Module pandare.qcows_internal

Module for fetching generic PANDA images and managing their metadata.

#!/usr/bin/env python3
'''
Module for fetching generic PANDA images and managing their metadata.
'''

import logging
import random
import hashlib
import json
from os import path, remove, makedirs
from subprocess import Popen, PIPE, CalledProcessError
from collections import namedtuple
from shlex import split as shlex_split
from sys import exit, stderr
from shutil import move

logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)

VM_DIR = path.join(path.expanduser("~"), ".panda")

class Image(namedtuple('Image', ['arch', 'os', 'prompt', 'cdrom', 'snapshot', 'url', 'alternate_urls', 'extra_files', 'qcow', 'default_mem', 'extra_args', 'hashes'])):
    '''
    The Image class stores information about a supported PANDA image

    Args:
        arch (str): Arch for the given architecture.
        os (str): an os string we can pass to panda with -os
        prompt (regex): a regex to detect a bash prompt after loading the snapshot and sending commands
        cdrom (str): name to use for cd-drive when inserting an ISO via monitor
        qcow (str): optional name to save qcow as
        url (str): url to download the qcow (e.g. https:// website.com/yourqcow.qcow2)
        default_mem (str): memory to use for the root snapshot (e.g. 1G)
        extra_files (list): other files (assumed to be in same directory on server) that we also need
        extra_args (list): Extra arguments to pass to PANDA (e.g. ['-display', 'none'])
        hashes (dict, optional): Mapping between qcow filenames and SHA1hashes they should match upon download
    '''

Image.__new__.__defaults__ = (None,) * len(Image._fields)

json_path = path.join(*[path.dirname(__file__), "qcows.json"])

with open(json_path, 'r') as f:
    images = json.load(f)

SUPPORTED_IMAGES = dict([
    (k, Image(**v)) for k, v in images.items()
])
"""
Dictionary of `Image` objects by name.
Generic values (underlying OS version may change) include:
    x86_64
    i386
    ppc
    arm
    aarch64
    mips
    mipsel
    mips64

You may also specify an exact arch/OS combination from the following exist:
    x86_64_ubuntu_1804
    i386_ubuntu_1604
    ppc_wheezy
    arm_wheezy
    aarch64 _focal
    mips_wheezy
    mips_buildroot5
    mipsel_wheezy
    mipsel_buildroot5
    mips64
""" # TODO: autogenerate values here

home_dir = path.expanduser('~')
for image in SUPPORTED_IMAGES:
    SUPPORTED_IMAGES[image] = (
        SUPPORTED_IMAGES[image]
            ._replace(
                extra_args=SUPPORTED_IMAGES[image].extra_args.replace('~', home_dir)
            )
    )

class Qcows():
    '''
    Helper library for managing qcows on your filesystem.
    Given an architecture, it can download a qcow from `panda.mit.edu` to `~/.panda/` and then use that.
    Alternatively, if a path to a qcow is provided, it can just use that.
    A qcow loaded by architecture can then be queried to get the name of the root snapshot or prompt.
    '''

    @staticmethod
    def get_qcow_info(name=None):
        '''
        Get information about supported image as specified by name.

        Args:
            name (str): String idenfifying a qcow supported

        Returns:
            Image: Instance of the Image class for a qcow
        '''
        if name is None:
            logger.warning("No qcow name provided. Defaulting to i386")
            name = "i386"

        if path.isfile(name):
            raise RuntimeError("TODO: can't automatically determine system info from custom qcows. Use one of: {}".format(", ".join(SUPPORTED_IMAGES.keys())))

        name = name.lower() # Case insensitive. Assumes supported_arches keys are lowercase
        if name not in SUPPORTED_IMAGES.keys():
            raise RuntimeError("Architecture {} is not in list of supported names: {}".format(name, ", ".join(SUPPORTED_IMAGES.keys())))

        r = SUPPORTED_IMAGES[name]
        # Move properties in .arch to being in the main object
        return r

    @staticmethod
    def get_qcow(name=None, download=True, _is_tty=True):
        '''
        Given a generic name of a qcow in `pandare.qcows.SUPPORTED_IMAGES` or a path to a qcow, return the path. Defaults to i386

        Args:
            name (str): generic name or path to qcow
            download (bool, default True): should the qcow be downloaded if necessary

        Returns:
            string: Path to qcow

        Raises:
            ValueError: if download is set to False and the qcow is not present
            RuntimeError: if the architecture is unsupported or the necessary files could not be downloaded
        '''
        if name is None:
            logger.warning("No qcow name provided. Defaulting to i386")
            name = "i386"

        if path.isfile(name):
            logger.debug("Provided qcow name appears to be a path, returning it directly: %s", name)
            return name

        name = name.lower() # Case insensitive. Assumes supported_images keys are lowercase
        if name not in SUPPORTED_IMAGES.keys():
            raise RuntimeError("Architecture {} is not in list of supported names: {}".format(name, ", ".join(SUPPORTED_IMAGES.keys())))

        image_data = SUPPORTED_IMAGES[name]
        qc = image_data.qcow
        if not qc: # Default, get name from url
            qc = image_data.url.split("/")[-1]
        qcow_path = path.join(VM_DIR,qc)
        makedirs(VM_DIR, exist_ok=True)

        # We need to downlaod if the QCOW or any extra files are missing
        # If the files are present on disk, assume they're okay
        needs_download = not path.isfile(qcow_path)

        if not needs_download:
            for extra_file in image_data.extra_files or []:
                extra_file_path = path.join(VM_DIR, extra_file)
                if not path.isfile(extra_file_path):
                    needs_download = True
                    break

        if needs_download and download:
            Qcows.download_qcow(image_data, qcow_path, _is_tty=_is_tty)
        elif needs_download:
            raise ValueError("Qcow is not on disk and download option is disabled")

        return qcow_path

    @staticmethod
    def get_file(urls, output_path, sha1hash=None, do_retry=True, _is_tty=True):
        assert len(urls) > 0
        url = random.choice([x for x in urls if x is not None])

        if _is_tty:
            print(f"Downloading required file: {url}")
            cmd = ["wget", '--show-progress', '--quiet', url, "-O", output_path+".tmp"]
        else:
            print(f"Please wait for download of required file: {url}", file=stderr)
            cmd = ["wget", "--quiet", url, "-O", output_path+".tmp"]

        try:
            with Popen(cmd, stdout=PIPE, bufsize=1, universal_newlines=True) as p:
                for line in p.stdout:
                    print(line, end='')

            if p.returncode != 0:
                raise CalledProcessError(p.returncode, p.args)

            # Check hash if we have one
            if sha1hash is not None:
                if _is_tty:
                    print(f"Validating file hash")
                sha1 = hashlib.sha1()

                with open(output_path+".tmp", 'rb') as f:
                    while True:
                        data = f.read(65536) #64kb chunks
                        if not data:
                            break
                        sha1.update(data)
                computed_hash = sha1.hexdigest()
                if computed_hash != sha1hash:
                    raise ValueError(f"{url} has hash {computed_hash} vs expected hash {sha1hash}")
                # Hash matches, move .tmp file to actual path
                move(output_path+".tmp", output_path)
            else:
                # No hash, move .tmp file to actual path
                move(output_path+".tmp", output_path)


        except Exception as e:
            logger.info("Download failed, deleting partial file: %s", output_path)
            remove(output_path+".tmp")

            if do_retry:
                if _is_tty and do_retry:
                    print("Hash mismatch - retrying")
                Qcows.get_file([url], output_path, sha1hash, do_retry=False, _is_tty=_is_tty)
            else:
                # Not retrying again, fatal - leave any partial files though
                raise RuntimeError(f"Unable to download expeted file from {url} even after retrying: {e}") from None
        logger.debug("Downloaded %s to %s", url, output_path)

    @staticmethod
    def download_qcow(image_data, output_path, _is_retry=False, _is_tty=True):
        '''
        Download the qcow described in the Image object in image_data
        Store to the output output_path.
        If the Image includes SHA1 hashes, validate the file was downloaded correctly, otherwise retry once
        '''

        # Check if we have a hash for the base qcow. Then download and vlidate with that hash
        qcow_base = image_data.url.split("/")[-1] if '/' in image_data.url else image_data.url
        base_hash = None

        if image_data.hashes is not None and qcow_base in image_data.hashes:
            base_hash = image_data.hashes[qcow_base]

        Qcows.get_file([image_data.url] + (image_data.alternate_urls if image_data.alternate_urls is not None else []), output_path, base_hash, _is_tty=_is_tty)

        # Download all extra files out of the same directory
        url_base = image_data.url[:image_data.url.rfind("/")] + "/"  # Truncate url to last /
        for extra_file in image_data.extra_files or []:
            extra_file_path = path.join(VM_DIR, extra_file)
            extra_hash = None
            if image_data.hashes is not None and extra_file in image_data.hashes:
                extra_hash = image_data.hashes[extra_file]
            Qcows.get_file([url_base + extra_file], extra_file_path, extra_hash, _is_tty=_is_tty) # TODO: support alternate URL here too? Won't work for some hosting options

    @staticmethod
    def qcow_from_arg(idx=1):
        '''
        Given an index into argv, call get_qcow with that arg if it exists, else with None

        Args:
            idx (int): an index into argv

        Returns:
            string: Path to qcow
        '''
        from sys import argv

        if len(argv) > idx:
            return Qcows.get_qcow(argv[idx])
        else:
            return Qcows.get_qcow()

    @staticmethod
    def remove_image(target):
        try:
            qcow = Qcows.get_qcow(target, download=False)
        except ValueError:
            # No QCOW, we're good!
            return

        try:
            image_data = SUPPORTED_IMAGES[target]
        except ValueError:
            # Not a valid image? I guess we're good
            return

        qc = image_data.qcow
        if not qc: # Default, get name from url
            qc = image_data.url.split("/")[-1]
        qcow_path = path.join(VM_DIR, qc)
        remove(qcow_path)

        for extra_file in image_data.extra_files or []:
            extra_file_path = path.join(VM_DIR, extra_file)
            if path.isfile(extra_file_path):
                remove(extra_file_path)

Global variables

var SUPPORTED_IMAGES

Dictionary of Image objects by name. Generic values (underlying OS version may change) include: x86_64 i386 ppc arm aarch64 mips mipsel mips64

You may also specify an exact arch/OS combination from the following exist: x86_64_ubuntu_1804 i386_ubuntu_1604 ppc_wheezy arm_wheezy aarch64 _focal mips_wheezy mips_buildroot5 mipsel_wheezy mipsel_buildroot5 mips64

Classes

class Image (arch=None, os=None, prompt=None, cdrom=None, snapshot=None, url=None, alternate_urls=None, extra_files=None, qcow=None, default_mem=None, extra_args=None, hashes=None)

The Image class stores information about a supported PANDA image

Args

arch : str

Arch for the given architecture.

os : str

an os string we can pass to panda with -os

prompt : regex

a regex to detect a bash prompt after loading the snapshot and sending commands

cdrom : str

name to use for cd-drive when inserting an ISO via monitor

qcow : str

optional name to save qcow as

url : str

url to download the qcow (e.g. https:// website.com/yourqcow.qcow2)

default_mem : str

memory to use for the root snapshot (e.g. 1G)

extra_files : list

other files (assumed to be in same directory on server) that we also need

extra_args : list

Extra arguments to pass to PANDA (e.g. ['-display', 'none'])

hashes : dict, optional

Mapping between qcow filenames and SHA1hashes they should match upon download

Expand source code

class Image(namedtuple('Image', ['arch', 'os', 'prompt', 'cdrom', 'snapshot', 'url', 'alternate_urls', 'extra_files', 'qcow', 'default_mem', 'extra_args', 'hashes'])):
    '''
    The Image class stores information about a supported PANDA image

    Args:
        arch (str): Arch for the given architecture.
        os (str): an os string we can pass to panda with -os
        prompt (regex): a regex to detect a bash prompt after loading the snapshot and sending commands
        cdrom (str): name to use for cd-drive when inserting an ISO via monitor
        qcow (str): optional name to save qcow as
        url (str): url to download the qcow (e.g. https:// website.com/yourqcow.qcow2)
        default_mem (str): memory to use for the root snapshot (e.g. 1G)
        extra_files (list): other files (assumed to be in same directory on server) that we also need
        extra_args (list): Extra arguments to pass to PANDA (e.g. ['-display', 'none'])
        hashes (dict, optional): Mapping between qcow filenames and SHA1hashes they should match upon download
    '''

Ancestors

• builtins.tuple

class Qcows

Helper library for managing qcows on your filesystem. Given an architecture, it can download a qcow from panda.mit.edu to ~/.panda/ and then use that. Alternatively, if a path to a qcow is provided, it can just use that. A qcow loaded by architecture can then be queried to get the name of the root snapshot or prompt.

Expand source code

class Qcows():
    '''
    Helper library for managing qcows on your filesystem.
    Given an architecture, it can download a qcow from `panda.mit.edu` to `~/.panda/` and then use that.
    Alternatively, if a path to a qcow is provided, it can just use that.
    A qcow loaded by architecture can then be queried to get the name of the root snapshot or prompt.
    '''

    @staticmethod
    def get_qcow_info(name=None):
        '''
        Get information about supported image as specified by name.

        Args:
            name (str): String idenfifying a qcow supported

        Returns:
            Image: Instance of the Image class for a qcow
        '''
        if name is None:
            logger.warning("No qcow name provided. Defaulting to i386")
            name = "i386"

        if path.isfile(name):
            raise RuntimeError("TODO: can't automatically determine system info from custom qcows. Use one of: {}".format(", ".join(SUPPORTED_IMAGES.keys())))

        name = name.lower() # Case insensitive. Assumes supported_arches keys are lowercase
        if name not in SUPPORTED_IMAGES.keys():
            raise RuntimeError("Architecture {} is not in list of supported names: {}".format(name, ", ".join(SUPPORTED_IMAGES.keys())))

        r = SUPPORTED_IMAGES[name]
        # Move properties in .arch to being in the main object
        return r

    @staticmethod
    def get_qcow(name=None, download=True, _is_tty=True):
        '''
        Given a generic name of a qcow in `pandare.qcows.SUPPORTED_IMAGES` or a path to a qcow, return the path. Defaults to i386

        Args:
            name (str): generic name or path to qcow
            download (bool, default True): should the qcow be downloaded if necessary

        Returns:
            string: Path to qcow

        Raises:
            ValueError: if download is set to False and the qcow is not present
            RuntimeError: if the architecture is unsupported or the necessary files could not be downloaded
        '''
        if name is None:
            logger.warning("No qcow name provided. Defaulting to i386")
            name = "i386"

        if path.isfile(name):
            logger.debug("Provided qcow name appears to be a path, returning it directly: %s", name)
            return name

        name = name.lower() # Case insensitive. Assumes supported_images keys are lowercase
        if name not in SUPPORTED_IMAGES.keys():
            raise RuntimeError("Architecture {} is not in list of supported names: {}".format(name, ", ".join(SUPPORTED_IMAGES.keys())))

        image_data = SUPPORTED_IMAGES[name]
        qc = image_data.qcow
        if not qc: # Default, get name from url
            qc = image_data.url.split("/")[-1]
        qcow_path = path.join(VM_DIR,qc)
        makedirs(VM_DIR, exist_ok=True)

        # We need to downlaod if the QCOW or any extra files are missing
        # If the files are present on disk, assume they're okay
        needs_download = not path.isfile(qcow_path)

        if not needs_download:
            for extra_file in image_data.extra_files or []:
                extra_file_path = path.join(VM_DIR, extra_file)
                if not path.isfile(extra_file_path):
                    needs_download = True
                    break

        if needs_download and download:
            Qcows.download_qcow(image_data, qcow_path, _is_tty=_is_tty)
        elif needs_download:
            raise ValueError("Qcow is not on disk and download option is disabled")

        return qcow_path

    @staticmethod
    def get_file(urls, output_path, sha1hash=None, do_retry=True, _is_tty=True):
        assert len(urls) > 0
        url = random.choice([x for x in urls if x is not None])

        if _is_tty:
            print(f"Downloading required file: {url}")
            cmd = ["wget", '--show-progress', '--quiet', url, "-O", output_path+".tmp"]
        else:
            print(f"Please wait for download of required file: {url}", file=stderr)
            cmd = ["wget", "--quiet", url, "-O", output_path+".tmp"]

        try:
            with Popen(cmd, stdout=PIPE, bufsize=1, universal_newlines=True) as p:
                for line in p.stdout:
                    print(line, end='')

            if p.returncode != 0:
                raise CalledProcessError(p.returncode, p.args)

            # Check hash if we have one
            if sha1hash is not None:
                if _is_tty:
                    print(f"Validating file hash")
                sha1 = hashlib.sha1()

                with open(output_path+".tmp", 'rb') as f:
                    while True:
                        data = f.read(65536) #64kb chunks
                        if not data:
                            break
                        sha1.update(data)
                computed_hash = sha1.hexdigest()
                if computed_hash != sha1hash:
                    raise ValueError(f"{url} has hash {computed_hash} vs expected hash {sha1hash}")
                # Hash matches, move .tmp file to actual path
                move(output_path+".tmp", output_path)
            else:
                # No hash, move .tmp file to actual path
                move(output_path+".tmp", output_path)


        except Exception as e:
            logger.info("Download failed, deleting partial file: %s", output_path)
            remove(output_path+".tmp")

            if do_retry:
                if _is_tty and do_retry:
                    print("Hash mismatch - retrying")
                Qcows.get_file([url], output_path, sha1hash, do_retry=False, _is_tty=_is_tty)
            else:
                # Not retrying again, fatal - leave any partial files though
                raise RuntimeError(f"Unable to download expeted file from {url} even after retrying: {e}") from None
        logger.debug("Downloaded %s to %s", url, output_path)

    @staticmethod
    def download_qcow(image_data, output_path, _is_retry=False, _is_tty=True):
        '''
        Download the qcow described in the Image object in image_data
        Store to the output output_path.
        If the Image includes SHA1 hashes, validate the file was downloaded correctly, otherwise retry once
        '''

        # Check if we have a hash for the base qcow. Then download and vlidate with that hash
        qcow_base = image_data.url.split("/")[-1] if '/' in image_data.url else image_data.url
        base_hash = None

        if image_data.hashes is not None and qcow_base in image_data.hashes:
            base_hash = image_data.hashes[qcow_base]

        Qcows.get_file([image_data.url] + (image_data.alternate_urls if image_data.alternate_urls is not None else []), output_path, base_hash, _is_tty=_is_tty)

        # Download all extra files out of the same directory
        url_base = image_data.url[:image_data.url.rfind("/")] + "/"  # Truncate url to last /
        for extra_file in image_data.extra_files or []:
            extra_file_path = path.join(VM_DIR, extra_file)
            extra_hash = None
            if image_data.hashes is not None and extra_file in image_data.hashes:
                extra_hash = image_data.hashes[extra_file]
            Qcows.get_file([url_base + extra_file], extra_file_path, extra_hash, _is_tty=_is_tty) # TODO: support alternate URL here too? Won't work for some hosting options

    @staticmethod
    def qcow_from_arg(idx=1):
        '''
        Given an index into argv, call get_qcow with that arg if it exists, else with None

        Args:
            idx (int): an index into argv

        Returns:
            string: Path to qcow
        '''
        from sys import argv

        if len(argv) > idx:
            return Qcows.get_qcow(argv[idx])
        else:
            return Qcows.get_qcow()

    @staticmethod
    def remove_image(target):
        try:
            qcow = Qcows.get_qcow(target, download=False)
        except ValueError:
            # No QCOW, we're good!
            return

        try:
            image_data = SUPPORTED_IMAGES[target]
        except ValueError:
            # Not a valid image? I guess we're good
            return

        qc = image_data.qcow
        if not qc: # Default, get name from url
            qc = image_data.url.split("/")[-1]
        qcow_path = path.join(VM_DIR, qc)
        remove(qcow_path)

        for extra_file in image_data.extra_files or []:
            extra_file_path = path.join(VM_DIR, extra_file)
            if path.isfile(extra_file_path):
                remove(extra_file_path)

Static methods

def download_qcow(image_data, output_path)

Download the qcow described in the Image object in image_data Store to the output output_path. If the Image includes SHA1 hashes, validate the file was downloaded correctly, otherwise retry once

Expand source code

@staticmethod
def download_qcow(image_data, output_path, _is_retry=False, _is_tty=True):
    '''
    Download the qcow described in the Image object in image_data
    Store to the output output_path.
    If the Image includes SHA1 hashes, validate the file was downloaded correctly, otherwise retry once
    '''

    # Check if we have a hash for the base qcow. Then download and vlidate with that hash
    qcow_base = image_data.url.split("/")[-1] if '/' in image_data.url else image_data.url
    base_hash = None

    if image_data.hashes is not None and qcow_base in image_data.hashes:
        base_hash = image_data.hashes[qcow_base]

    Qcows.get_file([image_data.url] + (image_data.alternate_urls if image_data.alternate_urls is not None else []), output_path, base_hash, _is_tty=_is_tty)

    # Download all extra files out of the same directory
    url_base = image_data.url[:image_data.url.rfind("/")] + "/"  # Truncate url to last /
    for extra_file in image_data.extra_files or []:
        extra_file_path = path.join(VM_DIR, extra_file)
        extra_hash = None
        if image_data.hashes is not None and extra_file in image_data.hashes:
            extra_hash = image_data.hashes[extra_file]
        Qcows.get_file([url_base + extra_file], extra_file_path, extra_hash, _is_tty=_is_tty) # TODO: support alternate URL here too? Won't work for some hosting options

def get_file(urls, output_path, sha1hash=None, do_retry=True)

Expand source code

@staticmethod
def get_file(urls, output_path, sha1hash=None, do_retry=True, _is_tty=True):
    assert len(urls) > 0
    url = random.choice([x for x in urls if x is not None])

    if _is_tty:
        print(f"Downloading required file: {url}")
        cmd = ["wget", '--show-progress', '--quiet', url, "-O", output_path+".tmp"]
    else:
        print(f"Please wait for download of required file: {url}", file=stderr)
        cmd = ["wget", "--quiet", url, "-O", output_path+".tmp"]

    try:
        with Popen(cmd, stdout=PIPE, bufsize=1, universal_newlines=True) as p:
            for line in p.stdout:
                print(line, end='')

        if p.returncode != 0:
            raise CalledProcessError(p.returncode, p.args)

        # Check hash if we have one
        if sha1hash is not None:
            if _is_tty:
                print(f"Validating file hash")
            sha1 = hashlib.sha1()

            with open(output_path+".tmp", 'rb') as f:
                while True:
                    data = f.read(65536) #64kb chunks
                    if not data:
                        break
                    sha1.update(data)
            computed_hash = sha1.hexdigest()
            if computed_hash != sha1hash:
                raise ValueError(f"{url} has hash {computed_hash} vs expected hash {sha1hash}")
            # Hash matches, move .tmp file to actual path
            move(output_path+".tmp", output_path)
        else:
            # No hash, move .tmp file to actual path
            move(output_path+".tmp", output_path)


    except Exception as e:
        logger.info("Download failed, deleting partial file: %s", output_path)
        remove(output_path+".tmp")

        if do_retry:
            if _is_tty and do_retry:
                print("Hash mismatch - retrying")
            Qcows.get_file([url], output_path, sha1hash, do_retry=False, _is_tty=_is_tty)
        else:
            # Not retrying again, fatal - leave any partial files though
            raise RuntimeError(f"Unable to download expeted file from {url} even after retrying: {e}") from None
    logger.debug("Downloaded %s to %s", url, output_path)

def get_qcow(name=None, download=True)

Given a generic name of a qcow in pandare.qcows.SUPPORTED_IMAGES or a path to a qcow, return the path. Defaults to i386

Args

name : str

generic name or path to qcow

download : bool, default True

should the qcow be downloaded if necessary

Returns

string

Path to qcow

Raises

ValueError

if download is set to False and the qcow is not present

RuntimeError

if the architecture is unsupported or the necessary files could not be downloaded

Expand source code

@staticmethod
def get_qcow(name=None, download=True, _is_tty=True):
    '''
    Given a generic name of a qcow in `pandare.qcows.SUPPORTED_IMAGES` or a path to a qcow, return the path. Defaults to i386

    Args:
        name (str): generic name or path to qcow
        download (bool, default True): should the qcow be downloaded if necessary

    Returns:
        string: Path to qcow

    Raises:
        ValueError: if download is set to False and the qcow is not present
        RuntimeError: if the architecture is unsupported or the necessary files could not be downloaded
    '''
    if name is None:
        logger.warning("No qcow name provided. Defaulting to i386")
        name = "i386"

    if path.isfile(name):
        logger.debug("Provided qcow name appears to be a path, returning it directly: %s", name)
        return name

    name = name.lower() # Case insensitive. Assumes supported_images keys are lowercase
    if name not in SUPPORTED_IMAGES.keys():
        raise RuntimeError("Architecture {} is not in list of supported names: {}".format(name, ", ".join(SUPPORTED_IMAGES.keys())))

    image_data = SUPPORTED_IMAGES[name]
    qc = image_data.qcow
    if not qc: # Default, get name from url
        qc = image_data.url.split("/")[-1]
    qcow_path = path.join(VM_DIR,qc)
    makedirs(VM_DIR, exist_ok=True)

    # We need to downlaod if the QCOW or any extra files are missing
    # If the files are present on disk, assume they're okay
    needs_download = not path.isfile(qcow_path)

    if not needs_download:
        for extra_file in image_data.extra_files or []:
            extra_file_path = path.join(VM_DIR, extra_file)
            if not path.isfile(extra_file_path):
                needs_download = True
                break

    if needs_download and download:
        Qcows.download_qcow(image_data, qcow_path, _is_tty=_is_tty)
    elif needs_download:
        raise ValueError("Qcow is not on disk and download option is disabled")

    return qcow_path

def get_qcow_info(name=None)

Get information about supported image as specified by name.

Args

name : str

String idenfifying a qcow supported

Returns

Image

Instance of the Image class for a qcow

Expand source code

@staticmethod
def get_qcow_info(name=None):
    '''
    Get information about supported image as specified by name.

    Args:
        name (str): String idenfifying a qcow supported

    Returns:
        Image: Instance of the Image class for a qcow
    '''
    if name is None:
        logger.warning("No qcow name provided. Defaulting to i386")
        name = "i386"

    if path.isfile(name):
        raise RuntimeError("TODO: can't automatically determine system info from custom qcows. Use one of: {}".format(", ".join(SUPPORTED_IMAGES.keys())))

    name = name.lower() # Case insensitive. Assumes supported_arches keys are lowercase
    if name not in SUPPORTED_IMAGES.keys():
        raise RuntimeError("Architecture {} is not in list of supported names: {}".format(name, ", ".join(SUPPORTED_IMAGES.keys())))

    r = SUPPORTED_IMAGES[name]
    # Move properties in .arch to being in the main object
    return r

def qcow_from_arg(idx=1)

Given an index into argv, call get_qcow with that arg if it exists, else with None

Args

idx : int

an index into argv

Returns

string

Path to qcow

Expand source code

@staticmethod
def qcow_from_arg(idx=1):
    '''
    Given an index into argv, call get_qcow with that arg if it exists, else with None

    Args:
        idx (int): an index into argv

    Returns:
        string: Path to qcow
    '''
    from sys import argv

    if len(argv) > idx:
        return Qcows.get_qcow(argv[idx])
    else:
        return Qcows.get_qcow()

def remove_image(target)

Expand source code

@staticmethod
def remove_image(target):
    try:
        qcow = Qcows.get_qcow(target, download=False)
    except ValueError:
        # No QCOW, we're good!
        return

    try:
        image_data = SUPPORTED_IMAGES[target]
    except ValueError:
        # Not a valid image? I guess we're good
        return

    qc = image_data.qcow
    if not qc: # Default, get name from url
        qc = image_data.url.split("/")[-1]
    qcow_path = path.join(VM_DIR, qc)
    remove(qcow_path)

    for extra_file in image_data.extra_files or []:
        extra_file_path = path.join(VM_DIR, extra_file)
        if path.isfile(extra_file_path):
            remove(extra_file_path)