feat: rag-embedding-ai-chat (#1)

Co-authored-by: Sharang Parnerkar <parnerkarsharang@gmail.com>
Reviewed-on: #1

compliance-agent/src/api/handlers/mod.rs

@@ -1,3 +1,4 @@
+pub mod chat;
 pub mod dast;
 pub mod graph;
 
@@ -5,7 +6,8 @@ use std::sync::Arc;
 
 #[allow(unused_imports)]
 use axum::extract::{Extension, Path, Query};
-use axum::http::StatusCode;
+use axum::http::{header, StatusCode};
+use axum::response::IntoResponse;
 use axum::Json;
 use mongodb::bson::doc;
 use serde::{Deserialize, Serialize};
@@ -89,6 +91,72 @@ pub struct UpdateStatusRequest {
     pub status: String,
 }
 
+#[derive(Deserialize)]
+pub struct SbomFilter {
+    #[serde(default)]
+    pub repo_id: Option<String>,
+    #[serde(default)]
+    pub package_manager: Option<String>,
+    #[serde(default)]
+    pub q: Option<String>,
+    #[serde(default)]
+    pub has_vulns: Option<bool>,
+    #[serde(default)]
+    pub license: Option<String>,
+    #[serde(default = "default_page")]
+    pub page: u64,
+    #[serde(default = "default_limit")]
+    pub limit: i64,
+}
+
+#[derive(Deserialize)]
+pub struct SbomExportParams {
+    pub repo_id: String,
+    #[serde(default = "default_export_format")]
+    pub format: String,
+}
+
+fn default_export_format() -> String {
+    "cyclonedx".to_string()
+}
+
+#[derive(Deserialize)]
+pub struct SbomDiffParams {
+    pub repo_a: String,
+    pub repo_b: String,
+}
+
+#[derive(Serialize)]
+pub struct LicenseSummary {
+    pub license: String,
+    pub count: u64,
+    pub is_copyleft: bool,
+    pub packages: Vec<String>,
+}
+
+#[derive(Serialize)]
+pub struct SbomDiffResult {
+    pub only_in_a: Vec<SbomDiffEntry>,
+    pub only_in_b: Vec<SbomDiffEntry>,
+    pub version_changed: Vec<SbomVersionDiff>,
+    pub common_count: u64,
+}
+
+#[derive(Serialize)]
+pub struct SbomDiffEntry {
+    pub name: String,
+    pub version: String,
+    pub package_manager: String,
+}
+
+#[derive(Serialize)]
+pub struct SbomVersionDiff {
+    pub name: String,
+    pub package_manager: String,
+    pub version_a: String,
+    pub version_b: String,
+}
+
 type AgentExt = Extension<Arc<ComplianceAgent>>;
 type ApiResult<T> = Result<Json<ApiResponse<T>>, StatusCode>;
 
@@ -235,6 +303,52 @@ pub async fn trigger_scan(
     Ok(Json(serde_json::json!({ "status": "scan_triggered" })))
 }
 
+pub async fn delete_repository(
+    Extension(agent): AgentExt,
+    Path(id): Path<String>,
+) -> Result<Json<serde_json::Value>, StatusCode> {
+    let oid = mongodb::bson::oid::ObjectId::parse_str(&id).map_err(|_| StatusCode::BAD_REQUEST)?;
+    let db = &agent.db;
+
+    // Delete the repository
+    let result = db
+        .repositories()
+        .delete_one(doc! { "_id": oid })
+        .await
+        .map_err(|_| StatusCode::INTERNAL_SERVER_ERROR)?;
+
+    if result.deleted_count == 0 {
+        return Err(StatusCode::NOT_FOUND);
+    }
+
+    // Cascade delete all related data
+    let _ = db.findings().delete_many(doc! { "repo_id": &id }).await;
+    let _ = db.sbom_entries().delete_many(doc! { "repo_id": &id }).await;
+    let _ = db.scan_runs().delete_many(doc! { "repo_id": &id }).await;
+    let _ = db.cve_alerts().delete_many(doc! { "repo_id": &id }).await;
+    let _ = db
+        .tracker_issues()
+        .delete_many(doc! { "repo_id": &id })
+        .await;
+    let _ = db.graph_nodes().delete_many(doc! { "repo_id": &id }).await;
+    let _ = db.graph_edges().delete_many(doc! { "repo_id": &id }).await;
+    let _ = db.graph_builds().delete_many(doc! { "repo_id": &id }).await;
+    let _ = db
+        .impact_analyses()
+        .delete_many(doc! { "repo_id": &id })
+        .await;
+    let _ = db
+        .code_embeddings()
+        .delete_many(doc! { "repo_id": &id })
+        .await;
+    let _ = db
+        .embedding_builds()
+        .delete_many(doc! { "repo_id": &id })
+        .await;
+
+    Ok(Json(serde_json::json!({ "status": "deleted" })))
+}
+
 pub async fn list_findings(
     Extension(agent): AgentExt,
     Query(filter): Query<FindingsFilter>,
@@ -322,21 +436,46 @@ pub async fn update_finding_status(
 
 pub async fn list_sbom(
     Extension(agent): AgentExt,
-    Query(params): Query<PaginationParams>,
+    Query(filter): Query<SbomFilter>,
 ) -> ApiResult<Vec<SbomEntry>> {
     let db = &agent.db;
-    let skip = (params.page.saturating_sub(1)) * params.limit as u64;
+    let mut query = doc! {};
+
+    if let Some(repo_id) = &filter.repo_id {
+        query.insert("repo_id", repo_id);
+    }
+    if let Some(pm) = &filter.package_manager {
+        query.insert("package_manager", pm);
+    }
+    if let Some(q) = &filter.q {
+        if !q.is_empty() {
+            query.insert("name", doc! { "$regex": q, "$options": "i" });
+        }
+    }
+    if let Some(has_vulns) = filter.has_vulns {
+        if has_vulns {
+            query.insert("known_vulnerabilities", doc! { "$exists": true, "$ne": [] });
+        } else {
+            query.insert("known_vulnerabilities", doc! { "$size": 0 });
+        }
+    }
+    if let Some(license) = &filter.license {
+        query.insert("license", license);
+    }
+
+    let skip = (filter.page.saturating_sub(1)) * filter.limit as u64;
     let total = db
         .sbom_entries()
-        .count_documents(doc! {})
+        .count_documents(query.clone())
         .await
         .unwrap_or(0);
 
     let entries = match db
         .sbom_entries()
-        .find(doc! {})
+        .find(query)
+        .sort(doc! { "name": 1 })
         .skip(skip)
-        .limit(params.limit)
+        .limit(filter.limit)
         .await
     {
         Ok(cursor) => collect_cursor_async(cursor).await,
@@ -346,7 +485,272 @@ pub async fn list_sbom(
     Ok(Json(ApiResponse {
         data: entries,
         total: Some(total),
-        page: Some(params.page),
+        page: Some(filter.page),
+    }))
+}
+
+pub async fn export_sbom(
+    Extension(agent): AgentExt,
+    Query(params): Query<SbomExportParams>,
+) -> Result<impl IntoResponse, StatusCode> {
+    let db = &agent.db;
+    let entries: Vec<SbomEntry> = match db
+        .sbom_entries()
+        .find(doc! { "repo_id": &params.repo_id })
+        .await
+    {
+        Ok(cursor) => collect_cursor_async(cursor).await,
+        Err(_) => Vec::new(),
+    };
+
+    let body = if params.format == "spdx" {
+        // SPDX 2.3 format
+        let packages: Vec<serde_json::Value> = entries
+            .iter()
+            .enumerate()
+            .map(|(i, e)| {
+                serde_json::json!({
+                    "SPDXID": format!("SPDXRef-Package-{i}"),
+                    "name": e.name,
+                    "versionInfo": e.version,
+                    "downloadLocation": "NOASSERTION",
+                    "licenseConcluded": e.license.as_deref().unwrap_or("NOASSERTION"),
+                    "externalRefs": e.purl.as_ref().map(|p| vec![serde_json::json!({
+                        "referenceCategory": "PACKAGE-MANAGER",
+                        "referenceType": "purl",
+                        "referenceLocator": p,
+                    })]).unwrap_or_default(),
+                })
+            })
+            .collect();
+
+        serde_json::json!({
+            "spdxVersion": "SPDX-2.3",
+            "dataLicense": "CC0-1.0",
+            "SPDXID": "SPDXRef-DOCUMENT",
+            "name": format!("sbom-{}", params.repo_id),
+            "documentNamespace": format!("https://compliance-scanner/sbom/{}", params.repo_id),
+            "packages": packages,
+        })
+    } else {
+        // CycloneDX 1.5 format
+        let components: Vec<serde_json::Value> = entries
+            .iter()
+            .map(|e| {
+                let mut comp = serde_json::json!({
+                    "type": "library",
+                    "name": e.name,
+                    "version": e.version,
+                    "group": e.package_manager,
+                });
+                if let Some(purl) = &e.purl {
+                    comp["purl"] = serde_json::Value::String(purl.clone());
+                }
+                if let Some(license) = &e.license {
+                    comp["licenses"] = serde_json::json!([{ "license": { "id": license } }]);
+                }
+                if !e.known_vulnerabilities.is_empty() {
+                    comp["vulnerabilities"] = serde_json::json!(
+                        e.known_vulnerabilities.iter().map(|v| serde_json::json!({
+                            "id": v.id,
+                            "source": { "name": v.source },
+                            "ratings": v.severity.as_ref().map(|s| vec![serde_json::json!({"severity": s})]).unwrap_or_default(),
+                        })).collect::<Vec<_>>()
+                    );
+                }
+                comp
+            })
+            .collect();
+
+        serde_json::json!({
+            "bomFormat": "CycloneDX",
+            "specVersion": "1.5",
+            "version": 1,
+            "metadata": {
+                "component": {
+                    "type": "application",
+                    "name": format!("repo-{}", params.repo_id),
+                }
+            },
+            "components": components,
+        })
+    };
+
+    let json_str =
+        serde_json::to_string_pretty(&body).map_err(|_| StatusCode::INTERNAL_SERVER_ERROR)?;
+    let filename = if params.format == "spdx" {
+        format!("sbom-{}-spdx.json", params.repo_id)
+    } else {
+        format!("sbom-{}-cyclonedx.json", params.repo_id)
+    };
+
+    let disposition = format!("attachment; filename=\"{filename}\"");
+    Ok((
+        [
+            (
+                header::CONTENT_TYPE,
+                header::HeaderValue::from_static("application/json"),
+            ),
+            (
+                header::CONTENT_DISPOSITION,
+                header::HeaderValue::from_str(&disposition)
+                    .unwrap_or_else(|_| header::HeaderValue::from_static("attachment")),
+            ),
+        ],
+        json_str,
+    ))
+}
+
+const COPYLEFT_LICENSES: &[&str] = &[
+    "GPL-2.0",
+    "GPL-2.0-only",
+    "GPL-2.0-or-later",
+    "GPL-3.0",
+    "GPL-3.0-only",
+    "GPL-3.0-or-later",
+    "AGPL-3.0",
+    "AGPL-3.0-only",
+    "AGPL-3.0-or-later",
+    "LGPL-2.1",
+    "LGPL-2.1-only",
+    "LGPL-2.1-or-later",
+    "LGPL-3.0",
+    "LGPL-3.0-only",
+    "LGPL-3.0-or-later",
+    "MPL-2.0",
+];
+
+pub async fn license_summary(
+    Extension(agent): AgentExt,
+    Query(params): Query<SbomFilter>,
+) -> ApiResult<Vec<LicenseSummary>> {
+    let db = &agent.db;
+    let mut query = doc! {};
+    if let Some(repo_id) = &params.repo_id {
+        query.insert("repo_id", repo_id);
+    }
+
+    let entries: Vec<SbomEntry> = match db.sbom_entries().find(query).await {
+        Ok(cursor) => collect_cursor_async(cursor).await,
+        Err(_) => Vec::new(),
+    };
+
+    let mut license_map: std::collections::HashMap<String, Vec<String>> =
+        std::collections::HashMap::new();
+    for entry in &entries {
+        let lic = entry.license.as_deref().unwrap_or("Unknown").to_string();
+        license_map.entry(lic).or_default().push(entry.name.clone());
+    }
+
+    let mut summaries: Vec<LicenseSummary> = license_map
+        .into_iter()
+        .map(|(license, packages)| {
+            let is_copyleft = COPYLEFT_LICENSES
+                .iter()
+                .any(|c| license.to_uppercase().contains(&c.to_uppercase()));
+            LicenseSummary {
+                license,
+                count: packages.len() as u64,
+                is_copyleft,
+                packages,
+            }
+        })
+        .collect();
+    summaries.sort_by(|a, b| b.count.cmp(&a.count));
+
+    Ok(Json(ApiResponse {
+        data: summaries,
+        total: None,
+        page: None,
+    }))
+}
+
+pub async fn sbom_diff(
+    Extension(agent): AgentExt,
+    Query(params): Query<SbomDiffParams>,
+) -> ApiResult<SbomDiffResult> {
+    let db = &agent.db;
+
+    let entries_a: Vec<SbomEntry> = match db
+        .sbom_entries()
+        .find(doc! { "repo_id": &params.repo_a })
+        .await
+    {
+        Ok(cursor) => collect_cursor_async(cursor).await,
+        Err(_) => Vec::new(),
+    };
+
+    let entries_b: Vec<SbomEntry> = match db
+        .sbom_entries()
+        .find(doc! { "repo_id": &params.repo_b })
+        .await
+    {
+        Ok(cursor) => collect_cursor_async(cursor).await,
+        Err(_) => Vec::new(),
+    };
+
+    // Build maps by (name, package_manager) -> version
+    let map_a: std::collections::HashMap<(String, String), String> = entries_a
+        .iter()
+        .map(|e| {
+            (
+                (e.name.clone(), e.package_manager.clone()),
+                e.version.clone(),
+            )
+        })
+        .collect();
+    let map_b: std::collections::HashMap<(String, String), String> = entries_b
+        .iter()
+        .map(|e| {
+            (
+                (e.name.clone(), e.package_manager.clone()),
+                e.version.clone(),
+            )
+        })
+        .collect();
+
+    let mut only_in_a = Vec::new();
+    let mut version_changed = Vec::new();
+    let mut common_count: u64 = 0;
+
+    for (key, ver_a) in &map_a {
+        match map_b.get(key) {
+            None => only_in_a.push(SbomDiffEntry {
+                name: key.0.clone(),
+                version: ver_a.clone(),
+                package_manager: key.1.clone(),
+            }),
+            Some(ver_b) if ver_a != ver_b => {
+                version_changed.push(SbomVersionDiff {
+                    name: key.0.clone(),
+                    package_manager: key.1.clone(),
+                    version_a: ver_a.clone(),
+                    version_b: ver_b.clone(),
+                });
+            }
+            Some(_) => common_count += 1,
+        }
+    }
+
+    let only_in_b: Vec<SbomDiffEntry> = map_b
+        .iter()
+        .filter(|(key, _)| !map_a.contains_key(key))
+        .map(|(key, ver)| SbomDiffEntry {
+            name: key.0.clone(),
+            version: ver.clone(),
+            package_manager: key.1.clone(),
+        })
+        .collect();
+
+    Ok(Json(ApiResponse {
+        data: SbomDiffResult {
+            only_in_a,
+            only_in_b,
+            version_changed,
+            common_count,
+        },
+        total: None,
+        page: None,
     }))
 }