modeling-users — For users: questions, support, announcements, etc.

Re: [Modeling-users] Insert inverse relationship reference problem

[Sebastien B. <sbi...@us...>]

        Hi,

  Sorry for the late answer.

Wara Songkran <war...@gm...> wrote:
> Hi
>=20
> I've try to test insertion Ojbect with refference to the others
[...]
> ec.setAutoInsertion(True)
> s =3D Staff()
> t =3D Task()
> s.addToTasks(t)
> ec.insert(s)
> ec.saveChanges()
>=20
> the staff foreign key in task row is Null. but if I execute
> t.setStaff(s). It's work normally.
>=20
>=20
> do I have to call t.setStaff(s) explicitly

In short: yes, you do.  As a general rule, you are responsible for keeping
your object graph consistent.

Now a bit longer: ec's autoInsertion feature only takes care of calling
ec.insert() on the objects that are:

  - not inserted in the ec yet,

  - and which are in relation to objects that are already under the
    control of an EC

However, the ec does not play with your relations , it always assumes
that the object graph is as the developper wants it to be.  If the
association is bidirectional, you have to set both relations.

Now you can also delegate the work to
addObjectToBothSidesOfRelationshipWithKey(), see
http://modeling.sf.net/UserGuide/customobject-relationshipmanipulation.html

In your case, you'll write:=20

  s.addObjectToBothSidesOfRelationshipWithKey(t, 'tasks')

or

  t.addObjectToBothSidesOfRelationshipWithKey(s,, 'staff')


-- S=E9bastien.




Wara Songkran <war...@gm...> wrote:
> Hi
>=20
> I've try to test insertion Ojbect with refference to the others
>=20
> My Entities are=20
>=20
> Entity('Staff',
>          properties=3D[ APrimaryKey('id', isClassProperty=3D0, isRequired=
=3D1),
>                       AString('firstName', isRequired=3D1,
> columnName=3D'firstName'),
>                       AString('lastName', isRequired=3D1, columnName=3D'l=
astName'),
>                       ],
>          ),
> Entity('Task',
>           properties=3D[ APrimaryKey('id', isClassProperty=3D0, isRequire=
d=3D1),
>                        ADateTime('startTime', columnName=3D'startTime',
> externalType=3D'datetime'),
>                        ADateTime('completeTime',
> columnName=3D'completeTime', externalType=3D'datetime'),
>                        AString('isCompleted',doc=3D'yes or no')
>                        ],
>          ),
>=20
> and my association is
>=20
>   Association('Task', 'Staff',=20
>               relations=3D['staff', 'tasks']),
>=20
> and when i execute
>=20
> ec.setAutoInsertion(True)
> s =3D Staff()
> t =3D Task()
> s.addToTasks(t)
> ec.insert(s)
> ec.saveChanges()
>=20
> the staff foreign key in task row is Null. but if I execute
> t.setStaff(s). It's work normally.
>=20
>=20
> do I have to call t.setStaff(s) explicitly
>=20
> Regards
> Wara Songkran

Re: [Modeling-users] slooow delete

[Sebastien B. <sbi...@us...>]

        Hi John,


John Lenton <jl...@gm...> wrote:
> I have a table that has ~28k rows, and doing a simple
>=20
>   ec.delete(obj)
>   ec.saveChanges()
>=20
> takes about 60 seconds. This is for a web app, so the delay is way
> beyond what a user would find reasonable. Is there any way to speed
> this up?
> --=20
> John Lenton (jl...@gm...) -- Random fortune:
> Don't anthropomorphise computers and cars, They hate that.

That's waaaaayyy to slow, even for a regular app.  Before investigating
any further, could you integrate the following patch and send the result
of ec.currentStateSummary() just before the saveChanges()?  I'd like to
see whether this is actually the only change or not, and how much
objects are handled by the ec.

Even better, if you can, change ec.delete/ec.saveChange into:

  ec.delete(obj)
  print ec.currentStateSummary() # print it or log it if this is a web app
  ec.processRecentChanges()
  print ec.currentStateSummary()
  ec.saveChanges()
  print ec.currentStateSummary()


-- S=E9bastien.

Re: [Modeling-users] slooow delete

[John L. <jl...@gm...>]

On 05 Mar 2005 16:31:38 +0100, Sebastien Bigaret
<sbi...@us...> wrote:
> 
> That's waaaaayyy to slow, even for a regular app.  Before investigating
> any further, could you integrate the following patch and send the result
> of ec.currentStateSummary() just before the saveChanges()?

I'm guessing you forgot the patch?

-- 
John Lenton (jl...@gm...) -- Random fortune:
Don't anthropomorphise computers and cars, They hate that.

[Modeling-users] slooow delete

[John L. <jl...@gm...>]

I have a table that has ~28k rows, and doing a simple

  ec.delete(obj)
  ec.saveChanges()

takes about 60 seconds. This is for a web app, so the delay is way
beyond what a user would find reasonable. Is there any way to speed
this up?

-- 
John Lenton (jl...@gm...) -- Random fortune:
Don't anthropomorphise computers and cars, They hate that.

[Modeling-users] Insert inverse relationship reference problem

[Wara S. <war...@gm...>]

Hi

I've try to test insertion Ojbect with refference to the others

My Entities are 

Entity('Staff',
         properties=[ APrimaryKey('id', isClassProperty=0, isRequired=1),
                      AString('firstName', isRequired=1,
columnName='firstName'),
                      AString('lastName', isRequired=1, columnName='lastName'),
                      ],
         ),
Entity('Task',
          properties=[ APrimaryKey('id', isClassProperty=0, isRequired=1),
                       ADateTime('startTime', columnName='startTime',
externalType='datetime'),
                       ADateTime('completeTime',
columnName='completeTime', externalType='datetime'),
                       AString('isCompleted',doc='yes or no')
                       ],
         ),

and my association is

  Association('Task', 'Staff', 
              relations=['staff', 'tasks']),

and when i execute

ec.setAutoInsertion(True)
s = Staff()
t = Task()
s.addToTasks(t)
ec.insert(s)
ec.saveChanges()

the staff foreign key in task row is Null. but if I execute
t.setStaff(s). It's work normally.


do I have to call t.setStaff(s) explicitly

Regards
Wara Songkran

Re: [Modeling-users] one-to-one association, howto

[Wara S. <war...@gm...>]

On 17 Feb 2005 12:34:43 +0100, Sebastien Bigaret
<sbi...@us...> wrote:
>=20
>        Hi,
>=20
> Wara Songkran <war...@gm...> wrote:
> > I've try to create one-to-one association from table Task to Supplier b=
y define
> >
> > model.association =3D [
> >   Association('Task', 'Supplier',multiplicity=3D[ [0, 1], [0, 1] ])
> > ]
> >
> > but the error said
> >
> > ValueError: invalid multiplicity dst->src: should be toMany
> >
> > I've also apply the patch from
> > http://sourceforge.net/mailarchive/message.php?msg_id=3D9401573
> >
> > Did I done something wrong?
>=20
> The problem is that one-to-one relationships are still not supported
> out of the box, at least not the way one think they should be :/
>=20
> For the moment one-to-one rels. should be modeled as one-to-many
> (http://modeling.sourceforge.net/UserGuide/design-rels.html)
>=20
> Back to your example, your model becomes::
>=20
>  model.association =3D [
>    ## this is in fact a one-to-one modeled as a one-to-many,
>    ## class 'Supplier' defines getTask() and setTask()
>    Association('Task', 'Supplier',multiplicity=3D[ [0, 1], [0, None] ],
>                relations=3D['supplier', 'tasks'] )
>  ]
>=20
> and then you add to the class Supplier the following code::
>=20
>   def validateTasks(self, aValue):
>     """
>     ``tasks`` is a to-one relationship modeled as a to-many (inverse of
>     the to-one rel ``Task.supplier``): it is valid iff it has no more
>     than one element
>=20
>     :exception Modeling.Validation.ValidationException: if more than one
>       task is registered
>     :see: `getTask()`, `setTask()`
>     """
>     # see http://modeling.sf.net/UserGuide/customobject-validation-by-key=
.html
>     # subsection 3.3.2.2.
>     from Modeling import Validation
>     if len(self.getTasks())>1:
>       raise Validation.ValidationException
>     return
>=20
>  def getTask(self):
>    if self.getTasks():
>        return self.getTasks()[0]
>    return None
>=20
>  def setTask(self, task):
>    if self.getTask():
>      self.removeFromTasks(self.getTask())
>    self.addToTasks(task)
>=20
>  # you can also define the corresponding property:
>  task =3D property(getTask, setTask, None, "to-one relationship to Task")
>=20
> Additional note: with this definition you'll have a foreign key in entity
>  'Task' pointing to the primary key in entity 'Supplier'.  It seems
>  reasonable, but in case you want the inverse, swap tasks and supplier in=
 the
>  association's definition and define get/setSupplier() in class Task.
>=20
>        Regards,
>=20
> -- S=E9bastien.
>=20
>=20

Thank for your reseponse. Now I know more limitation of Modeling. I'm
developing a relational <--> xml mapping (relational <--> object <-->
xml) using your framework and EaseXML http://xmlobject.base-art.net/
(nothing serius just a part of my school project)
Do you have any suggestion.

Regard
Wara Songkran

Re: [Modeling-users] one-to-one association, howto

[Sebastien B. <sbi...@us...>]

        Hi,

Wara Songkran <war...@gm...> wrote:
> I've try to create one-to-one association from table Task to Supplier by =
define
>=20
> model.association =3D [
>   Association('Task', 'Supplier',multiplicity=3D[ [0, 1], [0, 1] ])
> ]
>=20
> but the error said=20
>=20
> ValueError: invalid multiplicity dst->src: should be toMany
>=20
> I've also apply the patch from
> http://sourceforge.net/mailarchive/message.php?msg_id=3D9401573
>=20
> Did I done something wrong?


The problem is that one-to-one relationships are still not supported
out of the box, at least not the way one think they should be :/

For the moment one-to-one rels. should be modeled as one-to-many
(http://modeling.sourceforge.net/UserGuide/design-rels.html)

Back to your example, your model becomes::

  model.association =3D [
    ## this is in fact a one-to-one modeled as a one-to-many,
    ## class 'Supplier' defines getTask() and setTask()
    Association('Task', 'Supplier',multiplicity=3D[ [0, 1], [0, None] ],
                relations=3D['supplier', 'tasks'] )
  ]

and then you add to the class Supplier the following code::
=20=20
   def validateTasks(self, aValue):
     """
     ``tasks`` is a to-one relationship modeled as a to-many (inverse of
     the to-one rel ``Task.supplier``): it is valid iff it has no more
     than one element

     :exception Modeling.Validation.ValidationException: if more than one
       task is registered
     :see: `getTask()`, `setTask()`
     """
     # see http://modeling.sf.net/UserGuide/customobject-validation-by-key.=
html
     # subsection 3.3.2.2.
     from Modeling import Validation
     if len(self.getTasks())>1:
       raise Validation.ValidationException
     return
=20=20
  def getTask(self):
    if self.getTasks():
        return self.getTasks()[0]
    return None
=20=20
  def setTask(self, task):
    if self.getTask():
      self.removeFromTasks(self.getTask())
    self.addToTasks(task)

  # you can also define the corresponding property:
  task =3D property(getTask, setTask, None, "to-one relationship to Task")


Additional note: with this definition you'll have a foreign key in entity
  'Task' pointing to the primary key in entity 'Supplier'.  It seems
  reasonable, but in case you want the inverse, swap tasks and supplier in =
the
  association's definition and define get/setSupplier() in class Task.

        Regards,

-- S=E9bastien.

[Modeling-users] one-to-one association, howto

[Wara S. <war...@gm...>]

I've try to create one-to-one association from table Task to Supplier by define

model.association = [
  Association('Task', 'Supplier',multiplicity=[ [0, 1], [0, 1] ])
]

but the error said 

ValueError: invalid multiplicity dst->src: should be toMany

I've also apply the patch from
http://sourceforge.net/mailarchive/message.php?msg_id=9401573

Did I done something wrong?

Regrads
Wara Songkran

Re: [Modeling-users] Modeling performance for large number of objects?

[Wolfgang K. <wol...@gm...>]

> And I definitely agree w/ John, in that performance will highly depend on
> your particular use-case --maybe you could be more explicit on it?

The specific envisaged application case would be as a persistence framework
for a toolkit for extracting, transformating and forwarding of data from
various sources to various sinks - a re-implementation in 100% pure
(C)Python of the Retic toolkit currently which is currently implemented in
Jython.

But the question was also for general "enterprise" application cases like
for an ERP or CMMS system.

Best regards,

Wolfgang Keller

Re: [Modeling-users] Modeling performance for large number of objects?

[Wolfgang K. <wol...@gm...>]

Hello,

and thanks for your reply.

>> In fact my question was raised when I read the article about ERP5 on
>> pythonology.org, with the performance values they claimed for ZODB with
>> their ZSQLCatalog add-on. I would guess that their performance claims are
>> only valid if all the queried objects are in fact in memory...?
> 
> I didn't read that article (didn't search it either I admit, do you have
> the hyperlink at hand?), 

http://www.pythonology.org/success&story=nexedi

They claim that "Reading the Zope object database is 10 to 100 times faster
than retrieving a row from the fastest relational databases available on
the market today".

And about ZSQLCatalog in particular: "a Zope database with more than
2,000,000 objects can be queried with statistical methods in a few
milliseconds"

Best regards,

Wolfgang Keller

[Modeling-users] Some news before the next release

[Sebastien B. <sbi...@us...>]

sbi...@us... wrote:
[about optimization advices]
> Well, that's always hard to tell without specific use-cases, but the
> general advices are: [...]

  BTW, since I'm at it and that's quite a long time I've not
  given any news, I also suggest using the latest CVS version (see
  notes, below).

  I know it's a real shame I haven't made any release since february, I
  thought I could at least make one for xmas but obviously I ran out of
  time, it will be done next january.

Also note that section 3. Functionalities for Object Management in the
User's Guide has been rewritten and now includes full documentation for:

  - building modules and classes at runtime (3.1. From model to python
    code)

  - the framework's expectations in external code (section 3.2. The
    framework's requirements on python code)


Last, I just want to announce that the patch item #892454: Fetch slices
and sort orderings will be integrated into the next release --its
current limitations will remain the same, but at least the features it
provides will be available out of the box.
(https://sf.net/tracker/index.php?func=detail&aid=892454&group_id=58935&atid=489337)

    Cheers,

-- Sébastien.


Instructions to get the latest CVS are at 
  https://sourceforge.net/cvs/?group_id=58935

and the latest changes and the description of all bug fixes are at:
  http://cvs.sf.net/viewcvs.py/modeling/ProjectModeling/CHANGES?view=markup

Re: [Modeling-users] Modeling performance for large number of objects?

[Sebastien B. <sbi...@us...>]

  Hi Wolfgang, John and all,

Thanks John for giving the figures for the overhead induced by the
framework when creating/manipulating objects.  I'm currently away from
my computer and I only have a web access, so that's hard to try and
compare anything in those conditions ;)

Wolfgang Keller <wol...@gm...> wrote:
> The question is for me whether Modeling tries to and/or whether there
> would be some other way to cut the hourglass-display-time to the
> unavoidable minimum (dependend on the database) by some kind of smart
> caching of objects or by maintaining some kind of pool of pre-created
> object instances.

For the moment being the framework does not use any kind of pool of
pre-created objects.  On the other hand, it caches database snapshots so
that the data that have already been fetched is not fetched again (this
avoids a time-expensive round-trip to the database).

John Lenton <jl...@gm...> wrote:
> Modeling [...] only saves those objects that have effectively changed,
> so depending on your actual use cases you might be surprised at how
> well it works. The loading, modifying and saving of all the objects is
> pretty much the worse case; Modeling isn't meant (AFAICT) for that
> kind of batch processing. It certainly is convenient, though :)

Right: when saving changes the framework uses that cache to save only
the objects that were actually modified/deleted.  And I definitely agree
w/ John, in that performance will highly depend on your particular
use-case --maybe you could be more explicit on it?  When you say that
you need <<to process (create - transform - store) rather important
amounts of data >>, do you mean that every single fetched object will be
updated and stored back in the database?  If this is the case, as John
already pointed it out, this is the worst case and the most
time-comsuming process you'll have w/ the framework.

John Lenton <jl...@gm...> wrote:
> Of course, in both cases, writing the sql script would've taken a
> *lot* longer than the difference in run time, for me. However, it's
> obvious that there are cases where the runtime difference overpowers
> the developer time difference...

  ...and when runtime matters, you can also use the framework on sample
  data, extract the generated SQL statements and then directly use those
  statements in the real batch.


Wolfgang Keller <wol...@gm...> wrote:
> In fact my question was raised when I read the article about ERP5 on
> pythonology.org, with the performance values they claimed for ZODB with
> their ZSQLCatalog add-on. I would guess that their performance claims are
> only valid if all the queried objects are in fact in memory...?

I didn't read that article (didn't search it either I admit, do you have
the hyperlink at hand?), but I suspect that the performance mostly comes
from the fact that the ZODB.Persistent mixin class is written in C:
while the overhead for object creation is probably still the same, the
process of fully initializing an object (assign values to attributes) is
pretty much quicker (as far as I remember, it directly sets the object's
__dict__, so yes that's fast ;)

  The framework spend most of its initializing time in KeyValueCoding,
(http://modeling.sourceforge.net/UserGuide/customobject-key-value-coding.html)
examining objects and finding the correct way of setting the attributes.
While this allows a certain flexibility, I now tend to believe that most
applications pay the price for a feature they do not need (for example,
the way attributes'values are assigned by the framework should probably
be cached per class rather then systematically determined for every
object).  [1]

John Lenton <jl...@gm...> wrote:
> Of course, maybe Sébastien has a trick up his sleeve as to how one
> could go about using Modeling for batch processing...

  Well, that's always hard to tell without specific use-cases, but
  the general advices are:

  - use the latest python,

  - use new-style classes rather than old-styles ones,

  - activate MDL_ENABLE_SIMPLE_METHOD_CACHE
    (http://modeling.sourceforge.net/UserGuide/env-vars-core.html)

  - specifically for batch processing: be sure to read:
    http://modeling.sourceforge.net/UserGuide/ec-discard-changes.html


And of course, we'll be happy to examine your particular use-cases to
help you optimize the process.

-- Sébastien.


[1] and thinking a little more about this, I now realize that the way
    this is done in the framework at initialization time is pretty
    stupid (the KVC mechanism should definitely be cached somehow at
    this point: since the framework creates the object before
    initializing it there is absolutely no reason for different objects
    of the same class to behave differently wrt KVC at this point)...
    I'll get back on this for sure.  For the curious this is done in
    DatabaseContext.initializeObject(), lines 1588-1594.  For the
    records I'll add that the prepareForInitializationWithKeys() stuff
    is also not needed.

Re: [Modeling-users] Modeling performance for large number of objects?

[Wolfgang K. <wol...@gm...>]

>>> Usually, the penalty does not interfere with the job at hand.
>> 
>> Oops, why?

*snip*
 
> Usually you might
> *display* all the objects (where rawRows comes in handy), and then the
> user selects one of these objects to actually modify (so you fault the
> raw object into a real one, work on it, and saveChanges).

*click* Of course.

> Of course, maybe S=E9bastien has a trick up his sleeve as to how one
> could go about using Modeling for batch processing...

Not just for batch-processing...

In fact my question was raised when I read the article about ERP5 on
pythonology.org, with the performance values they claimed for ZODB with
their ZSQLCatalog add-on. I would guess that their performance claims are
only valid if all the queried objects are in fact in memory...?

Best regards

Wolfgang Keller

Re: [Modeling-users] Modeling performance for large number of objects?

[John L. <jl...@gm...>]

On Mon, 20 Dec 2004 14:21:40 +0100, Wolfgang Keller
<wol...@gm...> wrote:
> Hello,
>=20
> and thanks for your reply.
>=20
> > Usually, the penalty does not interfere with the job at hand.
>=20
> Oops, why?
>=20
> Obviously, when all objects get read into memory at startup of the
> application server, and written back only at night, then...

because this (reading in all the objects, modifying them all, and
saving them all) is not the usual use case. Usually you might
*display* all the objects (where rawRows comes in handy), and then the
user selects one of these objects to actually modify (so you fault the
raw object into a real one, work on it, and saveChanges). You still
have a 20x penalty, but it's much less then a second in this use case.

> > One reference point you might find useful is that when I loading 3000
> > objects from a database, modifying them, and then saving the changes,
> > on a 700MHz p3 notebook, the loading took about 40 seconds, and the
> > saving, 200. That's 20 times what a direct sql script would've taken.
>=20
> This gives me an idea, thanks. A multiplier of 20 is quite significant
> imho.
>=20
> > Of course, in both cases, writing the sql script would've taken a
> > *lot* longer than the difference in run time, for me. However, it's
> > obvious that there are cases where the runtime difference overpowers
> > the developer time difference...
>=20
> I was wondering whether Modeling would be suitable as a persistence layer
> for a Python application that needs to process (create - transform - stor=
e)
> rather important amounts of data.
>=20
> The question is for me whether Modeling tries to and/or whether there wou=
ld
> be some other way to cut the hourglass-display-time to the unavoidable
> minimum (dependend on the database) by some kind of smart caching of
> objects or by maintaining some kind of pool of pre-created object
> instances.

Modeling does cache the objects, and only saves those objects that
have effectively changed, so depending on your actual use cases you
might be surprised at how well it works. The loading, modifying and
saving of all the objects is pretty much the worse case; Modeling
isn't meant (AFAICT) for that kind of batch processing. It certainly
is convenient, though :)

Of course, maybe S=E9bastien has a trick up his sleeve as to how one
could go about using Modeling for batch processing...

--=20
John Lenton (jl...@gm...) -- Random fortune:
bash: fortune: command not found

Re: [Modeling-users] Modeling performance for large number of objects?

[Wolfgang K. <wol...@gm...>]

Hello,

and thanks for your reply.

> Usually, the penalty does not interfere with the job at hand. 

Oops, why?

Obviously, when all objects get read into memory at startup of the
application server, and written back only at night, then...

> One reference point you might find useful is that when I loading 3000
> objects from a database, modifying them, and then saving the changes,
> on a 700MHz p3 notebook, the loading took about 40 seconds, and the
> saving, 200. That's 20 times what a direct sql script would've taken.

This gives me an idea, thanks. A multiplier of 20 is quite significant
imho.

> Of course, in both cases, writing the sql script would've taken a
> *lot* longer than the difference in run time, for me. However, it's
> obvious that there are cases where the runtime difference overpowers
> the developer time difference...

I was wondering whether Modeling would be suitable as a persistence layer
for a Python application that needs to process (create - transform - store)
rather important amounts of data.

The question is for me whether Modeling tries to and/or whether there would
be some other way to cut the hourglass-display-time to the unavoidable
minimum (dependend on the database) by some kind of smart caching of
objects or by maintaining some kind of pool of pre-created object
instances.

Best regards,

Wolfgang Keller

Re: [Modeling-users] Modeling performance for large number of objects?

[John L. <jl...@gm...>]

On Mon, 20 Dec 2004 11:43:07 +0100, Wolfgang Keller
<wol...@gm...> wrote:
> 
> given the significant penalty for the creation of Python objects indicated
> by most benchmarks I have seen so far, I wonder how and how well Modeling
> deals with this issue...?

Usually, the penalty does not interfere with the job at hand. However,
when you actually need to manipulate a large number of objects, the
Modeling overhead can be quite noticeable; when said manipulation is
only for querying, using rawRows avoids most of the creation overhead,
but when you have to modify the objects in question, you might find
yourself waiting quite a long time for saveChanges to complete.

One reference point you might find useful is that when I loading 3000
objects from a database, modifying them, and then saving the changes,
on a 700MHz p3 notebook, the loading took about 40 seconds, and the
saving, 200. That's 20 times what a direct sql script would've taken.

On the other hand, loading 100000 objects using rawRows takes about 20
seconds on this same machine. That's 4 times what the straight sql
would've taken.

Of course, in both cases, writing the sql script would've taken a
*lot* longer than the difference in run time, for me. However, it's
obvious that there are cases where the runtime difference overpowers
the developer time difference...

-- 
John Lenton (jl...@gm...) -- Random fortune:
bash: fortune: command not found

[Modeling-users] Modeling performance for large number of objects?

[Wolfgang K. <wol...@gm...>]

Hello,

given the significant penalty for the creation of Python objects indicated
by most benchmarks I have seen so far, I wonder how and how well Modeling
deals with this issue...?

TIA,

Best regards

Wolfgang Keller

Re: [Modeling-users] CustomObject.globalID() problem

[<no...@ti...>]

On 08 Oct 2004 11:02:41 +0200
Sebastien Bigaret <sbi...@us...> wrote:

> Are you, by any chance, concurrently accessing the same EC without
> locking it?  If you do, then maybe you should probably re-read the
> dedicated section in the User's Guide :)
> http://modeling.sourceforge.net/UserGuide/framework-integration-MT-considerations.html

I remember that bit right while writing the previous email :) , I think thats where
the problem was.

thanks
--
Delio

Re: [Modeling-users] model for tests

[John L. <jl...@gm...>]

On Fri, 8 Oct 2004 08:33:07 -0300, John Lenton <jl...@gm...> wrote:
> On 08 Oct 2004 10:58:04 +0200, Sebastien Bigaret
> 
> 
> <sbi...@us...> wrote:
> >
> > John Lenton <jl...@gm...> wrote:
> > > to test one of the patches I'd need a model to have a relation to an
> > > entity that has subclasses; although StoreEmployee has such a relation
> > > (Holidays <<-> Employee), it doesn't define the toOne side of that
> > > relation, and I notice that is used to test exactly that feature.
> > > Should I create a new model that has these properties, or do I add a,
> > > say, 'Salary' class?
> >
> > Yes, go for it and modify the model.  Maybe a BankAccount class could
> > make more sense?  Okay, anyway that's a test model, so the semantics is
> > not that important :)
> 
> last night I uploaded a patch that called it 'Salary', but I'll rename
> it to  'BankAccount' in a while.

I forgot to say this: I uploaded a patch because I already had it
ready, and although I have time to create these patches, I don't have
a lot of time, so sometimes I get impatient and don't wait as much as
I should for the answers, and go with what I think best at the time; I
can always revisit the decision later, but this way I can move ahead
(still got a ways to go).

-- 
John Lenton (jl...@gm...) -- Random fortune:
bash: fortune: command not found

Re: [Modeling-users] model for tests

[John L. <jl...@gm...>]

On 08 Oct 2004 10:58:04 +0200, Sebastien Bigaret
<sbi...@us...> wrote:
> 
> John Lenton <jl...@gm...> wrote:
> > to test one of the patches I'd need a model to have a relation to an
> > entity that has subclasses; although StoreEmployee has such a relation
> > (Holidays <<-> Employee), it doesn't define the toOne side of that
> > relation, and I notice that is used to test exactly that feature.
> > Should I create a new model that has these properties, or do I add a,
> > say, 'Salary' class?
> 
> Yes, go for it and modify the model.  Maybe a BankAccount class could
> make more sense?  Okay, anyway that's a test model, so the semantics is
> not that important :)

last night I uploaded a patch that called it 'Salary', but I'll rename
it to  'BankAccount' in a while.

-- 
John Lenton (jl...@gm...) -- Random fortune:
bash: fortune: command not found

Re: [Modeling-users] CustomObject.globalID() problem

[Sebastien B. <sbi...@us...>]

no...@ti... wrote:
> On 05 Oct 2004 23:42:47 +0200
> Sebastien Bigaret <sbi...@us...> wrote:
>=20
> > Okay; if you observe that a freshly fetched object has None as globalId=
(),
> > this is definitely a bug, so do not hesitate to report!
>=20
> I have a deadline on friday and don't have time to write a test, it's nas=
ty
> ( or simply a problem with my code ) since it appears only if another thr=
ead is
> fetching the same object from the same EC while I try to retrieve the Glo=
balID.
> Wouldn't worry until I got some real evidence :-)

Are you, by any chance, concurrently accessing the same EC without
locking it?  If you do, then maybe you should probably re-read the
dedicated section in the User's Guide :)
http://modeling.sourceforge.net/UserGuide/framework-integration-MT-consider=
ations.html

  Operations on an EC is not thread-safe by default, and concurrent
  access w/o locking it first could lead to the behaviour you observe
  (where an object can be observed w/ a gID=3D=3DNone by a given thread
  while the fetch procedure is still processed in an other thread).


-- S=E9bastien.

Re: [Modeling-users] model for tests

[Sebastien B. <sbi...@us...>]

John Lenton <jl...@gm...> wrote:
> to test one of the patches I'd need a model to have a relation to an
> entity that has subclasses; although StoreEmployee has such a relation
> (Holidays <<-> Employee), it doesn't define the toOne side of that
> relation, and I notice that is used to test exactly that feature.
> Should I create a new model that has these properties, or do I add a,
> say, 'Salary' class?

Yes, go for it and modify the model.  Maybe a BankAccount class could
make more sense?  Okay, anyway that's a test model, so the semantics is
not that important :)

-- S=E9bastien.

[Modeling-users] model for tests

[John L. <jl...@gm...>]

to test one of the patches I'd need a model to have a relation to an
entity that has subclasses; although StoreEmployee has such a relation
(Holidays <<-> Employee), it doesn't define the toOne side of that
relation, and I notice that is used to test exactly that feature.
Should I create a new model that has these properties, or do I add a,
say, 'Salary' class?

-- 
John Lenton (jl...@gm...) -- Random fortune:
bash: fortune: command not found

Re: [Modeling-users] tests failing

[John L. <jl...@gm...>]

On 06 Oct 2004 00:03:03 +0200, Sebastien Bigaret
<sbi...@us...> wrote:
>   Before this is fixed, you can have all the tests run w/ the procedure
>   described in the README: ./run.py (NO -g), plus the three
>   ./test_EditingContext_XXX.py scripts.
> 
> --> The four test script should give zero error & zero warning w/
>     0.9pre17.1 (this is part of the procedure rnu before every release)

ok, except for the encoding errors, this is true here also.

> > am I doing something wrong? I'm guessing the README is outdated,
> > because I tried that and it was even worse; I ran
> > 'test_EditingContext_Global.py -r' and
> > 'test_EditingContext_Global_Inheritance.py -r', and then 'run.py -g'.
> > These two give errors every time, complaining that
> >
> > DROP INDEX WRITER_pkey
> > error: Couldn't evaluate expression DROP INDEX WRITER_pkey. Reason:
> > psycopg.ProgrammingError:ERROR:  cannot drop index writer_pkey because
> > constraint writer_pkey on table writer requires it
> > HINT:  You may drop constraint writer_pkey on table writer instead.
> 
> Another bug here. Workaround: to regenerate the test databases, connect
> to the admin. db template1, then
> 
>     DROP DATABASE "AUTHOR_BOOKS";
>     CREATE DATABASE "AUTHOR_BOOKS";
>     DROP DATABASE "STORE_EMPLOYEES";
>     CREATE DATABASE "STORE_EMPLOYEES"

createdb -E LATIN1 AUTHOR_BOOKS; createdb -E LATIN1 STORE_EMPLOYEES

works fine too (on debian, at least).

> and finally run the scripts:
>     ./test_EditingContext_Global.py -r
>     ./test_EditingContext_Global_Inheritance.py -r
> 
> Your test-dbs should then be okay.
> 
> > Also, one thing that isn't mentioned in the README is that the
> > database must be created with -E LATIN1, or it will produce an(other)
> > error.
> 
> Third bug --could you be more explicit about the error? What is the
> default encoding you use, and how can I reproduce it here (an other -E
> option?) ?
> 
> If I may ask and I you have some time, could you submit bug reports
> @sf.net please?

now I know which are bugs in Modeling and not bungles in my setup, I will.

-- 
John Lenton (jl...@gm...) -- Random fortune:
bash: fortune: command not found

Re: [Modeling-users] CustomObject.globalID() problem

[<no...@ti...>]

On 05 Oct 2004 23:42:47 +0200
Sebastien Bigaret <sbi...@us...> wrote:

> Okay; if you observe that a freshly fetched object has None as globalId(),
> this is definitely a bug, so do not hesitate to report!

I have a deadline on friday and don't have time to write a test, it's nasty
( or simply a problem with my code ) since it appears only if another thread is
fetching the same object from the same EC while I try to retrieve the GlobalID.
Wouldn't worry until I got some real evidence :-)

ciao
--
Delio