69 Commits

Author SHA1 Message Date
coissac 764ee27c0d Switch to the version 1.1.21 2015-07-13 15:31:20 +02:00
coissac 18d87aa1b1 Add a small work around patch for the --skip and --only options in
illuminapairedend
2015-07-11 23:38:47 +02:00
coissac 1c91c27ee3 Tag the version 1.1.20 2015-07-03 14:57:24 +02:00
coissac 22d343b46a Patch two small print bug in the new ecotag 2015-07-03 14:55:44 +02:00
coissac 2af94b9da7 Add docuentation for the new options and an option to manage the ecotag
cache size
2015-07-03 10:39:59 +02:00
coissac aa064dda57 Add two general options for limiting the analysis of a file to only a
sub-part of the data:

   - --skip <N>
   - --only <N>
   
They respectively allow to first skip the N first sequences, and to
analysize only the <N> next sequences.
2015-07-03 09:10:49 +02:00
coissac f38ccae698 Add a --minimum-circle option to ecotag and a cache on the self
alignment scores of 1000000 of pairwise scores
2015-07-02 16:14:22 +02:00
Celine Mercier b99881817a Closes #9 : adds a parser in obiaddtaxids for the UNITE 'general FASTA
release' format
2015-06-12 15:08:58 +02:00
Celine Mercier df093b3fa9 updated the UNITE parser of obiaddtaxids 2015-06-11 17:47:00 +02:00
coissac 9ba67985e9 Switch to the 1.1.18 version 2015-06-02 16:01:16 +02:00
coissac f74aeff002 Merge branch 'master' of git@git.metabarcoding.org:obitools/obitools.git 2015-06-02 15:59:57 +02:00
coissac b9fa718a2c tag the 1.1.17 version 2015-06-02 15:53:41 +02:00
coissac 51a7ecde21 Patch the installer 2015-06-02 15:52:18 +02:00
Celine Mercier 6bf6cc6612 fixing a documentation problem (issue #3) 2015-06-01 14:11:11 +02:00
Celine Mercier 191af9669a modified the obisilva script to deal with version names containing dots,
e.g. "119.1"
2015-06-01 14:07:19 +02:00
Celine Mercier 429fbbb61b switching branch OBITools-1.01 to master 2015-06-01 10:41:26 +02:00
Celine Mercier 18c3a7dc08 Fixes #3 2015-05-22 18:00:13 +02:00
Céline Mercier f40c2f5245 in obitaxonomy, fixed a problem when adding taxa from file with an associated path 2015-02-23 14:02:31 +00:00
Frédéric Boyer f596d3a4fb [ADD] order in location 2014-10-21 10:32:44 +00:00
Frédéric Boyer 4de712702d Bug sur merged_KEY lorsque la merged_KEY existe deja dans les sequences : la somme n'etait pas effectuee correctement 2014-09-30 09:21:32 +00:00
Frédéric Boyer 1dad3bf086 -r option fix 2014-05-15 14:41:16 +00:00
Frédéric Boyer b25091e8b0 Bug à l'affichage au chargement de la taxonomie 2014-04-30 15:04:23 +00:00
Frédéric Boyer a664ea5d77 assertion failed with progress bar when empty file 2014-04-25 08:51:17 +00:00
coissac a9a86d1067 remove memory leak 2014-04-23 14:00:48 +00:00
coissac 5a6c4d231c 2014-04-23 12:35:54 +00:00
Frédéric Boyer 62923e1cb8 [MOD] Erreur de lecture des fichiers (standard, pas les options) 2014-04-22 22:39:23 +00:00
Frédéric Boyer d5ce87f7f8 [MOD] Corrected a bug in initialization of 'word4table' that 'coredump' on linux 2014-03-10 18:19:55 +00:00
coissac 7c55ad6d32 Patch to fit the version 0.20 of cython 2014-02-11 16:25:06 +00:00
coissac eac5bc958d Patch to fit the version 0.20 of cython 2014-02-11 16:22:28 +00:00
Frédéric Boyer 847284267e 2014-01-31 16:51:20 +00:00
Frédéric Boyer 105cb4c8be 2014-01-31 16:45:18 +00:00
Frédéric Boyer acad67c67f 2014-01-31 16:26:44 +00:00
Frédéric Boyer 4daec682ff 2014-01-31 16:18:31 +00:00
Frédéric Boyer b09f7b4ef4 2014-01-31 10:39:48 +00:00
Frédéric Boyer 8b00463558 2014-01-31 10:32:21 +00:00
coissac bd75660e3d 2014-01-31 09:29:36 +00:00
Frédéric Boyer 2d432e1d51 doc 2014-01-30 16:20:36 +00:00
coissac 2f6c3c05db Change the name of the tags forward_primer and reverse_primer used in ecoPCR file conversion to forward_match and reverse_match to conform with names used in ngsfilter 2014-01-30 10:39:01 +00:00
Céline Mercier 4e22703ddb fixed spelling mistake 2013-12-10 13:59:51 +00:00
coissac 0675ea6770 Patch a bug related to the -n option of obitab 2013-12-10 08:11:07 +00:00
Frédéric Boyer f3d52f1007 [DOC] 2013-12-04 14:54:09 +00:00
Frédéric Boyer 9ae4a1b4c4 [MOD] direct_tag replaced by forward_tag 2013-12-04 13:41:31 +00:00
Frédéric Boyer a938c9bdc2 2013-12-04 11:02:02 +00:00
Frédéric Boyer 514cd02210 [MOD] nqual and junction removed 2013-12-04 11:00:27 +00:00
Frédéric Boyer f2f8d8805c 2013-12-04 10:18:00 +00:00
Céline Mercier f355cc7a9f 2013-11-20 15:42:07 +00:00
Frédéric Boyer 608db8b1c0 [MOD] removed a print 2013-11-15 21:53:40 +00:00
coissac 47e6b4108c New version of obiclean with clustering options 2013-11-14 15:03:11 +00:00
Frédéric Boyer 5917031d61 [MOD] now handle dictionaries, in such case, sequences can then be found in several files named after the keys of the dictionary 2013-11-05 14:14:24 +00:00
coissac 411c5134ae Path 2 error in the option helps 2013-10-29 14:02:24 +00:00
Céline Mercier 22217920a5 fixed obiaddtaxids bug 2013-10-24 14:32:27 +00:00
Céline Mercier 621b499cb8 fixed obitaxonomy bug 2013-10-24 13:32:44 +00:00
Céline Mercier 2e18284dc0 fixed obitaxonomy bug 2013-10-24 13:29:20 +00:00
Céline Mercier fc5d914093 obiaddtaxids now handle homonymes 2013-10-24 12:55:46 +00:00
Céline Mercier 7a7ee58a51 taxonomy databases now handle homonymes 2013-10-24 12:15:00 +00:00
Céline Mercier 82d92f8497 commented )D option in obitaxonomy because it doesn't work 2013-10-18 12:03:49 +00:00
Céline Mercier 47920a4687 obitaxonomy now checks if taxon already exists under the specific ancestor + changed -A option: if genus already existing it MUST be under restricting ancestor + fixed minor bug 2013-10-16 13:53:16 +00:00
coissac 12c080d7f4 rename --rank option to --seq-rank option 2013-09-19 01:58:11 +00:00
Céline Mercier 7b3886974d better handling of options in obiaddtaxids 2013-09-10 09:57:34 +00:00
Céline Mercier 0686029075 -w option in obitaxonomy 2013-09-04 12:07:16 +00:00
Céline Mercier ca36a4fbfe fixed bug 2013-09-04 10:28:56 +00:00
Céline Mercier cbf007b5a9 -w option in obitaxonomy 2013-09-04 10:27:02 +00:00
Céline Mercier 7e8835295a figuring out a bug 2013-09-04 10:11:10 +00:00
Céline Mercier e3aaa53be1 removed a print 2013-09-04 10:00:20 +00:00
Céline Mercier fbdf0db3c5 changed the UNITE parser in obiaddtaxids to remove empty fields ('-') 2013-09-04 09:46:26 +00:00
Frédéric Boyer 3b10f97587 [MOD] updated doc 2013-08-01 15:11:38 +00:00
coissac ff287fc31d Patch variance estimator for returning 0 for singleton 2013-06-22 08:31:52 +00:00
coissac 3cea0f40fe patch a bug in obisample on the sampling with replacement method 2013-06-18 08:42:48 +00:00
coissac 7bf04063f8 add doc to the manifest 2013-05-20 10:12:31 +00:00
386 changed files with 60493 additions and 152 deletions
+1 -1
View File
@@ -1,7 +1,7 @@
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<?eclipse-pydev version="1.0"?><pydev_project>
<pydev_pathproperty name="org.python.pydev.PROJECT_SOURCE_PATH">
<path>/OBITools-1.1/src</path>
<path>/obitools/src</path>
<path>/${PROJECT_DIR_NAME}/distutils.ext</path>
</pydev_pathproperty>
<pydev_property name="org.python.pydev.PYTHON_PROJECT_VERSION">python 2.7</pydev_property>
@@ -7,6 +7,7 @@ Created on 20 oct. 2012
from distutils.command.build import build as ori_build
from obidistutils.serenity.checksystem import is_mac_system
from distutils import log
class build(ori_build):
@@ -39,4 +40,8 @@ class build(ori_build):
('build_cexe', has_executables)] \
+ ori_build.sub_commands + \
[('build_sphinx',has_doc)]
def run(self):
log.info('\n\nRunning obidistutils build process\n\n')
ori_build.run(self)
@@ -96,7 +96,7 @@ class build_ext(ori_build_ext):
# from obidistutils.serenity.cython import get_a_cython_module
# from obidistutils.serenity.getcython import get_a_cython_module
# import imp
# import os.path
#
@@ -18,21 +18,21 @@ except ImportError:
log.info("Restarting installation with all dependencies ok")
rerun_with_anothe_python(os.path.realpath(sys.executable))
class build_sphinx(ori_build_sphinx):
'''
Build Sphinx documentation in html, epub and man formats
'''
description = __doc__
def run(self):
self.builder='html'
self.finalize_options()
ori_build_sphinx.run(self)
self.builder='epub'
self.finalize_options()
ori_build_sphinx.run(self)
self.builder='man'
self.finalize_options()
ori_build_sphinx.run(self)
class build_sphinx(ori_build_sphinx):
'''
Build Sphinx documentation in html, epub and man formats
'''
description = __doc__
def run(self):
self.builder='html'
self.finalize_options()
ori_build_sphinx.run(self)
self.builder='epub'
self.finalize_options()
ori_build_sphinx.run(self)
self.builder='man'
self.finalize_options()
ori_build_sphinx.run(self)
@@ -11,9 +11,15 @@ except ImportError:
from distutils.command.install import install as install_ori
has_setuptools=False
from distutils import log
class install(install_ori):
def __init__(self,dist):
install_ori.__init__(self, dist)
self.sub_commands.insert(0, ('build',lambda self: True))
self.sub_commands.append(('install_sphinx',lambda self: self.distribution.serenity))
def run(self):
log.info('\n\nRunning obidistutils install process\n\n')
install_ori.run(self)
@@ -8,7 +8,7 @@ import os
from obidistutils.command.build_exe import build_exe
from obidistutils.serenity.checksystem import is_mac_system
from distutils import log
class pidname(build_exe):
@@ -44,7 +44,9 @@ class pidname(build_exe):
def run(self):
if is_mac_system():
log.info("Building pidname...")
build_exe.run(self)
log.info("Done")
self.pidname=True
else:
self.pidname=False
+2
View File
@@ -201,5 +201,7 @@ def setup(**attrs):
if 'ext_modules' not in attrs:
attrs['ext_modules'] = EXTENTION
print "$$$$$$$$$$$$$$$$$$$$$$$$$$$",COMMANDS
ori_setup(**attrs)
@@ -5,10 +5,12 @@ Created on 2 oct. 2014
'''
from distutils import util
from distutils import log
def is_mac_system():
platform = util.get_platform().split('-')[0]
if platform=='macosx':
log.info('You are running on a Mac platform')
return platform=='macosx'
def is_windows_system():
Binary file not shown.
@@ -0,0 +1,277 @@
#!/usr/bin/env python
import os
import optparse
import sys
import re
from pip.exceptions import InstallationError, CommandError, PipError
from pip.log import logger
from pip.util import get_installed_distributions, get_prog
from pip.vcs import git, mercurial, subversion, bazaar # noqa
from pip.baseparser import ConfigOptionParser, UpdatingDefaultsHelpFormatter
from pip.commands import commands, get_summaries, get_similar_commands
# This fixes a peculiarity when importing via __import__ - as we are
# initialising the pip module, "from pip import cmdoptions" is recursive
# and appears not to work properly in that situation.
import pip.cmdoptions
cmdoptions = pip.cmdoptions
# The version as used in the setup.py and the docs conf.py
__version__ = "1.5.6"
def autocomplete():
"""Command and option completion for the main option parser (and options)
and its subcommands (and options).
Enable by sourcing one of the completion shell scripts (bash or zsh).
"""
# Don't complete if user hasn't sourced bash_completion file.
if 'PIP_AUTO_COMPLETE' not in os.environ:
return
cwords = os.environ['COMP_WORDS'].split()[1:]
cword = int(os.environ['COMP_CWORD'])
try:
current = cwords[cword - 1]
except IndexError:
current = ''
subcommands = [cmd for cmd, summary in get_summaries()]
options = []
# subcommand
try:
subcommand_name = [w for w in cwords if w in subcommands][0]
except IndexError:
subcommand_name = None
parser = create_main_parser()
# subcommand options
if subcommand_name:
# special case: 'help' subcommand has no options
if subcommand_name == 'help':
sys.exit(1)
# special case: list locally installed dists for uninstall command
if subcommand_name == 'uninstall' and not current.startswith('-'):
installed = []
lc = current.lower()
for dist in get_installed_distributions(local_only=True):
if dist.key.startswith(lc) and dist.key not in cwords[1:]:
installed.append(dist.key)
# if there are no dists installed, fall back to option completion
if installed:
for dist in installed:
print(dist)
sys.exit(1)
subcommand = commands[subcommand_name]()
options += [(opt.get_opt_string(), opt.nargs)
for opt in subcommand.parser.option_list_all
if opt.help != optparse.SUPPRESS_HELP]
# filter out previously specified options from available options
prev_opts = [x.split('=')[0] for x in cwords[1:cword - 1]]
options = [(x, v) for (x, v) in options if x not in prev_opts]
# filter options by current input
options = [(k, v) for k, v in options if k.startswith(current)]
for option in options:
opt_label = option[0]
# append '=' to options which require args
if option[1]:
opt_label += '='
print(opt_label)
else:
# show main parser options only when necessary
if current.startswith('-') or current.startswith('--'):
opts = [i.option_list for i in parser.option_groups]
opts.append(parser.option_list)
opts = (o for it in opts for o in it)
subcommands += [i.get_opt_string() for i in opts
if i.help != optparse.SUPPRESS_HELP]
print(' '.join([x for x in subcommands if x.startswith(current)]))
sys.exit(1)
def create_main_parser():
parser_kw = {
'usage': '\n%prog <command> [options]',
'add_help_option': False,
'formatter': UpdatingDefaultsHelpFormatter(),
'name': 'global',
'prog': get_prog(),
}
parser = ConfigOptionParser(**parser_kw)
parser.disable_interspersed_args()
pip_pkg_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
parser.version = 'pip %s from %s (python %s)' % (
__version__, pip_pkg_dir, sys.version[:3])
# add the general options
gen_opts = cmdoptions.make_option_group(cmdoptions.general_group, parser)
parser.add_option_group(gen_opts)
parser.main = True # so the help formatter knows
# create command listing for description
command_summaries = get_summaries()
description = [''] + ['%-27s %s' % (i, j) for i, j in command_summaries]
parser.description = '\n'.join(description)
return parser
def parseopts(args):
parser = create_main_parser()
# Note: parser calls disable_interspersed_args(), so the result of this call
# is to split the initial args into the general options before the
# subcommand and everything else.
# For example:
# args: ['--timeout=5', 'install', '--user', 'INITools']
# general_options: ['--timeout==5']
# args_else: ['install', '--user', 'INITools']
general_options, args_else = parser.parse_args(args)
# --version
if general_options.version:
sys.stdout.write(parser.version)
sys.stdout.write(os.linesep)
sys.exit()
# pip || pip help -> print_help()
if not args_else or (args_else[0] == 'help' and len(args_else) == 1):
parser.print_help()
sys.exit()
# the subcommand name
cmd_name = args_else[0].lower()
#all the args without the subcommand
cmd_args = args[:]
cmd_args.remove(args_else[0].lower())
if cmd_name not in commands:
guess = get_similar_commands(cmd_name)
msg = ['unknown command "%s"' % cmd_name]
if guess:
msg.append('maybe you meant "%s"' % guess)
raise CommandError(' - '.join(msg))
return cmd_name, cmd_args
def main(initial_args=None):
if initial_args is None:
initial_args = sys.argv[1:]
autocomplete()
try:
cmd_name, cmd_args = parseopts(initial_args)
except PipError:
e = sys.exc_info()[1]
sys.stderr.write("ERROR: %s" % e)
sys.stderr.write(os.linesep)
sys.exit(1)
command = commands[cmd_name]()
return command.main(cmd_args)
def bootstrap():
"""
Bootstrapping function to be called from install-pip.py script.
"""
pkgs = ['pip']
try:
import setuptools
except ImportError:
pkgs.append('setuptools')
return main(['install', '--upgrade'] + pkgs + sys.argv[1:])
############################################################
## Writing freeze files
class FrozenRequirement(object):
def __init__(self, name, req, editable, comments=()):
self.name = name
self.req = req
self.editable = editable
self.comments = comments
_rev_re = re.compile(r'-r(\d+)$')
_date_re = re.compile(r'-(20\d\d\d\d\d\d)$')
@classmethod
def from_dist(cls, dist, dependency_links, find_tags=False):
location = os.path.normcase(os.path.abspath(dist.location))
comments = []
from pip.vcs import vcs, get_src_requirement
if vcs.get_backend_name(location):
editable = True
try:
req = get_src_requirement(dist, location, find_tags)
except InstallationError:
ex = sys.exc_info()[1]
logger.warn("Error when trying to get requirement for VCS system %s, falling back to uneditable format" % ex)
req = None
if req is None:
logger.warn('Could not determine repository location of %s' % location)
comments.append('## !! Could not determine repository location')
req = dist.as_requirement()
editable = False
else:
editable = False
req = dist.as_requirement()
specs = req.specs
assert len(specs) == 1 and specs[0][0] == '=='
version = specs[0][1]
ver_match = cls._rev_re.search(version)
date_match = cls._date_re.search(version)
if ver_match or date_match:
svn_backend = vcs.get_backend('svn')
if svn_backend:
svn_location = svn_backend(
).get_location(dist, dependency_links)
if not svn_location:
logger.warn(
'Warning: cannot find svn location for %s' % req)
comments.append('## FIXME: could not find svn URL in dependency_links for this package:')
else:
comments.append('# Installing as editable to satisfy requirement %s:' % req)
if ver_match:
rev = ver_match.group(1)
else:
rev = '{%s}' % date_match.group(1)
editable = True
req = '%s@%s#egg=%s' % (svn_location, rev, cls.egg_name(dist))
return cls(dist.project_name, req, editable, comments)
@staticmethod
def egg_name(dist):
name = dist.egg_name()
match = re.search(r'-py\d\.\d$', name)
if match:
name = name[:match.start()]
return name
def __str__(self):
req = self.req
if self.editable:
req = '-e %s' % req
return '\n'.join(list(self.comments) + [str(req)]) + '\n'
if __name__ == '__main__':
exit = main()
if exit:
sys.exit(exit)
Binary file not shown.
@@ -0,0 +1,7 @@
import sys
from .runner import run
if __name__ == '__main__':
exit = run()
if exit:
sys.exit(exit)
Binary file not shown.
@@ -0,0 +1,8 @@
"""
pip._vendor is for vendoring dependencies of pip to prevent needing pip to
depend on something external.
Files inside of pip._vendor should be considered immutable and should only be
updated to versions from upstream.
"""
from __future__ import absolute_import
@@ -0,0 +1,16 @@
try:
import ast
from pip._vendor._markerlib.markers import default_environment, compile, interpret
except ImportError:
if 'ast' in globals():
raise
def default_environment():
return {}
def compile(marker):
def marker_fn(environment=None, override=None):
# 'empty markers are True' heuristic won't install extra deps.
return not marker.strip()
marker_fn.__doc__ = marker
return marker_fn
def interpret(marker, environment=None, override=None):
return compile(marker)()
@@ -0,0 +1,119 @@
# -*- coding: utf-8 -*-
"""Interpret PEP 345 environment markers.
EXPR [in|==|!=|not in] EXPR [or|and] ...
where EXPR belongs to any of those:
python_version = '%s.%s' % (sys.version_info[0], sys.version_info[1])
python_full_version = sys.version.split()[0]
os.name = os.name
sys.platform = sys.platform
platform.version = platform.version()
platform.machine = platform.machine()
platform.python_implementation = platform.python_implementation()
a free string, like '2.6', or 'win32'
"""
__all__ = ['default_environment', 'compile', 'interpret']
import ast
import os
import platform
import sys
import weakref
_builtin_compile = compile
try:
from platform import python_implementation
except ImportError:
if os.name == "java":
# Jython 2.5 has ast module, but not platform.python_implementation() function.
def python_implementation():
return "Jython"
else:
raise
# restricted set of variables
_VARS = {'sys.platform': sys.platform,
'python_version': '%s.%s' % sys.version_info[:2],
# FIXME parsing sys.platform is not reliable, but there is no other
# way to get e.g. 2.7.2+, and the PEP is defined with sys.version
'python_full_version': sys.version.split(' ', 1)[0],
'os.name': os.name,
'platform.version': platform.version(),
'platform.machine': platform.machine(),
'platform.python_implementation': python_implementation(),
'extra': None # wheel extension
}
for var in list(_VARS.keys()):
if '.' in var:
_VARS[var.replace('.', '_')] = _VARS[var]
def default_environment():
"""Return copy of default PEP 385 globals dictionary."""
return dict(_VARS)
class ASTWhitelist(ast.NodeTransformer):
def __init__(self, statement):
self.statement = statement # for error messages
ALLOWED = (ast.Compare, ast.BoolOp, ast.Attribute, ast.Name, ast.Load, ast.Str)
# Bool operations
ALLOWED += (ast.And, ast.Or)
# Comparison operations
ALLOWED += (ast.Eq, ast.Gt, ast.GtE, ast.In, ast.Is, ast.IsNot, ast.Lt, ast.LtE, ast.NotEq, ast.NotIn)
def visit(self, node):
"""Ensure statement only contains allowed nodes."""
if not isinstance(node, self.ALLOWED):
raise SyntaxError('Not allowed in environment markers.\n%s\n%s' %
(self.statement,
(' ' * node.col_offset) + '^'))
return ast.NodeTransformer.visit(self, node)
def visit_Attribute(self, node):
"""Flatten one level of attribute access."""
new_node = ast.Name("%s.%s" % (node.value.id, node.attr), node.ctx)
return ast.copy_location(new_node, node)
def parse_marker(marker):
tree = ast.parse(marker, mode='eval')
new_tree = ASTWhitelist(marker).generic_visit(tree)
return new_tree
def compile_marker(parsed_marker):
return _builtin_compile(parsed_marker, '<environment marker>', 'eval',
dont_inherit=True)
_cache = weakref.WeakValueDictionary()
def compile(marker):
"""Return compiled marker as a function accepting an environment dict."""
try:
return _cache[marker]
except KeyError:
pass
if not marker.strip():
def marker_fn(environment=None, override=None):
""""""
return True
else:
compiled_marker = compile_marker(parse_marker(marker))
def marker_fn(environment=None, override=None):
"""override updates environment"""
if override is None:
override = {}
if environment is None:
environment = default_environment()
environment.update(override)
return eval(compiled_marker, environment)
marker_fn.__doc__ = marker
_cache[marker] = marker_fn
return _cache[marker]
def interpret(marker, environment=None):
return compile(marker)(environment)
@@ -0,0 +1,7 @@
# Copyright Jonathan Hartley 2013. BSD 3-Clause license, see LICENSE file.
from .initialise import init, deinit, reinit
from .ansi import Fore, Back, Style
from .ansitowin32 import AnsiToWin32
__version__ = '0.3.1'
@@ -0,0 +1,50 @@
# Copyright Jonathan Hartley 2013. BSD 3-Clause license, see LICENSE file.
'''
This module generates ANSI character codes to printing colors to terminals.
See: http://en.wikipedia.org/wiki/ANSI_escape_code
'''
CSI = '\033['
def code_to_chars(code):
return CSI + str(code) + 'm'
class AnsiCodes(object):
def __init__(self, codes):
for name in dir(codes):
if not name.startswith('_'):
value = getattr(codes, name)
setattr(self, name, code_to_chars(value))
class AnsiFore:
BLACK = 30
RED = 31
GREEN = 32
YELLOW = 33
BLUE = 34
MAGENTA = 35
CYAN = 36
WHITE = 37
RESET = 39
class AnsiBack:
BLACK = 40
RED = 41
GREEN = 42
YELLOW = 43
BLUE = 44
MAGENTA = 45
CYAN = 46
WHITE = 47
RESET = 49
class AnsiStyle:
BRIGHT = 1
DIM = 2
NORMAL = 22
RESET_ALL = 0
Fore = AnsiCodes( AnsiFore )
Back = AnsiCodes( AnsiBack )
Style = AnsiCodes( AnsiStyle )
@@ -0,0 +1,190 @@
# Copyright Jonathan Hartley 2013. BSD 3-Clause license, see LICENSE file.
import re
import sys
from .ansi import AnsiFore, AnsiBack, AnsiStyle, Style
from .winterm import WinTerm, WinColor, WinStyle
from .win32 import windll
winterm = None
if windll is not None:
winterm = WinTerm()
def is_a_tty(stream):
return hasattr(stream, 'isatty') and stream.isatty()
class StreamWrapper(object):
'''
Wraps a stream (such as stdout), acting as a transparent proxy for all
attribute access apart from method 'write()', which is delegated to our
Converter instance.
'''
def __init__(self, wrapped, converter):
# double-underscore everything to prevent clashes with names of
# attributes on the wrapped stream object.
self.__wrapped = wrapped
self.__convertor = converter
def __getattr__(self, name):
return getattr(self.__wrapped, name)
def write(self, text):
self.__convertor.write(text)
class AnsiToWin32(object):
'''
Implements a 'write()' method which, on Windows, will strip ANSI character
sequences from the text, and if outputting to a tty, will convert them into
win32 function calls.
'''
ANSI_RE = re.compile('\033\[((?:\d|;)*)([a-zA-Z])')
def __init__(self, wrapped, convert=None, strip=None, autoreset=False):
# The wrapped stream (normally sys.stdout or sys.stderr)
self.wrapped = wrapped
# should we reset colors to defaults after every .write()
self.autoreset = autoreset
# create the proxy wrapping our output stream
self.stream = StreamWrapper(wrapped, self)
on_windows = sys.platform.startswith('win')
# should we strip ANSI sequences from our output?
if strip is None:
strip = on_windows
self.strip = strip
# should we should convert ANSI sequences into win32 calls?
if convert is None:
convert = on_windows and is_a_tty(wrapped)
self.convert = convert
# dict of ansi codes to win32 functions and parameters
self.win32_calls = self.get_win32_calls()
# are we wrapping stderr?
self.on_stderr = self.wrapped is sys.stderr
def should_wrap(self):
'''
True if this class is actually needed. If false, then the output
stream will not be affected, nor will win32 calls be issued, so
wrapping stdout is not actually required. This will generally be
False on non-Windows platforms, unless optional functionality like
autoreset has been requested using kwargs to init()
'''
return self.convert or self.strip or self.autoreset
def get_win32_calls(self):
if self.convert and winterm:
return {
AnsiStyle.RESET_ALL: (winterm.reset_all, ),
AnsiStyle.BRIGHT: (winterm.style, WinStyle.BRIGHT),
AnsiStyle.DIM: (winterm.style, WinStyle.NORMAL),
AnsiStyle.NORMAL: (winterm.style, WinStyle.NORMAL),
AnsiFore.BLACK: (winterm.fore, WinColor.BLACK),
AnsiFore.RED: (winterm.fore, WinColor.RED),
AnsiFore.GREEN: (winterm.fore, WinColor.GREEN),
AnsiFore.YELLOW: (winterm.fore, WinColor.YELLOW),
AnsiFore.BLUE: (winterm.fore, WinColor.BLUE),
AnsiFore.MAGENTA: (winterm.fore, WinColor.MAGENTA),
AnsiFore.CYAN: (winterm.fore, WinColor.CYAN),
AnsiFore.WHITE: (winterm.fore, WinColor.GREY),
AnsiFore.RESET: (winterm.fore, ),
AnsiBack.BLACK: (winterm.back, WinColor.BLACK),
AnsiBack.RED: (winterm.back, WinColor.RED),
AnsiBack.GREEN: (winterm.back, WinColor.GREEN),
AnsiBack.YELLOW: (winterm.back, WinColor.YELLOW),
AnsiBack.BLUE: (winterm.back, WinColor.BLUE),
AnsiBack.MAGENTA: (winterm.back, WinColor.MAGENTA),
AnsiBack.CYAN: (winterm.back, WinColor.CYAN),
AnsiBack.WHITE: (winterm.back, WinColor.GREY),
AnsiBack.RESET: (winterm.back, ),
}
def write(self, text):
if self.strip or self.convert:
self.write_and_convert(text)
else:
self.wrapped.write(text)
self.wrapped.flush()
if self.autoreset:
self.reset_all()
def reset_all(self):
if self.convert:
self.call_win32('m', (0,))
elif not self.wrapped.closed and is_a_tty(self.wrapped):
self.wrapped.write(Style.RESET_ALL)
def write_and_convert(self, text):
'''
Write the given text to our wrapped stream, stripping any ANSI
sequences from the text, and optionally converting them into win32
calls.
'''
cursor = 0
for match in self.ANSI_RE.finditer(text):
start, end = match.span()
self.write_plain_text(text, cursor, start)
self.convert_ansi(*match.groups())
cursor = end
self.write_plain_text(text, cursor, len(text))
def write_plain_text(self, text, start, end):
if start < end:
self.wrapped.write(text[start:end])
self.wrapped.flush()
def convert_ansi(self, paramstring, command):
if self.convert:
params = self.extract_params(paramstring)
self.call_win32(command, params)
def extract_params(self, paramstring):
def split(paramstring):
for p in paramstring.split(';'):
if p != '':
yield int(p)
return tuple(split(paramstring))
def call_win32(self, command, params):
if params == []:
params = [0]
if command == 'm':
for param in params:
if param in self.win32_calls:
func_args = self.win32_calls[param]
func = func_args[0]
args = func_args[1:]
kwargs = dict(on_stderr=self.on_stderr)
func(*args, **kwargs)
elif command in ('H', 'f'): # set cursor position
func = winterm.set_cursor_position
func(params, on_stderr=self.on_stderr)
elif command in ('J'):
func = winterm.erase_data
func(params, on_stderr=self.on_stderr)
elif command == 'A':
if params == () or params == None:
num_rows = 1
else:
num_rows = params[0]
func = winterm.cursor_up
func(num_rows, on_stderr=self.on_stderr)
@@ -0,0 +1,56 @@
# Copyright Jonathan Hartley 2013. BSD 3-Clause license, see LICENSE file.
import atexit
import sys
from .ansitowin32 import AnsiToWin32
orig_stdout = sys.stdout
orig_stderr = sys.stderr
wrapped_stdout = sys.stdout
wrapped_stderr = sys.stderr
atexit_done = False
def reset_all():
AnsiToWin32(orig_stdout).reset_all()
def init(autoreset=False, convert=None, strip=None, wrap=True):
if not wrap and any([autoreset, convert, strip]):
raise ValueError('wrap=False conflicts with any other arg=True')
global wrapped_stdout, wrapped_stderr
sys.stdout = wrapped_stdout = \
wrap_stream(orig_stdout, convert, strip, autoreset, wrap)
sys.stderr = wrapped_stderr = \
wrap_stream(orig_stderr, convert, strip, autoreset, wrap)
global atexit_done
if not atexit_done:
atexit.register(reset_all)
atexit_done = True
def deinit():
sys.stdout = orig_stdout
sys.stderr = orig_stderr
def reinit():
sys.stdout = wrapped_stdout
sys.stderr = wrapped_stdout
def wrap_stream(stream, convert, strip, autoreset, wrap):
if wrap:
wrapper = AnsiToWin32(stream,
convert=convert, strip=strip, autoreset=autoreset)
if wrapper.should_wrap():
stream = wrapper.stream
return stream
@@ -0,0 +1,137 @@
# Copyright Jonathan Hartley 2013. BSD 3-Clause license, see LICENSE file.
# from winbase.h
STDOUT = -11
STDERR = -12
import ctypes
from ctypes import LibraryLoader
try:
windll = LibraryLoader(ctypes.WinDLL)
from ctypes import wintypes
except (AttributeError, ImportError):
windll = None
SetConsoleTextAttribute = lambda *_: None
else:
from ctypes import (
byref, Structure, c_char, c_short, c_uint32, c_ushort, POINTER
)
class CONSOLE_SCREEN_BUFFER_INFO(Structure):
"""struct in wincon.h."""
_fields_ = [
("dwSize", wintypes._COORD),
("dwCursorPosition", wintypes._COORD),
("wAttributes", wintypes.WORD),
("srWindow", wintypes.SMALL_RECT),
("dwMaximumWindowSize", wintypes._COORD),
]
def __str__(self):
return '(%d,%d,%d,%d,%d,%d,%d,%d,%d,%d,%d)' % (
self.dwSize.Y, self.dwSize.X
, self.dwCursorPosition.Y, self.dwCursorPosition.X
, self.wAttributes
, self.srWindow.Top, self.srWindow.Left, self.srWindow.Bottom, self.srWindow.Right
, self.dwMaximumWindowSize.Y, self.dwMaximumWindowSize.X
)
_GetStdHandle = windll.kernel32.GetStdHandle
_GetStdHandle.argtypes = [
wintypes.DWORD,
]
_GetStdHandle.restype = wintypes.HANDLE
_GetConsoleScreenBufferInfo = windll.kernel32.GetConsoleScreenBufferInfo
_GetConsoleScreenBufferInfo.argtypes = [
wintypes.HANDLE,
POINTER(CONSOLE_SCREEN_BUFFER_INFO),
]
_GetConsoleScreenBufferInfo.restype = wintypes.BOOL
_SetConsoleTextAttribute = windll.kernel32.SetConsoleTextAttribute
_SetConsoleTextAttribute.argtypes = [
wintypes.HANDLE,
wintypes.WORD,
]
_SetConsoleTextAttribute.restype = wintypes.BOOL
_SetConsoleCursorPosition = windll.kernel32.SetConsoleCursorPosition
_SetConsoleCursorPosition.argtypes = [
wintypes.HANDLE,
wintypes._COORD,
]
_SetConsoleCursorPosition.restype = wintypes.BOOL
_FillConsoleOutputCharacterA = windll.kernel32.FillConsoleOutputCharacterA
_FillConsoleOutputCharacterA.argtypes = [
wintypes.HANDLE,
c_char,
wintypes.DWORD,
wintypes._COORD,
POINTER(wintypes.DWORD),
]
_FillConsoleOutputCharacterA.restype = wintypes.BOOL
_FillConsoleOutputAttribute = windll.kernel32.FillConsoleOutputAttribute
_FillConsoleOutputAttribute.argtypes = [
wintypes.HANDLE,
wintypes.WORD,
wintypes.DWORD,
wintypes._COORD,
POINTER(wintypes.DWORD),
]
_FillConsoleOutputAttribute.restype = wintypes.BOOL
handles = {
STDOUT: _GetStdHandle(STDOUT),
STDERR: _GetStdHandle(STDERR),
}
def GetConsoleScreenBufferInfo(stream_id=STDOUT):
handle = handles[stream_id]
csbi = CONSOLE_SCREEN_BUFFER_INFO()
success = _GetConsoleScreenBufferInfo(
handle, byref(csbi))
return csbi
def SetConsoleTextAttribute(stream_id, attrs):
handle = handles[stream_id]
return _SetConsoleTextAttribute(handle, attrs)
def SetConsoleCursorPosition(stream_id, position):
position = wintypes._COORD(*position)
# If the position is out of range, do nothing.
if position.Y <= 0 or position.X <= 0:
return
# Adjust for Windows' SetConsoleCursorPosition:
# 1. being 0-based, while ANSI is 1-based.
# 2. expecting (x,y), while ANSI uses (y,x).
adjusted_position = wintypes._COORD(position.Y - 1, position.X - 1)
# Adjust for viewport's scroll position
sr = GetConsoleScreenBufferInfo(STDOUT).srWindow
adjusted_position.Y += sr.Top
adjusted_position.X += sr.Left
# Resume normal processing
handle = handles[stream_id]
return _SetConsoleCursorPosition(handle, adjusted_position)
def FillConsoleOutputCharacter(stream_id, char, length, start):
handle = handles[stream_id]
char = c_char(char)
length = wintypes.DWORD(length)
num_written = wintypes.DWORD(0)
# Note that this is hard-coded for ANSI (vs wide) bytes.
success = _FillConsoleOutputCharacterA(
handle, char, length, start, byref(num_written))
return num_written.value
def FillConsoleOutputAttribute(stream_id, attr, length, start):
''' FillConsoleOutputAttribute( hConsole, csbi.wAttributes, dwConSize, coordScreen, &cCharsWritten )'''
handle = handles[stream_id]
attribute = wintypes.WORD(attr)
length = wintypes.DWORD(length)
num_written = wintypes.DWORD(0)
# Note that this is hard-coded for ANSI (vs wide) bytes.
return _FillConsoleOutputAttribute(
handle, attribute, length, start, byref(num_written))
@@ -0,0 +1,120 @@
# Copyright Jonathan Hartley 2013. BSD 3-Clause license, see LICENSE file.
from . import win32
# from wincon.h
class WinColor(object):
BLACK = 0
BLUE = 1
GREEN = 2
CYAN = 3
RED = 4
MAGENTA = 5
YELLOW = 6
GREY = 7
# from wincon.h
class WinStyle(object):
NORMAL = 0x00 # dim text, dim background
BRIGHT = 0x08 # bright text, dim background
class WinTerm(object):
def __init__(self):
self._default = win32.GetConsoleScreenBufferInfo(win32.STDOUT).wAttributes
self.set_attrs(self._default)
self._default_fore = self._fore
self._default_back = self._back
self._default_style = self._style
def get_attrs(self):
return self._fore + self._back * 16 + self._style
def set_attrs(self, value):
self._fore = value & 7
self._back = (value >> 4) & 7
self._style = value & WinStyle.BRIGHT
def reset_all(self, on_stderr=None):
self.set_attrs(self._default)
self.set_console(attrs=self._default)
def fore(self, fore=None, on_stderr=False):
if fore is None:
fore = self._default_fore
self._fore = fore
self.set_console(on_stderr=on_stderr)
def back(self, back=None, on_stderr=False):
if back is None:
back = self._default_back
self._back = back
self.set_console(on_stderr=on_stderr)
def style(self, style=None, on_stderr=False):
if style is None:
style = self._default_style
self._style = style
self.set_console(on_stderr=on_stderr)
def set_console(self, attrs=None, on_stderr=False):
if attrs is None:
attrs = self.get_attrs()
handle = win32.STDOUT
if on_stderr:
handle = win32.STDERR
win32.SetConsoleTextAttribute(handle, attrs)
def get_position(self, handle):
position = win32.GetConsoleScreenBufferInfo(handle).dwCursorPosition
# Because Windows coordinates are 0-based,
# and win32.SetConsoleCursorPosition expects 1-based.
position.X += 1
position.Y += 1
return position
def set_cursor_position(self, position=None, on_stderr=False):
if position is None:
#I'm not currently tracking the position, so there is no default.
#position = self.get_position()
return
handle = win32.STDOUT
if on_stderr:
handle = win32.STDERR
win32.SetConsoleCursorPosition(handle, position)
def cursor_up(self, num_rows=0, on_stderr=False):
if num_rows == 0:
return
handle = win32.STDOUT
if on_stderr:
handle = win32.STDERR
position = self.get_position(handle)
adjusted_position = (position.Y - num_rows, position.X)
self.set_cursor_position(adjusted_position, on_stderr)
def erase_data(self, mode=0, on_stderr=False):
# 0 (or None) should clear from the cursor to the end of the screen.
# 1 should clear from the cursor to the beginning of the screen.
# 2 should clear the entire screen. (And maybe move cursor to (1,1)?)
#
# At the moment, I only support mode 2. From looking at the API, it
# should be possible to calculate a different number of bytes to clear,
# and to do so relative to the cursor position.
if mode[0] not in (2,):
return
handle = win32.STDOUT
if on_stderr:
handle = win32.STDERR
# here's where we'll home the cursor
coord_screen = win32.COORD(0,0)
csbi = win32.GetConsoleScreenBufferInfo(handle)
# get the number of character cells in the current buffer
dw_con_size = csbi.dwSize.X * csbi.dwSize.Y
# fill the entire screen with blanks
win32.FillConsoleOutputCharacter(handle, ' ', dw_con_size, coord_screen)
# now set the buffer's attributes accordingly
win32.FillConsoleOutputAttribute(handle, self.get_attrs(), dw_con_size, coord_screen );
# put the cursor at (0, 0)
win32.SetConsoleCursorPosition(handle, (coord_screen.X, coord_screen.Y))
@@ -0,0 +1,23 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2012-2014 Vinay Sajip.
# Licensed to the Python Software Foundation under a contributor agreement.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
import logging
__version__ = '0.1.8'
class DistlibException(Exception):
pass
try:
from logging import NullHandler
except ImportError: # pragma: no cover
class NullHandler(logging.Handler):
def handle(self, record): pass
def emit(self, record): pass
def createLock(self): self.lock = None
logger = logging.getLogger(__name__)
logger.addHandler(NullHandler())
@@ -0,0 +1,6 @@
"""Modules copied from Python 3 standard libraries, for internal use only.
Individual classes and functions are found in d2._backport.misc. Intended
usage is to always import things missing from 3.1 from that module: the
built-in/stdlib objects will be used if found.
"""
@@ -0,0 +1,41 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2012 The Python Software Foundation.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
"""Backports for individual classes and functions."""
import os
import sys
__all__ = ['cache_from_source', 'callable', 'fsencode']
try:
from imp import cache_from_source
except ImportError:
def cache_from_source(py_file, debug=__debug__):
ext = debug and 'c' or 'o'
return py_file + ext
try:
callable = callable
except NameError:
from collections import Callable
def callable(obj):
return isinstance(obj, Callable)
try:
fsencode = os.fsencode
except AttributeError:
def fsencode(filename):
if isinstance(filename, bytes):
return filename
elif isinstance(filename, str):
return filename.encode(sys.getfilesystemencoding())
else:
raise TypeError("expect bytes or str, not %s" %
type(filename).__name__)
@@ -0,0 +1,761 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2012 The Python Software Foundation.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
"""Utility functions for copying and archiving files and directory trees.
XXX The functions here don't copy the resource fork or other metadata on Mac.
"""
import os
import sys
import stat
from os.path import abspath
import fnmatch
import collections
import errno
from . import tarfile
try:
import bz2
_BZ2_SUPPORTED = True
except ImportError:
_BZ2_SUPPORTED = False
try:
from pwd import getpwnam
except ImportError:
getpwnam = None
try:
from grp import getgrnam
except ImportError:
getgrnam = None
__all__ = ["copyfileobj", "copyfile", "copymode", "copystat", "copy", "copy2",
"copytree", "move", "rmtree", "Error", "SpecialFileError",
"ExecError", "make_archive", "get_archive_formats",
"register_archive_format", "unregister_archive_format",
"get_unpack_formats", "register_unpack_format",
"unregister_unpack_format", "unpack_archive", "ignore_patterns"]
class Error(EnvironmentError):
pass
class SpecialFileError(EnvironmentError):
"""Raised when trying to do a kind of operation (e.g. copying) which is
not supported on a special file (e.g. a named pipe)"""
class ExecError(EnvironmentError):
"""Raised when a command could not be executed"""
class ReadError(EnvironmentError):
"""Raised when an archive cannot be read"""
class RegistryError(Exception):
"""Raised when a registery operation with the archiving
and unpacking registeries fails"""
try:
WindowsError
except NameError:
WindowsError = None
def copyfileobj(fsrc, fdst, length=16*1024):
"""copy data from file-like object fsrc to file-like object fdst"""
while 1:
buf = fsrc.read(length)
if not buf:
break
fdst.write(buf)
def _samefile(src, dst):
# Macintosh, Unix.
if hasattr(os.path, 'samefile'):
try:
return os.path.samefile(src, dst)
except OSError:
return False
# All other platforms: check for same pathname.
return (os.path.normcase(os.path.abspath(src)) ==
os.path.normcase(os.path.abspath(dst)))
def copyfile(src, dst):
"""Copy data from src to dst"""
if _samefile(src, dst):
raise Error("`%s` and `%s` are the same file" % (src, dst))
for fn in [src, dst]:
try:
st = os.stat(fn)
except OSError:
# File most likely does not exist
pass
else:
# XXX What about other special files? (sockets, devices...)
if stat.S_ISFIFO(st.st_mode):
raise SpecialFileError("`%s` is a named pipe" % fn)
with open(src, 'rb') as fsrc:
with open(dst, 'wb') as fdst:
copyfileobj(fsrc, fdst)
def copymode(src, dst):
"""Copy mode bits from src to dst"""
if hasattr(os, 'chmod'):
st = os.stat(src)
mode = stat.S_IMODE(st.st_mode)
os.chmod(dst, mode)
def copystat(src, dst):
"""Copy all stat info (mode bits, atime, mtime, flags) from src to dst"""
st = os.stat(src)
mode = stat.S_IMODE(st.st_mode)
if hasattr(os, 'utime'):
os.utime(dst, (st.st_atime, st.st_mtime))
if hasattr(os, 'chmod'):
os.chmod(dst, mode)
if hasattr(os, 'chflags') and hasattr(st, 'st_flags'):
try:
os.chflags(dst, st.st_flags)
except OSError as why:
if (not hasattr(errno, 'EOPNOTSUPP') or
why.errno != errno.EOPNOTSUPP):
raise
def copy(src, dst):
"""Copy data and mode bits ("cp src dst").
The destination may be a directory.
"""
if os.path.isdir(dst):
dst = os.path.join(dst, os.path.basename(src))
copyfile(src, dst)
copymode(src, dst)
def copy2(src, dst):
"""Copy data and all stat info ("cp -p src dst").
The destination may be a directory.
"""
if os.path.isdir(dst):
dst = os.path.join(dst, os.path.basename(src))
copyfile(src, dst)
copystat(src, dst)
def ignore_patterns(*patterns):
"""Function that can be used as copytree() ignore parameter.
Patterns is a sequence of glob-style patterns
that are used to exclude files"""
def _ignore_patterns(path, names):
ignored_names = []
for pattern in patterns:
ignored_names.extend(fnmatch.filter(names, pattern))
return set(ignored_names)
return _ignore_patterns
def copytree(src, dst, symlinks=False, ignore=None, copy_function=copy2,
ignore_dangling_symlinks=False):
"""Recursively copy a directory tree.
The destination directory must not already exist.
If exception(s) occur, an Error is raised with a list of reasons.
If the optional symlinks flag is true, symbolic links in the
source tree result in symbolic links in the destination tree; if
it is false, the contents of the files pointed to by symbolic
links are copied. If the file pointed by the symlink doesn't
exist, an exception will be added in the list of errors raised in
an Error exception at the end of the copy process.
You can set the optional ignore_dangling_symlinks flag to true if you
want to silence this exception. Notice that this has no effect on
platforms that don't support os.symlink.
The optional ignore argument is a callable. If given, it
is called with the `src` parameter, which is the directory
being visited by copytree(), and `names` which is the list of
`src` contents, as returned by os.listdir():
callable(src, names) -> ignored_names
Since copytree() is called recursively, the callable will be
called once for each directory that is copied. It returns a
list of names relative to the `src` directory that should
not be copied.
The optional copy_function argument is a callable that will be used
to copy each file. It will be called with the source path and the
destination path as arguments. By default, copy2() is used, but any
function that supports the same signature (like copy()) can be used.
"""
names = os.listdir(src)
if ignore is not None:
ignored_names = ignore(src, names)
else:
ignored_names = set()
os.makedirs(dst)
errors = []
for name in names:
if name in ignored_names:
continue
srcname = os.path.join(src, name)
dstname = os.path.join(dst, name)
try:
if os.path.islink(srcname):
linkto = os.readlink(srcname)
if symlinks:
os.symlink(linkto, dstname)
else:
# ignore dangling symlink if the flag is on
if not os.path.exists(linkto) and ignore_dangling_symlinks:
continue
# otherwise let the copy occurs. copy2 will raise an error
copy_function(srcname, dstname)
elif os.path.isdir(srcname):
copytree(srcname, dstname, symlinks, ignore, copy_function)
else:
# Will raise a SpecialFileError for unsupported file types
copy_function(srcname, dstname)
# catch the Error from the recursive copytree so that we can
# continue with other files
except Error as err:
errors.extend(err.args[0])
except EnvironmentError as why:
errors.append((srcname, dstname, str(why)))
try:
copystat(src, dst)
except OSError as why:
if WindowsError is not None and isinstance(why, WindowsError):
# Copying file access times may fail on Windows
pass
else:
errors.extend((src, dst, str(why)))
if errors:
raise Error(errors)
def rmtree(path, ignore_errors=False, onerror=None):
"""Recursively delete a directory tree.
If ignore_errors is set, errors are ignored; otherwise, if onerror
is set, it is called to handle the error with arguments (func,
path, exc_info) where func is os.listdir, os.remove, or os.rmdir;
path is the argument to that function that caused it to fail; and
exc_info is a tuple returned by sys.exc_info(). If ignore_errors
is false and onerror is None, an exception is raised.
"""
if ignore_errors:
def onerror(*args):
pass
elif onerror is None:
def onerror(*args):
raise
try:
if os.path.islink(path):
# symlinks to directories are forbidden, see bug #1669
raise OSError("Cannot call rmtree on a symbolic link")
except OSError:
onerror(os.path.islink, path, sys.exc_info())
# can't continue even if onerror hook returns
return
names = []
try:
names = os.listdir(path)
except os.error:
onerror(os.listdir, path, sys.exc_info())
for name in names:
fullname = os.path.join(path, name)
try:
mode = os.lstat(fullname).st_mode
except os.error:
mode = 0
if stat.S_ISDIR(mode):
rmtree(fullname, ignore_errors, onerror)
else:
try:
os.remove(fullname)
except os.error:
onerror(os.remove, fullname, sys.exc_info())
try:
os.rmdir(path)
except os.error:
onerror(os.rmdir, path, sys.exc_info())
def _basename(path):
# A basename() variant which first strips the trailing slash, if present.
# Thus we always get the last component of the path, even for directories.
return os.path.basename(path.rstrip(os.path.sep))
def move(src, dst):
"""Recursively move a file or directory to another location. This is
similar to the Unix "mv" command.
If the destination is a directory or a symlink to a directory, the source
is moved inside the directory. The destination path must not already
exist.
If the destination already exists but is not a directory, it may be
overwritten depending on os.rename() semantics.
If the destination is on our current filesystem, then rename() is used.
Otherwise, src is copied to the destination and then removed.
A lot more could be done here... A look at a mv.c shows a lot of
the issues this implementation glosses over.
"""
real_dst = dst
if os.path.isdir(dst):
if _samefile(src, dst):
# We might be on a case insensitive filesystem,
# perform the rename anyway.
os.rename(src, dst)
return
real_dst = os.path.join(dst, _basename(src))
if os.path.exists(real_dst):
raise Error("Destination path '%s' already exists" % real_dst)
try:
os.rename(src, real_dst)
except OSError:
if os.path.isdir(src):
if _destinsrc(src, dst):
raise Error("Cannot move a directory '%s' into itself '%s'." % (src, dst))
copytree(src, real_dst, symlinks=True)
rmtree(src)
else:
copy2(src, real_dst)
os.unlink(src)
def _destinsrc(src, dst):
src = abspath(src)
dst = abspath(dst)
if not src.endswith(os.path.sep):
src += os.path.sep
if not dst.endswith(os.path.sep):
dst += os.path.sep
return dst.startswith(src)
def _get_gid(name):
"""Returns a gid, given a group name."""
if getgrnam is None or name is None:
return None
try:
result = getgrnam(name)
except KeyError:
result = None
if result is not None:
return result[2]
return None
def _get_uid(name):
"""Returns an uid, given a user name."""
if getpwnam is None or name is None:
return None
try:
result = getpwnam(name)
except KeyError:
result = None
if result is not None:
return result[2]
return None
def _make_tarball(base_name, base_dir, compress="gzip", verbose=0, dry_run=0,
owner=None, group=None, logger=None):
"""Create a (possibly compressed) tar file from all the files under
'base_dir'.
'compress' must be "gzip" (the default), "bzip2", or None.
'owner' and 'group' can be used to define an owner and a group for the
archive that is being built. If not provided, the current owner and group
will be used.
The output tar file will be named 'base_name' + ".tar", possibly plus
the appropriate compression extension (".gz", or ".bz2").
Returns the output filename.
"""
tar_compression = {'gzip': 'gz', None: ''}
compress_ext = {'gzip': '.gz'}
if _BZ2_SUPPORTED:
tar_compression['bzip2'] = 'bz2'
compress_ext['bzip2'] = '.bz2'
# flags for compression program, each element of list will be an argument
if compress is not None and compress not in compress_ext:
raise ValueError("bad value for 'compress', or compression format not "
"supported : {0}".format(compress))
archive_name = base_name + '.tar' + compress_ext.get(compress, '')
archive_dir = os.path.dirname(archive_name)
if not os.path.exists(archive_dir):
if logger is not None:
logger.info("creating %s", archive_dir)
if not dry_run:
os.makedirs(archive_dir)
# creating the tarball
if logger is not None:
logger.info('Creating tar archive')
uid = _get_uid(owner)
gid = _get_gid(group)
def _set_uid_gid(tarinfo):
if gid is not None:
tarinfo.gid = gid
tarinfo.gname = group
if uid is not None:
tarinfo.uid = uid
tarinfo.uname = owner
return tarinfo
if not dry_run:
tar = tarfile.open(archive_name, 'w|%s' % tar_compression[compress])
try:
tar.add(base_dir, filter=_set_uid_gid)
finally:
tar.close()
return archive_name
def _call_external_zip(base_dir, zip_filename, verbose=False, dry_run=False):
# XXX see if we want to keep an external call here
if verbose:
zipoptions = "-r"
else:
zipoptions = "-rq"
from distutils.errors import DistutilsExecError
from distutils.spawn import spawn
try:
spawn(["zip", zipoptions, zip_filename, base_dir], dry_run=dry_run)
except DistutilsExecError:
# XXX really should distinguish between "couldn't find
# external 'zip' command" and "zip failed".
raise ExecError("unable to create zip file '%s': "
"could neither import the 'zipfile' module nor "
"find a standalone zip utility") % zip_filename
def _make_zipfile(base_name, base_dir, verbose=0, dry_run=0, logger=None):
"""Create a zip file from all the files under 'base_dir'.
The output zip file will be named 'base_name' + ".zip". Uses either the
"zipfile" Python module (if available) or the InfoZIP "zip" utility
(if installed and found on the default search path). If neither tool is
available, raises ExecError. Returns the name of the output zip
file.
"""
zip_filename = base_name + ".zip"
archive_dir = os.path.dirname(base_name)
if not os.path.exists(archive_dir):
if logger is not None:
logger.info("creating %s", archive_dir)
if not dry_run:
os.makedirs(archive_dir)
# If zipfile module is not available, try spawning an external 'zip'
# command.
try:
import zipfile
except ImportError:
zipfile = None
if zipfile is None:
_call_external_zip(base_dir, zip_filename, verbose, dry_run)
else:
if logger is not None:
logger.info("creating '%s' and adding '%s' to it",
zip_filename, base_dir)
if not dry_run:
zip = zipfile.ZipFile(zip_filename, "w",
compression=zipfile.ZIP_DEFLATED)
for dirpath, dirnames, filenames in os.walk(base_dir):
for name in filenames:
path = os.path.normpath(os.path.join(dirpath, name))
if os.path.isfile(path):
zip.write(path, path)
if logger is not None:
logger.info("adding '%s'", path)
zip.close()
return zip_filename
_ARCHIVE_FORMATS = {
'gztar': (_make_tarball, [('compress', 'gzip')], "gzip'ed tar-file"),
'bztar': (_make_tarball, [('compress', 'bzip2')], "bzip2'ed tar-file"),
'tar': (_make_tarball, [('compress', None)], "uncompressed tar file"),
'zip': (_make_zipfile, [], "ZIP file"),
}
if _BZ2_SUPPORTED:
_ARCHIVE_FORMATS['bztar'] = (_make_tarball, [('compress', 'bzip2')],
"bzip2'ed tar-file")
def get_archive_formats():
"""Returns a list of supported formats for archiving and unarchiving.
Each element of the returned sequence is a tuple (name, description)
"""
formats = [(name, registry[2]) for name, registry in
_ARCHIVE_FORMATS.items()]
formats.sort()
return formats
def register_archive_format(name, function, extra_args=None, description=''):
"""Registers an archive format.
name is the name of the format. function is the callable that will be
used to create archives. If provided, extra_args is a sequence of
(name, value) tuples that will be passed as arguments to the callable.
description can be provided to describe the format, and will be returned
by the get_archive_formats() function.
"""
if extra_args is None:
extra_args = []
if not isinstance(function, collections.Callable):
raise TypeError('The %s object is not callable' % function)
if not isinstance(extra_args, (tuple, list)):
raise TypeError('extra_args needs to be a sequence')
for element in extra_args:
if not isinstance(element, (tuple, list)) or len(element) !=2:
raise TypeError('extra_args elements are : (arg_name, value)')
_ARCHIVE_FORMATS[name] = (function, extra_args, description)
def unregister_archive_format(name):
del _ARCHIVE_FORMATS[name]
def make_archive(base_name, format, root_dir=None, base_dir=None, verbose=0,
dry_run=0, owner=None, group=None, logger=None):
"""Create an archive file (eg. zip or tar).
'base_name' is the name of the file to create, minus any format-specific
extension; 'format' is the archive format: one of "zip", "tar", "bztar"
or "gztar".
'root_dir' is a directory that will be the root directory of the
archive; ie. we typically chdir into 'root_dir' before creating the
archive. 'base_dir' is the directory where we start archiving from;
ie. 'base_dir' will be the common prefix of all files and
directories in the archive. 'root_dir' and 'base_dir' both default
to the current directory. Returns the name of the archive file.
'owner' and 'group' are used when creating a tar archive. By default,
uses the current owner and group.
"""
save_cwd = os.getcwd()
if root_dir is not None:
if logger is not None:
logger.debug("changing into '%s'", root_dir)
base_name = os.path.abspath(base_name)
if not dry_run:
os.chdir(root_dir)
if base_dir is None:
base_dir = os.curdir
kwargs = {'dry_run': dry_run, 'logger': logger}
try:
format_info = _ARCHIVE_FORMATS[format]
except KeyError:
raise ValueError("unknown archive format '%s'" % format)
func = format_info[0]
for arg, val in format_info[1]:
kwargs[arg] = val
if format != 'zip':
kwargs['owner'] = owner
kwargs['group'] = group
try:
filename = func(base_name, base_dir, **kwargs)
finally:
if root_dir is not None:
if logger is not None:
logger.debug("changing back to '%s'", save_cwd)
os.chdir(save_cwd)
return filename
def get_unpack_formats():
"""Returns a list of supported formats for unpacking.
Each element of the returned sequence is a tuple
(name, extensions, description)
"""
formats = [(name, info[0], info[3]) for name, info in
_UNPACK_FORMATS.items()]
formats.sort()
return formats
def _check_unpack_options(extensions, function, extra_args):
"""Checks what gets registered as an unpacker."""
# first make sure no other unpacker is registered for this extension
existing_extensions = {}
for name, info in _UNPACK_FORMATS.items():
for ext in info[0]:
existing_extensions[ext] = name
for extension in extensions:
if extension in existing_extensions:
msg = '%s is already registered for "%s"'
raise RegistryError(msg % (extension,
existing_extensions[extension]))
if not isinstance(function, collections.Callable):
raise TypeError('The registered function must be a callable')
def register_unpack_format(name, extensions, function, extra_args=None,
description=''):
"""Registers an unpack format.
`name` is the name of the format. `extensions` is a list of extensions
corresponding to the format.
`function` is the callable that will be
used to unpack archives. The callable will receive archives to unpack.
If it's unable to handle an archive, it needs to raise a ReadError
exception.
If provided, `extra_args` is a sequence of
(name, value) tuples that will be passed as arguments to the callable.
description can be provided to describe the format, and will be returned
by the get_unpack_formats() function.
"""
if extra_args is None:
extra_args = []
_check_unpack_options(extensions, function, extra_args)
_UNPACK_FORMATS[name] = extensions, function, extra_args, description
def unregister_unpack_format(name):
"""Removes the pack format from the registery."""
del _UNPACK_FORMATS[name]
def _ensure_directory(path):
"""Ensure that the parent directory of `path` exists"""
dirname = os.path.dirname(path)
if not os.path.isdir(dirname):
os.makedirs(dirname)
def _unpack_zipfile(filename, extract_dir):
"""Unpack zip `filename` to `extract_dir`
"""
try:
import zipfile
except ImportError:
raise ReadError('zlib not supported, cannot unpack this archive.')
if not zipfile.is_zipfile(filename):
raise ReadError("%s is not a zip file" % filename)
zip = zipfile.ZipFile(filename)
try:
for info in zip.infolist():
name = info.filename
# don't extract absolute paths or ones with .. in them
if name.startswith('/') or '..' in name:
continue
target = os.path.join(extract_dir, *name.split('/'))
if not target:
continue
_ensure_directory(target)
if not name.endswith('/'):
# file
data = zip.read(info.filename)
f = open(target, 'wb')
try:
f.write(data)
finally:
f.close()
del data
finally:
zip.close()
def _unpack_tarfile(filename, extract_dir):
"""Unpack tar/tar.gz/tar.bz2 `filename` to `extract_dir`
"""
try:
tarobj = tarfile.open(filename)
except tarfile.TarError:
raise ReadError(
"%s is not a compressed or uncompressed tar file" % filename)
try:
tarobj.extractall(extract_dir)
finally:
tarobj.close()
_UNPACK_FORMATS = {
'gztar': (['.tar.gz', '.tgz'], _unpack_tarfile, [], "gzip'ed tar-file"),
'tar': (['.tar'], _unpack_tarfile, [], "uncompressed tar file"),
'zip': (['.zip'], _unpack_zipfile, [], "ZIP file")
}
if _BZ2_SUPPORTED:
_UNPACK_FORMATS['bztar'] = (['.bz2'], _unpack_tarfile, [],
"bzip2'ed tar-file")
def _find_unpack_format(filename):
for name, info in _UNPACK_FORMATS.items():
for extension in info[0]:
if filename.endswith(extension):
return name
return None
def unpack_archive(filename, extract_dir=None, format=None):
"""Unpack an archive.
`filename` is the name of the archive.
`extract_dir` is the name of the target directory, where the archive
is unpacked. If not provided, the current working directory is used.
`format` is the archive format: one of "zip", "tar", or "gztar". Or any
other registered format. If not provided, unpack_archive will use the
filename extension and see if an unpacker was registered for that
extension.
In case none is found, a ValueError is raised.
"""
if extract_dir is None:
extract_dir = os.getcwd()
if format is not None:
try:
format_info = _UNPACK_FORMATS[format]
except KeyError:
raise ValueError("Unknown unpack format '{0}'".format(format))
func = format_info[1]
func(filename, extract_dir, **dict(format_info[2]))
else:
# we need to look at the registered unpackers supported extensions
format = _find_unpack_format(filename)
if format is None:
raise ReadError("Unknown archive format '{0}'".format(filename))
func = _UNPACK_FORMATS[format][1]
kwargs = dict(_UNPACK_FORMATS[format][2])
func(filename, extract_dir, **kwargs)
@@ -0,0 +1,84 @@
[posix_prefix]
# Configuration directories. Some of these come straight out of the
# configure script. They are for implementing the other variables, not to
# be used directly in [resource_locations].
confdir = /etc
datadir = /usr/share
libdir = /usr/lib
statedir = /var
# User resource directory
local = ~/.local/{distribution.name}
stdlib = {base}/lib/python{py_version_short}
platstdlib = {platbase}/lib/python{py_version_short}
purelib = {base}/lib/python{py_version_short}/site-packages
platlib = {platbase}/lib/python{py_version_short}/site-packages
include = {base}/include/python{py_version_short}{abiflags}
platinclude = {platbase}/include/python{py_version_short}{abiflags}
data = {base}
[posix_home]
stdlib = {base}/lib/python
platstdlib = {base}/lib/python
purelib = {base}/lib/python
platlib = {base}/lib/python
include = {base}/include/python
platinclude = {base}/include/python
scripts = {base}/bin
data = {base}
[nt]
stdlib = {base}/Lib
platstdlib = {base}/Lib
purelib = {base}/Lib/site-packages
platlib = {base}/Lib/site-packages
include = {base}/Include
platinclude = {base}/Include
scripts = {base}/Scripts
data = {base}
[os2]
stdlib = {base}/Lib
platstdlib = {base}/Lib
purelib = {base}/Lib/site-packages
platlib = {base}/Lib/site-packages
include = {base}/Include
platinclude = {base}/Include
scripts = {base}/Scripts
data = {base}
[os2_home]
stdlib = {userbase}/lib/python{py_version_short}
platstdlib = {userbase}/lib/python{py_version_short}
purelib = {userbase}/lib/python{py_version_short}/site-packages
platlib = {userbase}/lib/python{py_version_short}/site-packages
include = {userbase}/include/python{py_version_short}
scripts = {userbase}/bin
data = {userbase}
[nt_user]
stdlib = {userbase}/Python{py_version_nodot}
platstdlib = {userbase}/Python{py_version_nodot}
purelib = {userbase}/Python{py_version_nodot}/site-packages
platlib = {userbase}/Python{py_version_nodot}/site-packages
include = {userbase}/Python{py_version_nodot}/Include
scripts = {userbase}/Scripts
data = {userbase}
[posix_user]
stdlib = {userbase}/lib/python{py_version_short}
platstdlib = {userbase}/lib/python{py_version_short}
purelib = {userbase}/lib/python{py_version_short}/site-packages
platlib = {userbase}/lib/python{py_version_short}/site-packages
include = {userbase}/include/python{py_version_short}
scripts = {userbase}/bin
data = {userbase}
[osx_framework_user]
stdlib = {userbase}/lib/python
platstdlib = {userbase}/lib/python
purelib = {userbase}/lib/python/site-packages
platlib = {userbase}/lib/python/site-packages
include = {userbase}/include
scripts = {userbase}/bin
data = {userbase}
@@ -0,0 +1,788 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2012 The Python Software Foundation.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
"""Access to Python's configuration information."""
import codecs
import os
import re
import sys
from os.path import pardir, realpath
try:
import configparser
except ImportError:
import ConfigParser as configparser
__all__ = [
'get_config_h_filename',
'get_config_var',
'get_config_vars',
'get_makefile_filename',
'get_path',
'get_path_names',
'get_paths',
'get_platform',
'get_python_version',
'get_scheme_names',
'parse_config_h',
]
def _safe_realpath(path):
try:
return realpath(path)
except OSError:
return path
if sys.executable:
_PROJECT_BASE = os.path.dirname(_safe_realpath(sys.executable))
else:
# sys.executable can be empty if argv[0] has been changed and Python is
# unable to retrieve the real program name
_PROJECT_BASE = _safe_realpath(os.getcwd())
if os.name == "nt" and "pcbuild" in _PROJECT_BASE[-8:].lower():
_PROJECT_BASE = _safe_realpath(os.path.join(_PROJECT_BASE, pardir))
# PC/VS7.1
if os.name == "nt" and "\\pc\\v" in _PROJECT_BASE[-10:].lower():
_PROJECT_BASE = _safe_realpath(os.path.join(_PROJECT_BASE, pardir, pardir))
# PC/AMD64
if os.name == "nt" and "\\pcbuild\\amd64" in _PROJECT_BASE[-14:].lower():
_PROJECT_BASE = _safe_realpath(os.path.join(_PROJECT_BASE, pardir, pardir))
def is_python_build():
for fn in ("Setup.dist", "Setup.local"):
if os.path.isfile(os.path.join(_PROJECT_BASE, "Modules", fn)):
return True
return False
_PYTHON_BUILD = is_python_build()
_cfg_read = False
def _ensure_cfg_read():
global _cfg_read
if not _cfg_read:
from ..resources import finder
backport_package = __name__.rsplit('.', 1)[0]
_finder = finder(backport_package)
_cfgfile = _finder.find('sysconfig.cfg')
assert _cfgfile, 'sysconfig.cfg exists'
with _cfgfile.as_stream() as s:
_SCHEMES.readfp(s)
if _PYTHON_BUILD:
for scheme in ('posix_prefix', 'posix_home'):
_SCHEMES.set(scheme, 'include', '{srcdir}/Include')
_SCHEMES.set(scheme, 'platinclude', '{projectbase}/.')
_cfg_read = True
_SCHEMES = configparser.RawConfigParser()
_VAR_REPL = re.compile(r'\{([^{]*?)\}')
def _expand_globals(config):
_ensure_cfg_read()
if config.has_section('globals'):
globals = config.items('globals')
else:
globals = tuple()
sections = config.sections()
for section in sections:
if section == 'globals':
continue
for option, value in globals:
if config.has_option(section, option):
continue
config.set(section, option, value)
config.remove_section('globals')
# now expanding local variables defined in the cfg file
#
for section in config.sections():
variables = dict(config.items(section))
def _replacer(matchobj):
name = matchobj.group(1)
if name in variables:
return variables[name]
return matchobj.group(0)
for option, value in config.items(section):
config.set(section, option, _VAR_REPL.sub(_replacer, value))
#_expand_globals(_SCHEMES)
# FIXME don't rely on sys.version here, its format is an implementation detail
# of CPython, use sys.version_info or sys.hexversion
_PY_VERSION = sys.version.split()[0]
_PY_VERSION_SHORT = sys.version[:3]
_PY_VERSION_SHORT_NO_DOT = _PY_VERSION[0] + _PY_VERSION[2]
_PREFIX = os.path.normpath(sys.prefix)
_EXEC_PREFIX = os.path.normpath(sys.exec_prefix)
_CONFIG_VARS = None
_USER_BASE = None
def _subst_vars(path, local_vars):
"""In the string `path`, replace tokens like {some.thing} with the
corresponding value from the map `local_vars`.
If there is no corresponding value, leave the token unchanged.
"""
def _replacer(matchobj):
name = matchobj.group(1)
if name in local_vars:
return local_vars[name]
elif name in os.environ:
return os.environ[name]
return matchobj.group(0)
return _VAR_REPL.sub(_replacer, path)
def _extend_dict(target_dict, other_dict):
target_keys = target_dict.keys()
for key, value in other_dict.items():
if key in target_keys:
continue
target_dict[key] = value
def _expand_vars(scheme, vars):
res = {}
if vars is None:
vars = {}
_extend_dict(vars, get_config_vars())
for key, value in _SCHEMES.items(scheme):
if os.name in ('posix', 'nt'):
value = os.path.expanduser(value)
res[key] = os.path.normpath(_subst_vars(value, vars))
return res
def format_value(value, vars):
def _replacer(matchobj):
name = matchobj.group(1)
if name in vars:
return vars[name]
return matchobj.group(0)
return _VAR_REPL.sub(_replacer, value)
def _get_default_scheme():
if os.name == 'posix':
# the default scheme for posix is posix_prefix
return 'posix_prefix'
return os.name
def _getuserbase():
env_base = os.environ.get("PYTHONUSERBASE", None)
def joinuser(*args):
return os.path.expanduser(os.path.join(*args))
# what about 'os2emx', 'riscos' ?
if os.name == "nt":
base = os.environ.get("APPDATA") or "~"
if env_base:
return env_base
else:
return joinuser(base, "Python")
if sys.platform == "darwin":
framework = get_config_var("PYTHONFRAMEWORK")
if framework:
if env_base:
return env_base
else:
return joinuser("~", "Library", framework, "%d.%d" %
sys.version_info[:2])
if env_base:
return env_base
else:
return joinuser("~", ".local")
def _parse_makefile(filename, vars=None):
"""Parse a Makefile-style file.
A dictionary containing name/value pairs is returned. If an
optional dictionary is passed in as the second argument, it is
used instead of a new dictionary.
"""
# Regexes needed for parsing Makefile (and similar syntaxes,
# like old-style Setup files).
_variable_rx = re.compile("([a-zA-Z][a-zA-Z0-9_]+)\s*=\s*(.*)")
_findvar1_rx = re.compile(r"\$\(([A-Za-z][A-Za-z0-9_]*)\)")
_findvar2_rx = re.compile(r"\${([A-Za-z][A-Za-z0-9_]*)}")
if vars is None:
vars = {}
done = {}
notdone = {}
with codecs.open(filename, encoding='utf-8', errors="surrogateescape") as f:
lines = f.readlines()
for line in lines:
if line.startswith('#') or line.strip() == '':
continue
m = _variable_rx.match(line)
if m:
n, v = m.group(1, 2)
v = v.strip()
# `$$' is a literal `$' in make
tmpv = v.replace('$$', '')
if "$" in tmpv:
notdone[n] = v
else:
try:
v = int(v)
except ValueError:
# insert literal `$'
done[n] = v.replace('$$', '$')
else:
done[n] = v
# do variable interpolation here
variables = list(notdone.keys())
# Variables with a 'PY_' prefix in the makefile. These need to
# be made available without that prefix through sysconfig.
# Special care is needed to ensure that variable expansion works, even
# if the expansion uses the name without a prefix.
renamed_variables = ('CFLAGS', 'LDFLAGS', 'CPPFLAGS')
while len(variables) > 0:
for name in tuple(variables):
value = notdone[name]
m = _findvar1_rx.search(value) or _findvar2_rx.search(value)
if m is not None:
n = m.group(1)
found = True
if n in done:
item = str(done[n])
elif n in notdone:
# get it on a subsequent round
found = False
elif n in os.environ:
# do it like make: fall back to environment
item = os.environ[n]
elif n in renamed_variables:
if (name.startswith('PY_') and
name[3:] in renamed_variables):
item = ""
elif 'PY_' + n in notdone:
found = False
else:
item = str(done['PY_' + n])
else:
done[n] = item = ""
if found:
after = value[m.end():]
value = value[:m.start()] + item + after
if "$" in after:
notdone[name] = value
else:
try:
value = int(value)
except ValueError:
done[name] = value.strip()
else:
done[name] = value
variables.remove(name)
if (name.startswith('PY_') and
name[3:] in renamed_variables):
name = name[3:]
if name not in done:
done[name] = value
else:
# bogus variable reference (e.g. "prefix=$/opt/python");
# just drop it since we can't deal
done[name] = value
variables.remove(name)
# strip spurious spaces
for k, v in done.items():
if isinstance(v, str):
done[k] = v.strip()
# save the results in the global dictionary
vars.update(done)
return vars
def get_makefile_filename():
"""Return the path of the Makefile."""
if _PYTHON_BUILD:
return os.path.join(_PROJECT_BASE, "Makefile")
if hasattr(sys, 'abiflags'):
config_dir_name = 'config-%s%s' % (_PY_VERSION_SHORT, sys.abiflags)
else:
config_dir_name = 'config'
return os.path.join(get_path('stdlib'), config_dir_name, 'Makefile')
def _init_posix(vars):
"""Initialize the module as appropriate for POSIX systems."""
# load the installed Makefile:
makefile = get_makefile_filename()
try:
_parse_makefile(makefile, vars)
except IOError as e:
msg = "invalid Python installation: unable to open %s" % makefile
if hasattr(e, "strerror"):
msg = msg + " (%s)" % e.strerror
raise IOError(msg)
# load the installed pyconfig.h:
config_h = get_config_h_filename()
try:
with open(config_h) as f:
parse_config_h(f, vars)
except IOError as e:
msg = "invalid Python installation: unable to open %s" % config_h
if hasattr(e, "strerror"):
msg = msg + " (%s)" % e.strerror
raise IOError(msg)
# On AIX, there are wrong paths to the linker scripts in the Makefile
# -- these paths are relative to the Python source, but when installed
# the scripts are in another directory.
if _PYTHON_BUILD:
vars['LDSHARED'] = vars['BLDSHARED']
def _init_non_posix(vars):
"""Initialize the module as appropriate for NT"""
# set basic install directories
vars['LIBDEST'] = get_path('stdlib')
vars['BINLIBDEST'] = get_path('platstdlib')
vars['INCLUDEPY'] = get_path('include')
vars['SO'] = '.pyd'
vars['EXE'] = '.exe'
vars['VERSION'] = _PY_VERSION_SHORT_NO_DOT
vars['BINDIR'] = os.path.dirname(_safe_realpath(sys.executable))
#
# public APIs
#
def parse_config_h(fp, vars=None):
"""Parse a config.h-style file.
A dictionary containing name/value pairs is returned. If an
optional dictionary is passed in as the second argument, it is
used instead of a new dictionary.
"""
if vars is None:
vars = {}
define_rx = re.compile("#define ([A-Z][A-Za-z0-9_]+) (.*)\n")
undef_rx = re.compile("/[*] #undef ([A-Z][A-Za-z0-9_]+) [*]/\n")
while True:
line = fp.readline()
if not line:
break
m = define_rx.match(line)
if m:
n, v = m.group(1, 2)
try:
v = int(v)
except ValueError:
pass
vars[n] = v
else:
m = undef_rx.match(line)
if m:
vars[m.group(1)] = 0
return vars
def get_config_h_filename():
"""Return the path of pyconfig.h."""
if _PYTHON_BUILD:
if os.name == "nt":
inc_dir = os.path.join(_PROJECT_BASE, "PC")
else:
inc_dir = _PROJECT_BASE
else:
inc_dir = get_path('platinclude')
return os.path.join(inc_dir, 'pyconfig.h')
def get_scheme_names():
"""Return a tuple containing the schemes names."""
return tuple(sorted(_SCHEMES.sections()))
def get_path_names():
"""Return a tuple containing the paths names."""
# xxx see if we want a static list
return _SCHEMES.options('posix_prefix')
def get_paths(scheme=_get_default_scheme(), vars=None, expand=True):
"""Return a mapping containing an install scheme.
``scheme`` is the install scheme name. If not provided, it will
return the default scheme for the current platform.
"""
_ensure_cfg_read()
if expand:
return _expand_vars(scheme, vars)
else:
return dict(_SCHEMES.items(scheme))
def get_path(name, scheme=_get_default_scheme(), vars=None, expand=True):
"""Return a path corresponding to the scheme.
``scheme`` is the install scheme name.
"""
return get_paths(scheme, vars, expand)[name]
def get_config_vars(*args):
"""With no arguments, return a dictionary of all configuration
variables relevant for the current platform.
On Unix, this means every variable defined in Python's installed Makefile;
On Windows and Mac OS it's a much smaller set.
With arguments, return a list of values that result from looking up
each argument in the configuration variable dictionary.
"""
global _CONFIG_VARS
if _CONFIG_VARS is None:
_CONFIG_VARS = {}
# Normalized versions of prefix and exec_prefix are handy to have;
# in fact, these are the standard versions used most places in the
# distutils2 module.
_CONFIG_VARS['prefix'] = _PREFIX
_CONFIG_VARS['exec_prefix'] = _EXEC_PREFIX
_CONFIG_VARS['py_version'] = _PY_VERSION
_CONFIG_VARS['py_version_short'] = _PY_VERSION_SHORT
_CONFIG_VARS['py_version_nodot'] = _PY_VERSION[0] + _PY_VERSION[2]
_CONFIG_VARS['base'] = _PREFIX
_CONFIG_VARS['platbase'] = _EXEC_PREFIX
_CONFIG_VARS['projectbase'] = _PROJECT_BASE
try:
_CONFIG_VARS['abiflags'] = sys.abiflags
except AttributeError:
# sys.abiflags may not be defined on all platforms.
_CONFIG_VARS['abiflags'] = ''
if os.name in ('nt', 'os2'):
_init_non_posix(_CONFIG_VARS)
if os.name == 'posix':
_init_posix(_CONFIG_VARS)
# Setting 'userbase' is done below the call to the
# init function to enable using 'get_config_var' in
# the init-function.
if sys.version >= '2.6':
_CONFIG_VARS['userbase'] = _getuserbase()
if 'srcdir' not in _CONFIG_VARS:
_CONFIG_VARS['srcdir'] = _PROJECT_BASE
else:
_CONFIG_VARS['srcdir'] = _safe_realpath(_CONFIG_VARS['srcdir'])
# Convert srcdir into an absolute path if it appears necessary.
# Normally it is relative to the build directory. However, during
# testing, for example, we might be running a non-installed python
# from a different directory.
if _PYTHON_BUILD and os.name == "posix":
base = _PROJECT_BASE
try:
cwd = os.getcwd()
except OSError:
cwd = None
if (not os.path.isabs(_CONFIG_VARS['srcdir']) and
base != cwd):
# srcdir is relative and we are not in the same directory
# as the executable. Assume executable is in the build
# directory and make srcdir absolute.
srcdir = os.path.join(base, _CONFIG_VARS['srcdir'])
_CONFIG_VARS['srcdir'] = os.path.normpath(srcdir)
if sys.platform == 'darwin':
kernel_version = os.uname()[2] # Kernel version (8.4.3)
major_version = int(kernel_version.split('.')[0])
if major_version < 8:
# On Mac OS X before 10.4, check if -arch and -isysroot
# are in CFLAGS or LDFLAGS and remove them if they are.
# This is needed when building extensions on a 10.3 system
# using a universal build of python.
for key in ('LDFLAGS', 'BASECFLAGS',
# a number of derived variables. These need to be
# patched up as well.
'CFLAGS', 'PY_CFLAGS', 'BLDSHARED'):
flags = _CONFIG_VARS[key]
flags = re.sub('-arch\s+\w+\s', ' ', flags)
flags = re.sub('-isysroot [^ \t]*', ' ', flags)
_CONFIG_VARS[key] = flags
else:
# Allow the user to override the architecture flags using
# an environment variable.
# NOTE: This name was introduced by Apple in OSX 10.5 and
# is used by several scripting languages distributed with
# that OS release.
if 'ARCHFLAGS' in os.environ:
arch = os.environ['ARCHFLAGS']
for key in ('LDFLAGS', 'BASECFLAGS',
# a number of derived variables. These need to be
# patched up as well.
'CFLAGS', 'PY_CFLAGS', 'BLDSHARED'):
flags = _CONFIG_VARS[key]
flags = re.sub('-arch\s+\w+\s', ' ', flags)
flags = flags + ' ' + arch
_CONFIG_VARS[key] = flags
# If we're on OSX 10.5 or later and the user tries to
# compiles an extension using an SDK that is not present
# on the current machine it is better to not use an SDK
# than to fail.
#
# The major usecase for this is users using a Python.org
# binary installer on OSX 10.6: that installer uses
# the 10.4u SDK, but that SDK is not installed by default
# when you install Xcode.
#
CFLAGS = _CONFIG_VARS.get('CFLAGS', '')
m = re.search('-isysroot\s+(\S+)', CFLAGS)
if m is not None:
sdk = m.group(1)
if not os.path.exists(sdk):
for key in ('LDFLAGS', 'BASECFLAGS',
# a number of derived variables. These need to be
# patched up as well.
'CFLAGS', 'PY_CFLAGS', 'BLDSHARED'):
flags = _CONFIG_VARS[key]
flags = re.sub('-isysroot\s+\S+(\s|$)', ' ', flags)
_CONFIG_VARS[key] = flags
if args:
vals = []
for name in args:
vals.append(_CONFIG_VARS.get(name))
return vals
else:
return _CONFIG_VARS
def get_config_var(name):
"""Return the value of a single variable using the dictionary returned by
'get_config_vars()'.
Equivalent to get_config_vars().get(name)
"""
return get_config_vars().get(name)
def get_platform():
"""Return a string that identifies the current platform.
This is used mainly to distinguish platform-specific build directories and
platform-specific built distributions. Typically includes the OS name
and version and the architecture (as supplied by 'os.uname()'),
although the exact information included depends on the OS; eg. for IRIX
the architecture isn't particularly important (IRIX only runs on SGI
hardware), but for Linux the kernel version isn't particularly
important.
Examples of returned values:
linux-i586
linux-alpha (?)
solaris-2.6-sun4u
irix-5.3
irix64-6.2
Windows will return one of:
win-amd64 (64bit Windows on AMD64 (aka x86_64, Intel64, EM64T, etc)
win-ia64 (64bit Windows on Itanium)
win32 (all others - specifically, sys.platform is returned)
For other non-POSIX platforms, currently just returns 'sys.platform'.
"""
if os.name == 'nt':
# sniff sys.version for architecture.
prefix = " bit ("
i = sys.version.find(prefix)
if i == -1:
return sys.platform
j = sys.version.find(")", i)
look = sys.version[i+len(prefix):j].lower()
if look == 'amd64':
return 'win-amd64'
if look == 'itanium':
return 'win-ia64'
return sys.platform
if os.name != "posix" or not hasattr(os, 'uname'):
# XXX what about the architecture? NT is Intel or Alpha,
# Mac OS is M68k or PPC, etc.
return sys.platform
# Try to distinguish various flavours of Unix
osname, host, release, version, machine = os.uname()
# Convert the OS name to lowercase, remove '/' characters
# (to accommodate BSD/OS), and translate spaces (for "Power Macintosh")
osname = osname.lower().replace('/', '')
machine = machine.replace(' ', '_')
machine = machine.replace('/', '-')
if osname[:5] == "linux":
# At least on Linux/Intel, 'machine' is the processor --
# i386, etc.
# XXX what about Alpha, SPARC, etc?
return "%s-%s" % (osname, machine)
elif osname[:5] == "sunos":
if release[0] >= "5": # SunOS 5 == Solaris 2
osname = "solaris"
release = "%d.%s" % (int(release[0]) - 3, release[2:])
# fall through to standard osname-release-machine representation
elif osname[:4] == "irix": # could be "irix64"!
return "%s-%s" % (osname, release)
elif osname[:3] == "aix":
return "%s-%s.%s" % (osname, version, release)
elif osname[:6] == "cygwin":
osname = "cygwin"
rel_re = re.compile(r'[\d.]+')
m = rel_re.match(release)
if m:
release = m.group()
elif osname[:6] == "darwin":
#
# For our purposes, we'll assume that the system version from
# distutils' perspective is what MACOSX_DEPLOYMENT_TARGET is set
# to. This makes the compatibility story a bit more sane because the
# machine is going to compile and link as if it were
# MACOSX_DEPLOYMENT_TARGET.
cfgvars = get_config_vars()
macver = cfgvars.get('MACOSX_DEPLOYMENT_TARGET')
if True:
# Always calculate the release of the running machine,
# needed to determine if we can build fat binaries or not.
macrelease = macver
# Get the system version. Reading this plist is a documented
# way to get the system version (see the documentation for
# the Gestalt Manager)
try:
f = open('/System/Library/CoreServices/SystemVersion.plist')
except IOError:
# We're on a plain darwin box, fall back to the default
# behaviour.
pass
else:
try:
m = re.search(r'<key>ProductUserVisibleVersion</key>\s*'
r'<string>(.*?)</string>', f.read())
finally:
f.close()
if m is not None:
macrelease = '.'.join(m.group(1).split('.')[:2])
# else: fall back to the default behaviour
if not macver:
macver = macrelease
if macver:
release = macver
osname = "macosx"
if ((macrelease + '.') >= '10.4.' and
'-arch' in get_config_vars().get('CFLAGS', '').strip()):
# The universal build will build fat binaries, but not on
# systems before 10.4
#
# Try to detect 4-way universal builds, those have machine-type
# 'universal' instead of 'fat'.
machine = 'fat'
cflags = get_config_vars().get('CFLAGS')
archs = re.findall('-arch\s+(\S+)', cflags)
archs = tuple(sorted(set(archs)))
if len(archs) == 1:
machine = archs[0]
elif archs == ('i386', 'ppc'):
machine = 'fat'
elif archs == ('i386', 'x86_64'):
machine = 'intel'
elif archs == ('i386', 'ppc', 'x86_64'):
machine = 'fat3'
elif archs == ('ppc64', 'x86_64'):
machine = 'fat64'
elif archs == ('i386', 'ppc', 'ppc64', 'x86_64'):
machine = 'universal'
else:
raise ValueError(
"Don't know machine value for archs=%r" % (archs,))
elif machine == 'i386':
# On OSX the machine type returned by uname is always the
# 32-bit variant, even if the executable architecture is
# the 64-bit variant
if sys.maxsize >= 2**32:
machine = 'x86_64'
elif machine in ('PowerPC', 'Power_Macintosh'):
# Pick a sane name for the PPC architecture.
# See 'i386' case
if sys.maxsize >= 2**32:
machine = 'ppc64'
else:
machine = 'ppc'
return "%s-%s-%s" % (osname, release, machine)
def get_python_version():
return _PY_VERSION_SHORT
def _print_dict(title, data):
for index, (key, value) in enumerate(sorted(data.items())):
if index == 0:
print('%s: ' % (title))
print('\t%s = "%s"' % (key, value))
def _main():
"""Display all information sysconfig detains."""
print('Platform: "%s"' % get_platform())
print('Python version: "%s"' % get_python_version())
print('Current installation scheme: "%s"' % _get_default_scheme())
print()
_print_dict('Paths', get_paths())
print()
_print_dict('Variables', get_config_vars())
if __name__ == '__main__':
_main()
File diff suppressed because it is too large Load Diff
File diff suppressed because it is too large Load Diff
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,488 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2013 Vinay Sajip.
# Licensed to the Python Software Foundation under a contributor agreement.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
import hashlib
import logging
import os
import shutil
import subprocess
import tempfile
try:
from threading import Thread
except ImportError:
from dummy_threading import Thread
from . import DistlibException
from .compat import (HTTPBasicAuthHandler, Request, HTTPPasswordMgr,
urlparse, build_opener, string_types)
from .util import cached_property, zip_dir, ServerProxy
logger = logging.getLogger(__name__)
DEFAULT_INDEX = 'https://pypi.python.org/pypi'
DEFAULT_REALM = 'pypi'
class PackageIndex(object):
"""
This class represents a package index compatible with PyPI, the Python
Package Index.
"""
boundary = b'----------ThIs_Is_tHe_distlib_index_bouNdaRY_$'
def __init__(self, url=None):
"""
Initialise an instance.
:param url: The URL of the index. If not specified, the URL for PyPI is
used.
"""
self.url = url or DEFAULT_INDEX
self.read_configuration()
scheme, netloc, path, params, query, frag = urlparse(self.url)
if params or query or frag or scheme not in ('http', 'https'):
raise DistlibException('invalid repository: %s' % self.url)
self.password_handler = None
self.ssl_verifier = None
self.gpg = None
self.gpg_home = None
self.rpc_proxy = None
with open(os.devnull, 'w') as sink:
for s in ('gpg2', 'gpg'):
try:
rc = subprocess.check_call([s, '--version'], stdout=sink,
stderr=sink)
if rc == 0:
self.gpg = s
break
except OSError:
pass
def _get_pypirc_command(self):
"""
Get the distutils command for interacting with PyPI configurations.
:return: the command.
"""
from distutils.core import Distribution
from distutils.config import PyPIRCCommand
d = Distribution()
return PyPIRCCommand(d)
def read_configuration(self):
"""
Read the PyPI access configuration as supported by distutils, getting
PyPI to do the acutal work. This populates ``username``, ``password``,
``realm`` and ``url`` attributes from the configuration.
"""
# get distutils to do the work
c = self._get_pypirc_command()
c.repository = self.url
cfg = c._read_pypirc()
self.username = cfg.get('username')
self.password = cfg.get('password')
self.realm = cfg.get('realm', 'pypi')
self.url = cfg.get('repository', self.url)
def save_configuration(self):
"""
Save the PyPI access configuration. You must have set ``username`` and
``password`` attributes before calling this method.
Again, distutils is used to do the actual work.
"""
self.check_credentials()
# get distutils to do the work
c = self._get_pypirc_command()
c._store_pypirc(self.username, self.password)
def check_credentials(self):
"""
Check that ``username`` and ``password`` have been set, and raise an
exception if not.
"""
if self.username is None or self.password is None:
raise DistlibException('username and password must be set')
pm = HTTPPasswordMgr()
_, netloc, _, _, _, _ = urlparse(self.url)
pm.add_password(self.realm, netloc, self.username, self.password)
self.password_handler = HTTPBasicAuthHandler(pm)
def register(self, metadata):
"""
Register a distribution on PyPI, using the provided metadata.
:param metadata: A :class:`Metadata` instance defining at least a name
and version number for the distribution to be
registered.
:return: The HTTP response received from PyPI upon submission of the
request.
"""
self.check_credentials()
metadata.validate()
d = metadata.todict()
d[':action'] = 'verify'
request = self.encode_request(d.items(), [])
response = self.send_request(request)
d[':action'] = 'submit'
request = self.encode_request(d.items(), [])
return self.send_request(request)
def _reader(self, name, stream, outbuf):
"""
Thread runner for reading lines of from a subprocess into a buffer.
:param name: The logical name of the stream (used for logging only).
:param stream: The stream to read from. This will typically a pipe
connected to the output stream of a subprocess.
:param outbuf: The list to append the read lines to.
"""
while True:
s = stream.readline()
if not s:
break
s = s.decode('utf-8').rstrip()
outbuf.append(s)
logger.debug('%s: %s' % (name, s))
stream.close()
def get_sign_command(self, filename, signer, sign_password):
"""
Return a suitable command for signing a file.
:param filename: The pathname to the file to be signed.
:param signer: The identifier of the signer of the file.
:param sign_password: The passphrase for the signer's
private key used for signing.
:return: The signing command as a list suitable to be
passed to :class:`subprocess.Popen`.
"""
cmd = [self.gpg, '--status-fd', '2', '--no-tty']
if self.gpg_home:
cmd.extend(['--homedir', self.gpg_home])
if sign_password is not None:
cmd.extend(['--batch', '--passphrase-fd', '0'])
td = tempfile.mkdtemp()
sf = os.path.join(td, os.path.basename(filename) + '.asc')
cmd.extend(['--detach-sign', '--armor', '--local-user',
signer, '--output', sf, filename])
logger.debug('invoking: %s', ' '.join(cmd))
return cmd, sf
def run_command(self, cmd, input_data=None):
"""
Run a command in a child process , passing it any input data specified.
:param cmd: The command to run.
:param input_data: If specified, this must be a byte string containing
data to be sent to the child process.
:return: A tuple consisting of the subprocess' exit code, a list of
lines read from the subprocess' ``stdout``, and a list of
lines read from the subprocess' ``stderr``.
"""
kwargs = {
'stdout': subprocess.PIPE,
'stderr': subprocess.PIPE,
}
if input_data is not None:
kwargs['stdin'] = subprocess.PIPE
stdout = []
stderr = []
p = subprocess.Popen(cmd, **kwargs)
# We don't use communicate() here because we may need to
# get clever with interacting with the command
t1 = Thread(target=self._reader, args=('stdout', p.stdout, stdout))
t1.start()
t2 = Thread(target=self._reader, args=('stderr', p.stderr, stderr))
t2.start()
if input_data is not None:
p.stdin.write(input_data)
p.stdin.close()
p.wait()
t1.join()
t2.join()
return p.returncode, stdout, stderr
def sign_file(self, filename, signer, sign_password):
"""
Sign a file.
:param filename: The pathname to the file to be signed.
:param signer: The identifier of the signer of the file.
:param sign_password: The passphrase for the signer's
private key used for signing.
:return: The absolute pathname of the file where the signature is
stored.
"""
cmd, sig_file = self.get_sign_command(filename, signer, sign_password)
rc, stdout, stderr = self.run_command(cmd,
sign_password.encode('utf-8'))
if rc != 0:
raise DistlibException('sign command failed with error '
'code %s' % rc)
return sig_file
def upload_file(self, metadata, filename, signer=None, sign_password=None,
filetype='sdist', pyversion='source'):
"""
Upload a release file to the index.
:param metadata: A :class:`Metadata` instance defining at least a name
and version number for the file to be uploaded.
:param filename: The pathname of the file to be uploaded.
:param signer: The identifier of the signer of the file.
:param sign_password: The passphrase for the signer's
private key used for signing.
:param filetype: The type of the file being uploaded. This is the
distutils command which produced that file, e.g.
``sdist`` or ``bdist_wheel``.
:param pyversion: The version of Python which the release relates
to. For code compatible with any Python, this would
be ``source``, otherwise it would be e.g. ``3.2``.
:return: The HTTP response received from PyPI upon submission of the
request.
"""
self.check_credentials()
if not os.path.exists(filename):
raise DistlibException('not found: %s' % filename)
metadata.validate()
d = metadata.todict()
sig_file = None
if signer:
if not self.gpg:
logger.warning('no signing program available - not signed')
else:
sig_file = self.sign_file(filename, signer, sign_password)
with open(filename, 'rb') as f:
file_data = f.read()
md5_digest = hashlib.md5(file_data).hexdigest()
sha256_digest = hashlib.sha256(file_data).hexdigest()
d.update({
':action': 'file_upload',
'protcol_version': '1',
'filetype': filetype,
'pyversion': pyversion,
'md5_digest': md5_digest,
'sha256_digest': sha256_digest,
})
files = [('content', os.path.basename(filename), file_data)]
if sig_file:
with open(sig_file, 'rb') as f:
sig_data = f.read()
files.append(('gpg_signature', os.path.basename(sig_file),
sig_data))
shutil.rmtree(os.path.dirname(sig_file))
request = self.encode_request(d.items(), files)
return self.send_request(request)
def upload_documentation(self, metadata, doc_dir):
"""
Upload documentation to the index.
:param metadata: A :class:`Metadata` instance defining at least a name
and version number for the documentation to be
uploaded.
:param doc_dir: The pathname of the directory which contains the
documentation. This should be the directory that
contains the ``index.html`` for the documentation.
:return: The HTTP response received from PyPI upon submission of the
request.
"""
self.check_credentials()
if not os.path.isdir(doc_dir):
raise DistlibException('not a directory: %r' % doc_dir)
fn = os.path.join(doc_dir, 'index.html')
if not os.path.exists(fn):
raise DistlibException('not found: %r' % fn)
metadata.validate()
name, version = metadata.name, metadata.version
zip_data = zip_dir(doc_dir).getvalue()
fields = [(':action', 'doc_upload'),
('name', name), ('version', version)]
files = [('content', name, zip_data)]
request = self.encode_request(fields, files)
return self.send_request(request)
def get_verify_command(self, signature_filename, data_filename):
"""
Return a suitable command for verifying a file.
:param signature_filename: The pathname to the file containing the
signature.
:param data_filename: The pathname to the file containing the
signed data.
:return: The verifying command as a list suitable to be
passed to :class:`subprocess.Popen`.
"""
cmd = [self.gpg, '--status-fd', '2', '--no-tty']
if self.gpg_home:
cmd.extend(['--homedir', self.gpg_home])
cmd.extend(['--verify', signature_filename, data_filename])
logger.debug('invoking: %s', ' '.join(cmd))
return cmd
def verify_signature(self, signature_filename, data_filename):
"""
Verify a signature for a file.
:param signature_filename: The pathname to the file containing the
signature.
:param data_filename: The pathname to the file containing the
signed data.
:return: True if the signature was verified, else False.
"""
if not self.gpg:
raise DistlibException('verification unavailable because gpg '
'unavailable')
cmd = self.get_verify_command(signature_filename, data_filename)
rc, stdout, stderr = self.run_command(cmd)
if rc not in (0, 1):
raise DistlibException('verify command failed with error '
'code %s' % rc)
return rc == 0
def download_file(self, url, destfile, digest=None, reporthook=None):
"""
This is a convenience method for downloading a file from an URL.
Normally, this will be a file from the index, though currently
no check is made for this (i.e. a file can be downloaded from
anywhere).
The method is just like the :func:`urlretrieve` function in the
standard library, except that it allows digest computation to be
done during download and checking that the downloaded data
matched any expected value.
:param url: The URL of the file to be downloaded (assumed to be
available via an HTTP GET request).
:param destfile: The pathname where the downloaded file is to be
saved.
:param digest: If specified, this must be a (hasher, value)
tuple, where hasher is the algorithm used (e.g.
``'md5'``) and ``value`` is the expected value.
:param reporthook: The same as for :func:`urlretrieve` in the
standard library.
"""
if digest is None:
digester = None
logger.debug('No digest specified')
else:
if isinstance(digest, (list, tuple)):
hasher, digest = digest
else:
hasher = 'md5'
digester = getattr(hashlib, hasher)()
logger.debug('Digest specified: %s' % digest)
# The following code is equivalent to urlretrieve.
# We need to do it this way so that we can compute the
# digest of the file as we go.
with open(destfile, 'wb') as dfp:
# addinfourl is not a context manager on 2.x
# so we have to use try/finally
sfp = self.send_request(Request(url))
try:
headers = sfp.info()
blocksize = 8192
size = -1
read = 0
blocknum = 0
if "content-length" in headers:
size = int(headers["Content-Length"])
if reporthook:
reporthook(blocknum, blocksize, size)
while True:
block = sfp.read(blocksize)
if not block:
break
read += len(block)
dfp.write(block)
if digester:
digester.update(block)
blocknum += 1
if reporthook:
reporthook(blocknum, blocksize, size)
finally:
sfp.close()
# check that we got the whole file, if we can
if size >= 0 and read < size:
raise DistlibException(
'retrieval incomplete: got only %d out of %d bytes'
% (read, size))
# if we have a digest, it must match.
if digester:
actual = digester.hexdigest()
if digest != actual:
raise DistlibException('%s digest mismatch for %s: expected '
'%s, got %s' % (hasher, destfile,
digest, actual))
logger.debug('Digest verified: %s', digest)
def send_request(self, req):
"""
Send a standard library :class:`Request` to PyPI and return its
response.
:param req: The request to send.
:return: The HTTP response from PyPI (a standard library HTTPResponse).
"""
handlers = []
if self.password_handler:
handlers.append(self.password_handler)
if self.ssl_verifier:
handlers.append(self.ssl_verifier)
opener = build_opener(*handlers)
return opener.open(req)
def encode_request(self, fields, files):
"""
Encode fields and files for posting to an HTTP server.
:param fields: The fields to send as a list of (fieldname, value)
tuples.
:param files: The files to send as a list of (fieldname, filename,
file_bytes) tuple.
"""
# Adapted from packaging, which in turn was adapted from
# http://code.activestate.com/recipes/146306
parts = []
boundary = self.boundary
for k, values in fields:
if not isinstance(values, (list, tuple)):
values = [values]
for v in values:
parts.extend((
b'--' + boundary,
('Content-Disposition: form-data; name="%s"' %
k).encode('utf-8'),
b'',
v.encode('utf-8')))
for key, filename, value in files:
parts.extend((
b'--' + boundary,
('Content-Disposition: form-data; name="%s"; filename="%s"' %
(key, filename)).encode('utf-8'),
b'',
value))
parts.extend((b'--' + boundary + b'--', b''))
body = b'\r\n'.join(parts)
ct = b'multipart/form-data; boundary=' + boundary
headers = {
'Content-type': ct,
'Content-length': str(len(body))
}
return Request(self.url, body, headers)
def search(self, terms, operator=None):
if isinstance(terms, string_types):
terms = {'name': terms}
if self.rpc_proxy is None:
self.rpc_proxy = ServerProxy(self.url, timeout=3.0)
return self.rpc_proxy.search(terms, operator or 'and')
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,364 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2012-2013 Python Software Foundation.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
"""
Class representing the list of files in a distribution.
Equivalent to distutils.filelist, but fixes some problems.
"""
import fnmatch
import logging
import os
import re
from . import DistlibException
from .compat import fsdecode
from .util import convert_path
__all__ = ['Manifest']
logger = logging.getLogger(__name__)
# a \ followed by some spaces + EOL
_COLLAPSE_PATTERN = re.compile('\\\w*\n', re.M)
_COMMENTED_LINE = re.compile('#.*?(?=\n)|\n(?=$)', re.M | re.S)
class Manifest(object):
"""A list of files built by on exploring the filesystem and filtered by
applying various patterns to what we find there.
"""
def __init__(self, base=None):
"""
Initialise an instance.
:param base: The base directory to explore under.
"""
self.base = os.path.abspath(os.path.normpath(base or os.getcwd()))
self.prefix = self.base + os.sep
self.allfiles = None
self.files = set()
#
# Public API
#
def findall(self):
"""Find all files under the base and set ``allfiles`` to the absolute
pathnames of files found.
"""
from stat import S_ISREG, S_ISDIR, S_ISLNK
self.allfiles = allfiles = []
root = self.base
stack = [root]
pop = stack.pop
push = stack.append
while stack:
root = pop()
names = os.listdir(root)
for name in names:
fullname = os.path.join(root, name)
# Avoid excess stat calls -- just one will do, thank you!
stat = os.stat(fullname)
mode = stat.st_mode
if S_ISREG(mode):
allfiles.append(fsdecode(fullname))
elif S_ISDIR(mode) and not S_ISLNK(mode):
push(fullname)
def add(self, item):
"""
Add a file to the manifest.
:param item: The pathname to add. This can be relative to the base.
"""
if not item.startswith(self.prefix):
item = os.path.join(self.base, item)
self.files.add(os.path.normpath(item))
def add_many(self, items):
"""
Add a list of files to the manifest.
:param items: The pathnames to add. These can be relative to the base.
"""
for item in items:
self.add(item)
def sorted(self, wantdirs=False):
"""
Return sorted files in directory order
"""
def add_dir(dirs, d):
dirs.add(d)
logger.debug('add_dir added %s', d)
if d != self.base:
parent, _ = os.path.split(d)
assert parent not in ('', '/')
add_dir(dirs, parent)
result = set(self.files) # make a copy!
if wantdirs:
dirs = set()
for f in result:
add_dir(dirs, os.path.dirname(f))
result |= dirs
return [os.path.join(*path_tuple) for path_tuple in
sorted(os.path.split(path) for path in result)]
def clear(self):
"""Clear all collected files."""
self.files = set()
self.allfiles = []
def process_directive(self, directive):
"""
Process a directive which either adds some files from ``allfiles`` to
``files``, or removes some files from ``files``.
:param directive: The directive to process. This should be in a format
compatible with distutils ``MANIFEST.in`` files:
http://docs.python.org/distutils/sourcedist.html#commands
"""
# Parse the line: split it up, make sure the right number of words
# is there, and return the relevant words. 'action' is always
# defined: it's the first word of the line. Which of the other
# three are defined depends on the action; it'll be either
# patterns, (dir and patterns), or (dirpattern).
action, patterns, thedir, dirpattern = self._parse_directive(directive)
# OK, now we know that the action is valid and we have the
# right number of words on the line for that action -- so we
# can proceed with minimal error-checking.
if action == 'include':
for pattern in patterns:
if not self._include_pattern(pattern, anchor=True):
logger.warning('no files found matching %r', pattern)
elif action == 'exclude':
for pattern in patterns:
if not self._exclude_pattern(pattern, anchor=True):
logger.warning('no previously-included files '
'found matching %r', pattern)
elif action == 'global-include':
for pattern in patterns:
if not self._include_pattern(pattern, anchor=False):
logger.warning('no files found matching %r '
'anywhere in distribution', pattern)
elif action == 'global-exclude':
for pattern in patterns:
if not self._exclude_pattern(pattern, anchor=False):
logger.warning('no previously-included files '
'matching %r found anywhere in '
'distribution', pattern)
elif action == 'recursive-include':
for pattern in patterns:
if not self._include_pattern(pattern, prefix=thedir):
logger.warning('no files found matching %r '
'under directory %r', pattern, thedir)
elif action == 'recursive-exclude':
for pattern in patterns:
if not self._exclude_pattern(pattern, prefix=thedir):
logger.warning('no previously-included files '
'matching %r found under directory %r',
pattern, thedir)
elif action == 'graft':
if not self._include_pattern(None, prefix=dirpattern):
logger.warning('no directories found matching %r',
dirpattern)
elif action == 'prune':
if not self._exclude_pattern(None, prefix=dirpattern):
logger.warning('no previously-included directories found '
'matching %r', dirpattern)
else: # pragma: no cover
# This should never happen, as it should be caught in
# _parse_template_line
raise DistlibException(
'invalid action %r' % action)
#
# Private API
#
def _parse_directive(self, directive):
"""
Validate a directive.
:param directive: The directive to validate.
:return: A tuple of action, patterns, thedir, dir_patterns
"""
words = directive.split()
if len(words) == 1 and words[0] not in ('include', 'exclude',
'global-include',
'global-exclude',
'recursive-include',
'recursive-exclude',
'graft', 'prune'):
# no action given, let's use the default 'include'
words.insert(0, 'include')
action = words[0]
patterns = thedir = dir_pattern = None
if action in ('include', 'exclude',
'global-include', 'global-exclude'):
if len(words) < 2:
raise DistlibException(
'%r expects <pattern1> <pattern2> ...' % action)
patterns = [convert_path(word) for word in words[1:]]
elif action in ('recursive-include', 'recursive-exclude'):
if len(words) < 3:
raise DistlibException(
'%r expects <dir> <pattern1> <pattern2> ...' % action)
thedir = convert_path(words[1])
patterns = [convert_path(word) for word in words[2:]]
elif action in ('graft', 'prune'):
if len(words) != 2:
raise DistlibException(
'%r expects a single <dir_pattern>' % action)
dir_pattern = convert_path(words[1])
else:
raise DistlibException('unknown action %r' % action)
return action, patterns, thedir, dir_pattern
def _include_pattern(self, pattern, anchor=True, prefix=None,
is_regex=False):
"""Select strings (presumably filenames) from 'self.files' that
match 'pattern', a Unix-style wildcard (glob) pattern.
Patterns are not quite the same as implemented by the 'fnmatch'
module: '*' and '?' match non-special characters, where "special"
is platform-dependent: slash on Unix; colon, slash, and backslash on
DOS/Windows; and colon on Mac OS.
If 'anchor' is true (the default), then the pattern match is more
stringent: "*.py" will match "foo.py" but not "foo/bar.py". If
'anchor' is false, both of these will match.
If 'prefix' is supplied, then only filenames starting with 'prefix'
(itself a pattern) and ending with 'pattern', with anything in between
them, will match. 'anchor' is ignored in this case.
If 'is_regex' is true, 'anchor' and 'prefix' are ignored, and
'pattern' is assumed to be either a string containing a regex or a
regex object -- no translation is done, the regex is just compiled
and used as-is.
Selected strings will be added to self.files.
Return True if files are found.
"""
# XXX docstring lying about what the special chars are?
found = False
pattern_re = self._translate_pattern(pattern, anchor, prefix, is_regex)
# delayed loading of allfiles list
if self.allfiles is None:
self.findall()
for name in self.allfiles:
if pattern_re.search(name):
self.files.add(name)
found = True
return found
def _exclude_pattern(self, pattern, anchor=True, prefix=None,
is_regex=False):
"""Remove strings (presumably filenames) from 'files' that match
'pattern'.
Other parameters are the same as for 'include_pattern()', above.
The list 'self.files' is modified in place. Return True if files are
found.
This API is public to allow e.g. exclusion of SCM subdirs, e.g. when
packaging source distributions
"""
found = False
pattern_re = self._translate_pattern(pattern, anchor, prefix, is_regex)
for f in list(self.files):
if pattern_re.search(f):
self.files.remove(f)
found = True
return found
def _translate_pattern(self, pattern, anchor=True, prefix=None,
is_regex=False):
"""Translate a shell-like wildcard pattern to a compiled regular
expression.
Return the compiled regex. If 'is_regex' true,
then 'pattern' is directly compiled to a regex (if it's a string)
or just returned as-is (assumes it's a regex object).
"""
if is_regex:
if isinstance(pattern, str):
return re.compile(pattern)
else:
return pattern
if pattern:
pattern_re = self._glob_to_re(pattern)
else:
pattern_re = ''
base = re.escape(os.path.join(self.base, ''))
if prefix is not None:
# ditch end of pattern character
empty_pattern = self._glob_to_re('')
prefix_re = self._glob_to_re(prefix)[:-len(empty_pattern)]
sep = os.sep
if os.sep == '\\':
sep = r'\\'
pattern_re = '^' + base + sep.join((prefix_re,
'.*' + pattern_re))
else: # no prefix -- respect anchor flag
if anchor:
pattern_re = '^' + base + pattern_re
return re.compile(pattern_re)
def _glob_to_re(self, pattern):
"""Translate a shell-like glob pattern to a regular expression.
Return a string containing the regex. Differs from
'fnmatch.translate()' in that '*' does not match "special characters"
(which are platform-specific).
"""
pattern_re = fnmatch.translate(pattern)
# '?' and '*' in the glob pattern become '.' and '.*' in the RE, which
# IMHO is wrong -- '?' and '*' aren't supposed to match slash in Unix,
# and by extension they shouldn't match such "special characters" under
# any OS. So change all non-escaped dots in the RE to match any
# character except the special characters (currently: just os.sep).
sep = os.sep
if os.sep == '\\':
# we're using a regex to manipulate a regex, so we need
# to escape the backslash twice
sep = r'\\\\'
escaped = r'\1[^%s]' % sep
pattern_re = re.sub(r'((?<!\\)(\\\\)*)\.', escaped, pattern_re)
return pattern_re
@@ -0,0 +1,190 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2012-2013 Vinay Sajip.
# Licensed to the Python Software Foundation under a contributor agreement.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
"""Parser for the environment markers micro-language defined in PEP 345."""
import ast
import os
import sys
import platform
from .compat import python_implementation, string_types
from .util import in_venv
__all__ = ['interpret']
class Evaluator(object):
"""
A limited evaluator for Python expressions.
"""
operators = {
'eq': lambda x, y: x == y,
'gt': lambda x, y: x > y,
'gte': lambda x, y: x >= y,
'in': lambda x, y: x in y,
'lt': lambda x, y: x < y,
'lte': lambda x, y: x <= y,
'not': lambda x: not x,
'noteq': lambda x, y: x != y,
'notin': lambda x, y: x not in y,
}
allowed_values = {
'sys_platform': sys.platform,
'python_version': '%s.%s' % sys.version_info[:2],
# parsing sys.platform is not reliable, but there is no other
# way to get e.g. 2.7.2+, and the PEP is defined with sys.version
'python_full_version': sys.version.split(' ', 1)[0],
'os_name': os.name,
'platform_in_venv': str(in_venv()),
'platform_release': platform.release(),
'platform_version': platform.version(),
'platform_machine': platform.machine(),
'platform_python_implementation': python_implementation(),
}
def __init__(self, context=None):
"""
Initialise an instance.
:param context: If specified, names are looked up in this mapping.
"""
self.context = context or {}
self.source = None
def get_fragment(self, offset):
"""
Get the part of the source which is causing a problem.
"""
fragment_len = 10
s = '%r' % (self.source[offset:offset + fragment_len])
if offset + fragment_len < len(self.source):
s += '...'
return s
def get_handler(self, node_type):
"""
Get a handler for the specified AST node type.
"""
return getattr(self, 'do_%s' % node_type, None)
def evaluate(self, node, filename=None):
"""
Evaluate a source string or node, using ``filename`` when
displaying errors.
"""
if isinstance(node, string_types):
self.source = node
kwargs = {'mode': 'eval'}
if filename:
kwargs['filename'] = filename
try:
node = ast.parse(node, **kwargs)
except SyntaxError as e:
s = self.get_fragment(e.offset)
raise SyntaxError('syntax error %s' % s)
node_type = node.__class__.__name__.lower()
handler = self.get_handler(node_type)
if handler is None:
if self.source is None:
s = '(source not available)'
else:
s = self.get_fragment(node.col_offset)
raise SyntaxError("don't know how to evaluate %r %s" % (
node_type, s))
return handler(node)
def get_attr_key(self, node):
assert isinstance(node, ast.Attribute), 'attribute node expected'
return '%s.%s' % (node.value.id, node.attr)
def do_attribute(self, node):
if not isinstance(node.value, ast.Name):
valid = False
else:
key = self.get_attr_key(node)
valid = key in self.context or key in self.allowed_values
if not valid:
raise SyntaxError('invalid expression: %s' % key)
if key in self.context:
result = self.context[key]
else:
result = self.allowed_values[key]
return result
def do_boolop(self, node):
result = self.evaluate(node.values[0])
is_or = node.op.__class__ is ast.Or
is_and = node.op.__class__ is ast.And
assert is_or or is_and
if (is_and and result) or (is_or and not result):
for n in node.values[1:]:
result = self.evaluate(n)
if (is_or and result) or (is_and and not result):
break
return result
def do_compare(self, node):
def sanity_check(lhsnode, rhsnode):
valid = True
if isinstance(lhsnode, ast.Str) and isinstance(rhsnode, ast.Str):
valid = False
#elif (isinstance(lhsnode, ast.Attribute)
# and isinstance(rhsnode, ast.Attribute)):
# klhs = self.get_attr_key(lhsnode)
# krhs = self.get_attr_key(rhsnode)
# valid = klhs != krhs
if not valid:
s = self.get_fragment(node.col_offset)
raise SyntaxError('Invalid comparison: %s' % s)
lhsnode = node.left
lhs = self.evaluate(lhsnode)
result = True
for op, rhsnode in zip(node.ops, node.comparators):
sanity_check(lhsnode, rhsnode)
op = op.__class__.__name__.lower()
if op not in self.operators:
raise SyntaxError('unsupported operation: %r' % op)
rhs = self.evaluate(rhsnode)
result = self.operators[op](lhs, rhs)
if not result:
break
lhs = rhs
lhsnode = rhsnode
return result
def do_expression(self, node):
return self.evaluate(node.body)
def do_name(self, node):
valid = False
if node.id in self.context:
valid = True
result = self.context[node.id]
elif node.id in self.allowed_values:
valid = True
result = self.allowed_values[node.id]
if not valid:
raise SyntaxError('invalid expression: %s' % node.id)
return result
def do_str(self, node):
return node.s
def interpret(marker, execution_context=None):
"""
Interpret a marker and return a result depending on environment.
:param marker: The marker to interpret.
:type marker: str
:param execution_context: The context used for name lookup.
:type execution_context: mapping
"""
return Evaluator(execution_context).evaluate(marker.strip())
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,317 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2013 Vinay Sajip.
# Licensed to the Python Software Foundation under a contributor agreement.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
from __future__ import unicode_literals
import bisect
import io
import logging
import os
import pkgutil
import shutil
import sys
import types
import zipimport
from . import DistlibException
from .util import cached_property, get_cache_base, path_to_cache_dir, Cache
logger = logging.getLogger(__name__)
cache = None # created when needed
class ResourceCache(Cache):
def __init__(self, base=None):
if base is None:
# Use native string to avoid issues on 2.x: see Python #20140.
base = os.path.join(get_cache_base(), str('resource-cache'))
super(ResourceCache, self).__init__(base)
def is_stale(self, resource, path):
"""
Is the cache stale for the given resource?
:param resource: The :class:`Resource` being cached.
:param path: The path of the resource in the cache.
:return: True if the cache is stale.
"""
# Cache invalidation is a hard problem :-)
return True
def get(self, resource):
"""
Get a resource into the cache,
:param resource: A :class:`Resource` instance.
:return: The pathname of the resource in the cache.
"""
prefix, path = resource.finder.get_cache_info(resource)
if prefix is None:
result = path
else:
result = os.path.join(self.base, self.prefix_to_dir(prefix), path)
dirname = os.path.dirname(result)
if not os.path.isdir(dirname):
os.makedirs(dirname)
if not os.path.exists(result):
stale = True
else:
stale = self.is_stale(resource, path)
if stale:
# write the bytes of the resource to the cache location
with open(result, 'wb') as f:
f.write(resource.bytes)
return result
class ResourceBase(object):
def __init__(self, finder, name):
self.finder = finder
self.name = name
class Resource(ResourceBase):
"""
A class representing an in-package resource, such as a data file. This is
not normally instantiated by user code, but rather by a
:class:`ResourceFinder` which manages the resource.
"""
is_container = False # Backwards compatibility
def as_stream(self):
"""
Get the resource as a stream.
This is not a property to make it obvious that it returns a new stream
each time.
"""
return self.finder.get_stream(self)
@cached_property
def file_path(self):
global cache
if cache is None:
cache = ResourceCache()
return cache.get(self)
@cached_property
def bytes(self):
return self.finder.get_bytes(self)
@cached_property
def size(self):
return self.finder.get_size(self)
class ResourceContainer(ResourceBase):
is_container = True # Backwards compatibility
@cached_property
def resources(self):
return self.finder.get_resources(self)
class ResourceFinder(object):
"""
Resource finder for file system resources.
"""
def __init__(self, module):
self.module = module
self.loader = getattr(module, '__loader__', None)
self.base = os.path.dirname(getattr(module, '__file__', ''))
def _adjust_path(self, path):
return os.path.realpath(path)
def _make_path(self, resource_name):
parts = resource_name.split('/')
parts.insert(0, self.base)
result = os.path.join(*parts)
return self._adjust_path(result)
def _find(self, path):
return os.path.exists(path)
def get_cache_info(self, resource):
return None, resource.path
def find(self, resource_name):
path = self._make_path(resource_name)
if not self._find(path):
result = None
else:
if self._is_directory(path):
result = ResourceContainer(self, resource_name)
else:
result = Resource(self, resource_name)
result.path = path
return result
def get_stream(self, resource):
return open(resource.path, 'rb')
def get_bytes(self, resource):
with open(resource.path, 'rb') as f:
return f.read()
def get_size(self, resource):
return os.path.getsize(resource.path)
def get_resources(self, resource):
def allowed(f):
return f != '__pycache__' and not f.endswith(('.pyc', '.pyo'))
return set([f for f in os.listdir(resource.path) if allowed(f)])
def is_container(self, resource):
return self._is_directory(resource.path)
_is_directory = staticmethod(os.path.isdir)
class ZipResourceFinder(ResourceFinder):
"""
Resource finder for resources in .zip files.
"""
def __init__(self, module):
super(ZipResourceFinder, self).__init__(module)
archive = self.loader.archive
self.prefix_len = 1 + len(archive)
# PyPy doesn't have a _files attr on zipimporter, and you can't set one
if hasattr(self.loader, '_files'):
self._files = self.loader._files
else:
self._files = zipimport._zip_directory_cache[archive]
self.index = sorted(self._files)
def _adjust_path(self, path):
return path
def _find(self, path):
path = path[self.prefix_len:]
if path in self._files:
result = True
else:
if path and path[-1] != os.sep:
path = path + os.sep
i = bisect.bisect(self.index, path)
try:
result = self.index[i].startswith(path)
except IndexError:
result = False
if not result:
logger.debug('_find failed: %r %r', path, self.loader.prefix)
else:
logger.debug('_find worked: %r %r', path, self.loader.prefix)
return result
def get_cache_info(self, resource):
prefix = self.loader.archive
path = resource.path[1 + len(prefix):]
return prefix, path
def get_bytes(self, resource):
return self.loader.get_data(resource.path)
def get_stream(self, resource):
return io.BytesIO(self.get_bytes(resource))
def get_size(self, resource):
path = resource.path[self.prefix_len:]
return self._files[path][3]
def get_resources(self, resource):
path = resource.path[self.prefix_len:]
if path and path[-1] != os.sep:
path += os.sep
plen = len(path)
result = set()
i = bisect.bisect(self.index, path)
while i < len(self.index):
if not self.index[i].startswith(path):
break
s = self.index[i][plen:]
result.add(s.split(os.sep, 1)[0]) # only immediate children
i += 1
return result
def _is_directory(self, path):
path = path[self.prefix_len:]
if path and path[-1] != os.sep:
path += os.sep
i = bisect.bisect(self.index, path)
try:
result = self.index[i].startswith(path)
except IndexError:
result = False
return result
_finder_registry = {
type(None): ResourceFinder,
zipimport.zipimporter: ZipResourceFinder
}
try:
import _frozen_importlib
_finder_registry[_frozen_importlib.SourceFileLoader] = ResourceFinder
_finder_registry[_frozen_importlib.FileFinder] = ResourceFinder
except (ImportError, AttributeError):
pass
def register_finder(loader, finder_maker):
_finder_registry[type(loader)] = finder_maker
_finder_cache = {}
def finder(package):
"""
Return a resource finder for a package.
:param package: The name of the package.
:return: A :class:`ResourceFinder` instance for the package.
"""
if package in _finder_cache:
result = _finder_cache[package]
else:
if package not in sys.modules:
__import__(package)
module = sys.modules[package]
path = getattr(module, '__path__', None)
if path is None:
raise DistlibException('You cannot get a finder for a module, '
'only for a package')
loader = getattr(module, '__loader__', None)
finder_maker = _finder_registry.get(type(loader))
if finder_maker is None:
raise DistlibException('Unable to locate finder for %r' % package)
result = finder_maker(module)
_finder_cache[package] = result
return result
_dummy_module = types.ModuleType(str('__dummy__'))
def finder_for_path(path):
"""
Return a resource finder for a path, which should represent a container.
:param path: The path.
:return: A :class:`ResourceFinder` instance for the path.
"""
result = None
# calls any path hooks, gets importer into cache
pkgutil.get_importer(path)
loader = sys.path_importer_cache.get(path)
finder = _finder_registry.get(type(loader))
if finder:
module = _dummy_module
module.__file__ = os.path.join(path, '')
module.__loader__ = loader
result = finder(module)
return result
@@ -0,0 +1,323 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2013 Vinay Sajip.
# Licensed to the Python Software Foundation under a contributor agreement.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
from io import BytesIO
import logging
import os
import re
import struct
import sys
from .compat import sysconfig, fsencode, detect_encoding, ZipFile
from .resources import finder
from .util import (FileOperator, get_export_entry, convert_path,
get_executable, in_venv)
logger = logging.getLogger(__name__)
_DEFAULT_MANIFEST = '''
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<assembly xmlns="urn:schemas-microsoft-com:asm.v1" manifestVersion="1.0">
<assemblyIdentity version="1.0.0.0"
processorArchitecture="X86"
name="%s"
type="win32"/>
<!-- Identify the application security requirements. -->
<trustInfo xmlns="urn:schemas-microsoft-com:asm.v3">
<security>
<requestedPrivileges>
<requestedExecutionLevel level="asInvoker" uiAccess="false"/>
</requestedPrivileges>
</security>
</trustInfo>
</assembly>'''.strip()
# check if Python is called on the first line with this expression
FIRST_LINE_RE = re.compile(b'^#!.*pythonw?[0-9.]*([ \t].*)?$')
SCRIPT_TEMPLATE = '''# -*- coding: utf-8 -*-
if __name__ == '__main__':
import sys, re
def _resolve(module, func):
__import__(module)
mod = sys.modules[module]
parts = func.split('.')
result = getattr(mod, parts.pop(0))
for p in parts:
result = getattr(result, p)
return result
try:
sys.argv[0] = re.sub(r'(-script\.pyw|\.exe)?$', '', sys.argv[0])
func = _resolve('%(module)s', '%(func)s')
rc = func() # None interpreted as 0
except Exception as e: # only supporting Python >= 2.6
sys.stderr.write('%%s\\n' %% e)
rc = 1
sys.exit(rc)
'''
class ScriptMaker(object):
"""
A class to copy or create scripts from source scripts or callable
specifications.
"""
script_template = SCRIPT_TEMPLATE
executable = None # for shebangs
def __init__(self, source_dir, target_dir, add_launchers=True,
dry_run=False, fileop=None):
self.source_dir = source_dir
self.target_dir = target_dir
self.add_launchers = add_launchers
self.force = False
self.clobber = False
# It only makes sense to set mode bits on POSIX.
self.set_mode = (os.name == 'posix')
self.variants = set(('', 'X.Y'))
self._fileop = fileop or FileOperator(dry_run)
def _get_alternate_executable(self, executable, options):
if options.get('gui', False) and os.name == 'nt':
dn, fn = os.path.split(executable)
fn = fn.replace('python', 'pythonw')
executable = os.path.join(dn, fn)
return executable
def _get_shebang(self, encoding, post_interp=b'', options=None):
enquote = True
if self.executable:
executable = self.executable
enquote = False # assume this will be taken care of
elif not sysconfig.is_python_build():
executable = get_executable()
elif in_venv():
executable = os.path.join(sysconfig.get_path('scripts'),
'python%s' % sysconfig.get_config_var('EXE'))
else:
executable = os.path.join(
sysconfig.get_config_var('BINDIR'),
'python%s%s' % (sysconfig.get_config_var('VERSION'),
sysconfig.get_config_var('EXE')))
if options:
executable = self._get_alternate_executable(executable, options)
# If the user didn't specify an executable, it may be necessary to
# cater for executable paths with spaces (not uncommon on Windows)
if enquote and ' ' in executable:
executable = '"%s"' % executable
executable = fsencode(executable)
shebang = b'#!' + executable + post_interp + b'\n'
# Python parser starts to read a script using UTF-8 until
# it gets a #coding:xxx cookie. The shebang has to be the
# first line of a file, the #coding:xxx cookie cannot be
# written before. So the shebang has to be decodable from
# UTF-8.
try:
shebang.decode('utf-8')
except UnicodeDecodeError:
raise ValueError(
'The shebang (%r) is not decodable from utf-8' % shebang)
# If the script is encoded to a custom encoding (use a
# #coding:xxx cookie), the shebang has to be decodable from
# the script encoding too.
if encoding != 'utf-8':
try:
shebang.decode(encoding)
except UnicodeDecodeError:
raise ValueError(
'The shebang (%r) is not decodable '
'from the script encoding (%r)' % (shebang, encoding))
return shebang
def _get_script_text(self, entry):
return self.script_template % dict(module=entry.prefix,
func=entry.suffix)
manifest = _DEFAULT_MANIFEST
def get_manifest(self, exename):
base = os.path.basename(exename)
return self.manifest % base
def _write_script(self, names, shebang, script_bytes, filenames, ext):
use_launcher = self.add_launchers and os.name == 'nt'
linesep = os.linesep.encode('utf-8')
if not use_launcher:
script_bytes = shebang + linesep + script_bytes
else:
if ext == 'py':
launcher = self._get_launcher('t')
else:
launcher = self._get_launcher('w')
stream = BytesIO()
with ZipFile(stream, 'w') as zf:
zf.writestr('__main__.py', script_bytes)
zip_data = stream.getvalue()
script_bytes = launcher + shebang + linesep + zip_data
for name in names:
outname = os.path.join(self.target_dir, name)
if use_launcher:
n, e = os.path.splitext(outname)
if e.startswith('.py'):
outname = n
outname = '%s.exe' % outname
try:
self._fileop.write_binary_file(outname, script_bytes)
except Exception:
# Failed writing an executable - it might be in use.
logger.warning('Failed to write executable - trying to '
'use .deleteme logic')
dfname = '%s.deleteme' % outname
if os.path.exists(dfname):
os.remove(dfname) # Not allowed to fail here
os.rename(outname, dfname) # nor here
self._fileop.write_binary_file(outname, script_bytes)
logger.debug('Able to replace executable using '
'.deleteme logic')
try:
os.remove(dfname)
except Exception:
pass # still in use - ignore error
else:
if os.name == 'nt' and not outname.endswith('.' + ext):
outname = '%s.%s' % (outname, ext)
if os.path.exists(outname) and not self.clobber:
logger.warning('Skipping existing file %s', outname)
continue
self._fileop.write_binary_file(outname, script_bytes)
if self.set_mode:
self._fileop.set_executable_mode([outname])
filenames.append(outname)
def _make_script(self, entry, filenames, options=None):
shebang = self._get_shebang('utf-8', options=options)
script = self._get_script_text(entry).encode('utf-8')
name = entry.name
scriptnames = set()
if '' in self.variants:
scriptnames.add(name)
if 'X' in self.variants:
scriptnames.add('%s%s' % (name, sys.version[0]))
if 'X.Y' in self.variants:
scriptnames.add('%s-%s' % (name, sys.version[:3]))
if options and options.get('gui', False):
ext = 'pyw'
else:
ext = 'py'
self._write_script(scriptnames, shebang, script, filenames, ext)
def _copy_script(self, script, filenames):
adjust = False
script = os.path.join(self.source_dir, convert_path(script))
outname = os.path.join(self.target_dir, os.path.basename(script))
if not self.force and not self._fileop.newer(script, outname):
logger.debug('not copying %s (up-to-date)', script)
return
# Always open the file, but ignore failures in dry-run mode --
# that way, we'll get accurate feedback if we can read the
# script.
try:
f = open(script, 'rb')
except IOError:
if not self.dry_run:
raise
f = None
else:
encoding, lines = detect_encoding(f.readline)
f.seek(0)
first_line = f.readline()
if not first_line:
logger.warning('%s: %s is an empty file (skipping)',
self.get_command_name(), script)
return
match = FIRST_LINE_RE.match(first_line.replace(b'\r\n', b'\n'))
if match:
adjust = True
post_interp = match.group(1) or b''
if not adjust:
if f:
f.close()
self._fileop.copy_file(script, outname)
if self.set_mode:
self._fileop.set_executable_mode([outname])
filenames.append(outname)
else:
logger.info('copying and adjusting %s -> %s', script,
self.target_dir)
if not self._fileop.dry_run:
shebang = self._get_shebang(encoding, post_interp)
if b'pythonw' in first_line:
ext = 'pyw'
else:
ext = 'py'
n = os.path.basename(outname)
self._write_script([n], shebang, f.read(), filenames, ext)
if f:
f.close()
@property
def dry_run(self):
return self._fileop.dry_run
@dry_run.setter
def dry_run(self, value):
self._fileop.dry_run = value
if os.name == 'nt':
# Executable launcher support.
# Launchers are from https://bitbucket.org/vinay.sajip/simple_launcher/
def _get_launcher(self, kind):
if struct.calcsize('P') == 8: # 64-bit
bits = '64'
else:
bits = '32'
name = '%s%s.exe' % (kind, bits)
# Issue 31: don't hardcode an absolute package name, but
# determine it relative to the current package
distlib_package = __name__.rsplit('.', 1)[0]
result = finder(distlib_package).find(name).bytes
return result
# Public API follows
def make(self, specification, options=None):
"""
Make a script.
:param specification: The specification, which is either a valid export
entry specification (to make a script from a
callable) or a filename (to make a script by
copying from a source location).
:param options: A dictionary of options controlling script generation.
:return: A list of all absolute pathnames written to.
"""
filenames = []
entry = get_export_entry(specification)
if entry is None:
self._copy_script(specification, filenames)
else:
self._make_script(entry, filenames, options=options)
return filenames
def make_multiple(self, specifications, options=None):
"""
Take a list of specifications and make scripts from them,
:param specifications: A list of specifications.
:return: A list of all absolute pathnames written to,
"""
filenames = []
for specification in specifications:
filenames.extend(self.make(specification, options))
return filenames
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,721 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2012-2013 The Python Software Foundation.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
"""
Implementation of a flexible versioning scheme providing support for PEP-386,
distribute-compatible and semantic versioning.
"""
import logging
import re
from .compat import string_types
__all__ = ['NormalizedVersion', 'NormalizedMatcher',
'LegacyVersion', 'LegacyMatcher',
'SemanticVersion', 'SemanticMatcher',
'UnsupportedVersionError', 'get_scheme']
logger = logging.getLogger(__name__)
class UnsupportedVersionError(ValueError):
"""This is an unsupported version."""
pass
class Version(object):
def __init__(self, s):
self._string = s = s.strip()
self._parts = parts = self.parse(s)
assert isinstance(parts, tuple)
assert len(parts) > 0
def parse(self, s):
raise NotImplementedError('please implement in a subclass')
def _check_compatible(self, other):
if type(self) != type(other):
raise TypeError('cannot compare %r and %r' % (self, other))
def __eq__(self, other):
self._check_compatible(other)
return self._parts == other._parts
def __ne__(self, other):
return not self.__eq__(other)
def __lt__(self, other):
self._check_compatible(other)
return self._parts < other._parts
def __gt__(self, other):
return not (self.__lt__(other) or self.__eq__(other))
def __le__(self, other):
return self.__lt__(other) or self.__eq__(other)
def __ge__(self, other):
return self.__gt__(other) or self.__eq__(other)
# See http://docs.python.org/reference/datamodel#object.__hash__
def __hash__(self):
return hash(self._parts)
def __repr__(self):
return "%s('%s')" % (self.__class__.__name__, self._string)
def __str__(self):
return self._string
@property
def is_prerelease(self):
raise NotImplementedError('Please implement in subclasses.')
class Matcher(object):
version_class = None
dist_re = re.compile(r"^(\w[\s\w'.-]*)(\((.*)\))?")
comp_re = re.compile(r'^(<=|>=|<|>|!=|==|~=)?\s*([^\s,]+)$')
num_re = re.compile(r'^\d+(\.\d+)*$')
# value is either a callable or the name of a method
_operators = {
'<': lambda v, c, p: v < c,
'>': lambda v, c, p: v > c,
'<=': lambda v, c, p: v == c or v < c,
'>=': lambda v, c, p: v == c or v > c,
'==': lambda v, c, p: v == c,
# by default, compatible => >=.
'~=': lambda v, c, p: v == c or v > c,
'!=': lambda v, c, p: v != c,
}
def __init__(self, s):
if self.version_class is None:
raise ValueError('Please specify a version class')
self._string = s = s.strip()
m = self.dist_re.match(s)
if not m:
raise ValueError('Not valid: %r' % s)
groups = m.groups('')
self.name = groups[0].strip()
self.key = self.name.lower() # for case-insensitive comparisons
clist = []
if groups[2]:
constraints = [c.strip() for c in groups[2].split(',')]
for c in constraints:
m = self.comp_re.match(c)
if not m:
raise ValueError('Invalid %r in %r' % (c, s))
groups = m.groups()
op = groups[0] or '~='
s = groups[1]
if s.endswith('.*'):
if op not in ('==', '!='):
raise ValueError('\'.*\' not allowed for '
'%r constraints' % op)
# Could be a partial version (e.g. for '2.*') which
# won't parse as a version, so keep it as a string
vn, prefix = s[:-2], True
if not self.num_re.match(vn):
# Just to check that vn is a valid version
self.version_class(vn)
else:
# Should parse as a version, so we can create an
# instance for the comparison
vn, prefix = self.version_class(s), False
clist.append((op, vn, prefix))
self._parts = tuple(clist)
def match(self, version):
"""
Check if the provided version matches the constraints.
:param version: The version to match against this instance.
:type version: Strring or :class:`Version` instance.
"""
if isinstance(version, string_types):
version = self.version_class(version)
for operator, constraint, prefix in self._parts:
f = self._operators.get(operator)
if isinstance(f, string_types):
f = getattr(self, f)
if not f:
msg = ('%r not implemented '
'for %s' % (operator, self.__class__.__name__))
raise NotImplementedError(msg)
if not f(version, constraint, prefix):
return False
return True
@property
def exact_version(self):
result = None
if len(self._parts) == 1 and self._parts[0][0] == '==':
result = self._parts[0][1]
return result
def _check_compatible(self, other):
if type(self) != type(other) or self.name != other.name:
raise TypeError('cannot compare %s and %s' % (self, other))
def __eq__(self, other):
self._check_compatible(other)
return self.key == other.key and self._parts == other._parts
def __ne__(self, other):
return not self.__eq__(other)
# See http://docs.python.org/reference/datamodel#object.__hash__
def __hash__(self):
return hash(self.key) + hash(self._parts)
def __repr__(self):
return "%s(%r)" % (self.__class__.__name__, self._string)
def __str__(self):
return self._string
PEP426_VERSION_RE = re.compile(r'^(\d+(\.\d+)*)((a|b|c|rc)(\d+))?'
r'(\.(post)(\d+))?(\.(dev)(\d+))?'
r'(-(\d+(\.\d+)?))?$')
def _pep426_key(s):
s = s.strip()
m = PEP426_VERSION_RE.match(s)
if not m:
raise UnsupportedVersionError('Not a valid version: %s' % s)
groups = m.groups()
nums = tuple(int(v) for v in groups[0].split('.'))
while len(nums) > 1 and nums[-1] == 0:
nums = nums[:-1]
pre = groups[3:5]
post = groups[6:8]
dev = groups[9:11]
local = groups[12]
if pre == (None, None):
pre = ()
else:
pre = pre[0], int(pre[1])
if post == (None, None):
post = ()
else:
post = post[0], int(post[1])
if dev == (None, None):
dev = ()
else:
dev = dev[0], int(dev[1])
if local is None:
local = ()
else:
local = tuple([int(s) for s in local.split('.')])
if not pre:
# either before pre-release, or final release and after
if not post and dev:
# before pre-release
pre = ('a', -1) # to sort before a0
else:
pre = ('z',) # to sort after all pre-releases
# now look at the state of post and dev.
if not post:
post = ('_',) # sort before 'a'
if not dev:
dev = ('final',)
#print('%s -> %s' % (s, m.groups()))
return nums, pre, post, dev, local
_normalized_key = _pep426_key
class NormalizedVersion(Version):
"""A rational version.
Good:
1.2 # equivalent to "1.2.0"
1.2.0
1.2a1
1.2.3a2
1.2.3b1
1.2.3c1
1.2.3.4
TODO: fill this out
Bad:
1 # mininum two numbers
1.2a # release level must have a release serial
1.2.3b
"""
def parse(self, s):
result = _normalized_key(s)
# _normalized_key loses trailing zeroes in the release
# clause, since that's needed to ensure that X.Y == X.Y.0 == X.Y.0.0
# However, PEP 440 prefix matching needs it: for example,
# (~= 1.4.5.0) matches differently to (~= 1.4.5.0.0).
m = PEP426_VERSION_RE.match(s) # must succeed
groups = m.groups()
self._release_clause = tuple(int(v) for v in groups[0].split('.'))
return result
PREREL_TAGS = set(['a', 'b', 'c', 'rc', 'dev'])
@property
def is_prerelease(self):
return any(t[0] in self.PREREL_TAGS for t in self._parts if t)
def _match_prefix(x, y):
x = str(x)
y = str(y)
if x == y:
return True
if not x.startswith(y):
return False
n = len(y)
return x[n] == '.'
class NormalizedMatcher(Matcher):
version_class = NormalizedVersion
# value is either a callable or the name of a method
_operators = {
'~=': '_match_compatible',
'<': '_match_lt',
'>': '_match_gt',
'<=': '_match_le',
'>=': '_match_ge',
'==': '_match_eq',
'!=': '_match_ne',
}
def _adjust_local(self, version, constraint, prefix):
if prefix:
strip_local = '-' not in constraint and version._parts[-1]
else:
# both constraint and version are
# NormalizedVersion instances.
# If constraint does not have a local component,
# ensure the version doesn't, either.
strip_local = not constraint._parts[-1] and version._parts[-1]
if strip_local:
s = version._string.split('-', 1)[0]
version = self.version_class(s)
return version, constraint
def _match_lt(self, version, constraint, prefix):
version, constraint = self._adjust_local(version, constraint, prefix)
if version >= constraint:
return False
release_clause = constraint._release_clause
pfx = '.'.join([str(i) for i in release_clause])
return not _match_prefix(version, pfx)
def _match_gt(self, version, constraint, prefix):
version, constraint = self._adjust_local(version, constraint, prefix)
if version <= constraint:
return False
release_clause = constraint._release_clause
pfx = '.'.join([str(i) for i in release_clause])
return not _match_prefix(version, pfx)
def _match_le(self, version, constraint, prefix):
version, constraint = self._adjust_local(version, constraint, prefix)
return version <= constraint
def _match_ge(self, version, constraint, prefix):
version, constraint = self._adjust_local(version, constraint, prefix)
return version >= constraint
def _match_eq(self, version, constraint, prefix):
version, constraint = self._adjust_local(version, constraint, prefix)
if not prefix:
result = (version == constraint)
else:
result = _match_prefix(version, constraint)
return result
def _match_ne(self, version, constraint, prefix):
version, constraint = self._adjust_local(version, constraint, prefix)
if not prefix:
result = (version != constraint)
else:
result = not _match_prefix(version, constraint)
return result
def _match_compatible(self, version, constraint, prefix):
version, constraint = self._adjust_local(version, constraint, prefix)
if version == constraint:
return True
if version < constraint:
return False
release_clause = constraint._release_clause
if len(release_clause) > 1:
release_clause = release_clause[:-1]
pfx = '.'.join([str(i) for i in release_clause])
return _match_prefix(version, pfx)
_REPLACEMENTS = (
(re.compile('[.+-]$'), ''), # remove trailing puncts
(re.compile(r'^[.](\d)'), r'0.\1'), # .N -> 0.N at start
(re.compile('^[.-]'), ''), # remove leading puncts
(re.compile(r'^\((.*)\)$'), r'\1'), # remove parentheses
(re.compile(r'^v(ersion)?\s*(\d+)'), r'\2'), # remove leading v(ersion)
(re.compile(r'^r(ev)?\s*(\d+)'), r'\2'), # remove leading v(ersion)
(re.compile('[.]{2,}'), '.'), # multiple runs of '.'
(re.compile(r'\b(alfa|apha)\b'), 'alpha'), # misspelt alpha
(re.compile(r'\b(pre-alpha|prealpha)\b'),
'pre.alpha'), # standardise
(re.compile(r'\(beta\)$'), 'beta'), # remove parentheses
)
_SUFFIX_REPLACEMENTS = (
(re.compile('^[:~._+-]+'), ''), # remove leading puncts
(re.compile('[,*")([\]]'), ''), # remove unwanted chars
(re.compile('[~:+_ -]'), '.'), # replace illegal chars
(re.compile('[.]{2,}'), '.'), # multiple runs of '.'
(re.compile(r'\.$'), ''), # trailing '.'
)
_NUMERIC_PREFIX = re.compile(r'(\d+(\.\d+)*)')
def _suggest_semantic_version(s):
"""
Try to suggest a semantic form for a version for which
_suggest_normalized_version couldn't come up with anything.
"""
result = s.strip().lower()
for pat, repl in _REPLACEMENTS:
result = pat.sub(repl, result)
if not result:
result = '0.0.0'
# Now look for numeric prefix, and separate it out from
# the rest.
#import pdb; pdb.set_trace()
m = _NUMERIC_PREFIX.match(result)
if not m:
prefix = '0.0.0'
suffix = result
else:
prefix = m.groups()[0].split('.')
prefix = [int(i) for i in prefix]
while len(prefix) < 3:
prefix.append(0)
if len(prefix) == 3:
suffix = result[m.end():]
else:
suffix = '.'.join([str(i) for i in prefix[3:]]) + result[m.end():]
prefix = prefix[:3]
prefix = '.'.join([str(i) for i in prefix])
suffix = suffix.strip()
if suffix:
#import pdb; pdb.set_trace()
# massage the suffix.
for pat, repl in _SUFFIX_REPLACEMENTS:
suffix = pat.sub(repl, suffix)
if not suffix:
result = prefix
else:
sep = '-' if 'dev' in suffix else '+'
result = prefix + sep + suffix
if not is_semver(result):
result = None
return result
def _suggest_normalized_version(s):
"""Suggest a normalized version close to the given version string.
If you have a version string that isn't rational (i.e. NormalizedVersion
doesn't like it) then you might be able to get an equivalent (or close)
rational version from this function.
This does a number of simple normalizations to the given string, based
on observation of versions currently in use on PyPI. Given a dump of
those version during PyCon 2009, 4287 of them:
- 2312 (53.93%) match NormalizedVersion without change
with the automatic suggestion
- 3474 (81.04%) match when using this suggestion method
@param s {str} An irrational version string.
@returns A rational version string, or None, if couldn't determine one.
"""
try:
_normalized_key(s)
return s # already rational
except UnsupportedVersionError:
pass
rs = s.lower()
# part of this could use maketrans
for orig, repl in (('-alpha', 'a'), ('-beta', 'b'), ('alpha', 'a'),
('beta', 'b'), ('rc', 'c'), ('-final', ''),
('-pre', 'c'),
('-release', ''), ('.release', ''), ('-stable', ''),
('+', '.'), ('_', '.'), (' ', ''), ('.final', ''),
('final', '')):
rs = rs.replace(orig, repl)
# if something ends with dev or pre, we add a 0
rs = re.sub(r"pre$", r"pre0", rs)
rs = re.sub(r"dev$", r"dev0", rs)
# if we have something like "b-2" or "a.2" at the end of the
# version, that is pobably beta, alpha, etc
# let's remove the dash or dot
rs = re.sub(r"([abc]|rc)[\-\.](\d+)$", r"\1\2", rs)
# 1.0-dev-r371 -> 1.0.dev371
# 0.1-dev-r79 -> 0.1.dev79
rs = re.sub(r"[\-\.](dev)[\-\.]?r?(\d+)$", r".\1\2", rs)
# Clean: 2.0.a.3, 2.0.b1, 0.9.0~c1
rs = re.sub(r"[.~]?([abc])\.?", r"\1", rs)
# Clean: v0.3, v1.0
if rs.startswith('v'):
rs = rs[1:]
# Clean leading '0's on numbers.
#TODO: unintended side-effect on, e.g., "2003.05.09"
# PyPI stats: 77 (~2%) better
rs = re.sub(r"\b0+(\d+)(?!\d)", r"\1", rs)
# Clean a/b/c with no version. E.g. "1.0a" -> "1.0a0". Setuptools infers
# zero.
# PyPI stats: 245 (7.56%) better
rs = re.sub(r"(\d+[abc])$", r"\g<1>0", rs)
# the 'dev-rNNN' tag is a dev tag
rs = re.sub(r"\.?(dev-r|dev\.r)\.?(\d+)$", r".dev\2", rs)
# clean the - when used as a pre delimiter
rs = re.sub(r"-(a|b|c)(\d+)$", r"\1\2", rs)
# a terminal "dev" or "devel" can be changed into ".dev0"
rs = re.sub(r"[\.\-](dev|devel)$", r".dev0", rs)
# a terminal "dev" can be changed into ".dev0"
rs = re.sub(r"(?![\.\-])dev$", r".dev0", rs)
# a terminal "final" or "stable" can be removed
rs = re.sub(r"(final|stable)$", "", rs)
# The 'r' and the '-' tags are post release tags
# 0.4a1.r10 -> 0.4a1.post10
# 0.9.33-17222 -> 0.9.33.post17222
# 0.9.33-r17222 -> 0.9.33.post17222
rs = re.sub(r"\.?(r|-|-r)\.?(\d+)$", r".post\2", rs)
# Clean 'r' instead of 'dev' usage:
# 0.9.33+r17222 -> 0.9.33.dev17222
# 1.0dev123 -> 1.0.dev123
# 1.0.git123 -> 1.0.dev123
# 1.0.bzr123 -> 1.0.dev123
# 0.1a0dev.123 -> 0.1a0.dev123
# PyPI stats: ~150 (~4%) better
rs = re.sub(r"\.?(dev|git|bzr)\.?(\d+)$", r".dev\2", rs)
# Clean '.pre' (normalized from '-pre' above) instead of 'c' usage:
# 0.2.pre1 -> 0.2c1
# 0.2-c1 -> 0.2c1
# 1.0preview123 -> 1.0c123
# PyPI stats: ~21 (0.62%) better
rs = re.sub(r"\.?(pre|preview|-c)(\d+)$", r"c\g<2>", rs)
# Tcl/Tk uses "px" for their post release markers
rs = re.sub(r"p(\d+)$", r".post\1", rs)
try:
_normalized_key(rs)
except UnsupportedVersionError:
rs = None
return rs
#
# Legacy version processing (distribute-compatible)
#
_VERSION_PART = re.compile(r'([a-z]+|\d+|[\.-])', re.I)
_VERSION_REPLACE = {
'pre': 'c',
'preview': 'c',
'-': 'final-',
'rc': 'c',
'dev': '@',
'': None,
'.': None,
}
def _legacy_key(s):
def get_parts(s):
result = []
for p in _VERSION_PART.split(s.lower()):
p = _VERSION_REPLACE.get(p, p)
if p:
if '0' <= p[:1] <= '9':
p = p.zfill(8)
else:
p = '*' + p
result.append(p)
result.append('*final')
return result
result = []
for p in get_parts(s):
if p.startswith('*'):
if p < '*final':
while result and result[-1] == '*final-':
result.pop()
while result and result[-1] == '00000000':
result.pop()
result.append(p)
return tuple(result)
class LegacyVersion(Version):
def parse(self, s):
return _legacy_key(s)
@property
def is_prerelease(self):
result = False
for x in self._parts:
if (isinstance(x, string_types) and x.startswith('*') and
x < '*final'):
result = True
break
return result
class LegacyMatcher(Matcher):
version_class = LegacyVersion
_operators = dict(Matcher._operators)
_operators['~='] = '_match_compatible'
numeric_re = re.compile('^(\d+(\.\d+)*)')
def _match_compatible(self, version, constraint, prefix):
if version < constraint:
return False
m = self.numeric_re.match(str(constraint))
if not m:
logger.warning('Cannot compute compatible match for version %s '
' and constraint %s', version, constraint)
return True
s = m.groups()[0]
if '.' in s:
s = s.rsplit('.', 1)[0]
return _match_prefix(version, s)
#
# Semantic versioning
#
_SEMVER_RE = re.compile(r'^(\d+)\.(\d+)\.(\d+)'
r'(-[a-z0-9]+(\.[a-z0-9-]+)*)?'
r'(\+[a-z0-9]+(\.[a-z0-9-]+)*)?$', re.I)
def is_semver(s):
return _SEMVER_RE.match(s)
def _semantic_key(s):
def make_tuple(s, absent):
if s is None:
result = (absent,)
else:
parts = s[1:].split('.')
# We can't compare ints and strings on Python 3, so fudge it
# by zero-filling numeric values so simulate a numeric comparison
result = tuple([p.zfill(8) if p.isdigit() else p for p in parts])
return result
m = is_semver(s)
if not m:
raise UnsupportedVersionError(s)
groups = m.groups()
major, minor, patch = [int(i) for i in groups[:3]]
# choose the '|' and '*' so that versions sort correctly
pre, build = make_tuple(groups[3], '|'), make_tuple(groups[5], '*')
return (major, minor, patch), pre, build
class SemanticVersion(Version):
def parse(self, s):
return _semantic_key(s)
@property
def is_prerelease(self):
return self._parts[1][0] != '|'
class SemanticMatcher(Matcher):
version_class = SemanticVersion
class VersionScheme(object):
def __init__(self, key, matcher, suggester=None):
self.key = key
self.matcher = matcher
self.suggester = suggester
def is_valid_version(self, s):
try:
self.matcher.version_class(s)
result = True
except UnsupportedVersionError:
result = False
return result
def is_valid_matcher(self, s):
try:
self.matcher(s)
result = True
except UnsupportedVersionError:
result = False
return result
def is_valid_constraint_list(self, s):
"""
Used for processing some metadata fields
"""
return self.is_valid_matcher('dummy_name (%s)' % s)
def suggest(self, s):
if self.suggester is None:
result = None
else:
result = self.suggester(s)
return result
_SCHEMES = {
'normalized': VersionScheme(_normalized_key, NormalizedMatcher,
_suggest_normalized_version),
'legacy': VersionScheme(_legacy_key, LegacyMatcher, lambda self, s: s),
'semantic': VersionScheme(_semantic_key, SemanticMatcher,
_suggest_semantic_version),
}
_SCHEMES['default'] = _SCHEMES['normalized']
def get_scheme(name):
if name not in _SCHEMES:
raise ValueError('unknown scheme name: %r' % name)
return _SCHEMES[name]
@@ -0,0 +1,958 @@
# -*- coding: utf-8 -*-
#
# Copyright (C) 2013-2014 Vinay Sajip.
# Licensed to the Python Software Foundation under a contributor agreement.
# See LICENSE.txt and CONTRIBUTORS.txt.
#
from __future__ import unicode_literals
import base64
import codecs
import datetime
import distutils.util
from email import message_from_file
import hashlib
import imp
import json
import logging
import os
import posixpath
import re
import shutil
import sys
import tempfile
import zipfile
from . import __version__, DistlibException
from .compat import sysconfig, ZipFile, fsdecode, text_type, filter
from .database import InstalledDistribution
from .metadata import Metadata, METADATA_FILENAME
from .util import (FileOperator, convert_path, CSVReader, CSVWriter, Cache,
cached_property, get_cache_base, read_exports, tempdir)
from .version import NormalizedVersion, UnsupportedVersionError
logger = logging.getLogger(__name__)
cache = None # created when needed
if hasattr(sys, 'pypy_version_info'):
IMP_PREFIX = 'pp'
elif sys.platform.startswith('java'):
IMP_PREFIX = 'jy'
elif sys.platform == 'cli':
IMP_PREFIX = 'ip'
else:
IMP_PREFIX = 'cp'
VER_SUFFIX = sysconfig.get_config_var('py_version_nodot')
if not VER_SUFFIX: # pragma: no cover
VER_SUFFIX = '%s%s' % sys.version_info[:2]
PYVER = 'py' + VER_SUFFIX
IMPVER = IMP_PREFIX + VER_SUFFIX
ARCH = distutils.util.get_platform().replace('-', '_').replace('.', '_')
ABI = sysconfig.get_config_var('SOABI')
if ABI and ABI.startswith('cpython-'):
ABI = ABI.replace('cpython-', 'cp')
else:
def _derive_abi():
parts = ['cp', VER_SUFFIX]
if sysconfig.get_config_var('Py_DEBUG'):
parts.append('d')
if sysconfig.get_config_var('WITH_PYMALLOC'):
parts.append('m')
if sysconfig.get_config_var('Py_UNICODE_SIZE') == 4:
parts.append('u')
return ''.join(parts)
ABI = _derive_abi()
del _derive_abi
FILENAME_RE = re.compile(r'''
(?P<nm>[^-]+)
-(?P<vn>\d+[^-]*)
(-(?P<bn>\d+[^-]*))?
-(?P<py>\w+\d+(\.\w+\d+)*)
-(?P<bi>\w+)
-(?P<ar>\w+)
\.whl$
''', re.IGNORECASE | re.VERBOSE)
NAME_VERSION_RE = re.compile(r'''
(?P<nm>[^-]+)
-(?P<vn>\d+[^-]*)
(-(?P<bn>\d+[^-]*))?$
''', re.IGNORECASE | re.VERBOSE)
SHEBANG_RE = re.compile(br'\s*#![^\r\n]*')
if os.sep == '/':
to_posix = lambda o: o
else:
to_posix = lambda o: o.replace(os.sep, '/')
class Mounter(object):
def __init__(self):
self.impure_wheels = {}
self.libs = {}
def add(self, pathname, extensions):
self.impure_wheels[pathname] = extensions
self.libs.update(extensions)
def remove(self, pathname):
extensions = self.impure_wheels.pop(pathname)
for k, v in extensions:
if k in self.libs:
del self.libs[k]
def find_module(self, fullname, path=None):
if fullname in self.libs:
result = self
else:
result = None
return result
def load_module(self, fullname):
if fullname in sys.modules:
result = sys.modules[fullname]
else:
if fullname not in self.libs:
raise ImportError('unable to find extension for %s' % fullname)
result = imp.load_dynamic(fullname, self.libs[fullname])
result.__loader__ = self
parts = fullname.rsplit('.', 1)
if len(parts) > 1:
result.__package__ = parts[0]
return result
_hook = Mounter()
class Wheel(object):
"""
Class to build and install from Wheel files (PEP 427).
"""
wheel_version = (1, 1)
hash_kind = 'sha256'
def __init__(self, filename=None, sign=False, verify=False):
"""
Initialise an instance using a (valid) filename.
"""
self.sign = sign
self.should_verify = verify
self.buildver = ''
self.pyver = [PYVER]
self.abi = ['none']
self.arch = ['any']
self.dirname = os.getcwd()
if filename is None:
self.name = 'dummy'
self.version = '0.1'
self._filename = self.filename
else:
m = NAME_VERSION_RE.match(filename)
if m:
info = m.groupdict('')
self.name = info['nm']
# Reinstate the local version separator
self.version = info['vn'].replace('_', '-')
self.buildver = info['bn']
self._filename = self.filename
else:
dirname, filename = os.path.split(filename)
m = FILENAME_RE.match(filename)
if not m:
raise DistlibException('Invalid name or '
'filename: %r' % filename)
if dirname:
self.dirname = os.path.abspath(dirname)
self._filename = filename
info = m.groupdict('')
self.name = info['nm']
self.version = info['vn']
self.buildver = info['bn']
self.pyver = info['py'].split('.')
self.abi = info['bi'].split('.')
self.arch = info['ar'].split('.')
@property
def filename(self):
"""
Build and return a filename from the various components.
"""
if self.buildver:
buildver = '-' + self.buildver
else:
buildver = ''
pyver = '.'.join(self.pyver)
abi = '.'.join(self.abi)
arch = '.'.join(self.arch)
# replace - with _ as a local version separator
version = self.version.replace('-', '_')
return '%s-%s%s-%s-%s-%s.whl' % (self.name, version, buildver,
pyver, abi, arch)
@property
def exists(self):
path = os.path.join(self.dirname, self.filename)
return os.path.isfile(path)
@property
def tags(self):
for pyver in self.pyver:
for abi in self.abi:
for arch in self.arch:
yield pyver, abi, arch
@cached_property
def metadata(self):
pathname = os.path.join(self.dirname, self.filename)
name_ver = '%s-%s' % (self.name, self.version)
info_dir = '%s.dist-info' % name_ver
wrapper = codecs.getreader('utf-8')
with ZipFile(pathname, 'r') as zf:
wheel_metadata = self.get_wheel_metadata(zf)
wv = wheel_metadata['Wheel-Version'].split('.', 1)
file_version = tuple([int(i) for i in wv])
if file_version < (1, 1):
fn = 'METADATA'
else:
fn = METADATA_FILENAME
try:
metadata_filename = posixpath.join(info_dir, fn)
with zf.open(metadata_filename) as bf:
wf = wrapper(bf)
result = Metadata(fileobj=wf)
except KeyError:
raise ValueError('Invalid wheel, because %s is '
'missing' % fn)
return result
def get_wheel_metadata(self, zf):
name_ver = '%s-%s' % (self.name, self.version)
info_dir = '%s.dist-info' % name_ver
metadata_filename = posixpath.join(info_dir, 'WHEEL')
with zf.open(metadata_filename) as bf:
wf = codecs.getreader('utf-8')(bf)
message = message_from_file(wf)
return dict(message)
@cached_property
def info(self):
pathname = os.path.join(self.dirname, self.filename)
with ZipFile(pathname, 'r') as zf:
result = self.get_wheel_metadata(zf)
return result
def process_shebang(self, data):
m = SHEBANG_RE.match(data)
if m:
data = b'#!python' + data[m.end():]
else:
cr = data.find(b'\r')
lf = data.find(b'\n')
if cr < 0 or cr > lf:
term = b'\n'
else:
if data[cr:cr + 2] == b'\r\n':
term = b'\r\n'
else:
term = b'\r'
data = b'#!python' + term + data
return data
def get_hash(self, data, hash_kind=None):
if hash_kind is None:
hash_kind = self.hash_kind
try:
hasher = getattr(hashlib, hash_kind)
except AttributeError:
raise DistlibException('Unsupported hash algorithm: %r' % hash_kind)
result = hasher(data).digest()
result = base64.urlsafe_b64encode(result).rstrip(b'=').decode('ascii')
return hash_kind, result
def write_record(self, records, record_path, base):
with CSVWriter(record_path) as writer:
for row in records:
writer.writerow(row)
p = to_posix(os.path.relpath(record_path, base))
writer.writerow((p, '', ''))
def write_records(self, info, libdir, archive_paths):
records = []
distinfo, info_dir = info
hasher = getattr(hashlib, self.hash_kind)
for ap, p in archive_paths:
with open(p, 'rb') as f:
data = f.read()
digest = '%s=%s' % self.get_hash(data)
size = os.path.getsize(p)
records.append((ap, digest, size))
p = os.path.join(distinfo, 'RECORD')
self.write_record(records, p, libdir)
ap = to_posix(os.path.join(info_dir, 'RECORD'))
archive_paths.append((ap, p))
def build_zip(self, pathname, archive_paths):
with ZipFile(pathname, 'w', zipfile.ZIP_DEFLATED) as zf:
for ap, p in archive_paths:
logger.debug('Wrote %s to %s in wheel', p, ap)
zf.write(p, ap)
def build(self, paths, tags=None, wheel_version=None):
"""
Build a wheel from files in specified paths, and use any specified tags
when determining the name of the wheel.
"""
if tags is None:
tags = {}
libkey = list(filter(lambda o: o in paths, ('purelib', 'platlib')))[0]
if libkey == 'platlib':
is_pure = 'false'
default_pyver = [IMPVER]
default_abi = [ABI]
default_arch = [ARCH]
else:
is_pure = 'true'
default_pyver = [PYVER]
default_abi = ['none']
default_arch = ['any']
self.pyver = tags.get('pyver', default_pyver)
self.abi = tags.get('abi', default_abi)
self.arch = tags.get('arch', default_arch)
libdir = paths[libkey]
name_ver = '%s-%s' % (self.name, self.version)
data_dir = '%s.data' % name_ver
info_dir = '%s.dist-info' % name_ver
archive_paths = []
# First, stuff which is not in site-packages
for key in ('data', 'headers', 'scripts'):
if key not in paths:
continue
path = paths[key]
if os.path.isdir(path):
for root, dirs, files in os.walk(path):
for fn in files:
p = fsdecode(os.path.join(root, fn))
rp = os.path.relpath(p, path)
ap = to_posix(os.path.join(data_dir, key, rp))
archive_paths.append((ap, p))
if key == 'scripts' and not p.endswith('.exe'):
with open(p, 'rb') as f:
data = f.read()
data = self.process_shebang(data)
with open(p, 'wb') as f:
f.write(data)
# Now, stuff which is in site-packages, other than the
# distinfo stuff.
path = libdir
distinfo = None
for root, dirs, files in os.walk(path):
if root == path:
# At the top level only, save distinfo for later
# and skip it for now
for i, dn in enumerate(dirs):
dn = fsdecode(dn)
if dn.endswith('.dist-info'):
distinfo = os.path.join(root, dn)
del dirs[i]
break
assert distinfo, '.dist-info directory expected, not found'
for fn in files:
# comment out next suite to leave .pyc files in
if fsdecode(fn).endswith(('.pyc', '.pyo')):
continue
p = os.path.join(root, fn)
rp = to_posix(os.path.relpath(p, path))
archive_paths.append((rp, p))
# Now distinfo. Assumed to be flat, i.e. os.listdir is enough.
files = os.listdir(distinfo)
for fn in files:
if fn not in ('RECORD', 'INSTALLER', 'SHARED'):
p = fsdecode(os.path.join(distinfo, fn))
ap = to_posix(os.path.join(info_dir, fn))
archive_paths.append((ap, p))
wheel_metadata = [
'Wheel-Version: %d.%d' % (wheel_version or self.wheel_version),
'Generator: distlib %s' % __version__,
'Root-Is-Purelib: %s' % is_pure,
]
for pyver, abi, arch in self.tags:
wheel_metadata.append('Tag: %s-%s-%s' % (pyver, abi, arch))
p = os.path.join(distinfo, 'WHEEL')
with open(p, 'w') as f:
f.write('\n'.join(wheel_metadata))
ap = to_posix(os.path.join(info_dir, 'WHEEL'))
archive_paths.append((ap, p))
# Now, at last, RECORD.
# Paths in here are archive paths - nothing else makes sense.
self.write_records((distinfo, info_dir), libdir, archive_paths)
# Now, ready to build the zip file
pathname = os.path.join(self.dirname, self.filename)
self.build_zip(pathname, archive_paths)
return pathname
def install(self, paths, maker, **kwargs):
"""
Install a wheel to the specified paths. If kwarg ``warner`` is
specified, it should be a callable, which will be called with two
tuples indicating the wheel version of this software and the wheel
version in the file, if there is a discrepancy in the versions.
This can be used to issue any warnings to raise any exceptions.
If kwarg ``lib_only`` is True, only the purelib/platlib files are
installed, and the headers, scripts, data and dist-info metadata are
not written.
The return value is a :class:`InstalledDistribution` instance unless
``options.lib_only`` is True, in which case the return value is ``None``.
"""
dry_run = maker.dry_run
warner = kwargs.get('warner')
lib_only = kwargs.get('lib_only', False)
pathname = os.path.join(self.dirname, self.filename)
name_ver = '%s-%s' % (self.name, self.version)
data_dir = '%s.data' % name_ver
info_dir = '%s.dist-info' % name_ver
metadata_name = posixpath.join(info_dir, METADATA_FILENAME)
wheel_metadata_name = posixpath.join(info_dir, 'WHEEL')
record_name = posixpath.join(info_dir, 'RECORD')
wrapper = codecs.getreader('utf-8')
with ZipFile(pathname, 'r') as zf:
with zf.open(wheel_metadata_name) as bwf:
wf = wrapper(bwf)
message = message_from_file(wf)
wv = message['Wheel-Version'].split('.', 1)
file_version = tuple([int(i) for i in wv])
if (file_version != self.wheel_version) and warner:
warner(self.wheel_version, file_version)
if message['Root-Is-Purelib'] == 'true':
libdir = paths['purelib']
else:
libdir = paths['platlib']
records = {}
with zf.open(record_name) as bf:
with CSVReader(stream=bf) as reader:
for row in reader:
p = row[0]
records[p] = row
data_pfx = posixpath.join(data_dir, '')
info_pfx = posixpath.join(info_dir, '')
script_pfx = posixpath.join(data_dir, 'scripts', '')
# make a new instance rather than a copy of maker's,
# as we mutate it
fileop = FileOperator(dry_run=dry_run)
fileop.record = True # so we can rollback if needed
bc = not sys.dont_write_bytecode # Double negatives. Lovely!
outfiles = [] # for RECORD writing
# for script copying/shebang processing
workdir = tempfile.mkdtemp()
# set target dir later
# we default add_launchers to False, as the
# Python Launcher should be used instead
maker.source_dir = workdir
maker.target_dir = None
try:
for zinfo in zf.infolist():
arcname = zinfo.filename
if isinstance(arcname, text_type):
u_arcname = arcname
else:
u_arcname = arcname.decode('utf-8')
# The signature file won't be in RECORD,
# and we don't currently don't do anything with it
if u_arcname.endswith('/RECORD.jws'):
continue
row = records[u_arcname]
if row[2] and str(zinfo.file_size) != row[2]:
raise DistlibException('size mismatch for '
'%s' % u_arcname)
if row[1]:
kind, value = row[1].split('=', 1)
with zf.open(arcname) as bf:
data = bf.read()
_, digest = self.get_hash(data, kind)
if digest != value:
raise DistlibException('digest mismatch for '
'%s' % arcname)
if lib_only and u_arcname.startswith((info_pfx, data_pfx)):
logger.debug('lib_only: skipping %s', u_arcname)
continue
is_script = (u_arcname.startswith(script_pfx)
and not u_arcname.endswith('.exe'))
if u_arcname.startswith(data_pfx):
_, where, rp = u_arcname.split('/', 2)
outfile = os.path.join(paths[where], convert_path(rp))
else:
# meant for site-packages.
if u_arcname in (wheel_metadata_name, record_name):
continue
outfile = os.path.join(libdir, convert_path(u_arcname))
if not is_script:
with zf.open(arcname) as bf:
fileop.copy_stream(bf, outfile)
outfiles.append(outfile)
# Double check the digest of the written file
if not dry_run and row[1]:
with open(outfile, 'rb') as bf:
data = bf.read()
_, newdigest = self.get_hash(data, kind)
if newdigest != digest:
raise DistlibException('digest mismatch '
'on write for '
'%s' % outfile)
if bc and outfile.endswith('.py'):
try:
pyc = fileop.byte_compile(outfile)
outfiles.append(pyc)
except Exception:
# Don't give up if byte-compilation fails,
# but log it and perhaps warn the user
logger.warning('Byte-compilation failed',
exc_info=True)
else:
fn = os.path.basename(convert_path(arcname))
workname = os.path.join(workdir, fn)
with zf.open(arcname) as bf:
fileop.copy_stream(bf, workname)
dn, fn = os.path.split(outfile)
maker.target_dir = dn
filenames = maker.make(fn)
fileop.set_executable_mode(filenames)
outfiles.extend(filenames)
if lib_only:
logger.debug('lib_only: returning None')
dist = None
else:
# Generate scripts
# Try to get pydist.json so we can see if there are
# any commands to generate. If this fails (e.g. because
# of a legacy wheel), log a warning but don't give up.
commands = None
file_version = self.info['Wheel-Version']
if file_version == '1.0':
# Use legacy info
ep = posixpath.join(info_dir, 'entry_points.txt')
try:
with zf.open(ep) as bwf:
epdata = read_exports(bwf)
commands = {}
for key in ('console', 'gui'):
k = '%s_scripts' % key
if k in epdata:
commands['wrap_%s' % key] = d = {}
for v in epdata[k].values():
s = '%s:%s' % (v.prefix, v.suffix)
if v.flags:
s += ' %s' % v.flags
d[v.name] = s
except Exception:
logger.warning('Unable to read legacy script '
'metadata, so cannot generate '
'scripts')
else:
try:
with zf.open(metadata_name) as bwf:
wf = wrapper(bwf)
commands = json.load(wf).get('commands')
except Exception:
logger.warning('Unable to read JSON metadata, so '
'cannot generate scripts')
if commands:
console_scripts = commands.get('wrap_console', {})
gui_scripts = commands.get('wrap_gui', {})
if console_scripts or gui_scripts:
script_dir = paths.get('scripts', '')
if not os.path.isdir(script_dir):
raise ValueError('Valid script path not '
'specified')
maker.target_dir = script_dir
for k, v in console_scripts.items():
script = '%s = %s' % (k, v)
filenames = maker.make(script)
fileop.set_executable_mode(filenames)
if gui_scripts:
options = {'gui': True }
for k, v in gui_scripts.items():
script = '%s = %s' % (k, v)
filenames = maker.make(script, options)
fileop.set_executable_mode(filenames)
p = os.path.join(libdir, info_dir)
dist = InstalledDistribution(p)
# Write SHARED
paths = dict(paths) # don't change passed in dict
del paths['purelib']
del paths['platlib']
paths['lib'] = libdir
p = dist.write_shared_locations(paths, dry_run)
if p:
outfiles.append(p)
# Write RECORD
dist.write_installed_files(outfiles, paths['prefix'],
dry_run)
return dist
except Exception: # pragma: no cover
logger.exception('installation failed.')
fileop.rollback()
raise
finally:
shutil.rmtree(workdir)
def _get_dylib_cache(self):
global cache
if cache is None:
# Use native string to avoid issues on 2.x: see Python #20140.
base = os.path.join(get_cache_base(), str('dylib-cache'),
sys.version[:3])
cache = Cache(base)
return cache
def _get_extensions(self):
pathname = os.path.join(self.dirname, self.filename)
name_ver = '%s-%s' % (self.name, self.version)
info_dir = '%s.dist-info' % name_ver
arcname = posixpath.join(info_dir, 'EXTENSIONS')
wrapper = codecs.getreader('utf-8')
result = []
with ZipFile(pathname, 'r') as zf:
try:
with zf.open(arcname) as bf:
wf = wrapper(bf)
extensions = json.load(wf)
cache = self._get_dylib_cache()
prefix = cache.prefix_to_dir(pathname)
cache_base = os.path.join(cache.base, prefix)
if not os.path.isdir(cache_base):
os.makedirs(cache_base)
for name, relpath in extensions.items():
dest = os.path.join(cache_base, convert_path(relpath))
if not os.path.exists(dest):
extract = True
else:
file_time = os.stat(dest).st_mtime
file_time = datetime.datetime.fromtimestamp(file_time)
info = zf.getinfo(relpath)
wheel_time = datetime.datetime(*info.date_time)
extract = wheel_time > file_time
if extract:
zf.extract(relpath, cache_base)
result.append((name, dest))
except KeyError:
pass
return result
def is_compatible(self):
"""
Determine if a wheel is compatible with the running system.
"""
return is_compatible(self)
def is_mountable(self):
"""
Determine if a wheel is asserted as mountable by its metadata.
"""
return True # for now - metadata details TBD
def mount(self, append=False):
pathname = os.path.abspath(os.path.join(self.dirname, self.filename))
if not self.is_compatible():
msg = 'Wheel %s not compatible with this Python.' % pathname
raise DistlibException(msg)
if not self.is_mountable():
msg = 'Wheel %s is marked as not mountable.' % pathname
raise DistlibException(msg)
if pathname in sys.path:
logger.debug('%s already in path', pathname)
else:
if append:
sys.path.append(pathname)
else:
sys.path.insert(0, pathname)
extensions = self._get_extensions()
if extensions:
if _hook not in sys.meta_path:
sys.meta_path.append(_hook)
_hook.add(pathname, extensions)
def unmount(self):
pathname = os.path.abspath(os.path.join(self.dirname, self.filename))
if pathname not in sys.path:
logger.debug('%s not in path', pathname)
else:
sys.path.remove(pathname)
if pathname in _hook.impure_wheels:
_hook.remove(pathname)
if not _hook.impure_wheels:
if _hook in sys.meta_path:
sys.meta_path.remove(_hook)
def verify(self):
pathname = os.path.join(self.dirname, self.filename)
name_ver = '%s-%s' % (self.name, self.version)
data_dir = '%s.data' % name_ver
info_dir = '%s.dist-info' % name_ver
metadata_name = posixpath.join(info_dir, METADATA_FILENAME)
wheel_metadata_name = posixpath.join(info_dir, 'WHEEL')
record_name = posixpath.join(info_dir, 'RECORD')
wrapper = codecs.getreader('utf-8')
with ZipFile(pathname, 'r') as zf:
with zf.open(wheel_metadata_name) as bwf:
wf = wrapper(bwf)
message = message_from_file(wf)
wv = message['Wheel-Version'].split('.', 1)
file_version = tuple([int(i) for i in wv])
# TODO version verification
records = {}
with zf.open(record_name) as bf:
with CSVReader(stream=bf) as reader:
for row in reader:
p = row[0]
records[p] = row
for zinfo in zf.infolist():
arcname = zinfo.filename
if isinstance(arcname, text_type):
u_arcname = arcname
else:
u_arcname = arcname.decode('utf-8')
if '..' in u_arcname:
raise DistlibException('invalid entry in '
'wheel: %r' % u_arcname)
# The signature file won't be in RECORD,
# and we don't currently don't do anything with it
if u_arcname.endswith('/RECORD.jws'):
continue
row = records[u_arcname]
if row[2] and str(zinfo.file_size) != row[2]:
raise DistlibException('size mismatch for '
'%s' % u_arcname)
if row[1]:
kind, value = row[1].split('=', 1)
with zf.open(arcname) as bf:
data = bf.read()
_, digest = self.get_hash(data, kind)
if digest != value:
raise DistlibException('digest mismatch for '
'%s' % arcname)
def update(self, modifier, dest_dir=None, **kwargs):
"""
Update the contents of a wheel in a generic way. The modifier should
be a callable which expects a dictionary argument: its keys are
archive-entry paths, and its values are absolute filesystem paths
where the contents the corresponding archive entries can be found. The
modifier is free to change the contents of the files pointed to, add
new entries and remove entries, before returning. This method will
extract the entire contents of the wheel to a temporary location, call
the modifier, and then use the passed (and possibly updated)
dictionary to write a new wheel. If ``dest_dir`` is specified, the new
wheel is written there -- otherwise, the original wheel is overwritten.
The modifier should return True if it updated the wheel, else False.
This method returns the same value the modifier returns.
"""
def get_version(path_map, info_dir):
version = path = None
key = '%s/%s' % (info_dir, METADATA_FILENAME)
if key not in path_map:
key = '%s/PKG-INFO' % info_dir
if key in path_map:
path = path_map[key]
version = Metadata(path=path).version
return version, path
def update_version(version, path):
updated = None
try:
v = NormalizedVersion(version)
i = version.find('-')
if i < 0:
updated = '%s-1' % version
else:
parts = [int(s) for s in version[i + 1:].split('.')]
parts[-1] += 1
updated = '%s-%s' % (version[:i],
'.'.join(str(i) for i in parts))
except UnsupportedVersionError:
logger.debug('Cannot update non-compliant (PEP-440) '
'version %r', version)
if updated:
md = Metadata(path=path)
md.version = updated
legacy = not path.endswith(METADATA_FILENAME)
md.write(path=path, legacy=legacy)
logger.debug('Version updated from %r to %r', version,
updated)
pathname = os.path.join(self.dirname, self.filename)
name_ver = '%s-%s' % (self.name, self.version)
info_dir = '%s.dist-info' % name_ver
record_name = posixpath.join(info_dir, 'RECORD')
with tempdir() as workdir:
with ZipFile(pathname, 'r') as zf:
path_map = {}
for zinfo in zf.infolist():
arcname = zinfo.filename
if isinstance(arcname, text_type):
u_arcname = arcname
else:
u_arcname = arcname.decode('utf-8')
if u_arcname == record_name:
continue
if '..' in u_arcname:
raise DistlibException('invalid entry in '
'wheel: %r' % u_arcname)
zf.extract(zinfo, workdir)
path = os.path.join(workdir, convert_path(u_arcname))
path_map[u_arcname] = path
# Remember the version.
original_version, _ = get_version(path_map, info_dir)
# Files extracted. Call the modifier.
modified = modifier(path_map, **kwargs)
if modified:
# Something changed - need to build a new wheel.
current_version, path = get_version(path_map, info_dir)
if current_version and (current_version == original_version):
# Add or update local version to signify changes.
update_version(current_version, path)
# Decide where the new wheel goes.
if dest_dir is None:
fd, newpath = tempfile.mkstemp(suffix='.whl',
prefix='wheel-update-',
dir=workdir)
os.close(fd)
else:
if not os.path.isdir(dest_dir):
raise DistlibException('Not a directory: %r' % dest_dir)
newpath = os.path.join(dest_dir, self.filename)
archive_paths = list(path_map.items())
distinfo = os.path.join(workdir, info_dir)
info = distinfo, info_dir
self.write_records(info, workdir, archive_paths)
self.build_zip(newpath, archive_paths)
if dest_dir is None:
shutil.copyfile(newpath, pathname)
return modified
def compatible_tags():
"""
Return (pyver, abi, arch) tuples compatible with this Python.
"""
versions = [VER_SUFFIX]
major = VER_SUFFIX[0]
for minor in range(sys.version_info[1] - 1, - 1, -1):
versions.append(''.join([major, str(minor)]))
abis = []
for suffix, _, _ in imp.get_suffixes():
if suffix.startswith('.abi'):
abis.append(suffix.split('.', 2)[1])
abis.sort()
if ABI != 'none':
abis.insert(0, ABI)
abis.append('none')
result = []
arches = [ARCH]
if sys.platform == 'darwin':
m = re.match('(\w+)_(\d+)_(\d+)_(\w+)$', ARCH)
if m:
name, major, minor, arch = m.groups()
minor = int(minor)
matches = [arch]
if arch in ('i386', 'ppc'):
matches.append('fat')
if arch in ('i386', 'ppc', 'x86_64'):
matches.append('fat3')
if arch in ('ppc64', 'x86_64'):
matches.append('fat64')
if arch in ('i386', 'x86_64'):
matches.append('intel')
if arch in ('i386', 'x86_64', 'intel', 'ppc', 'ppc64'):
matches.append('universal')
while minor >= 0:
for match in matches:
s = '%s_%s_%s_%s' % (name, major, minor, match)
if s != ARCH: # already there
arches.append(s)
minor -= 1
# Most specific - our Python version, ABI and arch
for abi in abis:
for arch in arches:
result.append((''.join((IMP_PREFIX, versions[0])), abi, arch))
# where no ABI / arch dependency, but IMP_PREFIX dependency
for i, version in enumerate(versions):
result.append((''.join((IMP_PREFIX, version)), 'none', 'any'))
if i == 0:
result.append((''.join((IMP_PREFIX, version[0])), 'none', 'any'))
# no IMP_PREFIX, ABI or arch dependency
for i, version in enumerate(versions):
result.append((''.join(('py', version)), 'none', 'any'))
if i == 0:
result.append((''.join(('py', version[0])), 'none', 'any'))
return set(result)
COMPATIBLE_TAGS = compatible_tags()
del compatible_tags
def is_compatible(wheel, tags=None):
if not isinstance(wheel, Wheel):
wheel = Wheel(wheel) # assume it's a filename
result = False
if tags is None:
tags = COMPATIBLE_TAGS
for ver, abi, arch in tags:
if ver in wheel.pyver and abi in wheel.abi and arch in wheel.arch:
result = True
break
return result
@@ -0,0 +1,23 @@
"""
HTML parsing library based on the WHATWG "HTML5"
specification. The parser is designed to be compatible with existing
HTML found in the wild and implements well-defined error recovery that
is largely compatible with modern desktop web browsers.
Example usage:
import html5lib
f = open("my_document.html")
tree = html5lib.parse(f)
"""
from __future__ import absolute_import, division, unicode_literals
from .html5parser import HTMLParser, parse, parseFragment
from .treebuilders import getTreeBuilder
from .treewalkers import getTreeWalker
from .serializer import serialize
__all__ = ["HTMLParser", "parse", "parseFragment", "getTreeBuilder",
"getTreeWalker", "serialize"]
__version__ = "1.0b3"
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,12 @@
from __future__ import absolute_import, division, unicode_literals
class Filter(object):
def __init__(self, source):
self.source = source
def __iter__(self):
return iter(self.source)
def __getattr__(self, name):
return getattr(self.source, name)
@@ -0,0 +1,20 @@
from __future__ import absolute_import, division, unicode_literals
from . import _base
try:
from collections import OrderedDict
except ImportError:
from ordereddict import OrderedDict
class Filter(_base.Filter):
def __iter__(self):
for token in _base.Filter.__iter__(self):
if token["type"] in ("StartTag", "EmptyTag"):
attrs = OrderedDict()
for name, value in sorted(token["data"].items(),
key=lambda x: x[0]):
attrs[name] = value
token["data"] = attrs
yield token
@@ -0,0 +1,65 @@
from __future__ import absolute_import, division, unicode_literals
from . import _base
class Filter(_base.Filter):
def __init__(self, source, encoding):
_base.Filter.__init__(self, source)
self.encoding = encoding
def __iter__(self):
state = "pre_head"
meta_found = (self.encoding is None)
pending = []
for token in _base.Filter.__iter__(self):
type = token["type"]
if type == "StartTag":
if token["name"].lower() == "head":
state = "in_head"
elif type == "EmptyTag":
if token["name"].lower() == "meta":
# replace charset with actual encoding
has_http_equiv_content_type = False
for (namespace, name), value in token["data"].items():
if namespace is not None:
continue
elif name.lower() == 'charset':
token["data"][(namespace, name)] = self.encoding
meta_found = True
break
elif name == 'http-equiv' and value.lower() == 'content-type':
has_http_equiv_content_type = True
else:
if has_http_equiv_content_type and (None, "content") in token["data"]:
token["data"][(None, "content")] = 'text/html; charset=%s' % self.encoding
meta_found = True
elif token["name"].lower() == "head" and not meta_found:
# insert meta into empty head
yield {"type": "StartTag", "name": "head",
"data": token["data"]}
yield {"type": "EmptyTag", "name": "meta",
"data": {(None, "charset"): self.encoding}}
yield {"type": "EndTag", "name": "head"}
meta_found = True
continue
elif type == "EndTag":
if token["name"].lower() == "head" and pending:
# insert meta into head (if necessary) and flush pending queue
yield pending.pop(0)
if not meta_found:
yield {"type": "EmptyTag", "name": "meta",
"data": {(None, "charset"): self.encoding}}
while pending:
yield pending.pop(0)
meta_found = True
state = "post_head"
if state == "in_head":
pending.append(token)
else:
yield token
@@ -0,0 +1,93 @@
from __future__ import absolute_import, division, unicode_literals
from gettext import gettext
_ = gettext
from . import _base
from ..constants import cdataElements, rcdataElements, voidElements
from ..constants import spaceCharacters
spaceCharacters = "".join(spaceCharacters)
class LintError(Exception):
pass
class Filter(_base.Filter):
def __iter__(self):
open_elements = []
contentModelFlag = "PCDATA"
for token in _base.Filter.__iter__(self):
type = token["type"]
if type in ("StartTag", "EmptyTag"):
name = token["name"]
if contentModelFlag != "PCDATA":
raise LintError(_("StartTag not in PCDATA content model flag: %(tag)s") % {"tag": name})
if not isinstance(name, str):
raise LintError(_("Tag name is not a string: %(tag)r") % {"tag": name})
if not name:
raise LintError(_("Empty tag name"))
if type == "StartTag" and name in voidElements:
raise LintError(_("Void element reported as StartTag token: %(tag)s") % {"tag": name})
elif type == "EmptyTag" and name not in voidElements:
raise LintError(_("Non-void element reported as EmptyTag token: %(tag)s") % {"tag": token["name"]})
if type == "StartTag":
open_elements.append(name)
for name, value in token["data"]:
if not isinstance(name, str):
raise LintError(_("Attribute name is not a string: %(name)r") % {"name": name})
if not name:
raise LintError(_("Empty attribute name"))
if not isinstance(value, str):
raise LintError(_("Attribute value is not a string: %(value)r") % {"value": value})
if name in cdataElements:
contentModelFlag = "CDATA"
elif name in rcdataElements:
contentModelFlag = "RCDATA"
elif name == "plaintext":
contentModelFlag = "PLAINTEXT"
elif type == "EndTag":
name = token["name"]
if not isinstance(name, str):
raise LintError(_("Tag name is not a string: %(tag)r") % {"tag": name})
if not name:
raise LintError(_("Empty tag name"))
if name in voidElements:
raise LintError(_("Void element reported as EndTag token: %(tag)s") % {"tag": name})
start_name = open_elements.pop()
if start_name != name:
raise LintError(_("EndTag (%(end)s) does not match StartTag (%(start)s)") % {"end": name, "start": start_name})
contentModelFlag = "PCDATA"
elif type == "Comment":
if contentModelFlag != "PCDATA":
raise LintError(_("Comment not in PCDATA content model flag"))
elif type in ("Characters", "SpaceCharacters"):
data = token["data"]
if not isinstance(data, str):
raise LintError(_("Attribute name is not a string: %(name)r") % {"name": data})
if not data:
raise LintError(_("%(type)s token with empty data") % {"type": type})
if type == "SpaceCharacters":
data = data.strip(spaceCharacters)
if data:
raise LintError(_("Non-space character(s) found in SpaceCharacters token: %(token)r") % {"token": data})
elif type == "Doctype":
name = token["name"]
if contentModelFlag != "PCDATA":
raise LintError(_("Doctype not in PCDATA content model flag: %(name)s") % {"name": name})
if not isinstance(name, str):
raise LintError(_("Tag name is not a string: %(tag)r") % {"tag": name})
# XXX: what to do with token["data"] ?
elif type in ("ParseError", "SerializeError"):
pass
else:
raise LintError(_("Unknown token type: %(type)s") % {"type": type})
yield token
@@ -0,0 +1,205 @@
from __future__ import absolute_import, division, unicode_literals
from . import _base
class Filter(_base.Filter):
def slider(self):
previous1 = previous2 = None
for token in self.source:
if previous1 is not None:
yield previous2, previous1, token
previous2 = previous1
previous1 = token
yield previous2, previous1, None
def __iter__(self):
for previous, token, next in self.slider():
type = token["type"]
if type == "StartTag":
if (token["data"] or
not self.is_optional_start(token["name"], previous, next)):
yield token
elif type == "EndTag":
if not self.is_optional_end(token["name"], next):
yield token
else:
yield token
def is_optional_start(self, tagname, previous, next):
type = next and next["type"] or None
if tagname in 'html':
# An html element's start tag may be omitted if the first thing
# inside the html element is not a space character or a comment.
return type not in ("Comment", "SpaceCharacters")
elif tagname == 'head':
# A head element's start tag may be omitted if the first thing
# inside the head element is an element.
# XXX: we also omit the start tag if the head element is empty
if type in ("StartTag", "EmptyTag"):
return True
elif type == "EndTag":
return next["name"] == "head"
elif tagname == 'body':
# A body element's start tag may be omitted if the first thing
# inside the body element is not a space character or a comment,
# except if the first thing inside the body element is a script
# or style element and the node immediately preceding the body
# element is a head element whose end tag has been omitted.
if type in ("Comment", "SpaceCharacters"):
return False
elif type == "StartTag":
# XXX: we do not look at the preceding event, so we never omit
# the body element's start tag if it's followed by a script or
# a style element.
return next["name"] not in ('script', 'style')
else:
return True
elif tagname == 'colgroup':
# A colgroup element's start tag may be omitted if the first thing
# inside the colgroup element is a col element, and if the element
# is not immediately preceeded by another colgroup element whose
# end tag has been omitted.
if type in ("StartTag", "EmptyTag"):
# XXX: we do not look at the preceding event, so instead we never
# omit the colgroup element's end tag when it is immediately
# followed by another colgroup element. See is_optional_end.
return next["name"] == "col"
else:
return False
elif tagname == 'tbody':
# A tbody element's start tag may be omitted if the first thing
# inside the tbody element is a tr element, and if the element is
# not immediately preceeded by a tbody, thead, or tfoot element
# whose end tag has been omitted.
if type == "StartTag":
# omit the thead and tfoot elements' end tag when they are
# immediately followed by a tbody element. See is_optional_end.
if previous and previous['type'] == 'EndTag' and \
previous['name'] in ('tbody', 'thead', 'tfoot'):
return False
return next["name"] == 'tr'
else:
return False
return False
def is_optional_end(self, tagname, next):
type = next and next["type"] or None
if tagname in ('html', 'head', 'body'):
# An html element's end tag may be omitted if the html element
# is not immediately followed by a space character or a comment.
return type not in ("Comment", "SpaceCharacters")
elif tagname in ('li', 'optgroup', 'tr'):
# A li element's end tag may be omitted if the li element is
# immediately followed by another li element or if there is
# no more content in the parent element.
# An optgroup element's end tag may be omitted if the optgroup
# element is immediately followed by another optgroup element,
# or if there is no more content in the parent element.
# A tr element's end tag may be omitted if the tr element is
# immediately followed by another tr element, or if there is
# no more content in the parent element.
if type == "StartTag":
return next["name"] == tagname
else:
return type == "EndTag" or type is None
elif tagname in ('dt', 'dd'):
# A dt element's end tag may be omitted if the dt element is
# immediately followed by another dt element or a dd element.
# A dd element's end tag may be omitted if the dd element is
# immediately followed by another dd element or a dt element,
# or if there is no more content in the parent element.
if type == "StartTag":
return next["name"] in ('dt', 'dd')
elif tagname == 'dd':
return type == "EndTag" or type is None
else:
return False
elif tagname == 'p':
# A p element's end tag may be omitted if the p element is
# immediately followed by an address, article, aside,
# blockquote, datagrid, dialog, dir, div, dl, fieldset,
# footer, form, h1, h2, h3, h4, h5, h6, header, hr, menu,
# nav, ol, p, pre, section, table, or ul, element, or if
# there is no more content in the parent element.
if type in ("StartTag", "EmptyTag"):
return next["name"] in ('address', 'article', 'aside',
'blockquote', 'datagrid', 'dialog',
'dir', 'div', 'dl', 'fieldset', 'footer',
'form', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6',
'header', 'hr', 'menu', 'nav', 'ol',
'p', 'pre', 'section', 'table', 'ul')
else:
return type == "EndTag" or type is None
elif tagname == 'option':
# An option element's end tag may be omitted if the option
# element is immediately followed by another option element,
# or if it is immediately followed by an <code>optgroup</code>
# element, or if there is no more content in the parent
# element.
if type == "StartTag":
return next["name"] in ('option', 'optgroup')
else:
return type == "EndTag" or type is None
elif tagname in ('rt', 'rp'):
# An rt element's end tag may be omitted if the rt element is
# immediately followed by an rt or rp element, or if there is
# no more content in the parent element.
# An rp element's end tag may be omitted if the rp element is
# immediately followed by an rt or rp element, or if there is
# no more content in the parent element.
if type == "StartTag":
return next["name"] in ('rt', 'rp')
else:
return type == "EndTag" or type is None
elif tagname == 'colgroup':
# A colgroup element's end tag may be omitted if the colgroup
# element is not immediately followed by a space character or
# a comment.
if type in ("Comment", "SpaceCharacters"):
return False
elif type == "StartTag":
# XXX: we also look for an immediately following colgroup
# element. See is_optional_start.
return next["name"] != 'colgroup'
else:
return True
elif tagname in ('thead', 'tbody'):
# A thead element's end tag may be omitted if the thead element
# is immediately followed by a tbody or tfoot element.
# A tbody element's end tag may be omitted if the tbody element
# is immediately followed by a tbody or tfoot element, or if
# there is no more content in the parent element.
# A tfoot element's end tag may be omitted if the tfoot element
# is immediately followed by a tbody element, or if there is no
# more content in the parent element.
# XXX: we never omit the end tag when the following element is
# a tbody. See is_optional_start.
if type == "StartTag":
return next["name"] in ['tbody', 'tfoot']
elif tagname == 'tbody':
return type == "EndTag" or type is None
else:
return False
elif tagname == 'tfoot':
# A tfoot element's end tag may be omitted if the tfoot element
# is immediately followed by a tbody element, or if there is no
# more content in the parent element.
# XXX: we never omit the end tag when the following element is
# a tbody. See is_optional_start.
if type == "StartTag":
return next["name"] == 'tbody'
else:
return type == "EndTag" or type is None
elif tagname in ('td', 'th'):
# A td element's end tag may be omitted if the td element is
# immediately followed by a td or th element, or if there is
# no more content in the parent element.
# A th element's end tag may be omitted if the th element is
# immediately followed by a td or th element, or if there is
# no more content in the parent element.
if type == "StartTag":
return next["name"] in ('td', 'th')
else:
return type == "EndTag" or type is None
return False
@@ -0,0 +1,12 @@
from __future__ import absolute_import, division, unicode_literals
from . import _base
from ..sanitizer import HTMLSanitizerMixin
class Filter(_base.Filter, HTMLSanitizerMixin):
def __iter__(self):
for token in _base.Filter.__iter__(self):
token = self.sanitize_token(token)
if token:
yield token
@@ -0,0 +1,38 @@
from __future__ import absolute_import, division, unicode_literals
import re
from . import _base
from ..constants import rcdataElements, spaceCharacters
spaceCharacters = "".join(spaceCharacters)
SPACES_REGEX = re.compile("[%s]+" % spaceCharacters)
class Filter(_base.Filter):
spacePreserveElements = frozenset(["pre", "textarea"] + list(rcdataElements))
def __iter__(self):
preserve = 0
for token in _base.Filter.__iter__(self):
type = token["type"]
if type == "StartTag" \
and (preserve or token["name"] in self.spacePreserveElements):
preserve += 1
elif type == "EndTag" and preserve:
preserve -= 1
elif not preserve and type == "SpaceCharacters" and token["data"]:
# Test on token["data"] above to not introduce spaces where there were not
token["data"] = " "
elif not preserve and type == "Characters":
token["data"] = collapse_spaces(token["data"])
yield token
def collapse_spaces(text):
return SPACES_REGEX.sub(' ', text)
File diff suppressed because it is too large Load Diff
@@ -0,0 +1,285 @@
from __future__ import absolute_import, division, unicode_literals
import re
import warnings
from .constants import DataLossWarning
baseChar = """
[#x0041-#x005A] | [#x0061-#x007A] | [#x00C0-#x00D6] | [#x00D8-#x00F6] |
[#x00F8-#x00FF] | [#x0100-#x0131] | [#x0134-#x013E] | [#x0141-#x0148] |
[#x014A-#x017E] | [#x0180-#x01C3] | [#x01CD-#x01F0] | [#x01F4-#x01F5] |
[#x01FA-#x0217] | [#x0250-#x02A8] | [#x02BB-#x02C1] | #x0386 |
[#x0388-#x038A] | #x038C | [#x038E-#x03A1] | [#x03A3-#x03CE] |
[#x03D0-#x03D6] | #x03DA | #x03DC | #x03DE | #x03E0 | [#x03E2-#x03F3] |
[#x0401-#x040C] | [#x040E-#x044F] | [#x0451-#x045C] | [#x045E-#x0481] |
[#x0490-#x04C4] | [#x04C7-#x04C8] | [#x04CB-#x04CC] | [#x04D0-#x04EB] |
[#x04EE-#x04F5] | [#x04F8-#x04F9] | [#x0531-#x0556] | #x0559 |
[#x0561-#x0586] | [#x05D0-#x05EA] | [#x05F0-#x05F2] | [#x0621-#x063A] |
[#x0641-#x064A] | [#x0671-#x06B7] | [#x06BA-#x06BE] | [#x06C0-#x06CE] |
[#x06D0-#x06D3] | #x06D5 | [#x06E5-#x06E6] | [#x0905-#x0939] | #x093D |
[#x0958-#x0961] | [#x0985-#x098C] | [#x098F-#x0990] | [#x0993-#x09A8] |
[#x09AA-#x09B0] | #x09B2 | [#x09B6-#x09B9] | [#x09DC-#x09DD] |
[#x09DF-#x09E1] | [#x09F0-#x09F1] | [#x0A05-#x0A0A] | [#x0A0F-#x0A10] |
[#x0A13-#x0A28] | [#x0A2A-#x0A30] | [#x0A32-#x0A33] | [#x0A35-#x0A36] |
[#x0A38-#x0A39] | [#x0A59-#x0A5C] | #x0A5E | [#x0A72-#x0A74] |
[#x0A85-#x0A8B] | #x0A8D | [#x0A8F-#x0A91] | [#x0A93-#x0AA8] |
[#x0AAA-#x0AB0] | [#x0AB2-#x0AB3] | [#x0AB5-#x0AB9] | #x0ABD | #x0AE0 |
[#x0B05-#x0B0C] | [#x0B0F-#x0B10] | [#x0B13-#x0B28] | [#x0B2A-#x0B30] |
[#x0B32-#x0B33] | [#x0B36-#x0B39] | #x0B3D | [#x0B5C-#x0B5D] |
[#x0B5F-#x0B61] | [#x0B85-#x0B8A] | [#x0B8E-#x0B90] | [#x0B92-#x0B95] |
[#x0B99-#x0B9A] | #x0B9C | [#x0B9E-#x0B9F] | [#x0BA3-#x0BA4] |
[#x0BA8-#x0BAA] | [#x0BAE-#x0BB5] | [#x0BB7-#x0BB9] | [#x0C05-#x0C0C] |
[#x0C0E-#x0C10] | [#x0C12-#x0C28] | [#x0C2A-#x0C33] | [#x0C35-#x0C39] |
[#x0C60-#x0C61] | [#x0C85-#x0C8C] | [#x0C8E-#x0C90] | [#x0C92-#x0CA8] |
[#x0CAA-#x0CB3] | [#x0CB5-#x0CB9] | #x0CDE | [#x0CE0-#x0CE1] |
[#x0D05-#x0D0C] | [#x0D0E-#x0D10] | [#x0D12-#x0D28] | [#x0D2A-#x0D39] |
[#x0D60-#x0D61] | [#x0E01-#x0E2E] | #x0E30 | [#x0E32-#x0E33] |
[#x0E40-#x0E45] | [#x0E81-#x0E82] | #x0E84 | [#x0E87-#x0E88] | #x0E8A |
#x0E8D | [#x0E94-#x0E97] | [#x0E99-#x0E9F] | [#x0EA1-#x0EA3] | #x0EA5 |
#x0EA7 | [#x0EAA-#x0EAB] | [#x0EAD-#x0EAE] | #x0EB0 | [#x0EB2-#x0EB3] |
#x0EBD | [#x0EC0-#x0EC4] | [#x0F40-#x0F47] | [#x0F49-#x0F69] |
[#x10A0-#x10C5] | [#x10D0-#x10F6] | #x1100 | [#x1102-#x1103] |
[#x1105-#x1107] | #x1109 | [#x110B-#x110C] | [#x110E-#x1112] | #x113C |
#x113E | #x1140 | #x114C | #x114E | #x1150 | [#x1154-#x1155] | #x1159 |
[#x115F-#x1161] | #x1163 | #x1165 | #x1167 | #x1169 | [#x116D-#x116E] |
[#x1172-#x1173] | #x1175 | #x119E | #x11A8 | #x11AB | [#x11AE-#x11AF] |
[#x11B7-#x11B8] | #x11BA | [#x11BC-#x11C2] | #x11EB | #x11F0 | #x11F9 |
[#x1E00-#x1E9B] | [#x1EA0-#x1EF9] | [#x1F00-#x1F15] | [#x1F18-#x1F1D] |
[#x1F20-#x1F45] | [#x1F48-#x1F4D] | [#x1F50-#x1F57] | #x1F59 | #x1F5B |
#x1F5D | [#x1F5F-#x1F7D] | [#x1F80-#x1FB4] | [#x1FB6-#x1FBC] | #x1FBE |
[#x1FC2-#x1FC4] | [#x1FC6-#x1FCC] | [#x1FD0-#x1FD3] | [#x1FD6-#x1FDB] |
[#x1FE0-#x1FEC] | [#x1FF2-#x1FF4] | [#x1FF6-#x1FFC] | #x2126 |
[#x212A-#x212B] | #x212E | [#x2180-#x2182] | [#x3041-#x3094] |
[#x30A1-#x30FA] | [#x3105-#x312C] | [#xAC00-#xD7A3]"""
ideographic = """[#x4E00-#x9FA5] | #x3007 | [#x3021-#x3029]"""
combiningCharacter = """
[#x0300-#x0345] | [#x0360-#x0361] | [#x0483-#x0486] | [#x0591-#x05A1] |
[#x05A3-#x05B9] | [#x05BB-#x05BD] | #x05BF | [#x05C1-#x05C2] | #x05C4 |
[#x064B-#x0652] | #x0670 | [#x06D6-#x06DC] | [#x06DD-#x06DF] |
[#x06E0-#x06E4] | [#x06E7-#x06E8] | [#x06EA-#x06ED] | [#x0901-#x0903] |
#x093C | [#x093E-#x094C] | #x094D | [#x0951-#x0954] | [#x0962-#x0963] |
[#x0981-#x0983] | #x09BC | #x09BE | #x09BF | [#x09C0-#x09C4] |
[#x09C7-#x09C8] | [#x09CB-#x09CD] | #x09D7 | [#x09E2-#x09E3] | #x0A02 |
#x0A3C | #x0A3E | #x0A3F | [#x0A40-#x0A42] | [#x0A47-#x0A48] |
[#x0A4B-#x0A4D] | [#x0A70-#x0A71] | [#x0A81-#x0A83] | #x0ABC |
[#x0ABE-#x0AC5] | [#x0AC7-#x0AC9] | [#x0ACB-#x0ACD] | [#x0B01-#x0B03] |
#x0B3C | [#x0B3E-#x0B43] | [#x0B47-#x0B48] | [#x0B4B-#x0B4D] |
[#x0B56-#x0B57] | [#x0B82-#x0B83] | [#x0BBE-#x0BC2] | [#x0BC6-#x0BC8] |
[#x0BCA-#x0BCD] | #x0BD7 | [#x0C01-#x0C03] | [#x0C3E-#x0C44] |
[#x0C46-#x0C48] | [#x0C4A-#x0C4D] | [#x0C55-#x0C56] | [#x0C82-#x0C83] |
[#x0CBE-#x0CC4] | [#x0CC6-#x0CC8] | [#x0CCA-#x0CCD] | [#x0CD5-#x0CD6] |
[#x0D02-#x0D03] | [#x0D3E-#x0D43] | [#x0D46-#x0D48] | [#x0D4A-#x0D4D] |
#x0D57 | #x0E31 | [#x0E34-#x0E3A] | [#x0E47-#x0E4E] | #x0EB1 |
[#x0EB4-#x0EB9] | [#x0EBB-#x0EBC] | [#x0EC8-#x0ECD] | [#x0F18-#x0F19] |
#x0F35 | #x0F37 | #x0F39 | #x0F3E | #x0F3F | [#x0F71-#x0F84] |
[#x0F86-#x0F8B] | [#x0F90-#x0F95] | #x0F97 | [#x0F99-#x0FAD] |
[#x0FB1-#x0FB7] | #x0FB9 | [#x20D0-#x20DC] | #x20E1 | [#x302A-#x302F] |
#x3099 | #x309A"""
digit = """
[#x0030-#x0039] | [#x0660-#x0669] | [#x06F0-#x06F9] | [#x0966-#x096F] |
[#x09E6-#x09EF] | [#x0A66-#x0A6F] | [#x0AE6-#x0AEF] | [#x0B66-#x0B6F] |
[#x0BE7-#x0BEF] | [#x0C66-#x0C6F] | [#x0CE6-#x0CEF] | [#x0D66-#x0D6F] |
[#x0E50-#x0E59] | [#x0ED0-#x0ED9] | [#x0F20-#x0F29]"""
extender = """
#x00B7 | #x02D0 | #x02D1 | #x0387 | #x0640 | #x0E46 | #x0EC6 | #x3005 |
#[#x3031-#x3035] | [#x309D-#x309E] | [#x30FC-#x30FE]"""
letter = " | ".join([baseChar, ideographic])
# Without the
name = " | ".join([letter, digit, ".", "-", "_", combiningCharacter,
extender])
nameFirst = " | ".join([letter, "_"])
reChar = re.compile(r"#x([\d|A-F]{4,4})")
reCharRange = re.compile(r"\[#x([\d|A-F]{4,4})-#x([\d|A-F]{4,4})\]")
def charStringToList(chars):
charRanges = [item.strip() for item in chars.split(" | ")]
rv = []
for item in charRanges:
foundMatch = False
for regexp in (reChar, reCharRange):
match = regexp.match(item)
if match is not None:
rv.append([hexToInt(item) for item in match.groups()])
if len(rv[-1]) == 1:
rv[-1] = rv[-1] * 2
foundMatch = True
break
if not foundMatch:
assert len(item) == 1
rv.append([ord(item)] * 2)
rv = normaliseCharList(rv)
return rv
def normaliseCharList(charList):
charList = sorted(charList)
for item in charList:
assert item[1] >= item[0]
rv = []
i = 0
while i < len(charList):
j = 1
rv.append(charList[i])
while i + j < len(charList) and charList[i + j][0] <= rv[-1][1] + 1:
rv[-1][1] = charList[i + j][1]
j += 1
i += j
return rv
# We don't really support characters above the BMP :(
max_unicode = int("FFFF", 16)
def missingRanges(charList):
rv = []
if charList[0] != 0:
rv.append([0, charList[0][0] - 1])
for i, item in enumerate(charList[:-1]):
rv.append([item[1] + 1, charList[i + 1][0] - 1])
if charList[-1][1] != max_unicode:
rv.append([charList[-1][1] + 1, max_unicode])
return rv
def listToRegexpStr(charList):
rv = []
for item in charList:
if item[0] == item[1]:
rv.append(escapeRegexp(chr(item[0])))
else:
rv.append(escapeRegexp(chr(item[0])) + "-" +
escapeRegexp(chr(item[1])))
return "[%s]" % "".join(rv)
def hexToInt(hex_str):
return int(hex_str, 16)
def escapeRegexp(string):
specialCharacters = (".", "^", "$", "*", "+", "?", "{", "}",
"[", "]", "|", "(", ")", "-")
for char in specialCharacters:
string = string.replace(char, "\\" + char)
return string
# output from the above
nonXmlNameBMPRegexp = re.compile('[\x00-,/:-@\\[-\\^`\\{-\xb6\xb8-\xbf\xd7\xf7\u0132-\u0133\u013f-\u0140\u0149\u017f\u01c4-\u01cc\u01f1-\u01f3\u01f6-\u01f9\u0218-\u024f\u02a9-\u02ba\u02c2-\u02cf\u02d2-\u02ff\u0346-\u035f\u0362-\u0385\u038b\u038d\u03a2\u03cf\u03d7-\u03d9\u03db\u03dd\u03df\u03e1\u03f4-\u0400\u040d\u0450\u045d\u0482\u0487-\u048f\u04c5-\u04c6\u04c9-\u04ca\u04cd-\u04cf\u04ec-\u04ed\u04f6-\u04f7\u04fa-\u0530\u0557-\u0558\u055a-\u0560\u0587-\u0590\u05a2\u05ba\u05be\u05c0\u05c3\u05c5-\u05cf\u05eb-\u05ef\u05f3-\u0620\u063b-\u063f\u0653-\u065f\u066a-\u066f\u06b8-\u06b9\u06bf\u06cf\u06d4\u06e9\u06ee-\u06ef\u06fa-\u0900\u0904\u093a-\u093b\u094e-\u0950\u0955-\u0957\u0964-\u0965\u0970-\u0980\u0984\u098d-\u098e\u0991-\u0992\u09a9\u09b1\u09b3-\u09b5\u09ba-\u09bb\u09bd\u09c5-\u09c6\u09c9-\u09ca\u09ce-\u09d6\u09d8-\u09db\u09de\u09e4-\u09e5\u09f2-\u0a01\u0a03-\u0a04\u0a0b-\u0a0e\u0a11-\u0a12\u0a29\u0a31\u0a34\u0a37\u0a3a-\u0a3b\u0a3d\u0a43-\u0a46\u0a49-\u0a4a\u0a4e-\u0a58\u0a5d\u0a5f-\u0a65\u0a75-\u0a80\u0a84\u0a8c\u0a8e\u0a92\u0aa9\u0ab1\u0ab4\u0aba-\u0abb\u0ac6\u0aca\u0ace-\u0adf\u0ae1-\u0ae5\u0af0-\u0b00\u0b04\u0b0d-\u0b0e\u0b11-\u0b12\u0b29\u0b31\u0b34-\u0b35\u0b3a-\u0b3b\u0b44-\u0b46\u0b49-\u0b4a\u0b4e-\u0b55\u0b58-\u0b5b\u0b5e\u0b62-\u0b65\u0b70-\u0b81\u0b84\u0b8b-\u0b8d\u0b91\u0b96-\u0b98\u0b9b\u0b9d\u0ba0-\u0ba2\u0ba5-\u0ba7\u0bab-\u0bad\u0bb6\u0bba-\u0bbd\u0bc3-\u0bc5\u0bc9\u0bce-\u0bd6\u0bd8-\u0be6\u0bf0-\u0c00\u0c04\u0c0d\u0c11\u0c29\u0c34\u0c3a-\u0c3d\u0c45\u0c49\u0c4e-\u0c54\u0c57-\u0c5f\u0c62-\u0c65\u0c70-\u0c81\u0c84\u0c8d\u0c91\u0ca9\u0cb4\u0cba-\u0cbd\u0cc5\u0cc9\u0cce-\u0cd4\u0cd7-\u0cdd\u0cdf\u0ce2-\u0ce5\u0cf0-\u0d01\u0d04\u0d0d\u0d11\u0d29\u0d3a-\u0d3d\u0d44-\u0d45\u0d49\u0d4e-\u0d56\u0d58-\u0d5f\u0d62-\u0d65\u0d70-\u0e00\u0e2f\u0e3b-\u0e3f\u0e4f\u0e5a-\u0e80\u0e83\u0e85-\u0e86\u0e89\u0e8b-\u0e8c\u0e8e-\u0e93\u0e98\u0ea0\u0ea4\u0ea6\u0ea8-\u0ea9\u0eac\u0eaf\u0eba\u0ebe-\u0ebf\u0ec5\u0ec7\u0ece-\u0ecf\u0eda-\u0f17\u0f1a-\u0f1f\u0f2a-\u0f34\u0f36\u0f38\u0f3a-\u0f3d\u0f48\u0f6a-\u0f70\u0f85\u0f8c-\u0f8f\u0f96\u0f98\u0fae-\u0fb0\u0fb8\u0fba-\u109f\u10c6-\u10cf\u10f7-\u10ff\u1101\u1104\u1108\u110a\u110d\u1113-\u113b\u113d\u113f\u1141-\u114b\u114d\u114f\u1151-\u1153\u1156-\u1158\u115a-\u115e\u1162\u1164\u1166\u1168\u116a-\u116c\u116f-\u1171\u1174\u1176-\u119d\u119f-\u11a7\u11a9-\u11aa\u11ac-\u11ad\u11b0-\u11b6\u11b9\u11bb\u11c3-\u11ea\u11ec-\u11ef\u11f1-\u11f8\u11fa-\u1dff\u1e9c-\u1e9f\u1efa-\u1eff\u1f16-\u1f17\u1f1e-\u1f1f\u1f46-\u1f47\u1f4e-\u1f4f\u1f58\u1f5a\u1f5c\u1f5e\u1f7e-\u1f7f\u1fb5\u1fbd\u1fbf-\u1fc1\u1fc5\u1fcd-\u1fcf\u1fd4-\u1fd5\u1fdc-\u1fdf\u1fed-\u1ff1\u1ff5\u1ffd-\u20cf\u20dd-\u20e0\u20e2-\u2125\u2127-\u2129\u212c-\u212d\u212f-\u217f\u2183-\u3004\u3006\u3008-\u3020\u3030\u3036-\u3040\u3095-\u3098\u309b-\u309c\u309f-\u30a0\u30fb\u30ff-\u3104\u312d-\u4dff\u9fa6-\uabff\ud7a4-\uffff]')
nonXmlNameFirstBMPRegexp = re.compile('[\x00-@\\[-\\^`\\{-\xbf\xd7\xf7\u0132-\u0133\u013f-\u0140\u0149\u017f\u01c4-\u01cc\u01f1-\u01f3\u01f6-\u01f9\u0218-\u024f\u02a9-\u02ba\u02c2-\u0385\u0387\u038b\u038d\u03a2\u03cf\u03d7-\u03d9\u03db\u03dd\u03df\u03e1\u03f4-\u0400\u040d\u0450\u045d\u0482-\u048f\u04c5-\u04c6\u04c9-\u04ca\u04cd-\u04cf\u04ec-\u04ed\u04f6-\u04f7\u04fa-\u0530\u0557-\u0558\u055a-\u0560\u0587-\u05cf\u05eb-\u05ef\u05f3-\u0620\u063b-\u0640\u064b-\u0670\u06b8-\u06b9\u06bf\u06cf\u06d4\u06d6-\u06e4\u06e7-\u0904\u093a-\u093c\u093e-\u0957\u0962-\u0984\u098d-\u098e\u0991-\u0992\u09a9\u09b1\u09b3-\u09b5\u09ba-\u09db\u09de\u09e2-\u09ef\u09f2-\u0a04\u0a0b-\u0a0e\u0a11-\u0a12\u0a29\u0a31\u0a34\u0a37\u0a3a-\u0a58\u0a5d\u0a5f-\u0a71\u0a75-\u0a84\u0a8c\u0a8e\u0a92\u0aa9\u0ab1\u0ab4\u0aba-\u0abc\u0abe-\u0adf\u0ae1-\u0b04\u0b0d-\u0b0e\u0b11-\u0b12\u0b29\u0b31\u0b34-\u0b35\u0b3a-\u0b3c\u0b3e-\u0b5b\u0b5e\u0b62-\u0b84\u0b8b-\u0b8d\u0b91\u0b96-\u0b98\u0b9b\u0b9d\u0ba0-\u0ba2\u0ba5-\u0ba7\u0bab-\u0bad\u0bb6\u0bba-\u0c04\u0c0d\u0c11\u0c29\u0c34\u0c3a-\u0c5f\u0c62-\u0c84\u0c8d\u0c91\u0ca9\u0cb4\u0cba-\u0cdd\u0cdf\u0ce2-\u0d04\u0d0d\u0d11\u0d29\u0d3a-\u0d5f\u0d62-\u0e00\u0e2f\u0e31\u0e34-\u0e3f\u0e46-\u0e80\u0e83\u0e85-\u0e86\u0e89\u0e8b-\u0e8c\u0e8e-\u0e93\u0e98\u0ea0\u0ea4\u0ea6\u0ea8-\u0ea9\u0eac\u0eaf\u0eb1\u0eb4-\u0ebc\u0ebe-\u0ebf\u0ec5-\u0f3f\u0f48\u0f6a-\u109f\u10c6-\u10cf\u10f7-\u10ff\u1101\u1104\u1108\u110a\u110d\u1113-\u113b\u113d\u113f\u1141-\u114b\u114d\u114f\u1151-\u1153\u1156-\u1158\u115a-\u115e\u1162\u1164\u1166\u1168\u116a-\u116c\u116f-\u1171\u1174\u1176-\u119d\u119f-\u11a7\u11a9-\u11aa\u11ac-\u11ad\u11b0-\u11b6\u11b9\u11bb\u11c3-\u11ea\u11ec-\u11ef\u11f1-\u11f8\u11fa-\u1dff\u1e9c-\u1e9f\u1efa-\u1eff\u1f16-\u1f17\u1f1e-\u1f1f\u1f46-\u1f47\u1f4e-\u1f4f\u1f58\u1f5a\u1f5c\u1f5e\u1f7e-\u1f7f\u1fb5\u1fbd\u1fbf-\u1fc1\u1fc5\u1fcd-\u1fcf\u1fd4-\u1fd5\u1fdc-\u1fdf\u1fed-\u1ff1\u1ff5\u1ffd-\u2125\u2127-\u2129\u212c-\u212d\u212f-\u217f\u2183-\u3006\u3008-\u3020\u302a-\u3040\u3095-\u30a0\u30fb-\u3104\u312d-\u4dff\u9fa6-\uabff\ud7a4-\uffff]')
# Simpler things
nonPubidCharRegexp = re.compile("[^\x20\x0D\x0Aa-zA-Z0-9\-\'()+,./:=?;!*#@$_%]")
class InfosetFilter(object):
replacementRegexp = re.compile(r"U[\dA-F]{5,5}")
def __init__(self, replaceChars=None,
dropXmlnsLocalName=False,
dropXmlnsAttrNs=False,
preventDoubleDashComments=False,
preventDashAtCommentEnd=False,
replaceFormFeedCharacters=True,
preventSingleQuotePubid=False):
self.dropXmlnsLocalName = dropXmlnsLocalName
self.dropXmlnsAttrNs = dropXmlnsAttrNs
self.preventDoubleDashComments = preventDoubleDashComments
self.preventDashAtCommentEnd = preventDashAtCommentEnd
self.replaceFormFeedCharacters = replaceFormFeedCharacters
self.preventSingleQuotePubid = preventSingleQuotePubid
self.replaceCache = {}
def coerceAttribute(self, name, namespace=None):
if self.dropXmlnsLocalName and name.startswith("xmlns:"):
warnings.warn("Attributes cannot begin with xmlns", DataLossWarning)
return None
elif (self.dropXmlnsAttrNs and
namespace == "http://www.w3.org/2000/xmlns/"):
warnings.warn("Attributes cannot be in the xml namespace", DataLossWarning)
return None
else:
return self.toXmlName(name)
def coerceElement(self, name, namespace=None):
return self.toXmlName(name)
def coerceComment(self, data):
if self.preventDoubleDashComments:
while "--" in data:
warnings.warn("Comments cannot contain adjacent dashes", DataLossWarning)
data = data.replace("--", "- -")
return data
def coerceCharacters(self, data):
if self.replaceFormFeedCharacters:
for i in range(data.count("\x0C")):
warnings.warn("Text cannot contain U+000C", DataLossWarning)
data = data.replace("\x0C", " ")
# Other non-xml characters
return data
def coercePubid(self, data):
dataOutput = data
for char in nonPubidCharRegexp.findall(data):
warnings.warn("Coercing non-XML pubid", DataLossWarning)
replacement = self.getReplacementCharacter(char)
dataOutput = dataOutput.replace(char, replacement)
if self.preventSingleQuotePubid and dataOutput.find("'") >= 0:
warnings.warn("Pubid cannot contain single quote", DataLossWarning)
dataOutput = dataOutput.replace("'", self.getReplacementCharacter("'"))
return dataOutput
def toXmlName(self, name):
nameFirst = name[0]
nameRest = name[1:]
m = nonXmlNameFirstBMPRegexp.match(nameFirst)
if m:
warnings.warn("Coercing non-XML name", DataLossWarning)
nameFirstOutput = self.getReplacementCharacter(nameFirst)
else:
nameFirstOutput = nameFirst
nameRestOutput = nameRest
replaceChars = set(nonXmlNameBMPRegexp.findall(nameRest))
for char in replaceChars:
warnings.warn("Coercing non-XML name", DataLossWarning)
replacement = self.getReplacementCharacter(char)
nameRestOutput = nameRestOutput.replace(char, replacement)
return nameFirstOutput + nameRestOutput
def getReplacementCharacter(self, char):
if char in self.replaceCache:
replacement = self.replaceCache[char]
else:
replacement = self.escapeChar(char)
return replacement
def fromXmlName(self, name):
for item in set(self.replacementRegexp.findall(name)):
name = name.replace(item, self.unescapeChar(item))
return name
def escapeChar(self, char):
replacement = "U%05X" % ord(char)
self.replaceCache[char] = replacement
return replacement
def unescapeChar(self, charcode):
return chr(int(charcode[1:], 16))
@@ -0,0 +1,881 @@
from __future__ import absolute_import, division, unicode_literals
from pip._vendor.six import text_type
import codecs
import re
from .constants import EOF, spaceCharacters, asciiLetters, asciiUppercase
from .constants import encodings, ReparseException
from . import utils
from io import StringIO
try:
from io import BytesIO
except ImportError:
BytesIO = StringIO
try:
from io import BufferedIOBase
except ImportError:
class BufferedIOBase(object):
pass
# Non-unicode versions of constants for use in the pre-parser
spaceCharactersBytes = frozenset([item.encode("ascii") for item in spaceCharacters])
asciiLettersBytes = frozenset([item.encode("ascii") for item in asciiLetters])
asciiUppercaseBytes = frozenset([item.encode("ascii") for item in asciiUppercase])
spacesAngleBrackets = spaceCharactersBytes | frozenset([b">", b"<"])
invalid_unicode_re = re.compile("[\u0001-\u0008\u000B\u000E-\u001F\u007F-\u009F\uD800-\uDFFF\uFDD0-\uFDEF\uFFFE\uFFFF\U0001FFFE\U0001FFFF\U0002FFFE\U0002FFFF\U0003FFFE\U0003FFFF\U0004FFFE\U0004FFFF\U0005FFFE\U0005FFFF\U0006FFFE\U0006FFFF\U0007FFFE\U0007FFFF\U0008FFFE\U0008FFFF\U0009FFFE\U0009FFFF\U000AFFFE\U000AFFFF\U000BFFFE\U000BFFFF\U000CFFFE\U000CFFFF\U000DFFFE\U000DFFFF\U000EFFFE\U000EFFFF\U000FFFFE\U000FFFFF\U0010FFFE\U0010FFFF]")
non_bmp_invalid_codepoints = set([0x1FFFE, 0x1FFFF, 0x2FFFE, 0x2FFFF, 0x3FFFE,
0x3FFFF, 0x4FFFE, 0x4FFFF, 0x5FFFE, 0x5FFFF,
0x6FFFE, 0x6FFFF, 0x7FFFE, 0x7FFFF, 0x8FFFE,
0x8FFFF, 0x9FFFE, 0x9FFFF, 0xAFFFE, 0xAFFFF,
0xBFFFE, 0xBFFFF, 0xCFFFE, 0xCFFFF, 0xDFFFE,
0xDFFFF, 0xEFFFE, 0xEFFFF, 0xFFFFE, 0xFFFFF,
0x10FFFE, 0x10FFFF])
ascii_punctuation_re = re.compile("[\u0009-\u000D\u0020-\u002F\u003A-\u0040\u005B-\u0060\u007B-\u007E]")
# Cache for charsUntil()
charsUntilRegEx = {}
class BufferedStream(object):
"""Buffering for streams that do not have buffering of their own
The buffer is implemented as a list of chunks on the assumption that
joining many strings will be slow since it is O(n**2)
"""
def __init__(self, stream):
self.stream = stream
self.buffer = []
self.position = [-1, 0] # chunk number, offset
def tell(self):
pos = 0
for chunk in self.buffer[:self.position[0]]:
pos += len(chunk)
pos += self.position[1]
return pos
def seek(self, pos):
assert pos <= self._bufferedBytes()
offset = pos
i = 0
while len(self.buffer[i]) < offset:
offset -= len(self.buffer[i])
i += 1
self.position = [i, offset]
def read(self, bytes):
if not self.buffer:
return self._readStream(bytes)
elif (self.position[0] == len(self.buffer) and
self.position[1] == len(self.buffer[-1])):
return self._readStream(bytes)
else:
return self._readFromBuffer(bytes)
def _bufferedBytes(self):
return sum([len(item) for item in self.buffer])
def _readStream(self, bytes):
data = self.stream.read(bytes)
self.buffer.append(data)
self.position[0] += 1
self.position[1] = len(data)
return data
def _readFromBuffer(self, bytes):
remainingBytes = bytes
rv = []
bufferIndex = self.position[0]
bufferOffset = self.position[1]
while bufferIndex < len(self.buffer) and remainingBytes != 0:
assert remainingBytes > 0
bufferedData = self.buffer[bufferIndex]
if remainingBytes <= len(bufferedData) - bufferOffset:
bytesToRead = remainingBytes
self.position = [bufferIndex, bufferOffset + bytesToRead]
else:
bytesToRead = len(bufferedData) - bufferOffset
self.position = [bufferIndex, len(bufferedData)]
bufferIndex += 1
rv.append(bufferedData[bufferOffset:bufferOffset + bytesToRead])
remainingBytes -= bytesToRead
bufferOffset = 0
if remainingBytes:
rv.append(self._readStream(remainingBytes))
return b"".join(rv)
def HTMLInputStream(source, encoding=None, parseMeta=True, chardet=True):
if hasattr(source, "read"):
isUnicode = isinstance(source.read(0), text_type)
else:
isUnicode = isinstance(source, text_type)
if isUnicode:
if encoding is not None:
raise TypeError("Cannot explicitly set an encoding with a unicode string")
return HTMLUnicodeInputStream(source)
else:
return HTMLBinaryInputStream(source, encoding, parseMeta, chardet)
class HTMLUnicodeInputStream(object):
"""Provides a unicode stream of characters to the HTMLTokenizer.
This class takes care of character encoding and removing or replacing
incorrect byte-sequences and also provides column and line tracking.
"""
_defaultChunkSize = 10240
def __init__(self, source):
"""Initialises the HTMLInputStream.
HTMLInputStream(source, [encoding]) -> Normalized stream from source
for use by html5lib.
source can be either a file-object, local filename or a string.
The optional encoding parameter must be a string that indicates
the encoding. If specified, that encoding will be used,
regardless of any BOM or later declaration (such as in a meta
element)
parseMeta - Look for a <meta> element containing encoding information
"""
# Craziness
if len("\U0010FFFF") == 1:
self.reportCharacterErrors = self.characterErrorsUCS4
self.replaceCharactersRegexp = re.compile("[\uD800-\uDFFF]")
else:
self.reportCharacterErrors = self.characterErrorsUCS2
self.replaceCharactersRegexp = re.compile("([\uD800-\uDBFF](?![\uDC00-\uDFFF])|(?<![\uD800-\uDBFF])[\uDC00-\uDFFF])")
# List of where new lines occur
self.newLines = [0]
self.charEncoding = ("utf-8", "certain")
self.dataStream = self.openStream(source)
self.reset()
def reset(self):
self.chunk = ""
self.chunkSize = 0
self.chunkOffset = 0
self.errors = []
# number of (complete) lines in previous chunks
self.prevNumLines = 0
# number of columns in the last line of the previous chunk
self.prevNumCols = 0
# Deal with CR LF and surrogates split over chunk boundaries
self._bufferedCharacter = None
def openStream(self, source):
"""Produces a file object from source.
source can be either a file object, local filename or a string.
"""
# Already a file object
if hasattr(source, 'read'):
stream = source
else:
stream = StringIO(source)
return stream
def _position(self, offset):
chunk = self.chunk
nLines = chunk.count('\n', 0, offset)
positionLine = self.prevNumLines + nLines
lastLinePos = chunk.rfind('\n', 0, offset)
if lastLinePos == -1:
positionColumn = self.prevNumCols + offset
else:
positionColumn = offset - (lastLinePos + 1)
return (positionLine, positionColumn)
def position(self):
"""Returns (line, col) of the current position in the stream."""
line, col = self._position(self.chunkOffset)
return (line + 1, col)
def char(self):
""" Read one character from the stream or queue if available. Return
EOF when EOF is reached.
"""
# Read a new chunk from the input stream if necessary
if self.chunkOffset >= self.chunkSize:
if not self.readChunk():
return EOF
chunkOffset = self.chunkOffset
char = self.chunk[chunkOffset]
self.chunkOffset = chunkOffset + 1
return char
def readChunk(self, chunkSize=None):
if chunkSize is None:
chunkSize = self._defaultChunkSize
self.prevNumLines, self.prevNumCols = self._position(self.chunkSize)
self.chunk = ""
self.chunkSize = 0
self.chunkOffset = 0
data = self.dataStream.read(chunkSize)
# Deal with CR LF and surrogates broken across chunks
if self._bufferedCharacter:
data = self._bufferedCharacter + data
self._bufferedCharacter = None
elif not data:
# We have no more data, bye-bye stream
return False
if len(data) > 1:
lastv = ord(data[-1])
if lastv == 0x0D or 0xD800 <= lastv <= 0xDBFF:
self._bufferedCharacter = data[-1]
data = data[:-1]
self.reportCharacterErrors(data)
# Replace invalid characters
# Note U+0000 is dealt with in the tokenizer
data = self.replaceCharactersRegexp.sub("\ufffd", data)
data = data.replace("\r\n", "\n")
data = data.replace("\r", "\n")
self.chunk = data
self.chunkSize = len(data)
return True
def characterErrorsUCS4(self, data):
for i in range(len(invalid_unicode_re.findall(data))):
self.errors.append("invalid-codepoint")
def characterErrorsUCS2(self, data):
# Someone picked the wrong compile option
# You lose
skip = False
for match in invalid_unicode_re.finditer(data):
if skip:
continue
codepoint = ord(match.group())
pos = match.start()
# Pretty sure there should be endianness issues here
if utils.isSurrogatePair(data[pos:pos + 2]):
# We have a surrogate pair!
char_val = utils.surrogatePairToCodepoint(data[pos:pos + 2])
if char_val in non_bmp_invalid_codepoints:
self.errors.append("invalid-codepoint")
skip = True
elif (codepoint >= 0xD800 and codepoint <= 0xDFFF and
pos == len(data) - 1):
self.errors.append("invalid-codepoint")
else:
skip = False
self.errors.append("invalid-codepoint")
def charsUntil(self, characters, opposite=False):
""" Returns a string of characters from the stream up to but not
including any character in 'characters' or EOF. 'characters' must be
a container that supports the 'in' method and iteration over its
characters.
"""
# Use a cache of regexps to find the required characters
try:
chars = charsUntilRegEx[(characters, opposite)]
except KeyError:
if __debug__:
for c in characters:
assert(ord(c) < 128)
regex = "".join(["\\x%02x" % ord(c) for c in characters])
if not opposite:
regex = "^%s" % regex
chars = charsUntilRegEx[(characters, opposite)] = re.compile("[%s]+" % regex)
rv = []
while True:
# Find the longest matching prefix
m = chars.match(self.chunk, self.chunkOffset)
if m is None:
# If nothing matched, and it wasn't because we ran out of chunk,
# then stop
if self.chunkOffset != self.chunkSize:
break
else:
end = m.end()
# If not the whole chunk matched, return everything
# up to the part that didn't match
if end != self.chunkSize:
rv.append(self.chunk[self.chunkOffset:end])
self.chunkOffset = end
break
# If the whole remainder of the chunk matched,
# use it all and read the next chunk
rv.append(self.chunk[self.chunkOffset:])
if not self.readChunk():
# Reached EOF
break
r = "".join(rv)
return r
def unget(self, char):
# Only one character is allowed to be ungotten at once - it must
# be consumed again before any further call to unget
if char is not None:
if self.chunkOffset == 0:
# unget is called quite rarely, so it's a good idea to do
# more work here if it saves a bit of work in the frequently
# called char and charsUntil.
# So, just prepend the ungotten character onto the current
# chunk:
self.chunk = char + self.chunk
self.chunkSize += 1
else:
self.chunkOffset -= 1
assert self.chunk[self.chunkOffset] == char
class HTMLBinaryInputStream(HTMLUnicodeInputStream):
"""Provides a unicode stream of characters to the HTMLTokenizer.
This class takes care of character encoding and removing or replacing
incorrect byte-sequences and also provides column and line tracking.
"""
def __init__(self, source, encoding=None, parseMeta=True, chardet=True):
"""Initialises the HTMLInputStream.
HTMLInputStream(source, [encoding]) -> Normalized stream from source
for use by html5lib.
source can be either a file-object, local filename or a string.
The optional encoding parameter must be a string that indicates
the encoding. If specified, that encoding will be used,
regardless of any BOM or later declaration (such as in a meta
element)
parseMeta - Look for a <meta> element containing encoding information
"""
# Raw Stream - for unicode objects this will encode to utf-8 and set
# self.charEncoding as appropriate
self.rawStream = self.openStream(source)
HTMLUnicodeInputStream.__init__(self, self.rawStream)
self.charEncoding = (codecName(encoding), "certain")
# Encoding Information
# Number of bytes to use when looking for a meta element with
# encoding information
self.numBytesMeta = 512
# Number of bytes to use when using detecting encoding using chardet
self.numBytesChardet = 100
# Encoding to use if no other information can be found
self.defaultEncoding = "windows-1252"
# Detect encoding iff no explicit "transport level" encoding is supplied
if (self.charEncoding[0] is None):
self.charEncoding = self.detectEncoding(parseMeta, chardet)
# Call superclass
self.reset()
def reset(self):
self.dataStream = codecs.getreader(self.charEncoding[0])(self.rawStream,
'replace')
HTMLUnicodeInputStream.reset(self)
def openStream(self, source):
"""Produces a file object from source.
source can be either a file object, local filename or a string.
"""
# Already a file object
if hasattr(source, 'read'):
stream = source
else:
stream = BytesIO(source)
try:
stream.seek(stream.tell())
except:
stream = BufferedStream(stream)
return stream
def detectEncoding(self, parseMeta=True, chardet=True):
# First look for a BOM
# This will also read past the BOM if present
encoding = self.detectBOM()
confidence = "certain"
# If there is no BOM need to look for meta elements with encoding
# information
if encoding is None and parseMeta:
encoding = self.detectEncodingMeta()
confidence = "tentative"
# Guess with chardet, if avaliable
if encoding is None and chardet:
confidence = "tentative"
try:
try:
from charade.universaldetector import UniversalDetector
except ImportError:
from chardet.universaldetector import UniversalDetector
buffers = []
detector = UniversalDetector()
while not detector.done:
buffer = self.rawStream.read(self.numBytesChardet)
assert isinstance(buffer, bytes)
if not buffer:
break
buffers.append(buffer)
detector.feed(buffer)
detector.close()
encoding = detector.result['encoding']
self.rawStream.seek(0)
except ImportError:
pass
# If all else fails use the default encoding
if encoding is None:
confidence = "tentative"
encoding = self.defaultEncoding
# Substitute for equivalent encodings:
encodingSub = {"iso-8859-1": "windows-1252"}
if encoding.lower() in encodingSub:
encoding = encodingSub[encoding.lower()]
return encoding, confidence
def changeEncoding(self, newEncoding):
assert self.charEncoding[1] != "certain"
newEncoding = codecName(newEncoding)
if newEncoding in ("utf-16", "utf-16-be", "utf-16-le"):
newEncoding = "utf-8"
if newEncoding is None:
return
elif newEncoding == self.charEncoding[0]:
self.charEncoding = (self.charEncoding[0], "certain")
else:
self.rawStream.seek(0)
self.reset()
self.charEncoding = (newEncoding, "certain")
raise ReparseException("Encoding changed from %s to %s" % (self.charEncoding[0], newEncoding))
def detectBOM(self):
"""Attempts to detect at BOM at the start of the stream. If
an encoding can be determined from the BOM return the name of the
encoding otherwise return None"""
bomDict = {
codecs.BOM_UTF8: 'utf-8',
codecs.BOM_UTF16_LE: 'utf-16-le', codecs.BOM_UTF16_BE: 'utf-16-be',
codecs.BOM_UTF32_LE: 'utf-32-le', codecs.BOM_UTF32_BE: 'utf-32-be'
}
# Go to beginning of file and read in 4 bytes
string = self.rawStream.read(4)
assert isinstance(string, bytes)
# Try detecting the BOM using bytes from the string
encoding = bomDict.get(string[:3]) # UTF-8
seek = 3
if not encoding:
# Need to detect UTF-32 before UTF-16
encoding = bomDict.get(string) # UTF-32
seek = 4
if not encoding:
encoding = bomDict.get(string[:2]) # UTF-16
seek = 2
# Set the read position past the BOM if one was found, otherwise
# set it to the start of the stream
self.rawStream.seek(encoding and seek or 0)
return encoding
def detectEncodingMeta(self):
"""Report the encoding declared by the meta element
"""
buffer = self.rawStream.read(self.numBytesMeta)
assert isinstance(buffer, bytes)
parser = EncodingParser(buffer)
self.rawStream.seek(0)
encoding = parser.getEncoding()
if encoding in ("utf-16", "utf-16-be", "utf-16-le"):
encoding = "utf-8"
return encoding
class EncodingBytes(bytes):
"""String-like object with an associated position and various extra methods
If the position is ever greater than the string length then an exception is
raised"""
def __new__(self, value):
assert isinstance(value, bytes)
return bytes.__new__(self, value.lower())
def __init__(self, value):
self._position = -1
def __iter__(self):
return self
def __next__(self):
p = self._position = self._position + 1
if p >= len(self):
raise StopIteration
elif p < 0:
raise TypeError
return self[p:p + 1]
def next(self):
# Py2 compat
return self.__next__()
def previous(self):
p = self._position
if p >= len(self):
raise StopIteration
elif p < 0:
raise TypeError
self._position = p = p - 1
return self[p:p + 1]
def setPosition(self, position):
if self._position >= len(self):
raise StopIteration
self._position = position
def getPosition(self):
if self._position >= len(self):
raise StopIteration
if self._position >= 0:
return self._position
else:
return None
position = property(getPosition, setPosition)
def getCurrentByte(self):
return self[self.position:self.position + 1]
currentByte = property(getCurrentByte)
def skip(self, chars=spaceCharactersBytes):
"""Skip past a list of characters"""
p = self.position # use property for the error-checking
while p < len(self):
c = self[p:p + 1]
if c not in chars:
self._position = p
return c
p += 1
self._position = p
return None
def skipUntil(self, chars):
p = self.position
while p < len(self):
c = self[p:p + 1]
if c in chars:
self._position = p
return c
p += 1
self._position = p
return None
def matchBytes(self, bytes):
"""Look for a sequence of bytes at the start of a string. If the bytes
are found return True and advance the position to the byte after the
match. Otherwise return False and leave the position alone"""
p = self.position
data = self[p:p + len(bytes)]
rv = data.startswith(bytes)
if rv:
self.position += len(bytes)
return rv
def jumpTo(self, bytes):
"""Look for the next sequence of bytes matching a given sequence. If
a match is found advance the position to the last byte of the match"""
newPosition = self[self.position:].find(bytes)
if newPosition > -1:
# XXX: This is ugly, but I can't see a nicer way to fix this.
if self._position == -1:
self._position = 0
self._position += (newPosition + len(bytes) - 1)
return True
else:
raise StopIteration
class EncodingParser(object):
"""Mini parser for detecting character encoding from meta elements"""
def __init__(self, data):
"""string - the data to work on for encoding detection"""
self.data = EncodingBytes(data)
self.encoding = None
def getEncoding(self):
methodDispatch = (
(b"<!--", self.handleComment),
(b"<meta", self.handleMeta),
(b"</", self.handlePossibleEndTag),
(b"<!", self.handleOther),
(b"<?", self.handleOther),
(b"<", self.handlePossibleStartTag))
for byte in self.data:
keepParsing = True
for key, method in methodDispatch:
if self.data.matchBytes(key):
try:
keepParsing = method()
break
except StopIteration:
keepParsing = False
break
if not keepParsing:
break
return self.encoding
def handleComment(self):
"""Skip over comments"""
return self.data.jumpTo(b"-->")
def handleMeta(self):
if self.data.currentByte not in spaceCharactersBytes:
# if we have <meta not followed by a space so just keep going
return True
# We have a valid meta element we want to search for attributes
hasPragma = False
pendingEncoding = None
while True:
# Try to find the next attribute after the current position
attr = self.getAttribute()
if attr is None:
return True
else:
if attr[0] == b"http-equiv":
hasPragma = attr[1] == b"content-type"
if hasPragma and pendingEncoding is not None:
self.encoding = pendingEncoding
return False
elif attr[0] == b"charset":
tentativeEncoding = attr[1]
codec = codecName(tentativeEncoding)
if codec is not None:
self.encoding = codec
return False
elif attr[0] == b"content":
contentParser = ContentAttrParser(EncodingBytes(attr[1]))
tentativeEncoding = contentParser.parse()
if tentativeEncoding is not None:
codec = codecName(tentativeEncoding)
if codec is not None:
if hasPragma:
self.encoding = codec
return False
else:
pendingEncoding = codec
def handlePossibleStartTag(self):
return self.handlePossibleTag(False)
def handlePossibleEndTag(self):
next(self.data)
return self.handlePossibleTag(True)
def handlePossibleTag(self, endTag):
data = self.data
if data.currentByte not in asciiLettersBytes:
# If the next byte is not an ascii letter either ignore this
# fragment (possible start tag case) or treat it according to
# handleOther
if endTag:
data.previous()
self.handleOther()
return True
c = data.skipUntil(spacesAngleBrackets)
if c == b"<":
# return to the first step in the overall "two step" algorithm
# reprocessing the < byte
data.previous()
else:
# Read all attributes
attr = self.getAttribute()
while attr is not None:
attr = self.getAttribute()
return True
def handleOther(self):
return self.data.jumpTo(b">")
def getAttribute(self):
"""Return a name,value pair for the next attribute in the stream,
if one is found, or None"""
data = self.data
# Step 1 (skip chars)
c = data.skip(spaceCharactersBytes | frozenset([b"/"]))
assert c is None or len(c) == 1
# Step 2
if c in (b">", None):
return None
# Step 3
attrName = []
attrValue = []
# Step 4 attribute name
while True:
if c == b"=" and attrName:
break
elif c in spaceCharactersBytes:
# Step 6!
c = data.skip()
break
elif c in (b"/", b">"):
return b"".join(attrName), b""
elif c in asciiUppercaseBytes:
attrName.append(c.lower())
elif c is None:
return None
else:
attrName.append(c)
# Step 5
c = next(data)
# Step 7
if c != b"=":
data.previous()
return b"".join(attrName), b""
# Step 8
next(data)
# Step 9
c = data.skip()
# Step 10
if c in (b"'", b'"'):
# 10.1
quoteChar = c
while True:
# 10.2
c = next(data)
# 10.3
if c == quoteChar:
next(data)
return b"".join(attrName), b"".join(attrValue)
# 10.4
elif c in asciiUppercaseBytes:
attrValue.append(c.lower())
# 10.5
else:
attrValue.append(c)
elif c == b">":
return b"".join(attrName), b""
elif c in asciiUppercaseBytes:
attrValue.append(c.lower())
elif c is None:
return None
else:
attrValue.append(c)
# Step 11
while True:
c = next(data)
if c in spacesAngleBrackets:
return b"".join(attrName), b"".join(attrValue)
elif c in asciiUppercaseBytes:
attrValue.append(c.lower())
elif c is None:
return None
else:
attrValue.append(c)
class ContentAttrParser(object):
def __init__(self, data):
assert isinstance(data, bytes)
self.data = data
def parse(self):
try:
# Check if the attr name is charset
# otherwise return
self.data.jumpTo(b"charset")
self.data.position += 1
self.data.skip()
if not self.data.currentByte == b"=":
# If there is no = sign keep looking for attrs
return None
self.data.position += 1
self.data.skip()
# Look for an encoding between matching quote marks
if self.data.currentByte in (b'"', b"'"):
quoteMark = self.data.currentByte
self.data.position += 1
oldPosition = self.data.position
if self.data.jumpTo(quoteMark):
return self.data[oldPosition:self.data.position]
else:
return None
else:
# Unquoted value
oldPosition = self.data.position
try:
self.data.skipUntil(spaceCharactersBytes)
return self.data[oldPosition:self.data.position]
except StopIteration:
# Return the whole remaining value
return self.data[oldPosition:]
except StopIteration:
return None
def codecName(encoding):
"""Return the python codec name corresponding to an encoding or None if the
string doesn't correspond to a valid encoding."""
if isinstance(encoding, bytes):
try:
encoding = encoding.decode("ascii")
except UnicodeDecodeError:
return None
if encoding:
canonicalName = ascii_punctuation_re.sub("", encoding).lower()
return encodings.get(canonicalName, None)
else:
return None
@@ -0,0 +1,271 @@
from __future__ import absolute_import, division, unicode_literals
import re
from xml.sax.saxutils import escape, unescape
from .tokenizer import HTMLTokenizer
from .constants import tokenTypes
class HTMLSanitizerMixin(object):
""" sanitization of XHTML+MathML+SVG and of inline style attributes."""
acceptable_elements = ['a', 'abbr', 'acronym', 'address', 'area',
'article', 'aside', 'audio', 'b', 'big', 'blockquote', 'br', 'button',
'canvas', 'caption', 'center', 'cite', 'code', 'col', 'colgroup',
'command', 'datagrid', 'datalist', 'dd', 'del', 'details', 'dfn',
'dialog', 'dir', 'div', 'dl', 'dt', 'em', 'event-source', 'fieldset',
'figcaption', 'figure', 'footer', 'font', 'form', 'header', 'h1',
'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'i', 'img', 'input', 'ins',
'keygen', 'kbd', 'label', 'legend', 'li', 'm', 'map', 'menu', 'meter',
'multicol', 'nav', 'nextid', 'ol', 'output', 'optgroup', 'option',
'p', 'pre', 'progress', 'q', 's', 'samp', 'section', 'select',
'small', 'sound', 'source', 'spacer', 'span', 'strike', 'strong',
'sub', 'sup', 'table', 'tbody', 'td', 'textarea', 'time', 'tfoot',
'th', 'thead', 'tr', 'tt', 'u', 'ul', 'var', 'video']
mathml_elements = ['maction', 'math', 'merror', 'mfrac', 'mi',
'mmultiscripts', 'mn', 'mo', 'mover', 'mpadded', 'mphantom',
'mprescripts', 'mroot', 'mrow', 'mspace', 'msqrt', 'mstyle', 'msub',
'msubsup', 'msup', 'mtable', 'mtd', 'mtext', 'mtr', 'munder',
'munderover', 'none']
svg_elements = ['a', 'animate', 'animateColor', 'animateMotion',
'animateTransform', 'clipPath', 'circle', 'defs', 'desc', 'ellipse',
'font-face', 'font-face-name', 'font-face-src', 'g', 'glyph', 'hkern',
'linearGradient', 'line', 'marker', 'metadata', 'missing-glyph',
'mpath', 'path', 'polygon', 'polyline', 'radialGradient', 'rect',
'set', 'stop', 'svg', 'switch', 'text', 'title', 'tspan', 'use']
acceptable_attributes = ['abbr', 'accept', 'accept-charset', 'accesskey',
'action', 'align', 'alt', 'autocomplete', 'autofocus', 'axis',
'background', 'balance', 'bgcolor', 'bgproperties', 'border',
'bordercolor', 'bordercolordark', 'bordercolorlight', 'bottompadding',
'cellpadding', 'cellspacing', 'ch', 'challenge', 'char', 'charoff',
'choff', 'charset', 'checked', 'cite', 'class', 'clear', 'color',
'cols', 'colspan', 'compact', 'contenteditable', 'controls', 'coords',
'data', 'datafld', 'datapagesize', 'datasrc', 'datetime', 'default',
'delay', 'dir', 'disabled', 'draggable', 'dynsrc', 'enctype', 'end',
'face', 'for', 'form', 'frame', 'galleryimg', 'gutter', 'headers',
'height', 'hidefocus', 'hidden', 'high', 'href', 'hreflang', 'hspace',
'icon', 'id', 'inputmode', 'ismap', 'keytype', 'label', 'leftspacing',
'lang', 'list', 'longdesc', 'loop', 'loopcount', 'loopend',
'loopstart', 'low', 'lowsrc', 'max', 'maxlength', 'media', 'method',
'min', 'multiple', 'name', 'nohref', 'noshade', 'nowrap', 'open',
'optimum', 'pattern', 'ping', 'point-size', 'poster', 'pqg', 'preload',
'prompt', 'radiogroup', 'readonly', 'rel', 'repeat-max', 'repeat-min',
'replace', 'required', 'rev', 'rightspacing', 'rows', 'rowspan',
'rules', 'scope', 'selected', 'shape', 'size', 'span', 'src', 'start',
'step', 'style', 'summary', 'suppress', 'tabindex', 'target',
'template', 'title', 'toppadding', 'type', 'unselectable', 'usemap',
'urn', 'valign', 'value', 'variable', 'volume', 'vspace', 'vrml',
'width', 'wrap', 'xml:lang']
mathml_attributes = ['actiontype', 'align', 'columnalign', 'columnalign',
'columnalign', 'columnlines', 'columnspacing', 'columnspan', 'depth',
'display', 'displaystyle', 'equalcolumns', 'equalrows', 'fence',
'fontstyle', 'fontweight', 'frame', 'height', 'linethickness', 'lspace',
'mathbackground', 'mathcolor', 'mathvariant', 'mathvariant', 'maxsize',
'minsize', 'other', 'rowalign', 'rowalign', 'rowalign', 'rowlines',
'rowspacing', 'rowspan', 'rspace', 'scriptlevel', 'selection',
'separator', 'stretchy', 'width', 'width', 'xlink:href', 'xlink:show',
'xlink:type', 'xmlns', 'xmlns:xlink']
svg_attributes = ['accent-height', 'accumulate', 'additive', 'alphabetic',
'arabic-form', 'ascent', 'attributeName', 'attributeType',
'baseProfile', 'bbox', 'begin', 'by', 'calcMode', 'cap-height',
'class', 'clip-path', 'color', 'color-rendering', 'content', 'cx',
'cy', 'd', 'dx', 'dy', 'descent', 'display', 'dur', 'end', 'fill',
'fill-opacity', 'fill-rule', 'font-family', 'font-size',
'font-stretch', 'font-style', 'font-variant', 'font-weight', 'from',
'fx', 'fy', 'g1', 'g2', 'glyph-name', 'gradientUnits', 'hanging',
'height', 'horiz-adv-x', 'horiz-origin-x', 'id', 'ideographic', 'k',
'keyPoints', 'keySplines', 'keyTimes', 'lang', 'marker-end',
'marker-mid', 'marker-start', 'markerHeight', 'markerUnits',
'markerWidth', 'mathematical', 'max', 'min', 'name', 'offset',
'opacity', 'orient', 'origin', 'overline-position',
'overline-thickness', 'panose-1', 'path', 'pathLength', 'points',
'preserveAspectRatio', 'r', 'refX', 'refY', 'repeatCount',
'repeatDur', 'requiredExtensions', 'requiredFeatures', 'restart',
'rotate', 'rx', 'ry', 'slope', 'stemh', 'stemv', 'stop-color',
'stop-opacity', 'strikethrough-position', 'strikethrough-thickness',
'stroke', 'stroke-dasharray', 'stroke-dashoffset', 'stroke-linecap',
'stroke-linejoin', 'stroke-miterlimit', 'stroke-opacity',
'stroke-width', 'systemLanguage', 'target', 'text-anchor', 'to',
'transform', 'type', 'u1', 'u2', 'underline-position',
'underline-thickness', 'unicode', 'unicode-range', 'units-per-em',
'values', 'version', 'viewBox', 'visibility', 'width', 'widths', 'x',
'x-height', 'x1', 'x2', 'xlink:actuate', 'xlink:arcrole',
'xlink:href', 'xlink:role', 'xlink:show', 'xlink:title', 'xlink:type',
'xml:base', 'xml:lang', 'xml:space', 'xmlns', 'xmlns:xlink', 'y',
'y1', 'y2', 'zoomAndPan']
attr_val_is_uri = ['href', 'src', 'cite', 'action', 'longdesc', 'poster',
'xlink:href', 'xml:base']
svg_attr_val_allows_ref = ['clip-path', 'color-profile', 'cursor', 'fill',
'filter', 'marker', 'marker-start', 'marker-mid', 'marker-end',
'mask', 'stroke']
svg_allow_local_href = ['altGlyph', 'animate', 'animateColor',
'animateMotion', 'animateTransform', 'cursor', 'feImage', 'filter',
'linearGradient', 'pattern', 'radialGradient', 'textpath', 'tref',
'set', 'use']
acceptable_css_properties = ['azimuth', 'background-color',
'border-bottom-color', 'border-collapse', 'border-color',
'border-left-color', 'border-right-color', 'border-top-color', 'clear',
'color', 'cursor', 'direction', 'display', 'elevation', 'float', 'font',
'font-family', 'font-size', 'font-style', 'font-variant', 'font-weight',
'height', 'letter-spacing', 'line-height', 'overflow', 'pause',
'pause-after', 'pause-before', 'pitch', 'pitch-range', 'richness',
'speak', 'speak-header', 'speak-numeral', 'speak-punctuation',
'speech-rate', 'stress', 'text-align', 'text-decoration', 'text-indent',
'unicode-bidi', 'vertical-align', 'voice-family', 'volume',
'white-space', 'width']
acceptable_css_keywords = ['auto', 'aqua', 'black', 'block', 'blue',
'bold', 'both', 'bottom', 'brown', 'center', 'collapse', 'dashed',
'dotted', 'fuchsia', 'gray', 'green', '!important', 'italic', 'left',
'lime', 'maroon', 'medium', 'none', 'navy', 'normal', 'nowrap', 'olive',
'pointer', 'purple', 'red', 'right', 'solid', 'silver', 'teal', 'top',
'transparent', 'underline', 'white', 'yellow']
acceptable_svg_properties = ['fill', 'fill-opacity', 'fill-rule',
'stroke', 'stroke-width', 'stroke-linecap', 'stroke-linejoin',
'stroke-opacity']
acceptable_protocols = ['ed2k', 'ftp', 'http', 'https', 'irc',
'mailto', 'news', 'gopher', 'nntp', 'telnet', 'webcal',
'xmpp', 'callto', 'feed', 'urn', 'aim', 'rsync', 'tag',
'ssh', 'sftp', 'rtsp', 'afs']
# subclasses may define their own versions of these constants
allowed_elements = acceptable_elements + mathml_elements + svg_elements
allowed_attributes = acceptable_attributes + mathml_attributes + svg_attributes
allowed_css_properties = acceptable_css_properties
allowed_css_keywords = acceptable_css_keywords
allowed_svg_properties = acceptable_svg_properties
allowed_protocols = acceptable_protocols
# Sanitize the +html+, escaping all elements not in ALLOWED_ELEMENTS, and
# stripping out all # attributes not in ALLOWED_ATTRIBUTES. Style
# attributes are parsed, and a restricted set, # specified by
# ALLOWED_CSS_PROPERTIES and ALLOWED_CSS_KEYWORDS, are allowed through.
# attributes in ATTR_VAL_IS_URI are scanned, and only URI schemes specified
# in ALLOWED_PROTOCOLS are allowed.
#
# sanitize_html('<script> do_nasty_stuff() </script>')
# => &lt;script> do_nasty_stuff() &lt;/script>
# sanitize_html('<a href="javascript: sucker();">Click here for $100</a>')
# => <a>Click here for $100</a>
def sanitize_token(self, token):
# accommodate filters which use token_type differently
token_type = token["type"]
if token_type in list(tokenTypes.keys()):
token_type = tokenTypes[token_type]
if token_type in (tokenTypes["StartTag"], tokenTypes["EndTag"],
tokenTypes["EmptyTag"]):
if token["name"] in self.allowed_elements:
return self.allowed_token(token, token_type)
else:
return self.disallowed_token(token, token_type)
elif token_type == tokenTypes["Comment"]:
pass
else:
return token
def allowed_token(self, token, token_type):
if "data" in token:
attrs = dict([(name, val) for name, val in
token["data"][::-1]
if name in self.allowed_attributes])
for attr in self.attr_val_is_uri:
if attr not in attrs:
continue
val_unescaped = re.sub("[`\000-\040\177-\240\s]+", '',
unescape(attrs[attr])).lower()
# remove replacement characters from unescaped characters
val_unescaped = val_unescaped.replace("\ufffd", "")
if (re.match("^[a-z0-9][-+.a-z0-9]*:", val_unescaped) and
(val_unescaped.split(':')[0] not in
self.allowed_protocols)):
del attrs[attr]
for attr in self.svg_attr_val_allows_ref:
if attr in attrs:
attrs[attr] = re.sub(r'url\s*\(\s*[^#\s][^)]+?\)',
' ',
unescape(attrs[attr]))
if (token["name"] in self.svg_allow_local_href and
'xlink:href' in attrs and re.search('^\s*[^#\s].*',
attrs['xlink:href'])):
del attrs['xlink:href']
if 'style' in attrs:
attrs['style'] = self.sanitize_css(attrs['style'])
token["data"] = [[name, val] for name, val in list(attrs.items())]
return token
def disallowed_token(self, token, token_type):
if token_type == tokenTypes["EndTag"]:
token["data"] = "</%s>" % token["name"]
elif token["data"]:
attrs = ''.join([' %s="%s"' % (k, escape(v)) for k, v in token["data"]])
token["data"] = "<%s%s>" % (token["name"], attrs)
else:
token["data"] = "<%s>" % token["name"]
if token.get("selfClosing"):
token["data"] = token["data"][:-1] + "/>"
if token["type"] in list(tokenTypes.keys()):
token["type"] = "Characters"
else:
token["type"] = tokenTypes["Characters"]
del token["name"]
return token
def sanitize_css(self, style):
# disallow urls
style = re.compile('url\s*\(\s*[^\s)]+?\s*\)\s*').sub(' ', style)
# gauntlet
if not re.match("""^([:,;#%.\sa-zA-Z0-9!]|\w-\w|'[\s\w]+'|"[\s\w]+"|\([\d,\s]+\))*$""", style):
return ''
if not re.match("^\s*([-\w]+\s*:[^:;]*(;\s*|$))*$", style):
return ''
clean = []
for prop, value in re.findall("([-\w]+)\s*:\s*([^:;]*)", style):
if not value:
continue
if prop.lower() in self.allowed_css_properties:
clean.append(prop + ': ' + value + ';')
elif prop.split('-')[0].lower() in ['background', 'border', 'margin',
'padding']:
for keyword in value.split():
if not keyword in self.acceptable_css_keywords and \
not re.match("^(#[0-9a-f]+|rgb\(\d+%?,\d*%?,?\d*%?\)?|\d{0,2}\.?\d{0,2}(cm|em|ex|in|mm|pc|pt|px|%|,|\))?)$", keyword):
break
else:
clean.append(prop + ': ' + value + ';')
elif prop.lower() in self.allowed_svg_properties:
clean.append(prop + ': ' + value + ';')
return ' '.join(clean)
class HTMLSanitizer(HTMLTokenizer, HTMLSanitizerMixin):
def __init__(self, stream, encoding=None, parseMeta=True, useChardet=True,
lowercaseElementName=False, lowercaseAttrName=False, parser=None):
# Change case matching defaults as we only output lowercase html anyway
# This solution doesn't seem ideal...
HTMLTokenizer.__init__(self, stream, encoding, parseMeta, useChardet,
lowercaseElementName, lowercaseAttrName, parser=parser)
def __iter__(self):
for token in HTMLTokenizer.__iter__(self):
token = self.sanitize_token(token)
if token:
yield token

Some files were not shown because too many files have changed in this diff Show More