| xText: any text rule [message #375966] |
Wed, 21 February 2007 12:10  |
Eclipse User |
|
|
|
Hi.
I have read short documentation on xText framework, server-side article
and have seen screen casts.
But I have not found answer on following question.
Is there is any possibility to create rule "Any text"?
In my model domain text to be analyzed is arbitrary but I need to
proceed in some logic way text fragments like
{action name='my_name'}
So for me it doesn't matter what text is before and after.
For example of text:
<!-- fdfdf -->
#sdfsdf
*dddd {action name='my_name1'} 090909
*dddd {action name='my_name2'} 090909
<!-- fdfdf -->
The only interesting parts for me are
{action name='my_name1'}
{action name='my_name2'}
Is there is some built-in rule/token like ANY_TEXT?
If there is no such rule could you help me to prototype xText grammar
for such language?
Thanks in advance.
Ilya
|
|
|
|
|
|
| Re: xText: any text rule [message #376305 is a reply to message #375985] |
Mon, 12 March 2007 09:02  |
| Eclipse User |
|
|
|
Hi Bernd,
3 weeks passed since your reply but there is no new oaw release.
Could I ask you when it will be released?
Thanks
Ilya
Bernd Kolb wrote:
> Ilya Klyuchnikov wrote:
>> It's very cool that I'll be able to create Antlr rules!
>>
>> BTW, will docs/tutorials be upgraded in short terms?
>
>
> Hopefully. As soon as we find the time. We are always grateful for help ;-)
>
> Bernd
>
>
>>
>> Thanks
>>
>> Ilya
>>
>> Bernd Kolb wrote:
>>> Hi Ilya,
>>>
>>> sorry for the late reply.
>>>
>>> We'll release a new version of oaw before the EclipseCon (I think mid
>>> to end next week)
>>>
>>> In the new release we ported the grammar from ANTLR 2 to 3 and added
>>> some features. The new version is backward compatible.
>>>
>>> With the current version you can't solve your problem.
>>> With the new version you can :-)
>>>
>>> It'll be possible to specify a 'NativeLexerRule'. In there you'll be
>>> able to write your own rule in Antlr format.
>>>
>>> The syntax will look like this:
>>> 'NativeLexerRule' ruleName ':' STRING ';'
>>>
>>>
>>> Hope that helps.
>>>
>>> Best regards,
>>>
>>> Bernd
>>>
>>>
>>>
>>>
>>> Ilya Klyuchnikov wrote:
>>>> Hi.
>>>>
>>>> I have read short documentation on xText framework, server-side
>>>> article and have seen screen casts.
>>>>
>>>> But I have not found answer on following question.
>>>>
>>>> Is there is any possibility to create rule "Any text"?
>>>>
>>>> In my model domain text to be analyzed is arbitrary but I need to
>>>> proceed in some logic way text fragments like
>>>>
>>>> {action name='my_name'}
>>>>
>>>> So for me it doesn't matter what text is before and after.
>>>> For example of text:
>>>>
>>>> <!-- fdfdf -->
>>>> #sdfsdf
>>>> *dddd {action name='my_name1'} 090909
>>>> *dddd {action name='my_name2'} 090909
>>>> <!-- fdfdf -->
>>>>
>>>> The only interesting parts for me are
>>>>
>>>> {action name='my_name1'}
>>>> {action name='my_name2'}
>>>>
>>>> Is there is some built-in rule/token like ANY_TEXT?
>>>>
>>>> If there is no such rule could you help me to prototype xText
>>>> grammar for such language?
>>>>
>>>> Thanks in advance.
>>>>
>>>> Ilya
>>>>
|
|
|
| Re: xText: any text rule [message #571629 is a reply to message #375966] |
Fri, 23 February 2007 03:05 |
| Eclipse User |
|
|
|
Hi Ilya,
sorry for the late reply.
We'll release a new version of oaw before the EclipseCon (I think mid to
end next week)
In the new release we ported the grammar from ANTLR 2 to 3 and added
some features. The new version is backward compatible.
With the current version you can't solve your problem.
With the new version you can :-)
It'll be possible to specify a 'NativeLexerRule'. In there you'll be
able to write your own rule in Antlr format.
The syntax will look like this:
'NativeLexerRule' ruleName ':' STRING ';'
Hope that helps.
Best regards,
Bernd
Ilya Klyuchnikov wrote:
> Hi.
>
> I have read short documentation on xText framework, server-side article
> and have seen screen casts.
>
> But I have not found answer on following question.
>
> Is there is any possibility to create rule "Any text"?
>
> In my model domain text to be analyzed is arbitrary but I need to
> proceed in some logic way text fragments like
>
> {action name='my_name'}
>
> So for me it doesn't matter what text is before and after.
> For example of text:
>
> <!-- fdfdf -->
> #sdfsdf
> *dddd {action name='my_name1'} 090909
> *dddd {action name='my_name2'} 090909
> <!-- fdfdf -->
>
> The only interesting parts for me are
>
> {action name='my_name1'}
> {action name='my_name2'}
>
> Is there is some built-in rule/token like ANY_TEXT?
>
> If there is no such rule could you help me to prototype xText grammar
> for such language?
>
> Thanks in advance.
>
> Ilya
>
|
|
|
| Re: xText: any text rule [message #571655 is a reply to message #375969] |
Fri, 23 February 2007 04:10 |
| Eclipse User |
|
|
|
It's very cool that I'll be able to create Antlr rules!
BTW, will docs/tutorials be upgraded in short terms?
Thanks
Ilya
Bernd Kolb wrote:
> Hi Ilya,
>
> sorry for the late reply.
>
> We'll release a new version of oaw before the EclipseCon (I think mid to
> end next week)
>
> In the new release we ported the grammar from ANTLR 2 to 3 and added
> some features. The new version is backward compatible.
>
> With the current version you can't solve your problem.
> With the new version you can :-)
>
> It'll be possible to specify a 'NativeLexerRule'. In there you'll be
> able to write your own rule in Antlr format.
>
> The syntax will look like this:
> 'NativeLexerRule' ruleName ':' STRING ';'
>
>
> Hope that helps.
>
> Best regards,
>
> Bernd
>
>
>
>
> Ilya Klyuchnikov wrote:
>> Hi.
>>
>> I have read short documentation on xText framework, server-side
>> article and have seen screen casts.
>>
>> But I have not found answer on following question.
>>
>> Is there is any possibility to create rule "Any text"?
>>
>> In my model domain text to be analyzed is arbitrary but I need to
>> proceed in some logic way text fragments like
>>
>> {action name='my_name'}
>>
>> So for me it doesn't matter what text is before and after.
>> For example of text:
>>
>> <!-- fdfdf -->
>> #sdfsdf
>> *dddd {action name='my_name1'} 090909
>> *dddd {action name='my_name2'} 090909
>> <!-- fdfdf -->
>>
>> The only interesting parts for me are
>>
>> {action name='my_name1'}
>> {action name='my_name2'}
>>
>> Is there is some built-in rule/token like ANY_TEXT?
>>
>> If there is no such rule could you help me to prototype xText grammar
>> for such language?
>>
>> Thanks in advance.
>>
>> Ilya
>>
|
|
|
| Re: xText: any text rule [message #571845 is a reply to message #375971] |
Fri, 23 February 2007 13:27 |
| Eclipse User |
|
|
|
Ilya Klyuchnikov wrote:
> It's very cool that I'll be able to create Antlr rules!
>
> BTW, will docs/tutorials be upgraded in short terms?
Hopefully. As soon as we find the time. We are always grateful for help ;-)
Bernd
>
> Thanks
>
> Ilya
>
> Bernd Kolb wrote:
>> Hi Ilya,
>>
>> sorry for the late reply.
>>
>> We'll release a new version of oaw before the EclipseCon (I think mid
>> to end next week)
>>
>> In the new release we ported the grammar from ANTLR 2 to 3 and added
>> some features. The new version is backward compatible.
>>
>> With the current version you can't solve your problem.
>> With the new version you can :-)
>>
>> It'll be possible to specify a 'NativeLexerRule'. In there you'll be
>> able to write your own rule in Antlr format.
>>
>> The syntax will look like this:
>> 'NativeLexerRule' ruleName ':' STRING ';'
>>
>>
>> Hope that helps.
>>
>> Best regards,
>>
>> Bernd
>>
>>
>>
>>
>> Ilya Klyuchnikov wrote:
>>> Hi.
>>>
>>> I have read short documentation on xText framework, server-side
>>> article and have seen screen casts.
>>>
>>> But I have not found answer on following question.
>>>
>>> Is there is any possibility to create rule "Any text"?
>>>
>>> In my model domain text to be analyzed is arbitrary but I need to
>>> proceed in some logic way text fragments like
>>>
>>> {action name='my_name'}
>>>
>>> So for me it doesn't matter what text is before and after.
>>> For example of text:
>>>
>>> <!-- fdfdf -->
>>> #sdfsdf
>>> *dddd {action name='my_name1'} 090909
>>> *dddd {action name='my_name2'} 090909
>>> <!-- fdfdf -->
>>>
>>> The only interesting parts for me are
>>>
>>> {action name='my_name1'}
>>> {action name='my_name2'}
>>>
>>> Is there is some built-in rule/token like ANY_TEXT?
>>>
>>> If there is no such rule could you help me to prototype xText grammar
>>> for such language?
>>>
>>> Thanks in advance.
>>>
>>> Ilya
>>>
|
|
|
| Re: xText: any text rule [message #578022 is a reply to message #375985] |
Mon, 12 March 2007 09:02 |
| Eclipse User |
|
|
|
Hi Bernd,
3 weeks passed since your reply but there is no new oaw release.
Could I ask you when it will be released?
Thanks
Ilya
Bernd Kolb wrote:
> Ilya Klyuchnikov wrote:
>> It's very cool that I'll be able to create Antlr rules!
>>
>> BTW, will docs/tutorials be upgraded in short terms?
>
>
> Hopefully. As soon as we find the time. We are always grateful for help ;-)
>
> Bernd
>
>
>>
>> Thanks
>>
>> Ilya
>>
>> Bernd Kolb wrote:
>>> Hi Ilya,
>>>
>>> sorry for the late reply.
>>>
>>> We'll release a new version of oaw before the EclipseCon (I think mid
>>> to end next week)
>>>
>>> In the new release we ported the grammar from ANTLR 2 to 3 and added
>>> some features. The new version is backward compatible.
>>>
>>> With the current version you can't solve your problem.
>>> With the new version you can :-)
>>>
>>> It'll be possible to specify a 'NativeLexerRule'. In there you'll be
>>> able to write your own rule in Antlr format.
>>>
>>> The syntax will look like this:
>>> 'NativeLexerRule' ruleName ':' STRING ';'
>>>
>>>
>>> Hope that helps.
>>>
>>> Best regards,
>>>
>>> Bernd
>>>
>>>
>>>
>>>
>>> Ilya Klyuchnikov wrote:
>>>> Hi.
>>>>
>>>> I have read short documentation on xText framework, server-side
>>>> article and have seen screen casts.
>>>>
>>>> But I have not found answer on following question.
>>>>
>>>> Is there is any possibility to create rule "Any text"?
>>>>
>>>> In my model domain text to be analyzed is arbitrary but I need to
>>>> proceed in some logic way text fragments like
>>>>
>>>> {action name='my_name'}
>>>>
>>>> So for me it doesn't matter what text is before and after.
>>>> For example of text:
>>>>
>>>> <!-- fdfdf -->
>>>> #sdfsdf
>>>> *dddd {action name='my_name1'} 090909
>>>> *dddd {action name='my_name2'} 090909
>>>> <!-- fdfdf -->
>>>>
>>>> The only interesting parts for me are
>>>>
>>>> {action name='my_name1'}
>>>> {action name='my_name2'}
>>>>
>>>> Is there is some built-in rule/token like ANY_TEXT?
>>>>
>>>> If there is no such rule could you help me to prototype xText
>>>> grammar for such language?
>>>>
>>>> Thanks in advance.
>>>>
>>>> Ilya
>>>>
|
|
|
Powered by
FUDForum. Page generated in 0.04253 seconds
Home