[llvm-symbolizer] Switch command line parsing from llvm::cl to OptTable

for the advantage outlined by D83639 ([OptTable] Support grouped short options)

Some behavior changes:

* -i={0,false} is removed. Use --no-inlines instead.
* --demangle={0,false} is removed. Use --no-demangle instead
* -untag-addresses={0,false} is removed. Use --no-untag-addresses instead

Added a higher level API OptTable::parseArgs which handles optional
initial options populated from an environment variable, expands response
files recursively, and parses options.

Reviewed By: jhenderson

Differential Revision: https://reviews.llvm.org/D83530

Author: MaskRay

llvm/docs/CommandGuide/llvm-symbolizer.rst

@@ -220,16 +220,16 @@ OPTIONS
 
   Show help and usage for this command.
 
-.. option:: --help-list
-
-  Show help and usage for this command without grouping the options into categories.
-
 .. _llvm-symbolizer-opt-i:
 
 .. option:: --inlining, --inlines, -i
 
   If a source code location is in an inlined function, prints all the inlined
-  frames. Defaults to true.
+  frames. This is the default.
+
+.. option:: --no-inlines
+
+  Don't print inlined frames.
 
 .. option:: --no-demangle
 
@@ -267,17 +267,17 @@ OPTIONS
 
     foo() at /tmp/test.cpp:6:3
 
-    $ llvm-symbolizer --output-style=LLVM --obj=inlined.elf 0x4004be 0x400486 -p -i=0
+    $ llvm-symbolizer --output-style=LLVM --obj=inlined.elf 0x4004be 0x400486 -p --no-inlines
     main at /tmp/test.cpp:11:18
 
     foo() at /tmp/test.cpp:6:3
 
-    $ llvm-symbolizer --output-style=GNU --obj=inlined.elf 0x4004be 0x400486 -p -i=0
+    $ llvm-symbolizer --output-style=GNU --obj=inlined.elf 0x4004be 0x400486 -p --no-inlines
     baz() at /tmp/test.cpp:11
     foo() at /tmp/test.cpp:6
 
     $ clang -g -fdebug-info-for-profiling test.cpp -o profiling.elf
-    $ llvm-symbolizer --output-style=GNU --obj=profiling.elf 0x401167 -p -i=0
+    $ llvm-symbolizer --output-style=GNU --obj=profiling.elf 0x401167 -p --no-inlines
     main at /tmp/test.cpp:15 (discriminator 2)
 
 .. option:: --pretty-print, -p

llvm/include/llvm/Option/OptTable.h

@@ -13,13 +13,15 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/StringSet.h"
 #include "llvm/Option/OptSpecifier.h"
+#include "llvm/Support/StringSaver.h"
 #include <cassert>
 #include <string>
 #include <vector>
 
 namespace llvm {
 
 class raw_ostream;
+template <typename Fn> class function_ref;
 
 namespace opt {
 
@@ -60,6 +62,7 @@ class OptTable {
   std::vector<Info> OptionInfos;
   bool IgnoreCase;
   bool GroupedShortOptions = false;
+  const char *EnvVar = nullptr;
 
   unsigned TheInputOptionID = 0;
   unsigned TheUnknownOptionID = 0;
@@ -123,6 +126,9 @@ class OptTable {
     return getInfo(id).MetaVar;
   }
 
+  /// Specify the environment variable where initial options should be read.
+  void setInitialOptionsFromEnvironment(const char *E) { EnvVar = E; }
+
   /// Support grouped short options. e.g. -ab represents -a -b.
   void setGroupedShortOptions(bool Value) { GroupedShortOptions = Value; }
 
@@ -219,6 +225,18 @@ class OptTable {
                          unsigned &MissingArgCount, unsigned FlagsToInclude = 0,
                          unsigned FlagsToExclude = 0) const;
 
+  /// A convenience helper which handles optional initial options populated from
+  /// an environment variable, expands response files recursively and parses
+  /// options.
+  ///
+  /// \param ErrorFn - Called on a formatted error message for missing arguments
+  /// or unknown options.
+  /// \return An InputArgList; on error this will contain all the options which
+  /// could be parsed.
+  InputArgList parseArgs(int Argc, char *const *Argv, OptSpecifier Unknown,
+                         StringSaver &Saver,
+                         function_ref<void(StringRef)> ErrorFn) const;
+
   /// Render the help text for an option table.
   ///
   /// \param OS - The stream to write the help text to.

llvm/include/llvm/Support/CommandLine.h

@@ -2085,6 +2085,14 @@ bool ExpandResponseFiles(
     llvm::vfs::FileSystem &FS = *llvm::vfs::getRealFileSystem(),
     llvm::Optional<llvm::StringRef> CurrentDir = llvm::None);
 
+/// A convenience helper which concatenates the options specified by the
+/// environment variable EnvVar and command line options, then expands response
+/// files recursively. The tokenizer is a predefined GNU or Windows one.
+/// \return true if all @files were expanded successfully or there were none.
+bool expandResponseFiles(int Argc, const char *const *Argv, const char *EnvVar,
+                         StringSaver &Saver,
+                         SmallVectorImpl<const char *> &NewArgv);
+
 /// Mark all options not part of this category as cl::ReallyHidden.
 ///
 /// \param Category the category of options to keep displaying

llvm/lib/Option/OptTable.cpp

@@ -6,14 +6,15 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include "llvm/Option/OptTable.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/StringSet.h"
 #include "llvm/Option/Arg.h"
 #include "llvm/Option/ArgList.h"
-#include "llvm/Option/Option.h"
 #include "llvm/Option/OptSpecifier.h"
-#include "llvm/Option/OptTable.h"
+#include "llvm/Option/Option.h"
+#include "llvm/Support/CommandLine.h" // for expandResponseFiles
 #include "llvm/Support/Compiler.h"
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/raw_ostream.h"
@@ -490,6 +491,33 @@ InputArgList OptTable::ParseArgs(ArrayRef<const char *> ArgArr,
   return Args;
 }
 
+InputArgList OptTable::parseArgs(int Argc, char *const *Argv,
+                                 OptSpecifier Unknown, StringSaver &Saver,
+                                 function_ref<void(StringRef)> ErrorFn) const {
+  SmallVector<const char *, 0> NewArgv;
+  // The environment variable specifies initial options which can be overridden
+  // by commnad line options.
+  cl::expandResponseFiles(Argc, Argv, EnvVar, Saver, NewArgv);
+
+  unsigned MAI, MAC;
+  opt::InputArgList Args = ParseArgs(makeArrayRef(NewArgv), MAI, MAC);
+  if (MAC)
+    ErrorFn((Twine(Args.getArgString(MAI)) + ": missing argument").str());
+
+  // For each unknwon option, call ErrorFn with a formatted error message. The
+  // message includes a suggested alternative option spelling if available.
+  std::string Nearest;
+  for (const opt::Arg *A : Args.filtered(Unknown)) {
+    std::string Spelling = A->getAsString(Args);
+    if (findNearest(Spelling, Nearest) > 1)
+      ErrorFn("unknown argument '" + A->getAsString(Args) + "'");
+    else
+      ErrorFn("unknown argument '" + A->getAsString(Args) +
+              "', did you mean '" + Nearest + "'?");
+  }
+  return Args;
+}
+
 static std::string getOptionHelpName(const OptTable &Opts, OptSpecifier Id) {
   const Option O = Opts.getOption(Id);
   std::string Name = O.getPrefixedName();

llvm/lib/Support/CommandLine.cpp

@@ -1251,6 +1251,22 @@ bool cl::ExpandResponseFiles(StringSaver &Saver, TokenizerCallback Tokenizer,
   return AllExpanded;
 }
 
+bool cl::expandResponseFiles(int Argc, const char *const *Argv,
+                             const char *EnvVar, StringSaver &Saver,
+                             SmallVectorImpl<const char *> &NewArgv) {
+  auto Tokenize = Triple(sys::getProcessTriple()).isOSWindows()
+                      ? cl::TokenizeWindowsCommandLine
+                      : cl::TokenizeGNUCommandLine;
+  // The environment variable specifies initial options.
+  if (EnvVar)
+    if (llvm::Optional<std::string> EnvValue = sys::Process::GetEnv(EnvVar))
+      Tokenize(*EnvValue, Saver, NewArgv, /*MarkEOLs=*/false);
+
+  // Command line options can override the environment variable.
+  NewArgv.append(Argv + 1, Argv + Argc);
+  return ExpandResponseFiles(Saver, Tokenize, NewArgv);
+}
+
 bool cl::readConfigFile(StringRef CfgFile, StringSaver &Saver,
                         SmallVectorImpl<const char *> &Argv) {
   SmallString<128> AbsPath;

llvm/test/DebugInfo/debuglineinfo-path.ll

@@ -8,9 +8,9 @@
 ; RUN: llvm-nm --radix=o %t | grep posix_absolute_func > %t.posix_absolute_func
 ; RUN: llvm-nm --radix=o %t | grep posix_relative_func > %t.posix_relative_func
 ; RUN: llvm-nm --radix=o %t | grep win_func > %t.win_func
-; RUN: llvm-symbolizer --functions=linkage --inlining --demangle=false --obj %t < %t.posix_absolute_func | FileCheck %s --check-prefix=POSIX_A
-; RUN: llvm-symbolizer --functions=linkage --inlining --demangle=false --obj %t < %t.posix_relative_func | FileCheck %s --check-prefix=POSIX_R
-; RUN: llvm-symbolizer --functions=linkage --inlining --demangle=false --obj %t < %t.win_func | FileCheck %s --check-prefix=WIN
+; RUN: llvm-symbolizer --functions=linkage --inlining --no-demangle --obj %t < %t.posix_absolute_func | FileCheck %s --check-prefix=POSIX_A
+; RUN: llvm-symbolizer --functions=linkage --inlining --no-demangle --obj %t < %t.posix_relative_func | FileCheck %s --check-prefix=POSIX_R
+; RUN: llvm-symbolizer --functions=linkage --inlining --no-demangle --obj %t < %t.win_func | FileCheck %s --check-prefix=WIN
 
 ;POSIX_A: posix_absolute_func
 ;POSIX_A: /absolute/posix/path{{[\/]}}posix.c

llvm/test/tools/llvm-symbolizer/basic.s

@@ -17,6 +17,7 @@ foo:
 
 # Check --obj aliases --exe, -e
 # RUN: llvm-symbolizer 0xa 0xb --exe=%t.o | FileCheck %s
+# RUN: llvm-symbolizer 0xa 0xb --exe %t.o | FileCheck %s
 # RUN: llvm-symbolizer 0xa 0xb -e %t.o | FileCheck %s
 # RUN: llvm-symbolizer 0xa 0xb -e=%t.o | FileCheck %s
 # RUN: llvm-symbolizer 0xa 0xb -e%t.o | FileCheck %s

llvm/test/tools/llvm-symbolizer/help.test

@@ -4,9 +4,9 @@ RUN: llvm-addr2line -h | FileCheck %s --check-prefix=ADDR2LINE
 RUN: llvm-addr2line --help | FileCheck %s --check-prefix=ADDR2LINE
 
 SYMBOLIZER: OVERVIEW: llvm-symbolizer
-SYMBOLIZER: USAGE: llvm-symbolizer{{(.exe)?}} [options] <input addresses>...
+SYMBOLIZER: USAGE: llvm-symbolizer{{(.exe)?}} [options] addresses...
 SYMBOLIZER: @FILE
 
 ADDR2LINE: OVERVIEW: llvm-addr2line
-ADDR2LINE: USAGE: llvm-addr2line{{(.exe)?}} [options] <input addresses>...
+ADDR2LINE: USAGE: llvm-addr2line{{(.exe)?}} [options] addresses...
 ADDR2LINE: @FILE

llvm/test/tools/llvm-symbolizer/output-style-inlined.test

@@ -1,16 +1,16 @@
-This test checks that when inlined frames are not shown (-i=0) and the output
+This test checks that when inlined frames are not shown (--no-inlines) and the output
 style is set to GNU (--output-style=GNU) the name of an inlined function is not
 replaced with the name of the top caller function. At the same time, the current
 behavior of llvm-symbolizer is preserved with --output-style=LLVM or when
 the option is not specified.
 
-RUN: llvm-symbolizer -i=0 -e %p/Inputs/addr.exe 0x40054d \
+RUN: llvm-symbolizer --no-inlines -e %p/Inputs/addr.exe 0x40054d \
 RUN:   | FileCheck %s --check-prefix=LLVM --implicit-check-not=inctwo
 
-RUN: llvm-symbolizer --output-style=LLVM -i=0 -e %p/Inputs/addr.exe 0x40054d \
+RUN: llvm-symbolizer --output-style=LLVM --no-inlines -e %p/Inputs/addr.exe 0x40054d \
 RUN:   | FileCheck %s --check-prefix=LLVM --implicit-check-not=inctwo
 
-RUN: llvm-symbolizer --output-style=GNU -i=0 -e %p/Inputs/addr.exe 0x40054d \
+RUN: llvm-symbolizer --output-style=GNU --no-inlines -e %p/Inputs/addr.exe 0x40054d \
 RUN:   | FileCheck %s --check-prefix=GNU --implicit-check-not=main
 
 RUN: llvm-addr2line -f -e %p/Inputs/addr.exe 0x40054d \

llvm/test/tools/llvm-symbolizer/split-dwarf.test

@@ -4,22 +4,22 @@ RUN: mkdir -p %t
 RUN: cp %p/Inputs/split-dwarf-test.dwo %t
 
 RUN: cd %t
-RUN: llvm-symbolizer --functions=linkage --inlining --demangle=false \
+RUN: llvm-symbolizer --functions=linkage --inlining --no-demangle \
 RUN:    --obj=%p/Inputs/split-dwarf-test 0x400504 0x4004f4 | FileCheck --check-prefixes=SPLIT,DWO %s
 
 Ensure we get the same results in the absence of gmlt-like data in the executable but the presence of a .dwo file
 
 RUN: echo "%p/Inputs/split-dwarf-test-nogmlt 0x400504" >> %t.input
 RUN: echo "%p/Inputs/split-dwarf-test-nogmlt 0x4004f4" >> %t.input
-RUN: llvm-symbolizer --functions=linkage --inlining --demangle=false \
+RUN: llvm-symbolizer --functions=linkage --inlining --no-demangle \
 RUN:    --default-arch=i386 --obj=%p/Inputs/split-dwarf-test-nogmlt 0x400504 0x4004f4 | FileCheck --check-prefixes=SPLIT,DWO %s
 
 Ensure we get gmlt like results in the absence of a .dwo file but the presence of gmlt-like data in the executable
 
 RUN: rm %t/split-dwarf-test.dwo
 RUN: echo "%p/Inputs/split-dwarf-test 0x400504" >> %t.input
 RUN: echo "%p/Inputs/split-dwarf-test 0x4004f4" >> %t.input
-RUN: llvm-symbolizer --functions=linkage --inlining --demangle=false \
+RUN: llvm-symbolizer --functions=linkage --inlining --no-demangle \
 RUN:    --default-arch=i386 --obj=%p/Inputs/split-dwarf-test 0x400504 0x4004f4 | FileCheck --check-prefixes=SPLIT,NODWO %s
 
 DWO: _Z2f2v

llvm/test/tools/llvm-symbolizer/unknown-argument.test

@@ -0,0 +1,12 @@
+# RUN: not llvm-symbolizer -x --flag 2>&1 | FileCheck %s
+
+# CHECK:      error: unknown argument '-x'{{$}}
+# CHECK-NEXT: error: unknown argument '--flag'{{$}}
+
+# RUN: not llvm-symbolizer --inline 2>&1 | FileCheck %s --check-prefix=SUGGEST
+
+# SUGGEST: error: unknown argument '--inline', did you mean '--inlines'?
+
+# RUN: not llvm-symbolizer -e 2>&1 | FileCheck %s --check-prefix=MISSING
+
+# MISSING: error: -e: missing argument

llvm/test/tools/llvm-symbolizer/untag-addresses.test

@@ -2,7 +2,7 @@
 
 # RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t.o
 # RUN: echo DATA %t.o 0 | llvm-symbolizer | FileCheck --check-prefix=UNTAG %s
-# RUN: echo DATA %t.o 0 | llvm-symbolizer -untag-addresses=0 | FileCheck --check-prefix=NOUNTAG %s
+# RUN: echo DATA %t.o 0 | llvm-symbolizer --no-untag-addresses | FileCheck --check-prefix=NOUNTAG %s
 # RUN: echo DATA %t.o 0 | llvm-addr2line | FileCheck --check-prefix=NOUNTAG %s
 
 # UNTAG: foo

llvm/tools/llvm-symbolizer/CMakeLists.txt

@@ -3,17 +3,24 @@
 # This means that we need LLVM libraries to be compiled for these
 # targets as well. Currently, there is no support for such a build strategy.
 
+set(LLVM_TARGET_DEFINITIONS Opts.td)
+tablegen(LLVM Opts.inc -gen-opt-parser-defs)
+add_public_tablegen_target(SymbolizerOptsTableGen)
+
 set(LLVM_LINK_COMPONENTS
   DebugInfoDWARF
   DebugInfoPDB
   Demangle
   Object
+  Option
   Support
   Symbolize
   )
 
 add_llvm_tool(llvm-symbolizer
   llvm-symbolizer.cpp
+  DEPENDS
+  SymbolizerOptsTableGen
   )
 
 add_llvm_tool_symlink(llvm-addr2line llvm-symbolizer)

llvm/tools/llvm-symbolizer/Opts.td

@@ -0,0 +1,60 @@
+include "llvm/Option/OptParser.td"
+
+multiclass B<string name, string help1, string help2> {
+  def NAME: Flag<["--", "-"], name>, HelpText<help1>;
+  def no_ # NAME: Flag<["--", "-"], "no-" # name>, HelpText<help2>;
+}
+
+multiclass Eq<string name, string help> {
+  def NAME #_EQ : Joined<["--", "-"], name #"=">,
+                  HelpText<help>;
+  def : Separate<["--", "-"], name>, Alias<!cast<Joined>(NAME #_EQ)>;
+}
+
+class F<string name, string help>: Flag<["--", "-"], name>, HelpText<help>;
+
+def addresses : F<"addresses", "Show address before line information">;
+defm adjust_vma
+    : Eq<"adjust-vma", "Add specified offset to object file addresses">,
+      MetaVarName<"<offset>">;
+def basenames : Flag<["--"], "basenames">, HelpText<"Strip directory names from paths">;
+defm debug_file_directory : Eq<"debug-file-directory", "Path to directory where to look for debug files">, MetaVarName<"<dir>">;
+defm default_arch : Eq<"default-arch", "Default architecture (for multi-arch objects)">;
+defm demangle : B<"demangle", "Demangle function names", "Don't demangle function names">;
+def functions : F<"functions", "Print function name for a given address">;
+def functions_EQ : Joined<["--"], "functions=">, HelpText<"Print function name for a given address">, Values<"none,short,linkage">;
+def help : F<"help", "Display this help">;
+defm dwp : Eq<"dwp", "Path to DWP file to be use for any split CUs">, MetaVarName<"<file>">;
+defm dsym_hint : Eq<"dsym-hint", "Path to .dSYM bundles to search for debug info for the object files">, MetaVarName<"<dir>">;
+defm fallback_debug_path : Eq<"fallback-debug-path", "Fallback path for debug binaries">, MetaVarName<"<dir>">;
+defm inlines : B<"inlines", "Print all inlined frames for a given address",
+                 "Do not print inlined frames">;
+defm obj
+    : Eq<"obj", "Path to object file to be symbolized (if not provided, "
+                "object file should be specified for each input line)">, MetaVarName<"<file>">;
+defm output_style
+    : Eq<"output-style", "Specify print style. Supported styles: LLVM, GNU">,
+      MetaVarName<"style">,
+      Values<"LLVM,GNU">;
+def pretty_print : F<"pretty-print", "Make the output more human friendly">;
+defm print_source_context_lines : Eq<"print-source-context-lines", "Print N lines of source file context">;
+def relative_address : F<"relative-address", "Interpret addresses as addresses relative to the image base">;
+def relativenames : F<"relativenames", "Strip the compilation directory from paths">;
+defm untag_addresses : B<"untag-addresses", "", "Remove memory tags from addresses before symbolization">;
+def use_native_pdb_reader : F<"use-native-pdb-reader", "Use native PDB functionality">;
+def verbose : F<"verbose", "Print verbose line info">;
+
+def : Flag<["-"], "a">, Alias<addresses>, HelpText<"Alias for --addresses">;
+def : F<"print-address", "Alias for --addresses">, Alias<addresses>;
+def : Flag<["-"], "C">, Alias<demangle>, HelpText<"Alias for --demangle">;
+def : Joined<["--"], "exe=">, Alias<obj_EQ>, HelpText<"Alias for --obj">, MetaVarName<"<file>">;
+def : Separate<["--"], "exe">, Alias<obj_EQ>, HelpText<"Alias for --obj">, MetaVarName<"<file>">;
+def : JoinedOrSeparate<["-"], "e">, Alias<obj_EQ>, HelpText<"Alias for --obj">, MetaVarName<"<file>">;
+def : Joined<["-"], "e=">, Alias<obj_EQ>, HelpText<"Alias for --obj">, MetaVarName<"<file>">;
+def : Flag<["-"], "f">, Alias<functions>, HelpText<"Alias for --functions">;
+def : Joined<["-"], "f=">, Alias<functions_EQ>, HelpText<"Alias for --functions=">;
+def : Flag<["-"], "h">, Alias<help>;
+def : Flag<["-"], "i">, Alias<inlines>, HelpText<"Alias for --inlines">;
+def : F<"inlining", "Alias for --inlines">, Alias<inlines>;
+def : Flag<["-"], "p">, Alias<pretty_print>, HelpText<"Alias for --pretty-print">;
+def : Flag<["-"], "s">, Alias<basenames>, HelpText<"Alias for --basenames">;