You will have to use a search engine library to add search abilities to your application.
As i read yours and your team mate's queries i thought i could be helpful with my findings on using a cell base DB: First of all you will have to decide the logical entry point to your data usage. For example, Sachin mentioned about an Employee table in the other post with the rowkey as id. Let's say you want to authorize employees to a specific area of your app via a password. Your employee table will look like this: Rowkey ColumnFamily Value Doug Password <md5pass> Sachin Password <md5pass> Sanmaya Password <md5pass> Then to access those users datas you will need a unique id in an other column for each of them: Rowkey ColumnFamily Value Doug UUID 123456 Sachin UUID 654321 Sanmaya UUID 789456 As you got to a point where you have an id for a specific user you can then access to datas of that user, for example: (EmployeeMonthlyWorkHour Table) Rowkey ColumnFamily:Qualifier Value 123456 MWH:December2008 32 123456 MWH:March2009 45 123456 MWH:April2009 12 789456 MWH:March2009 55 Most probably you will need to search in column values for some scenarios such as (EmployeeAddressInfo Table) Rowkey ColumnFamily Value 123456 Address xyz street etc etc 789456 Address asd street etc etc To find all employees who's address is xyz street you will need to use a search engine library such as Xapian (www.xapian.org). You will pass Address value to the indexer where document id will be 123456. So when you query "xyz street" it will return 123456. This what I understood and applied until now for my needs and i don't regret any SQL functionality for now. Ufuk (BTW, sorry for the bad grammar) On Jun 4, 6:27 pm, Vicaya <[email protected]> wrote: > On Wed, Jun 3, 2009 at 11:58 PM, Sachin <[email protected]> wrote: > > > Hi, > > > In our application, we are going to give serach facility. User can > > enter any word and the application will search this word in any of the > > columns. Here we will need like operator, as the word to be search can > > be anywhere, in start, in middle or at the end. > > > For this type of search, can you please advice me how to go ahead? > > > I know I am thinking right now in SQL Server fashion. But need some > > guideline, so that I can think on this differently. > > What you really need is full-text search for this application. Using > SQL/HQL is a kludge that would not work very well. Solr, Sphinx etc is > what you need to look into. Basically you write a data import query in > them and you send your word query to these search engines. > > __Luke > > > > > Regards, > > Sachin > > > On Jun 3, 2:03 pm, Vicaya <[email protected]> wrote: > >> >> 3) Is there any operators in hypertable as like 'IN' and 'LIKE' as in > >> >> SQL? > > >> > Not yet. LIKE should be fairly easy to implement. IN may be a little > >> > more > >> > challenging to do efficiently. > > >> IN on rowkey should be fairly easy to implement though, since the > >> client api already support list of intervals. > > >> __Luke --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Hypertable Development" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/hypertable-dev?hl=en -~----------~----~----~----~------~----~------~--~---
