I got the same issue, delete operations by id work locally, but not on
gae (dev_appserver or live).
Modifying from .count() to len(..) appears to fix this for me. I
locally modified the delete method in gql.py, line 692 as follows:
def delete(self):
self._db['_lastsql'] = 'DELETE WHERE %
Hi all -
I wonder if anyone else has hit the file-size upload limits on GAE?
This limit is supposed to be 10MB which for my needs would be fine.
The GAE file system is read-only so I save upload data in the database
like this:
Field('file','upload',uploadfield='file_binary', ...
Field('file_bina
ython supports multiple
>
> files in memory.
>
> However this will be very expensive to do, if you have a 10MB file it
> will most likely take at "least" double the ram to perform this
> operation.
>
> -Thadeus
>
>
>
> On Mon, Feb 22, 2010 at 9:29 AM, wha
Hi Snaky -
Identical source code can be deployed both to GAE and elsewhere
without editing, you don't need to edit connection strings before
deploying to GAE. The default db.py code you get with a new project
detects if gae is running with the line
if request.env.web2py_runtime_gae:
...
using thi
I had not heard of the blobstore service before - thank you for
pointing that out. I agree - well worth considering, up to 1GB files
the service is still free.
On Feb 22, 6:22 pm, villas wrote:
> For anything above 1Mb, I think Google are trying their best to
> promote their new 'blobstore' whic
I got web2py working with blobstore, have written up details on the
steps I took here: http://www.web2pyslices.com/main/slices/take_slice/63
- Alex
On Feb 22, 7:42 pm, what_ho wrote:
> I had not heard of the blobstore service before - thank you for
> pointing that out. I agree - well
I got the blobstore service working with web2py for up to 50MB file
support on GAE. Have written up the steps I took here:
http://www.web2pyslices.com/main/slices/take_slice/63
- Alex
On Feb 22, 7:42 pm, what_ho wrote:
> I had not heard of the blobstore service before - thank you
no problem!
With some informal testing I found the blobstore absolutely flies
along, serves up files as fast as whatever internet connection I tried
could handle. I did not load test with multiple users, but I don't
doubt this service is built to scale.
The blobstore is meant for file upload and
I have used css in forms to position the default submit button as the
first button declared in the form html before, even though on-screen
it may appear anywhere in the form. If you need a javscript-free
solution this arrangement works across all the browsers.
The basic premise is:
As an example of a multiple select field definition:
db.define_table('fruit',Field('name'))
db.define_table('fruit_basket',Field('contents'))
db.fruit_basket.contents.requires=IS_IN_DB(db,'fruit.id','%
(name)s',multiple=True)
db.fruit.insert('apple')
db.fruit.insert('orange')
db.fruit.insert('ban
u suggest.
>
> Anyway |1|2|11| allows you to search for |1| or |2| or |11|.
>
> This 1|2|11 would create problems when searhing for 1, in
> particularing considering that values can be non-numeric.
>
> On Mar 21, 5:48 pm, what_ho wrote:
>
>
>
> > As an exam
This is a breaking change I support. For other UI widgets like the
text widget, the default behaviour is not to submit any value during a
form submit without explicit user input. It makes sense to align the
default behaviour of other UI widgets as much as possible to the
same.
When other framework
Yes - correct, sorry I didn't make that clear. I support the
historical breaking change to zero=''
On Mar 22, 2:12 pm, mdipierro wrote:
> You say
>
> On Mar 22, 7:01 am, what_ho wrote:
>
> > This is a breaking change I support. For other UI widgets like
Intrigued by the recommendation to put code in modules instead of
models if possible.
At present I have db.define_table .. method calls in a model file. The
database structure will stay the same between releases, so it does not
feel optimal at present to have such definitions run on every page
re
I did some quick and dirty performance tests to measure the time taken
by db.define_table calls.
Test was as follows:
- Run web2py locally, with built-in web server
- Create new application 'perftest' in web2py admin interface
- Add 5 dummy table definitions to db.py, uncomment mail and auth
sett
:
not compiled: 22.4s
compiled: 16s
- Alex
On Aug 24, 11:19 am, mdipierro wrote:
> thanks for this test. Would you be able to also test them setting
>
> db.define_table(,migrate=False)
>
> ?
>
> which is what people should do on production.
>
> Massimo
>
> O
no problem!
correct - 12ms per request is the time I recorded to completely
process one webpage request, from the start point of the http request
headers being sent to the server, to the last byte of the HTTP 200
response being successfully received by the client.
This is using all default setti
Hi all - just had time this weekend to look further into this.
Thank you to suggestions from Iceberg, Yarko and Massimo. I have tried
putting table definitions into a module, and also run some timing
tests against a Postgres database
As a note, for my previous tests coming out at 12ms page respo
components.
Cheers,
-Alex
On Aug 23, 9:04 pm, Yarko Tymciurak wrote:
> On Sun, Aug 23, 2009 at 1:55 PM, what_ho wrote:
>
> > Intrigued by the recommendation to put code in modules instead of
> > models if possible.
>
> > At present I have db.define_table .. method ca
, 1:28 pm, Iceberg wrote:
> Thanks for the interesting test, Pearson (What_Ho).
>
> Besides of trying Yarko's "putting table definitions in gluon just as
> a quick hack to compare performance", would you mind try a more
> general way?
>
> A. Put table definitio
20 matches
Mail list logo
