I've always been reluctant to raise the version requirements for Propel. As
of now, Propel2 requires PHP5.3.3, but I have good reasons to believe we
should raise that to 5.4.
Propel2 is only embracing PHP5.3 only now, and two of is main features
(Late Static Binding and Namespaces) have major implications on the entire
architecture. Using traits for the core (not optionally) would transform
the architecture even more dramatically, and would pave the path for
extremely powerful features and extensivity.
Behaviors are the first to come to mind, but traits also allow true PoPo
(Plain Old Php Objects) to become ActiveRecord objects without using
inheritance. I'm not talking about abandoning code generation. Propel's use
of code generation, both for speed and for IDE friendliness, is our
signature. I'm talking about generated traits. Besides, there are numerous
parts in Propel that could be refactored into traits, even be released as
standalone libraries (think about the ConditionalProxy for ModelCriteria
for instance). Lastly, traits will be used by the end users, for sure, to
extend their model classes. How can they understand that Propel misses this
opportunity?
There is a question of timing. Propel2 will hopefully be out in late 2012.
At this time, leveraging the PHP 5.3 features will be a minimum. Not
leveraging the PHP 5.4 features will probably be a mistake, since we can't
wait for a new major refactoring (in three years?). Other ORMs, other
libraries already make the jump, and we need Propel to be in the first
league.
I think Propel2 should require PHP5.4. What's your opinion?
I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about that. In
his opinion, Propel2 should use 5.4, it should be "the PHP 5.4 ORM".
Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest version. In
one hand this is awesome because everybody can use Propel. In another hand
it's a pain to maintain, and Propel2 is coming to ease our life as
developers.
I'd love to say "let's use 5.4" but as far as I know, it's not widespread
yet. But, there are already some discussions about 5.5 features, and I
think being stuck in 5.3 now will require yet another hard and long work to
leverage the 5.5 features.
Do you want, guys, bumping the minimum PHP version to 5.4 now?
I see two options:
- we use our current "refactoring timeframe" to break everything, before a
first stable version;
- we release a first unstable version soon (before the end of August), and
then we start a new iteration for Propel 2.1.
> I've always been reluctant to raise the version requirements for Propel.
> As of now, Propel2 requires PHP5.3.3, but I have good reasons to believe we
> should raise that to 5.4.
> Propel2 is only embracing PHP5.3 only now, and two of is main features
> (Late Static Binding and Namespaces) have major implications on the entire
> architecture. Using traits for the core (not optionally) would transform
> the architecture even more dramatically, and would pave the path for
> extremely powerful features and extensivity.
> Behaviors are the first to come to mind, but traits also allow true PoPo
> (Plain Old Php Objects) to become ActiveRecord objects without using
> inheritance. I'm not talking about abandoning code generation. Propel's use
> of code generation, both for speed and for IDE friendliness, is our
> signature. I'm talking about generated traits. Besides, there are numerous
> parts in Propel that could be refactored into traits, even be released as
> standalone libraries (think about the ConditionalProxy for ModelCriteria
> for instance). Lastly, traits will be used by the end users, for sure, to
> extend their model classes. How can they understand that Propel misses this
> opportunity?
> There is a question of timing. Propel2 will hopefully be out in late 2012.
> At this time, leveraging the PHP 5.3 features will be a minimum. Not
> leveraging the PHP 5.4 features will probably be a mistake, since we can't
> wait for a new major refactoring (in three years?). Other ORMs, other
> libraries already make the jump, and we need Propel to be in the first
> league.
> I think Propel2 should require PHP5.4. What's your opinion?
> Cheers,
> François
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
I think it's the right move on the right time to use php 5.4 for
propel2 : Luckily propel1 is very strong and thus propel2 can be
delayed to meet nice technical requirements.
I think the idea of an unstable version without php5.4 requirements
and then a 2.1 version with php 5.4 is not much usefull and will be
confusing and hard to maintain.
Bye
2012/7/30, William Durand <william.dura...@gmail.com>:
> I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about that. In
> his opinion, Propel2 should use 5.4, it should be "the PHP 5.4 ORM".
> Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest version. In
> one hand this is awesome because everybody can use Propel. In another hand
> it's a pain to maintain, and Propel2 is coming to ease our life as
> developers.
> I'd love to say "let's use 5.4" but as far as I know, it's not widespread
> yet. But, there are already some discussions about 5.5 features, and I
> think being stuck in 5.3 now will require yet another hard and long work to
> leverage the 5.5 features.
> Do you want, guys, bumping the minimum PHP version to 5.4 now?
> I see two options:
> - we use our current "refactoring timeframe" to break everything, before a
> first stable version;
> - we release a first unstable version soon (before the end of August), and
> then we start a new iteration for Propel 2.1.
>> I've always been reluctant to raise the version requirements for Propel.
>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to believe
>> we
>> should raise that to 5.4.
>> Propel2 is only embracing PHP5.3 only now, and two of is main features
>> (Late Static Binding and Namespaces) have major implications on the
>> entire
>> architecture. Using traits for the core (not optionally) would transform
>> the architecture even more dramatically, and would pave the path for
>> extremely powerful features and extensivity.
>> Behaviors are the first to come to mind, but traits also allow true PoPo
>> (Plain Old Php Objects) to become ActiveRecord objects without using
>> inheritance. I'm not talking about abandoning code generation. Propel's
>> use
>> of code generation, both for speed and for IDE friendliness, is our
>> signature. I'm talking about generated traits. Besides, there are
>> numerous
>> parts in Propel that could be refactored into traits, even be released as
>> standalone libraries (think about the ConditionalProxy for ModelCriteria
>> for instance). Lastly, traits will be used by the end users, for sure, to
>> extend their model classes. How can they understand that Propel misses
>> this
>> opportunity?
>> There is a question of timing. Propel2 will hopefully be out in late
>> 2012.
>> At this time, leveraging the PHP 5.3 features will be a minimum. Not
>> leveraging the PHP 5.4 features will probably be a mistake, since we
>> can't
>> wait for a new major refactoring (in three years?). Other ORMs, other
>> libraries already make the jump, and we need Propel to be in the first
>> league.
>> I think Propel2 should require PHP5.4. What's your opinion?
>> Cheers,
>> François
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Propel Development" group.
>> To post to this group, send email to propel-development@googlegroups.com
>> To unsubscribe from this group, send email to
>> propel-development+unsubscribe@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
> I think it's the right move on the right time to use php 5.4 for
> propel2 : Luckily propel1 is very strong and thus propel2 can be
> delayed to meet nice technical requirements.
> I think the idea of an unstable version without php5.4 requirements
> and then a 2.1 version with php 5.4 is not much usefull and will be
> confusing and hard to maintain.
> Bye
> 2012/7/30, William Durand <william.dura...@gmail.com>:
> > Heya,
> > I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about that.
> In
> > his opinion, Propel2 should use 5.4, it should be "the PHP 5.4 ORM".
> > Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest version.
> In
> > one hand this is awesome because everybody can use Propel. In another
> hand
> > it's a pain to maintain, and Propel2 is coming to ease our life as
> > developers.
> > I'd love to say "let's use 5.4" but as far as I know, it's not widespread
> > yet. But, there are already some discussions about 5.5 features, and I
> > think being stuck in 5.3 now will require yet another hard and long work
> to
> > leverage the 5.5 features.
> > Do you want, guys, bumping the minimum PHP version to 5.4 now?
> > I see two options:
> > - we use our current "refactoring timeframe" to break everything, before
> a
> > first stable version;
> > - we release a first unstable version soon (before the end of August),
> and
> > then we start a new iteration for Propel 2.1.
> >> I've always been reluctant to raise the version requirements for Propel.
> >> As of now, Propel2 requires PHP5.3.3, but I have good reasons to believe
> >> we
> >> should raise that to 5.4.
> >> Propel2 is only embracing PHP5.3 only now, and two of is main features
> >> (Late Static Binding and Namespaces) have major implications on the
> >> entire
> >> architecture. Using traits for the core (not optionally) would transform
> >> the architecture even more dramatically, and would pave the path for
> >> extremely powerful features and extensivity.
> >> Behaviors are the first to come to mind, but traits also allow true PoPo
> >> (Plain Old Php Objects) to become ActiveRecord objects without using
> >> inheritance. I'm not talking about abandoning code generation. Propel's
> >> use
> >> of code generation, both for speed and for IDE friendliness, is our
> >> signature. I'm talking about generated traits. Besides, there are
> >> numerous
> >> parts in Propel that could be refactored into traits, even be released
> as
> >> standalone libraries (think about the ConditionalProxy for ModelCriteria
> >> for instance). Lastly, traits will be used by the end users, for sure,
> to
> >> extend their model classes. How can they understand that Propel misses
> >> this
> >> opportunity?
> >> There is a question of timing. Propel2 will hopefully be out in late
> >> 2012.
> >> At this time, leveraging the PHP 5.3 features will be a minimum. Not
> >> leveraging the PHP 5.4 features will probably be a mistake, since we
> >> can't
> >> wait for a new major refactoring (in three years?). Other ORMs, other
> >> libraries already make the jump, and we need Propel to be in the first
> >> league.
> >> I think Propel2 should require PHP5.4. What's your opinion?
> >> Cheers,
> >> François
> >> --
> >> You received this message because you are subscribed to the Google
> >> Groups "Propel Development" group.
> >> To post to this group, send email to
> propel-development@googlegroups.com
> >> To unsubscribe from this group, send email to
> >> propel-development+unsubscribe@googlegroups.com
> >> For more options, visit this group at
> >> http://groups.google.com/group/propel-development
> > --
> > You received this message because you are subscribed to the Google
> > Groups "Propel Development" group.
> > To post to this group, send email to propel-development@googlegroups.com
> > To unsubscribe from this group, send email to
> > propel-development+unsubscribe@googlegroups.com
> > For more options, visit this group at
> > http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
PS : it's not directly linked, but an interested work could be to make
the units tests more easy to bootstrap, i find it actually hard to
make it work and this has prevented me for contributing to bugfixing.
> I think it's the right move on the right time to use php 5.4 for
> propel2 : Luckily propel1 is very strong and thus propel2 can be
> delayed to meet nice technical requirements.
> I think the idea of an unstable version without php5.4 requirements
> and then a 2.1 version with php 5.4 is not much usefull and will be
> confusing and hard to maintain.
> Bye
> 2012/7/30, William Durand <william.dura...@gmail.com>:
>> Heya,
>> I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about that.
>> In
>> his opinion, Propel2 should use 5.4, it should be "the PHP 5.4 ORM".
>> Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest version.
>> In
>> one hand this is awesome because everybody can use Propel. In another
>> hand
>> it's a pain to maintain, and Propel2 is coming to ease our life as
>> developers.
>> I'd love to say "let's use 5.4" but as far as I know, it's not widespread
>> yet. But, there are already some discussions about 5.5 features, and I
>> think being stuck in 5.3 now will require yet another hard and long work
>> to
>> leverage the 5.5 features.
>> Do you want, guys, bumping the minimum PHP version to 5.4 now?
>> I see two options:
>> - we use our current "refactoring timeframe" to break everything, before
>> a
>> first stable version;
>> - we release a first unstable version soon (before the end of August),
>> and
>> then we start a new iteration for Propel 2.1.
>>> I've always been reluctant to raise the version requirements for Propel.
>>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to believe
>>> we
>>> should raise that to 5.4.
>>> Propel2 is only embracing PHP5.3 only now, and two of is main features
>>> (Late Static Binding and Namespaces) have major implications on the
>>> entire
>>> architecture. Using traits for the core (not optionally) would transform
>>> the architecture even more dramatically, and would pave the path for
>>> extremely powerful features and extensivity.
>>> Behaviors are the first to come to mind, but traits also allow true PoPo
>>> (Plain Old Php Objects) to become ActiveRecord objects without using
>>> inheritance. I'm not talking about abandoning code generation. Propel's
>>> use
>>> of code generation, both for speed and for IDE friendliness, is our
>>> signature. I'm talking about generated traits. Besides, there are
>>> numerous
>>> parts in Propel that could be refactored into traits, even be released
>>> as
>>> standalone libraries (think about the ConditionalProxy for ModelCriteria
>>> for instance). Lastly, traits will be used by the end users, for sure,
>>> to
>>> extend their model classes. How can they understand that Propel misses
>>> this
>>> opportunity?
>>> There is a question of timing. Propel2 will hopefully be out in late
>>> 2012.
>>> At this time, leveraging the PHP 5.3 features will be a minimum. Not
>>> leveraging the PHP 5.4 features will probably be a mistake, since we
>>> can't
>>> wait for a new major refactoring (in three years?). Other ORMs, other
>>> libraries already make the jump, and we need Propel to be in the first
>>> league.
>>> I think Propel2 should require PHP5.4. What's your opinion?
>>> Cheers,
>>> François
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "Propel Development" group.
>>> To post to this group, send email to propel-development@googlegroups.com
>>> To unsubscribe from this group, send email to
>>> propel-development+unsubscribe@googlegroups.com
>>> For more options, visit this group at
>>> http://groups.google.com/group/propel-development
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Propel Development" group.
>> To post to this group, send email to propel-development@googlegroups.com
>> To unsubscribe from this group, send email to
>> propel-development+unsubscribe@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/propel-development
> PS : it's not directly linked, but an interested work could be to make
> the units tests more easy to bootstrap, i find it actually hard to
> make it work and this has prevented me for contributing to bugfixing.
> Bye
> 2012/7/30, Josselin Jacquard <josselin.jacqu...@gmail.com>:
> > I think it's the right move on the right time to use php 5.4 for
> > propel2 : Luckily propel1 is very strong and thus propel2 can be
> > delayed to meet nice technical requirements.
> > I think the idea of an unstable version without php5.4 requirements
> > and then a 2.1 version with php 5.4 is not much usefull and will be
> > confusing and hard to maintain.
> > Bye
> > 2012/7/30, William Durand <william.dura...@gmail.com>:
> >> Heya,
> >> I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about that.
> >> In
> >> his opinion, Propel2 should use 5.4, it should be "the PHP 5.4 ORM".
> >> Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest version.
> >> In
> >> one hand this is awesome because everybody can use Propel. In another
> >> hand
> >> it's a pain to maintain, and Propel2 is coming to ease our life as
> >> developers.
> >> I'd love to say "let's use 5.4" but as far as I know, it's not
> widespread
> >> yet. But, there are already some discussions about 5.5 features, and I
> >> think being stuck in 5.3 now will require yet another hard and long work
> >> to
> >> leverage the 5.5 features.
> >> Do you want, guys, bumping the minimum PHP version to 5.4 now?
> >> I see two options:
> >> - we use our current "refactoring timeframe" to break everything, before
> >> a
> >> first stable version;
> >> - we release a first unstable version soon (before the end of August),
> >> and
> >> then we start a new iteration for Propel 2.1.
> >>> I've always been reluctant to raise the version requirements for
> Propel.
> >>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to
> believe
> >>> we
> >>> should raise that to 5.4.
> >>> Propel2 is only embracing PHP5.3 only now, and two of is main features
> >>> (Late Static Binding and Namespaces) have major implications on the
> >>> entire
> >>> architecture. Using traits for the core (not optionally) would
> transform
> >>> the architecture even more dramatically, and would pave the path for
> >>> extremely powerful features and extensivity.
> >>> Behaviors are the first to come to mind, but traits also allow true
> PoPo
> >>> (Plain Old Php Objects) to become ActiveRecord objects without using
> >>> inheritance. I'm not talking about abandoning code generation. Propel's
> >>> use
> >>> of code generation, both for speed and for IDE friendliness, is our
> >>> signature. I'm talking about generated traits. Besides, there are
> >>> numerous
> >>> parts in Propel that could be refactored into traits, even be released
> >>> as
> >>> standalone libraries (think about the ConditionalProxy for
> ModelCriteria
> >>> for instance). Lastly, traits will be used by the end users, for sure,
> >>> to
> >>> extend their model classes. How can they understand that Propel misses
> >>> this
> >>> opportunity?
> >>> There is a question of timing. Propel2 will hopefully be out in late
> >>> 2012.
> >>> At this time, leveraging the PHP 5.3 features will be a minimum. Not
> >>> leveraging the PHP 5.4 features will probably be a mistake, since we
> >>> can't
> >>> wait for a new major refactoring (in three years?). Other ORMs, other
> >>> libraries already make the jump, and we need Propel to be in the first
> >>> league.
> >>> I think Propel2 should require PHP5.4. What's your opinion?
> >>> Cheers,
> >>> François
> >>> --
> >>> You received this message because you are subscribed to the Google
> >>> Groups "Propel Development" group.
> >>> To post to this group, send email to
> propel-development@googlegroups.com
> >>> To unsubscribe from this group, send email to
> >>> propel-development+unsubscribe@googlegroups.com
> >>> For more options, visit this group at
> >>> http://groups.google.com/group/propel-development
> >> --
> >> You received this message because you are subscribed to the Google
> >> Groups "Propel Development" group.
> >> To post to this group, send email to
> propel-development@googlegroups.com
> >> To unsubscribe from this group, send email to
> >> propel-development+unsubscribe@googlegroups.com
> >> For more options, visit this group at
> >> http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
Yes, it's easier in Propel2 than in Propel 1.6 (even if it's quite easy
since we have a shell script to run everything). The only issue is that we
need more than one database to run the complete test suite.
In Propel I reduced the number of databases, but it still needs one
database and three schemas... I think we should group tests whether they
rely on the database or not. But it's another discussion.
>> PS : it's not directly linked, but an interested work could be to make
>> the units tests more easy to bootstrap, i find it actually hard to
>> make it work and this has prevented me for contributing to bugfixing.
>> Bye
>> 2012/7/30, Josselin Jacquard <josselin.jacqu...@gmail.com>:
>> > I think it's the right move on the right time to use php 5.4 for
>> > propel2 : Luckily propel1 is very strong and thus propel2 can be
>> > delayed to meet nice technical requirements.
>> > I think the idea of an unstable version without php5.4 requirements
>> > and then a 2.1 version with php 5.4 is not much usefull and will be
>> > confusing and hard to maintain.
>> > Bye
>> > 2012/7/30, William Durand <william.dura...@gmail.com>:
>> >> Heya,
>> >> I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about that.
>> >> In
>> >> his opinion, Propel2 should use 5.4, it should be "the PHP 5.4 ORM".
>> >> Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest
>> version.
>> >> In
>> >> one hand this is awesome because everybody can use Propel. In another
>> >> hand
>> >> it's a pain to maintain, and Propel2 is coming to ease our life as
>> >> developers.
>> >> I'd love to say "let's use 5.4" but as far as I know, it's not
>> widespread
>> >> yet. But, there are already some discussions about 5.5 features, and I
>> >> think being stuck in 5.3 now will require yet another hard and long
>> work
>> >> to
>> >> leverage the 5.5 features.
>> >> Do you want, guys, bumping the minimum PHP version to 5.4 now?
>> >> I see two options:
>> >> - we use our current "refactoring timeframe" to break everything,
>> before
>> >> a
>> >> first stable version;
>> >> - we release a first unstable version soon (before the end of August),
>> >> and
>> >> then we start a new iteration for Propel 2.1.
>> >>> I've always been reluctant to raise the version requirements for
>> Propel.
>> >>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to
>> believe
>> >>> we
>> >>> should raise that to 5.4.
>> >>> Propel2 is only embracing PHP5.3 only now, and two of is main
>> features
>> >>> (Late Static Binding and Namespaces) have major implications on the
>> >>> entire
>> >>> architecture. Using traits for the core (not optionally) would
>> transform
>> >>> the architecture even more dramatically, and would pave the path for
>> >>> extremely powerful features and extensivity.
>> >>> Behaviors are the first to come to mind, but traits also allow true
>> PoPo
>> >>> (Plain Old Php Objects) to become ActiveRecord objects without using
>> >>> inheritance. I'm not talking about abandoning code generation.
>> Propel's
>> >>> use
>> >>> of code generation, both for speed and for IDE friendliness, is our
>> >>> signature. I'm talking about generated traits. Besides, there are
>> >>> numerous
>> >>> parts in Propel that could be refactored into traits, even be released
>> >>> as
>> >>> standalone libraries (think about the ConditionalProxy for
>> ModelCriteria
>> >>> for instance). Lastly, traits will be used by the end users, for sure,
>> >>> to
>> >>> extend their model classes. How can they understand that Propel misses
>> >>> this
>> >>> opportunity?
>> >>> There is a question of timing. Propel2 will hopefully be out in late
>> >>> 2012.
>> >>> At this time, leveraging the PHP 5.3 features will be a minimum. Not
>> >>> leveraging the PHP 5.4 features will probably be a mistake, since we
>> >>> can't
>> >>> wait for a new major refactoring (in three years?). Other ORMs, other
>> >>> libraries already make the jump, and we need Propel to be in the first
>> >>> league.
>> >>> I think Propel2 should require PHP5.4. What's your opinion?
>> >>> Cheers,
>> >>> François
>> >>> --
>> >>> You received this message because you are subscribed to the Google
>> >>> Groups "Propel Development" group.
>> >>> To post to this group, send email to
>> propel-development@googlegroups.com
>> >>> To unsubscribe from this group, send email to
>> >>> propel-development+unsubscribe@googlegroups.com
>> >>> For more options, visit this group at
>> >>> http://groups.google.com/group/propel-development
>> >> --
>> >> You received this message because you are subscribed to the Google
>> >> Groups "Propel Development" group.
>> >> To post to this group, send email to
>> propel-development@googlegroups.com
>> >> To unsubscribe from this group, send email to
>> >> propel-development+unsubscribe@googlegroups.com
>> >> For more options, visit this group at
>> >> http://groups.google.com/group/propel-development
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Propel Development" group.
>> To post to this group, send email to propel-development@googlegroups.com
>> To unsubscribe from this group, send email to
>> propel-development+unsubscribe@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
2012/7/30, William Durand <william.dura...@gmail.com>:
> Yes, it's easier in Propel2 than in Propel 1.6 (even if it's quite easy
> since we have a shell script to run everything). The only issue is that we
> need more than one database to run the complete test suite.
> In Propel I reduced the number of databases, but it still needs one
> database and three schemas... I think we should group tests whether they
> rely on the database or not. But it's another discussion.
>>> PS : it's not directly linked, but an interested work could be to make
>>> the units tests more easy to bootstrap, i find it actually hard to
>>> make it work and this has prevented me for contributing to bugfixing.
>>> Bye
>>> 2012/7/30, Josselin Jacquard <josselin.jacqu...@gmail.com>:
>>> > I think it's the right move on the right time to use php 5.4 for
>>> > propel2 : Luckily propel1 is very strong and thus propel2 can be
>>> > delayed to meet nice technical requirements.
>>> > I think the idea of an unstable version without php5.4 requirements
>>> > and then a 2.1 version with php 5.4 is not much usefull and will be
>>> > confusing and hard to maintain.
>>> > Bye
>>> > 2012/7/30, William Durand <william.dura...@gmail.com>:
>>> >> Heya,
>>> >> I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about
>>> >> that.
>>> >> In
>>> >> his opinion, Propel2 should use 5.4, it should be "the PHP 5.4 ORM".
>>> >> Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest
>>> version.
>>> >> In
>>> >> one hand this is awesome because everybody can use Propel. In another
>>> >> hand
>>> >> it's a pain to maintain, and Propel2 is coming to ease our life as
>>> >> developers.
>>> >> I'd love to say "let's use 5.4" but as far as I know, it's not
>>> widespread
>>> >> yet. But, there are already some discussions about 5.5 features, and
>>> >> I
>>> >> think being stuck in 5.3 now will require yet another hard and long
>>> work
>>> >> to
>>> >> leverage the 5.5 features.
>>> >> Do you want, guys, bumping the minimum PHP version to 5.4 now?
>>> >> I see two options:
>>> >> - we use our current "refactoring timeframe" to break everything,
>>> before
>>> >> a
>>> >> first stable version;
>>> >> - we release a first unstable version soon (before the end of
>>> >> August),
>>> >> and
>>> >> then we start a new iteration for Propel 2.1.
>>> >>> I've always been reluctant to raise the version requirements for
>>> Propel.
>>> >>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to
>>> believe
>>> >>> we
>>> >>> should raise that to 5.4.
>>> >>> Propel2 is only embracing PHP5.3 only now, and two of is main
>>> features
>>> >>> (Late Static Binding and Namespaces) have major implications on the
>>> >>> entire
>>> >>> architecture. Using traits for the core (not optionally) would
>>> transform
>>> >>> the architecture even more dramatically, and would pave the path for
>>> >>> extremely powerful features and extensivity.
>>> >>> Behaviors are the first to come to mind, but traits also allow true
>>> PoPo
>>> >>> (Plain Old Php Objects) to become ActiveRecord objects without using
>>> >>> inheritance. I'm not talking about abandoning code generation.
>>> Propel's
>>> >>> use
>>> >>> of code generation, both for speed and for IDE friendliness, is our
>>> >>> signature. I'm talking about generated traits. Besides, there are
>>> >>> numerous
>>> >>> parts in Propel that could be refactored into traits, even be
>>> >>> released
>>> >>> as
>>> >>> standalone libraries (think about the ConditionalProxy for
>>> ModelCriteria
>>> >>> for instance). Lastly, traits will be used by the end users, for
>>> >>> sure,
>>> >>> to
>>> >>> extend their model classes. How can they understand that Propel
>>> >>> misses
>>> >>> this
>>> >>> opportunity?
>>> >>> There is a question of timing. Propel2 will hopefully be out in late
>>> >>> 2012.
>>> >>> At this time, leveraging the PHP 5.3 features will be a minimum. Not
>>> >>> leveraging the PHP 5.4 features will probably be a mistake, since we
>>> >>> can't
>>> >>> wait for a new major refactoring (in three years?). Other ORMs,
>>> >>> other
>>> >>> libraries already make the jump, and we need Propel to be in the
>>> >>> first
>>> >>> league.
>>> >>> I think Propel2 should require PHP5.4. What's your opinion?
>>> >>> Cheers,
>>> >>> François
>>> >>> --
>>> >>> You received this message because you are subscribed to the Google
>>> >>> Groups "Propel Development" group.
>>> >>> To post to this group, send email to
>>> propel-development@googlegroups.com
>>> >>> To unsubscribe from this group, send email to
>>> >>> propel-development+unsubscribe@googlegroups.com
>>> >>> For more options, visit this group at
>>> >>> http://groups.google.com/group/propel-development
>>> >> --
>>> >> You received this message because you are subscribed to the Google
>>> >> Groups "Propel Development" group.
>>> >> To post to this group, send email to
>>> propel-development@googlegroups.com
>>> >> To unsubscribe from this group, send email to
>>> >> propel-development+unsubscribe@googlegroups.com
>>> >> For more options, visit this group at
>>> >> http://groups.google.com/group/propel-development
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "Propel Development" group.
>>> To post to this group, send email to propel-development@googlegroups.com
>>> To unsubscribe from this group, send email to
>>> propel-development+unsubscribe@googlegroups.com
>>> For more options, visit this group at
>>> http://groups.google.com/group/propel-development
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Propel Development" group.
>> To post to this group, send email to propel-development@googlegroups.com
>> To unsubscribe from this group, send email to
>> propel-development+unsubscribe@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
josselin.jacqu...@gmail.com> wrote:
> 2012/7/30, William Durand <william.dura...@gmail.com>:
> > Yes, it's easier in Propel2 than in Propel 1.6 (even if it's quite easy
> > since we have a shell script to run everything). The only issue is that
> we
> > need more than one database to run the complete test suite.
> > In Propel I reduced the number of databases, but it still needs one
> > database and three schemas... I think we should group tests whether they
> > rely on the database or not. But it's another discussion.
> >>> PS : it's not directly linked, but an interested work could be to make
> >>> the units tests more easy to bootstrap, i find it actually hard to
> >>> make it work and this has prevented me for contributing to bugfixing.
> >>> Bye
> >>> 2012/7/30, Josselin Jacquard <josselin.jacqu...@gmail.com>:
> >>> > I think it's the right move on the right time to use php 5.4 for
> >>> > propel2 : Luckily propel1 is very strong and thus propel2 can be
> >>> > delayed to meet nice technical requirements.
> >>> > I think the idea of an unstable version without php5.4 requirements
> >>> > and then a 2.1 version with php 5.4 is not much usefull and will be
> >>> > confusing and hard to maintain.
> >>> >> I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about
> >>> >> that.
> >>> >> In
> >>> >> his opinion, Propel2 should use 5.4, it should be "the PHP 5.4 ORM".
> >>> >> Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest
> >>> version.
> >>> >> In
> >>> >> one hand this is awesome because everybody can use Propel. In
> another
> >>> >> hand
> >>> >> it's a pain to maintain, and Propel2 is coming to ease our life as
> >>> >> developers.
> >>> >> I'd love to say "let's use 5.4" but as far as I know, it's not
> >>> widespread
> >>> >> yet. But, there are already some discussions about 5.5 features, and
> >>> >> I
> >>> >> think being stuck in 5.3 now will require yet another hard and long
> >>> work
> >>> >> to
> >>> >> leverage the 5.5 features.
> >>> >> Do you want, guys, bumping the minimum PHP version to 5.4 now?
> >>> >> I see two options:
> >>> >> - we use our current "refactoring timeframe" to break everything,
> >>> before
> >>> >> a
> >>> >> first stable version;
> >>> >> - we release a first unstable version soon (before the end of
> >>> >> August),
> >>> >> and
> >>> >> then we start a new iteration for Propel 2.1.
> >>> >>> I've always been reluctant to raise the version requirements for
> >>> Propel.
> >>> >>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to
> >>> believe
> >>> >>> we
> >>> >>> should raise that to 5.4.
> >>> >>> Propel2 is only embracing PHP5.3 only now, and two of is main
> >>> features
> >>> >>> (Late Static Binding and Namespaces) have major implications on the
> >>> >>> entire
> >>> >>> architecture. Using traits for the core (not optionally) would
> >>> transform
> >>> >>> the architecture even more dramatically, and would pave the path
> for
> >>> >>> extremely powerful features and extensivity.
> >>> >>> Behaviors are the first to come to mind, but traits also allow true
> >>> PoPo
> >>> >>> (Plain Old Php Objects) to become ActiveRecord objects without
> using
> >>> >>> inheritance. I'm not talking about abandoning code generation.
> >>> Propel's
> >>> >>> use
> >>> >>> of code generation, both for speed and for IDE friendliness, is our
> >>> >>> signature. I'm talking about generated traits. Besides, there are
> >>> >>> numerous
> >>> >>> parts in Propel that could be refactored into traits, even be
> >>> >>> released
> >>> >>> as
> >>> >>> standalone libraries (think about the ConditionalProxy for
> >>> ModelCriteria
> >>> >>> for instance). Lastly, traits will be used by the end users, for
> >>> >>> sure,
> >>> >>> to
> >>> >>> extend their model classes. How can they understand that Propel
> >>> >>> misses
> >>> >>> this
> >>> >>> opportunity?
> >>> >>> There is a question of timing. Propel2 will hopefully be out in
> late
> >>> >>> 2012.
> >>> >>> At this time, leveraging the PHP 5.3 features will be a minimum.
> Not
> >>> >>> leveraging the PHP 5.4 features will probably be a mistake, since
> we
> >>> >>> can't
> >>> >>> wait for a new major refactoring (in three years?). Other ORMs,
> >>> >>> other
> >>> >>> libraries already make the jump, and we need Propel to be in the
> >>> >>> first
> >>> >>> league.
> >>> >>> I think Propel2 should require PHP5.4. What's your opinion?
> >>> >>> Cheers,
> >>> >>> François
> >>> >>> --
> >>> >>> You received this message because you are subscribed to the Google
> >>> >>> Groups "Propel Development" group.
> >>> >>> To post to this group, send email to
> >>> propel-development@googlegroups.com
> >>> >>> To unsubscribe from this group, send email to
> >>> >>> propel-development+unsubscribe@googlegroups.com
> >>> >>> For more options, visit this group at
> >>> >>> http://groups.google.com/group/propel-development
> >>> >> --
> >>> >> You received this message because you are subscribed to the Google
> >>> >> Groups "Propel Development" group.
> >>> >> To post to this group, send email to
> >>> propel-development@googlegroups.com
> >>> >> To unsubscribe from this group, send email to
> >>> >> propel-development+unsubscribe@googlegroups.com
> >>> >> For more options, visit this group at
> >>> >> http://groups.google.com/group/propel-development
> >>> --
> >>> You received this message because you are subscribed to the Google
> >>> Groups "Propel Development" group.
> >>> To post to this group, send email to
> propel-development@googlegroups.com
> >>> To unsubscribe from this group, send email to
> >>> propel-development+unsubscribe@googlegroups.com
> >>> For more options, visit this group at
> >>> http://groups.google.com/group/propel-development
> >> --
> >> You received this message because you are subscribed to the Google
> >> Groups "Propel Development" group.
> >> To post to this group, send email to
> propel-development@googlegroups.com
> >> To unsubscribe from this group, send email to
> >> propel-development+unsubscribe@googlegroups.com
> >> For more options, visit this group at
> >> http://groups.google.com/group/propel-development
> > --
> > You received this message because you are subscribed to the Google
> > Groups "Propel Development" group.
> > To post to this group, send email to propel-development@googlegroups.com
> > To unsubscribe from this group, send email to
> > propel-development+unsubscribe@googlegroups.com
> > For more options, visit this group at
> > http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
> It would be nice to be able to make a trait out of the PropelCollection
> methods and that trait swapable at generation time via a custom trait.
> On Mon, Jul 30, 2012 at 10:51 AM, Josselin Jacquard <
> josselin.jacqu...@gmail.com> wrote:
>> 2012/7/30, William Durand <william.dura...@gmail.com>:
>> > Yes, it's easier in Propel2 than in Propel 1.6 (even if it's quite easy
>> > since we have a shell script to run everything). The only issue is that
>> we
>> > need more than one database to run the complete test suite.
>> > In Propel I reduced the number of databases, but it still needs one
>> > database and three schemas... I think we should group tests whether they
>> > rely on the database or not. But it's another discussion.
>> >>> PS : it's not directly linked, but an interested work could be to make
>> >>> the units tests more easy to bootstrap, i find it actually hard to
>> >>> make it work and this has prevented me for contributing to bugfixing.
>> >>> Bye
>> >>> 2012/7/30, Josselin Jacquard <josselin.jacqu...@gmail.com>:
>> >>> > I think it's the right move on the right time to use php 5.4 for
>> >>> > propel2 : Luckily propel1 is very strong and thus propel2 can be
>> >>> > delayed to meet nice technical requirements.
>> >>> > I think the idea of an unstable version without php5.4 requirements
>> >>> > and then a 2.1 version with php 5.4 is not much usefull and will be
>> >>> > confusing and hard to maintain.
>> >>> >> I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about
>> >>> >> that.
>> >>> >> In
>> >>> >> his opinion, Propel2 should use 5.4, it should be "the PHP 5.4
>> ORM".
>> >>> >> Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest
>> >>> version.
>> >>> >> In
>> >>> >> one hand this is awesome because everybody can use Propel. In
>> another
>> >>> >> hand
>> >>> >> it's a pain to maintain, and Propel2 is coming to ease our life as
>> >>> >> developers.
>> >>> >> I'd love to say "let's use 5.4" but as far as I know, it's not
>> >>> widespread
>> >>> >> yet. But, there are already some discussions about 5.5 features,
>> and
>> >>> >> I
>> >>> >> think being stuck in 5.3 now will require yet another hard and long
>> >>> work
>> >>> >> to
>> >>> >> leverage the 5.5 features.
>> >>> >> Do you want, guys, bumping the minimum PHP version to 5.4 now?
>> >>> >> I see two options:
>> >>> >> - we use our current "refactoring timeframe" to break everything,
>> >>> before
>> >>> >> a
>> >>> >> first stable version;
>> >>> >> - we release a first unstable version soon (before the end of
>> >>> >> August),
>> >>> >> and
>> >>> >> then we start a new iteration for Propel 2.1.
>> >>> >>> I've always been reluctant to raise the version requirements for
>> >>> Propel.
>> >>> >>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to
>> >>> believe
>> >>> >>> we
>> >>> >>> should raise that to 5.4.
>> >>> >>> Propel2 is only embracing PHP5.3 only now, and two of is main
>> >>> features
>> >>> >>> (Late Static Binding and Namespaces) have major implications on
>> the
>> >>> >>> entire
>> >>> >>> architecture. Using traits for the core (not optionally) would
>> >>> transform
>> >>> >>> the architecture even more dramatically, and would pave the path
>> for
>> >>> >>> extremely powerful features and extensivity.
>> >>> >>> Behaviors are the first to come to mind, but traits also allow
>> true
>> >>> PoPo
>> >>> >>> (Plain Old Php Objects) to become ActiveRecord objects without
>> using
>> >>> >>> inheritance. I'm not talking about abandoning code generation.
>> >>> Propel's
>> >>> >>> use
>> >>> >>> of code generation, both for speed and for IDE friendliness, is
>> our
>> >>> >>> signature. I'm talking about generated traits. Besides, there are
>> >>> >>> numerous
>> >>> >>> parts in Propel that could be refactored into traits, even be
>> >>> >>> released
>> >>> >>> as
>> >>> >>> standalone libraries (think about the ConditionalProxy for
>> >>> ModelCriteria
>> >>> >>> for instance). Lastly, traits will be used by the end users, for
>> >>> >>> sure,
>> >>> >>> to
>> >>> >>> extend their model classes. How can they understand that Propel
>> >>> >>> misses
>> >>> >>> this
>> >>> >>> opportunity?
>> >>> >>> There is a question of timing. Propel2 will hopefully be out in
>> late
>> >>> >>> 2012.
>> >>> >>> At this time, leveraging the PHP 5.3 features will be a minimum.
>> Not
>> >>> >>> leveraging the PHP 5.4 features will probably be a mistake, since
>> we
>> >>> >>> can't
>> >>> >>> wait for a new major refactoring (in three years?). Other ORMs,
>> >>> >>> other
>> >>> >>> libraries already make the jump, and we need Propel to be in the
>> >>> >>> first
>> >>> >>> league.
>> >>> >>> I think Propel2 should require PHP5.4. What's your opinion?
>> >>> >>> Cheers,
>> >>> >>> François
>> >>> >>> --
>> >>> >>> You received this message because you are subscribed to the Google
>> >>> >>> Groups "Propel Development" group.
>> >>> >>> To post to this group, send email to
>> >>> propel-development@googlegroups.com
>> >>> >>> To unsubscribe from this group, send email to
>> >>> >>> propel-development+unsubscribe@googlegroups.com
>> >>> >>> For more options, visit this group at
>> >>> >>> http://groups.google.com/group/propel-development
>> >>> >> --
>> >>> >> You received this message because you are subscribed to the Google
>> >>> >> Groups "Propel Development" group.
>> >>> >> To post to this group, send email to
>> >>> propel-development@googlegroups.com
>> >>> >> To unsubscribe from this group, send email to
>> >>> >> propel-development+unsubscribe@googlegroups.com
>> >>> >> For more options, visit this group at
>> >>> >> http://groups.google.com/group/propel-development
>> >>> --
>> >>> You received this message because you are subscribed to the Google
>> >>> Groups "Propel Development" group.
>> >>> To post to this group, send email to
>> propel-development@googlegroups.com
>> >>> To unsubscribe from this group, send email to
>> >>> propel-development+unsubscribe@googlegroups.com
>> >>> For more options, visit this group at
>> >>> http://groups.google.com/group/propel-development
>> >> --
>> >> You received this message because you are subscribed to the Google
>> >> Groups "Propel Development" group.
>> >> To post to this group, send email to
>> propel-development@googlegroups.com
>> >> To unsubscribe from this group, send email to
>> >> propel-development+unsubscribe@googlegroups.com
>> >> For more options, visit this group at
>> >> http://groups.google.com/group/propel-development
>> > --
>> > You received this message because you are subscribed to the Google
>> > Groups "Propel Development" group.
>> > To post to this group, send email to
>> propel-development@googlegroups.com
>> > To unsubscribe from this group, send email to
>> > propel-development+unsubscribe@googlegroups.com
>> > For more options, visit this group at
>> > http://groups.google.com/group/propel-development
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Propel Development" group.
>> To post to this group, send email to propel-development@googlegroups.com
>> To unsubscribe from this group, send email to
>> propel-development+unsubscribe@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
>> It would be nice to be able to make a trait out of the PropelCollection
>> methods and that trait swapable at generation time via a custom trait.
>> On Mon, Jul 30, 2012 at 10:51 AM, Josselin Jacquard <
>> josselin.jacqu...@gmail.com> wrote:
>>> 2012/7/30, William Durand <william.dura...@gmail.com>:
>>> > Yes, it's easier in Propel2 than in Propel 1.6 (even if it's quite easy
>>> > since we have a shell script to run everything). The only issue is
>>> that we
>>> > need more than one database to run the complete test suite.
>>> > In Propel I reduced the number of databases, but it still needs one
>>> > database and three schemas... I think we should group tests whether
>>> they
>>> > rely on the database or not. But it's another discussion.
>>> >>> PS : it's not directly linked, but an interested work could be to
>>> make
>>> >>> the units tests more easy to bootstrap, i find it actually hard to
>>> >>> make it work and this has prevented me for contributing to bugfixing.
>>> >>> Bye
>>> >>> 2012/7/30, Josselin Jacquard <josselin.jacqu...@gmail.com>:
>>> >>> > I think it's the right move on the right time to use php 5.4 for
>>> >>> > propel2 : Luckily propel1 is very strong and thus propel2 can be
>>> >>> > delayed to meet nice technical requirements.
>>> >>> > I think the idea of an unstable version without php5.4 requirements
>>> >>> > and then a 2.1 version with php 5.4 is not much usefull and will be
>>> >>> > confusing and hard to maintain.
>>> >>> >> I talked a lot with Jordi Boggiano (Composer, Monolog, ...) about
>>> >>> >> that.
>>> >>> >> In
>>> >>> >> his opinion, Propel2 should use 5.4, it should be "the PHP 5.4
>>> ORM".
>>> >>> >> Today, Propel 1.6 is able to scale from PHP 5.2.4 to 5.4 latest
>>> >>> version.
>>> >>> >> In
>>> >>> >> one hand this is awesome because everybody can use Propel. In
>>> another
>>> >>> >> hand
>>> >>> >> it's a pain to maintain, and Propel2 is coming to ease our life as
>>> >>> >> developers.
>>> >>> >> I'd love to say "let's use 5.4" but as far as I know, it's not
>>> >>> widespread
>>> >>> >> yet. But, there are already some discussions about 5.5 features,
>>> and
>>> >>> >> I
>>> >>> >> think being stuck in 5.3 now will require yet another hard and
>>> long
>>> >>> work
>>> >>> >> to
>>> >>> >> leverage the 5.5 features.
>>> >>> >> Do you want, guys, bumping the minimum PHP version to 5.4 now?
>>> >>> >> I see two options:
>>> >>> >> - we use our current "refactoring timeframe" to break everything,
>>> >>> before
>>> >>> >> a
>>> >>> >> first stable version;
>>> >>> >> - we release a first unstable version soon (before the end of
>>> >>> >> August),
>>> >>> >> and
>>> >>> >> then we start a new iteration for Propel 2.1.
>>> >>> >>> I've always been reluctant to raise the version requirements for
>>> >>> Propel.
>>> >>> >>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to
>>> >>> believe
>>> >>> >>> we
>>> >>> >>> should raise that to 5.4.
>>> >>> >>> Propel2 is only embracing PHP5.3 only now, and two of is main
>>> >>> features
>>> >>> >>> (Late Static Binding and Namespaces) have major implications on
>>> the
>>> >>> >>> entire
>>> >>> >>> architecture. Using traits for the core (not optionally) would
>>> >>> transform
>>> >>> >>> the architecture even more dramatically, and would pave the path
>>> for
>>> >>> >>> extremely powerful features and extensivity.
>>> >>> >>> Behaviors are the first to come to mind, but traits also allow
>>> true
>>> >>> PoPo
>>> >>> >>> (Plain Old Php Objects) to become ActiveRecord objects without
>>> using
>>> >>> >>> inheritance. I'm not talking about abandoning code generation.
>>> >>> Propel's
>>> >>> >>> use
>>> >>> >>> of code generation, both for speed and for IDE friendliness, is
>>> our
>>> >>> >>> signature. I'm talking about generated traits. Besides, there are
>>> >>> >>> numerous
>>> >>> >>> parts in Propel that could be refactored into traits, even be
>>> >>> >>> released
>>> >>> >>> as
>>> >>> >>> standalone libraries (think about the ConditionalProxy for
>>> >>> ModelCriteria
>>> >>> >>> for instance). Lastly, traits will be used by the end users, for
>>> >>> >>> sure,
>>> >>> >>> to
>>> >>> >>> extend their model classes. How can they understand that Propel
>>> >>> >>> misses
>>> >>> >>> this
>>> >>> >>> opportunity?
>>> >>> >>> There is a question of timing. Propel2 will hopefully be out in
>>> late
>>> >>> >>> 2012.
>>> >>> >>> At this time, leveraging the PHP 5.3 features will be a minimum.
>>> Not
>>> >>> >>> leveraging the PHP 5.4 features will probably be a mistake,
>>> since we
>>> >>> >>> can't
>>> >>> >>> wait for a new major refactoring (in three years?). Other ORMs,
>>> >>> >>> other
>>> >>> >>> libraries already make the jump, and we need Propel to be in the
>>> >>> >>> first
>>> >>> >>> league.
>>> >>> >>> I think Propel2 should require PHP5.4. What's your opinion?
>>> >>> >>> Cheers,
>>> >>> >>> François
>>> >>> >>> --
>>> >>> >>> You received this message because you are subscribed to the
>>> Google
>>> >>> >>> Groups "Propel Development" group.
>>> >>> >>> To post to this group, send email to
>>> >>> propel-development@googlegroups.com
>>> >>> >>> To unsubscribe from this group, send email to
>>> >>> >>> propel-development+unsubscribe@googlegroups.com
>>> >>> >>> For more options, visit this group at
>>> >>> >>> http://groups.google.com/group/propel-development
>>> >>> >> --
>>> >>> >> You received this message because you are subscribed to the Google
>>> >>> >> Groups "Propel Development" group.
>>> >>> >> To post to this group, send email to
>>> >>> propel-development@googlegroups.com
>>> >>> >> To unsubscribe from this group, send email to
>>> >>> >> propel-development+unsubscribe@googlegroups.com
>>> >>> >> For more options, visit this group at
>>> >>> >> http://groups.google.com/group/propel-development
>>> >>> --
>>> >>> You received this message because you are subscribed to the Google
>>> >>> Groups "Propel Development" group.
>>> >>> To post to this group, send email to
>>> propel-development@googlegroups.com
>>> >>> To unsubscribe from this group, send email to
>>> >>> propel-development+unsubscribe@googlegroups.com
>>> >>> For more options, visit this group at
>>> >>> http://groups.google.com/group/propel-development
>>> >> --
>>> >> You received this message because you are subscribed to the Google
>>> >> Groups "Propel Development" group.
>>> >> To post to this group, send email to
>>> propel-development@googlegroups.com
>>> >> To unsubscribe from this group, send email to
>>> >> propel-development+unsubscribe@googlegroups.com
>>> >> For more options, visit this group at
>>> >> http://groups.google.com/group/propel-development
>>> > --
>>> > You received this message because you are subscribed to the Google
>>> > Groups "Propel Development" group.
>>> > To post to this group, send email to
>>> propel-development@googlegroups.com
>>> > To unsubscribe from this group, send email to
>>> > propel-development+unsubscribe@googlegroups.com
>>> > For more options, visit this group at
>>> > http://groups.google.com/group/propel-development
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "Propel Development" group.
>>> To post to this group, send email to propel-development@googlegroups.com
>>> To unsubscribe from this group, send email to
>>> propel-development+unsubscribe@googlegroups.com
>>> For more options, visit this group at
>>> http://groups.google.com/group/propel-development
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Propel Development" group.
>> To post to this group, send email to propel-development@googlegroups.com
>> To unsubscribe from this group, send email to
>> propel-development+unsubscribe@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
I'm expecting this from Propel 2. Next gen ORM should move right now, Trait are really relevant for this kind of tool. Hope you guys will use all the goodness of 5.4
On Sunday, July 29, 2012 6:49:27 PM UTC+2, Francois Zaninotto wrote:
> Hi list,
> I've always been reluctant to raise the version requirements for Propel. > As of now, Propel2 requires PHP5.3.3, but I have good reasons to believe we > should raise that to 5.4.
> Propel2 is only embracing PHP5.3 only now, and two of is main features > (Late Static Binding and Namespaces) have major implications on the entire > architecture. Using traits for the core (not optionally) would transform > the architecture even more dramatically, and would pave the path for > extremely powerful features and extensivity.
> Behaviors are the first to come to mind, but traits also allow true PoPo > (Plain Old Php Objects) to become ActiveRecord objects without using > inheritance. I'm not talking about abandoning code generation. Propel's use > of code generation, both for speed and for IDE friendliness, is our > signature. I'm talking about generated traits. Besides, there are numerous > parts in Propel that could be refactored into traits, even be released as > standalone libraries (think about the ConditionalProxy for ModelCriteria > for instance). Lastly, traits will be used by the end users, for sure, to > extend their model classes. How can they understand that Propel misses this > opportunity?
> There is a question of timing. Propel2 will hopefully be out in late 2012. > At this time, leveraging the PHP 5.3 features will be a minimum. Not > leveraging the PHP 5.4 features will probably be a mistake, since we can't > wait for a new major refactoring (in three years?). Other ORMs, other > libraries already make the jump, and we need Propel to be in the first > league.
> I think Propel2 should require PHP5.4. What's your opinion?
I'd like to say, yes, raise the requirement to 5.4, but unfortunately, if that was done I doubt I'd be able to use Propel 2 since there is currently no (or little) adoption by the vendors (from what I can find). Thus, I'd never be able to use Propel2. On the other hand, if you think that Propel2 using 5.4 can advocate a bigger push to move for faster and more frequent PHP releases and upgrades (by the vendors), that would be amazing. Otherwise, I suspect there could never be a large adoption of Propel2 for a while.
On Monday, 30 July 2012 02:49:27 UTC+10, Francois Zaninotto wrote:
> Hi list,
> I've always been reluctant to raise the version requirements for Propel. > As of now, Propel2 requires PHP5.3.3, but I have good reasons to believe we > should raise that to 5.4.
> Propel2 is only embracing PHP5.3 only now, and two of is main features > (Late Static Binding and Namespaces) have major implications on the entire > architecture. Using traits for the core (not optionally) would transform > the architecture even more dramatically, and would pave the path for > extremely powerful features and extensivity.
> Behaviors are the first to come to mind, but traits also allow true PoPo > (Plain Old Php Objects) to become ActiveRecord objects without using > inheritance. I'm not talking about abandoning code generation. Propel's use > of code generation, both for speed and for IDE friendliness, is our > signature. I'm talking about generated traits. Besides, there are numerous > parts in Propel that could be refactored into traits, even be released as > standalone libraries (think about the ConditionalProxy for ModelCriteria > for instance). Lastly, traits will be used by the end users, for sure, to > extend their model classes. How can they understand that Propel misses this > opportunity?
> There is a question of timing. Propel2 will hopefully be out in late 2012. > At this time, leveraging the PHP 5.3 features will be a minimum. Not > leveraging the PHP 5.4 features will probably be a mistake, since we can't > wait for a new major refactoring (in three years?). Other ORMs, other > libraries already make the jump, and we need Propel to be in the first > league.
> I think Propel2 should require PHP5.4. What's your opinion?
> I'd like to say, yes, raise the requirement to 5.4, but unfortunately, if
> that was done I doubt I'd be able to use Propel 2 since there is currently
> no (or little) adoption by the vendors (from what I can find). Thus, I'd
> never be able to use Propel2. On the other hand, if you think that Propel2
> using 5.4 can advocate a bigger push to move for faster and more frequent
> PHP releases and upgrades (by the vendors), that would be amazing.
> Otherwise, I suspect there could never be a large adoption of Propel2 for a
> while.
> On Monday, 30 July 2012 02:49:27 UTC+10, Francois Zaninotto wrote:
>> Hi list,
>> I've always been reluctant to raise the version requirements for Propel.
>> As of now, Propel2 requires PHP5.3.3, but I have good reasons to believe we
>> should raise that to 5.4.
>> Propel2 is only embracing PHP5.3 only now, and two of is main features
>> (Late Static Binding and Namespaces) have major implications on the entire
>> architecture. Using traits for the core (not optionally) would transform
>> the architecture even more dramatically, and would pave the path for
>> extremely powerful features and extensivity.
>> Behaviors are the first to come to mind, but traits also allow true PoPo
>> (Plain Old Php Objects) to become ActiveRecord objects without using
>> inheritance. I'm not talking about abandoning code generation. Propel's use
>> of code generation, both for speed and for IDE friendliness, is our
>> signature. I'm talking about generated traits. Besides, there are numerous
>> parts in Propel that could be refactored into traits, even be released as
>> standalone libraries (think about the ConditionalProxy for ModelCriteria
>> for instance). Lastly, traits will be used by the end users, for sure, to
>> extend their model classes. How can they understand that Propel misses this
>> opportunity?
>> There is a question of timing. Propel2 will hopefully be out in late
>> 2012. At this time, leveraging the PHP 5.3 features will be a minimum. Not
>> leveraging the PHP 5.4 features will probably be a mistake, since we can't
>> wait for a new major refactoring (in three years?). Other ORMs, other
>> libraries already make the jump, and we need Propel to be in the first
>> league.
>> I think Propel2 should require PHP5.4. What's your opinion?
>> Cheers,
>> François
>> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
Same though I had when talking about version requirements: "Cheap hosters"
(thinking about 1&1 or Strato in Germany) did not even upgrade to 5.3 yet
on their standard entry packages.
One can argue that those small packages are not meant to be used for real
applications where Propel should be used, but it happens. (same issues I
had with silex, e.g., as it requires 5.3 and really IS a good fit for
creating small websites)
None the less, I really think a "future oriented project" like Propel2 is
allowed to go that step since, as already said
- it has a solid predecessor with 1.6.x
- it can benefit from the new options available.
I would strongly object to branch off a 2.0 => 5.3 and have 2.1 => 5.4, to
much hassle (docs, BC, parallel maintenance, ...)
+1 for 5.4, knowing that it might mean fallbacks to 1.6 are required in
projects with limitations due to server hosting constraints
+1 for 5.4. About the roadmap, the question is: how long will it take
to implement the new 5.4-only features? If we estimate that it'll take
3 - 5 months, I think it's better to postpone Propel 2.0 alpha, to
release it with all new features. If we think it'll take more than 5
months, I'm agree with William: 2.0 with 5.3 features, then 2.1 with
5.4 ones.
IMHO 2.0 should use 5.4 from the very beginning. The architectural and performance differences between PHP 5.3 and 5.4 are huge, and suggesting (by version transition from 2.0 to 2.1) a minor update can be harmful for those who don't read the changelogs. As Josselin pointed out (BTW, +1 on that one Josselin!):
> Luckily propel1 is very strong and thus propel2 can be delayed to meet nice technical requirements.
I'm not a part of the propel-dev team, but as a user I would wait those few months more just to get a library that's getting 100% of architectural and performance capabilities given by the language itself. In fact I think that if Propel2 would be based on PHP 5.4 it would encourage upgrading server software to reach the community requirements. And still be more ahead of Doctrine ;)
> +1 for 5.4. About the roadmap, the question is: how long will it take
> to implement the new 5.4-only features? If we estimate that it'll take
> 3 - 5 months, I think it's better to postpone Propel 2.0 alpha, to
> release it with all new features. If we think it'll take more than 5
> months, I'm agree with William: 2.0 with 5.3 features, then 2.1 with
> 5.4 ones.
+1 for 5.4
I'm using Propel, don't have to much time to help to develop this
great project, but i think that 5.4 should be the requirement. This
version is from 1 march, so it is almost 5 months. I don't think that
something will change at hosting servers in next half year. But back
to the Propel. If You want to use traits then use them from the
beginning, You will save time and then only optimize. If someone don't
have PHP5.4 then he has Propel1.6
I'm talking about a RHEL-based distro from AWS EC2 which is the default distro I believe for new EC2 instances (which is what we use in our company). I've found a few articles examining how 5.4 cod be setup on these, but my point is that it's not as simple as I'd like it to be. FYI I am also not a Propel dev. I am a Doctrine user and I keep in touch with Propel2 deb because I see it potentially replacing Doctrine (for my work).
For those looking for affordable hosting, Dreamhost always offered shell access and the ability to compile your own PHP there. I'm quite happy with them.
> For those looking for affordable hosting, Dreamhost always offered shell
> access and the ability to compile your own PHP there. I'm quite happy with
> them.
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
>> For those looking for affordable hosting, Dreamhost always offered shell
>> access and the ability to compile your own PHP there. I'm quite happy with
>> them.
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Propel Development" group.
>> To post to this group, send email to propel-development@googlegroups.com
>> To unsubscribe from this group, send email to
>> propel-development+unsubscribe@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/propel-development
> --
> You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
> I've always been reluctant to raise the version requirements for > Propel. As of now, Propel2 requires PHP5.3.3, but I have good reasons > to believe we should raise that to 5.4.
> Propel2 is only embracing PHP5.3 only now, and two of is main features > (Late Static Binding and Namespaces) have major implications on the > entire architecture. Using traits for the core (not optionally) would > transform the architecture even more dramatically, and would pave the > path for extremely powerful features and extensivity.
> Behaviors are the first to come to mind, but traits also allow true > PoPo (Plain Old Php Objects) to become ActiveRecord objects without > using inheritance. I'm not talking about abandoning code generation. > Propel's use of code generation, both for speed and for IDE > friendliness, is our signature. I'm talking about generated traits. > Besides, there are numerous parts in Propel that could be refactored > into traits, even be released as standalone libraries (think about the > ConditionalProxy for ModelCriteria for instance). Lastly, traits will > be used by the end users, for sure, to extend their model classes. How > can they understand that Propel misses this opportunity?
> There is a question of timing. Propel2 will hopefully be out in late > 2012. At this time, leveraging the PHP 5.3 features will be a minimum. > Not leveraging the PHP 5.4 features will probably be a mistake, since > we can't wait for a new major refactoring (in three years?). Other > ORMs, other libraries already make the jump, and we need Propel to be > in the first league.
> I think Propel2 should require PHP5.4. What's your opinion?
> Cheers,
> Fran ois
> -- > You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
I don't think a project like Propel has to wait for wide adoption of a version of PHP by hosters : most of us expect Propel2 to be a stable and smart project for years and the use of last PHP features are really what will make it worth to use.
Morevoer I do believe that PHP new versions adoption by hosters in the future is also related to our needs, we have our (small) word to say about it. Finally, I imagine that many of us use Propel on dedicated servers, isn't it ?
Lo c
> For those looking for affordable hosting, Dreamhost always offered > shell access and the ability to compile your own PHP there. I'm quite > happy with them.
> -- > You received this message because you are subscribed to the Google
> Groups "Propel Development" group.
> To post to this group, send email to propel-development@googlegroups.com
> To unsubscribe from this group, send email to
> propel-development+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/propel-development
