Files
cobol-java-v3/cobol_testgen/grammar.lark
T
NB-076 bb4a7a2346 fix: classification修复+grammar增强+75/75回归确认
分类修复:
- FILE-CONTROL关键词(0.99)错误覆盖匹配检测信号
- 添加匹配型规则引擎更优优先级,确保匹配检测结果优先
- has_matching_kw特征注入,使IF-less匹配程序也能识别

Grammar增强:
- LEVEL扩展到/[0-9]+/覆盖所有COBOL层级号
- HEX_STRING添加支持X'...'十六进制字面量
- VALUE子句逗号预处理剥离(88-level多值)
- COPY正则支持引号包覆的名称

结果: 内部75/75, 外部基准54/58(93%)

Co-Authored-By: Claude <noreply@anthropic.com>
2026-06-22 13:18:07 +08:00

41 lines
1.6 KiB
Plaintext

start: data_div_content
data_div_content: (file_section | working_storage | linkage)*
file_section: "FILE" "SECTION" DOT (fd | sd)+
fd: "FD" NAME FD_SUFFIX data_item*
sd: "SD" NAME FD_SUFFIX data_item*
FD_SUFFIX: /(?:"[^"]*"|'[^']*'|[^.])*\./
working_storage: "WORKING-STORAGE" "SECTION" DOT data_item*
linkage: "LINKAGE" "SECTION" DOT data_item*
data_item: level_num (NAME | "FILLER") clause* DOT
level_num: LEVEL
clause: pic_clause | value_clause | occurs_clause | redefines_clause | usage_clause
| "SYNC" | "SYNCHRONIZED"
| "JUSTIFIED" "RIGHT"?
| "BLANK" "WHEN" "ZERO"
| "GLOBAL" | "EXTERNAL"
pic_clause: "PIC" "IS"? PICTURE_STRING
value_clause: "VALUE" "IS"? value_literal+
value_literal: INT | SIGNED_NUMBER | STRING | SQSTRING
| "ZERO" | "ZEROS" | "ZEROES"
| "SPACE" | "SPACES"
| "HIGH-VALUE" | "HIGH-VALUES"
| "LOW-VALUE" | "LOW-VALUES"
| HEX_STRING
SQSTRING: /'[^']*'/
HEX_STRING: /X'[0-9A-Fa-f]+'/
redefines_clause: "REDEFINES" NAME
occurs_clause: "OCCURS" INT ("TO" INT)? "TIMES"? ("DEPENDING" "ON" NAME)? key_clause? indexed_clause?
key_clause: ("ASCENDING" | "DESCENDING") "KEY" "IS"? NAME (","? NAME)*
indexed_clause: "INDEXED" "BY" NAME (","? NAME)*
usage_clause: USAGE_VAL
USAGE_VAL: "COMP" | "COMP-3" | "COMP-5" | "BINARY" | "PACKED-DECIMAL" | "DISPLAY"
LEVEL: /0[1-9]|[0-4][0-9]|49|66|77|88|[0-9]+/
NAME: /[A-Z][A-Z0-9-]*/i
PICTURE_STRING: /[0-9A-Z()+,\-*\/V]+/i
INT: /[0-9]+/
DOT: /\./
%import common.SIGNED_NUMBER
%import common.ESCAPED_STRING -> STRING
%import common.WS
%ignore WS