[fix](tvf) Restrict local paths and require explicit startup opt-in

be/src/io/fs/local_file_system.cpp

@@ -421,28 +421,48 @@ const std::shared_ptr<LocalFileSystem>& global_local_filesystem() {
 Status LocalFileSystem::canonicalize_local_file(const std::string& dir,
                                                 const std::string& file_path,
                                                 std::string* full_path) {
-    const std::string absolute_path = dir + "/" + file_path;
+    // An empty root would turn a relative request into an unrestricted absolute path.
+    if (dir.empty()) {
+        return Status::InvalidArgument("user_files_secure_path must not be empty");
+    }
+    // Reject embedded NULs before passing paths to filesystem APIs that use C strings.
+    if (file_path.empty() || file_path.find('\0') != std::string::npos) {
+        return Status::InvalidArgument("local file path must not be empty or contain NUL");
+    }
+    Path local_path;
+    // Convert file: URIs to filesystem paths and reject remote schemes before any reader opens them.
+    RETURN_IF_ERROR(convert_to_abs_path(file_path, local_path));
+    std::string secure_path;
+    RETURN_IF_ERROR(canonicalize(dir, &secure_path));
     std::string canonical_path;
-    RETURN_IF_ERROR(canonicalize(absolute_path, &canonical_path));
-    if (!contain_path(dir, canonical_path)) {
-        return Status::InvalidArgument("file path is not allowed: {}", canonical_path);
+    const auto status = canonicalize(Path(secure_path) / local_path, &canonical_path);
+    // A glob result can reveal external names even when resolution fails on a dangling symlink.
+    // Reject both failed resolution and out-of-root targets without exposing expanded names.
+    if (!status.ok() || !contain_path(secure_path, canonical_path)) {
+        return Status::InvalidArgument("file path is not allowed");
     }
-
-    *full_path = canonical_path;
+    *full_path = std::move(canonical_path);
     return Status::OK();
 }
 
 Status LocalFileSystem::safe_glob(const std::string& path, std::vector<FileInfo>* res) {
     if (path.find("..") != std::string::npos) {
         return Status::InvalidArgument("can not contain '..' in path");
     }
-    std::string full_path = config::user_files_secure_path + "/" + path;
+    // An empty root would turn the prefix below into "/" and allow any local file.
+    if (config::user_files_secure_path.empty()) {
+        return Status::InvalidArgument("user_files_secure_path must not be empty");
+    }
+    std::string secure_path;
+    RETURN_IF_ERROR(canonicalize(config::user_files_secure_path, &secure_path));
+    std::string full_path = secure_path + "/" + path;
     std::vector<std::string> files;
     RETURN_IF_ERROR(_glob(full_path, &files));
     for (auto& file : files) {
         FileInfo fi;
         fi.is_file = true;
-        RETURN_IF_ERROR(canonicalize_local_file("", file, &(fi.file_name)));
+        // Glob classes and symlinks can escape without a literal ".." in the pattern.
+        RETURN_IF_ERROR(canonicalize_local_file(secure_path, file, &(fi.file_name)));
         RETURN_IF_ERROR(file_size_impl(fi.file_name, &(fi.file_size)));
         res->push_back(std::move(fi));
     }

be/src/io/fs/local_file_system.h

@@ -66,6 +66,8 @@ class LocalFileSystem final : public FileSystem {
 
     static std::filesystem::perms PERMS_OWNER_RW;
 
+    // Resolve a local path relative to a nonempty secure root, then check canonical containment.
+    // Absolute paths and file: URIs must resolve inside the same root; remote schemes are rejected.
     Status canonicalize_local_file(const std::string& dir, const std::string& file_path,
                                    std::string* full_path);
 

be/test/io/fs/local_file_system_test.cpp

@@ -29,11 +29,13 @@
 #include <filesystem>
 #include <vector>
 
+#include "common/config.h"
 #include "common/status.h"
 #include "cpp/sync_point.h"
 #include "gtest/gtest_pred_impl.h"
 #include "io/fs/file_reader.h"
 #include "io/fs/file_writer.h"
+#include "util/defer_op.h"
 #include "util/slice.h"
 
 namespace doris {
@@ -386,6 +388,10 @@ TEST_F(LocalFileSystemTest, AbnormalWriteRead) {
 }
 
 TEST_F(LocalFileSystemTest, TestGlob) {
+    const std::string original_secure_path = config::user_files_secure_path;
+    DEFER(config::user_files_secure_path = original_secure_path);
+    // Bind this fixture to its own files, independent of the runner's build-profile directory.
+    config::user_files_secure_path = std::filesystem::absolute("./be/ut_build_ASAN/test").string();
     std::string path = "./be/ut_build_ASAN/test/file_path/";
     EXPECT_TRUE(io::global_local_filesystem()->delete_directory(path).ok());
     EXPECT_TRUE(io::global_local_filesystem()
@@ -417,6 +423,101 @@ TEST_F(LocalFileSystemTest, TestGlob) {
     EXPECT_TRUE(io::global_local_filesystem()->delete_directory(path).ok());
 }
 
+TEST_F(LocalFileSystemTest, TestSafeGlobPathContainment) {
+    auto fs = io::global_local_filesystem();
+    const std::string original_secure_path = config::user_files_secure_path;
+    DEFER(config::user_files_secure_path = original_secure_path);
+    const auto root = std::filesystem::canonical(test_dir);
+    const auto secure = root / "secure";
+    const auto outside = root / "secure-other";
+    std::filesystem::create_directories(secure / "nested");
+    std::filesystem::create_directory(outside);
+    ASSERT_TRUE(save_string_file((secure / "nested/inside.csv").string(), "inside").ok());
+    ASSERT_TRUE(save_string_file((outside / "outside.csv").string(), "outside").ok());
+    std::filesystem::create_directory_symlink(outside, secure / "escape");
+    std::filesystem::create_symlink(outside / "outside.csv", secure / "outside.csv");
+    std::filesystem::create_symlink(outside / "missing.csv", outside / "broken.csv");
+    std::filesystem::create_directory_symlink(secure / "nested", secure / "inside");
+    config::user_files_secure_path = secure.string();
+
+    for (const auto* pattern : {".[.]/secure-other/outside.csv", ".?/secure-other/outside.csv",
+                                ".[.-.]/secure-other/outside.csv", "escape/*.csv", "outside.csv",
+                                ".[.]/secure-other/*.csv", "escape/broken*.csv"}) {
+        std::vector<io::FileInfo> files;
+        const auto status = fs->safe_glob(pattern, &files);
+        EXPECT_FALSE(status.ok()) << pattern;
+        EXPECT_NE(status.to_string().find("file path is not allowed"), std::string::npos) << status;
+        // The pattern need not name the matched file, so errors must not reveal that name.
+        const auto message = status.to_string();
+        EXPECT_EQ(std::string::npos, message.find(root.string())) << message;
+        for (const auto* name : {"outside.csv", "broken.csv", "missing.csv"}) {
+            EXPECT_EQ(std::string::npos, message.find(name)) << message;
+        }
+        EXPECT_TRUE(files.empty());
+    }
+
+    for (const auto* pattern : {"nested/*.csv", "inside/*.csv"}) {
+        std::vector<io::FileInfo> files;
+        ASSERT_TRUE(fs->safe_glob(pattern, &files).ok()) << pattern;
+        ASSERT_EQ(1, files.size());
+        EXPECT_EQ((secure / "nested/inside.csv").string(), files[0].file_name);
+        EXPECT_EQ(6, files[0].file_size);
+    }
+}
+
+TEST_F(LocalFileSystemTest, TestSafeGlobSecureRoot) {
+    auto fs = io::global_local_filesystem();
+    const std::string original_secure_path = config::user_files_secure_path;
+    DEFER(config::user_files_secure_path = original_secure_path);
+    const auto root = std::filesystem::canonical(test_dir);
+    std::filesystem::create_directory(root / "secure");
+    std::filesystem::create_directory_symlink(root / "secure", root / "secure-link");
+    ASSERT_TRUE(save_string_file((root / "secure/inside.csv").string(), "inside").ok());
+
+    for (const auto& secure : {std::string(test_dir) + "/secure", (root / "secure/").string(),
+                               (root / "secure-link").string()}) {
+        config::user_files_secure_path = secure;
+        std::vector<io::FileInfo> files;
+        ASSERT_TRUE(fs->safe_glob("*.csv", &files).ok()) << secure;
+        ASSERT_EQ(1, files.size());
+        EXPECT_EQ((root / "secure/inside.csv").string(), files[0].file_name);
+    }
+    config::user_files_secure_path = "";
+    std::vector<io::FileInfo> files;
+    EXPECT_FALSE(fs->safe_glob("*.csv", &files).ok());
+    config::user_files_secure_path = (root / "missing").string();
+    EXPECT_FALSE(fs->safe_glob("*.csv", &files).ok());
+}
+
+TEST_F(LocalFileSystemTest, TestCanonicalizeSecureLocalPath) {
+    auto fs = io::global_local_filesystem();
+    const auto root = std::filesystem::canonical(test_dir);
+    const auto secure = root / "secure";
+    std::filesystem::create_directories(secure / "nested");
+    std::filesystem::create_directory(root / "secure-other");
+    std::filesystem::create_directory_symlink(root / "secure-other", secure / "escape");
+    std::filesystem::create_directory_symlink(secure / "nested", secure / "inside");
+    for (const auto& path :
+         {std::string("nested"), std::string("nested/../inside"), (secure / "nested").string(),
+          "file://" + (secure / "nested").string()}) {
+        std::string resolved;
+        ASSERT_TRUE(fs->canonicalize_local_file(secure.string(), path, &resolved).ok()) << path;
+        EXPECT_EQ((secure / "nested").string(), resolved);
+    }
+    for (const auto& path :
+         {std::string("../secure-other"), std::string("escape"), (root / "secure-other").string(),
+          std::string("https://invalid/dataset"), std::string("s3://bucket/dataset"),
+          std::string("http://invalid/dataset"), std::string(""),
+          std::string("nested\0ignored", 14)}) {
+        std::string resolved;
+        const auto status = fs->canonicalize_local_file(secure.string(), path, &resolved);
+        EXPECT_FALSE(status.ok()) << path;
+        EXPECT_EQ(std::string::npos, status.to_string().find(root.string())) << status;
+    }
+    std::string resolved;
+    EXPECT_FALSE(fs->canonicalize_local_file("", secure.string(), &resolved).ok());
+}
+
 TEST_F(LocalFileSystemTest, TestConvertToAbsPath) {
     io::Path abs_path;
     Status st;

fe/fe-common/src/main/java/org/apache/doris/common/Config.java

@@ -2732,6 +2732,10 @@ public class Config extends ConfigBase {
     })
     public static boolean enable_java_udf = true;
 
+    @ConfField(description = "Whether to enable LOCAL TVF and FILE TVF with local storage. Disabled by default. "
+            + "Set this in fe.conf and restart every FE to enable it. Existing access checks still apply.")
+    public static boolean enable_local_tvf = false;
+
     @ConfField(mutable = true, masterOnly = true, description = {
         "开启后，可以在导入时，利用创建的全局java_udf函数处理数据, 默认为false。",
         "When enabled, data can be processed using the globally created java_udf function during import."

fe/fe-core/src/main/java/org/apache/doris/tablefunction/LocalTableValuedFunction.java

@@ -21,8 +21,11 @@
 import org.apache.doris.analysis.StorageBackend.StorageType;
 import org.apache.doris.catalog.Env;
 import org.apache.doris.common.AnalysisException;
+import org.apache.doris.common.Config;
+import org.apache.doris.common.ErrorCode;
 import org.apache.doris.common.UserException;
 import org.apache.doris.datasource.property.storage.StorageProperties;
+import org.apache.doris.mysql.privilege.PrivPredicate;
 import org.apache.doris.proto.InternalService;
 import org.apache.doris.proto.InternalService.PGlobResponse;
 import org.apache.doris.proto.InternalService.PGlobResponse.PFileInfo;
@@ -62,6 +65,14 @@ public class LocalTableValuedFunction extends ExternalFileTableValuedFunction {
     private boolean sharedStorage = false;
 
     public LocalTableValuedFunction(Map<String, String> properties) throws AnalysisException {
+        if (!Config.enable_local_tvf) {
+            throw new org.apache.doris.nereids.exceptions.AnalysisException(
+                    "LOCAL TVF is disabled. Set enable_local_tvf=true in fe.conf and restart every FE to enable it.");
+        }
+
+        // Check privileges before listing files or discovering their schema.
+        checkAuth(ConnectContext.get());
+
         // 1. analyze common properties
         Map<String, String> props = super.parseCommonProperties(properties);
 
@@ -132,6 +143,16 @@ private void getFileListFromBackend() throws AnalysisException {
         }
     }
 
+    @Override
+    public void checkAuth(ConnectContext ctx) {
+        if (!Env.getCurrentEnv().getAccessManager().checkGlobalPriv(ctx, PrivPredicate.ADMIN)
+                && !Env.getCurrentEnv().getAccessManager().checkGlobalPriv(ctx, PrivPredicate.OPERATOR)) {
+            String message = ErrorCode.ERR_SPECIFIC_ACCESS_DENIED_ERROR.formatErrorMsg("ADMIN/OPERATOR");
+            throw new org.apache.doris.nereids.exceptions.AnalysisException(message);
+        }
+        super.checkAuth(ctx);
+    }
+
     @Override
     public TFileType getTFileType() {
         return TFileType.FILE_LOCAL;