!'); | |||
| $replacementMap = array(' AND '=>' && ', ' OR '=>' || ', ' NOT '=>' !'); | |||
There was a problem hiding this comment.
wouldn't be a regular expression better for this?
There was a problem hiding this comment.
Do you mean something like
$reservedWords = array('/\sAND\s/u' => ' && ', '/\sOR\s/u' => ' || ', '/\sNOT\s/u' => ' !');
$result = preg_replace(array_keys($reservedWords), array_values($reservedWords), $term);?
There was a problem hiding this comment.
I guess it might make no difference for ES, but I was thinking
$reservedWords = array('/\s+AND\s+/u' => ' && ', '/\s+OR\s+/u' => ' || ', '/\s+NOT\s+/u' => ' !');
There was a problem hiding this comment.
I didn't study the Regexp in detail. Would the main difference be performance or is there an other difference? After the ! no space is needed?
There was a problem hiding this comment.
@ruflin no space is needed
curl -XGET 'http://localhost:9200/_all/_search?q=hello+!kitty' is correct
I tested performance Regexp vs. strtr and differences were minimal. Regex is better for multibyte white spaces.
I prefer (w/o \s+)
$reservedWords = array('/\sAND\s/u' => ' && ', '/\sOR\s/u' => ' || ', '/\sNOT\s/u' => ' !');because your solution (with \s+) replace prefered words & "repair" query (blue AND red => blue && red). And it is the task for analyzer, I mean.
|
@jkbnerad Can you also update the changes.txt? |
|
@ruflin Yes. |
Fixed replacing reserved words.
|
Merged, thx. |
|
Thank you. |
I solved problem with uppercase queries.
Fixed:
"TIMBERLAND Men
s" was converted to "TIMBERL&& Mens"