[lit] Add readfile substitution

This patch adds a new %{readfile:<file name>} substitution to lit. This
is needed for porting a couple of tests to lit's internal shell. These
tests are all using subshells to pass some option to a command are not
feasible to run within the internal shell without this functionality.

Reviewers: petrhosek, jh7370, ilovepi, cmtice

Reviewed By: jh7370, cmtice

Pull Request: llvm/llvm-project#158441

Author: boomanaiden154

lit/TestRunner.py

@@ -720,6 +720,30 @@ def processRedirects(cmd, stdin_source, cmd_shenv, opened_files):
     return std_fds
 
 
+def _expandLateSubstitutions(cmd, arguments, cwd):
+    for i, arg in enumerate(arguments):
+        if not isinstance(arg, str):
+            continue
+
+        def _replaceReadFile(match):
+            filePath = match.group(1)
+            if not os.path.isabs(filePath):
+                filePath = os.path.join(cwd, filePath)
+            try:
+                with open(filePath) as fileHandle:
+                    return fileHandle.read()
+            except FileNotFoundError:
+                raise InternalShellError(
+                    cmd,
+                    "File specified in readfile substitution does not exist: %s"
+                    % filePath,
+                )
+
+        arguments[i] = re.sub(r"%{readfile:([^}]*)}", _replaceReadFile, arg)
+
+    return arguments
+
+
 def _executeShCmd(cmd, shenv, results, timeoutHelper):
     if timeoutHelper.timeoutReached():
         # Prevent further recursion if the timeout has been hit
@@ -834,6 +858,9 @@ def _executeShCmd(cmd, shenv, results, timeoutHelper):
         # Ensure args[0] is hashable.
         args[0] = expand_glob(args[0], cmd_shenv.cwd)[0]
 
+        # Expand all late substitutions.
+        args = _expandLateSubstitutions(j, args, cmd_shenv.cwd)
+
         inproc_builtin = inproc_builtins.get(args[0], None)
         if inproc_builtin and (args[0] != "echo" or len(cmd.commands) == 1):
             # env calling an in-process builtin is useless, so we take the safe

tests/Inputs/shtest-readfile/absolute-paths.txt

@@ -0,0 +1,6 @@
+## Tests that readfile works with absolute paths.
+# RUN: echo -n "hello" > %t
+# RUN: echo %{readfile:%t}
+
+## Fail the test so we can assert on the output.
+# RUN: not echo return

tests/Inputs/shtest-readfile/file-does-not-exist.txt

@@ -0,0 +1,4 @@
+## Test that readfile reports information appropriately when the file specified
+## does not exist.
+
+# RUN: echo %{readfile:/file/does/not/exist}

tests/Inputs/shtest-readfile/lit.cfg

@@ -0,0 +1,7 @@
+import lit.formats
+
+config.name = "shtest-readfile"
+config.suffixes = [".txt"]
+config.test_format = lit.formats.ShTest(execute_external=False)
+config.test_source_root = None
+config.test_exec_root = None

tests/Inputs/shtest-readfile/relative-paths.txt

@@ -0,0 +1,7 @@
+## Tests that readfile works with relative paths.
+# RUN: mkdir -p rel_path_test_folder
+# RUN: echo -n "hello" > rel_path_test_folder/test_file
+# RUN: echo %{readfile:rel_path_test_folder/test_file}
+
+## Fail the test so we can assert on the output.
+# RUN: not echo return

tests/Inputs/shtest-readfile/two-same-line.txt

@@ -0,0 +1,8 @@
+## Tests that readfile works with two substitutions on the same line to ensure the
+## regular expressions are set up correctly.
+# RUN: echo -n "hello" > %t.1
+# RUN: echo -n "bye" > %t.2
+# RUN: echo %{readfile:%t.1} %{readfile:%t.2}
+
+## Fail the test so we can assert on the output.
+# RUN: not echo return

tests/shtest-readfile.py

@@ -0,0 +1,21 @@
+## Tests the readfile substitution.
+
+# RUN: not %{lit} -a -v %{inputs}/shtest-readfile | FileCheck -match-full-lines -DTEMP_PATH=%S%{fs-sep}Inputs%{fs-sep}shtest-readfile%{fs-sep}Output %s
+
+# CHECK: -- Testing: 4 tests{{.*}}
+
+# CHECK-LABEL: FAIL: shtest-readfile :: absolute-paths.txt ({{[^)]*}})
+# CHECK: echo hello
+# CHECK: # executed command: echo '%{readfile:[[TEMP_PATH]]{{[\\\/]}}absolute-paths.txt.tmp}'
+
+# CHECK-LABEL: FAIL: shtest-readfile :: file-does-not-exist.txt ({{[^)]*}})
+# CHECK: # executed command: @echo 'echo %{readfile:/file/does/not/exist}'
+# CHECK: # | File specified in readfile substitution does not exist: /file/does/not/exist
+
+# CHECK-LABEL: FAIL: shtest-readfile :: relative-paths.txt ({{[^)]*}})
+# CHECK: echo hello
+# CHECK: # executed command: echo '%{readfile:rel_path_test_folder/test_file}'
+
+# CHECK-LABEL: FAIL: shtest-readfile :: two-same-line.txt ({{[^)]*}})
+# CHECK: echo hello bye
+# CHECK: # executed command: echo '%{readfile:[[TEMP_PATH]]{{[\\\/]}}two-same-line.txt.tmp.1}' '%{readfile:[[TEMP_PATH]]{{[\\\/]}}two-same-line.txt.tmp.2}'