LexParseException: No viable alternative at input `(Regex(#"[ -퟿-�𐀀-􏿿\r\n\t]`

[masp]

Version

Documents/cfnpkl » pkl -v
Pkl 0.25.1 (macOS 14.1, native)

break.pkl

A = #"[ -퟿-�𐀀-􏿿
	]*"#

See playground: https://pkl-playground.vercel.app/?share=individual-pure-stomach

Expected behavior

I would expect using #"..."# would avoid any such issues with lexing, even if the characters are unprintable.

Here's a stack trace from pkl project package:

An unexpected error has occurred. Would you mind filing a bug report?

org.pkl.core.parser.LexParseException$ParseError: No viable alternative at input `(Regex(#"[ -퟿-�𐀀-􏿿\r\n\t]`.
        at org.pkl.core.parser.Parser$1.syntaxError(Parser.java:201)
        at org.pkl.core.parser.Parser$1.syntaxError(Parser.java:180)
        at org.antlr.v4.runtime.ProxyErrorListener.syntaxError(ProxyErrorListener.java:44)
        at org.antlr.v4.runtime.Parser.notifyErrorListeners(Parser.java:543)
        at org.pkl.core.parser.ErrorStrategy.reportNoViableAlternative(ErrorStrategy.java:41)
        at org.antlr.v4.runtime.DefaultErrorStrategy.reportError(DefaultErrorStrategy.java:139)
        at org.pkl.core.parser.antlr.PklParser.expr(PklParser.java:4159)
        at org.pkl.core.parser.antlr.PklParser.expr(PklParser.java:3996)
        at org.pkl.core.parser.antlr.PklParser.type(PklParser.java:2614)
        at org.pkl.core.parser.antlr.PklParser.typeAnnotation(PklParser.java:1951)
        at org.pkl.core.parser.antlr.PklParser.classProperty(PklParser.java:1360)
        at org.pkl.core.parser.antlr.PklParser.classBody(PklParser.java:1073)
        at org.pkl.core.parser.antlr.PklParser.clazz(PklParser.java:859)
        at org.pkl.core.parser.antlr.PklParser.module(PklParser.java:454)
        at org.pkl.core.parser.Parser.parseProduction(Parser.java:91)
        at org.pkl.core.parser.Parser.parseModule(Parser.java:44)
        at org.pkl.core.parser.Parser.parseModule(Parser.java:49)
        at org.pkl.core.ast.builder.ImportsAndReadsParser.parse(ImportsAndReadsParser.java:59)
        at org.pkl.core.project.ProjectPackager.getImportsAndReads(ProjectPackager.java:423)
        at org.pkl.core.project.ProjectPackager.validateImportsAndReads(ProjectPackager.java:381)
        at org.pkl.core.project.ProjectPackager.validatePklImportsAndReads(ProjectPackager.java:324)
        at org.pkl.core.project.ProjectPackager.doPackage(ProjectPackager.java:149)
        at org.pkl.core.project.ProjectPackager.createPackages(ProjectPackager.java:122)
        at org.pkl.cli.CliProjectPackager.doRun(CliProjectPackager.kt:88)
        at org.pkl.commons.cli.CliCommand.run(CliCommand.kt:45)
        at org.pkl.cli.commands.ProjectCommand$Companion$PackageCommand.run(ProjectCommand.kt:145)
        at com.github.ajalt.clikt.parsers.Parser.parse(Parser.kt:198)
        at com.github.ajalt.clikt.parsers.Parser.parse(Parser.kt:211)
        at com.github.ajalt.clikt.parsers.Parser.parse(Parser.kt:211)
        at com.github.ajalt.clikt.parsers.Parser.parse(Parser.kt:18)
        at com.github.ajalt.clikt.core.CliktCommand.parse(CliktCommand.kt:400)
        at com.github.ajalt.clikt.core.CliktCommand.parse$default(CliktCommand.kt:397)
        at com.github.ajalt.clikt.core.CliktCommand.main(CliktCommand.kt:415)
        at com.github.ajalt.clikt.core.CliktCommand.main(CliktCommand.kt:440)
        at org.pkl.cli.Main$main$1.invoke(Main.kt:52)
        at org.pkl.cli.Main$main$1.invoke(Main.kt:40)
        at org.pkl.commons.cli.CliMainKt.cliMain(CliMain.kt:31)
        at org.pkl.cli.Main.main(Main.kt:40)

[masp]

Looking at the language spec more, I realize the issue is the \n in the raw literal. To do a multiline raw literal, you need to use #"""..."""# instead. I'm leaving this open in case the error message can be improved to match what's reported with pkl eval.

[stackoverflow]

We plan to move out of ANTLR at some point which should improve parser errors. You are right that the message could be better.

[bioball]

I'm going to close this as it's behaving correctly. We will improve these error messages when we manage to get off of ANTLR as our parser as Islon mentioned.