Please, can you share some parts of your models? Specially an example of a computed field declaration and what are you using in compute=?, also it is important to have an example of what types the the other fields.
It is important to reproduce the problem for fixing (if a bug) also, can you open a ticket in google code? code.google.com/p/web2py Sometimes little things can break a little, but if it is detected as a bug or break of compatibility, for sure it will be fixed in few hours. Thanks On Wed, Nov 16, 2011 at 11:39 PM, David Manns <dgma...@gmail.com> wrote: > One of the recent releases has broken computed fields. > > My application uses several instances of computed fields. They all > worked perfectly in 1.94.6 to which I have now reverted. > > In 1.99.2 (and possibly intervening releases) the behavior of computed > fields is unpredictable. In my case, they are all computed correctly > when a new record is created (crud.create). However, when a record is > updated (crud.update) some of them are recomputed but others are not, > thus becoming inconsistent with the fields used in the computation. > > This is in addition to application failures caused in 1.99.2 due to > the acknowledged replace_id problem (see > > http://groups.google.com/group/web2py/browse_thread/thread/ddc384b33ec1a7cc/3935ecc0a6785e9e?lnk=gst&q=dgmanns#3935ecc0a6785e9e > ) > which caused me to spend lots of time finding a workaround. > > This is all very alarming in a framework which boasts of "always > maintaining backward compatibility" - quote taken from the preface of > "the book". > > Perhaps there is too much of a rush to add new features: not breaking > existing applications would seem to be of paramount importance. > > A pity because in general, it is an outstanding platform. -- Bruno Rocha [http://rochacbruno.com.br]
