Change the handling of invalid single quote to improve error messages. (#8370)

* add value_not_exposed report

* Change handling of invalid single-quote.

* fix typo

src/canonicalize/Can.zig

@@ -2943,7 +2943,8 @@ fn canonicalizeDeclWithAnnotation(
 
 fn parseSingleQuoteCodepoint(
     inner_text: []const u8,
-) ?u21 {
+) u21 {
+    // tokenizer checks for valid single quote codepoints, so every error case is unreachable here
     const escaped = inner_text[0] == '\\';
 
     if (escaped) {
@@ -2951,13 +2952,9 @@ fn parseSingleQuoteCodepoint(
         switch (c) {
             'u' => {
                 const hex_code = inner_text[3 .. inner_text.len - 1];
-                const codepoint = std.fmt.parseInt(u21, hex_code, 16) catch {
+                const codepoint = std.fmt.parseInt(u21, hex_code, 16) catch unreachable;
-                    return null;
-                };
 
-                if (!std.unicode.utf8ValidCodepoint(codepoint)) {
+                std.debug.assert(std.unicode.utf8ValidCodepoint(codepoint));
-                    return null;
-                }
 
                 return codepoint;
             },
@@ -2973,26 +2970,16 @@ fn parseSingleQuoteCodepoint(
             't' => {
                 return '\t';
             },
-            else => {
+            else => unreachable,
-                return null;
-            },
         }
     } else {
-        const view = std.unicode.Utf8View.init(inner_text) catch |err| switch (err) {
+        const view = std.unicode.Utf8View.init(inner_text) catch unreachable;
-            error.InvalidUtf8 => {
-                return null;
-            },
-        };
 
         var iterator = view.iterator();
 
-        if (iterator.nextCodepoint()) |codepoint| {
+        const codepoint = iterator.nextCodepoint().?;
-            std.debug.assert(iterator.nextCodepoint() == null);
+        std.debug.assert(iterator.nextCodepoint() == null);
-            return codepoint;
+        return codepoint;
-        } else {
-            // only single valid utf8 codepoint can be here after tokenization
-            unreachable;
-        }
     }
 }
 
@@ -3034,34 +3021,30 @@ fn canonicalizeSingleQuote(
 
     // Resolve to a string slice from the source
     const token_text = self.parse_ir.resolve(token);
+    std.debug.assert(token_text[0] == '\'' and token_text[token_text.len - 1] == '\'');
 
-    if (parseSingleQuoteCodepoint(token_text[1 .. token_text.len - 1])) |codepoint| {
+    const codepoint = parseSingleQuoteCodepoint(token_text[1 .. token_text.len - 1]);
-        const value_content = CIR.IntValue{
+    const value_content = CIR.IntValue{
-            .bytes = @bitCast(@as(u128, @intCast(codepoint))),
+        .bytes = @bitCast(@as(u128, @intCast(codepoint))),
-            .kind = .u128,
+        .kind = .u128,
-        };
+    };
-        if (comptime Idx == Expr.Idx) {
+    if (comptime Idx == Expr.Idx) {
-            const expr_idx = try self.env.addExpr(CIR.Expr{
+        const expr_idx = try self.env.addExpr(CIR.Expr{
-                .e_num = .{
+            .e_num = .{
-                    .value = value_content,
-                    .kind = .int_unbound,
-                },
-            }, region);
-            return expr_idx;
-        } else if (comptime Idx == Pattern.Idx) {
-            const pat_idx = try self.env.addPattern(Pattern{ .num_literal = .{
                 .value = value_content,
                 .kind = .int_unbound,
-            } }, region);
+            },
-            return pat_idx;
+        }, region);
-        } else {
+        return expr_idx;
-            @compileError("Unsupported Idx type");
+    } else if (comptime Idx == Pattern.Idx) {
-        }
+        const pat_idx = try self.env.addPattern(Pattern{ .num_literal = .{
+            .value = value_content,
+            .kind = .int_unbound,
+        } }, region);
+        return pat_idx;
+    } else {
+        @compileError("Unsupported Idx type");
     }
-
-    return try self.env.pushMalformed(Idx, Diagnostic{ .invalid_single_quote = .{
-        .region = region,
-    } });
 }
 
 fn canonicalizeRecordField(

src/canonicalize/Diagnostic.zig

@@ -30,9 +30,6 @@ pub const Diagnostic = union(enum) {
     invalid_num_literal: struct {
         region: Region,
     },
-    invalid_single_quote: struct {
-        region: Region,
-    },
     empty_tuple: struct {
         region: Region,
     },

src/canonicalize/ModuleEnv.zig

@@ -954,7 +954,7 @@ pub fn diagnosticToReport(self: *Self, diagnostic: CIR.Diagnostic, allocator: st
             // Format the message to match origin/main
             try report.document.addText("The type ");
             try report.document.addInlineCode(type_name);
-            try report.document.addReflowingText(" is not an exposed by the module ");
+            try report.document.addReflowingText(" is not exposed by the module ");
             try report.document.addInlineCode(module_name);
             try report.document.addReflowingText(".");
             try report.document.addLineBreak();
@@ -973,6 +973,33 @@ pub fn diagnosticToReport(self: *Self, diagnostic: CIR.Diagnostic, allocator: st
 
             break :blk report;
         },
+        .value_not_exposed => |data| blk: {
+            const region_info = self.calcRegionInfo(data.region);
+
+            var report = Report.init(allocator, "VALUE NOT EXPOSED", .runtime_error);
+
+            // Format the message to match origin/main
+            try report.document.addText("The value ");
+            try report.document.addInlineCode(self.getIdent(data.value_name));
+            try report.document.addReflowingText(" is not exposed by the module ");
+            try report.document.addInlineCode(self.getIdent(data.module_name));
+            try report.document.addReflowingText(".");
+            try report.document.addLineBreak();
+            try report.document.addLineBreak();
+
+            try report.document.addReflowingText("You're attempting to use this value here:");
+            try report.document.addLineBreak();
+            const owned_filename = try report.addOwnedString(filename);
+            try report.document.addSourceRegion(
+                region_info,
+                .error_highlight,
+                owned_filename,
+                self.getSourceAll(),
+                self.getLineStartsAll(),
+            );
+
+            break :blk report;
+        },
         .module_not_found => |data| blk: {
             const region_info = self.calcRegionInfo(data.region);
 

src/canonicalize/Node.zig

@@ -151,7 +151,6 @@ pub const Tag = enum {
     // diagnostic indices stored in malformed nodes.
     diag_not_implemented,
     diag_invalid_num_literal,
-    diag_invalid_single_quote,
     diag_empty_single_quote,
     diag_empty_tuple,
     diag_ident_already_in_scope,

src/canonicalize/NodeStore.zig

@@ -128,7 +128,7 @@ pub fn deinit(store: *NodeStore) void {
 /// when adding/removing variants from ModuleEnv unions. Update these when modifying the unions.
 ///
 /// Count of the diagnostic nodes in the ModuleEnv
-pub const MODULEENV_DIAGNOSTIC_NODE_COUNT = 59;
+pub const MODULEENV_DIAGNOSTIC_NODE_COUNT = 58;
 /// Count of the expression nodes in the ModuleEnv
 pub const MODULEENV_EXPR_NODE_COUNT = 35;
 /// Count of the statement nodes in the ModuleEnv
@@ -2670,10 +2670,6 @@ pub fn addDiagnostic(store: *NodeStore, reason: CIR.Diagnostic) Allocator.Error!
             node.tag = .diag_invalid_num_literal;
             region = r.region;
         },
-        .invalid_single_quote => |r| {
-            node.tag = .diag_invalid_single_quote;
-            region = r.region;
-        },
         .empty_tuple => |r| {
             node.tag = .diag_empty_tuple;
             region = r.region;
@@ -3029,9 +3025,6 @@ pub fn getDiagnostic(store: *const NodeStore, diagnostic: CIR.Diagnostic.Idx) CI
         .diag_invalid_num_literal => return CIR.Diagnostic{ .invalid_num_literal = .{
             .region = store.getRegionAt(node_idx),
         } },
-        .diag_invalid_single_quote => return CIR.Diagnostic{ .invalid_single_quote = .{
-            .region = store.getRegionAt(node_idx),
-        } },
         .diag_empty_tuple => return CIR.Diagnostic{ .empty_tuple = .{
             .region = store.getRegionAt(node_idx),
         } },

src/canonicalize/test/node_store_test.zig

@@ -686,12 +686,6 @@ test "NodeStore round trip - Diagnostics" {
         },
     });
 
-    try diagnostics.append(gpa, CIR.Diagnostic{
-        .invalid_single_quote = .{
-            .region = rand_region(),
-        },
-    });
-
     try diagnostics.append(gpa, CIR.Diagnostic{
         .f64_pattern_literal = .{
             .region = rand_region(),

src/parse/AST.zig

@@ -142,6 +142,9 @@ pub fn tokenizeDiagnosticToReport(self: *AST, diagnostic: tokenize.Diagnostic, a
         .NonPrintableUnicodeInStrLiteral => "NON-PRINTABLE UNICODE IN STRING-LIKE LITERAL",
         .InvalidUtf8InSource => "INVALID UTF-8",
         .DollarInMiddleOfIdentifier => "STRAY DOLLAR SIGN",
+        .SingleQuoteTooLong => "SINGLE QUOTE TOO LONG",
+        .SingleQuoteEmpty => "SINGLE QUOTE EMPTY",
+        .SingleQuoteUnclosed => "UNCLOSED SINGLE QUOTE",
     };
 
     const body = switch (diagnostic.tag) {
@@ -155,6 +158,8 @@ pub fn tokenizeDiagnosticToReport(self: *AST, diagnostic: tokenize.Diagnostic, a
         .NonPrintableUnicodeInStrLiteral => "Non-printable Unicode characters are not allowed in string-like literals.",
         .InvalidUtf8InSource => "Invalid UTF-8 encoding found in source code. Roc source files must be valid UTF-8.",
         .DollarInMiddleOfIdentifier => "Dollar sign ($) is only allowed at the very beginning of a name, not in the middle or at the end.",
+        .SingleQuoteTooLong, .SingleQuoteEmpty => "Single-quoted literals must contain exactly one valid UTF-8 codepoint.",
+        .SingleQuoteUnclosed => "This single-quoted literal is missing a closing quote.",
     };
 
     var report = reporting.Report.init(allocator, title, .runtime_error);

src/parse/tokenize.zig

@@ -50,10 +50,7 @@ pub const Token = struct {
         StringPart,
         MalformedStringPart, // malformed, but should be treated similar to a StringPart in the parser
         SingleQuote,
-        MalformedSingleQuoteUnclosed, // malformed, but should be treated similar to a SingleQuote in the parser
+        MalformedSingleQuote, // malformed, but should be treated similar to a SingleQuote in the parser
-        MalformedSingleQuoteEmpty, // malformed, but should be treated similar to a SingleQuote in the parser
-        MalformedSingleQuoteTooLong, // malformed, but should be treated similar to a SingleQuote in the parser
-        MalformedSingleQuoteInvalidEscapeSequence, // malformed, but should be treated similar to a SingleQuote in the parser
         Int,
         MalformedNumberBadSuffix, // malformed, but should be treated similar to an int in the parser
         MalformedNumberUnicodeSuffix, // malformed, but should be treated similar to an int in the parser
@@ -315,10 +312,7 @@ pub const Token = struct {
                 .MalformedOpaqueNameWithoutName,
                 .MalformedUnicodeIdent,
                 .MalformedUnknownToken,
-                .MalformedSingleQuoteUnclosed,
+                .MalformedSingleQuote,
-                .MalformedSingleQuoteEmpty,
-                .MalformedSingleQuoteTooLong,
-                .MalformedSingleQuoteInvalidEscapeSequence,
                 .MalformedStringPart,
                 => true,
             };
@@ -486,6 +480,9 @@ pub const Diagnostic = struct {
         NonPrintableUnicodeInStrLiteral,
         InvalidUtf8InSource,
         DollarInMiddleOfIdentifier,
+        SingleQuoteTooLong,
+        SingleQuoteEmpty,
+        SingleQuoteUnclosed,
     };
 };
 
@@ -924,6 +921,16 @@ pub const Cursor = struct {
                         return error.InvalidUnicodeEscapeSequence;
                     }
                 }
+                const hex_code = self.buf[hex_start .. self.pos - 1];
+                const codepoint = std.fmt.parseInt(u21, hex_code, 16) catch {
+                    self.pushMessage(.InvalidUnicodeEscapeSequence, escape_start, self.pos);
+                    return error.InvalidUnicodeEscapeSequence;
+                };
+
+                if (!std.unicode.utf8ValidCodepoint(codepoint)) {
+                    self.pushMessage(.InvalidUnicodeEscapeSequence, escape_start, self.pos);
+                    return error.InvalidUnicodeEscapeSequence;
+                }
             },
             else => {
                 // Include the character after the backslash in the error region
@@ -941,7 +948,9 @@ pub const Cursor = struct {
             TooLong,
             Invalid,
         };
+
         std.debug.assert(self.peek() == '\'');
+        const start = self.pos;
 
         // Skip the initial quote.
         self.pos += 1;
@@ -959,7 +968,8 @@ pub const Cursor = struct {
             switch (state) {
                 .Empty => switch (c) {
                     '\'' => {
-                        return .MalformedSingleQuoteEmpty;
+                        self.pushMessage(.SingleQuoteEmpty, start, self.pos);
+                        return .MalformedSingleQuote;
                     },
                     '\\' => {
                         state = .Enough;
@@ -983,20 +993,22 @@ pub const Cursor = struct {
                 },
                 .TooLong => switch (c) {
                     '\'' => {
-                        return .MalformedSingleQuoteTooLong;
+                        self.pushMessage(.SingleQuoteTooLong, start, self.pos);
+                        return .MalformedSingleQuote;
                     },
                     else => {},
                 },
                 .Invalid => switch (c) {
                     '\'' => {
-                        return .MalformedSingleQuoteInvalidEscapeSequence;
+                        return .MalformedSingleQuote;
                     },
                     else => {},
                 },
             }
         }
 
-        return .MalformedSingleQuoteUnclosed;
+        self.pushMessage(.SingleQuoteUnclosed, start, self.pos);
+        return .MalformedSingleQuote;
     }
 
     /// Chomps a UTF-8 codepoint and advances the cursor position.
@@ -2277,10 +2289,7 @@ fn rebuildBufferForTesting(buf: []const u8, tokens: *TokenizedBuffer, alloc: std
             .MalformedNamedUnderscoreUnicode,
             .MalformedOpaqueNameUnicode,
             .MalformedOpaqueNameWithoutName,
-            .MalformedSingleQuoteEmpty,
+            .MalformedSingleQuote,
-            .MalformedSingleQuoteTooLong,
-            .MalformedSingleQuoteUnclosed,
-            .MalformedSingleQuoteInvalidEscapeSequence,
             .MalformedStringPart,
             => {
                 return error.Unsupported;

test/snapshots/expr/unicode_overflow_str.md

@@ -8,27 +8,34 @@ type=expr
 "\u(FFFFFF)"
 ~~~
 # EXPECTED
-NIL
+INVALID UNICODE ESCAPE SEQUENCE - unicode_overflow_str.md:1:2:1:12
 # PROBLEMS
-NIL
+**INVALID UNICODE ESCAPE SEQUENCE**
+This Unicode escape sequence is not valid.
+
+**unicode_overflow_str.md:1:2:1:12:**
+```roc
+"\u(FFFFFF)"
+```
+ ^^^^^^^^^^
+
+
 # TOKENS
 ~~~zig
-StringStart,StringPart,StringEnd,
+StringStart,MalformedStringPart,StringEnd,
 EndOfFile,
 ~~~
 # PARSE
 ~~~clojure
-(e-string
+(e-string)
-	(e-string-part (raw "\u(FFFFFF)")))
 ~~~
 # FORMATTED
 ~~~roc
-NO CHANGE
+""
 ~~~
 # CANONICALIZE
 ~~~clojure
-(e-string
+(e-string)
-	(e-literal (string "\u(FFFFFF)")))
 ~~~
 # TYPES
 ~~~clojure

test/snapshots/fuzz_crash/fuzz_crash_031.md

@@ -11,6 +11,7 @@ mule []
 vavar t= '
 ~~~
 # EXPECTED
+UNCLOSED SINGLE QUOTE - fuzz_crash_031.md:4:10:4:11
 PARSE ERROR - fuzz_crash_031.md:1:1:1:5
 PARSE ERROR - fuzz_crash_031.md:1:6:1:7
 PARSE ERROR - fuzz_crash_031.md:1:7:1:8
@@ -19,6 +20,16 @@ UNEXPECTED TOKEN IN EXPRESSION - fuzz_crash_031.md:4:10:4:11
 UNRECOGNIZED SYNTAX - fuzz_crash_031.md:4:10:4:11
 MISSING MAIN! FUNCTION - fuzz_crash_031.md:1:1:4:11
 # PROBLEMS
+**UNCLOSED SINGLE QUOTE**
+This single-quoted literal is missing a closing quote.
+
+**fuzz_crash_031.md:4:10:4:11:**
+```roc
+vavar t= '
+```
+         ^
+
+
 **PARSE ERROR**
 A parsing error occurred: `statement_unexpected_token`
 This is an unexpected parsing error. Please check your syntax.
@@ -104,7 +115,7 @@ vavar t= '
 # TOKENS
 ~~~zig
 LowerIdent,OpenSquare,CloseSquare,
-LowerIdent,LowerIdent,OpAssign,MalformedSingleQuoteUnclosed,
+LowerIdent,LowerIdent,OpAssign,MalformedSingleQuote,
 EndOfFile,
 ~~~
 # PARSE

test/snapshots/fuzz_crash/fuzz_crash_039.md

@@ -9,9 +9,20 @@ module[}('
 )
 ~~~
 # EXPECTED
+UNCLOSED SINGLE QUOTE - fuzz_crash_039.md:1:10:1:11
 PARSE ERROR - fuzz_crash_039.md:1:8:1:9
 PARSE ERROR - fuzz_crash_039.md:3:1:3:1
 # PROBLEMS
+**UNCLOSED SINGLE QUOTE**
+This single-quoted literal is missing a closing quote.
+
+**fuzz_crash_039.md:1:10:1:11:**
+```roc
+module[}('
+```
+         ^
+
+
 **PARSE ERROR**
 A parsing error occurred: `exposed_item_unexpected_token`
 This is an unexpected parsing error. Please check your syntax.
@@ -36,7 +47,7 @@ This is an unexpected parsing error. Please check your syntax.
 
 # TOKENS
 ~~~zig
-KwModule,OpenSquare,CloseCurly,NoSpaceOpenRound,MalformedSingleQuoteUnclosed,
+KwModule,OpenSquare,CloseCurly,NoSpaceOpenRound,MalformedSingleQuote,
 CloseRound,
 EndOfFile,
 ~~~

test/snapshots/unicode_single_quotes.md

@@ -14,6 +14,7 @@ x = (
     '\u(',
     '\u()',
     '\u(1F680)',
+    '\u(EDA0B5)'
     '\u(K)',
     '\\',
     '\'',
@@ -32,27 +33,24 @@ INVALID UNICODE ESCAPE SEQUENCE - unicode_single_quotes.md:5:6:5:8
 INVALID UNICODE ESCAPE SEQUENCE - unicode_single_quotes.md:6:6:6:8
 INVALID UNICODE ESCAPE SEQUENCE - unicode_single_quotes.md:7:6:7:9
 INVALID UNICODE ESCAPE SEQUENCE - unicode_single_quotes.md:8:6:8:10
-INVALID UNICODE ESCAPE SEQUENCE - unicode_single_quotes.md:10:6:10:11
+INVALID UNICODE ESCAPE SEQUENCE - unicode_single_quotes.md:10:6:10:16
-INVALID ESCAPE SEQUENCE - unicode_single_quotes.md:21:2:22:1
+INVALID UNICODE ESCAPE SEQUENCE - unicode_single_quotes.md:11:6:11:11
+SINGLE QUOTE EMPTY - unicode_single_quotes.md:14:5:14:7
+SINGLE QUOTE TOO LONG - unicode_single_quotes.md:15:5:15:11
+UNCLOSED SINGLE QUOTE - unicode_single_quotes.md:16:5:16:9
+UNCLOSED SINGLE QUOTE - unicode_single_quotes.md:19:5:19:7
+INVALID ESCAPE SEQUENCE - unicode_single_quotes.md:22:2:23:1
+UNCLOSED SINGLE QUOTE - unicode_single_quotes.md:22:1:22:3
 UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:5:5:5:9
 UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:6:5:6:10
 UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:7:5:7:10
 UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:8:5:8:11
-UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:10:5:10:12
+UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:10:5:10:17
-UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:13:5:13:7
+PARSE ERROR - unicode_single_quotes.md:17:1:17:2
-UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:14:5:14:11
+UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:19:5:19:7
-UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:15:5:15:9
+PARSE ERROR - unicode_single_quotes.md:22:1:22:3
-UNEXPECTED TOKEN IN EXPRESSION - unicode_single_quotes.md:18:5:18:7
+UNRECOGNIZED SYNTAX - unicode_single_quotes.md:17:1:17:2
-PARSE ERROR - unicode_single_quotes.md:21:1:21:3
+UNRECOGNIZED SYNTAX - unicode_single_quotes.md:19:5:19:7
-INVALID TUPLE ELEMENT - :0:0:0:0
-INVALID TUPLE ELEMENT - :0:0:0:0
-INVALID TUPLE ELEMENT - :0:0:0:0
-INVALID TUPLE ELEMENT - :0:0:0:0
-INVALID TUPLE ELEMENT - :0:0:0:0
-INVALID TUPLE ELEMENT - :0:0:0:0
-INVALID TUPLE ELEMENT - :0:0:0:0
-INVALID TUPLE ELEMENT - :0:0:0:0
-UNRECOGNIZED SYNTAX - unicode_single_quotes.md:18:5:18:7
 # PROBLEMS
 **INVALID UNICODE ESCAPE SEQUENCE**
 This Unicode escape sequence is not valid.
@@ -97,23 +95,83 @@ This Unicode escape sequence is not valid.
 **INVALID UNICODE ESCAPE SEQUENCE**
 This Unicode escape sequence is not valid.
 
-**unicode_single_quotes.md:10:6:10:11:**
+**unicode_single_quotes.md:10:6:10:16:**
+```roc
+    '\u(EDA0B5)'
+```
+     ^^^^^^^^^^
+
+
+**INVALID UNICODE ESCAPE SEQUENCE**
+This Unicode escape sequence is not valid.
+
+**unicode_single_quotes.md:11:6:11:11:**
 ```roc
     '\u(K)',
 ```
      ^^^^^
 
 
+**SINGLE QUOTE EMPTY**
+Single-quoted literals must contain exactly one valid UTF-8 codepoint.
+
+**unicode_single_quotes.md:14:5:14:7:**
+```roc
+    '',
+```
+    ^^
+
+
+**SINGLE QUOTE TOO LONG**
+Single-quoted literals must contain exactly one valid UTF-8 codepoint.
+
+**unicode_single_quotes.md:15:5:15:11:**
+```roc
+    'long',
+```
+    ^^^^^^
+
+
+**UNCLOSED SINGLE QUOTE**
+This single-quoted literal is missing a closing quote.
+
+**unicode_single_quotes.md:16:5:16:9:**
+```roc
+    '\',
+```
+    ^^^^
+
+
+**UNCLOSED SINGLE QUOTE**
+This single-quoted literal is missing a closing quote.
+
+**unicode_single_quotes.md:19:5:19:7:**
+```roc
+y = 'u
+```
+    ^^
+
+
 **INVALID ESCAPE SEQUENCE**
 This escape sequence is not recognized.
 
-**unicode_single_quotes.md:21:2:22:1:**
+**unicode_single_quotes.md:22:2:23:1:**
 ```roc
 '\
 
 ```
 
 
+**UNCLOSED SINGLE QUOTE**
+This single-quoted literal is missing a closing quote.
+
+**unicode_single_quotes.md:22:1:22:3:**
+```roc
+'\
+```
+^^
+
+
 **UNEXPECTED TOKEN IN EXPRESSION**
 The token **'\u'** is not expected in an expression.
 Expressions can be identifiers, literals, function calls, or operators.
@@ -159,54 +217,32 @@ Expressions can be identifiers, literals, function calls, or operators.
 
 
 **UNEXPECTED TOKEN IN EXPRESSION**
-The token **'\u(K)'** is not expected in an expression.
+The token **'\u(EDA0B5)'** is not expected in an expression.
 Expressions can be identifiers, literals, function calls, or operators.
 
-**unicode_single_quotes.md:10:5:10:12:**
+**unicode_single_quotes.md:10:5:10:17:**
 ```roc
-    '\u(K)',
+    '\u(EDA0B5)'
 ```
-    ^^^^^^^
+    ^^^^^^^^^^^^
 
 
-**UNEXPECTED TOKEN IN EXPRESSION**
+**PARSE ERROR**
-The token **''** is not expected in an expression.
+A parsing error occurred: `expected_expr_close_round_or_comma`
-Expressions can be identifiers, literals, function calls, or operators.
+This is an unexpected parsing error. Please check your syntax.
 
-**unicode_single_quotes.md:13:5:13:7:**
+**unicode_single_quotes.md:17:1:17:2:**
 ```roc
-    '',
+)
 ```
-    ^^
+^
-
-
-**UNEXPECTED TOKEN IN EXPRESSION**
-The token **'long'** is not expected in an expression.
-Expressions can be identifiers, literals, function calls, or operators.
-
-**unicode_single_quotes.md:14:5:14:11:**
-```roc
-    'long',
-```
-    ^^^^^^
-
-
-**UNEXPECTED TOKEN IN EXPRESSION**
-The token **'\',** is not expected in an expression.
-Expressions can be identifiers, literals, function calls, or operators.
-
-**unicode_single_quotes.md:15:5:15:9:**
-```roc
-    '\',
-```
-    ^^^^
 
 
 **UNEXPECTED TOKEN IN EXPRESSION**
 The token **'u** is not expected in an expression.
 Expressions can be identifiers, literals, function calls, or operators.
 
-**unicode_single_quotes.md:18:5:18:7:**
+**unicode_single_quotes.md:19:5:19:7:**
 ```roc
 y = 'u
 ```
@@ -217,41 +253,28 @@ y = 'u
 A parsing error occurred: `statement_unexpected_token`
 This is an unexpected parsing error. Please check your syntax.
 
-**unicode_single_quotes.md:21:1:21:3:**
+**unicode_single_quotes.md:22:1:22:3:**
 ```roc
 '\
 ```
 ^^
 
 
-**INVALID TUPLE ELEMENT**
+**UNRECOGNIZED SYNTAX**
-This tuple element is malformed or contains invalid syntax.
+I don't recognize this syntax.
 
-**INVALID TUPLE ELEMENT**
+**unicode_single_quotes.md:17:1:17:2:**
-This tuple element is malformed or contains invalid syntax.
+```roc
+)
+```
+^
 
-**INVALID TUPLE ELEMENT**
+This might be a syntax error, an unsupported language feature, or a typo.
-This tuple element is malformed or contains invalid syntax.
-
-**INVALID TUPLE ELEMENT**
-This tuple element is malformed or contains invalid syntax.
-
-**INVALID TUPLE ELEMENT**
-This tuple element is malformed or contains invalid syntax.
-
-**INVALID TUPLE ELEMENT**
-This tuple element is malformed or contains invalid syntax.
-
-**INVALID TUPLE ELEMENT**
-This tuple element is malformed or contains invalid syntax.
-
-**INVALID TUPLE ELEMENT**
-This tuple element is malformed or contains invalid syntax.
 
 **UNRECOGNIZED SYNTAX**
 I don't recognize this syntax.
 
-**unicode_single_quotes.md:18:5:18:7:**
+**unicode_single_quotes.md:19:5:19:7:**
 ```roc
 y = 'u
 ```
@@ -265,20 +288,21 @@ LowerIdent,OpAssign,OpenRound,
 SingleQuote,Comma,
 SingleQuote,Comma,
 SingleQuote,Comma,
-MalformedSingleQuoteInvalidEscapeSequence,Comma,
+MalformedSingleQuote,Comma,
-MalformedSingleQuoteInvalidEscapeSequence,Comma,
+MalformedSingleQuote,Comma,
-MalformedSingleQuoteInvalidEscapeSequence,Comma,
+MalformedSingleQuote,Comma,
-MalformedSingleQuoteInvalidEscapeSequence,Comma,
+MalformedSingleQuote,Comma,
 SingleQuote,Comma,
-MalformedSingleQuoteInvalidEscapeSequence,Comma,
+MalformedSingleQuote,
+MalformedSingleQuote,Comma,
 SingleQuote,Comma,
 SingleQuote,Comma,
-MalformedSingleQuoteEmpty,Comma,
+MalformedSingleQuote,Comma,
-MalformedSingleQuoteTooLong,Comma,
+MalformedSingleQuote,Comma,
-MalformedSingleQuoteUnclosed,
+MalformedSingleQuote,
 CloseRound,
-LowerIdent,OpAssign,MalformedSingleQuoteUnclosed,
+LowerIdent,OpAssign,MalformedSingleQuote,
-MalformedSingleQuoteUnclosed,
+MalformedSingleQuote,
 EndOfFile,
 ~~~
 # PARSE
@@ -288,21 +312,7 @@ EndOfFile,
 	(statements
 		(s-decl
 			(p-ident (raw "x"))
-			(e-tuple
+			(e-malformed (reason "expected_expr_close_round_or_comma")))
-				(e-single-quote (raw "'a'"))
-				(e-single-quote (raw "'é'"))
-				(e-single-quote (raw "'🚀'"))
-				(e-malformed (reason "expr_unexpected_token"))
-				(e-malformed (reason "expr_unexpected_token"))
-				(e-malformed (reason "expr_unexpected_token"))
-				(e-malformed (reason "expr_unexpected_token"))
-				(e-single-quote (raw "'\u(1F680)'"))
-				(e-malformed (reason "expr_unexpected_token"))
-				(e-single-quote (raw "'\\'"))
-				(e-single-quote (raw "'\''"))
-				(e-malformed (reason "expr_unexpected_token"))
-				(e-malformed (reason "expr_unexpected_token"))
-				(e-malformed (reason "expr_unexpected_token"))))
 		(s-decl
 			(p-ident (raw "y"))
 			(e-malformed (reason "expr_unexpected_token")))
@@ -310,22 +320,8 @@ EndOfFile,
 ~~~
 # FORMATTED
 ~~~roc
-x = (
+x = 
-	'a',
+	
-	'é',
-	'🚀',
-	,
-	,
-	,
-	,
-	'\u(1F680)',
-	,
-	'\\',
-	'\'',
-	,
-	,
-	,
-)
 
 y = 
 
@@ -336,22 +332,7 @@ y =
 (can-ir
 	(d-let
 		(p-assign (ident "x"))
-		(e-tuple
+		(e-runtime-error (tag "expr_not_canonicalized")))
-			(elems
-				(e-num (value "97"))
-				(e-num (value "233"))
-				(e-num (value "128640"))
-				(e-runtime-error (tag "tuple_elem_not_canonicalized"))
-				(e-runtime-error (tag "tuple_elem_not_canonicalized"))
-				(e-runtime-error (tag "tuple_elem_not_canonicalized"))
-				(e-runtime-error (tag "tuple_elem_not_canonicalized"))
-				(e-num (value "128640"))
-				(e-runtime-error (tag "tuple_elem_not_canonicalized"))
-				(e-num (value "92"))
-				(e-num (value "39"))
-				(e-runtime-error (tag "tuple_elem_not_canonicalized"))
-				(e-runtime-error (tag "tuple_elem_not_canonicalized"))
-				(e-runtime-error (tag "tuple_elem_not_canonicalized")))))
 	(d-let
 		(p-assign (ident "y"))
 		(e-runtime-error (tag "expr_not_canonicalized"))))
@@ -360,9 +341,9 @@ y =
 ~~~clojure
 (inferred-types
 	(defs
-		(patt (type "(Num(Int(_size)), Num(Int(_size2)), Num(Int(_size3)), Error, Error, Error, Error, Num(Int(_size4)), Error, Num(Int(_size5)), Num(Int(_size6)), Error, Error, Error)"))
+		(patt (type "Error"))
 		(patt (type "Error")))
 	(expressions
-		(expr (type "(Num(Int(_size)), Num(Int(_size2)), Num(Int(_size3)), Error, Error, Error, Error, Num(Int(_size4)), Error, Num(Int(_size5)), Num(Int(_size6)), Error, Error, Error)"))
+		(expr (type "Error"))
 		(expr (type "Error"))))
 ~~~