| Forum: TMF (Xtext) |
|---|
| Topic: Serialization & formatting |
|---|
| Serialization & formatting [message #499988] |
Tue, 24 November 2009 11:22 |
 Remi Melisson
Messages: 1
Registered: November 2009 |
Junior Member |
|
|
Hi everyone,
I got some problems with white spaces when the serialization process occurs.
I use a FullyQualifiedName rule like this:
grammar org.ow2.Fractal with org.eclipse.xtext.common.Terminals hidden(WS, ML_COMMENT, SL_COMMENT)
ImportDefinition :
'import' importName=FullyQualifiedName ';' ;
FullyQualifiedName :
ID('.'ID)*;
With this kind of definition :
import a;
import b;
import c;
For example, when I update a name with the EMF tree (b -> z), and save it, the result is :
import a;
importz;
importc;
Another detail: I let the serialization automatically process without any custom classes. And I don't use any other terminal rule.
The missing spaces could be related with DefaultHiddenTokenMerger or FormattingConfigBasedStream or with my grammar.
So, I've got two question :
* Is it possible to use my formatting configuration (which is able to format my definition when I use ctrl+shift+f) or the OneWhitespaceFormatter instead of the FormattingConfigBasedStream ? And, independently, is it possible to automatically launch my formatter after all ?
* Do you have any idea about this behavior, because debugging make me feel a little bit nervous ^^ ?
Thanks
|
|
|
| Topic: Using a DSL to parse an EString attribute |
|---|
|
| Topic: [Announce] Eclipse/OMG Symposium 2010 |
|---|
| [Announce] Eclipse/OMG Symposium 2010 [message #496347] |
Mon, 09 November 2009 13:35 |
Kenn Hussey
Messages: 812
Registered: July 2009 |
Senior Member |
|
|
In 2008, Eclipse and the OMG (Object Management Group) jointly ran two
one-day symposia to promote and build on the partnership between
Eclipse's open source software and OMG's open specifications. Following
the success of these events, OMG and Eclipse will hold a further one-day
symposium during the OMG Technical Meeting in Minneapolis, Minnesota on
23rd June 2010. The symposium will be organized as a series of
discussion sessions on OMG specifications and corresponding Eclipse
projects, to discuss the alignment between current specifications and
implemented software, and identify areas where the cooperation could be
further improved in the future.
Eclipse and OMG invite position papers of up to 600 words on any of the
topic areas below, or any other area where an OMG specification relates
to Eclipse software. The Program Committee will publish all submitted
position papers, and invite some of the paper authors to lead individual
discussion sessions. This symposium is a further opportunity to
participate in shaping the joint future of the Eclipse Open Source
community and the OMG Open Specifications community. Please join us for
stimulating technical planning and discussion in this exciting area.
Topics may include (but are not limited to) the following:
- Extensibility of specifications and/or open source implementations to
support
> conforming commercial implementations
> new application domains new
> requirements
- Processes for evolving specifications and/or open source implementations
- Open source implementations for specifications
- Standardized models or interfaces for open source implementations
- Collaboration between open source implementations
Papers that deal with specific problems and solutions that may benefit
from a wider discussion than that available though Eclipse Bug Reports
or OMG Issues are particularly welcome. Interested individuals or
organizations are invited to submit a brief (up to 600 words) position
paper by 24th February 2010 using the form at
http://www.omg.org/abstracts (please select "Session Presentation").
The Program Committee will send invitations to prospective leaders for
each session in early March 2010. The final symposium agenda and
registration details will be available on 17th March 2010 and posted at
http://www.omg.org/news/meetings/tc/mn/special-events/Eclips e.htm.
We hope to see you at this special event.
|
|
|
| Topic: Eclipse Modeling Days |
|---|
| Eclipse Modeling Days [message #496222] |
Mon, 09 November 2009 06:57 |
Ed Merks
Messages: 15894
Registered: July 2009 |
Senior Member |
|
|
This is a multi-part message in MIME format.
--------------050606080005090201040308
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Hi,
As I blogged this morning
< http://ed-merks.blogspot.com/2009/11/do-it-better-with-model .html>, the
dead line for registering for Eclipse Modeling Day
<http://wiki.eclipse.org/Eclipse_Modeling_Day> is rapidly approaching,
i.e., the 12th for the New York event on the 16th and the 13th for the
Toronto event on the 18th. The Toronto event is followed by the Eclipse
Runtime Day <http://wiki.eclipse.org/EclipseRT_Day> on the 19th, so it's
almost like an Eclipse Summit only completely free.
Cheers,
Ed
--------------050606080005090201040308
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
</head>
<body bgcolor="#ffffff" text="#000000">
Hi,<br>
<br>
As I <a
href=" http://ed-merks.blogspot.com/2009/11/do-it-better-with-model .html">blogged
this morning</a>, the dead line for registering for <a
href="http://wiki.eclipse.org/Eclipse_Modeling_Day">Eclipse Modeling
Day</a> is rapidly approaching, i.e., the 12th for the New York event
on the 16th and the 13th for the Toronto event on the 18th. The
Toronto event is followed by the <a
href="http://wiki.eclipse.org/EclipseRT_Day">Eclipse Runtime Day</a>
on the 19th, so it's almost like an Eclipse Summit only completely free.<br>
<br>
Cheers,<br>
Ed<br>
</body>
</html>
--------------050606080005090201040308--
|
|
|
| Topic: Mistake in the Xtext User Guide / Getting Started Example |
|---|
| Mistake in the Xtext User Guide / Getting Started Example [message #495896] |
Fri, 06 November 2009 09:41 |
Raffaele Merico
Messages: 1
Registered: November 2009 |
Junior Member |
|
|
The code provided for the Main.xpt file contains twice the expressions:
- this.types.typeSelect(Entity)
In order to work with the Model definition of the "Getting Started Example" this expressions must be changed to:
- this.elements.typeSelect(Entity)
Raffaele
|
|
|
| Topic: Xtext Presentation on Banglaore DemoCamp |
|---|
|
| Topic: Getting errors and warnings in "compiler style" when running in batch headless? |
|---|
| Getting errors and warnings in "compiler style" when running in batch headless? [message #493082] |
Thu, 22 October 2009 22:52 |
Michael Scharf
Messages: 243
Registered: July 2009 |
Senior Member |
|
|
Hi,
is there a way to get warnings and errors of a
xtext parser on stdout/stderror (in a format that
something like emacs can consume or a the eclipse
console)?
I am sure there is a something I can inject but that
is hard to find for me....
Michael
--
Michael Scharf
Wind River Systems GmbH
http://www.WindRiver.com
http://MichaelScharf.blogspot.com/
|
|
|
| Topic: [Xtext] how to serialize a model? |
|---|
| [Xtext] how to serialize a model? [message #492991] |
Thu, 22 October 2009 11:33 |
Michael Burkhardt
Messages: 53
Registered: July 2009 |
Member |
|
|
Hallo *,
has anybody some lines of code to show, how a model can be serialized with TMF?
Parsing DSL code to DSL ecore model (found in mailing list)
=== begin listing ===
Injector injector = new DslStandaloneSetup().createInjectorAndDoEMFRegistration();
XtextResourceSet xtextResourceSet = injector.getInstance(XtextResourceSet.class);
xtextResourceSet.addLoadOption(XtextResource.OPTION_RESOLVE_ ALL, Boolean.TRUE);
Map<?, ?> xtextOptions = xtextResourceSet.getLoadOptions();
Resource resource = xtextResourceSet.createResource("any uri with the file
extension of my dsl");
resource.load(new ByteArrayInputStream("my dsl source code") , xtextOptions);
/* my dsl source will be parsed and a abstract syntax tree (or model) will be
created and put at index 0 of contents in my EMF resource */
Model ret = (Model) resource.getContents().get(0);
=== end listing ===
OK, I have some tree in 'Model'. Now I navigate through my model. Some of the
nodes in this model I want to serialize.
How does this work?
For this example, use this EBNF:
Model : A* B;
A: X* Y*;
B: "b";
X: "x";
Y: "y";
For example, I want to print all X.
Which lines of code should do I use for a stand alone setup app?
Ciao, Micha
|
|
|
| Topic: [Xtext] Anybody working (or planning to work) on an ANSI-C grammar for Xtext |
|---|
| [Xtext] Anybody working (or planning to work) on an ANSI-C grammar for Xtext [message #490395] |
Thu, 08 October 2009 10:21 |
Steven Derrien
Messages: 17
Registered: July 2009 |
Junior Member |
|
|
Hello,
I am thinking about implementing a ANSI-C grammar with Xtext, but I am a
bit scared of the amount of work it represents since I am not really an
expert in parser writing.
I know of ANTLR versions of an ANSI compliant C grammar, but I really
have no idea how easy it would be to port the ANTLR existing stuff to Xtext.
Is there anybody in this working on (or thinking of) something similar ?
If so I'd be *really* interested in sharing ideas and efforts to provide
an open-source ANSI-C Xtext implementation.
Steven
|
|
|
| Topic: Ignoring parts of source code |
|---|
| Ignoring parts of source code [message #488623] |
Tue, 29 September 2009 09:56 |
Roland S.
Messages: 6
Registered: September 2009 |
Junior Member |
|
|
Hello everyone,
is it possible to let the parser ignore parts of source code? For example:
Rule : "keyword" "(" Inore_This ")" ";" ;
The parser should ignore everything (except the keyword ");" of course) between the left and right parenthesis.
Roland
|
|
|
| Topic: Replacing String in xText Grammar |
|---|
| Replacing String in xText Grammar [message #487419] |
Wed, 23 September 2009 04:07 |
No Real Name
Messages: 3
Registered: September 2009 |
Junior Member |
|
|
Hello,
I want to build a grammar in xText where it is possible to have tokens
within strings.
For this I have overwritten STRING and build my own rule for it. But my rule
does not work always as expected.
In the example all lines except for the last one work as expected. The "
TOKEN#" is treated like a keyword where it should be treated like normal
text for my understanding.
The grammar for this example is below the text example.
Regards,
Martin
######## TEXT EXAMPLE ##########
#TOKEN 1 TOKEN#
"TOKEN##TOKEN 1 TOKEN#"
"2TOKEN##TOKEN 1 TOKEN#"
"tTOKEN##TOKEN 1 TOKEN#"
" TOKEN#"
#########################
######## MATCHING GRAMMAR ########
grammar org.xtext.example.MyDsl with org.eclipse.xtext.common.Terminals
generate myDsl "http://www.xtext.org/example/MyDsl"
Model :
(elements+=(Token))*;
Token:
MyString | SomeToken;
terminal STRING:
'î';MyString:
'" '(((tokens+=SomeToken)|(mytext+=text)|(number+=INT)|(char+=' #')))+' "';
terminal text:
('a'|'b'|'c'|'d'|'e'|'f'|'g'|'h'|'i'|'j'|'k'|'l'|'m'|'n'|'o' |'p'|'q'|'r'|'s'|'t'|'u'|'v'|'w'
|'x'|'y'|'z'|'A'|'B'|'C'|'D'|'E'|'F'|'G'|'H'|'I'|'J'|'K'|'L' |'M'|'N'|'O'|'P'|'Q'|'R'|'S'|'T'|'U'|
'V'|'W'|'X'|'Y'|'Z'|' ')+;
SomeToken:
'#TOKEN ' number+=INT ' TOKEN#';
#########################
|
|
|
| Topic: [TCS] two-step reference |
|---|
| [TCS] two-step reference [message #487262] |
Tue, 22 September 2009 11:30 |
|
Originally posted by: ivanomalavolta.gmail.com
Hello everybody, I m having problems to get elements reltaed by a
two-stpe reference.
This is an examplary metamodel in km3:
package pack {
class A {
attribute name : String;
reference r1[*] container : B oppositeOf r2;
}
class B {
reference r2[1-1] : A oppositeOf r1;
reference r3[*] container : C oppositeOf r4;
}
class C {
reference r4[1-1] : B oppositeOf r3;
}
}
----
in TCS I want to refer the name attribute of A while matching
instances of the C metaclass, like this:
template A
: ...
;
template B
: ...
;
template C
: r4.r2{refersTo = name, autoCreate = never}
;
I know that this is not syntactically correct, but my question is:
is it possible to do something like that??
An alternative could be to access the "owner" of an element, like for
example using something like refImmediateComposite() in ATL... is there
something similar in TCS?
Best regards,
Ivano
|
|
|
| Topic: Re: How to invoke ATL transformation from Xtext workflow? |
|---|
| Re: How to invoke ATL transformation from Xtext workflow? [message #483107] |
Mon, 31 August 2009 03:14 |
Sebastian Zarnekow
Messages: 700
Registered: July 2009 |
Senior Member |
|
|
Hi Adalbert,
as your question is related to MWE, I'll add the emft newsgroup to this
reply.
I guess that you'll have to implement your own workflow component which
is documented in the eclipse helf of mwe.
Regards,
Sebastian
--
Need professional support for Eclipse Modeling?
Go visit: http://xtext.itemis.com
| Adalbert Gorecki wrote on Sat, 29 August 2009 17:29 | Hello
everyone,
I think the subject says it all :)
I'm using ATL for M2M transformation and I'd like to run that translation
from an xtext workflow file. I can run other workflow files from my main
workflow with
<component file="path/to/my/workflow/file.mwe" />
Does anyone know how I can do something like this with an atl file?
Regards,
Adalbert
|
|
|
|
| Topic: How to invoke ATL transformation from Xtext workflow? |
|---|
| How to invoke ATL transformation from Xtext workflow? [message #483040] |
Sat, 29 August 2009 17:29 |
Adalbert Gorecki
Messages: 10
Registered: July 2009 |
Junior Member |
|
|
Hello everyone,
I think the subject says it all :)
I'm using ATL for M2M transformation and I'd like to run that translation
from an xtext workflow file. I can run other workflow files from my main
workflow with
<component file="path/to/my/workflow/file.mwe" />
Does anyone know how I can do something like this with an atl file?
Regards,
Adalbert
|
|
|
| Topic: [TCS] loosing Data after model Injection and extraction |
|---|
| [TCS] loosing Data after model Injection and extraction [message #482347] |
Wed, 26 August 2009 06:31 |
|
Originally posted by: kanso.marwa.gmail.com
Hello,
Is it normal to loose data once I inject a model file into a .xmi file and
when I extract the .xmi file into a model file again ?
thanks for ur help,
Marwa
|
|
|
| Topic: [Announce] TMF XTEXT 0.7.2RC1 is available |
|---|
|
| Topic: [XText] parseResult is null |
|---|
| [XText] parseResult is null [message #478863] |
Fri, 07 August 2009 05:50 |
Michael Burkhardt
Messages: 53
Registered: July 2009 |
Member |
|
|
Hello *,
I've tried to use more CrossReference, but now my editor has errors while code
completion and the Outline-view is empty.
Error Message (in Error Log ;-):
"Content Assist" did not complete normally. Please see the log for more
information.
Stack Trace:
An exception stack trace is not available.
What went wrong?
Ciao, Micha
|
|
|
| Topic: DSL editor incl. error and warning pane in standalone (RCP/PDE) application |
|---|
| DSL editor incl. error and warning pane in standalone (RCP/PDE) application [message #468128] |
Tue, 04 August 2009 06:15 |
Marcel Merkel
Messages: 2
Registered: July 2009 |
Junior Member |
|
|
Hi,
I'd like to ship my dsl including an editor to the customer later. This
shipment should basically include only the XText editor for my dsl, a
panel which shows errors and warnings (like the one in eclipse). The
editor should have the same features (highlighting, Keyboard shortcuts,
...) as the one eclipse offers. What would be the easiest way to do this ?
Xtext already sets up an UI project which uses RCP. I guess I could make
my own RCP product and import the editor somehow. Loading the dsl in a
standalone java application has been greatly discussed in this newsgroup
but I think this is not quite what I want. If I understand it correctly
then I would need to add the panel which shows the errors and warnings
manually. Afterwards I might have to deal with keyboard shortcuts and so
on.
However I think there is a far easier way for that. It is basically
everything already in eclipse present. All I have to do is to generate a
product / possibly a pde which only holds the necessary components. A
stripped down version of eclipse which only has the editor, the warnings
and error panel and the possibility to add a custom menu or two to three
buttons would be enough for my purpose. I just don't know how to start on
this. Could someone give me some hints/suggestions on that ?
Any help would be appreciated.
Marcel
|
|
|
| Topic: [Announce] TMF XTEXT 0.7.1 I200907291635 is available |
|---|
|
| Topic: Web-Frontend for this Newsgroup |
|---|
|
| Topic: Xtext replace internal model instance |
|---|
| Xtext replace internal model instance [message #361787] |
Tue, 28 July 2009 09:52 |
Hauke Fuhrmann
Messages: 184
Registered: July 2009 |
Senior Member |
|
|
Hi there!
is there a way to get programmatically the actual ecore model instance
that an Xtext editor (or even the parser used standalone without UI)
works on?
Or the other way round: can I change it? Can I swap the current model
instance by another one?
Rationale: I have an GMF graphical editor that edits the model instance.
I want to use an Xtext parser only as parser for labels in the GMF
editor which should edit only small parts of the model instance. Could
that be possible?
For now my workaround is, to create a completely new model instance by
the Xtext parser and manually fumble that into the GMF model instance.
Not very convenient, especially where elements are only referenced.
Cheers,
Hauke
|
|
|
| Topic: xtend SystemCommand in mwe |
|---|
| xtend SystemCommand in mwe [message #63160] |
Mon, 27 July 2009 09:38 |
Andreas Graf
Messages: 2
Registered: July 2009 |
Junior Member |
|
|
Hi,
I am trying to migrate a oaw4 to Eclipse 3.5 TMF. The .mwe has the
following fragment:
<component class="org.eclipse.xtend.util.stdlib.SystemCommand">
<directory value="src-gen"/>
<command value="cmd"/>
<arg value="/c ..\\AStyle.exe -n -R *"/>
</component>
This worked fine in oaw. However, with Galileo, it complains
[ERROR]: Cannot convert from String to 'java.io.File'. You need to
register a corresponding Converter.(
If I remove the <directory ...> The error message always is:
[ERROR]: Runtime error: null(Element: -UNKNOWN-; Reported by: -UNKNOWN-)
Any help is appreciated.
|
|
|
| Topic: [XText] Generator properties |
|---|
| [XText] Generator properties [message #62907] |
Thu, 23 July 2009 05:01 |
Michael Burkhardt
Messages: 53
Registered: July 2009 |
Member |
|
|
Hello *,
I want to split the XText main project into two subprojects:
o parser project
I think, this is known as XText main project with its "src" folder.
o grammar project
This is a project with the EMF file. I think, this is known as "src-gen"
folder of the main project.
Second, I want to create other src-folders. "src" and "src-gen" is nice, but I
need something like "src/main/java" and "src-gen/main/java". Reason: we use
Maven2 (http://maven.apache.org) for project management.
However, today I search for some properties for the generator / workflow. The
XText documentation (chapter 3) says something about beans and components, but I
found no named properties and possible values.
Is there an other doc?
Ciao, Micha
|
|
|
| Topic: [Announce] TMF XTEXT 0.7.1 is available |
|---|
|
| Topic: Re: Modifying Content assist, and notifying linking service |
|---|
| Re: Modifying Content assist, and notifying linking service [message #58608] |
Tue, 14 July 2009 15:55 |
Sebastian Zarnekow
Messages: 700
Registered: July 2009 |
Senior Member |
|
|
Hi Durga,
as this is a TMF related question, I added the right newsgroup to the
recipients of this reply.
You actually had the right idea: Your problem is not only related to
content assist but to the linking stage, too. Xtext has a special
service that solves the somewhat complicated task of collecting all
referable objects. It is the IScopeProvider.
What you actually have to do is to implement a custom scope provider
service, that will create an IScope based on the content of your repository.
If you have any further questions regarding the scope provider
interface, please don't hesitate to ask them in the TMF newsgroup.
Regards,
Sebastian
Am 14.07.2009 12:57 Uhr, schrieb Durga Prasana Sahoo:
> Hi All
>
> I want to modify/enhance my tmf editor's content assist in a manner
> specific to my requirements.
>
> Actually, I have a grammar, that I'm using in fragements for managing
> model elements.It's like, I have a navigation view that shows the
> various model elements in the repository and when one clicks on any
> particular element, I invoke the editor that shows up the code for that
> particular element, instead of showing the concrete model for the entire
> model.
>
> So, I need to somehow modify the content assist, So as to show the list
> of possible elements that can be crossreferenced from the repository,
> while editing the code for any particular attribute of an element.
>
> ____________________________________________________________ _______________
> public void complete[TypeName]_[FeatureName](
> EObject model, Assignment assignment, ContentAssistContext context,
> ICompletionProposalAcceptor acceptor) {
> // clients may override
> }
>
> public void complete_[RuleName](
> EObject model, RuleCall ruleCall, ContentAssistContext context,
> ICompletionProposalAcceptor acceptor) {
> // clients may override
> }
> ____________________________________________________________ _______________
>
> I was going through the XText docs, and it seems if I can redefine the
> above prototype methods in EdificeProposalProvider.java('Edifice' is my
> grammar),
> then I can possibly get the names of possible elements in repository
> while invoking content assist for an attribute.
>
> Cud anyone pls tell me how to feed in the list of valid elements that
> can be crossreferenced for an attribute.
>
> ____________________________________________________________ _______________
>
> say, I have an (of kind MessageType){Object[] objs} array(this means
> that there are actually objects of type MessageType in the repository
> which can be crossreferenced).
> and I want to feed in the names of these objects for the content assist of
> attrib input/output of Operation:
>
> Operation returns refn_staticstructure::Operation:
> 'Operation' id=STRING ':' name=STRING '{'
> //parent attrib(s)
> 'doc:' documentation=STRING';'
> //class specific attributes
> ('input:' input=[refn_staticstructure::MessageType])? ';'
> ('output:' output=[refn_staticstructure::MessageType])? ';'
> 'throwsExceptions:' (throwsExceptions+=OperationException)* ';'
> '}'
>
> ;
>
> ____________________________________________________________ ________________
>
>
> how do I modify:
>
> public void completeOperation_input(
> EObject model, Assignment assignment, ContentAssistContext context,
> ICompletionProposalAcceptor acceptor) {
>
> //code snippet to feed in the names of valid crossreferenceable objects ??
>
> }
>
>
> Any suggestions
>
> ____________________________________________________________ ________________
>
>
> In addition to it, I need 2 somehow transmit this information to the
> default
> linking service that indeed doesn't know that these elements can indeed
> be crossreferenced, and avoid showing error marks.
>
> (Since every time an element from repository view is invoked, I create a
> concrete model file pertaining to that element from the repository.xml
> using XPand2, So the entire concrete model is actually split into multiple
> <element_name>.edifice files).
>
> any advice??
> :)
>
>
> Thanks.
>
> Durga
>
|
|
|
| Topic: [xText] Pulling up features into common super class |
|---|
| [xText] Pulling up features into common super class [message #58352] |
Tue, 14 July 2009 07:47 |
|
Originally posted by: c.krause.cwi.nl
Hi all,
I want xText to generate a common superclass for two types and to pull
up a feature into this superclass. More specifically, I have my grammar
for arithmetical expressions:
/* ----- N-ary Operations ----- */
Addition returns Expression:
Subtraction ( {Addition.operands += current} ('+' operands +=
Subtraction)+)?;
Multiplication returns Expression:
Division ( {Multiplication.operands += current} ('*' operands +=
Division)+)?;
/* ----- Binary Operations ----- */
Subtraction returns Expression:
Multiplication ( {Subtraction.left = current} '-' right =
Multiplication)?;
Division returns Expression:
Operation ( {Division.left = current} '/' right = Operation)?;
Operation returns Expression:
Literal | '(' Addition ')';
....and I would like to have the following type hierarchy generated:
* Operation, Literal -> Expression
* BinaryOperation, NaryOperation -> Operation
* Addition, Multiplication -> NaryOperation
* Subtraction, Division -> BinaryOperation
In all my attempts I ended up with a cyclic hierarchy. What I basically
tried to do is to add a dummy rule:
Operation:
BinaryOperation | NaryOperation;
I believe the problem is that in my current model Operation is not a type.
If anybody has a hint, I'd appreciate any help. Anyway, I will try to
get it working somehow..
Cheers,
Christian
|
|
|
| Topic: Eclipse xtext projects to Central Maven Repository |
|---|
|
| Topic: Maven Support for XTEXT |
|---|
| Maven Support for XTEXT [message #52684] |
Wed, 24 June 2009 12:52 |
Martin Kuhn
Messages: 33
Registered: July 2009 |
Member |
|
|
Hi,
I recognized the release of TMF XTEXT 0.7.0
Congratulation for the really good work you have done.
One question about maven:
I found a sample project config for maven (which uses 0.7.0-M6-SNAPSHOT
artifacts located on
http://www.fornax-platform.org/archiva/repository/snapshots).
Are there maven artifacts available for the current release?
It would be a great help for all of us which have to use maven!
Could anybody give me a hint?
TIA
Martin
|
|
|
| Topic: An invalid XML character |
|---|
| An invalid XML character [message #52142] |
Wed, 17 June 2009 22:11 |
Erol Akarsu
Messages: 2
Registered: July 2009 |
Junior Member |
|
|
When I run workflow for a DSL, then I am getting this error.
0 [main] INFO eclipse.emf.mwe.core.WorkflowRunner -
------------------------------------------------------------ --------------------------
13 [main] INFO eclipse.emf.mwe.core.WorkflowRunner - EMF Modeling
Workflow Engine 0.7.0, Build v200906160748
13 [main] INFO eclipse.emf.mwe.core.WorkflowRunner - (c) 2005-2009
openarchitectureware.org and contributors
13 [main] INFO eclipse.emf.mwe.core.WorkflowRunner -
------------------------------------------------------------ --------------------------
13 [main] INFO eclipse.emf.mwe.core.WorkflowRunner - running workflow:
/home/eakarsu/workspace-gal/org.xtext.example.mydsl1/src/org /xtext/example/GenerateMyDsl1.mwe
13 [main] INFO eclipse.emf.mwe.core.WorkflowRunner -
848 [main] INFO lipse.emf.mwe.utils.StandaloneSetup - Registering
platform uri '/home/eakarsu/workspace-gal'
5819 [main] INFO e.core.container.CompositeComponent - DirectoryCleaner:
cleaning directory '../org.xtext.example.mydsl1/src-gen'
5819 [main] INFO ipse.emf.mwe.utils.DirectoryCleaner - Cleaning
/home/eakarsu/workspace-gal/org.xtext.example.mydsl1/../org. xtext.example.mydsl1/src-gen
5823 [main] INFO e.core.container.CompositeComponent - DirectoryCleaner:
cleaning directory '../org.xtext.example.mydsl1.ui/src-gen'
5823 [main] INFO ipse.emf.mwe.utils.DirectoryCleaner - Cleaning
/home/eakarsu/workspace-gal/org.xtext.example.mydsl1/../org. xtext.example.mydsl1.ui/src-gen
5825 [main] INFO e.core.container.CompositeComponent - Generator
5875 [main] INFO ipse.xtext.generator.LanguageConfig - generating
infrastructure for org.xtext.example.MyDsl1 with fragments :
ImplicitRuntimeFragment, ImplicitUiFragment, GrammarAccessFragment,
EcoreGeneratorFragment, ParseTreeConstructorFragment,
ResourceFactoryFragment, AntlrDelegatingFragment, JavaValidatorFragment,
JavaScopingFragment, FormatterFragment, LabelProviderFragment,
TransformerFragment, OutlineNodeAdapterFactoryFragment,
JavaBasedContentAssistFragment, DelegatingGeneratorFragment
16296 [main] ERROR enerator.CompositeGeneratorFragment - An invalid XML
character (Unicode: 0xc) was found in the element content:
java.lang.RuntimeException: An invalid XML character (Unicode: 0xc) was
found in the element content:
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl$Escape.convert(XM LSaveImpl.java:3280)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.getDatatypeValue( XMLSaveImpl.java:3064)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveDataTypeSingl e(XMLSaveImpl.java:1660)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveFeatures(XMLS aveImpl.java:1247)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElementID(XML SaveImpl.java:2667)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElement(XMLSa veImpl.java:1156)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElement(XMLSa veImpl.java:1017)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveContainedMany (XMLSaveImpl.java:2368)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveFeatures(XMLS aveImpl.java:1515)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElementID(XML SaveImpl.java:2667)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElement(XMLSa veImpl.java:1156)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElement(XMLSa veImpl.java:1017)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveContainedSing le(XMLSaveImpl.java:2354)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveFeatures(XMLS aveImpl.java:1509)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElementID(XML SaveImpl.java:2667)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElement(XMLSa veImpl.java:1156)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElement(XMLSa veImpl.java:1017)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveContainedMany (XMLSaveImpl.java:2368)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveFeatures(XMLS aveImpl.java:1515)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.saveElementID(XML SaveImpl.java:2667)
at
org.eclipse.emf.ecore.xmi.impl.XMISaveImpl.writeTopObjects(X MISaveImpl.java:90)
at
org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.traverse(XMLSaveI mpl.java:592)
at org.eclipse.emf.ecore.xmi.impl.XMLSaveImpl.save(XMLSaveImpl. java:256)
at
org.eclipse.emf.ecore.xmi.impl.XMLResourceImpl.doSave(XMLRes ourceImpl.java:206)
at
org.eclipse.emf.ecore.resource.impl.ResourceImpl.save(Resour ceImpl.java:1406)
at
org.eclipse.emf.ecore.resource.impl.ResourceImpl.save(Resour ceImpl.java:993)
at
org.eclipse.xtext.generator.grammarAccess.GrammarAccessFragm ent.generate(GrammarAccessFragment.java:61)
at
org.eclipse.xtext.generator.CompositeGeneratorFragment.gener ate(CompositeGeneratorFragment.java:77)
at
org.eclipse.xtext.generator.LanguageConfig.generate(Language Config.java:68)
at org.eclipse.xtext.generator.Generator.generate(Generator.jav a:264)
at org.eclipse.xtext.generator.Generator.invokeInternal(Generat or.java:80)
at
org.eclipse.emf.mwe.core.lib.AbstractWorkflowComponent.invok e(AbstractWorkflowComponent.java:118)
at
org.eclipse.emf.mwe.core.container.CompositeComponent.intern alInvoke(CompositeComponent.java:101)
at
org.eclipse.emf.mwe.core.container.CompositeComponent.invoke (CompositeComponent.java:86)
at
org.eclipse.emf.mwe.core.WorkflowRunner.executeWorkflow(Work flowRunner.java:412)
at org.eclipse.emf.mwe.core.WorkflowRunner.run(WorkflowRunner.j ava:292)
at org.eclipse.emf.mwe.core.WorkflowRunner.main(WorkflowRunner. java:239)
|
|
|
| Topic: [Announce] TMF XTEXT 0.7.0RC5 is available |
|---|
|
| Topic: [Xtext] import scoping |
|---|
| [Xtext] import scoping [message #51795] |
Tue, 16 June 2009 04:41 |
Derek Palma
Messages: 126
Registered: July 2009 |
Senior Member |
|
|
Hi,
I am trying to determine the easiest way to control import scope. I have
been using the default scoping for entities at the file level (like classes)
so I have not taken over the this processing since it works well. My DSL
allows qualifying class references with the containing file's namespace. It
looks like AbstractDeclarativeScopingProvider ends up creating a
LazyReferenceResourceScope with the appropriate entities for the imported
file. It seems if I do this in the scoping provider I will need to duplicate
much of this functionality just adding the filtering by the imported
namespace. This is fine, but I was just wondering if there was a simpler way
so I don't have to take over all the class resolution.
For my DSL I have a syntax like <File>::<Class>. I just need to return the
classes in the imported module so the classes filtered against the
appropriate import name space. Name spaces can be hierachical (imagine one
imported file importing others and DefaultScope.createParent handles this).
Thanks in advance!
Derek
|
|
|
| Topic: Bug in codecompletion? |
|---|
| Bug in codecompletion? [message #51740] |
Tue, 16 June 2009 02:14 |
Michael Baedorf
Messages: 3
Registered: July 2009 |
Junior Member |
|
|
I've a question concerning the codecompletion feature respectively
deriving from TerminalsProposalProvider.
My model looks like
Application Library {
Module media {
Entity Library {
String name
}
abstract Entity Media {
String title
}
Entity Book extends Media {
String isbn
}
ValueObject Engagement {
String role
}
}
}
I defined a dsl like this
...
Model :
app=Application;
Application :
"Application" name=ID "{"
"basePackage" "=" basePackage=JavaIdentifier
(modules+=Module)*
"}";
Module :
"Module" name=ID "{"
(services+=Service)*
(domainObjects+=DomainObject)*
"}";
Attribute :
type=Type name=ID;
DomainObject :
Entity | ValueObject;
Service :
"Service" name=ID "{"
"}";
Entity :
(abstract?="abstract")? "Entity" name=ID ("extends" extends=[Entity])?
"{"
(attributes+=Attribute)+
"}"
;
ValueObject :
(abstract?="abstract")? "ValueObject" name=ID ("extends"
extends=[ValueObject])? "{"
(attributes+=Attribute)+
"}"
;
...
Now I'd like to add codecompletion for the service name:
public void completeService_Name(Service service, Assignment assignment,
ContentAssistContext context, ICompletionProposalAcceptor
acceptor) {
super.completeService_Name(service, assignment, context, acceptor);
...
ICompletionProposal completionProposal =
createServiceNameDefaultProposal(service, context);
acceptor.accept(completionProposal);
}
I noticed two things:
1. Codecompletion is only available if Service is followed by a
ValueObject.
2. Without only user input the service name is type of the following
object(ValueObject, Entity).
Are these bugs or are there any mistakes implementing the codecompletion?
|
|
|
| Topic: xtext generator fails when calling from maven |
|---|
| xtext generator fails when calling from maven [message #50085] |
Thu, 11 June 2009 08:44 |
 Christoph Kulla
Messages: 24
Registered: July 2009 |
Junior Member |
|
|
Hi,
Im trying to mavenize my xtext project with the fornax plugin. I run in
difficulties with a hierarchical module and directory layout. I have a
parent module and in the parent module are submodules which represent the
xtext and the xtext.ui project. When I build only the xtext project
everything works fine. But when starting from the parent project it fails
when reading the grammar files. But the grammar file is correctly
generated!?!
2219 [main] INFO ipse.xtext.generator.LanguageConfig generating
infrastructure for com.XXX.XXX.metamodel.type.Type with fragments :
ImplicitRuntimeFragment, ImplicitUiFragment, GrammarAccessFragment,
EcoreGeneratorFragment, ParseTreeConstructorFragment,
ResourceFactoryFragment, DelegatingGeneratorFragment,
JavaValidatorFragment, JavaScopingFragment, FormatterFragment,
LabelProviderFragment, TransformerFragment,
OutlineNodeAdapterFactoryFragment, JavaBasedContentAssistFragment,
DelegatingGeneratorFragment
error(7): cannot find or open file:
../metamodel.type/src-gen/com/XXX/XXX/metamodel/type/parser/ antlr/internal/InternalType.g
I have no clue what the exact meaning of the "pathRtProject" parameter is.
Is this a relative path? How to put an absolute path there? At which point
of the code is the path to the grammar file specified? Any ideas how to
solve this?
Regards
Christoph
|
|
|
| Topic: [Announce] TMF XTEXT 0.7.0RC4 is available |
|---|
|
| Topic: Running Xtext generator from maven |
|---|
| Running Xtext generator from maven [message #49637] |
Tue, 09 June 2009 05:42 |
Christoph Kulla
Messages: 24
Registered: July 2009 |
Junior Member |
|
|
Hi,
has anyone tried to run the xtext generator from a maven build?
I'm following the blog post
http://kthoms.wordpress.com/2009/05/19/fornax-oaw-plugin-ver sion-3-0-0-released/.
I'm using xtext 0.7 RC2 an getting following error when running the fornax
maven plugin
org.eclipse.emf.common.util.WrappedException: java.io.IOException: The
path
'/metamodel.type/src-gen/com/myproject/metamodel/type/TypePa ckage.java' is
unmapped
What does "is unmapped" mean?
Regards
Christoph
|
|
|
| Topic: [Xtext] Checks not recognized |
|---|
| [Xtext] Checks not recognized [message #48475] |
Fri, 05 June 2009 09:41 |
Ralf Heydenreich
Messages: 32
Registered: July 2009 |
Member |
|
|
Hi all,
I try to activate the check checkTypeNameStartsWithCapital() in the
default example, but it doesn't work. The check isn' called although the
Validator is bound (in UIModule) and no errors are in the projects. I've
activated the check in MyProjectJavaValidator. Do I have to set some
other stuff? The curious thing is if I start the application in the
debugger it works if I have a breakpoint that let the app wait for some
seconds... Any ideas?
TIA,
Ralf.
|
|
|
| Topic: [Announce] TMF XTEXT 0.7.0RC3 is available |
|---|
|
| Topic: [Announce] TMF XTEXT 0.7.0RC2 is available |
|---|
|
| Topic: [TCS] What features shall be installed by a TCS project? |
|---|
| [TCS] What features shall be installed by a TCS project? [message #43865] |
Fri, 15 May 2009 12:36 |
Mikai Yang
Messages: 140
Registered: July 2009 |
Senior Member |
|
|
Hi,
As we know, there is no binary installation for TCS. So can you give me a hint about what features are needed by TCS and how I can download them from?
By the way, I have read the wiki of TCS; there is a workd quoted as follows:
------------------------------------------------------------ ----------
The source code of TCS plugins is available from Eclipse.org. Binary downloads will be available soon. Note that you need to install ATL as well. The simplest way is to fetch all ATL plugins from
CVS using branch R2_1_0_dev.
------------------------------------------------------------ ----------
I am a newbie of eclipse and have no idea about what is brach R2_1_0_dev and how I can find it. Thanks.
Regards.
Michael.
|
|
|
| Topic: [Announce] TMF XTEXT 0.7.0 I200904271049 is available |
|---|
|
Help
Login
Home
) and this is my situation: