node/tools/wafadmin/Utils.py

#!/usr/bin/env python
# encoding: utf-8
# Thomas Nagy, 2005 (ita)

"""
Utilities, the stable ones are the following:

* h_file: compute a unique value for a file (hash), it uses
  the module fnv if it is installed (see waf/utils/fnv & http://code.google.com/p/waf/wiki/FAQ)
  else, md5 (see the python docs)

  For large projects (projects with more than 15000 files) or slow hard disks and filesystems (HFS)
  it is possible to use a hashing based on the path and the size (may give broken cache results)
  The method h_file MUST raise an OSError if the file is a folder

	import stat
	def h_file(filename):
		st = os.stat(filename)
		if stat.S_ISDIR(st[stat.ST_MODE]): raise IOError('not a file')
		m = Utils.md5()
		m.update(str(st.st_mtime))
		m.update(str(st.st_size))
		m.update(filename)
		return m.digest()

	To replace the function in your project, use something like this:
	import Utils
	Utils.h_file = h_file

* h_list
* h_fun
* get_term_cols
* ordered_dict

"""

import os, sys, imp, string, errno, traceback, inspect, re, shutil, datetime, gc

# In python 3.0 we can get rid of all this
try: from UserDict import UserDict
except ImportError: from collections import UserDict
if sys.hexversion >= 0x2060000 or os.name == 'java':
	import subprocess as pproc
else:
	import pproc
import Logs
from Constants import *

try:
	from collections import deque
except ImportError:
	class deque(list):
		def popleft(self):
			return self.pop(0)

is_win32 = sys.platform == 'win32'

try:
	# defaultdict in python 2.5
	from collections import defaultdict as DefaultDict
except ImportError:
	class DefaultDict(dict):
		def __init__(self, default_factory):
			super(DefaultDict, self).__init__()
			self.default_factory = default_factory
		def __getitem__(self, key):
			try:
				return super(DefaultDict, self).__getitem__(key)
			except KeyError:
				value = self.default_factory()
				self[key] = value
				return value

class WafError(Exception):
	def __init__(self, *args):
		self.args = args
		try:
			self.stack = traceback.extract_stack()
		except:
			pass
		Exception.__init__(self, *args)
	def __str__(self):
		return str(len(self.args) == 1 and self.args[0] or self.args)

class WscriptError(WafError):
	def __init__(self, message, wscript_file=None):
		if wscript_file:
			self.wscript_file = wscript_file
			self.wscript_line = None
		else:
			try:
				(self.wscript_file, self.wscript_line) = self.locate_error()
			except:
				(self.wscript_file, self.wscript_line) = (None, None)

		msg_file_line = ''
		if self.wscript_file:
			msg_file_line = "%s:" % self.wscript_file
			if self.wscript_line:
				msg_file_line += "%s:" % self.wscript_line
		err_message = "%s error: %s" % (msg_file_line, message)
		WafError.__init__(self, err_message)

	def locate_error(self):
		stack = traceback.extract_stack()
		stack.reverse()
		for frame in stack:
			file_name = os.path.basename(frame[0])
			is_wscript = (file_name == WSCRIPT_FILE or file_name == WSCRIPT_BUILD_FILE)
			if is_wscript:
				return (frame[0], frame[1])
		return (None, None)

indicator = is_win32 and '\x1b[A\x1b[K%s%s%s\r' or '\x1b[K%s%s%s\r'

try:
	from fnv import new as md5
	import Constants
	Constants.SIG_NIL = 'signofnv'

	def h_file(filename):
		m = md5()
		try:
			m.hfile(filename)
			x = m.digest()
			if x is None: raise OSError("not a file")
			return x
		except SystemError:
			raise OSError("not a file" + filename)

except ImportError:
	try:
		from hashlib import md5
	except ImportError:
		from md5 import md5

	def h_file(filename):
		f = open(filename, 'rb')
		m = md5()
		while (filename):
			filename = f.read(100000)
			m.update(filename)
		f.close()
		return m.digest()

class ordered_dict(UserDict):
	def __init__(self, dict = None):
		self.allkeys = []
		UserDict.__init__(self, dict)

	def __delitem__(self, key):
		self.allkeys.remove(key)
		UserDict.__delitem__(self, key)

	def __setitem__(self, key, item):
		if key not in self.allkeys: self.allkeys.append(key)
		UserDict.__setitem__(self, key, item)

def exec_command(s, **kw):
	if 'log' in kw:
		kw['stdout'] = kw['stderr'] = kw['log']
		del(kw['log'])
	kw['shell'] = isinstance(s, str)

	try:
		proc = pproc.Popen(s, **kw)
		return proc.wait()
	except OSError:
		return -1

if is_win32:
	def exec_command(s, **kw):
		if 'log' in kw:
			kw['stdout'] = kw['stderr'] = kw['log']
			del(kw['log'])
		kw['shell'] = isinstance(s, str)

		if len(s) > 2000:
			startupinfo = pproc.STARTUPINFO()
			startupinfo.dwFlags |= pproc.STARTF_USESHOWWINDOW
			kw['startupinfo'] = startupinfo

		try:
			if 'stdout' not in kw:
				kw['stdout'] = pproc.PIPE
				kw['stderr'] = pproc.PIPE
				proc = pproc.Popen(s,**kw)
				(stdout, stderr) = proc.communicate()
				Logs.info(stdout)
				if stderr:
					Logs.error(stderr)
				return proc.returncode
			else:
				proc = pproc.Popen(s,**kw)
				return proc.wait()
		except OSError:
			return -1

listdir = os.listdir
if is_win32:
	def listdir_win32(s):
		if re.match('^[A-Za-z]:$', s):
			# os.path.isdir fails if s contains only the drive name... (x:)
			s += os.sep
		if not os.path.isdir(s):
			e = OSError()
			e.errno = errno.ENOENT
			raise e
		return os.listdir(s)
	listdir = listdir_win32

def waf_version(mini = 0x010000, maxi = 0x100000):
	"Halts if the waf version is wrong"
	ver = HEXVERSION
	try: min_val = mini + 0
	except TypeError: min_val = int(mini.replace('.', '0'), 16)

	if min_val > ver:
		Logs.error("waf version should be at least %s (%s found)" % (mini, ver))
		sys.exit(0)

	try: max_val = maxi + 0
	except TypeError: max_val = int(maxi.replace('.', '0'), 16)

	if max_val < ver:
		Logs.error("waf version should be at most %s (%s found)" % (maxi, ver))
		sys.exit(0)

def python_24_guard():
	if sys.hexversion < 0x20400f0 or sys.hexversion >= 0x3000000:
		raise ImportError("Waf requires Python >= 2.3 but the raw source requires Python 2.4, 2.5 or 2.6")

def ex_stack():
	exc_type, exc_value, tb = sys.exc_info()
	if Logs.verbose > 1:
		exc_lines = traceback.format_exception(exc_type, exc_value, tb)
		return ''.join(exc_lines)
	return str(exc_value)

def to_list(sth):
	if isinstance(sth, str):
		return sth.split()
	else:
		return sth

g_loaded_modules = {}
"index modules by absolute path"

g_module=None
"the main module is special"

def load_module(file_path, name=WSCRIPT_FILE):
	"this function requires an absolute path"
	try:
		return g_loaded_modules[file_path]
	except KeyError:
		pass

	module = imp.new_module(name)

	try:
		code = readf(file_path, m='rU')
	except (IOError, OSError):
		raise WscriptError('Could not read the file %r' % file_path)

	module.waf_hash_val = code

	sys.path.insert(0, os.path.dirname(file_path))
	try:
		exec(compile(code, file_path, 'exec'), module.__dict__)
	except Exception:
		raise WscriptError(traceback.format_exc(), file_path)
	sys.path.pop(0)

	g_loaded_modules[file_path] = module

	return module

def set_main_module(file_path):
	"Load custom options, if defined"
	global g_module
	g_module = load_module(file_path, 'wscript_main')
	g_module.root_path = file_path

	# note: to register the module globally, use the following:
	# sys.modules['wscript_main'] = g_module

def to_hashtable(s):
	"used for importing env files"
	tbl = {}
	lst = s.split('\n')
	for line in lst:
		if not line: continue
		mems = line.split('=')
		tbl[mems[0]] = mems[1]
	return tbl

def get_term_cols():
	"console width"
	return 80
try:
	import struct, fcntl, termios
except ImportError:
	pass
else:
	if Logs.got_tty:
		def myfun():
			dummy_lines, cols = struct.unpack("HHHH", \
			fcntl.ioctl(sys.stderr.fileno(),termios.TIOCGWINSZ , \
			struct.pack("HHHH", 0, 0, 0, 0)))[:2]
			return cols
		# we actually try the function once to see if it is suitable
		try:
			myfun()
		except IOError:
			pass
		else:
			get_term_cols = myfun

rot_idx = 0
rot_chr = ['\\', '|', '/', '-']
"the rotation character in the progress bar"


def split_path(path):
	return path.split('/')

def split_path_cygwin(path):
	if path.startswith('//'):
		ret = path.split('/')[2:]
		ret[0] = '/' + ret[0]
		return ret
	return path.split('/')

re_sp = re.compile('[/\\\\]')
def split_path_win32(path):
	if path.startswith('\\\\'):
		ret = re.split(re_sp, path)[2:]
		ret[0] = '\\' + ret[0]
		return ret
	return re.split(re_sp, path)

if sys.platform == 'cygwin':
	split_path = split_path_cygwin
elif is_win32:
	split_path = split_path_win32

def copy_attrs(orig, dest, names, only_if_set=False):
	for a in to_list(names):
		u = getattr(orig, a, ())
		if u or not only_if_set:
			setattr(dest, a, u)

def def_attrs(cls, **kw):
	'''
	set attributes for class.
	@param cls [any class]: the class to update the given attributes in.
	@param kw [dictionary]: dictionary of attributes names and values.

	if the given class hasn't one (or more) of these attributes, add the attribute with its value to the class.
	'''
	for k, v in kw.iteritems():
		if not hasattr(cls, k):
			setattr(cls, k, v)

def quote_define_name(path):
	fu = re.compile("[^a-zA-Z0-9]").sub("_", path)
	fu = fu.upper()
	return fu

def quote_whitespace(path):
	return (path.strip().find(' ') > 0 and '"%s"' % path or path).replace('""', '"')

def trimquotes(s):
	if not s: return ''
	s = s.rstrip()
	if s[0] == "'" and s[-1] == "'": return s[1:-1]
	return s

def h_list(lst):
	m = md5()
	m.update(str(lst))
	return m.digest()

def h_fun(fun):
	try:
		return fun.code
	except AttributeError:
		try:
			h = inspect.getsource(fun)
		except IOError:
			h = "nocode"
		try:
			fun.code = h
		except AttributeError:
			pass
		return h

def pprint(col, str, label='', sep=os.linesep):
	"print messages in color"
	sys.stderr.write("%s%s%s %s%s" % (Logs.colors(col), str, Logs.colors.NORMAL, label, sep))

def check_dir(dir):
	"""If a folder doesn't exists, create it."""
	try:
		os.stat(dir)
	except OSError:
		try:
			os.makedirs(dir)
		except OSError, e:
			raise WafError("Cannot create folder '%s' (original error: %s)" % (dir, e))

def cmd_output(cmd, **kw):

	silent = False
	if 'silent' in kw:
		silent = kw['silent']
		del(kw['silent'])

	if 'e' in kw:
		tmp = kw['e']
		del(kw['e'])
		kw['env'] = tmp

	kw['shell'] = isinstance(cmd, str)
	kw['stdout'] = pproc.PIPE
	if silent:
		kw['stderr'] = pproc.PIPE

	try:
		p = pproc.Popen(cmd, **kw)
		output = p.communicate()[0]
	except OSError, e:
		raise ValueError(str(e))

	if p.returncode:
		if not silent:
			msg = "command execution failed: %s -> %r" % (cmd, str(output))
			raise ValueError(msg)
		output = ''
	return output

reg_subst = re.compile(r"(\\\\)|(\$\$)|\$\{([^}]+)\}")
def subst_vars(expr, params):
	"substitute ${PREFIX}/bin in /usr/local/bin"
	def repl_var(m):
		if m.group(1):
			return '\\'
		if m.group(2):
			return '$'
		try:
			# environments may contain lists
			return params.get_flat(m.group(3))
		except AttributeError:
			return params[m.group(3)]
	return reg_subst.sub(repl_var, expr)

def unversioned_sys_platform_to_binary_format(unversioned_sys_platform):
	"infers the binary format from the unversioned_sys_platform name."

	if unversioned_sys_platform in ('linux', 'freebsd', 'netbsd', 'openbsd', 'sunos'):
		return 'elf'
	elif unversioned_sys_platform == 'darwin':
		return 'mac-o'
	elif unversioned_sys_platform in ('win32', 'cygwin', 'uwin', 'msys'):
		return 'pe'
	# TODO we assume all other operating systems are elf, which is not true.
	# we may set this to 'unknown' and have ccroot and other tools handle the case "gracefully" (whatever that means).
	return 'elf'

def unversioned_sys_platform():
	"""returns an unversioned name from sys.platform.
	sys.plaform is not very well defined and depends directly on the python source tree.
	The version appended to the names is unreliable as it's taken from the build environment at the time python was built,
	i.e., it's possible to get freebsd7 on a freebsd8 system.
	So we remove the version from the name, except for special cases where the os has a stupid name like os2 or win32.
	Some possible values of sys.platform are, amongst others:
		aix3 aix4 atheos beos5 darwin freebsd2 freebsd3 freebsd4 freebsd5 freebsd6 freebsd7
		generic irix5 irix6 linux2 mac netbsd1 next3 os2emx riscos sunos5 unixware7
	Investigating the python source tree may reveal more values.
	"""
	s = sys.platform
	if s == 'java':
		# The real OS is hidden under the JVM.
		from java.lang import System
		s = System.getProperty('os.name')
		# see http://lopica.sourceforge.net/os.html for a list of possible values
		if s == 'Mac OS X':
			return 'darwin'
		elif s.startswith('Windows '):
			return 'win32'
		elif s == 'OS/2':
			return 'os2'
		elif s == 'HP-UX':
			return 'hpux'
		elif s in ('SunOS', 'Solaris'):
			return 'sunos'
		else: s = s.lower()
	if s == 'win32' or s.endswith('os2') and s != 'sunos2': return s
	return re.split('\d+$', s)[0]

#@deprecated('use unversioned_sys_platform instead')
def detect_platform():
	"""this function has been in the Utils module for some time.
	It's hard to guess what people have used it for.
	It seems its goal is to return an unversionned sys.platform, but it's not handling all platforms.
	For example, the version is not removed on freebsd and netbsd, amongst others.
	"""
	s = sys.platform

	# known POSIX
	for x in 'cygwin linux irix sunos hpux aix darwin'.split():
		# sys.platform may be linux2
		if s.find(x) >= 0:
			return x

	# unknown POSIX
	if os.name in 'posix java os2'.split():
		return os.name

	return s

def load_tool(tool, tooldir=None):
	'''
	load_tool: import a Python module, optionally using several directories.
	@param tool [string]: name of tool to import.
	@param tooldir [list]: directories to look for the tool.
	@return: the loaded module.

	Warning: this function is not thread-safe: plays with sys.path,
					 so must run in sequence.
	'''
	if tooldir:
		assert isinstance(tooldir, list)
		sys.path = tooldir + sys.path
	try:
		try:
			return __import__(tool)
		except ImportError, e:
			Logs.error('Could not load the tool %r in %r:\n%s' % (tool, sys.path, e))
			raise
	finally:
		if tooldir:
			sys.path = sys.path[len(tooldir):]

def readf(fname, m='r'):
	"get the contents of a file, it is not used anywhere for the moment"
	f = open(fname, m)
	try:
		txt = f.read()
	finally:
		f.close()
	return txt

def nada(*k, **kw):
	"""A function that does nothing"""
	pass

def diff_path(top, subdir):
	"""difference between two absolute paths"""
	top = os.path.normpath(top).replace('\\', '/').split('/')
	subdir = os.path.normpath(subdir).replace('\\', '/').split('/')
	if len(top) == len(subdir): return ''
	diff = subdir[len(top) - len(subdir):]
	return os.path.join(*diff)

class Context(object):
	"""A base class for commands to be executed from Waf scripts"""

	def set_curdir(self, dir):
		self.curdir_ = dir

	def get_curdir(self):
		try:
			return self.curdir_
		except AttributeError:
			self.curdir_ = os.getcwd()
			return self.get_curdir()

	curdir = property(get_curdir, set_curdir)

	def recurse(self, dirs, name=''):
		"""The function for calling scripts from folders, it tries to call wscript + function_name
		and if that file does not exist, it will call the method 'function_name' from a file named wscript
		the dirs can be a list of folders or a string containing space-separated folder paths
		"""
		if not name:
			name = inspect.stack()[1][3]

		if isinstance(dirs, str):
			dirs = to_list(dirs)

		for x in dirs:
			if os.path.isabs(x):
				nexdir = x
			else:
				nexdir = os.path.join(self.curdir, x)

			base = os.path.join(nexdir, WSCRIPT_FILE)
			file_path = base + '_' + name

			try:
				txt = readf(file_path, m='rU')
			except (OSError, IOError):
				try:
					module = load_module(base)
				except OSError:
					raise WscriptError('No such script %s' % base)

				try:
					f = module.__dict__[name]
				except KeyError:
					raise WscriptError('No function %s defined in %s' % (name, base))

				if getattr(self.__class__, 'pre_recurse', None):
					self.pre_recurse(f, base, nexdir)
				old = self.curdir
				self.curdir = nexdir
				try:
					f(self)
				finally:
					self.curdir = old
				if getattr(self.__class__, 'post_recurse', None):
					self.post_recurse(module, base, nexdir)
			else:
				dc = {'ctx': self}
				if getattr(self.__class__, 'pre_recurse', None):
					dc = self.pre_recurse(txt, file_path, nexdir)
				old = self.curdir
				self.curdir = nexdir
				try:
					try:
						exec(compile(txt, file_path, 'exec'), dc)
					except Exception:
						raise WscriptError(traceback.format_exc(), base)
				finally:
					self.curdir = old
				if getattr(self.__class__, 'post_recurse', None):
					self.post_recurse(txt, file_path, nexdir)

if is_win32:
	old = shutil.copy2
	def copy2(src, dst):
		old(src, dst)
		shutil.copystat(src, src)
	setattr(shutil, 'copy2', copy2)

def zip_folder(dir, zip_file_name, prefix):
	"""
	prefix represents the app to add in the archive
	"""
	import zipfile
	zip = zipfile.ZipFile(zip_file_name, 'w', compression=zipfile.ZIP_DEFLATED)
	base = os.path.abspath(dir)

	if prefix:
		if prefix[-1] != os.sep:
			prefix += os.sep

	n = len(base)
	for root, dirs, files in os.walk(base):
		for f in files:
			archive_name = prefix + root[n:] + os.sep + f
			zip.write(root + os.sep + f, archive_name, zipfile.ZIP_DEFLATED)
	zip.close()

def get_elapsed_time(start):
	"Format a time delta (datetime.timedelta) using the format DdHhMmS.MSs"
	delta = datetime.datetime.now() - start
	# cast to int necessary for python 3.0
	days = int(delta.days)
	hours = int(delta.seconds / 3600)
	minutes = int((delta.seconds - hours * 3600) / 60)
	seconds = delta.seconds - hours * 3600 - minutes * 60 \
		+ float(delta.microseconds) / 1000 / 1000
	result = ''
	if days:
		result += '%dd' % days
	if days or hours:
		result += '%dh' % hours
	if days or hours or minutes:
		result += '%dm' % minutes
	return '%s%.3fs' % (result, seconds)

if os.name == 'java':
	# For Jython (they should really fix the inconsistency)
	try:
		gc.disable()
		gc.enable()
	except NotImplementedError:
		gc.disable = gc.enable