Skip to content

Commit 4d62902

Browse files
snazysnazy
committed
Fix parse error of system default /usr/share/nano/*.nanorc
(Recent) `nano` packages in Ubuntu come with some `.nanorc` files preinstalled. jline's `NanorcParser` sadly fails parsing a couple of the regular expressions. This change translates the regular expressions to Java regular expressions. The differences are described in `org.jline.builtins.SyntaxHighlighter#posixToJavaRegex`: * The first `]` in a bracket expression does not need to be escaped in Posix,translate to `\]`. * Same as above for a negating bracket expression like `[^][]`, translate to `[^\]\[]`. * Any `[` in a bracket expression does not need to be escaped in Posix, translate to `\[`. * Any `]` not in a bracket expression is valid in both Posix and Java, no translation. * A backslash before the closing bracket like `[.f\]` is not an escape of the closing bracket, the backslash needs to be escaped for Java, translate to `[.f\\]`. * Do not perform the above translations within an escape via `\`. * Do not perform the above translations for Posix "classes" like `[[:word:]]` or `[[:digit:]]` and their negation `[-[:word]]`. * Do not perform the above translations for single-bracket Posix classes like `[:digit:]`, and handle the case of single-bracket Posix classes inside bracket expressions, like `[[:digit:]-.]`. Test cases have been added. There are however two regexes that still don't work, but those look invalid. To let jnano not trip over these, any `PatternSyntaxException` lets jnano just ignore the particular rule. A warning is logged in such cases. Fixes jline#1156
1 parent 4504b46 commit 4d62902

File tree

3 files changed

+775
-31
lines changed

3 files changed

+775
-31
lines changed

‎builtins/pom.xml

+5
Original file line numberDiff line numberDiff line change
@@ -47,6 +47,11 @@
4747
<artifactId>junit-jupiter-api</artifactId>
4848
<scope>test</scope>
4949
</dependency>
50+
<dependency>
51+
<groupId>org.junit.jupiter</groupId>
52+
<artifactId>junit-jupiter-params</artifactId>
53+
<scope>test</scope>
54+
</dependency>
5055
</dependencies>
5156

5257
<build>

‎builtins/src/main/java/org/jline/builtins/SyntaxHighlighter.java

+173-31
Original file line numberDiff line numberDiff line change
@@ -656,47 +656,189 @@ private void addHighlightRule(String reference, List<String> parts, boolean case
656656
Styles.StyleCompiler sh = new Styles.StyleCompiler(spec, true);
657657
AttributedStyle style = new StyleResolver(sh::getStyle).resolve("." + reference);
658658

659-
if (HighlightRule.evalRuleType(parts) == HighlightRule.RuleType.PATTERN) {
660-
if (parts.size() == 2) {
661-
highlightRules.get(tokenName).add(new HighlightRule(style, doPattern(".*", caseInsensitive)));
662-
} else {
663-
for (int i = 2; i < parts.size(); i++) {
664-
highlightRules
665-
.get(tokenName)
666-
.add(new HighlightRule(style, doPattern(parts.get(i), caseInsensitive)));
659+
try {
660+
if (HighlightRule.evalRuleType(parts) == HighlightRule.RuleType.PATTERN) {
661+
if (parts.size() == 2) {
662+
highlightRules.get(tokenName).add(new HighlightRule(style, doPattern(".*", caseInsensitive)));
663+
} else {
664+
for (int i = 2; i < parts.size(); i++) {
665+
highlightRules
666+
.get(tokenName)
667+
.add(new HighlightRule(style, doPattern(parts.get(i), caseInsensitive)));
668+
}
667669
}
670+
} else if (HighlightRule.evalRuleType(parts) == HighlightRule.RuleType.START_END) {
671+
String s = parts.get(2);
672+
String e = parts.get(3);
673+
highlightRules
674+
.get(tokenName)
675+
.add(new HighlightRule(
676+
style,
677+
doPattern(s.substring(7, s.length() - 1), caseInsensitive),
678+
doPattern(e.substring(5, e.length() - 1), caseInsensitive)));
679+
} else if (HighlightRule.evalRuleType(parts) == HighlightRule.RuleType.PARSER_START_WITH) {
680+
highlightRules
681+
.get(tokenName)
682+
.add(new HighlightRule(
683+
HighlightRule.RuleType.PARSER_START_WITH,
684+
style,
685+
parts.get(2).substring(10)));
686+
} else if (HighlightRule.evalRuleType(parts) == HighlightRule.RuleType.PARSER_CONTINUE_AS) {
687+
highlightRules
688+
.get(tokenName)
689+
.add(new HighlightRule(
690+
HighlightRule.RuleType.PARSER_CONTINUE_AS,
691+
style,
692+
parts.get(2).substring(11)));
668693
}
669-
} else if (HighlightRule.evalRuleType(parts) == HighlightRule.RuleType.START_END) {
670-
String s = parts.get(2);
671-
String e = parts.get(3);
672-
highlightRules
673-
.get(tokenName)
674-
.add(new HighlightRule(
675-
style,
676-
doPattern(s.substring(7, s.length() - 1), caseInsensitive),
677-
doPattern(e.substring(5, e.length() - 1), caseInsensitive)));
678-
} else if (HighlightRule.evalRuleType(parts) == HighlightRule.RuleType.PARSER_START_WITH) {
679-
highlightRules
680-
.get(tokenName)
681-
.add(new HighlightRule(
682-
HighlightRule.RuleType.PARSER_START_WITH,
683-
style,
684-
parts.get(2).substring(10)));
685-
} else if (HighlightRule.evalRuleType(parts) == HighlightRule.RuleType.PARSER_CONTINUE_AS) {
686-
highlightRules
687-
.get(tokenName)
688-
.add(new HighlightRule(
689-
HighlightRule.RuleType.PARSER_CONTINUE_AS,
690-
style,
691-
parts.get(2).substring(11)));
694+
} catch (PatternSyntaxException e) {
695+
Log.warn("Invalid highlight regex", reference, parts, e);
696+
} catch (Exception e) {
697+
Log.warn("Failure while handling highlight regex", reference, parts, e);
692698
}
693699
}
694700

695701
private Pattern doPattern(String regex, boolean caseInsensitive) {
702+
regex = posixToJavaRegex(regex);
696703
return caseInsensitive ? Pattern.compile(regex, Pattern.CASE_INSENSITIVE) : Pattern.compile(regex);
697704
}
698705
}
699706

707+
/**
708+
* Posix regex is different from Java regex. This function parses the given Posix regex and escapes according to these rules:
709+
*
710+
* <p>The first {@code ]} in a bracket expression does not need to be escaped in Posix,translate to {@code \]}.
711+
*
712+
* <p>Same as above for a negating bracket expression like {@code [^][]}, translate to {@code [^\]\[]}.
713+
*
714+
* <p>Any {@code [} in a bracket expression does not need to be escaped in Posix, translate to {@code \[}.
715+
*
716+
* <p>Any {@code ]} not in a bracket expression is valid in both Posix and Java, no translation.
717+
*
718+
* <p>A backslash before the closing bracket like {@code [.f\]} is not an escape of the closing bracket,
719+
* the backslash needs to be escaped for Java, translate to {@code [.f\\]}.
720+
*
721+
* <p>Do not perform the above translations within an escape via {@code \}.
722+
*
723+
* <p>Do not perform the above translations for Posix "classes" like {@code [[:word:]]} or {@code [[:digit:]]}
724+
* and their negation {@code [-[:word]]}.
725+
*
726+
* <p>Do not perform the above translations for single-bracket Posix classes like {@code [:digit:]},
727+
* and handle the case of single-bracket Posix classes inside bracket expressions, like
728+
* @code {[[:digit:]-.]}.
729+
*
730+
* @param posix Posix regex
731+
* @return Java regex
732+
*/
733+
static String posixToJavaRegex(String posix) {
734+
int len = posix.length();
735+
StringBuilder java = new StringBuilder();
736+
737+
boolean inBracketExpression = false;
738+
739+
int i = 0;
740+
char next;
741+
try {
742+
for (; i < len; i++) {
743+
char c = posix.charAt(i);
744+
745+
switch (c) {
746+
case '\\':
747+
next = posix.charAt(++i);
748+
// Don't translate anything after the \ character escape
749+
if (inBracketExpression && next == ']') {
750+
inBracketExpression = false;
751+
java.append("\\\\").append(next);
752+
} else {
753+
java.append(c).append(next);
754+
}
755+
break;
756+
case '[':
757+
if (i == len - 1) {
758+
throw new IllegalArgumentException("Lone [ at the end of (index " + i + "): " + posix);
759+
}
760+
// Handle "double bracket" Posix "classes" like [[:word:]] or [[:digit:]] and their negations
761+
// starting with [-[:
762+
if (posix.regionMatches(i, "[[:", 0, 3) || posix.regionMatches(i, "[-[:", 0, 4)) {
763+
int afterClass = nextAfterClass(posix, i + 3);
764+
if (posix.regionMatches(afterClass, ":]]", 0, 3)) {
765+
java.append(posix, i, afterClass + 3);
766+
i = afterClass + 2;
767+
break;
768+
} else if (posix.regionMatches(afterClass, ":]", 0, 2)) {
769+
if (inBracketExpression) {
770+
throw new IllegalArgumentException("Unclear bracket expression");
771+
}
772+
// Handles character patterns like [[:alpha:]_-]
773+
java.append(posix, i, afterClass + 2);
774+
i = afterClass + 1;
775+
inBracketExpression = true;
776+
break;
777+
} else {
778+
throw new IllegalArgumentException("Invalid character class");
779+
}
780+
}
781+
// Handle "single bracket" Posix "classes" like [:word:]
782+
else if (posix.charAt(i + 1) == ':') {
783+
int afterClass = nextAfterClass(posix, i + 2);
784+
if (!posix.regionMatches(afterClass, ":]", 0, 2)) {
785+
java.append("[:");
786+
i++;
787+
inBracketExpression = true;
788+
} else {
789+
java.append(posix, i, afterClass + 2);
790+
i = afterClass + 1;
791+
}
792+
break;
793+
}
794+
if (inBracketExpression) {
795+
// Translate lone [ to \[
796+
java.append('\\').append(c);
797+
} else {
798+
inBracketExpression = true;
799+
java.append(c);
800+
next = posix.charAt(i + 1);
801+
if (next == ']') {
802+
i++;
803+
java.append("\\]");
804+
} else if (next == '^' && posix.charAt(i + 2) == ']') {
805+
i += 2;
806+
java.append("^\\]");
807+
}
808+
}
809+
break;
810+
case ']':
811+
if (inBracketExpression) {
812+
inBracketExpression = false;
813+
}
814+
java.append(c);
815+
break;
816+
default:
817+
java.append(c);
818+
break;
819+
}
820+
}
821+
} catch (Exception e) {
822+
throw new IllegalArgumentException(
823+
"Posix-to-Java regex translation failed around index " + i + " of: " + posix, e);
824+
}
825+
return java.toString();
826+
}
827+
828+
private static int nextAfterClass(String s, int idx) {
829+
if (s.charAt(idx) == ':') {
830+
idx++;
831+
}
832+
while (true) {
833+
char c = s.charAt(idx);
834+
if (!Character.isLetterOrDigit(c)) {
835+
break;
836+
}
837+
idx++;
838+
}
839+
return idx;
840+
}
841+
700842
protected static class RuleSplitter {
701843
protected static List<String> split(String s) {
702844
List<String> out = new ArrayList<>();

0 commit comments

Comments
 (0)