emscripten/emcc

#!/usr/bin/env python

'''
emcc - compiler helper script
=============================

emcc is a drop-in replacement for a compiler like gcc or clang.

Tell your build system to use this instead of the compiler, linker, ar and
ranlib. All the normal build commands will be sent to this script, which
will proxy them to the appropriate LLVM build commands, in order to
generate proper code for Emscripten to later process.

For example, compilation will be translated into calls to clang
with -emit-llvm, and linking will be translated into calls to llvm-link,
and so forth.

emcc is only meant to *COMPILE* source code into LLVM bitcode. It does
not do optimizations (in fact, it will disable -Ox flags and warn you
about that). The reason is that doing such optimizations early can lead
to bitcode that Emscripten cannot process properly, or can process but
not fully optimize. You can (and probably should) still run LLVM
optimizations though, by telling emscripten.py to do so (or run LLVM
opt yourself, but be careful with the parameters you pass).

Example uses:

 * For configure, instead of ./configure, cmake, etc., run emconfiguren.py
   with that command as an argument, for example

    emconfiguren.py ./configure [options]
  
   emconfiguren.py is a tiny script that just sets some environment vars
   as a convenience. The command just shown is equivalent to

    EMMAKEN_JUST_CONFIGURE=1 RANLIB=PATH/emcc AR=PATH/emcc CXX=PATH/em++ CC=PATH/emcc ./configure [options]

   where PATH is the path to this file.

   EMMAKEN_JUST_CONFIGURE tells emcc that it is being run in ./configure,
   so it should relay everything to gcc/g++. You should not define that when
   running make, of course.

 * With CMake, the same command will work (with cmake instead of ./configure). You may also be
   able to do the following in your CMakeLists.txt:

    SET(CMAKE_C_COMPILER "PATH/emcc")
    SET(CMAKE_CXX_COMPILER "PATH/em++")
    SET(CMAKE_LINKER "PATH/emcc")
    SET(CMAKE_CXX_LINKER "PATH/emcc")
    SET(CMAKE_C_LINK_EXECUTABLE "PATH/emcc")
    SET(CMAKE_CXX_LINK_EXECUTABLE "PATH/emcc")
    SET(CMAKE_AR "PATH/emcc")
    SET(CMAKE_RANLIB "PATH/emcc")

 * For SCons the shared.py can be imported like so:
    __file__ = str(Dir('#/project_path_to_emscripten/dummy/dummy'))
    __rootpath__ = os.path.abspath(os.path.dirname(os.path.dirname(__file__)))
    def path_from_root(*pathelems):
      return os.path.join(__rootpath__, *pathelems)
    exec(open(path_from_root('tools', 'shared.py'), 'r').read())
    
   For using the Emscripten compilers/linkers/etc. you can do:
    env = Environment()
    ...
    env.Append(CCFLAGS = COMPILER_OPTS)
    env.Replace(LINK = LLVM_LD)
    env.Replace(LD   = LLVM_LD)
   TODO: Document all relevant setup changes

After setting that up, run your build system normally. It should generate
LLVM instead of the normal output, and end up with .ll files that you can
give to Emscripten. Note that this tool doesn't run Emscripten itself. Note
also that you may need to do some manual fiddling later, for example to
link files that weren't linked, and them llvm-dis them.

Note the appearance of em++ instead of emcc
for the C++ compiler. This is needed for cases where we get
a C++ file with a C extension, in which case CMake can be told
to run g++ on it despite the .c extension, see

  https://github.com/kripken/emscripten/issues/6

(If a similar situation occurs with ./configure, you can do the same there too.)

emcc can be influenced by a few environment variables:

  EMMAKEN_NO_SDK - Will tell emcc *not* to use the emscripten headers. Instead
                   your system headers will be used.

  EMMAKEN_COMPILER - The compiler to be used, if you don't want the default clang.

'''

import os, sys, shutil
from subprocess import Popen, PIPE, STDOUT
from tools import shared

DEBUG = 1

################### XXX
print >> sys.stderr, '\n***This is a WORK IN PROGRESS***'
print >> sys.stderr, '***[%s]***\n' % str(sys.argv)
################### XXX

if DEBUG: print >> sys.stderr, 'emcc: ', ' '.join(sys.argv)

# Handle some global flags

if len(sys.argv) == 1:
  print 'emcc: no input files'
  exit(0)

if sys.argv[1] == '--version':
  print '''emcc (Emscripten GCC-like replacement) 2.0
Copyright (C) 2011 the Emscripten authors.
This is free and open source software under the MIT license.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
  '''
  exit(0)
elif sys.argv[1] == '--help':
  this = os.path.basename('em++' if os.environ.get('EMMAKEN_CXX') else 'emcc')

  print '''%s [options] file...

Most normal gcc/g++ options will work, for example:
  --help                   Display this information
  --version                Display compiler version information

Options that are modified or new in %s include:
  -O0                      [..] default
  -OX                      TODO
  -s OPTION=VALUE          JavaScript code generation option
                           passed into the emscripten compiler
  --typed-arrays <mode>    0: no typed arrays
                           1: parallel typed arrays
                           2: shared typed arrays (default)
  --llvm-opts <mode>       0: none (default)
                           1: safe/portable
                           2: unsafe/unportable

The target file, if specified (-o <target>), defines what will
be generated:

  <name>.js                JavaScript (default)
  <name>.o                 LLVM bitcode
  <name>.bc                LLVM bitcode
  <name>.html              HTML with embedded JavaScript

The -c option (which tells gcc not to run the linker) will
also cause LLVM bitcode to be generated, as %s only generates
JavaScript in the final linking stage of building.

''' % (this, this, this)
  exit(0)

# If this is a configure-type thing, just do that
CONFIGURE_CONFIG = os.environ.get('EMMAKEN_JUST_CONFIGURE')
CMAKE_CONFIG = 'CMakeFiles/cmTryCompileExec.dir' in ' '.join(sys.argv)# or 'CMakeCCompilerId' in ' '.join(sys.argv)
if CONFIGURE_CONFIG or CMAKE_CONFIG:
  compiler = 'g++' if 'CXXCompiler' in ' '.join(sys.argv) or os.environ.get('EMMAKEN_CXX') else 'gcc'
  cmd = [compiler] + EMSDK_OPTS + sys.argv[1:]
  if DEBUG: print >> sys.stderr, 'emcc, just configuring: ', cmd
  exit(os.execvp(compiler, cmd))

if os.environ.get('EMMAKEN_COMPILER'):
  CXX = os.environ['EMMAKEN_COMPILER']
else:
  CXX = shared.CLANG

CC = shared.to_cc(CXX)

# If we got here from a redirection through emmakenxx.py, then force a C++ compiler here
if os.environ.get('EMMAKEN_CXX'):
  CC = CXX

CC_ADDITIONAL_ARGS = shared.COMPILER_OPTS # + ['-g']?
ALLOWED_LINK_ARGS = ['-f', '-help', '-o', '-print-after', '-print-after-all', '-print-before',
                     '-print-before-all', '-time-passes', '-v', '-verify-dom-info', '-version' ]
TWO_PART_DISALLOWED_LINK_ARGS = ['-L'] # Ignore thingsl like |-L .|

EMMAKEN_CFLAGS = os.environ.get('EMMAKEN_CFLAGS')
if EMMAKEN_CFLAGS: CC_ADDITIONAL_ARGS += EMMAKEN_CFLAGS.split(' ')

# ----------------  End configs -------------

if len(sys.argv) == 1 or sys.argv[1] in ['x', 't']:
  # noop ar
  if DEBUG: print >> sys.stderr, 'emcc, just ar'
  sys.exit(0)

use_cxx = True
use_linker = True
header = False # pre-compiled headers. We fake that by just copying the file

opts = []
files = []
for i in range(1, len(sys.argv)):
  arg = sys.argv[i]
  if arg.startswith('-'):
    opts.append(arg)
  else:
    files.append(arg)
    if arg.endswith('.c'):
      use_cxx = False
    if arg.endswith(('.c', '.cc', '.cpp', '.dT')):
      use_linker = False
    if arg.endswith('.h') and sys.argv[i-1] != '-include':
      header = True
      use_linker = False

if '--version' in opts:
  use_linker = False

use_compiler = not use_linker and not header

if set(sys.argv[1]).issubset(set('-cruqs')): # ar
  sys.argv = sys.argv[:1] + sys.argv[3:] + ['-o='+sys.argv[2]]
  assert use_linker, 'Linker should be used in this case'

# Check if a target is specified
target = None
for i in range(len(sys.argv)-1):
  if sys.argv[i] == '-o':
    target = sys.argv[i+1]
    sys.argv = sys.argv[:i] + sys.argv[i+2:]
    break

if use_linker:
  call = LLVM_LD
  newargs = ['-disable-opt']
  i = 0
  while i < len(sys.argv)-1:
    i += 1
    arg = sys.argv[i]
    if arg.startswith('-'):
      prefix = arg.split('=')[0]
      if prefix in ALLOWED_LINK_ARGS:
        newargs.append(arg)
      if arg in TWO_PART_DISALLOWED_LINK_ARGS:
        i += 1
    elif arg.endswith('.so'):
      continue # .so's do not exist yet, in many cases
    else:
      # not option, so just append
      newargs.append(arg)
  if target:
    newargs.append('-o=' + target)

  if DEBUG: print >> sys.stderr, "Running:", call, ' '.join(newargs)
  Popen([call] + newargs).communicate()
  exit(0)

elif use_compiler:
  call = CXX if use_cxx else CC
  newargs = sys.argv[1:]
  for i in range(len(newargs)):
    if newargs[i].startswith('-O'):
      if DEBUG: print >> sys.stderr, 'emcc: WARNING: Optimization flags (-Ox) are ignored in emcc. Tell emscripten.py to do that, or run LLVM opt.'
      newargs[i] = ''
  newargs = [ arg for arg in newargs if arg is not '' ] + CC_ADDITIONAL_ARGS

  if target is None:
    # No explicit -o specified, so do the most natural thing, compile to .js
    target = 'a.out.js'

  target_basename = '.'.join(target.split('.')[:-1])

  if '-c' in newargs: # -c means do not link in gcc, and for us, the parallel is to not go all the way to JS, but stop at bitcode
    target = target_basename + '.bc'

  final_suffix = target.split('.')[-1]

  # First, generate LLVM bitcode TODO: handle |emcc a.cpp b.cpp -c| which generate *two* bitcode files
  newargs = newargs + ['-emit-llvm', '-c', '-o', target_basename + '.bc']

  if DEBUG: print >> sys.stderr, "Running:", call, ' '.join(newargs)
  Popen([call] + newargs).communicate()

  # If we were just asked to generate bitcode, stop there
  if final_suffix in ['o', 'bc']:
    if final_suffix == 'o':
      shutil.move(target_basename + '.bc', target_basename + '.o')
    exit(0)

  # Continue on to create JavaScript
  temp_files = shared.TempFiles()
  temp_files.note(target_basename + '.bc')
  try:
    shared.Building.emscripten(target_basename + '.bc', append_ext=False)
    shutil.move(target_basename + '.bc.o.js', target_basename + '.js')

    # If we were asked to also generate HTML, do that
    if final_suffix == 'html':
      shell = open(shared.path_from_root('src', 'shell.html')).read()
      html = open(target_basename + '.html', 'w')
      html.write(shell.replace('{{{ SCRIPT_CODE }}}', open(target_basename + '.js').read()))
      html.close()
      temp_files.note(target_basename + '.js')

  finally:
    temp_files.clean()

  exit(0)

else: # header or such
  if DEBUG: print >> sys.stderr, 'Just copy.'
  shutil.copy(sys.argv[-1], sys.argv[-2])
  exit(0)