[
https://issues.apache.org/jira/browse/UIMA-3598?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Martin Toepfer updated UIMA-3598:
---------------------------------
Description:
text span:
"0.3-0.7"
rules:
{noformat}
DECLARE Number;
DECLARE NumberExpression;
NUM {-PARTOF(Number)} ((COMMA | PERIOD) NUM)? {->MARK(Number,1,2)};
((Number? ("<" | ">" | "=" | "==" | "-" | "+"))? {-PARTOF(NumberExpression)}
@Number) {-PARTOF(NumberExpression)->MARK(NumberExpression)};
{noformat}
result:
-0.7 annotated as number expression
expected:
0.3 as number expression
[or 0.3-0.7 if manual anchoring tries to match at the end of the document
first... depends on rule inference]
the rules should not be used for parsing of number expressions, but the
behaviour should be consistent with the expectations for the rule inference
was:
text span:
"0.3-0.7"
rules:
{noformat}
DECLARE Number;
DECLARE NumberExpression;
NUM {-PARTOF(Number)} ((COMMA | PERIOD) NUM)? {->MARK(Number,1,2)};
((Number? ("<" | ">" | "=" | "==" | "-" | "+"))? {-PARTOF(NumberExpression)}
@Number) {-PARTOF(NumberExpression)->MARK(NumberExpression)};
{noformat}
result:
-0.7 annotated as number expression
expected:
0.3 as number expression
[or 0.3-0.7 if manual anchoring tries to match at the end of the document
first... depends on rule inference]
> Ruta: matching with manual anchor
> ---------------------------------
>
> Key: UIMA-3598
> URL: https://issues.apache.org/jira/browse/UIMA-3598
> Project: UIMA
> Issue Type: Bug
> Components: ruta
> Reporter: Martin Toepfer
> Assignee: Peter Klügl
>
> text span:
> "0.3-0.7"
> rules:
> {noformat}
> DECLARE Number;
> DECLARE NumberExpression;
> NUM {-PARTOF(Number)} ((COMMA | PERIOD) NUM)? {->MARK(Number,1,2)};
> ((Number? ("<" | ">" | "=" | "==" | "-" | "+"))? {-PARTOF(NumberExpression)}
> @Number) {-PARTOF(NumberExpression)->MARK(NumberExpression)};
> {noformat}
> result:
> -0.7 annotated as number expression
> expected:
> 0.3 as number expression
> [or 0.3-0.7 if manual anchoring tries to match at the end of the document
> first... depends on rule inference]
> the rules should not be used for parsing of number expressions, but the
> behaviour should be consistent with the expectations for the rule inference
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
