diff --git a/common/json-schema-to-grammar.cpp b/common/json-schema-to-grammar.cpp index 2f233e2e7..4b19ff5a1 100644 --- a/common/json-schema-to-grammar.cpp +++ b/common/json-schema-to-grammar.cpp @@ -316,7 +316,7 @@ std::unordered_map GRAMMAR_LITERAL_ESCAPES = { }; std::unordered_set NON_LITERAL_SET = {'|', '.', '(', ')', '[', ']', '{', '}', '*', '+', '?'}; -std::unordered_set ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = {'[', ']', '(', ')', '|', '{', '}', '*', '+', '?'}; +std::unordered_set ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = {'^', '$', '.', '[', ']', '(', ')', '|', '{', '}', '*', '+', '?'}; template std::string join(Iterator begin, Iterator end, const std::string & separator) { diff --git a/examples/json_schema_to_grammar.py b/examples/json_schema_to_grammar.py index 92f6e3d47..ad23cdb1c 100755 --- a/examples/json_schema_to_grammar.py +++ b/examples/json_schema_to_grammar.py @@ -231,7 +231,7 @@ GRAMMAR_RANGE_LITERAL_ESCAPE_RE = re.compile(r'[\r\n"\]\-\\]') GRAMMAR_LITERAL_ESCAPES = {'\r': '\\r', '\n': '\\n', '"': '\\"', '-': '\\-', ']': '\\]'} NON_LITERAL_SET = set('|.()[]{}*+?') -ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = set('[]()|{}*+?') +ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = set('^$.[]()|{}*+?') class SchemaConverter: diff --git a/examples/server/public/json-schema-to-grammar.mjs b/examples/server/public/json-schema-to-grammar.mjs index 06d76edde..837569889 100644 --- a/examples/server/public/json-schema-to-grammar.mjs +++ b/examples/server/public/json-schema-to-grammar.mjs @@ -259,7 +259,7 @@ const GRAMMAR_RANGE_LITERAL_ESCAPE_RE = /[\n\r"\]\-\\]/g; const GRAMMAR_LITERAL_ESCAPES = { '\r': '\\r', '\n': '\\n', '"': '\\"', '-': '\\-', ']': '\\]' }; const NON_LITERAL_SET = new Set('|.()[]{}*+?'); -const ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = new Set('[]()|{}*+?'); +const ESCAPED_IN_REGEXPS_BUT_NOT_IN_LITERALS = new Set('^$.[]()|{}*+?'); export class SchemaConverter { constructor(options) { diff --git a/tests/test-grammar-integration.cpp b/tests/test-grammar-integration.cpp index 0e21dc795..30484e97d 100644 --- a/tests/test-grammar-integration.cpp +++ b/tests/test-grammar-integration.cpp @@ -993,6 +993,40 @@ static void test_json_schema() { } ); + test_schema( + "simple pattern", + // Schema + R"""({ + "pattern": "^[a-zA-Z0-9_-]*$" + })""", + // Passing strings + { + R"""("")""", + R"""("He_llo-12")""", + }, + // Failing strings + { + R"""("!")""", + R"""("Hello World")""", + } + ); + + test_schema( + "pattern with escapes", + // Schema + R"""({ + "pattern": "^a\\^\\$\\.\\[\\]\\(\\)\\|\\{\\}\\*\\+\\?b$" + })""", + // Passing strings + { + R"""("a^$.[]()|{}*+?b")""", + }, + // Failing strings + { + R"""("ab")""", + } + ); + test_schema( "", // Schema