From 0eb8d5168062a10b2e433dd3ea239cf37c3ada98 Mon Sep 17 00:00:00 2001
From: Bhaskar Brahma <bhbrahma@microsoft.com>
Date: Thu, 6 Oct 2022 12:02:55 -0700
Subject: [PATCH] Fixes ExtensionUtils.run_command_get_output() (#1604)

* Fixes ExtensionUtils.run_command_get_output()

ExtensionUtils.run_command_get_output() now decodes the output of
commands into a utf-8 string.

Fixes VMAccess extension's inability to accept non latin characters as
password.

* Addressed PR comments

* Updated extension version
---
 .vscode/launch.json                 | 20 ++++++++++++++++++++
 Utils/extensionutils.py             | 10 +++++-----
 Utils/test/non_latin_characters.txt |  1 +
 Utils/test/test_encode.py           | 15 +++++++++++++++
 VMAccess/manifest.xml               |  2 +-
 5 files changed, 42 insertions(+), 6 deletions(-)
 create mode 100644 .vscode/launch.json
 create mode 100644 Utils/test/non_latin_characters.txt

diff --git a/.vscode/launch.json b/.vscode/launch.json
new file mode 100644
index 000000000..e55b75059
--- /dev/null
+++ b/.vscode/launch.json
@@ -0,0 +1,20 @@
+{
+    // Use IntelliSense to learn about possible attributes.
+    // Hover to view descriptions of existing attributes.
+    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
+    "version": "0.2.0",
+    "configurations": [
+        {
+            "name": "Python: test_encode.py",
+            "type": "python",
+            "request": "launch",
+            "program": "test_encode.py",
+            "console": "integratedTerminal",
+            "justMyCode": true,
+            "cwd": "${workspaceFolder}/Utils/test",
+            "env" : {
+                "PYTHONPATH": "${workspaceFolder}"
+            }
+        }
+    ]
+}
\ No newline at end of file
diff --git a/Utils/extensionutils.py b/Utils/extensionutils.py
index 2eac5885e..f0393a228 100644
--- a/Utils/extensionutils.py
+++ b/Utils/extensionutils.py
@@ -192,15 +192,15 @@ def run_command_get_output(cmd, chk_err=True, log_cmd=True):
             logger.error('CalledProcessError.  Error Code is ' + str(e.returncode))
             logger.error('CalledProcessError.  Command string was ' + str(cmd))
             logger.error(
-                'CalledProcessError.  Command result was ' + (e.output[:-1]).decode('latin-1'))
-        return e.returncode, e.output.decode('latin-1')
+                'CalledProcessError.  Command result was ' + (e.output[:-1]).decode('utf-8'))
+        return e.returncode, e.output.decode('utf-8')
     except EnvironmentError as e:
         if chk_err and log_cmd:
             logger.error(
                 'CalledProcessError.  Error message is ' + str(e))
             return e.errno, str(e)
     # noinspection PyUnboundLocalVariable
-    return 0, output.decode('latin-1')
+    return 0, output.decode('utf-8')
 
 
 def run(cmd, chk_err=True):
@@ -238,8 +238,8 @@ def run_send_stdin(cmd, cmd_input, chk_err=True, log_cmd=True):
         logger.error('CalledProcessError.  Error Code is ' + str(me.returncode))
         logger.error('CalledProcessError.  Command was ' + str(cmd))
         logger.error(
-            'CalledProcessError.  Command result was ' + output[0].decode('latin-1'))
-    return me.returncode, output[0].decode('latin-1')
+            'CalledProcessError.  Command result was ' + output[0].decode('utf-8'))
+    return me.returncode, output[0].decode('utf-8')
 
 
 def get_line_starting_with(prefix, filepath):
diff --git a/Utils/test/non_latin_characters.txt b/Utils/test/non_latin_characters.txt
new file mode 100644
index 000000000..99fe40df2
--- /dev/null
+++ b/Utils/test/non_latin_characters.txt
@@ -0,0 +1 @@
+ü
\ No newline at end of file
diff --git a/Utils/test/test_encode.py b/Utils/test/test_encode.py
index 7808e8edf..e9de56a20 100644
--- a/Utils/test/test_encode.py
+++ b/Utils/test/test_encode.py
@@ -26,6 +26,21 @@ def test_encode(self):
         known_non_ascii_character = b"%c" % encoded_contents[2353]
         self.assertEqual(known_non_ascii_character, b'\x9d')
 
+class TestRunCommandGetOutput(unittest.TestCase):
+    def test_output(self):
+        cmd = ["cat", "non_latin_characters.txt"]
+        return_code, output_string = eu.run_command_get_output(cmd)
+        self.assertEqual(0, return_code)
+        expected_character_byte = b'\xc3\xbc'
+        expected_character = expected_character_byte.decode("utf-8")
+        self.assertEqual(expected_character, output_string[0])
+
+    def test_stdin(self):
+        cmd = ['bash', '-c', 'read ; echo $REPLY']
+        cmd_input = b'\xc3\xbc' # ü character
+        return_code, output_string = eu.run_send_stdin(cmd, cmd_input)
+        self.assertEqual(0, return_code)
+        self.assertEqual(cmd_input.decode('utf-8'), output_string[0])
 
 if __name__ == '__main__':
     unittest.main()
diff --git a/VMAccess/manifest.xml b/VMAccess/manifest.xml
index 407adbaaa..396b0248d 100644
--- a/VMAccess/manifest.xml
+++ b/VMAccess/manifest.xml
@@ -2,7 +2,7 @@
 <ExtensionImage xmlns="http://schemas.microsoft.com/windowsazure">
   <ProviderNameSpace>Microsoft.OSTCExtensions</ProviderNameSpace>
   <Type>VMAccessForLinux</Type>
-  <Version>1.5.11</Version>
+  <Version>1.5.12</Version>
   <Label>Microsoft Azure VM Access Extension for Linux Virtual Machines</Label>
   <HostingResources>VmRole</HostingResources>
   <MediaLink></MediaLink>