[LLDB] Update Shell lit config to handle c8031c3dd743

[bulbazord]

No description provided.

[llvmbot]

@llvm/pr-subscribers-lldb

Author: Alex Langford (bulbazord)

Changes

---

Full diff: https://github.com/llvm/llvm-project/pull/170225.diff

1 Files Affected:

• (modified) lldb/test/Shell/helper/toolchain.py (+1-1)

diff --git a/lldb/test/Shell/helper/toolchain.py b/lldb/test/Shell/helper/toolchain.py
index b9e7dd7c196ab..b0d4e272d5646 100644
--- a/lldb/test/Shell/helper/toolchain.py
+++ b/lldb/test/Shell/helper/toolchain.py
@@ -226,7 +226,7 @@ def use_support_substitutions(config):
         except OSError:
             res = -1
         if res == 0 and out:
-            sdk_path = lit.util.to_string(out)
+            sdk_path = str(out)
             llvm_config.lit_config.note("using SDKROOT: %r" % sdk_path)
             host_flags += ["-isysroot", sdk_path]
     elif sys.platform != "win32":

[medismailben]

🚢

[felipepiovezan]

This is breaking all the shell tests in CI, so I'm going to revert it for now.

[delcypher]

@bulbazord / @medismailben This fix wasn't quite right. Here's an example showing why:

>>> test = b'xxx'
>>> type(test)
<class 'bytes'>
>>> str(test)
"b'xxx'"
>>> test.decode()
'xxx'

If you do str() on a bytes object you end up with b'...' in the string because of the object being a bytes object. You need to call .decode() on the bytes object to get the str representation.

[medismailben]

@bulbazord / @medismailben This fix wasn't quite right. Here's an example showing why:

>>> test = b'xxx'
>>> type(test)
<class 'bytes'>
>>> str(test)
"b'xxx'"
>>> test.decode()
'xxx'

If you do str() on a bytes object you end up with b'...' in the string because of the object being a bytes object. You need to call .decode() on the bytes object to get the str representation.

Good catch, @felipepiovezan already reverted this and landed a proper fix 😅