"""This module provides classes representing grammars."""
import json
from typing import Any, Dict, List, Optional, Tuple, Type, Union
from pydantic import BaseModel, Field
from .base import XGRObject, _core
[docs]class StructuralTagItem(BaseModel):
"""A structural tag item. See Grammar.from_structural_tag() for more details."""
begin: str
"""The begin tag."""
schema_: Union[str, Type[BaseModel], Dict[str, Any]] = Field(alias="schema")
"""The schema."""
end: str
"""The end tag."""
def _convert_schema_to_str(schema: Union[str, Type[BaseModel], Dict[str, Any]]) -> str:
"""Convert a schema to a string representation.
This function handles different schema input types and converts them to a JSON string:
- Pydantic models are converted using their schema methods
- String inputs are returned as-is (assumed to be valid JSON)
- Dictionary inputs are converted to JSON strings
Parameters
----------
schema : Union[str, Type[BaseModel], Dict[str, Any]]
The schema to convert, which can be a Pydantic model class,
a JSON schema string, or a dictionary representing a JSON schema.
Returns
-------
str
The JSON schema as a string.
Raises
------
ValueError, TypeError
If the schema type is not supported, or the dictionary is not serializable.
"""
if isinstance(schema, type) and issubclass(schema, BaseModel):
if hasattr(schema, "model_json_schema"):
return json.dumps(schema.model_json_schema())
if hasattr(schema, "schema_json"):
return json.dumps(schema.schema_json())
else:
raise ValueError("The schema should have a model_json_schema or json_schema method.")
elif isinstance(schema, str):
return schema
elif isinstance(schema, dict):
return json.dumps(schema)
else:
raise ValueError("The schema should be a string or a Pydantic model.")
[docs]class Grammar(XGRObject):
"""This class represents a grammar object in XGrammar, and can be used later in the
grammar-guided generation.
The Grammar object supports context-free grammar (CFG). EBNF (extended Backus-Naur Form) is
used as the format of the grammar. There are many specifications for EBNF in the literature,
and we follow the specification of GBNF (GGML BNF) in
https://github.com/ggerganov/llama.cpp/blob/master/grammars/README.md.
When printed, the grammar will be converted to GBNF format.
"""
def __str__(self) -> str:
"""Print the BNF grammar to a string, in EBNF format.
Returns
-------
grammar_string : str
The BNF grammar string.
"""
return self._handle.to_string()
[docs] @staticmethod
def from_ebnf(ebnf_string: str, *, root_rule_name: str = "root") -> "Grammar":
"""Construct a grammar from EBNF string. The EBNF string should follow the format
in https://github.com/ggerganov/llama.cpp/blob/master/grammars/README.md.
Parameters
----------
ebnf_string : str
The grammar string in EBNF format.
root_rule_name : str, default: "root"
The name of the root rule in the grammar.
Raises
------
RuntimeError
When converting the regex pattern fails, with details about the parsing error.
"""
return Grammar._create_from_handle(_core.Grammar.from_ebnf(ebnf_string, root_rule_name))
[docs] @staticmethod
def from_json_schema(
schema: Union[str, Type[BaseModel], Dict[str, Any]],
*,
any_whitespace: bool = True,
indent: Optional[int] = None,
separators: Optional[Tuple[str, str]] = None,
strict_mode: bool = True,
print_converted_ebnf: bool = False,
) -> "Grammar":
"""Construct a grammar from JSON schema. Pydantic model or JSON schema string can be
used to specify the schema.
It allows any whitespace by default. If user want to specify the format of the JSON,
set `any_whitespace` to False and use the `indent` and `separators` parameters. The
meaning and the default values of the parameters follows the convention in json.dumps().
It internally converts the JSON schema to a EBNF grammar.
Parameters
----------
schema : Union[str, Type[BaseModel], Dict[str, Any]]
The schema string or Pydantic model or JSON schema dict.
any_whitespace : bool, default: True
Whether to use any whitespace. If True, the generated grammar will ignore the
indent and separators parameters, and allow any whitespace.
indent : Optional[int], default: None
The number of spaces for indentation. If None, the output will be in one line.
Note that specifying the indentation means forcing the LLM to generate JSON strings
strictly formatted. However, some models may tend to generate JSON strings that
are not strictly formatted. In this case, forcing the LLM to generate strictly
formatted JSON strings may degrade the generation quality. See
<https://github.com/sgl-project/sglang/issues/2216#issuecomment-2516192009> for more
details.
separators : Optional[Tuple[str, str]], default: None
Two separators used in the schema: comma and colon. Examples: (",", ":"), (", ", ": ").
If None, the default separators will be used: (",", ": ") when the indent is not None,
and (", ", ": ") otherwise.
strict_mode : bool, default: True
Whether to use strict mode. In strict mode, the generated grammar will not allow
properties and items that is not specified in the schema. This is equivalent to
setting unevaluatedProperties and unevaluatedItems to false.
This helps LLM to generate accurate output in the grammar-guided generation with JSON
schema.
print_converted_ebnf : bool, default: False
If True, the converted EBNF string will be printed. For debugging purposes.
Returns
-------
grammar : Grammar
The constructed grammar.
Raises
------
RuntimeError
When converting the json schema fails, with details about the parsing error.
"""
schema_str = _convert_schema_to_str(schema)
return Grammar._create_from_handle(
_core.Grammar.from_json_schema(
schema_str, any_whitespace, indent, separators, strict_mode, print_converted_ebnf
)
)
[docs] @staticmethod
def from_regex(regex_string: str, *, print_converted_ebnf: bool = False) -> "Grammar":
"""Create a grammar from a regular expression string.
Parameters
----------
regex_string : str
The regular expression pattern to create the grammar from.
print_converted_ebnf : bool, default: False
This method will convert the regex pattern to EBNF first. If this is true, the converted
EBNF string will be printed. For debugging purposes. Default: False.
Returns
-------
grammar : Grammar
The constructed grammar from the regex pattern.
Raises
------
RuntimeError
When parsing the regex pattern fails, with details about the parsing error.
"""
return Grammar._create_from_handle(
_core.Grammar.from_regex(regex_string, print_converted_ebnf)
)
[docs] @staticmethod
def from_structural_tag(tags: List[StructuralTagItem], triggers: List[str]) -> "Grammar":
"""Create a grammar from structural tags. The structural tag handles the dispatching
of different grammars based on the tags and triggers: it initially allows any output,
until a trigger is encountered, then dispatch to the corresponding tag; when the end tag
is encountered, the grammar will allow any following output, until the next trigger is
encountered.
The tags parameter is used to specify the output pattern. It is especially useful for LLM
function calling, where the pattern is:
<function=func_name>{"arg1": ..., "arg2": ...}</function>.
This pattern consists of three parts: a begin tag (<function=func_name>), a parameter list
according to some schema ({"arg1": ..., "arg2": ...}), and an end tag (</function>). This
pattern can be described in a StructuralTagItem with a begin tag, a schema, and an end tag.
The structural tag is able to handle multiple such patterns by passing them into multiple
tags.
The triggers parameter is used to trigger the dispatching of different grammars. The trigger
should be a prefix of a provided begin tag. When the trigger is encountered, the
corresponding tag should be used to constrain the following output. There can be multiple
tags matching the same trigger. Then if the trigger is encountered, the following output
should match one of the tags. For example, in function calling, the triggers can be
["<function="]. Then if "<function=" is encountered, the following output must match one
of the tags (e.g. <function=get_weather>{"city": "Beijing"}</function>).
The corrrespondence of tags and triggers is automatically determined: all tags with the
same trigger will be grouped together. User should make sure any trigger is not a prefix
of another trigger: then the corrrespondence of tags and triggers will be ambiguous.
To use this grammar in grammar-guided generation, the GrammarMatcher constructed from
structural tag will generate a mask for each token. When the trigger is not encountered,
the mask will likely be all-1 and not have to be used (fill_next_token_bitmask returns
False, meaning no token is masked). When a trigger is encountered, the mask should be
enforced (fill_next_token_bitmask will return True, meaning some token is masked) to the
output logits.
The benefit of this method is the token boundary between tags and triggers is automatically
handled. The user does not need to worry about the token boundary.
Parameters
----------
tags : List[StructuralTagItem]
The structural tags.
triggers : List[str]
The triggers.
Examples
--------
>>> class Schema1(BaseModel):
>>> arg1: str
>>> arg2: int
>>> class Schema2(BaseModel):
>>> arg3: float
>>> arg4: List[str]
>>> tags = [
>>> StructuralTagItem(begin="<function=f>", schema=Schema1, end="</function>"),
>>> StructuralTagItem(begin="<function=g>", schema=Schema2, end="</function>"),
>>> ]
>>> triggers = ["<function="]
>>> grammar = Grammar.from_structural_tag(tags, triggers)
"""
tags_tuple = [(tag.begin, _convert_schema_to_str(tag.schema_), tag.end) for tag in tags]
return Grammar._create_from_handle(_core.Grammar.from_structural_tag(tags_tuple, triggers))
[docs] @staticmethod
def builtin_json_grammar() -> "Grammar":
"""Get the grammar of standard JSON. This is compatible with the official JSON grammar
specification in https://www.json.org/json-en.html.
Returns
-------
grammar : Grammar
The JSON grammar.
"""
return Grammar._create_from_handle(_core.Grammar.builtin_json_grammar())
[docs] @staticmethod
def concat(*grammars: "Grammar") -> "Grammar":
"""Create a grammar that matches the concatenation of the grammars in the list. That is
equivalent to using the `+` operator to concatenate the grammars in the list.
Parameters
----------
grammars : List[Grammar]
The grammars to create the concatenation of.
Returns
-------
grammar : Grammar
The concatenation of the grammars.
"""
grammar_handles = [grammar._handle for grammar in grammars]
return Grammar._create_from_handle(_core.Grammar.concat(grammar_handles))
[docs] @staticmethod
def union(*grammars: "Grammar") -> "Grammar":
"""Create a grammar that matches any of the grammars in the list. That is equivalent to
using the `|` operator to concatenate the grammars in the list.
Parameters
----------
grammars : List[Grammar]
The grammars to create the union of.
Returns
-------
grammar : Grammar
The union of the grammars.
"""
grammar_handles = [grammar._handle for grammar in grammars]
return Grammar._create_from_handle(_core.Grammar.union(grammar_handles))