Re: deprecating ipaddressfield (at some time)

2013-08-31 Thread Some Developer
On 25/08/2013 11:25, Aymeric Augustin wrote: On 18 mai 2013, at 14:07, Erik Romijn wrote: As far as I know, IPAddressField has no current features that are not also available in a GenericIPAddressField. Therefore, I suggest that we, some time from now, deprecate IPAddressField, in favour of

Re: deprecating ipaddressfield

2013-08-25 Thread Aymeric Augustin
On 25 août 2013, at 19:14, Erik Romijn wrote: > To sum it up, the overhead may be an issue for users for which storage is > very critical, and who will only need to store IPv4 addresses. Users with such stringent storage requirements would be better off storing IPv4 addresses as 32 bits intege

Re: deprecating ipaddressfield

2013-08-25 Thread Marc Tamlyn
+1 to deprecating the old field. This caused me some confusion. Now migrations are in master, the upgrade path will be much simpler. Is the legacy field releasable as a 3rd party package? It seems to be this would be difficult due to differing behaviour across DBs. Marc On 25 Aug 2013 18:14, "Eri

Re: deprecating ipaddressfield

2013-08-25 Thread Erik Romijn
Hi Michael, On Aug 25, 2013, at 6:27 PM, Michael Manfre wrote: > The code on master doesn't agree with your statement about mysql and sqlite. > sqlite - > https://github.com/django/django/blob/master/django/db/backends/sqlite3/creation.py#L26 SQLite doesn't actually have a CHAR type - it will c

Re: deprecating ipaddressfield

2013-08-25 Thread Michael Manfre
On Sun, Aug 25, 2013 at 6:13 AM, Erik Romijn wrote: > I understand your comparison, but the situation is a little different > here. The integer fields you refer to actually have different storage size > requirements in many database backends. But for GenericIPAddressField in > PostgreSQL and SQLi

Re: deprecating ipaddressfield (at some time)

2013-08-25 Thread Aymeric Augustin
On 18 mai 2013, at 14:07, Erik Romijn wrote: > As far as I know, IPAddressField has no current features that are not also > available in a GenericIPAddressField. Therefore, I suggest that we, some time > from now, deprecate IPAddressField, in favour of GenericIPAddressField. Yes. Having both o

Re: deprecating ipaddressfield

2013-08-25 Thread Erik Romijn
Hi Michael, On Aug 25, 2013, at 12:07 AM, Michael Manfre wrote: > IPAddressField is meant for IPv4 addresses and GenericIPAddressField is for > both IPv4 and IPv6. Most backends define different database data types for > each of those fields. E.g. mysql is char(15) vs char(39). Forcing the lar

Re: deprecating ipaddressfield

2013-08-25 Thread Florian Apolloner
On Sunday, August 25, 2013 12:07:11 AM UTC+2, Michael Manfre wrote: > > IPAddressField is meant for IPv4 addresses and GenericIPAddressField is > for both IPv4 and IPv6. Most backends define different database data types > for each of those fields. E.g. mysql is char(15) vs char(39). Forcing the

Re: deprecating ipaddressfield

2013-08-24 Thread Michael Manfre
l mail last May, I haven't seen any opinions against > deprecating IPAddressField, on this list or in #20439 [1]. > > I therefore propose that we start the deprecation of IPAddressField in > 1.7, in favour of GenericIPAddressField (which was introduced in 1.4). That > means PendingD

Re: deprecating ipaddressfield

2013-08-24 Thread Erik Romijn
ressField. Since this initial mail last May, I haven't seen any opinions against deprecating IPAddressField, on this list or in #20439 [1]. I therefore propose that we start the deprecation of IPAddressField in 1.7, in favour of GenericIPAddressField (which was introduced in 1.

deprecating ipaddressfield (at some time)

2013-05-18 Thread Erik Romijn
Hello all, Since Django 1.4, we've added GenericIPAddressField, next to IPAddressField. The new GenericIPAddressField supports IPv4 as well as IPv6 addresses, and does normalisation of IPv6 addresses. It can also be configured to only accept IPv4 or IPv6 addresses. As far as I know, IPAddressF