A single cell table seems extremely lightweight to me compared to adding a column to my object tables. Alternatively you could use a file or timestamp on a file. Right now I don't want to add the column because the initial update takes a very long time and will lock up my database of 20M rows. That's a huge amount of overhead for something that could be easily done with this other method. I may create my own method based on this once I get tired of the cron+window technique I'm using now :) Seems pretty straightforward to implement.
Thanks. On Jun 9, 10:45 am, Pat Allan <[email protected]> wrote: > I guess it's viable - adding a table is extra overhead, just like a > delta column is overhead, so it's much of a muchness. I've not had > others request this though, so I'm not sure if I'd merge it in. > Perhaps it's worth having a look at Dan Pickett's Workling Delta code, > and build your own approach in the same manner, making it easy for > others to use. > > http://github.com/dpickett/workling_delta_indexer/tree/master > > Cheers > > -- > Pat > > On 09/06/2009, at 10:27 AM, Alex wrote: > > > > > I don't want to add a delta column to my table. Why not use the > > update_at column and keep track of the last scan for updated records, > > instead of doing the "interval method" that risks missing records and > > reindexing records? > > > I.e. create a table with one cell with the last index date and index > > records updated since then. That seems a lot more reliable and > > efficient than the other non-delta column methods. > > > Any reason not to add this a delta method? > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Thinking Sphinx" 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/thinking-sphinx?hl=en -~----------~----~----~----~------~----~------~--~---
