Thanks Marcus! I got it.
BR, Mark. On Fri, Mar 19, 2010 at 5:50 AM, Markus Jelsma <mar...@buyways.nl> wrote: > > On Thursday 18 March 2010 17:47:45 Mark Fletcher wrote: > > Hi, > > > > Thanks for the mail. I had tried the WIKI. > > > > My doubts remaining were mainly:- > > > > 1. > > If we have synonyms specified and they replace your search keyword with > the > > ones specified wouldn't we face a risk of our original keyword missed > out. > > What i meant is if I have a keyword for search say "agriculture" and I > > replace it with some synonyms, will I never again be able to search > > directly for "agriculture". ie suppose I have a document which has the > > term agriculture and none of the synonyms in it. Will that document be > > retrieved when i search for agriculture as I have now mapped it to other > > terms. > > It depends whether you let them be replaced. If you omit the => sign, the > terms simlpy will be expanded to whatever synonyms you specified. I could > not > explain it any better than the wiki. > > > 2. > > I am still a bit confused about the interpretation of:- > > a\=>a => b\=>b > > > > a\,a => b\,b > > > > abc def rose\, my cap , rose flower > > > > Can you pls give a one linere explanation for the above. There are > some > > sample entries in the synonyms.txt > > This is escaping otherwise meaningful characters. The , and => are > meaninful > to the SynonymFilterFactory and therefore need to be escaped as you also > would > escape certain characters in any language or whatever. You need to escape > qoutes in many languages and you must escape the : sign a.o. in you Lucene > queries. > > > > 3. If I get some help me with the above 3 it will help me understand the > backslash "\" also better. > > Thanks, > > Mark. > > > > > > On Thu, Mar 18, 2010 at 12:19 PM, Markus Jelsma <mar...@buyways.nl> > wrote: > > Hi, > > > > > > Check out the wiki page on the SynonymFilterFactory. It gives a decent > > explantion on the subject. The backslash is just for escaping otherwise > > meaningful characters. > > > > > > [1]: > http://wiki.apache.org/solr/AnalyzersTokenizersTokenFilters#solr.Synony > > mFilterFactory > > > > > > Cheers, > > > > On Thursday 18 March 2010 17:10:56 Mark Fletcher wrote: > > > Hi, > > > > > > Just needed some help to understand the following synonym mappings:- > > > > > > 1. aaa => aaaa > > > does it mean:- > > > if the user queries for aaa it is replaced with aaaa and > > > documents matching aaaa are searched for > > > or does it mean > > > if the user queries for aaa, documents with aaa as well as > aaaa > > > are looked for > > > > > > > > > 2. bbb => bbbb1 bbbb2 > > > does it mean that if the user queries for bbb, SOLR will look > for > > > documents that contain bbbb1 bbbb2 > > > > > > > > > 3. ccc => cccc1,cccc2 > > > does it mean that if the user queries for ccc, SOLR will look > for > > > documents that contain cccc1 or cccc2 > > > > > > 4. a\=>a => b\=>b > > > First of all my doubt is what does the "\" do there. Does it > > > have any special significance. > > > Can someone help me interpret the above > > > > > > 5. a\,a => b\,b > > > Can some one help me with this also > > > > > > 6. fooaaa,baraaa,bazaaa > > > does this mean that if any of fooaaa or baraaa or bazaaa > > > comes as the search keyword, SOLR will look for documents that contain > > > fooaaa > > > > > > 7. abc def rose\, my cap , rose flower > > > > > > > > > does this mean a query for any of the above 3 will always be > > > replaced by a query for abc def rose\ > > > > > > Can some one pls extend some help at your earliest convenience. > > > > > > Thank you. > > > Mark. > > > > Markus Jelsma - Technisch Architect - Buyways BV > > http://www.linkedin.com/in/markus17 > > 050-8536620 / 06-50258350 > > > > Markus Jelsma - Technisch Architect - Buyways BV > http://www.linkedin.com/in/markus17 > 050-8536620 / 06-50258350 > >
