apt Update signature fail + info in thread (IMPORTANT
Hello there,
I’ve been doing r3v.3r.se 3ngin.3ri.ng and d1.gi.t@l F0rE.ns1c.s for quite a while… Started with $tuxN3t back in the 200-9.
I am fl@gg3.d or someone got me really good and I am working on current 3xpl01.ts aff3cting BIOS (may be hardware for BIOS), Unix-based systems Linux and MacOS (some systems just take longer for the migration to complete), iOS and Android. In my book it’s enought to raise eyebrows and if this is not completely cross-platform I am definitely fl@gg3.d. This is why I try to reduce identifiable k3y.words to a minimum using the most basic type of plain text cryp.t. I need help now after 3 1nfect1ng iOS, 1 MacOS, 4 laptops, 2 desktops. Help needed at j.s.m.i.o.u.s.s.e@agi.sky.net (remove all dots in the first part of the mail and the dot between agi and sky)
Now for the update signature fail bug and sorry I cannot upload files since I can’t use the commandline and permissions and denied using the GUI…
@
h3. amnesia@amnesia:~$ sudo apt update && sudo apt upgrade
We trust you have received the usual lecture from the local System
Administrator. It usually boils down to these three things:
#1) Respect the privacy of others.
#2) Think before you type.
#3) With great power comes great responsibility.
[sudo] password for amnesia:
Get:1 tor+http://sdscoq7snqtznauu.onion/torproject.org stretch InRelease [4,965 B]
Get:2 tor+http://sdscoq7snqtznauu.onion/torproject.org stretch/main amd64 Packages [3,496 B]
Get:3 tor+http://jenw7xbd6tf7vfhp.onion 3.11 InRelease [6,858 B]
Err:3 tor+http://jenw7xbd6tf7vfhp.onion 3.11 InRelease
The following signatures were invalid: EXPKEYSIG C7988EA7A358D82E deb.tails.boum.org archive signing key
Get:4 tor+http://sgvtcaew4bxjd7ln.onion stretch/updates InRelease [94.3 kB]
Get:5 tor+http://sgvtcaew4bxjd7ln.onion stretch/updates/main amd64 Packages [487 kB]
Ign:6 tor+http://vwakviie2ienjx6t.onion/debian stretch InRelease
Get:7 tor+http://vwakviie2ienjx6t.onion/debian sid InRelease [247 kB]
Get:8 tor+http://sgvtcaew4bxjd7ln.onion stretch/updates/main Translation-en [216 kB]
Get:9 tor+http://sgvtcaew4bxjd7ln.onion stretch/updates/contrib amd64 Packages [1,760 B]
Get:10 tor+http://sgvtcaew4bxjd7ln.onion stretch/updates/contrib Translation-en [1,759 B]
Get:11 tor+http://vwakviie2ienjx6t.onion/debian stretch-backports InRelease [91.8 kB]
Get:12 tor+http://vwakviie2ienjx6t.onion/debian stretch Release [118 kB]
Get:13 tor+http://vwakviie2ienjx6t.onion/debian sid/main amd64 Packages [8,283 kB]
Get:14 tor+http://vwakviie2ienjx6t.onion/debian sid/main Translation-en [6,320 kB]
Get:15 tor+http://vwakviie2ienjx6t.onion/debian sid/contrib amd64 Packages [60.1 kB]
Get:16 tor+http://vwakviie2ienjx6t.onion/debian sid/contrib Translation-en [50.4 kB]
Get:17 tor+http://vwakviie2ienjx6t.onion/debian stretch-backports/main amd64 Packages [593 kB]
Get:18 tor+http://vwakviie2ienjx6t.onion/debian stretch-backports/main Translation-en [454 kB]
Get:19 tor+http://vwakviie2ienjx6t.onion/debian stretch-backports/contrib amd64 Packages [11.1 kB]
Get:20 tor+http://vwakviie2ienjx6t.onion/debian stretch-backports/contrib Translation-en [7,540 B]
Get:21 tor+http://vwakviie2ienjx6t.onion/debian stretch Release.gpg [2,434 B]
Get:22 tor+http://vwakviie2ienjx6t.onion/debian stretch/main amd64 Packages [7,082 kB]
Get:23 tor+http://vwakviie2ienjx6t.onion/debian stretch/main Translation-en [5,384 kB]
Get:24 tor+http://vwakviie2ienjx6t.onion/debian stretch/contrib amd64 Packages [50.9 kB]
Get:25 tor+http://vwakviie2ienjx6t.onion/debian stretch/contrib Translation-en [45.9 kB]
Reading package lists... Done
W: GPG error: tor+http://jenw7xbd6tf7vfhp.onion 3.11 InRelease: The following signatures were invalid: EXPKEYSIG C7988EA7A358D82E deb.tails.boum.org archive signing key
E: The repository 'tor+http://jenw7xbd6tf7vfhp.onion 3.11 InRelease' is not signed.
N: Updating from such a repository can't be done securely, and is therefore disabled by default.
N: See apt-secure(8) manpage for repository creation and user configuration details.
amnesia@amnesia:~$@ Build information:
3.11 - 20181210 (though I made this key not so long ago… anyway)
aa5bd4c38cc82abd590ba4a956ab441937f6b306
live-build: 3.0.5+really+is+2.0.12-0.tails5
live-boot: 1:20170112
live-config: 5.20170112+deb9u1
I found this nice python script and a more elaborated remote upgrader for people like me following doesn’t seem to be tails programming @ _datasource.py
<code class="python">
from __future__ import absolute_import, print_function
import glob
import logging
import os.path
import re
import shutil
import time
import apt_pkg
from .distinfo import DistInfo
#from apt_pkg import gettext as _
# some global helpers
__all__ = ['is_mirror', 'SourceEntry', 'NullMatcher', 'SourcesList',
'SourceEntryMatcher']
def is_mirror(master_uri, compare_uri):
""" check if the given add_url is idential or a mirror of orig_uri e.g.:
master_uri = archive.ubuntu.com
compare_uri = de.archive.ubuntu.com
-> True
"""
# remove traling spaces and "/"
compare_uri = compare_uri.rstrip("/ ")
master_uri = master_uri.rstrip("/ ")
# uri is identical
if compare_uri == master_uri:
#print "Identical"
return True
# add uri is a master site and orig_uri has the from "XX.mastersite"
# (e.g. de.archive.ubuntu.com)
try:
compare_srv = compare_uri.split("//")[1]
master_srv = master_uri.split("//")[1]
#print "%s == %s " % (add_srv, orig_srv)
except IndexError: # ok, somethings wrong here
#print "IndexError"
return False
# remove the leading "<country>." (if any) and see if that helps
if "." in compare_srv and \
compare_srv[compare_srv.index(".") + 1:] == master_srv:
#print "Mirror"
return True
return False
def uniq(s):
""" simple and efficient way to return uniq collection
This is not intended for use with a SourceList. It is provided
for internal use only. It does not have a leading underscore to
not break any old code that uses it; but it should not be used
in new code (and is not listed in __all__)."""
return list(set(s))
class SourceEntry(object):
""" single sources.list entry """
def __init__(self, line, file=None):
self.invalid = False # is the source entry valid
self.disabled = False # is it disabled ('#' in front)
self.type = "" # what type (deb, deb-src)
self.architectures = [] # architectures
self.trusted = None # Trusted
self.uri = "" # base-uri
self.dist = "" # distribution (dapper, edgy, etc)
self.comps = [] # list of available componetns (may empty)
self.comment = "" # (optional) comment
self.line = line # the original sources.list line
if file is None:
file = apt_pkg.config.find_dir(
"Dir::Etc") + apt_pkg.config.find("Dir::Etc::sourcelist")
self.file = file # the file that the entry is located in
self.parse(line)
self.template = None # type DistInfo.Suite
self.children = []
def __eq__(self, other):
""" equal operator for two sources.list entries """
return (self.disabled == other.disabled and
self.type == other.type and
self.uri == other.uri and
self.dist == other.dist and
self.comps == other.comps)
def mysplit(self, line):
""" a split() implementation that understands the sources.list
format better and takes [] into account (for e.g. cdroms) """
line = line.strip()
pieces = []
tmp = ""
# we are inside a [..] block
p_found = False
space_found = False
for i in range(len(line)):
if line[i] == "[":
if space_found:
space_found = False
p_found = True
pieces.append(tmp)
tmp = line[i]
else:
p_found = True
tmp += line[i]
elif line[i] == "]":
p_found = False
tmp += line[i]
elif space_found and not line[i].isspace():
# we skip one or more space
space_found = False
pieces.append(tmp)
tmp = line[i]
elif line[i].isspace() and not p_found:
# found a whitespace
space_found = True
else:
tmp += line[i]
# append last piece
if len(tmp) > 0:
pieces.append(tmp)
return pieces
def parse(self, line):
""" parse a given sources.list (textual) line and break it up
into the field we have """
line = self.line.strip()
#print line
# check if the source is enabled/disabled
if line == "" or line == "#": # empty line
self.invalid = True
return
if line[0] == "#":
self.disabled = True
pieces = line[1:].strip().split()
# if it looks not like a disabled deb line return
if not pieces[0] in ("rpm", "rpm-src", "deb", "deb-src"):
self.invalid = True
return
else:
line = line[1:]
# check for another "#" in the line (this is treated as a comment)
i = line.find("#")
if i > 0:
self.comment = line[i + 1:]
line = line[:i]
# source is ok, split it and see what we have
pieces = self.mysplit(line)
# Sanity check
if len(pieces) < 3:
self.invalid = True
return
# Type, deb or deb-src
self.type = pieces[0].strip()
# Sanity check
if self.type not in ("deb", "deb-src", "rpm", "rpm-src"):
-----> self.invalid = True
return
if pieces[1].strip()[0] == "[":
options = pieces.pop(1).strip("[]").split()
for option in options:
try:
key, value = option.split("=", 1)
except Exception:
self.invalid = True
else:
if key == "arch":
self.architectures = value.split(",")
elif key == "trusted":
self.trusted = apt_pkg.string_to_bool(value)
else:
self.invalid = True
# URI
self.uri = pieces[1].strip()
if len(self.uri) < 1:
self.invalid = True
# distro and components (optional)
# Directory or distro
self.dist = pieces[2].strip()
if len(pieces) > 3:
# List of components
self.comps = pieces[3:]
else:
self.comps = []
def set_enabled(self, new_value):
""" set a line to enabled or disabled """
self.disabled = not new_value
# enable, remove all "#" from the start of the line
if new_value:
self.line = self.line.lstrip().lstrip('#')
else:
# disabled, add a "#"
if self.line.strip()[0] != "#":
self.line = "#" + self.line
def __str__(self):
""" debug helper """
return self.str().strip()
def str(self):
""" return the current line as string """
if self.invalid:
return self.line
line = ""
if self.disabled:
line = "# "
line += self.type
if self.architectures and self.trusted is not None:
line += " [arch=%s trusted=%s]" % (
",".join(self.architectures), "yes" if self.trusted else "no")
elif self.trusted is not None:
line += " [trusted=%s]" % ("yes" if self.trusted else "no")
elif self.architectures:
line += " [arch=%s]" % ",".join(self.architectures)
line += " %s %s" % (self.uri, self.dist)
if len(self.comps) > 0:
line += " " + " ".join(self.comps)
if self.comment != "":
line += " #" + self.comment
line += "\n"
return line
class NullMatcher(object):
""" a Matcher that does nothing """
def match(self, s):
return True
class SourcesList(object):
""" represents the full sources.list + sources.list.d file """
def __init__(self,
withMatcher=True,
matcherPath="/usr/share/python-apt/templates/"):
self.list = [] # the actual SourceEntries Type
if withMatcher:
self.matcher = SourceEntryMatcher(matcherPath)
else:
self.matcher = NullMatcher()
self.refresh()
def refresh(self):
""" update the list of known entries """
self.list = []
# read sources.list
file = apt_pkg.config.find_file("Dir::Etc::sourcelist")
self.load(file)
# read sources.list.d
partsdir = apt_pkg.config.find_dir("Dir::Etc::sourceparts")
for file in glob.glob("%s/*.list" % partsdir):
self.load(file)
# check if the source item fits a predefined template
for source in self.list:
if not source.invalid:
self.matcher.match(source)
def __iter__(self):
""" simple iterator to go over self.list, returns SourceEntry
types """
for entry in self.list:
yield entry
def __find(self, *predicates, **attrs):
for source in self.list:
if (all(getattr(source, key) == attrs[key] for key in attrs) and
all(predicate(source) for predicate in predicates)):
yield source
def add(self, type, uri, dist, orig_comps, comment="", pos=-1, file=None,
architectures=[]):
"""
Add a new source to the sources.list.
The method will search for existing matching repos and will try to
reuse them as far as possible
"""
architectures = set(architectures)
# create a working copy of the component list so that
# we can modify it later
comps = orig_comps[:]
sources = self.__find(lambda s: set(s.architectures) == architectures,
disabled=False, invalid=False, type=type,
uri=uri, dist=dist)
# check if we have this source already in the sources.list
for source in sources:
for new_comp in comps:
if new_comp in source.comps:
# we have this component already, delete it
# from the new_comps list
del comps[comps.index(new_comp)]
if len(comps) == 0:
return source
sources = self.__find(lambda s: set(s.architectures) == architectures,
invalid=False, type=type, uri=uri, dist=dist)
for source in sources:
# if there is a repo with the same (type, uri, dist) just add the
# components
if source.disabled and set(source.comps) == set(comps):
source.disabled = False
return source
elif not source.disabled:
source.comps = uniq(source.comps + comps)
return source
# there isn't any matching source, so create a new line and parse it
line = type
if architectures:
line += " [arch=%s]" % ",".join(architectures)
line += " %s %s" % (uri, dist)
for c in comps:
line = line + " " + c
if comment != "":
line = "%s #%s\n" % (line, comment)
line = line + "\n"
new_entry = SourceEntry(line)
if file is not None:
new_entry.file = file
self.matcher.match(new_entry)
self.list.insert(pos, new_entry)
return new_entry
def remove(self, source_entry):
""" remove the specified entry from the sources.list """
self.list.remove(source_entry)
def restore_backup(self, backup_ext):
" restore sources.list files based on the backup extension "
file = apt_pkg.config.find_file("Dir::Etc::sourcelist")
if os.path.exists(file + backup_ext) and os.path.exists(file):
shutil.copy(file + backup_ext, file)
# now sources.list.d
partsdir = apt_pkg.config.find_dir("Dir::Etc::sourceparts")
for file in glob.glob("%s/*.list" % partsdir):
if os.path.exists(file + backup_ext):
shutil.copy(file + backup_ext, file)
def backup(self, backup_ext=None):
""" make a backup of the current source files, if no backup extension
is given, the current date/time is used (and returned) """
already_backuped = set()
if backup_ext is None:
backup_ext = time.strftime("%y%m%d.%H%M")
for source in self.list:
if (source.file not in already_backuped and
os.path.exists(source.file)):
shutil.copy(source.file, "%s%s" % (source.file, backup_ext))
return backup_ext
def load(self, file):
""" (re)load the current sources """
try:
with open(file, "r") as f:
for line in f:
source = SourceEntry(line, file)
self.list.append(source)
except:
logging.warning("could not open file '%s'\n" % file)
def save(self):
""" save the current sources """
files = {}
# write an empty default config file if there aren't any sources
if len(self.list) == 0:
path = apt_pkg.config.find_file("Dir::Etc::sourcelist")
header = (
"## See sources.list(5) for more information, especialy\n"
"# Remember that you can only use http, ftp or file URIs\n"
"# CDROMs are managed through the apt-cdrom tool.\n")
with open(path, "w") as f:
f.write(header)
return
try:
for source in self.list:
if source.file not in files:
files[source.file] = open(source.file, "w")
files[source.file].write(source.str())
finally:
for f in files:
files[f].close()
def check_for_relations(self, sources_list):
"""get all parent and child channels in the sources list"""
parents = []
used_child_templates = {}
for source in sources_list:
# try to avoid checking uninterressting sources
if source.template is None:
continue
# set up a dict with all used child templates and corresponding
# source entries
if source.template.child:
key = source.template
if key not in used_child_templates:
used_child_templates[key] = []
temp = used_child_templates[key]
temp.append(source)
else:
# store each source with children aka. a parent :)
if len(source.template.children) > 0:
parents.append(source)
#print self.used_child_templates
#print self.parents
return (parents, used_child_templates)
class SourceEntryMatcher(object):
""" matcher class to make a source entry look nice
lots of predefined matchers to make it i18n/gettext friendly
"""
def __init__(self, matcherPath):
self.templates = []
# Get the human readable channel and comp names from the channel .infos
spec_files = glob.glob("%s/*.info" % matcherPath)
for f in spec_files:
f = os.path.basename(f)
i = f.find(".info")
f = f[0:i]
dist = DistInfo(f, base_dir=matcherPath)
for template in dist.templates:
if template.match_uri is not None:
self.templates.append(template)
return
def match(self, source):
"""Add a matching template to the source"""
found = False
for template in self.templates:
if (re.search(template.match_uri, source.uri) and
re.match(template.match_name, source.dist) and
# deb is a valid fallback for deb-src (if that is not
# definied, see #760035
(source.type == template.type or template.type == "deb")):
found = True
source.template = template
break
elif (template.is_mirror(source.uri) and
re.match(template.match_name, source.dist)):
found = True
source.template = template
break
return found
# some simple tests
if __name__ == "__main__":
apt_pkg.init_config()
sources = SourcesList()
for entry in sources:
logging.info("entry %s" % entry.str())
#print entry.uri
mirror = is_mirror("http://archive.ubuntu.com/ubuntu/",
"http://de.archive.ubuntu.com/ubuntu/")
logging.info("is_mirror(): %s" % mirror)
logging.info(is_mirror("http://archive.ubuntu.com/ubuntu",
"http://de.archive.ubuntu.com/ubuntu/"))
logging.info(is_mirror("http://archive.ubuntu.com/ubuntu/",
"http://de.archive.ubuntu.com/ubuntu"))
</code><
the other one I find really interesting is this one
<pre><code class="python">
from __future__ import division, absolute_import, print_function
import os
import sys
import shutil
_open = open
# Using a class instead of a module-level dictionary
# to reduce the inital 'import numpy' overhead by
# deferring the import of bz2 and gzip until needed
# TODO: .zip support, .tar support?
class _FileOpeners(object):
"""
Container for different methods to open (un-)compressed files.
`_FileOpeners` contains a dictionary that holds one method for each
supported file format. Attribute lookup is implemented in such a way
that an instance of `_FileOpeners` itself can be indexed with the keys
of that dictionary. Currently uncompressed files as well as files
compressed with ``gzip`` or ``bz2`` compression are supported.
Notes
-----
`_file_openers`, an instance of `_FileOpeners`, is made available for
use in the `_datasource` module.
Examples
--------
>>> np.lib._datasource._file_openers.keys()
[None, '.bz2', '.gz']
>>> np.lib._datasource._file_openers['.gz'] is gzip.open
True
"""
def __init__(self):
self._loaded = False
self._file_openers = {None: open}
def _load(self):
if self._loaded:
return
try:
import bz2
self._file_openers[".bz2"] = bz2.BZ2File
except ImportError:
pass
try:
import gzip
self._file_openers[".gz"] = gzip.open
except ImportError:
pass
self._loaded = True
def keys(self):
"""
Return the keys of currently supported file openers.
Parameters
----------
None
Returns
-------
keys : list
The keys are None for uncompressed files and the file extension
strings (i.e. ``'.gz'``, ``'.bz2'``) for supported compression
methods.
"""
self._load()
return list(self._file_openers.keys())
def __getitem__(self, key):
self._load()
return self._file_openers[key]
_file_openers = _FileOpeners()
def open(path, mode='r', destpath=os.curdir):
"""
Open `path` with `mode` and return the file object.
If ``path`` is an URL, it will be downloaded, stored in the
`DataSource` `destpath` directory and opened from there.
Parameters
----------
path : str
Local file path or URL to open.
mode : str, optional
Mode to open `path`. Mode 'r' for reading, 'w' for writing, 'a' to
append. Available modes depend on the type of object specified by
path. Default is 'r'.
destpath : str, optional
Path to the directory where the source file gets downloaded to for
use. If `destpath` is None, a temporary directory will be created.
The default path is the current directory.
Returns
-------
out : file object
The opened file.
Notes
-----
This is a convenience function that instantiates a `DataSource` and
returns the file object from ``DataSource.open(path)``.
"""
ds = DataSource(destpath)
return ds.open(path, mode)
class DataSource (object):
"""
DataSource(destpath='.')
A generic data source file (file, http, ftp, ...).
DataSources can be local files or remote files/URLs. The files may
also be compressed or uncompressed. DataSource hides some of the
low-level details of downloading the file, allowing you to simply pass
in a valid file path (or URL) and obtain a file object.
Parameters
----------
destpath : str or None, optional
Path to the directory where the source file gets downloaded to for
use. If `destpath` is None, a temporary directory will be created.
The default path is the current directory.
Notes
-----
URLs require a scheme string (``http://``) to be used, without it they
will fail::
>>> repos = DataSource()
>>> repos.exists('www.google.com/index.html')
False
>>> repos.exists('http://www.google.com/index.html')
True
Temporary directories are deleted when the DataSource is deleted.
Examples
--------
::
>>> ds = DataSource('/home/guido')
>>> urlname = 'http://www.google.com/index.html'
>>> gfile = ds.open('http://www.google.com/index.html') # remote file
>>> ds.abspath(urlname)
'/home/guido/www.google.com/site/index.html'
>>> ds = DataSource(None) # use with temporary file
>>> ds.open('/home/guido/foobar.txt')
<open file '/home/guido.foobar.txt', mode 'r' at 0x91d4430>
>>> ds.abspath('/home/guido/foobar.txt')
'/tmp/tmpy4pgsP/home/guido/foobar.txt'
"""
def __init__(self, destpath=os.curdir):
"""Create a DataSource with a local path at destpath."""
if destpath:
self._destpath = os.path.abspath(destpath)
self._istmpdest = False
else:
import tempfile # deferring import to improve startup time
self._destpath = tempfile.mkdtemp()
self._istmpdest = True
def __del__(self):
# Remove temp directories
if self._istmpdest:
shutil.rmtree(self._destpath)
def _iszip(self, filename):
"""Test if the filename is a zip file by looking at the file extension.
"""
fname, ext = os.path.splitext(filename)
return ext in _file_openers.keys()
def _iswritemode(self, mode):
"""Test if the given mode will open a file for writing."""
# Currently only used to test the bz2 files.
_writemodes = ("w", "+")
for c in mode:
if c in _writemodes:
return True
return False
def _splitzipext(self, filename):
"""Split zip extension from filename and return filename.
*Returns*:
base, zip_ext : {tuple}
"""
if self._iszip(filename):
return os.path.splitext(filename)
else:
return filename, None
def _possible_names(self, filename):
"""Return a tuple containing compressed filename variations."""
names = [filename]
if not self._iszip(filename):
for zipext in _file_openers.keys():
if zipext:
names.append(filename+zipext)
return names
def _isurl(self, path):
"""Test if path is a net location. Tests the scheme and netloc."""
# We do this here to reduce the 'import numpy' initial import time.
if sys.version_info[0] >= 3:
from urllib.parse import urlparse
else:
from urlparse import urlparse
# BUG : URLs require a scheme string ('http://') to be used.
# www.google.com will fail.
# Should we prepend the scheme for those that don't have it and
# test that also? Similar to the way we append .gz and test for
# for compressed versions of files.
scheme, netloc, upath, uparams, uquery, ufrag = urlparse(path)
return bool(scheme and netloc)
def _cache(self, path):
"""Cache the file specified by path.
Creates a copy of the file in the datasource cache.
"""
# We import these here because importing urllib2 is slow and
# a significant fraction of numpy's total import time.
if sys.version_info[0] >= 3:
from urllib.request import urlopen
from urllib.error import URLError
else:
from urllib2 import urlopen
from urllib2 import URLError
upath = self.abspath(path)
# ensure directory exists
if not os.path.exists(os.path.dirname(upath)):
os.makedirs(os.path.dirname(upath))
# TODO: Doesn't handle compressed files!
if self._isurl(path):
try:
openedurl = urlopen(path)
f = _open(upath, 'wb')
try:
shutil.copyfileobj(openedurl, f)
finally:
f.close()
openedurl.close()
except URLError:
raise URLError("URL not found: %s" % path)
else:
shutil.copyfile(path, upath)
return upath
def _findfile(self, path):
"""Searches for ``path`` and returns full path if found.
If path is an URL, _findfile will cache a local copy and return the
path to the cached file. If path is a local file, _findfile will
return a path to that local file.
The search will include possible compressed versions of the file
and return the first occurrence found.
"""
# Build list of possible local file paths
if not self._isurl(path):
# Valid local paths
filelist = self._possible_names(path)
# Paths in self._destpath
filelist += self._possible_names(self.abspath(path))
else:
# Cached URLs in self._destpath
filelist = self._possible_names(self.abspath(path))
# Remote URLs
filelist = filelist + self._possible_names(path)
for name in filelist:
if self.exists(name):
if self._isurl(name):
name = self._cache(name)
return name
return None
def abspath(self, path):
"""
Return absolute path of file in the DataSource directory.
If `path` is an URL, then `abspath` will return either the location
the file exists locally or the location it would exist when opened
using the `open` method.
Parameters
----------
path : str
Can be a local file or a remote URL.
Returns
-------
out : str
Complete path, including the `DataSource` destination directory.
Notes
-----
The functionality is based on `os.path.abspath`.
"""
# We do this here to reduce the 'import numpy' initial import time.
if sys.version_info[0] >= 3:
from urllib.parse import urlparse
else:
from urlparse import urlparse
# TODO: This should be more robust. Handles case where path includes
# the destpath, but not other sub-paths. Failing case:
# path = /home/guido/datafile.txt
# destpath = /home/alex/
# upath = self.abspath(path)
# upath == '/home/alex/home/guido/datafile.txt'
# handle case where path includes self._destpath
splitpath = path.split(self._destpath, 2)
if len(splitpath) > 1:
path = splitpath[1]
scheme, netloc, upath, uparams, uquery, ufrag = urlparse(path)
netloc = self._sanitize_relative_path(netloc)
upath = self._sanitize_relative_path(upath)
return os.path.join(self._destpath, netloc, upath)
def _sanitize_relative_path(self, path):
"""Return a sanitised relative path for which
os.path.abspath(os.path.join(base, path)).startswith(base)
"""
last = None
path = os.path.normpath(path)
while path != last:
last = path
# Note: os.path.join treats '/' as os.sep on Windows
path = path.lstrip(os.sep).lstrip('/')
path = path.lstrip(os.pardir).lstrip('..')
drive, path = os.path.splitdrive(path) # for Windows
return path
def exists(self, path):
"""
Test if path exists.
Test if `path` exists as (and in this order):
- a local file.
- a remote URL that has been downloaded and stored locally in the
`DataSource` directory.
- a remote URL that has not been downloaded, but is valid and
accessible.
Parameters
----------
path : str
Can be a local file or a remote URL.
Returns
-------
out : bool
True if `path` exists.
Notes
-----
When `path` is an URL, `exists` will return True if it's either
stored locally in the `DataSource` directory, or is a valid remote
URL. `DataSource` does not discriminate between the two, the file
is accessible if it exists in either location.
"""
# We import this here because importing urllib2 is slow and
# a significant fraction of numpy's total import time.
if sys.version_info[0] >= 3:
from urllib.request import urlopen
from urllib.error import URLError
else:
from urllib2 import urlopen
from urllib2 import URLError
# Test local path
if os.path.exists(path):
return True
# Test cached url
upath = self.abspath(path)
if os.path.exists(upath):
return True
# Test remote url
if self._isurl(path):
try:
netfile = urlopen(path)
netfile.close()
del(netfile)
return True
except URLError:
return False
return False
def open(self, path, mode='r'):
"""
Open and return file-like object.
If `path` is an URL, it will be downloaded, stored in the
`DataSource` directory and opened from there.
Parameters
----------
path : str
Local file path or URL to open.
mode : {'r', 'w', 'a'}, optional
Mode to open `path`. Mode 'r' for reading, 'w' for writing,
'a' to append. Available modes depend on the type of object
specified by `path`. Default is 'r'.
Returns
-------
out : file object
File object.
"""
# TODO: There is no support for opening a file for writing which
# doesn't exist yet (creating a file). Should there be?
# TODO: Add a ``subdir`` parameter for specifying the subdirectory
# used to store URLs in self._destpath.
if self._isurl(path) and self._iswritemode(mode):
raise ValueError("URLs are not writeable")
# NOTE: _findfile will fail on a new file opened for writing.
found = self._findfile(path)
if found:
_fname, ext = self._splitzipext(found)
if ext == 'bz2':
mode.replace("+", "")
return _file_openers[ext](found, mode=mode)
else:
raise IOError("%s not found." % path)
class Repository (DataSource):
"""
Repository(baseurl, destpath='.')
A data repository where multiple DataSource's share a base
URL/directory.
`Repository` extends `DataSource` by prepending a base URL (or
directory) to all the files it handles. Use `Repository` when you will
be working with multiple files from one base URL. Initialize
`Repository` with the base URL, then refer to each file by its filename
only.
Parameters
----------
baseurl : str
Path to the local directory or remote location that contains the
data files.
destpath : str or None, optional
Path to the directory where the source file gets downloaded to for
use. If `destpath` is None, a temporary directory will be created.
The default path is the current directory.
Examples
--------
To analyze all files in the repository, do something like this
(note: this is not self-contained code)::
>>> repos = np.lib._datasource.Repository('/home/user/data/dir/')
>>> for filename in filelist:
... fp = repos.open(filename)
... fp.analyze()
... fp.close()
Similarly you could use a URL for a repository::
>>> repos = np.lib._datasource.Repository('http://www.xyz.edu/data')
"""
def __init__(self, baseurl, destpath=os.curdir):
"""Create a Repository with a shared url or directory of baseurl."""
DataSource.__init__(self, destpath=destpath)
self._baseurl = baseurl
def __del__(self):
DataSource.__del__(self)
def _fullpath(self, path):
"""Return complete path for path. Prepends baseurl if necessary."""
splitpath = path.split(self._baseurl, 2)
if len(splitpath) == 1:
result = os.path.join(self._baseurl, path)
else:
result = path # path contains baseurl already
return result
def _findfile(self, path):
"""Extend DataSource method to prepend baseurl to ``path``."""
return DataSource._findfile(self, self._fullpath(path))
def abspath(self, path):
"""
Return absolute path of file in the Repository directory.
If `path` is an URL, then `abspath` will return either the location
the file exists locally or the location it would exist when opened
using the `open` method.
Parameters
----------
path : str
Can be a local file or a remote URL. This may, but does not
have to, include the `baseurl` with which the `Repository` was
initialized.
Returns
-------
out : str
Complete path, including the `DataSource` destination directory.
"""
return DataSource.abspath(self, self._fullpath(path))
def exists(self, path):
"""
Test if path exists prepending Repository base URL to path.
Test if `path` exists as (and in this order):
- a local file.
- a remote URL that has been downloaded and stored locally in the
`DataSource` directory.
- a remote URL that has not been downloaded, but is valid and
accessible.
Parameters
----------
path : str
Can be a local file or a remote URL. This may, but does not
have to, include the `baseurl` with which the `Repository` was
initialized.
Returns
-------
out : bool
True if `path` exists.
Notes
-----
When `path` is an URL, `exists` will return True if it's either
stored locally in the `DataSource` directory, or is a valid remote
URL. `DataSource` does not discriminate between the two, the file
is accessible if it exists in either location.
"""
return DataSource.exists(self, self._fullpath(path))
def open(self, path, mode='r'):
"""
Open and return file-like object prepending Repository base URL.
If `path` is an URL, it will be downloaded, stored in the
DataSource directory and opened from there.
Parameters
----------
path : str
Local file path or URL to open. This may, but does not have to,
include the `baseurl` with which the `Repository` was
initialized.
mode : {'r', 'w', 'a'}, optional
Mode to open `path`. Mode 'r' for reading, 'w' for writing,
'a' to append. Available modes depend on the type of object
specified by `path`. Default is 'r'.
Returns
-------
out : file object
File object.
"""
return DataSource.open(self, self._fullpath(path), mode)
def listdir(self):
"""
List files in the source Repository.
Returns
-------
files : list of str
List of file names (not containing a directory part).
Notes
-----
Does not currently work for remote repositories.
"""
if self._isurl(self._baseurl):
raise NotImplementedError(
"Directory listing of URLs, not supported yet.")
else:
return os.listdir(self._baseurl)
</code>Feature Branch: Help needed
Original created by @g00gl3vil on 16682 (Redmine)