Eclipse Community Forums: TMF (Xtext) » Possible Unicode Bug

Home » Modeling » TMF (Xtext) » Possible Unicode Bug - Greek Polytonic(Although acute accented letters included in UTF8 range '\u1F00'..'\u1FFF' for Polytonic Greek, XText shows error )

Show: Today's Messages :: Show Polls :: Message Navigator

Possible Unicode Bug - Greek Polytonic [message #894157]

Sat, 07 July 2012 01:55

Kokongi Nei

Messages: 5
Registered: July 2012

Junior Member

The body below contains UTF8 Polytonic Greek.

In order to allow for IDs using polytonic Greek, have defined the following rule in my grammar:

terminal ID: '^'?('a'..'z'|'A'..'Z'|'α'..'ω'|'Α'..'Ω'|'ύ'|'\u1F00'..'\u1FFF')('a'..'z'|'A'..'Z'|'α'..'ω'|'Α'..'Ω'|'\u1F00'..'\u1FFF'|'0'..'9')*
;

The generated grammar correctly handles accented Greek characters in the range of 1F00 to 1FFF except for certain letters with an acute accent mark. Examples include 1F71 ά and 1F7B ύ both of which are within the specified range.

When using an XText editor with the generated grammar, these characters with acute accent marks are marked as an error, and the error message states "extraneous input expecting RULE_STRING"

I have also tried explicitly declaring ύ in the terminal ID rule, but I get the same error.

Is this a bug?

Attachment: Error Message.png
(Size: 270.91KB, Downloaded 182 times)

[Updated on: Sat, 07 July 2012 18:33]

Report message to a moderator

Re: Possible Unicode Bug - Greek Polytonic [message #894217 is a reply to message #894157]

Sat, 07 July 2012 18:30

Jan Koehnlein

Messages: 760
Registered: July 2009
Location: Hamburg

Senior Member

Did you double check the encoding of all files?
We do nothing special in Xtext here, so this could also be a Java or
Antlr issue. A bug with a reproducible example would be welcome.

Am 07.07.12 03:55, schrieb Kokongi Nei:
> The body below contains UTF8 Polytonic Greek.
>
> In order to allow for IDs using polytonic Greek, have defined the
> following rule in my grammar:
>
> terminal ID:
> '^'?('a'..'z'|'A'..'Z'|'α'..'ω'|'Α'..'Ω'|'ύ'|'\u1F00'..'\u1FFF')('a'..'z'|'A'..'Z'|'α'..'ω'|'Α'..'Ω'|'\u1F00'..'\u1FFF'|'0'..'9')*
>
> ;
>
> The generated grammar correctly handles accented Greek characters in the
> range of 1F00 to 1FFF except for certain letters with an acute accent
> mark. Examples include 1F71 ά and 1F7B ύ both of which are within the
> specified range.
>
> When using an XText editor with the generated grammar, these characters
> with acute accent marks are marked as an error, and the error message
> states "extraneous input expecting RULE_STRING"
>
> I have also tried explicitly declaring ύ in the terminal ID rule, but I
> get the same error.
>
> Is this a bug?
>
>

--
Need professional support for Eclipse Modeling?
Go visit: http://xtext.itemis.com

---
Get professional support from the Xtext committers at www.typefox.io

Report message to a moderator