From 8918bdc38388527f06661293dd7e3ccb304cd014 Mon Sep 17 00:00:00 2001
From: "Mr. Outis" <mroutis@protonmail.com>
Date: Mon, 19 Aug 2019 16:47:15 -0500
Subject: [PATCH] http: add exception when trying to traverse remote (#2411)

* http: add exception when trying to traverse remote

Fix #2380

* http: exception raise on `list_cache_paths` instead of `all`

`NotImplementedError` -> `RemoteActionNotImplemented`

* http: raise exception on `__init__`

* set default no_traverse to True on remote base

* tests: explicitly use traverse on data cloud and no traverse with HTTP

* remote: define a new RemoteConfigError

* remote: reuse ConfigError instead of RemoteConfigError

* tests: set no_traverse to False with SSHMocked

* deepsource: fix minor issue with spacing

* remove blankspace

Co-Authored-By: Ruslan Kuprieiev <kupruser@gmail.com>
---
 dvc/remote/base.py             |  2 +-
 dvc/remote/http.py             |  8 +++++++-
 tests/func/test_data_cloud.py  | 15 +++++++++++++++
 tests/unit/remote/test_http.py | 14 ++++++++++++++
 4 files changed, 37 insertions(+), 2 deletions(-)
 create mode 100644 tests/unit/remote/test_http.py

diff --git a/dvc/remote/base.py b/dvc/remote/base.py
index 050c16659..88c52fbd3 100644
--- a/dvc/remote/base.py
+++ b/dvc/remote/base.py
@@ -116,7 +116,7 @@ class RemoteBASE(object):
         )
 
         self.protected = False
-        self.no_traverse = config.get(Config.SECTION_REMOTE_NO_TRAVERSE)
+        self.no_traverse = config.get(Config.SECTION_REMOTE_NO_TRAVERSE, True)
         self._dir_info = {}
 
     def __repr__(self):
diff --git a/dvc/remote/http.py b/dvc/remote/http.py
index f377f0ddc..300d0f896 100644
--- a/dvc/remote/http.py
+++ b/dvc/remote/http.py
@@ -10,7 +10,7 @@ import logging
 
 from dvc.progress import progress
 from dvc.exceptions import DvcException
-from dvc.config import Config
+from dvc.config import Config, ConfigError
 from dvc.remote.base import RemoteBASE
 
 
@@ -42,6 +42,12 @@ class RemoteHTTP(RemoteBASE):
         url = config.get(Config.SECTION_REMOTE_URL)
         self.path_info = self.path_cls(url) if url else None
 
+        if not self.no_traverse:
+            raise ConfigError(
+                "HTTP doesn't support traversing the remote to list existing "
+                "files. Use: `dvc remote modify <name> no_traverse true`"
+            )
+
     def _download(self, from_info, to_file, name=None, no_progress_bar=False):
         callback = None
         if not no_progress_bar:
diff --git a/tests/func/test_data_cloud.py b/tests/func/test_data_cloud.py
index f7bf341fd..780755c73 100644
--- a/tests/func/test_data_cloud.py
+++ b/tests/func/test_data_cloud.py
@@ -457,6 +457,21 @@ class TestRemoteSSHMocked(TestDataCloudBase):
         self.ssh_server = ssh_server
         self.method_name = request.function.__name__
 
+    def _setup_cloud(self):
+        self._ensure_should_run()
+
+        repo = self._get_url()
+        keyfile = self._get_keyfile()
+
+        config = copy.deepcopy(TEST_CONFIG)
+        config[TEST_SECTION][Config.SECTION_REMOTE_URL] = repo
+        config[TEST_SECTION][Config.SECTION_REMOTE_KEY_FILE] = keyfile
+        config[TEST_SECTION][Config.SECTION_REMOTE_NO_TRAVERSE] = False
+        self.dvc.config.config = config
+        self.cloud = DataCloud(self.dvc)
+
+        self.assertIsInstance(self.cloud.get_remote(), self._get_cloud_class())
+
     def _get_url(self):
         user = self.ssh_server.test_creds["username"]
         return get_ssh_url_mocked(user, self.ssh_server.port)
diff --git a/tests/unit/remote/test_http.py b/tests/unit/remote/test_http.py
new file mode 100644
index 000000000..b2d82bd8e
--- /dev/null
+++ b/tests/unit/remote/test_http.py
@@ -0,0 +1,14 @@
+import pytest
+from dvc.remote.http import RemoteHTTP
+from dvc.config import ConfigError
+
+
+def test_no_traverse_compatibility(dvc_repo):
+    config = {
+        "url": "http://example.com/",
+        "path_info": "file.html",
+        "no_traverse": False,
+    }
+
+    with pytest.raises(ConfigError):
+        RemoteHTTP(dvc_repo, config)