You can use both solutions but if you do not need the scalability features of GAE and, instead, you need control of your data (for example transaction integrity, the ability to run data mining scripts in background,do full-text search), than do not use GAE just because it free.
Massimo On Feb 25, 3:07 pm, "Ron K. Jeffries" <[email protected]> wrote: > I need to create an application to manage a small (fewer than 1,000 > members) association database. The usual name address phone etc > fields, plus transaction data on payments made etc. > > I plan to use web2py, and fortunately know a Python friend who is willing > to help me get started. He has never used web2py and has not tried GAE > > Is GAE the right choice, or would I be better advised to use a > commercial web hosting site such ashttp://bluehost.com? > > If someone has already created a membership management database app > and is willing to share, that would be most awesome. > > be well > --- > Ron K. Jeffries > [email protected] > > > > > > > > On Fri, Feb 25, 2011 at 12:53, <[email protected]> wrote: > > Today's Topic Summary > > > Group:http://groups.google.com/group/web2py/topics > > > Creating Tables That Allow NULL with MSSQL [1 Update] > > Table Alias [6 Updates] > > More general question: "matchmaking" based on survey/quiz results [1 Update] > > IS_DATE_IN_RANGE & IS_NOT_IN_DB [4 Updates] > > newb - a postgres database view [1 Update] > > Crud onvalidation [1 Update] > > web2py database administration [4 Updates] > > web2py getting some recognition [1 Update] > > The two halves of cache management - what's the recommended method of > > flushing? [1 Update] > > need help on ajax() post [1 Update] > > background operations? [1 Update] > > Redirect & response.flash [1 Update] > > > Topic: Creating Tables That Allow NULL with MSSQL > > > Ross Peoples <[email protected]> Feb 25 12:41PM -0800 ^ > > > I am running into a problem where web2py creates all the tables in my MSSQL > > database with fields that do not allow NULL values. This is regardless of > > the notnull=False setting on the fields. > > > I am using mssql:// as the adapter and having web2py create table > > structures. A test table definition: > > > db.define_table('test_table', > > Field('field_allows_null', notnull=False), > > Field('field_rejects_null', notnull=True) > > ) > > > In MSSQL, both fields will get created to reject NULL values. > > > Topic: Table Alias > > > DenesL <[email protected]> Feb 25 08:27AM -0800 ^ > > > Instead of using db.GlobalSettings you can assign it to a variable > > > global_settings = db.GlobalSettings > > > and use the var in place for all practical purposes e.g. > > > db().select(db.GlobalSettings.ALL) > > > becomes > > > db().select(global_settings.ALL) > > > Ross Peoples <[email protected]> Feb 25 11:32AM -0500 ^ > > > True, but I'd still have the issue of: > > > db().select(global_settings.SettingName) > > > when using any field names. > > > On Feb 25, 2011, at 11:27 AM, DenesL wrote: > > > Vasile Ermicioi <[email protected]> Feb 25 06:42PM +0200 ^ > > >http://weblogs.sqlteam.com/mladenp/archive/2006/12/07/44176.aspx > > > if you use a case insensitive collation you can use lower case, but without > > underscore > > > db.define_table('globalsettings', Field('settingname')) > > > rows = db(db.globalsettings).select() > > > Ross Peoples <[email protected]> Feb 25 11:59AM -0500 ^ > > > I believe we are using case insensitive collation, but I we weren't we'd be > > screwed...as would any PostgreSQL user with forced capitals in the table > > names. What is the web2py convention for naming tables and fields anyways? > > > On Feb 25, 2011, at 11:42 AM, Vasile Ermicioi wrote: > > > DenesL <[email protected]> Feb 25 12:03PM -0800 ^ > > > What issue?. > > > Ross Peoples <[email protected]> Feb 25 03:33PM -0500 ^ > > > I guess this was more of a feature request, since I will soon have to tackle > > an application using PostgreSQL that forces case-sensitive naming. > > > On Feb 25, 2011, at 3:03 PM, DenesL wrote: > > > Topic: More general question: "matchmaking" based on survey/quiz results > > > Lorin Rivers <[email protected]> Feb 25 02:07PM -0600 ^ > > > I want to connect users on a site based on their responses to a series of > > questions (some sample questions): > > Smoker? [y/n] (possibly with a flip-side question: Tolerate smokers? [y/n]) > > Carnivore or Vegetarian? > > > and so on, with a goal of finding other "compatible" people who've taken the > > survey. > > > Can you think of a light-weight solution? > > > Thanks! > > > -- > > Lorin Rivers > > Mosasaur: Killer Technical Marketing <http://www.mosasaur.com> > > <mailto:[email protected]> > > 512/203.3198 (m) > > > Topic: IS_DATE_IN_RANGE & IS_NOT_IN_DB > > > greenpoise <[email protected]> Feb 25 11:01AM -0800 ^ > > > How could I combine these two? I have a table that accepts two dates. > > I would not like duplicates between in those entry dates. > > > Basically a From Date and a To Date > > > Thanks > > > Dan > > > Jonathan Lundell <[email protected]> Feb 25 11:14AM -0800 ^ > > > On Feb 25, 2011, at 11:01 AM, greenpoise wrote: > > >> How could I combine these two? I have a table that accepts two dates. > >> I would not like duplicates between in those entry dates. > > >> Basically a From Date and a To Date > > > So the restriction is essentially: no overlapping date ranges in the > > database? > > > greenpoise <[email protected]> Feb 25 11:23AM -0800 ^ > > > exactly, no overlapping date ranges > > > Jonathan Lundell <[email protected]> Feb 25 12:00PM -0800 ^ > > > On Feb 25, 2011, at 11:23 AM, greenpoise wrote: > > >> exactly, no overlapping date ranges > > > If I were doing it, I think I'd start with a (renamed) copy of > > gluon.validators.IS_NOT_IN_DB in my model file, and change it to do a range > > check on two fields, instead of an equality check on one. > > > I'm sort of fuzzy on how you do date comparisons, but presumably there's a > > way. > > > Topic: newb - a postgres database view > > > pbreit <[email protected]> Feb 25 11:55AM -0800 ^ > > > A simple "thank you" would suffice. > > > Topic: Crud onvalidation > > > greenpoise <[email protected]> Feb 25 11:53AM -0800 ^ > > > I have a simple validation according to the books page 296. Is > > actually simpler and it is not even being evaluated, no errors > > either... Here it is: > > > def f(form): > > if form.vars.suns>form.vars.sunv: form.errors.sunv="cant be > > more than 9" > > > return dict(employee=employee,form=form,onvalidation=f) > > > Topic: web2py database administration > > > stargate <[email protected]> Feb 25 08:50AM -0800 ^ > > > But how do you make it so that it will generate a Registration ID > > automatically. When i create a new application with the db wizard and > > connect to the sqllite database I don't have any of these problems. I > > can create a new user with out giving a registration id > > > Anthony <[email protected]> Feb 25 10:30AM -0800 ^ > > > On Friday, February 25, 2011 11:50:40 AM UTC-5, stargate wrote: > > >> automatically. When i create a new application with the db wizard and > >> connect to the sqllite database I don't have any of these problems. I > >> can create a new user with out giving a registration id > > > web2py doesn't generate the registration_id -- it is something received from > > Janrain when a user logs in via Janrain. If the login isn't happening via > > Janrain, then there won't be any registration_id. If you want to avoid that > > error, I think you have three options: > > > 1. Enter your own unique registration_id for each user in appadmin (I > > think any unique string should work, including just repeating the username > > or email); > > 2. Add new users via the regular application registration form (i.e., > > /[your_app]/default/user/register) instead of via appadmin. The standard > > registration form doesn't show the registration_id field (because readable = > > writable = False), so I don't think it ends up getting validated, and > > therefore there is no validation error; or > > 3. Remove the following line: > > > db.auth_user.registration_id.requires = IS_NOT_IN_DB(db, > > db.auth_user.registration_id) > > > As far as I can tell, the registration_id error you're getting in appadmin > > should happen any time you use the wizard to create an app with auth, even > > using SQLite (I get the same error when I try it). Anyway, it should be easy > > enough to avoid via the above options. > > > Anthony > > > stargate <[email protected]> Feb 25 11:19AM -0800 ^ > > > Thanks for all the help it triggered a idea I will try > > > Anthony <[email protected]> Feb 25 11:27AM -0800 ^ > > > On Friday, February 25, 2011 2:19:17 PM UTC-5, stargate wrote: > > >> Thanks for all the help it triggered a idea I will try > > > No problem. Let us know what you do and how it works. > > > Anthony > > > Topic: web2py getting some recognition > > > ron_m <[email protected]> Feb 25 09:26AM -0800 ^ > > >https://groups.google.com/d/msg/comp.lang.python/iEV5oyLApHI/laV-P_Q6... > > > Topic: The two halves of cache management - what's the recommended method > > of flushing? > > > Carl <[email protected]> Feb 25 09:18AM -0800 ^ > > > Web2py has an elegant interface to using cache (ram/disk/memcache) but > > is there a hole when it is applied to database actions? > > > When using database .select() statements one can specify the use of > > caching using... > > e.g., ... .select(cache=(cache.ram, 60 * 60 * 24)) # cache result for > > a day > > > To flush a cache I can't find a clean/targeted solution. > > > I've read postings (mdipierro, 9mar2010): > > ... after insert/update/delete you do > > db(...).select(cache=(cache.disk,0)) > > > or: > > cache.ram.clear(db._uri + '/SELECT') > > > The former flushes all the contents of the cache and the latter, using > > the private variable _uri, restricts the flush to all database > > selects. > > > Is there a better way extend Web2py's cache system to supporting > > databases? > > > Can we get one more step of specificity? Namely, can we limit the > > flush to a specific table or tables? > > > Topic: need help on ajax() post > > > DenesL <[email protected]> Feb 25 08:51AM -0800 ^ > > > Please tell us which web2py version, which browser, which OS, etc. > > > Topic: background operations? > > > mika <[email protected]> Feb 25 08:22AM -0800 ^ > > > Hello, My application needs to download ON DEMAND some data from external > > xml. I would like user to can start the process, and just see that the job > > was started, and then to use application normally. The downloading and > > parsing xml happens in the background, after this user is notified (flash > > popup?). How to achieve this? I checked about cron, I checked about services > > and background process but I am not sure is it proper way to do that. BR, > > Mika > > > Topic: Redirect & response.flash > > > Jonathan Lundell <[email protected]> Feb 25 08:11AM -0800 ^ > > > On Feb 25, 2011, at 2:02 AM, SergeyPo wrote: > >> redirect(URL(r=request, f='headers', args=(request.args[1]))) > > >> response.flash does not keep its value when you redirect. Is it > >> possible to fix or workaround? > > > As others have already said, you want to use session.flash here. It's useful > > to understand why. > > > Redirect is itself a response, sending a redirection code and the new URL to > > the requesting browser. When that happens, everything in response, including > > response.flash, is discarded--the redirection becomes the entire response. > > > On every request from the browser, web2py initializes a new response > > dictionary (along with request and some other stuff). It sets response.flash > > = session.flash, and then sets session.flash = None. So anything you store > > in session.flash is used as response.flash for the *next* request/response, > > which is exactly what you want when you're calling redirect.
