Most of the users are not supposed to use get/put calls generally. Direct
mutation is convenient and obvious way. There is not much point in that apis
at all.

On Thu, Aug 26, 2010 at 12:47 AM, Luke Lu <[email protected]> wrote:

> Well, the put_* api is analogous to the ubiquitous/standard file api
> where fputc/fputs] on FILE * is buffered by default until a flush.
> Changing the api name breaks existing applications and doesn't seem to
> gain anything (except for subjective clarity, for worse for me as it
> breaks the get/put symmetry.)
>
> Improving documentation on various options of Mutator/MutateSpec would
> be a more productive effort. IMO.
>
> On Wed, Aug 25, 2010 at 7:56 AM, Doug Judd <[email protected]> wrote:
> > Yes, depending on the type of Mutator defined in the MutateSpec, the data
> > won't appear until a subsequent flush happens.  Also, if there was some
> sort
> > of failure (e.g. ROW OVERFLOW), then the client won't get notified of the
> > failure.  This API is really for situations where data loss can be
> > tolerated.
> > - Doug
> >
> > On Wed, Aug 25, 2010 at 7:37 AM, Stanislav Yudin <[email protected]>
> wrote:
> >>
> >> > The shared mutator will buffer the cells and flush them at a later
> >> > time.
> >> So modified cells are not available right after submit?
> >> On Wed, Aug 25, 2010 at 7:09 AM, Doug Judd <[email protected]> wrote:
> >>>
> >>> I'm considering changing the names of the put_ methods of the Thrift
> >>> interface from
> >>> put_cell
> >>> put_cell_as_array
> >>> put_cells
> >>> put_cells_as_arrays
> >>> to ...
> >>> submit_cell
> >>> submit_cell_as_array
> >>> submit_cells
> >>> submit_cells_as_arrays
> >>> The reason is that these methods merely add cells to a shared mutator
> and
> >>> return immediately.  The shared mutator will buffer the cells and flush
> them
> >>> at a later time.  The term "put" doesn't adequately convey this fact,
> >>> whereas "submit" implies that the cells will be submitted to the system
> >>> without waiting for the result.
> >>> Any objections?
> >>> - Doug
> >>>
> >>> --
> >>> 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]<hypertable-dev%[email protected]>
> .
> >>> For more options, visit this group at
> >>> http://groups.google.com/group/hypertable-dev?hl=en.
> >>
> >> --
> >> 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]<hypertable-dev%[email protected]>
> .
> >> For more options, visit this group at
> >> http://groups.google.com/group/hypertable-dev?hl=en.
> >
> > --
> > 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]<hypertable-dev%[email protected]>
> .
> > For more options, visit this group at
> > http://groups.google.com/group/hypertable-dev?hl=en.
> >
>
> --
> 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]<hypertable-dev%[email protected]>
> .
> For more options, visit this group at
> http://groups.google.com/group/hypertable-dev?hl=en.
>
>

-- 
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.

Reply via email to