Repository: incubator-systemml
Updated Branches:
refs/heads/master d853d74b2 -> 1b48ca0a9
Fixed incorrect handling of escape characters in the parser
The escape characters were not propogated by the parser to the runtime.
Consider the following script:
print("Foo:\t" + "blah\n---")
The above script will output:
Foo:\tblah\n---
Handling escape characters is especially important when the user wants
to provide tab seperated file. For example:
X = read("testInput.csv", format="csv", sep="\t")
Project: http://git-wip-us.apache.org/repos/asf/incubator-systemml/repo
Commit:
http://git-wip-us.apache.org/repos/asf/incubator-systemml/commit/1b48ca0a
Tree: http://git-wip-us.apache.org/repos/asf/incubator-systemml/tree/1b48ca0a
Diff: http://git-wip-us.apache.org/repos/asf/incubator-systemml/diff/1b48ca0a
Branch: refs/heads/master
Commit: 1b48ca0a91997f1efa72210dbb1d289359f043e6
Parents: d853d74
Author: Niketan Pansare <[email protected]>
Authored: Mon Apr 25 12:16:52 2016 -0700
Committer: Niketan Pansare <[email protected]>
Committed: Mon Apr 25 12:16:52 2016 -0700
----------------------------------------------------------------------
.../parser/common/CommonSyntacticValidator.java | 21 ++++++++++++++------
.../java/org/apache/sysml/parser/dml/Dml.g4 | 2 +-
.../java/org/apache/sysml/parser/pydml/Pydml.g4 | 2 +-
3 files changed, 17 insertions(+), 8 deletions(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/1b48ca0a/src/main/java/org/apache/sysml/parser/common/CommonSyntacticValidator.java
----------------------------------------------------------------------
diff --git
a/src/main/java/org/apache/sysml/parser/common/CommonSyntacticValidator.java
b/src/main/java/org/apache/sysml/parser/common/CommonSyntacticValidator.java
index a7da73a..d7fdc5e 100644
--- a/src/main/java/org/apache/sysml/parser/common/CommonSyntacticValidator.java
+++ b/src/main/java/org/apache/sysml/parser/common/CommonSyntacticValidator.java
@@ -282,16 +282,25 @@ public abstract class CommonSyntacticValidator {
}
}
- protected void constStringIdExpressionHelper(ParserRuleContext ctx,
ExpressionInfo me) {
- String val = "";
- String text = ctx.getText();
+ protected String extractStringInQuotes(String text) {
+ String val = null;
if( (text.startsWith("\"") && text.endsWith("\"")) ||
(text.startsWith("\'") && text.endsWith("\'"))) {
if(text.length() > 2) {
- val = text.substring(1, text.length()-1);
+ val = text.substring(1, text.length()-1)
+ .replaceAll("\\\\b","\b")
+ .replaceAll("\\\\t","\t")
+ .replaceAll("\\\\n","\n")
+ .replaceAll("\\\\f","\f")
+ .replaceAll("\\\\r","\r");
}
}
- else {
+ return val;
+ }
+
+ protected void constStringIdExpressionHelper(ParserRuleContext ctx,
ExpressionInfo me) {
+ String val = extractStringInQuotes(ctx.getText());
+ if(val == null) {
notifyErrorListeners("incorrect string literal ",
ctx.start);
return;
}
@@ -425,7 +434,7 @@ public abstract class CommonSyntacticValidator {
String text = varValue;
if( (text.startsWith("\"") && text.endsWith("\"")) ||
(text.startsWith("\'") && text.endsWith("\'"))) {
if(text.length() > 2) {
- val = text.substring(1, text.length()-1);
+ val = extractStringInQuotes(text);
}
}
else {
http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/1b48ca0a/src/main/java/org/apache/sysml/parser/dml/Dml.g4
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/parser/dml/Dml.g4
b/src/main/java/org/apache/sysml/parser/dml/Dml.g4
index 0ee12c6..31fcec0 100644
--- a/src/main/java/org/apache/sysml/parser/dml/Dml.g4
+++ b/src/main/java/org/apache/sysml/parser/dml/Dml.g4
@@ -210,7 +210,7 @@ COMMANDLINE_POSITION_ID: '$' DIGIT+;
// supports single and double quoted string with escape characters
STRING: '"' ( ESC | ~[\\"] )*? '"' | '\'' ( ESC | ~[\\'] )*? '\'';
-fragment ESC : '\\' [abtnfrv"'\\] ;
+fragment ESC : '\\' [btnfr"'\\] ;
// Comments, whitespaces and new line
LINE_COMMENT : '#' .*? '\r'? '\n' -> skip ;
MULTILINE_BLOCK_COMMENT : '/*' .*? '*/' -> skip ;
http://git-wip-us.apache.org/repos/asf/incubator-systemml/blob/1b48ca0a/src/main/java/org/apache/sysml/parser/pydml/Pydml.g4
----------------------------------------------------------------------
diff --git a/src/main/java/org/apache/sysml/parser/pydml/Pydml.g4
b/src/main/java/org/apache/sysml/parser/pydml/Pydml.g4
index 9072f74..fda5c06 100644
--- a/src/main/java/org/apache/sysml/parser/pydml/Pydml.g4
+++ b/src/main/java/org/apache/sysml/parser/pydml/Pydml.g4
@@ -325,7 +325,7 @@ COMMANDLINE_POSITION_ID: '$' DIGIT+;
// supports single and double quoted string with escape characters
STRING: '"' ( ESC | ~[\\"] )*? '"' | '\'' ( ESC | ~[\\'] )*? '\'';
-fragment ESC : '\\' [abtnfrv"'\\] ;
+fragment ESC : '\\' [btnfr"'\\] ;
// Comments, whitespaces and new line
// LINE_COMMENT : '#' .*? '\r'? '\n' -> skip ;
// MULTILINE_BLOCK_COMMENT : '/*' .*? '*/' -> skip ;
