feat: 1.21.5 text component and SNBT changes (#306)

Author: misode

mecha/ast.py

@@ -832,6 +832,7 @@ class AstNbtByteArray(AstNbt):
     elements: AstChildren[AstNbt] = required_field()
 
     parser = None
+    array_prefix = "B"
 
     def evaluate(self) -> Any:
         return ByteArray([element.evaluate() for element in self.elements])  # type: ignore
@@ -844,6 +845,7 @@ class AstNbtIntArray(AstNbt):
     elements: AstChildren[AstNbt] = required_field()
 
     parser = None
+    array_prefix = "I"
 
     def evaluate(self) -> Any:
         return IntArray([element.evaluate() for element in self.elements])  # type: ignore
@@ -856,6 +858,7 @@ class AstNbtLongArray(AstNbt):
     elements: AstChildren[AstNbt] = required_field()
 
     parser = None
+    array_prefix = "L"
 
     def evaluate(self) -> Any:
         return LongArray([element.evaluate() for element in self.elements])  # type: ignore

mecha/parse.py

@@ -184,7 +184,7 @@
 from .config import CommandTree
 from .error import MechaError
 from .spec import CommandSpec, Parser
-from .utils import JsonQuoteHelper, QuoteHelper, string_to_number
+from .utils import JsonQuoteHelper, NbtQuoteHelper, QuoteHelper, string_to_number
 
 NUMBER_PATTERN: str = r"-?(?:\d+\.?\d*|\.\d+)"
 
@@ -493,8 +493,8 @@ def get_default_parsers() -> Dict[str, Parser]:
             ]
         ),
         "command:argument:minecraft:column_pos": delegate("column_pos"),
-        "command:argument:minecraft:component": MultilineParser(delegate("json")),
-        "command:argument:minecraft:style": MultilineParser(delegate("json")),
+        "command:argument:minecraft:component": MultilineParser(delegate("nbt")),
+        "command:argument:minecraft:style": MultilineParser(delegate("nbt")),
         "command:argument:minecraft:dimension": delegate("resource_location"),
         "command:argument:minecraft:entity": delegate("entity"),
         "command:argument:minecraft:entity_anchor": delegate("entity_anchor"),
@@ -583,6 +583,12 @@ def get_parsers(version: VersionNumber = LATEST_MINECRAFT_VERSION) -> Dict[str,
     if version < (1, 20):
         parsers["scoreboard_slot"] = BasicLiteralParser(AstLegacyScoreboardSlot)
 
+    if version < (1, 21):
+        parsers["command:argument:minecraft:component"] = MultilineParser(
+            delegate("json")
+        )
+        parsers["command:argument:minecraft:style"] = MultilineParser(delegate("json"))
+
     return parsers
 
 
@@ -1165,13 +1171,7 @@ class NbtParser:
         }
     )
 
-    quote_helper: QuoteHelper = field(
-        default_factory=lambda: QuoteHelper(
-            escape_sequences={
-                r"\\": "\\",
-            }
-        )
-    )
+    quote_helper: QuoteHelper = field(default_factory=NbtQuoteHelper)
 
     def __post_init__(self):
         self.compound_entry_parser = self.parse_compound_entry
@@ -1234,24 +1234,16 @@ def __call__(self, stream: TokenStream) -> AstNbt:
                         node = AstNbtLongArray(elements=AstChildren(elements))
                         element_type = Long  # type: ignore
                         msg = "Expected all elements to be long integers."
+
+                    node = set_location(node, array, stream.current)
+
+                    for element in node.elements:
+                        if isinstance(element, AstNbtValue):
+                            if type(element.value) is not element_type:
+                                raise element.emit_error(InvalidSyntax(msg))
                 else:
                     node = AstNbtList(elements=AstChildren(elements))
-                    element_type = None
-                    msg = "Expected all elements to have the same type."
-
-                node = set_location(node, bracket or array, stream.current)
-
-                for element in node.elements:
-                    if isinstance(element, AstNbtValue):
-                        if not element_type:
-                            element_type = type(element.value)
-                        elif type(element.value) is not element_type:
-                            raise element.emit_error(InvalidSyntax(msg))
-                    elif isinstance(element, AstNbt):  # type: ignore
-                        if not element_type:
-                            element_type = type(element)
-                        elif type(element) is not element_type:
-                            raise element.emit_error(InvalidSyntax(msg))
+                    node = set_location(node, bracket, stream.current)
 
                 return node
 

mecha/serialize.py

@@ -32,12 +32,17 @@
     AstMacroLineText,
     AstMacroLineVariable,
     AstMessage,
-    AstNbt,
     AstNbtBool,
+    AstNbtByteArray,
     AstNbtCompound,
+    AstNbtCompoundKey,
+    AstNbtIntArray,
+    AstNbtList,
+    AstNbtLongArray,
     AstNbtPath,
     AstNbtPathKey,
     AstNbtPathSubscript,
+    AstNbtValue,
     AstNode,
     AstNumber,
     AstParticle,
@@ -61,9 +66,10 @@
 from .database import CompilationDatabase
 from .dispatch import Visitor, rule
 from .spec import CommandSpec
-from .utils import QuoteHelper, number_to_string
+from .utils import NbtQuoteHelper, QuoteHelper, number_to_string
 
 REGEX_COMMENTS = re.compile(r"^(?:(\s*#.*)|.+)", re.MULTILINE)
+UNQUOTED_COMPOUND_KEY = re.compile(r"^[a-zA-Z0-9._+-]+$")
 
 
 class FormattingOptions(BaseModel):
@@ -97,6 +103,7 @@ class Serializer(Visitor):
             }
         )
     )
+    nbt_quote_helper: QuoteHelper = field(default_factory=NbtQuoteHelper)
 
     def __call__(self, node: AstNode, **kwargs: Any) -> str:  # type: ignore
         result: List[str] = []
@@ -237,14 +244,68 @@ def json(self, node: AstJson, result: List[str]):
             )
         )
 
-    @rule(AstNbt)
-    def nbt(self, node: AstNbt, result: List[str]):
-        result.append(node.evaluate().snbt(compact=self.formatting.nbt_compact))
+    @rule(AstNbtValue)
+    def nbt_value(self, node: AstNbtValue, result: List[str]):
+        if isinstance(node.value, str):
+            result.append(self.nbt_quote_helper.quote_string(node.value))
+        else:
+            result.append(node.evaluate().snbt(compact=self.formatting.nbt_compact))
 
     @rule(AstNbtBool)
     def nbt_bool(self, node: AstNbtBool, result: List[str]):
         result.append("true" if node.value else "false")
 
+    @rule(AstNbtList)
+    def nbt_list(self, node: AstNbtList, result: List[str]):
+        result.append("[")
+        comma = "," if self.formatting.nbt_compact else ", "
+        sep = ""
+        for element in node.elements:
+            result.append(sep)
+            sep = comma
+            yield element
+        result.append("]")
+
+    @rule(AstNbtCompoundKey)
+    def nbt_compound_key(self, node: AstNbtCompoundKey, result: List[str]):
+        if UNQUOTED_COMPOUND_KEY.match(node.value):
+            result.append(node.value)
+        else:
+            result.append(self.nbt_quote_helper.quote_string(node.value))
+
+    @rule(AstNbtCompound)
+    def nbt_compound(self, node: AstNbtCompound, result: List[str]):
+        result.append("{")
+        comma, colon = (",", ":") if self.formatting.nbt_compact else (", ", ": ")
+        sep = ""
+        for entry in node.entries:
+            result.append(sep)
+            sep = comma
+            yield entry.key
+            result.append(colon)
+            yield entry.value
+        result.append("}")
+
+    @rule(AstNbtByteArray)
+    @rule(AstNbtIntArray)
+    @rule(AstNbtLongArray)
+    def nbt_array(
+        self,
+        node: Union[AstNbtByteArray, AstNbtIntArray, AstNbtLongArray],
+        result: List[str],
+    ):
+        result.append("[")
+        result.append(node.array_prefix)
+        semicolon = ";" if self.formatting.nbt_compact else "; "
+        result.append(semicolon)
+        comma = "," if self.formatting.nbt_compact else ", "
+        sep = ""
+        for element in node.elements:
+            result.append(sep)
+            sep = comma
+            yield element
+        result.append("]")
+
     @rule(AstResourceLocation)
     def resource_location(self, node: AstResourceLocation, result: List[str]):
         result.append(node.get_value())

mecha/utils.py

@@ -2,6 +2,7 @@
     "QuoteHelper",
     "QuoteHelperWithUnicode",
     "JsonQuoteHelper",
+    "NbtQuoteHelper",
     "InvalidEscapeSequence",
     "normalize_whitespace",
     "string_to_number",
@@ -14,7 +15,7 @@
 import re
 from dataclasses import dataclass, field
 from pathlib import Path
-from typing import Any, Dict, Optional, Union
+from typing import Any, Dict, List, Optional, Union
 
 from beet import File
 from beet.core.utils import FileSystemPath
@@ -25,6 +26,7 @@
 ESCAPE_REGEX = re.compile(r"\\.")
 UNICODE_ESCAPE_REGEX = re.compile(r"\\(?:u([0-9a-fA-F]{4})|.)")
 AVOID_QUOTES_REGEX = re.compile(r"^[0-9A-Za-z_\.\+\-]+$")
+QUOTE_REGEX = re.compile(r"\"|'")
 
 WHITESPACE_REGEX = re.compile(r"\s+")
 
@@ -70,10 +72,15 @@ class QuoteHelper:
     avoid_quotes_regex: "re.Pattern[str]" = AVOID_QUOTES_REGEX
 
     escape_sequences: Dict[str, str] = field(default_factory=dict)
+    unquote_only_escape_characters: List[str] = field(default_factory=list)
     escape_characters: Dict[str, str] = field(init=False)
 
     def __post_init__(self):
-        self.escape_characters = {v: k for k, v in self.escape_sequences.items()}
+        self.escape_characters = {
+            v: k
+            for k, v in self.escape_sequences.items()
+            if not k in self.unquote_only_escape_characters
+        }
 
     def unquote_string(self, token: Token) -> str:
         """Remove quotes and substitute escaped characters."""
@@ -106,9 +113,14 @@ def quote_string(self, value: str, quote: str = '"') -> str:
         """Wrap the string in quotes if it can't be represented unquoted."""
         if self.avoid_quotes_regex.match(value):
             return value
+        value = self.handle_quoting(value)
+        return quote + value.replace(quote, "\\" + quote) + quote
+
+    def handle_quoting(self, value: str) -> str:
+        """Handle escape characters during quoting."""
         for match, seq in self.escape_characters.items():
             value = value.replace(match, seq)
-        return quote + value.replace(quote, "\\" + quote) + quote
+        return value
 
 
 @dataclass
@@ -122,6 +134,17 @@ def handle_substitution(self, token: Token, match: "re.Match[str]") -> str:
             return chr(int(unicode_hex, 16))
         return super().handle_substitution(token, match)
 
+    def handle_quoting(self, value: str) -> str:
+        value = super().handle_quoting(value)
+
+        def escape_char(char: str) -> str:
+            codepoint = ord(char)
+            if codepoint < 128:
+                return char
+            return f"\\u{codepoint:04x}"
+
+        return "".join(escape_char(c) for c in value)
+
 
 @dataclass
 class JsonQuoteHelper(QuoteHelperWithUnicode):
@@ -138,6 +161,31 @@ class JsonQuoteHelper(QuoteHelperWithUnicode):
     )
 
 
+@dataclass
+class NbtQuoteHelper(QuoteHelperWithUnicode):
+    """Quote helper used for snbt."""
+
+    escape_sequences: Dict[str, str] = field(
+        default_factory=lambda: {
+            r"\\": "\\",
+            r"\b": "\b",
+            r"\f": "\f",
+            r"\n": "\n",
+            r"\r": "\r",
+            r"\s": " ",
+            r"\t": "\t",
+        }
+    )
+    unquote_only_escape_characters: List[str] = field(default_factory=lambda: [r"\s"])
+
+    def quote_string(self, value: str, quote: Optional[str] = None) -> str:
+        if not quote:
+            found = QUOTE_REGEX.search(value)
+            quote = "'" if found and found.group() == '"' else '"'
+        value = super().handle_quoting(value)
+        return quote + value.replace(quote, "\\" + quote) + quote
+
+
 def underline_code(
     source: str,
     location: SourceLocation,

tests/resources/argument_examples.json

@@ -145,12 +145,13 @@
         "{\"text\":\"hello world\"}",
         "[\"\"]",
         "\"\\u2588\"",
-        "{\"\\u2588\\\"\\n\":[123]}"
+        "{\"\\u2588\\\"\\n\":[123]}",
+        "wat"
       ]
     },
     {
       "invalid": true,
-      "examples": ["wat", "{\"hey\": ]}"]
+      "examples": ["{\"hey\": ]}"]
     }
   ],
   "minecraft:dimension": [
@@ -415,18 +416,18 @@
         "[L;1l,2l]",
         "[a, b, c]",
         "[[], [1], [foo]]",
-        "[[],[[[[[[[[[[],[[[[5,1]],[]]]]]]]]]]],[[[[]]]]]]"
+        "[[],[[[[[[[[[[],[[[[5,1]],[]]]]]]]]]]],[[[[]]]]]]",
+        "\"\\n\"",
+        "[a,1]",
+        "[[],[],1b]"
       ]
     },
     {
       "invalid": true,
       "examples": [
         "\"\\\"",
-        "\"\\n\"",
         "\"\\\\\\\"",
         "{\"\\\":1}",
-        "[a,1]",
-        "[[],[],1b]",
         "[B;5l,4l,3]",
         "[I;5l,4l,3]",
         "[L;5l,4l,3]",

tests/snapshots/examples__build_basic_formatting__0.pack.md

@@ -20,7 +20,7 @@
 ```mcfunction
 say this is a test
 execute as @a[nbt={SelectedItem:{id:"minecraft:diamond",Count:64b}}] at @s run setblock ~ ~ ~ repeater[delay=3,facing=south]
-tellraw @a ["",{"text":"hello","color":"red"},"not ascii: \u00b6"]
+tellraw @a ["",{text:"hello",color:"red"},"not ascii: \u00b6"]
 say goodbye
 ```
 
@@ -29,7 +29,7 @@ say goodbye
 ```mcfunction
 say this is a test
 execute as @a[nbt={SelectedItem: {id: "minecraft:diamond", Count: 64b}}] at @s run setblock ~ ~ ~ repeater[delay=3, facing=south]
-tellraw @a ["", {"text": "hello", "color": "red"}, "not ascii: \u00b6"]
+tellraw @a ["", {text: "hello", color: "red"}, "not ascii: \u00b6"]
 say goodbye
 ```
 
@@ -44,7 +44,7 @@ execute as @a[nbt={SelectedItem: {id: "minecraft:diamond", Count: 64b}}] at @s r
 
 
 
-tellraw @a ["", {"text": "hello", "color": "red"}, "not ascii: \u00b6"]
+tellraw @a ["", {text: "hello", color: "red"}, "not ascii: \u00b6"]
 
 
 
@@ -63,7 +63,7 @@ execute as @a[nbt={SelectedItem:{id:"minecraft:diamond",Count:64b}}] at @s run s
 
 
 
-tellraw @a ["", {"color": "red", "text": "hello"}, "not ascii: ¶"]
+tellraw @a ["",{text:"hello",color:"red"},"not ascii: \u00b6"]