Sybase NNTP forums - End Of Life (EOL)

The NNTP forums from Sybase - forums.sybase.com - are now closed.

All new questions should be directed to the appropriate forum at the SAP Community Network (SCN).

Individual products have links to the respective forums on SCN, or you can go to SCN and search for your product in the search box (upper right corner) to find your specific developer center.

Fulltext search/SQL Anywhere 11 - full wild card support?

7 posts in General Discussion Last posting was on 2008-08-13 14:11:14.0Z
Keith Doty Posted on 2008-08-11 17:03:47.0Z
Sender: 5ef7.48a062ff.1804289383@sybase.com
From: Keith Doty
Newsgroups: ianywhere.public.general
Subject: Fulltext search/SQL Anywhere 11 - full wild card support?
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <48a070f3.6104.1681692777@sybase.com>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 11 Aug 2008 10:03:47 -0700
X-Trace: forums-1-dub 1218474227 10.22.241.41 (11 Aug 2008 10:03:47 -0700)
X-Original-Trace: 11 Aug 2008 10:03:47 -0700, 10.22.241.41
Lines: 2
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub ianywhere.public.general:7015
Article PK: 5239

Are there any plans for full wildcard full text search,
i.e., prefix, postfix and within, e.g., *e?onic* or *.doc?


Matthew Young-Lai (Sybase iAnywhere) Posted on 2008-08-12 13:22:22.0Z
From: "Matthew Young-Lai \(Sybase iAnywhere\)" <none>
Newsgroups: ianywhere.public.general
References: <48a070f3.6104.1681692777@sybase.com>
Subject: Re: Fulltext search/SQL Anywhere 11 - full wild card support?
Lines: 33
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2900.5512
X-RFC2646: Format=Flowed; Original
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.5512
NNTP-Posting-Host: vip152.sybase.com
X-Original-NNTP-Posting-Host: vip152.sybase.com
Message-ID: <48a18e8e$1@forums-1-dub>
Date: 12 Aug 2008 06:22:22 -0700
X-Trace: forums-1-dub 1218547342 10.22.241.152 (12 Aug 2008 06:22:22 -0700)
X-Original-Trace: 12 Aug 2008 06:22:22 -0700, vip152.sybase.com
X-Authenticated-User: techsupp
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub ianywhere.public.general:7016
Article PK: 5240

Prefix patterns are already supported. For example:

select * from t
where contains( c, 'doc*' )

More general patterns are only supported with regular expressions or LIKE
predicates that can't directly use a text index. One way to support more
complex pattern matching with a text index is to construct a text query from
another query on the text index vocabulary. For example:

select list(term, ' or ' )
from sa_text_index_vocab(
index_name, tab_name, user_name)
where term like '%e_onic%'

It's also possible use a text query to return candidate results and filter
them further with an additional predicate. For example:

select * from t
where contains( c, 'doc*' )
and c like '%.doc%'

There aren't specific plans for adding more powerful pattern matching
against text indexes. However, it's certainly possible. Can you give any
more details about what you're looking for?

Matthew

<Keith Doty> wrote in message news:48a070f3.6104.1681692777@sybase.com...
> Are there any plans for full wildcard full text search,
> i.e., prefix, postfix and within, e.g., *e?onic* or *.doc?


Keith Doty (Symantec) Posted on 2008-08-12 14:57:55.0Z
Sender: 5ef7.48a062ff.1804289383@sybase.com
From: Keith Doty (Symantec)
Newsgroups: ianywhere.public.general
Subject: Re: Fulltext search/SQL Anywhere 11 - full wild card support?
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <48a1a4f3.a3b.1681692777@sybase.com>
References: <48a18e8e$1@forums-1-dub>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 12 Aug 2008 07:57:55 -0700
X-Trace: forums-1-dub 1218553075 10.22.241.41 (12 Aug 2008 07:57:55 -0700)
X-Original-Trace: 12 Aug 2008 07:57:55 -0700, 10.22.241.41
Lines: 47
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub ianywhere.public.general:7019
Article PK: 5243

Matthew,
Thanks for your response. The use case we have is searching
file paths, so postfix wildcarding (find all Word Doc files
- *.doc) would be the first priority, but full wildcard
support (wildcard anywhere in the pattern) would be
desirable as well.
Thanks again for your quick response.
Keith

> Prefix patterns are already supported. For example:
>
> select * from t
> where contains( c, 'doc*' )
>
> More general patterns are only supported with regular
> expressions or LIKE predicates that can't directly use a
> text index. One way to support more complex pattern
> matching with a text index is to construct a text query
> from another query on the text index vocabulary. For
> example:
>
> select list(term, ' or ' )
> from sa_text_index_vocab(
> index_name, tab_name, user_name)
> where term like '%e_onic%'
>
> It's also possible use a text query to return candidate
> results and filter them further with an additional
> predicate. For example:
>
> select * from t
> where contains( c, 'doc*' )
> and c like '%.doc%'
>
> There aren't specific plans for adding more powerful
> pattern matching against text indexes. However, it's
> certainly possible. Can you give any more details about
> what you're looking for?
>
> Matthew
>
> <Keith Doty> wrote in message
> > news:48a070f3.6104.1681692777@sybase.com... Are there
> > any plans for full wildcard full text search, i.e.,
> prefix, postfix and within, e.g., *e?onic* or *.doc?
>
>


Matthew Young-Lai (Sybase iAnywhere) Posted on 2008-08-12 17:50:41.0Z
From: "Matthew Young-Lai \(Sybase iAnywhere\)" <none>
Newsgroups: ianywhere.public.general
References: <48a18e8e$1@forums-1-dub> <48a1a4f3.a3b.1681692777@sybase.com>
Subject: Re: Fulltext search/SQL Anywhere 11 - full wild card support?
Lines: 82
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2900.5512
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.5512
X-RFC2646: Format=Flowed; Original
NNTP-Posting-Host: vip152.sybase.com
X-Original-NNTP-Posting-Host: vip152.sybase.com
Message-ID: <48a1cd71$1@forums-1-dub>
Date: 12 Aug 2008 10:50:41 -0700
X-Trace: forums-1-dub 1218563441 10.22.241.152 (12 Aug 2008 10:50:41 -0700)
X-Original-Trace: 12 Aug 2008 10:50:41 -0700, vip152.sybase.com
X-Authenticated-User: techsupp
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub ianywhere.public.general:7022
Article PK: 5245

You could create a separate column that stores just the filename extension,
create an index on it (a normal index rather than a text index), and use
that to support either a col='doc' search or a col like 'doc%' search.

There's another variation of the trick of using a text index to find
candidate matches and then filtering with a LIKE, REGEXP, or SIMILAR
predicate. You can create an NGRAM text index on the column, break up the
wildcard pattern, and then search for all the n-grams that you find. For
example, with a 3-gram index, the following query

select list( term, ' AND ' ) from
sa_char_terms( '*e?onic*.doc' )

will ignore the special characters and return 'onic AND doc'. Then you can
search the text index and filter out the rows that don't match the real
pattern:

select * from t
where contains( c, 'onic AND doc' )
and c like '%e_onic%.doc'

Note that with the a 3-gram index, the 'onic' term is internally converted
to a phrase so that the query is effectively ' "oni nic" AND doc '.

One limitation of this is that you have to choose an N (which was 3 in the
above example). A pattern then needs to include a substring of at least N
alphanumeric characters or the text index can't be used to filter anything.

Matthew

<Keith Doty (Symantec)> wrote in message
news:48a1a4f3.a3b.1681692777@sybase.com...
> Matthew,
> Thanks for your response. The use case we have is searching
> file paths, so postfix wildcarding (find all Word Doc files
> - *.doc) would be the first priority, but full wildcard
> support (wildcard anywhere in the pattern) would be
> desirable as well.
> Thanks again for your quick response.
> Keith
>
>> Prefix patterns are already supported. For example:
>>
>> select * from t
>> where contains( c, 'doc*' )
>>
>> More general patterns are only supported with regular
>> expressions or LIKE predicates that can't directly use a
>> text index. One way to support more complex pattern
>> matching with a text index is to construct a text query
>> from another query on the text index vocabulary. For
>> example:
>>
>> select list(term, ' or ' )
>> from sa_text_index_vocab(
>> index_name, tab_name, user_name)
>> where term like '%e_onic%'
>>
>> It's also possible use a text query to return candidate
>> results and filter them further with an additional
>> predicate. For example:
>>
>> select * from t
>> where contains( c, 'doc*' )
>> and c like '%.doc%'
>>
>> There aren't specific plans for adding more powerful
>> pattern matching against text indexes. However, it's
>> certainly possible. Can you give any more details about
>> what you're looking for?
>>
>> Matthew
>>
>> <Keith Doty> wrote in message
>> > news:48a070f3.6104.1681692777@sybase.com... Are there
>> > any plans for full wildcard full text search, i.e.,
>> prefix, postfix and within, e.g., *e?onic* or *.doc?
>>
>>


Keith Doty Posted on 2008-08-12 18:22:28.0Z
Sender: 5ef7.48a062ff.1804289383@sybase.com
From: Keith Doty
Newsgroups: ianywhere.public.general
Subject: Re: Fulltext search/SQL Anywhere 11 - full wild card support?
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <48a1d4e4.18f4.1681692777@sybase.com>
References: <48a1cd71$1@forums-1-dub>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 12 Aug 2008 11:22:28 -0700
X-Trace: forums-1-dub 1218565348 10.22.241.41 (12 Aug 2008 11:22:28 -0700)
X-Original-Trace: 12 Aug 2008 11:22:28 -0700, 10.22.241.41
Lines: 97
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub ianywhere.public.general:7023
Article PK: 5246

Matthew,
Thanks again for the additional suggestions. One more
question ...
Do you have any data (percent increase for e.g.) on the
additional overhead the NGRAM term breaker approach might
take on both the indexing and search side of things?
Thanks,
Keith

> You could create a separate column that stores just the
> filename extension, create an index on it (a normal index
> rather than a text index), and use that to support either
> a col='doc' search or a col like 'doc%' search.
>
> There's another variation of the trick of using a text
> index to find candidate matches and then filtering with a
> LIKE, REGEXP, or SIMILAR predicate. You can create an
> NGRAM text index on the column, break up the wildcard
> pattern, and then search for all the n-grams that you
> find. For example, with a 3-gram index, the following
> query
>
> select list( term, ' AND ' ) from
> sa_char_terms( '*e?onic*.doc' )
>
> will ignore the special characters and return 'onic AND
> doc'. Then you can search the text index and filter out
> the rows that don't match the real pattern:
>
> select * from t
> where contains( c, 'onic AND doc' )
> and c like '%e_onic%.doc'
>
> Note that with the a 3-gram index, the 'onic' term is
> internally converted to a phrase so that the query is
> effectively ' "oni nic" AND doc '.
>
> One limitation of this is that you have to choose an N
> (which was 3 in the above example). A pattern then needs
> to include a substring of at least N alphanumeric
> characters or the text index can't be used to filter
> anything.
>
> Matthew
>
>
> <Keith Doty (Symantec)> wrote in message
> news:48a1a4f3.a3b.1681692777@sybase.com...
> > Matthew,
> > Thanks for your response. The use case we have is
> > searching file paths, so postfix wildcarding (find all
> > Word Doc files - *.doc) would be the first priority, but
> > full wildcard support (wildcard anywhere in the pattern)
> > would be desirable as well.
> > Thanks again for your quick response.
> > Keith
> >
> >> Prefix patterns are already supported. For example:
> >>
> >> select * from t
> >> where contains( c, 'doc*' )
> >>
> >> More general patterns are only supported with regular
> >> expressions or LIKE predicates that can't directly use
> a >> text index. One way to support more complex pattern
> >> matching with a text index is to construct a text query
> >> from another query on the text index vocabulary. For
> >> example:
> >>
> >> select list(term, ' or ' )
> >> from sa_text_index_vocab(
> >> index_name, tab_name, user_name)
> >> where term like '%e_onic%'
> >>
> >> It's also possible use a text query to return candidate
> >> results and filter them further with an additional
> >> predicate. For example:
> >>
> >> select * from t
> >> where contains( c, 'doc*' )
> >> and c like '%.doc%'
> >>
> >> There aren't specific plans for adding more powerful
> >> pattern matching against text indexes. However, it's
> >> certainly possible. Can you give any more details
> about >> what you're looking for?
> >>
> >> Matthew
> >>
> >> <Keith Doty> wrote in message
> >> > news:48a070f3.6104.1681692777@sybase.com... Are there
> >> > any plans for full wildcard full text search, i.e.,
> >> prefix, postfix and within, e.g., *e?onic* or *.doc?
> >>
> >>
>
>


Matthew Young-Lai (Sybase iAnywhere) Posted on 2008-08-13 13:33:28.0Z
From: "Matthew Young-Lai \(Sybase iAnywhere\)" <none>
Newsgroups: ianywhere.public.general
References: <48a1cd71$1@forums-1-dub> <48a1d4e4.18f4.1681692777@sybase.com>
Subject: Re: Fulltext search/SQL Anywhere 11 - full wild card support?
Lines: 125
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2900.5512
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.5512
X-RFC2646: Format=Flowed; Original
NNTP-Posting-Host: vip152.sybase.com
X-Original-NNTP-Posting-Host: vip152.sybase.com
Message-ID: <48a2e2a8$1@forums-1-dub>
Date: 13 Aug 2008 06:33:28 -0700
X-Trace: forums-1-dub 1218634408 10.22.241.152 (13 Aug 2008 06:33:28 -0700)
X-Original-Trace: 13 Aug 2008 06:33:28 -0700, vip152.sybase.com
X-Authenticated-User: techsupp
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub ianywhere.public.general:7027
Article PK: 5251

NGRAM indexes are bigger than term-based text indexes because they index
more locations in the text. For the same reason, they cost more to update
when you insert, update, or delete a row. I don't have any numbers, and
they depend on the characteristics of the data anyway, but an approximate
way to think of it is that inserting or deleting a row with K ngrams in it
will cost as much as K regular index updates. File names are relatively
short, so this overhead may be reasonable. If it isn't, and allowing the
index to become temporarily out-of-date index is acceptable, then an AUTO or
MANUAL refresh index may be worth considering instead of the default
IMMEDIATE refresh type.

For search, the cost will depend on how selective the query is. If you use
2-grams and search for a pattern with a single 2-gram 'xx', and 'xx' appears
in every file name, then the query will end up scanning the entire table,
and you might as well just use a regular expression. If you just want a few
matches quickly, then you can consider using TOP N instead of fetching all
rows. Another strategy is to use TOP N with some cut-off point like 1000
and say "too many matches" if you get the full 1000, which means that there
are >= 1000 matches in the table.

Those are few things to think about. I can say more if any of the above is
unclear, or if you were looking for something different.

Matthew

<Keith Doty> wrote in message news:48a1d4e4.18f4.1681692777@sybase.com...
> Matthew,
> Thanks again for the additional suggestions. One more
> question ...
> Do you have any data (percent increase for e.g.) on the
> additional overhead the NGRAM term breaker approach might
> take on both the indexing and search side of things?
> Thanks,
> Keith
>
>> You could create a separate column that stores just the
>> filename extension, create an index on it (a normal index
>> rather than a text index), and use that to support either
>> a col='doc' search or a col like 'doc%' search.
>>
>> There's another variation of the trick of using a text
>> index to find candidate matches and then filtering with a
>> LIKE, REGEXP, or SIMILAR predicate. You can create an
>> NGRAM text index on the column, break up the wildcard
>> pattern, and then search for all the n-grams that you
>> find. For example, with a 3-gram index, the following
>> query
>>
>> select list( term, ' AND ' ) from
>> sa_char_terms( '*e?onic*.doc' )
>>
>> will ignore the special characters and return 'onic AND
>> doc'. Then you can search the text index and filter out
>> the rows that don't match the real pattern:
>>
>> select * from t
>> where contains( c, 'onic AND doc' )
>> and c like '%e_onic%.doc'
>>
>> Note that with the a 3-gram index, the 'onic' term is
>> internally converted to a phrase so that the query is
>> effectively ' "oni nic" AND doc '.
>>
>> One limitation of this is that you have to choose an N
>> (which was 3 in the above example). A pattern then needs
>> to include a substring of at least N alphanumeric
>> characters or the text index can't be used to filter
>> anything.
>>
>> Matthew
>>
>>
>> <Keith Doty (Symantec)> wrote in message
>> news:48a1a4f3.a3b.1681692777@sybase.com...
>> > Matthew,
>> > Thanks for your response. The use case we have is
>> > searching file paths, so postfix wildcarding (find all
>> > Word Doc files - *.doc) would be the first priority, but
>> > full wildcard support (wildcard anywhere in the pattern)
>> > would be desirable as well.
>> > Thanks again for your quick response.
>> > Keith
>> >
>> >> Prefix patterns are already supported. For example:
>> >>
>> >> select * from t
>> >> where contains( c, 'doc*' )
>> >>
>> >> More general patterns are only supported with regular
>> >> expressions or LIKE predicates that can't directly use
>> a >> text index. One way to support more complex pattern
>> >> matching with a text index is to construct a text query
>> >> from another query on the text index vocabulary. For
>> >> example:
>> >>
>> >> select list(term, ' or ' )
>> >> from sa_text_index_vocab(
>> >> index_name, tab_name, user_name)
>> >> where term like '%e_onic%'
>> >>
>> >> It's also possible use a text query to return candidate
>> >> results and filter them further with an additional
>> >> predicate. For example:
>> >>
>> >> select * from t
>> >> where contains( c, 'doc*' )
>> >> and c like '%.doc%'
>> >>
>> >> There aren't specific plans for adding more powerful
>> >> pattern matching against text indexes. However, it's
>> >> certainly possible. Can you give any more details
>> about >> what you're looking for?
>> >>
>> >> Matthew
>> >>
>> >> <Keith Doty> wrote in message
>> >> > news:48a070f3.6104.1681692777@sybase.com... Are there
>> >> > any plans for full wildcard full text search, i.e.,
>> >> prefix, postfix and within, e.g., *e?onic* or *.doc?
>> >>
>> >>
>>
>>


Keith Doty Posted on 2008-08-13 14:11:14.0Z
Sender: 4c7f.48a2eafa.1804289383@sybase.com
From: Keith Doty
Newsgroups: ianywhere.public.general
Subject: Re: Fulltext search/SQL Anywhere 11 - full wild card support?
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <48a2eb82.4c97.1681692777@sybase.com>
References: <48a2e2a8$1@forums-1-dub>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 13 Aug 2008 07:11:14 -0700
X-Trace: forums-1-dub 1218636674 10.22.241.41 (13 Aug 2008 07:11:14 -0700)
X-Original-Trace: 13 Aug 2008 07:11:14 -0700, 10.22.241.41
Lines: 137
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub ianywhere.public.general:7028
Article PK: 5252

Matthew,
Thanks for the info/discussion - certainly some things to
consider.
Regards,
Keith

> NGRAM indexes are bigger than term-based text indexes
> because they index more locations in the text. For the
> same reason, they cost more to update when you insert,
> update, or delete a row. I don't have any numbers, and
> they depend on the characteristics of the data anyway, but
> an approximate way to think of it is that inserting or
> deleting a row with K ngrams in it will cost as much as K
> regular index updates. File names are relatively short,
> so this overhead may be reasonable. If it isn't, and
> allowing the index to become temporarily out-of-date
> index is acceptable, then an AUTO or MANUAL refresh index
> may be worth considering instead of the default IMMEDIATE
> refresh type.
>
> For search, the cost will depend on how selective the
> query is. If you use 2-grams and search for a pattern
> with a single 2-gram 'xx', and 'xx' appears in every file
> name, then the query will end up scanning the entire table
> , and you might as well just use a regular expression.
> If you just want a few matches quickly, then you can
> consider using TOP N instead of fetching all rows.
> Another strategy is to use TOP N with some cut-off point
> like 1000 and say "too many matches" if you get the full
> 1000, which means that there are >= 1000 matches in the
> table.
>
> Those are few things to think about. I can say more if
> any of the above is unclear, or if you were looking for
> something different.
>
> Matthew
>
> <Keith Doty> wrote in message
> > news:48a1d4e4.18f4.1681692777@sybase.com... Matthew,
> > Thanks again for the additional suggestions. One more
> > question ...
> > Do you have any data (percent increase for e.g.) on the
> > additional overhead the NGRAM term breaker approach
> > might take on both the indexing and search side of
> > things? Thanks,
> > Keith
> >
> >> You could create a separate column that stores just the
> >> filename extension, create an index on it (a normal
> index >> rather than a text index), and use that to
> support either >> a col='doc' search or a col like 'doc%'
> search. >>
> >> There's another variation of the trick of using a text
> >> index to find candidate matches and then filtering
> with a >> LIKE, REGEXP, or SIMILAR predicate. You can
> create an >> NGRAM text index on the column, break up the
> wildcard >> pattern, and then search for all the n-grams
> that you >> find. For example, with a 3-gram index, the
> following >> query
> >>
> >> select list( term, ' AND ' ) from
> >> sa_char_terms( '*e?onic*.doc' )
> >>
> >> will ignore the special characters and return 'onic AND
> >> doc'. Then you can search the text index and filter
> out >> the rows that don't match the real pattern:
> >>
> >> select * from t
> >> where contains( c, 'onic AND doc' )
> >> and c like '%e_onic%.doc'
> >>
> >> Note that with the a 3-gram index, the 'onic' term is
> >> internally converted to a phrase so that the query is
> >> effectively ' "oni nic" AND doc '.
> >>
> >> One limitation of this is that you have to choose an N
> >> (which was 3 in the above example). A pattern then
> needs >> to include a substring of at least N
> alphanumeric >> characters or the text index can't be used
> to filter >> anything.
> >>
> >> Matthew
> >>
> >>
> >> <Keith Doty (Symantec)> wrote in message
> >> news:48a1a4f3.a3b.1681692777@sybase.com...
> >> > Matthew,
> >> > Thanks for your response. The use case we have is
> >> > searching file paths, so postfix wildcarding (find
> all >> > Word Doc files - *.doc) would be the first
> priority, but >> > full wildcard support (wildcard
> anywhere in the pattern) >> > would be desirable as well.
> >> > Thanks again for your quick response.
> >> > Keith
> >> >
> >> >> Prefix patterns are already supported. For example:
> >> >>
> >> >> select * from t
> >> >> where contains( c, 'doc*' )
> >> >>
> >> >> More general patterns are only supported with
> regular >> >> expressions or LIKE predicates that can't
> directly use >> a >> text index. One way to support more
> complex pattern >> >> matching with a text index is to
> construct a text query >> >> from another query on the
> text index vocabulary. For >> >> example:
> >> >>
> >> >> select list(term, ' or ' )
> >> >> from sa_text_index_vocab(
> >> >> index_name, tab_name, user_name)
> >> >> where term like '%e_onic%'
> >> >>
> >> >> It's also possible use a text query to return
> candidate >> >> results and filter them further with an
> additional >> >> predicate. For example:
> >> >>
> >> >> select * from t
> >> >> where contains( c, 'doc*' )
> >> >> and c like '%.doc%'
> >> >>
> >> >> There aren't specific plans for adding more powerful
> >> >> pattern matching against text indexes. However,
> it's >> >> certainly possible. Can you give any more
> details >> about >> what you're looking for?
> >> >>
> >> >> Matthew
> >> >>
> >> >> <Keith Doty> wrote in message
> >> >> > news:48a070f3.6104.1681692777@sybase.com... Are
> there >> >> > any plans for full wildcard full text search
> , i.e., >> >> prefix, postfix and within, e.g., *e?onic*
> or *.doc? >> >>
> >> >>
> >>
> >>
>
>