but i only need a way to delete an entity by one command or action. not by a program(and it doesn't always works), it is certainly not a matter of how it is implemented, but a feature of the platform.
also 1000 index limit makes it not possible to fetcher older data on paging. for if we need an indexed page more than 10000 items, it would cost us a lot of cpu time to calculate the base for GQL to fetch the data with index less than 1000. how can the platform be scalable, for currently the platform aches with only 10000 items. On Dec 9, 1:25 pm, Tim Hoffman <[email protected]> wrote: > imho I think you do need to understand any new platform to a certain > degree if you really want to take advantage of it. > It is completely different from SQL/RDBMS which means if you don't > change your thinking and adapt to the platform > it "can" only be a toy for you. If you can't get your head around the > fact that its not a rdbms then possibly it is the wrong > tool for whatever you are trying to do. (That doesn't mean app engine > couldn't use a whole range of improvements/features) > > For the rest of us its is a very compelling delivery platform for a > range of applications. > > Rgds > > T > > On Dec 9, 10:55 am, ajaxer <[email protected]> wrote: > > > > > thanks for the explanation. > > but I have no interest to learn such things as big table or something. > > the only reason that i keep an eye on this project is it may bring me > > convenience in my web development > > not that it will bring me some knowledge of science or technology. > > > before the data manipulation tool becomes good enough, I will only > > keep it as a toy. > > > On Dec 7, 11:00 am, OvermindDL1 <[email protected]> wrote: > > > > On Sun, Dec 6, 2009 at 7:36 PM, ajaxer <[email protected]> wrote: > > > > yes, i have tried. > > > > but i alway get timeouts. > > > > and the amount is more than millions maybe > > > > I uploading this data by bulk uploading for more than a day. > > > > > I think it is very important to be able delete a table in a single act > > > > on the panel > > > > You are thinking about this wrong. > > > GQL does not have a relational database design, it is an indexed blob > > > database, there is no concept of "table" as there is in SQL and such. > > > Instead you create 'kinds', which are more analogous to a message in > > > Google Protocol Buffers. The things you specify as 'indexes' are > > > actually just those things pulled from the 'message'. Each index you > > > specify creates a whole new 'table' with some duplicated data between > > > all the 'tables'. So for any given kind you will have multiple > > > 'tables' (this is not a SQL like table, this is more like a map/dict, > > > with ordering and such...). Each index type can also have multiple > > > kinds in it (such as when you change the format of a kind, you are > > > creating a new one, but all still indexed together, meaning if you try > > > to access an old one from your new one, you get a nice exception > > > thrown, hence why you should always change your kind name when you > > > change your kind too). > > > > So, a kind can have multiple indexes(tables), and each index(table) > > > can have multiple kinds. > > > > GQL is not like SQL, you need to relearn things, starting with getting > > > the concept of tables out of your mind. :) > > > > P.S. The above description is not perfectly accurate as to how it > > > works, but close enough to get the idea across. :) -- You received this message because you are subscribed to the Google Groups "Google App Engine" 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/google-appengine?hl=en.
