Merge 718b80336c into 6bb99767ff

gitnexus/src/core/group/config-parser.ts

@@ -37,6 +37,7 @@ const DEFAULT_DETECT = {
   topics: true,
   includes: false,
   workspace_deps: false,
+  sofa: false,
 };
 
 const DEFAULT_MATCHING = {

gitnexus/src/core/group/extractors/sofa-extractor.ts

@@ -0,0 +1,136 @@
+import { glob } from 'glob';
+import { createIgnoreFilter } from '../../../config/ignore-service.js';
+import type { ContractExtractor, CypherExecutor } from '../contract-extractor.js';
+import type { ExtractedContract, RepoHandle } from '../types.js';
+import { readSafe } from './fs-utils.js';
+
+/**
+ * SOFA framework extractor for RPC contract discovery from XML files.
+ *
+ * SOFA RPC (Alibaba SOFA framework) uses XML declarations:
+ *   - `<sofa:service interface="com.xxx.YyyService">` + `<sofa:binding.tr/>` → provider
+ *   - `<sofa:reference interface="com.xxx.YyyService">` + `<sofa:binding.tr/>` → consumer
+ *
+ * SOFAMQ topic extraction is handled by the Java tree-sitter topic extractor
+ * (topic-patterns/java.ts), which detects OpenMessaging API calls:
+ *   - `consumer.subscribe("topic", "tag", listener)` → topic consumer
+ *   - `new Message("topic", "tag", body)` → topic provider
+ *
+ * XML `sofa:consumer` + `sofa:channel` + `sofa:binding.msg_broker` declarations
+ * are NOT extracted because they represent distributed scheduled task dispatching
+ * (e.g. `TP_F_SC` channel), not real message pub/sub relationships.
+ *
+ * Because SOFA XML uses a distinctive namespace prefix, regex extraction is
+ * reliable and avoids the need for a full XML tree-sitter grammar.
+ */
+
+const SOFA_XML_GLOB = '**/*.xml';
+
+// ─── Regex patterns for SOFA XML elements ───────────────────────────
+
+/**
+ * Match <sofa:service interface="...">…</sofa:service> blocks.
+ * Captures the interface attribute and the full block content.
+ * Note: self-closing <sofa:service …/> is not matched; SOFA XML always
+ * uses the closing tag because it contains binding children.
+ */
+const SOFA_SERVICE_RE = /<sofa:service\s+[^>]*interface="([^"]+)"[^>]*>([\s\S]*?)<\/sofa:service>/g;
+
+/**
+ * Match <sofa:reference interface="…" …> (may be self-closing or have body).
+ * Captures the interface attribute.
+ */
+const SOFA_REFERENCE_RE = /<sofa:reference\s+[^>]*interface="([^"]+)"[^>]*\/?>/g;
+
+/**
+ * Detect TR (bolt) binding inside a sofa:service block.
+ */
+const BINDING_TR_RE = /sofa:binding\.tr/;
+
+// ─── Helper: build an ExtractedContract ──────────────────────────────
+
+function makeRpcContract(
+  interfaceName: string,
+  role: 'provider' | 'consumer',
+  filePath: string,
+): ExtractedContract {
+  return {
+    contractId: `custom::${interfaceName}`,
+    type: 'custom',
+    role,
+    symbolUid: '',
+    symbolRef: { filePath: filePath.replace(/\\/g, '/'), name: interfaceName },
+    symbolName: interfaceName,
+    confidence: 0.85,
+    meta: {
+      framework: 'sofa-rpc',
+      interface: interfaceName,
+      extractionStrategy: 'regex_xml',
+    },
+  };
+}
+
+// ─── Extractor class ─────────────────────────────────────────────────
+
+export class SofaExtractor implements ContractExtractor {
+  type = 'custom' as const;
+
+  async canExtract(_repo: RepoHandle): Promise<boolean> {
+    return true;
+  }
+
+  async extract(
+    _dbExecutor: CypherExecutor | null,
+    repoPath: string,
+    _repo: RepoHandle,
+  ): Promise<ExtractedContract[]> {
+    const baseFilter = await createIgnoreFilter(repoPath);
+    const files = await glob(SOFA_XML_GLOB, {
+      cwd: repoPath,
+      ignore: baseFilter,
+      nodir: true,
+    });
+
+    const out: ExtractedContract[] = [];
+
+    for (const rel of files) {
+      const content = readSafe(repoPath, rel);
+      if (!content) continue;
+
+      // Skip files that don't contain any SOFA namespace declarations
+      if (!content.includes('sofa:')) continue;
+
+      // ── SOFA RPC: sofa:service (provider) ──
+      let m: RegExpExecArray | null;
+      SOFA_SERVICE_RE.lastIndex = 0;
+      while ((m = SOFA_SERVICE_RE.exec(content)) !== null) {
+        const interfaceName = m[1];
+        const body = m[2];
+        if (BINDING_TR_RE.test(body)) {
+          out.push(makeRpcContract(interfaceName, 'provider', rel));
+        }
+      }
+
+      // ── SOFA RPC: sofa:reference (consumer) ──
+      SOFA_REFERENCE_RE.lastIndex = 0;
+      while ((m = SOFA_REFERENCE_RE.exec(content)) !== null) {
+        const interfaceName = m[1];
+        out.push(makeRpcContract(interfaceName, 'consumer', rel));
+      }
+    }
+
+    return this.dedupe(out);
+  }
+
+  private dedupe(items: ExtractedContract[]): ExtractedContract[] {
+    const seen = new Set<string>();
+    const out: ExtractedContract[] = [];
+    for (const c of items) {
+      const k = `${c.contractId}|${c.role}|${c.symbolRef.filePath}`;
+      if (seen.has(k)) continue;
+      seen.add(k);
+      out.push(c);
+    }
+    return out;
+  }
+}

gitnexus/src/core/group/extractors/topic-patterns/java.ts

@@ -5,11 +5,14 @@ import type { TopicMeta } from './types.js';
 /**
  * Java topic extraction patterns.
  *
- * Detects Kafka and RabbitMQ (Spring conventions) producer/consumer APIs:
+ * Detects Kafka, RabbitMQ (Spring conventions) and SOFAMQ (OpenMessaging API)
+ * producer/consumer APIs:
  *   - `@KafkaListener(topics = "xxx")`
  *   - `@RabbitListener(queues = "xxx")`
  *   - `kafkaTemplate.send("xxx", ...)`
  *   - `rabbitTemplate.convertAndSend("xxx", ...)`
+ *   - `consumer.subscribe("topic", "tag", listener)`     (SOFAMQ consumer)
+ *   - `new Message("topic", "tag", body)`                (SOFAMQ producer)
  *
  * Every query MUST bind `@value` to the topic literal node.
  */
@@ -77,6 +80,33 @@ const JAVA_TOPIC_SPEC: LanguagePatterns<TopicMeta> = {
           arguments: (argument_list . (string_literal) @value))
       `,
     },
+    {
+      meta: {
+        role: 'consumer',
+        broker: 'sofamq',
+        confidence: 0.7,
+        symbolName: 'openMessaging.Consumer.subscribe',
+      },
+      query: `
+        (method_invocation
+          object: (identifier) @obj (#eq? @obj "consumer")
+          name: (identifier) @method (#eq? @method "subscribe")
+          arguments: (argument_list . (string_literal) @value))
+      `,
+    },
+    {
+      meta: {
+        role: 'provider',
+        broker: 'sofamq',
+        confidence: 0.7,
+        symbolName: 'openMessaging.Message.constructor',
+      },
+      query: `
+        (object_creation_expression
+          type: (type_identifier) @cls (#eq? @cls "Message")
+          arguments: (argument_list . (string_literal) @value))
+      `,
+    },
   ],
 };
 

gitnexus/src/core/group/extractors/topic-patterns/types.ts

@@ -8,7 +8,7 @@
  * query text directly — that's the whole point of the split.
  */
 
-export type Broker = 'kafka' | 'rabbitmq' | 'nats';
+export type Broker = 'kafka' | 'rabbitmq' | 'nats' | 'sofamq';
 
 /**
  * Per-pattern payload every topic plugin attaches to its query. Whatever

gitnexus/src/core/group/sync.ts

@@ -36,6 +36,7 @@ import { GraphqlExtractor } from './extractors/graphql-extractor.js';
 import { GrpcExtractor } from './extractors/grpc-extractor.js';
 import { ThriftExtractor } from './extractors/thrift-extractor.js';
 import { TopicExtractor } from './extractors/topic-extractor.js';
+import { SofaExtractor } from './extractors/sofa-extractor.js';
 import { IncludeExtractor } from './extractors/include-extractor.js';
 import { ManifestExtractor } from './extractors/manifest-extractor.js';
 import { discoverWorkspaceLinks } from './extractors/workspace-extractor.js';
@@ -349,6 +350,7 @@ export async function syncGroup(config: GroupConfig, opts?: SyncOptions): Promis
       const grpcEx = new GrpcExtractor();
       const thriftEx = new ThriftExtractor();
       const topicEx = new TopicExtractor();
+      const sofaEx = new SofaExtractor();
       const includeEx = new IncludeExtractor();
 
       for (const [groupPath, regName] of Object.entries(config.repos)) {
@@ -452,6 +454,17 @@ export async function syncGroup(config: GroupConfig, opts?: SyncOptions): Promis
             }
           }
 
+          if (config.detect.sofa) {
+            const extracted = await sofaEx.extract(executor, handle.repoPath, handle);
+            for (const c of extracted) {
+              autoContracts.push({
+                ...c,
+                repo: groupPath,
+                service: assignService(c.symbolRef.filePath, boundaries),
+              });
+            }
+          }
+
           if (config.detect.includes) {
             const extracted = await includeEx.extract(executor, handle.repoPath, handle);
             for (const c of extracted) {

gitnexus/src/core/group/types.ts

@@ -40,6 +40,8 @@ export interface DetectConfig {
   topics: boolean;
   includes: boolean;
   workspace_deps: boolean;
+  /** SOFA framework (SOFA RPC + SOFAMQ) XML contract extraction. Opt-in. */
+  sofa: boolean;
 }
 
 export interface MatchingConfig {