新增修改

This commit is contained in:
2026-04-23 12:48:59 +08:00
parent df46812eff
commit 92b856b622
60 changed files with 18623 additions and 398 deletions

86
final/config/config.json Normal file
View File

@@ -0,0 +1,86 @@
{
"source": "json",
"json": {
"input_file": "data/raw_data_sample.json",
"output_file": "data/nodes.json",
"report_file": "report/detailed_cleaning_report.json"
},
"triplet": {
"enabled": true,
"input_file": null,
"output_file": "data/triples.json",
"report_file": "report/triplet_cleaning_report.json",
"clean": {
"deduplicate": true,
"remove_null_core": true,
"normalize_relation_type": true,
"filter_orphans": false
}
},
"db": {
"host": "127.0.0.1",
"port": 5080,
"user": "SYSDBA",
"schema": "SYSDBA",
"table": "OBJECTIVE_INSTANCE",
"query": null,
"triplet_table": "RELATION_INSTANCE",
"triplet_query": null
},
"formal": {
"enabled": false,
"user_id": "ADMIN",
"task_id": "TASK_FULL_KB",
"source_sys_type": "关系推理后的体系",
"target_sys_type": "形式化后的体系",
"source_query": null,
"user_column": "USER_ID",
"task_column": "TASK_ID",
"sys_type_column": "SYS_TYPE",
"skip_sys_type_validation": false,
"entity_id_column": "TARGET_ID",
"objective_table": "OBJECTIVE_INSTANCE",
"triplet_table": "RELATION_INSTANCE",
"triplet_primary_key": "RELATION_INSTANCE_ID",
"triplet_omit_primary_key_on_insert": true,
"triple_batch_column": "BATCH_ID",
"attribute_table": "METRIC_VALUE_INSTANCE",
"metric_id_column": "METRIC_ID",
"metric_value_column": "METRIC_VALUE",
"metric_insert_static": null,
"formal_result_table": "RELATION_INSTANCE",
"analysis_table": "RES_FORMAL_ANALYSIS",
"analysis_insert_columns": null,
"analysis_static_columns": null,
"rules_summary": null,
"required_attrs": [
"f_IC",
"f_IA",
"f_CS",
"f_IT",
"f_DP",
"f_CPS",
"f_CC",
"position",
"effective_radius",
"response_time",
"cycle_time",
"data_age",
"time_window",
"core_performance",
"survivability",
"mtbf",
"protocol_list",
"format_list",
"interface_list",
"security_level",
"org_unit",
"nation",
"history_success"
]
},
"eval": {
"nodes_file": "data/nodes.json",
"triples_file": "data/triples.json"
}
}

96
final/config/config.yaml Normal file
View File

@@ -0,0 +1,96 @@
# 数据清洗系统配置文件
# 通过 source 切换json开发 / db生产-达梦)
source: json # 开发用 json部署达梦时改为 db或直接改 config.json
# JSON 文件模式source=json 时生效)
json:
input_file: data/raw_data_sample.json
output_file: data/nodes.json
report_file: report/detailed_cleaning_report.json
# 网络关系/效能评估读取路径main_227 在清洗后加载)
eval:
nodes_file: data/nodes.json # 与 json.output_file 一致(清洗输出)
triples_file: data/triples.json # 与 triplet.output_file 一致run_all 输出)
# 三元组run_all 时一并处理db 模式从库读json 模式可配 input_file
triplet:
enabled: true # false 可关闭三元组处理
input_file: null # json 模式下可选,如 data/triplets_raw.json
output_file: data/triples.json # 与 eval.triples_file 一致,便于 main_227 直接读取
report_file: report/triplet_cleaning_report.json
# 清洗配置clean: false 关闭清洗true 使用默认;或自定义子项
clean:
deduplicate: true # 按 (HEAD_ID, RELATION_TYPE, TAIL_ID) 去重
remove_null_core: true # 移除 HEAD_ID/TAIL_ID/RELATION_TYPE 为空的记录
normalize_relation_type: true # 规范化 RELATION_TYPE去空格
filter_orphans: false # 过滤悬空引用(需加载实体表,稍慢)
# 达梦数据库模式source=db 时生效)
# 密码通过环境变量 DM_PASSWORD 传入,切勿写入配置文件
db:
host: 127.0.0.1
port: 5080
user: SYSDBA
schema: SYSDBA
# OBJECTIVE_INSTANCE 表结构与 raw_data_sample.json 一致,直接查询即可
table: OBJECTIVE_INSTANCE
query: null
# 三元组表RELATION_INSTANCE: HEAD_ID, RELATION_TYPE, TAIL_ID
triplet_table: RELATION_INSTANCE
triplet_query: null # 可选:自定义 SQL
# 模块1.3 形式化入库流程source=db 时生效)
formal:
enabled: false
user_id: ADMIN
task_id: TASK_FULL_KB
source_sys_type: 关系推理后的体系
target_sys_type: 形式化后的体系
# 推荐:显式 SQL 只读「关系推理后的体系」,列名与库一致时最稳(为 null 则用下方列名拼 WHERE
source_query: null
user_column: USER_ID
task_column: TASK_ID
sys_type_column: SYS_TYPE
skip_sys_type_validation: false
entity_id_column: TARGET_ID
objective_table: OBJECTIVE_INSTANCE
triplet_table: RELATION_INSTANCE
triplet_primary_key: RELATION_INSTANCE_ID
triplet_omit_primary_key_on_insert: true
triple_batch_column: BATCH_ID
attribute_table: METRIC_VALUE_INSTANCE
metric_id_column: METRIC_ID
metric_value_column: METRIC_VALUE
metric_insert_static: null
formal_result_table: RELATION_INSTANCE
analysis_table: RES_FORMAL_ANALYSIS
# 若自动匹配列失败,在此列出 RES_FORMAL_ANALYSIS 的实际列名(顺序与 VALUES 一致)
analysis_insert_columns: null
analysis_static_columns: null
rules_summary: null
required_attrs:
- f_IC
- f_IA
- f_CS
- f_IT
- f_DP
- f_CPS
- f_CC
- position
- effective_radius
- response_time
- cycle_time
- data_age
- time_window
- core_performance
- survivability
- mtbf
- protocol_list
- format_list
- interface_list
- security_level
- org_unit
- nation
- history_success