On Nov 22, 1:05 am, sharekey <[EMAIL PROTECTED]> wrote:
> I can insert and select records from GAE Development Console. and I
> can insert record from
> http://127.0.0.1:8080/myapp/appadmin/insert/db/mytable,
> select record fromhttp://127.0.0.1:8080/myapp/default/index. But I
> can not select
I can insert and select records from GAE Development Console. and I
can insert record from http://127.0.0.1:8080/myapp/appadmin/insert/db/mytable,
select record from http://127.0.0.1:8080/myapp/default/index. But I
can not select from http://127.0.0.1:8080/myapp/appadmin/select/db/mytable,
I get
Yes!
That is the problem. Your session_data is 'text' while it should be
'bytea'. web2py is attempting a migration losing data and postgres
does not allow copy text into a bytea.
Massimo
On Nov 22, 12:26 am, morningovermidnight <[EMAIL PROTECTED]>
wrote:
> Ok, with version 1.51 running I added
Ok, with version 1.51 running I added the line to gluon/globals.py.
Here's the message returned:
Traceback (most recent call last):
File "/home/basic/workbench/sites/newsite/gluon/restricted.py", line
62, in restricted
exec ccode in environment
File "/home/basic/workbench/sites/newsite/ap
svn version works fine now - thx :)
On Fri, Nov 21, 2008 at 6:41 PM, mdipierro <[EMAIL PROTECTED]> wrote:
>
> Thank you Wes for insisting. You are right. The bug was reintroduced
> when the "resolve conflict" feature was introduced.
> I just reposted trunk and it should be fixed.
>
> Massimo
>
>
I did a diff of 1.51 vs 1.47. As I remembered the only think that
changed in session management is storing the pickle from 'text' to
'blob'. This actually fixed a problem with sessions in mysql, as
pointed out by CJ.
Please do a test for me. In gluon/globals.py
replace these lines with:
Thank you Wes for insisting. You are right. The bug was reintroduced
when the "resolve conflict" feature was introduced.
I just reposted trunk and it should be fixed.
Massimo
On Nov 21, 7:33 pm, "Wes James" <[EMAIL PROTECTED]> wrote:
> But this is latest svn version.
>
> On 11/21/08, mdipierro <
But this is latest svn version.
On 11/21/08, mdipierro <[EMAIL PROTECTED]> wrote:
>
> You have an old version of web2py with removes ALL trailing spaced.
> If the last line is a comment it is a problem because python does not
> like comments not ending in \n.
>
> This was fixed in 1.48 I believe.
That is because a form is reset to default upon accept, unless you
accept with keepvalues=True.
Massimo
On Nov 21, 5:41 pm, billf <[EMAIL PROTECTED]> wrote:
> I have been going back to basics and running my own test - just web2py
> - and I agree that it does all seem to work ok (correct data on
You have an old version of web2py with removes ALL trailing spaced.
If the last line is a comment it is a problem because python does not
like comments not ending in \n.
This was fixed in 1.48 I believe.
Massimo
On Nov 21, 5:36 pm, "Wes James" <[EMAIL PROTECTED]> wrote:
> When a file is saved i
we agree Bill.
On Nov 21, 5:15 pm, billf <[EMAIL PROTECTED]> wrote:
> I did suggest at the outset (18/11) that creating a new version of
> itemize was probably the easiest solution :-).
>
> The in-memory database was just a response to "how can I pass data to
> itemize?" - not a recommendation.
>
I have been going back to basics and running my own test - just web2py
- and I agree that it does all seem to work ok (correct data on db and
correct data in form) with one exception (see next para). I think I
overlooked the fact that, for standard form processing, update() only
updates the field
When a file is saved in the gui (i.e. db.py), i get a syntax error at
the end of the db.py file as a ticket
when i save a file with wingIDE this does not happen
latest svn
-wj
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Goog
I did suggest at the outset (18/11) that creating a new version of
itemize was probably the easiest solution :-).
The in-memory database was just a response to "how can I pass data to
itemize?" - not a recommendation.
On Nov 21, 10:18 pm, "Yarko Tymciurak" <[EMAIL PROTECTED]> wrote:
> On Fri, No
btw, i'm using sqlite right now
On Fri, Nov 21, 2008 at 3:48 PM, Wes James <[EMAIL PROTECTED]> wrote:
> thx for the tip. This helped me get in to my app with latest svn
> version. It had a session error otherwise.
>
> -wj
>
> On Fri, Nov 21, 2008 at 1:31 PM, kev <[EMAIL PROTECTED]> wrote:
thx for the tip. This helped me get in to my app with latest svn
version. It had a session error otherwise.
-wj
On Fri, Nov 21, 2008 at 1:31 PM, kev <[EMAIL PROTECTED]> wrote:
>
> Try deleting all the session from the database.
--~--~-~--~~~---~--~~
You receive
Can you try using the sqlite database?
Can you send me your app privately?
Massimo
On Nov 21, 4:25 pm, morningovermidnight <[EMAIL PROTECTED]>
wrote:
> Ok, I tried it out...removed the .table files, dropped the session
> table, removed cookies, still same problem...nothing being returned
> from
Ok, I tried it out...removed the .table files, dropped the session
table, removed cookies, still same problem...nothing being returned
from sessions. Specifically, I have a login form. When the wrong
username is entered, there is a session flash that says "invalid
username". This doesn't show with
On Fri, Nov 21, 2008 at 4:11 PM, mdipierro <[EMAIL PROTECTED]> wrote:
>
> Please check now. The DNS got messed up.
+1 ;-)
lookin' good!
>
>
> Massimo
>
> On Nov 21, 4:03 pm, "Yarko Tymciurak" <[EMAIL PROTECTED]> wrote:
> > yeah - I too am getting the eternal "...waiting on
> mdp.cti.depaul.ed
On Fri, Nov 21, 2008 at 4:15 PM, mdipierro <[EMAIL PROTECTED]> wrote:
>
> The fact we can store a database in memory does not mean this is the
> right solution to your problem.
>
> I would just take the code out of t2 itemize and adapt to your
> purposes. It is relatively small.
+1
>
>
> itemi
I need a patch to do the check of version in background using $.ajax
so that this problem does not arise again.
Massimo
On Nov 21, 4:09 pm, "Yarko Tymciurak" <[EMAIL PROTECTED]> wrote:
> yes - this must be slow note because of web2py, but
> becausewww.web2py.com(forwhatever reason) is not respo
The fact we can store a database in memory does not mean this is the
right solution to your problem.
I would just take the code out of t2 itemize and adapt to your
purposes. It is relatively small.
itemize does very very little for you so I doubt you need it.
Massimo
On Nov 21, 4:07 pm, Baron
Please check now. The DNS got messed up.
Massimo
On Nov 21, 4:03 pm, "Yarko Tymciurak" <[EMAIL PROTECTED]> wrote:
> yeah - I too am getting the eternal "...waiting on mdp.cti.depaul.edu"
> when trying to get towww.web2py.com
>
> On Fri, Nov 21, 2008 at 3:36 PM, kev <[EMAIL PROTECTED]> wrote:
Good point. There is no way to get rid of the delete box in T2. we
need a pacth!
The exposes is in the t2 manual. At the end.
On Nov 21, 3:40 pm, "Wes James" <[EMAIL PROTECTED]> wrote:
> Massimo,
>
> thx, this works. I noticed with sqlform that I can do deletable=False
> (i guess to get rid of
yes - this must be slow note because of web2py, but because
www.web2py.com(for whatever reason) is not responding.
On Fri, Nov 21, 2008 at 3:50 PM, Wes James <[EMAIL PROTECTED]> wrote:
>
> ceej,
>
> Thx!! i also see a timout param in there. I added a 0 on to one of
> the 60's as i've been bitten
sorry, I may have my object types wrong.
To summarize:
I'm trying to pass data directly to t2.itemize() without having to
write to a database.
If I understood right, Bill was hinting that writing to an in-memory
database was the easiest way, so I was wondering what was the harder
alternative way
yeah - I too am getting the eternal "...waiting on mdp.cti.depaul.edu"
when trying to get to www.web2py.com
On Fri, Nov 21, 2008 at 3:36 PM, kev <[EMAIL PROTECTED]> wrote:
>
> Hey,
> Thanks for reply. I tested the site right now and it is still down.
> (according to my internet and siteuptime
I am using PostgreSQL, but I will try what you recommend and let you
know how it works out.
On Nov 21, 3:57 pm, mdipierro <[EMAIL PROTECTED]> wrote:
> Are you using mysql? The problem may be a bug in web2py 1.47 that has
> been fixed in an non-backward incompatible way. In 1.47 a session was
> pi
ceej,
Thx!! i also see a timout param in there. I added a 0 on to one of
the 60's as i've been bitten a few times with editting in the gui and
click on save and it asks for the prompt to login and I lose my edits.
I learned to log in and then back arrow to my edits and save them.
This expire ch
It's because web2py checks web2py.com to see if your version matches
with the latest version and Massimo just updated the server. In your
app you can create a model named 0.py and add CHECK_VERSION=False to
stop this :)
On Nov 21, 1:34 pm, "mr.freeze" <[EMAIL PROTECTED]> wrote:
> Yes, the trunk v
Massimo,
thx, this works. I noticed with sqlform that I can do deletable=False
(i guess to get rid of delete checkbox)
How is this done with t2?
Also, I looked for exposed and exposes in the t2/web2py manuals and I
don't see these.
-wj
On Fri, Nov 21, 2008 at 1:47 PM, mdipierro <[EMAIL PROTE
Hey,
Thanks for reply. I tested the site right now and it is still down.
(according to my internet and siteuptime)
Kev
On Nov 21, 3:49 pm, mdipierro <[EMAIL PROTECTED]> wrote:
> I have moved the web site to a different virtual machine today. It
> should be up now.
> No. there is no delete record
Yes, the trunk version last night was fast, but it is slow for me to
login today.
On Nov 21, 1:20 pm, "Wes James" <[EMAIL PROTECTED]> wrote:
> Can anyone else verify if with the latest svn version of web2py if it
> takes a long time to log in the admin area.
>
> it's doing it on mac pro and mac b
Oh! So when I use exposes on say 10 out of 20 fields then
{{=form}} will only show those 10 fields and leave the other 10 alone
- that would be good with the checkbox issue. How do I format my
output then so =form does do its own table format? Is this what
sqlform() does?
On Fri, Nov 21, 2
yes.
rationale:
T2 does a lot of stuff for you. db.table.exposes tells T2 which fields
you want in forms. If you don't it assumes all fields. Yet you are
manually extracting one field from the form object and placing only
that one into the HTML form. web2py has no way of knowing that you are
dis
Are you using mysql? The problem may be a bug in web2py 1.47 that has
been fixed in an non-backward incompatible way. In 1.47 a session was
picked in a text field. In mysql text fields cannot be longer than
2**16 bytes. So since 1.48 they are pickled in a blob.
Try 1.51 again but please drop the
I'm don't know why that is evil (would you explain?). Anyway, I'm
using that to iterate through all the check boxes (and also place
other components) on the page manually - so I can format their
placement how I want.
So the exposes will fix the checkbox issue?
On Fri, Nov 21, 2008 at 1:47 PM,
Wes,
let's take this privately from now on. Some of your questions may
confuse other people.
The "id" is the "id" of the field you are editing {{form.record_id}}.
You should let web2py give you values for fields, not hard code them
in forms.
Massimo
On Nov 21, 2:36 pm, "Wes James" <[EMAIL PROT
I have moved the web site to a different virtual machine today. It
should be up now.
No. there is no delete record you have to do
db(db.table.id==someobject.id).delete()
Massimo
On Nov 21, 2:29 pm, kev <[EMAIL PROTECTED]> wrote:
> Hello,
> Is there a way to do something like someobject.delete_
Thanks for the reply. I tried deleting all sessions from the database,
still did not solve the problem for me. In doing some investigation to
solve the problem, I changed my web2py version from 1.51 to version
1.47 (replaced files with the 1.47 source). Now, everything works
perfectly. Not sure if
You are doing something evil in page1.html: {{=form[0][0][1]}}
Anyway, given your custom forms and I assume you do not want to change
them. you need to replace
@t2.requires_login(next='login')
def page1():
stus=db(db.student.id).select()
if stus:
form=t2.update(db.student,db.stud
Also, I'm wondering; in the forms at the bottom you get something like
this with a {{=form}}:
is the id always 1 for the hidden id input field? I have this in my
layout as static text and I'm wondering how that might change with
different users. I did look at it with a different user and
Try deleting all the session from the database.
On Nov 21, 2:51 pm, morningovermidnight <[EMAIL PROTECTED]>
wrote:
> I'm not sure what's happening. It was all working before. I was
> storing sessions in the database and working with them throughout my
> application with no problem. Now, it seems
Hello,
Is there a way to do something like someobject.delete_record() just
like someobject.update_record()?
Also the website web2py.com seems to be down atm for me.
Thanks,
Kev
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Googl
OK. Here is a demo of the checkbox issue:
http://paste.pocoo.org/show/92051/
Put the code above in to a new app.
Select page 2 and type some text in to the text box and select any
number of checkboxes and click submit.
Now go to page 1 and click on save
Now go to page 2, you will see that th
I'm not sure what's happening. It was all working before. I was
storing sessions in the database and working with them throughout my
application with no problem. Now, it seems suddenly, sessions have
stopped working. I don't get a session.flash or any other session
objects returned. When I comment
Another glitch I noticed:
The latest version in trunk occassionally displays the "newer version
available" message.
Daniel
On Fri, Nov 21, 2008 at 20:20, Wes James <[EMAIL PROTECTED]> wrote:
>
> Can anyone else verify if with the latest svn version of web2py if it
> takes a long time to log in
Can anyone else verify if with the latest svn version of web2py if it
takes a long time to log in the admin area.
it's doing it on mac pro and mac book pro that i just did itunes
update - but i'm not sure that is what started this.
thx,
-wj
--~--~-~--~~~---~--~~
Yes, I use something similar ATM, but I'm concerned about the overhead
if you have a large number of tables...
On Nov 21, 7:26 pm, mdipierro <[EMAIL PROTECTED]> wrote:
> if not db(db.table.id>0).count(): db.table.import_from_csv_file(open
> ('mytable.csv'))
>
> Massimo
>
> On Nov 21, 12:24 pm, ac
And while on a database migration topic, I would also very much
welcome some index functionality. The easiest way to implement that in
a backwards compatible way would probably to have an additional
attribute in the sqlfield statement. Example:
SQLField('name','string',index=True,indexlength=5)
if not db(db.table.id>0).count(): db.table.import_from_csv_file(open
('mytable.csv'))
Massimo
On Nov 21, 12:24 pm, achipa <[EMAIL PROTECTED]> wrote:
> We have talked about this previously, but the CSV import/export in the
> previous topic reminded me - wouldn't it be nice to have a sort of
> SQL
We have talked about this previously, but the CSV import/export in the
previous topic reminded me - wouldn't it be nice to have a sort of
SQLDefaultData('mytable.csv') entry in the model, which would contain
the default contents of the table ? If the migrator decides it needs
to create the table,
Sorry for all this hassle.
I will do some more testing, but for now this is what I have working.
On my first page I have:
{{for cbi in range(26,76):}}
{{pass}}
I have hidden actual checkbox elements using css on the div
(#hidethis{display:none;}
The values are being kept now in all cases.
I am sure there is no bug here. Wes, I suspect you are trying to
reinvent something that exist but I cannot figure out what.
If you just want to hide fields you can do:
$(document).ready(function() {
$('#tablename_fieldname__row').hide();
// list other fields...
});
On Nov 21, 11:03 am, "Y
thanks for directing me to look at the model. that is what was causing
the problem...
after trying to look at what was causing problems in my own model, i
decided to see if it would work in other working applications. I
uploaded reddish and added the test.py module and tried running the
module. i
Wes -
If you're doing things a "hard" way when there's a direct way, it would be
nice for Massimo to point that out (and maybe that will help document
something better);
If there's a bug that we don't understand, then I'd like the loop to be
closed so that can be addressed - either a patch from b
Yarko,
so far, i can get it to work if the checkbox fields are not hidden on
the form. My problem now is trying to make 'hidden' checkbox form
elements. I don't see any way to hide checkbox elements.
On Fri, Nov 21, 2008 at 9:39 AM, Yarko Tymciurak <[EMAIL PROTECTED]> wrote:
> Wes -
>
> I thin
*sigh*
On Fri, Nov 21, 2008 at 11:02 AM, Wes James <[EMAIL PROTECTED]> wrote:
>
> oh, i forgot, i can hide with css
>
> i'm going to try that.
>
> On Fri, Nov 21, 2008 at 10:00 AM, Wes James <[EMAIL PROTECTED]> wrote:
> > Yarko,
> >
> > so far, i can get it to work if the checkbox fields are
i'll build a simple app with a text input and a couple of check boxs
and see how that works.
-wes
On Fri, Nov 21, 2008 at 10:02 AM, Yarko Tymciurak <[EMAIL PROTECTED]> wrote:
> Understood - but Bill thinks there's a bug, and is proposing patches;
> Massimo is thinking perhaps you're doing someth
oh, i forgot, i can hide with css
i'm going to try that.
On Fri, Nov 21, 2008 at 10:00 AM, Wes James <[EMAIL PROTECTED]> wrote:
> Yarko,
>
> so far, i can get it to work if the checkbox fields are not hidden on
> the form. My problem now is trying to make 'hidden' checkbox form
> elements.
Understood - but Bill thinks there's a bug, and is proposing patches;
Massimo is thinking perhaps you're doing something wrong and no patch should
be needed - having an example - even if you want to send privately to
Massimo - would help close this disconnect.
Thanks.
On Fri, Nov 21, 2008 at 11:0
Wes -
...
So your student table has boolean fields called "checkbox1", "checkbox2" and
"checkbox3"?
Those names sound off ... not like things that describe a student data
record.
I think we need more context.
Something we can reproduce - complete example would really be useful at this
point
if i put:
db.student.exposes=['checkbox1','checkbox2','checkbox3','checkbox4']
in db.py
then i get this error:
Traceback (most recent call last):
File "gluon/restricted.pyc", line 62, in restricted
File
"/Users/Shared/web2py.app/Contents/Resources/applications/cnrsa/views/default/profile.
Wes -
I think we need a complete - if minimized example which others can use to
reproduce. Can you provide?
Thanks,
Yarko
On Fri, Nov 21, 2008 at 10:15 AM, mdipierro <[EMAIL PROTECTED]> wrote:
>
> SQLFORM(...,fields=[...])
>
> should be used to determine which fields are expected and which sho
SQLFORM(...,fields=[...])
should be used to determine which fields are expected and which should
be ignored.
For T2 use
db.table.exposes=[] # list of field names
Massimo
On Nov 21, 9:52 am, billf <[EMAIL PROTECTED]> wrote:
> Massimo
>
> I think there is a typo in your point 1) and it shou
well, this does not work either, I tried using the default value
'checked' then tried 'true', but when I save the form the value is not
retained:
{{for cbi in range(26,76):}}
{{pass}}
It seems that there is no such think as a hidden checkbox, because:
when submitted does not keep its value,
Massimo
I think there is a typo in your point 1) and it should read "update
form and it was *checked* in the original record". The behaviour of
the setting the checkbox value is not in question.
I believe the problem Wes originally raised is that a boolean field on
the database gets set to Fals
You should not be doing that. You are clearly doing something wrong in
the controller. Can you post the relevent parts of your controller and
custom views?
Massimo
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"web
like this since the one line was an output line to let me see what was
happending:
{{for cbi in range(26,27):}}
{{pass}
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"web2py Web Framework" group.
To post to th
This is how I have my fields set:
SQLField("fieldname","boolean")
I'll need to do what billf said, loop through my checkbox fields and
set them as hidden.
something like this:
{{for cbi in range(26,27):}}
{{=form[0][cbi][1][0]['_checked']}}
{{pass}
On Fri, Nov 21, 2008 at 7:10 AM, mdipierro
wait
yes, a functions can return multiple objects in a tuple
no, a controller action should not do that. an action returns a string
(the web page) or a dictionary (defining the symbols exposed to the
view).
Massimo
On Nov 21, 8:47 am, kaj <[EMAIL PROTECTED]> wrote:
> Hi - and thanks for yo
Hi - and thanks for your reply;
Yes, I realized this is a python issue, and found my python manual
telling me that I could combine several dicts in the return statement
by making a tuple out of them, just listing them one by one, separated
by a comma.
Regards,
Kaj R.
On Nov 21, 3:18 am, "mr.fre
It was mentioned in one of the previous emails. I am not sure it helps
with the current problem. I am not sure I understand the current
problem.
checkboxes are always unchecked unless:
1) update form and it was unchecked in the original record
2) create form and corresponding field has default=Tr
Massimo
In what way are you suggesting "SQLFORM and and FORM take a parameter
hidden=dict(key='value')" would be used?
Bill
On Nov 21, 2:10 pm, mdipierro <[EMAIL PROTECTED]> wrote:
> Do you have default=False in the database?
>
> SQLFORM and and FORM take a parameter hidden=dict(key='value')
>
Yes you need the -M in order to call "print db.tables" but now we know
that the bug is in your model file(s). If you post it we may be able
to help.
Massimo
On Nov 21, 7:33 am, shlo <[EMAIL PROTECTED]> wrote:
> taking out the -M works great, but i am now getting the following in
> the traceback:
Do you have default=False in the database?
SQLFORM and and FORM take a parameter hidden=dict(key='value')
Massimo
On Nov 21, 6:32 am, billf <[EMAIL PROTECTED]> wrote:
> I am investigating a possible solution to the above problem that would
> return fieldname=OFF if a displayed checkbox is unche
That is not necessarily a problem. Do not trust appadmin on GAE. can
you insert and select records?
On Nov 21, 1:47 am, sharekey <[EMAIL PROTECTED]> wrote:
> Hi Massimo,
> Thank you very much, I get my answer from your point three of case
> two(there are no tables either, until data is stored
You have the option of using form=SQLFORM and then form[0].insert(2,TR
(...)) the rows from the other table.
It may be easier to create the form using form_factory.
Massimo
On Nov 21, 12:34 am, mmstud <[EMAIL PROTECTED]> wrote:
> Im thinking to create an insert and update form with fields taken
taking out the -M works great, but i am now getting the following in
the traceback:
global name 'db' is not defined
this is my revised module:
import time
try:
while 1:
time.sleep(60)
print db.tables
except Exception,e:
print e
thanks again!
On Nov 20, 9:44 pm, mdipier
I am investigating a possible solution to the above problem that would
return fieldname=OFF if a displayed checkbox is unchecked.
This would enable web2py to retain existing values for a boolean field
if a request.var was not present.
However, I recognise that some people may consider returning
The following is a proposed patch to sqlhtml.py so that record fields
of type 'boolean' are added as hidden form fields if they are not
already included in the form. In this way, the status of a checked/
true field is maintained even if the field is not included in the form
as a checkbox.
In the
If the following explanation is correct it is a big coincidence
because I have just been looking at this behaviour.
When a form is submitted, if there is no parameter for a field then
one of 2 things happen (simplified):
If it is NOT a checkable form control (checkable=check box or radio
button)
82 matches
Mail list logo
