GRCodecStream>>isStream ??

> 2010/10/5 Norbert Hartl <[hidden email]>:
>> I would like to have implemented
>>
>> GRCodecStream>>isStream
>>        ^ true
>>
>> isStream is implemented at least in Object to return false. We some trouble using grease streams with the xml parser.
>
> I also thought about this one once. The trouble is that we don't
> implement the full stream protocol so you might end up wrapping them
> again. OTOH it can't be more broken than it currently is.

>
> On 06.10.2010, at 08:56, Philippe Marschall wrote:
>
>> 2010/10/5 Norbert Hartl <[hidden email]>:
>>> I would like to have implemented
>>>
>>> GRCodecStream>>isStream
>>>        ^ true
>>>
>>> isStream is implemented at least in Object to return false. We some trouble using grease streams with the xml parser.
>>
>> I also thought about this one once. The trouble is that we don't
>> implement the full stream protocol so you might end up wrapping them
>> again. OTOH it can't be more broken than it currently is.
>
> I think there is difference between incomplete and broken ;) I mean if you start implementing a class that has stream in its name and that implements the same methods it only makes sense if most of the protocol is there, right? Or to say it in other words. If this is the case there will always be people like me asking why this and that isn't available.
> I asked about Stream>>tab a while ago. Last time I just changed the code not to use Stream>>tab instead of putting tab into grease stream. The question about isStream is more easy to tackle. The question is what would be the replacement/workaround for isStream? I don't see a good one.

>
> On 06.10.2010, at 10:18, Philippe Marschall wrote:
>
> 2010/10/6 Norbert Hartl <[hidden email]>:
>
> On 06.10.2010, at 08:56, Philippe Marschall wrote:
>
> 2010/10/5 Norbert Hartl <[hidden email]>:
>
> I would like to have implemented
>
> GRCodecStream>>isStream
>
>        ^ true
>
> isStream is implemented at least in Object to return false. We some trouble
> using grease streams with the xml parser.
>
> I also thought about this one once. The trouble is that we don't
>
> implement the full stream protocol so you might end up wrapping them
>
> again. OTOH it can't be more broken than it currently is.
>
> I think there is difference between incomplete and broken ;) I mean if you
> start implementing a class that has stream in its name and that implements
> the same methods it only makes sense if most of the protocol is there,
> right? Or to say it in other words. If this is the case there will always be
> people like me asking why this and that isn't available.
>
> I asked about Stream>>tab a while ago. Last time I just changed the code not
> to use Stream>>tab instead of putting tab into grease stream. The question
> about isStream is more easy to tackle. The question is what would be the
> replacement/workaround for isStream? I don't see a good one.
>
> Implement a wrapper around GRCodecStream that has the full stream
> protocol. But I guess at one point or another we'll need to add more
> of the stream protocol anyway so we can as well start now.
>
> I wouldn't implement a wrapper around GRCodecStream I would replace it for
> my purpose. And that is probably not one of the goals of grease. The only
> thing  I need to do is to read a XML file in a platform independent manner.
> This doesn't work with grease and that is a pitty. I don't even recognize
> the need for discussing this thoroughly. I'm just doing it because I don't
> have commit rights in the repo.
> So it would be nice to hear something definite from the core developers. If
> you are willing to add this to grease that would be fine.

> 2010/10/6 Norbert Hartl <[hidden email]>:
>>
>> On 06.10.2010, at 10:18, Philippe Marschall wrote:
>>
>> 2010/10/6 Norbert Hartl <[hidden email]>:
>>
>> On 06.10.2010, at 08:56, Philippe Marschall wrote:
>>
>> 2010/10/5 Norbert Hartl <[hidden email]>:
>>
>> I would like to have implemented
>>
>> GRCodecStream>>isStream
>>
>>        ^ true
>>
>> isStream is implemented at least in Object to return false. We some trouble
>> using grease streams with the xml parser.
>>
>> I also thought about this one once. The trouble is that we don't
>>
>> implement the full stream protocol so you might end up wrapping them
>>
>> again. OTOH it can't be more broken than it currently is.
>>
>> I think there is difference between incomplete and broken ;) I mean if you
>> start implementing a class that has stream in its name and that implements
>> the same methods it only makes sense if most of the protocol is there,
>> right? Or to say it in other words. If this is the case there will always be
>> people like me asking why this and that isn't available.
>>
>> I asked about Stream>>tab a while ago. Last time I just changed the code not
>> to use Stream>>tab instead of putting tab into grease stream. The question
>> about isStream is more easy to tackle. The question is what would be the
>> replacement/workaround for isStream? I don't see a good one.
>>
>> Implement a wrapper around GRCodecStream that has the full stream
>> protocol. But I guess at one point or another we'll need to add more
>> of the stream protocol anyway so we can as well start now.
>>
>> I wouldn't implement a wrapper around GRCodecStream I would replace it for
>> my purpose. And that is probably not one of the goals of grease. The only
>> thing  I need to do is to read a XML file in a platform independent manner.
>> This doesn't work with grease and that is a pitty. I don't even recognize
>> the need for discussing this thoroughly. I'm just doing it because I don't
>> have commit rights in the repo.
>> So it would be nice to hear something definite from the core developers. If
>> you are willing to add this to grease that would be fine.
>
> Would be fine with me.

Ok,here is the changeset,

>
> On 06.10.2010, at 23:53, Julian Fitzell wrote:
>
>> On Wed, Oct 6, 2010 at 2:49 PM, Philippe Marschall
>> <[hidden email]> wrote:
>>> 2010/10/6 Norbert Hartl <[hidden email]>:
>>>>
>>>> On 06.10.2010, at 10:18, Philippe Marschall wrote:
>>>>
>>>> 2010/10/6 Norbert Hartl <[hidden email]>:
>>>>
>>>> On 06.10.2010, at 08:56, Philippe Marschall wrote:
>>>>
>>>> 2010/10/5 Norbert Hartl <[hidden email]>:
>>>>
>>>> I would like to have implemented
>>>>
>>>> GRCodecStream>>isStream
>>>>
>>>>        ^ true
>>>>
>>>> isStream is implemented at least in Object to return false. We some trouble
>>>> using grease streams with the xml parser.
>>>>
>>>> I also thought about this one once. The trouble is that we don't
>>>>
>>>> implement the full stream protocol so you might end up wrapping them
>>>>
>>>> again. OTOH it can't be more broken than it currently is.
>>>>
>>>> I think there is difference between incomplete and broken ;) I mean if you
>>>> start implementing a class that has stream in its name and that implements
>>>> the same methods it only makes sense if most of the protocol is there,
>>>> right? Or to say it in other words. If this is the case there will always be
>>>> people like me asking why this and that isn't available.
>>>>
>>>> I asked about Stream>>tab a while ago. Last time I just changed the code not
>>>> to use Stream>>tab instead of putting tab into grease stream. The question
>>>> about isStream is more easy to tackle. The question is what would be the
>>>> replacement/workaround for isStream? I don't see a good one.
>>>>
>>>> Implement a wrapper around GRCodecStream that has the full stream
>>>> protocol. But I guess at one point or another we'll need to add more
>>>> of the stream protocol anyway so we can as well start now.
>>>>
>>>> I wouldn't implement a wrapper around GRCodecStream I would replace it for
>>>> my purpose. And that is probably not one of the goals of grease. The only
>>>> thing  I need to do is to read a XML file in a platform independent manner.
>>>> This doesn't work with grease and that is a pitty. I don't even recognize
>>>> the need for discussing this thoroughly. I'm just doing it because I don't
>>>> have commit rights in the repo.
>>>> So it would be nice to hear something definite from the core developers. If
>>>> you are willing to add this to grease that would be fine.
>>>
>>> Would be fine with me.
>>
>> I don't feel strongly either way. So, sure, it's fine with me.
>>
> Ok,here is the changeset,