reset upstream subtrees to yocto 2.6
Reset the following subtrees on thud HEAD:
poky: 87e3a9739d
meta-openembedded: 6094ae18c8
meta-security: 31dc4e7532
meta-raspberrypi: a48743dc36
meta-xilinx: c42016e2e6
Also re-apply backports that didn't make it into thud:
poky:
17726d0 systemd-systemctl-native: handle Install wildcards
meta-openembedded:
4321a5d libtinyxml2: update to 7.0.1
042f0a3 libcereal: Add native and nativesdk classes
e23284f libcereal: Allow empty package
030e8d4 rsyslog: curl-less build with fmhttp PACKAGECONFIG
179a1b9 gtest: update to 1.8.1
Squashed OpenBMC subtree compatibility updates:
meta-aspeed:
Brad Bishop (1):
aspeed: add yocto 2.6 compatibility
meta-ibm:
Brad Bishop (1):
ibm: prepare for yocto 2.6
meta-ingrasys:
Brad Bishop (1):
ingrasys: set layer compatibility to yocto 2.6
meta-openpower:
Brad Bishop (1):
openpower: set layer compatibility to yocto 2.6
meta-phosphor:
Brad Bishop (3):
phosphor: set layer compatibility to thud
phosphor: libgpg-error: drop patches
phosphor: react to fitimage artifact rename
Ed Tanous (4):
Dropbear: upgrade options for latest upgrade
yocto2.6: update openssl options
busybox: remove upstream watchdog patch
systemd: Rebase CONFIG_CGROUP_BPF patch
Change-Id: I7b1fe71cca880d0372a82d94b5fd785323e3a9e7
Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com>
diff --git a/poky/bitbake/lib/layerindexlib/__init__.py b/poky/bitbake/lib/layerindexlib/__init__.py
new file mode 100644
index 0000000..cb79cb3
--- /dev/null
+++ b/poky/bitbake/lib/layerindexlib/__init__.py
@@ -0,0 +1,1363 @@
+# Copyright (C) 2016-2018 Wind River Systems, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License version 2 as
+# published by the Free Software Foundation.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+# See the GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+
+import datetime
+
+import logging
+import imp
+
+from collections import OrderedDict
+from layerindexlib.plugin import LayerIndexPluginUrlError
+
+logger = logging.getLogger('BitBake.layerindexlib')
+
+# Exceptions
+
+class LayerIndexException(Exception):
+ '''LayerIndex Generic Exception'''
+ def __init__(self, message):
+ self.msg = message
+ Exception.__init__(self, message)
+
+ def __str__(self):
+ return self.msg
+
+class LayerIndexUrlError(LayerIndexException):
+ '''Exception raised when unable to access a URL for some reason'''
+ def __init__(self, url, message=""):
+ if message:
+ msg = "Unable to access layerindex url %s: %s" % (url, message)
+ else:
+ msg = "Unable to access layerindex url %s" % url
+ self.url = url
+ LayerIndexException.__init__(self, msg)
+
+class LayerIndexFetchError(LayerIndexException):
+ '''General layerindex fetcher exception when something fails'''
+ def __init__(self, url, message=""):
+ if message:
+ msg = "Unable to fetch layerindex url %s: %s" % (url, message)
+ else:
+ msg = "Unable to fetch layerindex url %s" % url
+ self.url = url
+ LayerIndexException.__init__(self, msg)
+
+
+# Interface to the overall layerindex system
+# the layer may contain one or more individual indexes
+class LayerIndex():
+ def __init__(self, d):
+ if not d:
+ raise LayerIndexException("Must be initialized with bb.data.")
+
+ self.data = d
+
+ # List of LayerIndexObj
+ self.indexes = []
+
+ self.plugins = []
+
+ import bb.utils
+ bb.utils.load_plugins(logger, self.plugins, os.path.dirname(__file__))
+ for plugin in self.plugins:
+ if hasattr(plugin, 'init'):
+ plugin.init(self)
+
+ def __add__(self, other):
+ newIndex = LayerIndex(self.data)
+
+ if self.__class__ != newIndex.__class__ or \
+ other.__class__ != newIndex.__class__:
+ raise TypeException("Can not add different types.")
+
+ for indexEnt in self.indexes:
+ newIndex.indexes.append(indexEnt)
+
+ for indexEnt in other.indexes:
+ newIndex.indexes.append(indexEnt)
+
+ return newIndex
+
+ def _parse_params(self, params):
+ '''Take a parameter list, return a dictionary of parameters.
+
+ Expected to be called from the data of urllib.parse.urlparse(url).params
+
+ If there are two conflicting parameters, last in wins...
+ '''
+
+ param_dict = {}
+ for param in params.split(';'):
+ if not param:
+ continue
+ item = param.split('=', 1)
+ logger.debug(1, item)
+ param_dict[item[0]] = item[1]
+
+ return param_dict
+
+ def _fetch_url(self, url, username=None, password=None, debuglevel=0):
+ '''Fetch data from a specific URL.
+
+ Fetch something from a specific URL. This is specifically designed to
+ fetch data from a layerindex-web instance, but may be useful for other
+ raw fetch actions.
+
+ It is not designed to be used to fetch recipe sources or similar. the
+ regular fetcher class should used for that.
+
+ It is the responsibility of the caller to check BB_NO_NETWORK and related
+ BB_ALLOWED_NETWORKS.
+ '''
+
+ if not url:
+ raise LayerIndexUrlError(url, "empty url")
+
+ import urllib
+ from urllib.request import urlopen, Request
+ from urllib.parse import urlparse
+
+ up = urlparse(url)
+
+ if username:
+ logger.debug(1, "Configuring authentication for %s..." % url)
+ password_mgr = urllib.request.HTTPPasswordMgrWithDefaultRealm()
+ password_mgr.add_password(None, "%s://%s" % (up.scheme, up.netloc), username, password)
+ handler = urllib.request.HTTPBasicAuthHandler(password_mgr)
+ opener = urllib.request.build_opener(handler, urllib.request.HTTPSHandler(debuglevel=debuglevel))
+ else:
+ opener = urllib.request.build_opener(urllib.request.HTTPSHandler(debuglevel=debuglevel))
+
+ urllib.request.install_opener(opener)
+
+ logger.debug(1, "Fetching %s (%s)..." % (url, ["without authentication", "with authentication"][bool(username)]))
+
+ try:
+ res = urlopen(Request(url, headers={'User-Agent': 'Mozilla/5.0 (bitbake/lib/layerindex)'}, unverifiable=True))
+ except urllib.error.HTTPError as e:
+ logger.debug(1, "HTTP Error: %s: %s" % (e.code, e.reason))
+ logger.debug(1, " Requested: %s" % (url))
+ logger.debug(1, " Actual: %s" % (e.geturl()))
+
+ if e.code == 404:
+ logger.debug(1, "Request not found.")
+ raise LayerIndexFetchError(url, e)
+ else:
+ logger.debug(1, "Headers:\n%s" % (e.headers))
+ raise LayerIndexFetchError(url, e)
+ except OSError as e:
+ error = 0
+ reason = ""
+
+ # Process base OSError first...
+ if hasattr(e, 'errno'):
+ error = e.errno
+ reason = e.strerror
+
+ # Process gaierror (socket error) subclass if available.
+ if hasattr(e, 'reason') and hasattr(e.reason, 'errno') and hasattr(e.reason, 'strerror'):
+ error = e.reason.errno
+ reason = e.reason.strerror
+ if error == -2:
+ raise LayerIndexFetchError(url, "%s: %s" % (e, reason))
+
+ if error and error != 0:
+ raise LayerIndexFetchError(url, "Unexpected exception: [Error %s] %s" % (error, reason))
+ else:
+ raise LayerIndexFetchError(url, "Unable to fetch OSError exception: %s" % e)
+
+ finally:
+ logger.debug(1, "...fetching %s (%s), done." % (url, ["without authentication", "with authentication"][bool(username)]))
+
+ return res
+
+
+ def load_layerindex(self, indexURI, load=['layerDependencies', 'recipes', 'machines', 'distros'], reload=False):
+ '''Load the layerindex.
+
+ indexURI - An index to load. (Use multiple calls to load multiple indexes)
+
+ reload - If reload is True, then any previously loaded indexes will be forgotten.
+
+ load - List of elements to load. Default loads all items.
+ Note: plugs may ignore this.
+
+The format of the indexURI:
+
+ <url>;branch=<branch>;cache=<cache>;desc=<description>
+
+ Note: the 'branch' parameter if set can select multiple branches by using
+ comma, such as 'branch=master,morty,pyro'. However, many operations only look
+ at the -first- branch specified!
+
+ The cache value may be undefined, in this case a network failure will
+ result in an error, otherwise the system will look for a file of the cache
+ name and load that instead.
+
+ For example:
+
+ http://layers.openembedded.org/layerindex/api/;branch=master;desc=OpenEmbedded%20Layer%20Index
+ cooker://
+'''
+ if reload:
+ self.indexes = []
+
+ logger.debug(1, 'Loading: %s' % indexURI)
+
+ if not self.plugins:
+ raise LayerIndexException("No LayerIndex Plugins available")
+
+ for plugin in self.plugins:
+ # Check if the plugin was initialized
+ logger.debug(1, 'Trying %s' % plugin.__class__)
+ if not hasattr(plugin, 'type') or not plugin.type:
+ continue
+ try:
+ # TODO: Implement 'cache', for when the network is not available
+ indexEnt = plugin.load_index(indexURI, load)
+ break
+ except LayerIndexPluginUrlError as e:
+ logger.debug(1, "%s doesn't support %s" % (plugin.type, e.url))
+ except NotImplementedError:
+ pass
+ else:
+ logger.debug(1, "No plugins support %s" % indexURI)
+ raise LayerIndexException("No plugins support %s" % indexURI)
+
+ # Mark CONFIG data as something we've added...
+ indexEnt.config['local'] = []
+ indexEnt.config['local'].append('config')
+
+ # No longer permit changes..
+ indexEnt.lockData()
+
+ self.indexes.append(indexEnt)
+
+ def store_layerindex(self, indexURI, index=None):
+ '''Store one layerindex
+
+Typically this will be used to create a local cache file of a remote index.
+
+ file://<path>;branch=<branch>
+
+We can write out in either the restapi or django formats. The split option
+will write out the individual elements split by layer and related components.
+'''
+ if not index:
+ logger.warning('No index to write, nothing to do.')
+ return
+
+ if not self.plugins:
+ raise LayerIndexException("No LayerIndex Plugins available")
+
+ for plugin in self.plugins:
+ # Check if the plugin was initialized
+ logger.debug(1, 'Trying %s' % plugin.__class__)
+ if not hasattr(plugin, 'type') or not plugin.type:
+ continue
+ try:
+ plugin.store_index(indexURI, index)
+ break
+ except LayerIndexPluginUrlError as e:
+ logger.debug(1, "%s doesn't support %s" % (plugin.type, e.url))
+ except NotImplementedError:
+ logger.debug(1, "Store not implemented in %s" % plugin.type)
+ pass
+ else:
+ logger.debug(1, "No plugins support %s" % url)
+ raise LayerIndexException("No plugins support %s" % url)
+
+
+ def is_empty(self):
+ '''Return True or False if the index has any usable data.
+
+We check the indexes entries to see if they have a branch set, as well as
+layerBranches set. If not, they are effectively blank.'''
+
+ found = False
+ for index in self.indexes:
+ if index.__bool__():
+ found = True
+ break
+ return not found
+
+
+ def find_vcs_url(self, vcs_url, branch=None):
+ '''Return the first layerBranch with the given vcs_url
+
+ If a branch has not been specified, we will iterate over the branches in
+ the default configuration until the first vcs_url/branch match.'''
+
+ for index in self.indexes:
+ logger.debug(1, ' searching %s' % index.config['DESCRIPTION'])
+ layerBranch = index.find_vcs_url(vcs_url, [branch])
+ if layerBranch:
+ return layerBranch
+ return None
+
+ def find_collection(self, collection, version=None, branch=None):
+ '''Return the first layerBranch with the given collection name
+
+ If a branch has not been specified, we will iterate over the branches in
+ the default configuration until the first collection/branch match.'''
+
+ logger.debug(1, 'find_collection: %s (%s) %s' % (collection, version, branch))
+
+ if branch:
+ branches = [branch]
+ else:
+ branches = None
+
+ for index in self.indexes:
+ logger.debug(1, ' searching %s' % index.config['DESCRIPTION'])
+ layerBranch = index.find_collection(collection, version, branches)
+ if layerBranch:
+ return layerBranch
+ else:
+ logger.debug(1, 'Collection %s (%s) not found for branch (%s)' % (collection, version, branch))
+ return None
+
+ def find_layerbranch(self, name, branch=None):
+ '''Return the layerBranch item for a given name and branch
+
+ If a branch has not been specified, we will iterate over the branches in
+ the default configuration until the first name/branch match.'''
+
+ if branch:
+ branches = [branch]
+ else:
+ branches = None
+
+ for index in self.indexes:
+ layerBranch = index.find_layerbranch(name, branches)
+ if layerBranch:
+ return layerBranch
+ return None
+
+ def find_dependencies(self, names=None, layerbranches=None, ignores=None):
+ '''Return a tuple of all dependencies and valid items for the list of (layer) names
+
+ The dependency scanning happens depth-first. The returned
+ dependencies should be in the best order to define bblayers.
+
+ names - list of layer names (searching layerItems)
+ branches - when specified (with names) only this list of branches are evaluated
+
+ layerbranches - list of layerbranches to resolve dependencies
+
+ ignores - list of layer names to ignore
+
+ return: (dependencies, invalid)
+
+ dependencies[LayerItem.name] = [ LayerBranch, LayerDependency1, LayerDependency2, ... ]
+ invalid = [ LayerItem.name1, LayerItem.name2, ... ]
+ '''
+
+ invalid = []
+
+ # Convert name/branch to layerbranches
+ if layerbranches is None:
+ layerbranches = []
+
+ for name in names:
+ if ignores and name in ignores:
+ continue
+
+ for index in self.indexes:
+ layerbranch = index.find_layerbranch(name)
+ if not layerbranch:
+ # Not in this index, hopefully it's in another...
+ continue
+ layerbranches.append(layerbranch)
+ break
+ else:
+ invalid.append(name)
+
+
+ def _resolve_dependencies(layerbranches, ignores, dependencies, invalid):
+ for layerbranch in layerbranches:
+ if ignores and layerbranch.layer.name in ignores:
+ continue
+
+ # Get a list of dependencies and then recursively process them
+ for layerdependency in layerbranch.index.layerDependencies_layerBranchId[layerbranch.id]:
+ deplayerbranch = layerdependency.dependency_layerBranch
+
+ if ignores and deplayerbranch.layer.name in ignores:
+ continue
+
+ # This little block is why we can't re-use the LayerIndexObj version,
+ # we must be able to satisfy each dependencies across layer indexes and
+ # use the layer index order for priority. (r stands for replacement below)
+
+ # If this is the primary index, we can fast path and skip this
+ if deplayerbranch.index != self.indexes[0]:
+ # Is there an entry in a prior index for this collection/version?
+ rdeplayerbranch = self.find_collection(
+ collection=deplayerbranch.collection,
+ version=deplayerbranch.version
+ )
+ if rdeplayerbranch != deplayerbranch:
+ logger.debug(1, 'Replaced %s:%s:%s with %s:%s:%s' % \
+ (deplayerbranch.index.config['DESCRIPTION'],
+ deplayerbranch.branch.name,
+ deplayerbranch.layer.name,
+ rdeplayerbranch.index.config['DESCRIPTION'],
+ rdeplayerbranch.branch.name,
+ rdeplayerbranch.layer.name))
+ deplayerbranch = rdeplayerbranch
+
+ # New dependency, we need to resolve it now... depth-first
+ if deplayerbranch.layer.name not in dependencies:
+ (dependencies, invalid) = _resolve_dependencies([deplayerbranch], ignores, dependencies, invalid)
+
+ if deplayerbranch.layer.name not in dependencies:
+ dependencies[deplayerbranch.layer.name] = [deplayerbranch, layerdependency]
+ else:
+ if layerdependency not in dependencies[deplayerbranch.layer.name]:
+ dependencies[deplayerbranch.layer.name].append(layerdependency)
+
+ return (dependencies, invalid)
+
+ # OK, resolve this one...
+ dependencies = OrderedDict()
+ (dependencies, invalid) = _resolve_dependencies(layerbranches, ignores, dependencies, invalid)
+
+ for layerbranch in layerbranches:
+ if layerbranch.layer.name not in dependencies:
+ dependencies[layerbranch.layer.name] = [layerbranch]
+
+ return (dependencies, invalid)
+
+
+ def list_obj(self, object):
+ '''Print via the plain logger object information
+
+This function is used to implement debugging and provide the user info.
+'''
+ for lix in self.indexes:
+ if not hasattr(lix, object):
+ continue
+
+ logger.plain ('')
+ logger.plain ('Index: %s' % lix.config['DESCRIPTION'])
+
+ output = []
+
+ if object == 'branches':
+ logger.plain ('%s %s %s' % ('{:26}'.format('branch'), '{:34}'.format('description'), '{:22}'.format('bitbake branch')))
+ logger.plain ('{:-^80}'.format(""))
+ for branchid in lix.branches:
+ output.append('%s %s %s' % (
+ '{:26}'.format(lix.branches[branchid].name),
+ '{:34}'.format(lix.branches[branchid].short_description),
+ '{:22}'.format(lix.branches[branchid].bitbake_branch)
+ ))
+ for line in sorted(output):
+ logger.plain (line)
+
+ continue
+
+ if object == 'layerItems':
+ logger.plain ('%s %s' % ('{:26}'.format('layer'), '{:34}'.format('description')))
+ logger.plain ('{:-^80}'.format(""))
+ for layerid in lix.layerItems:
+ output.append('%s %s' % (
+ '{:26}'.format(lix.layerItems[layerid].name),
+ '{:34}'.format(lix.layerItems[layerid].summary)
+ ))
+ for line in sorted(output):
+ logger.plain (line)
+
+ continue
+
+ if object == 'layerBranches':
+ logger.plain ('%s %s %s' % ('{:26}'.format('layer'), '{:34}'.format('description'), '{:19}'.format('collection:version')))
+ logger.plain ('{:-^80}'.format(""))
+ for layerbranchid in lix.layerBranches:
+ output.append('%s %s %s' % (
+ '{:26}'.format(lix.layerBranches[layerbranchid].layer.name),
+ '{:34}'.format(lix.layerBranches[layerbranchid].layer.summary),
+ '{:19}'.format("%s:%s" %
+ (lix.layerBranches[layerbranchid].collection,
+ lix.layerBranches[layerbranchid].version)
+ )
+ ))
+ for line in sorted(output):
+ logger.plain (line)
+
+ continue
+
+ if object == 'layerDependencies':
+ logger.plain ('%s %s %s %s' % ('{:19}'.format('branch'), '{:26}'.format('layer'), '{:11}'.format('dependency'), '{:26}'.format('layer')))
+ logger.plain ('{:-^80}'.format(""))
+ for layerDependency in lix.layerDependencies:
+ if not lix.layerDependencies[layerDependency].dependency_layerBranch:
+ continue
+
+ output.append('%s %s %s %s' % (
+ '{:19}'.format(lix.layerDependencies[layerDependency].layerbranch.branch.name),
+ '{:26}'.format(lix.layerDependencies[layerDependency].layerbranch.layer.name),
+ '{:11}'.format('requires' if lix.layerDependencies[layerDependency].required else 'recommends'),
+ '{:26}'.format(lix.layerDependencies[layerDependency].dependency_layerBranch.layer.name)
+ ))
+ for line in sorted(output):
+ logger.plain (line)
+
+ continue
+
+ if object == 'recipes':
+ logger.plain ('%s %s %s' % ('{:20}'.format('recipe'), '{:10}'.format('version'), 'layer'))
+ logger.plain ('{:-^80}'.format(""))
+ output = []
+ for recipe in lix.recipes:
+ output.append('%s %s %s' % (
+ '{:30}'.format(lix.recipes[recipe].pn),
+ '{:30}'.format(lix.recipes[recipe].pv),
+ lix.recipes[recipe].layer.name
+ ))
+ for line in sorted(output):
+ logger.plain (line)
+
+ continue
+
+ if object == 'machines':
+ logger.plain ('%s %s %s' % ('{:24}'.format('machine'), '{:34}'.format('description'), '{:19}'.format('layer')))
+ logger.plain ('{:-^80}'.format(""))
+ for machine in lix.machines:
+ output.append('%s %s %s' % (
+ '{:24}'.format(lix.machines[machine].name),
+ '{:34}'.format(lix.machines[machine].description)[:34],
+ '{:19}'.format(lix.machines[machine].layerbranch.layer.name)
+ ))
+ for line in sorted(output):
+ logger.plain (line)
+
+ continue
+
+ if object == 'distros':
+ logger.plain ('%s %s %s' % ('{:24}'.format('distro'), '{:34}'.format('description'), '{:19}'.format('layer')))
+ logger.plain ('{:-^80}'.format(""))
+ for distro in lix.distros:
+ output.append('%s %s %s' % (
+ '{:24}'.format(lix.distros[distro].name),
+ '{:34}'.format(lix.distros[distro].description)[:34],
+ '{:19}'.format(lix.distros[distro].layerbranch.layer.name)
+ ))
+ for line in sorted(output):
+ logger.plain (line)
+
+ continue
+
+ logger.plain ('')
+
+
+# This class holds a single layer index instance
+# The LayerIndexObj is made up of dictionary of elements, such as:
+# index['config'] - configuration data for this index
+# index['branches'] - dictionary of Branch objects, by id number
+# index['layerItems'] - dictionary of layerItem objects, by id number
+# ...etc... (See: http://layers.openembedded.org/layerindex/api/)
+#
+# The class needs to manage the 'index' entries and allow easily adding
+# of new items, as well as simply loading of the items.
+class LayerIndexObj():
+ def __init__(self):
+ super().__setattr__('_index', {})
+ super().__setattr__('_lock', False)
+
+ def __bool__(self):
+ '''False if the index is effectively empty
+
+ We check the index to see if it has a branch set, as well as
+ layerbranches set. If not, it is effectively blank.'''
+
+ if not bool(self._index):
+ return False
+
+ try:
+ if self.branches and self.layerBranches:
+ return True
+ except AttributeError:
+ pass
+
+ return False
+
+ def __getattr__(self, name):
+ if name.startswith('_'):
+ return super().__getattribute__(name)
+
+ if name not in self._index:
+ raise AttributeError('%s not in index datastore' % name)
+
+ return self._index[name]
+
+ def __setattr__(self, name, value):
+ if self.isLocked():
+ raise TypeError("Can not set attribute '%s': index is locked" % name)
+
+ if name.startswith('_'):
+ super().__setattr__(name, value)
+ return
+
+ self._index[name] = value
+
+ def __delattr__(self, name):
+ if self.isLocked():
+ raise TypeError("Can not delete attribute '%s': index is locked" % name)
+
+ if name.startswith('_'):
+ super().__delattr__(name)
+
+ self._index.pop(name)
+
+ def lockData(self):
+ '''Lock data object (make it readonly)'''
+ super().__setattr__("_lock", True)
+
+ def unlockData(self):
+ '''unlock data object (make it readonly)'''
+ super().__setattr__("_lock", False)
+
+ # When the data is unlocked, we have to clear the caches, as
+ # modification is allowed!
+ del(self._layerBranches_layerId_branchId)
+ del(self._layerDependencies_layerBranchId)
+ del(self._layerBranches_vcsUrl)
+
+ def isLocked(self):
+ '''Is this object locked (readonly)?'''
+ return self._lock
+
+ def add_element(self, indexname, objs):
+ '''Add a layer index object to index.<indexname>'''
+ if indexname not in self._index:
+ self._index[indexname] = {}
+
+ for obj in objs:
+ if obj.id in self._index[indexname]:
+ if self._index[indexname][obj.id] == obj:
+ continue
+ raise LayerIndexError('Conflict adding object %s(%s) to index' % (indexname, obj.id))
+ self._index[indexname][obj.id] = obj
+
+ def add_raw_element(self, indexname, objtype, rawobjs):
+ '''Convert a raw layer index data item to a layer index item object and add to the index'''
+ objs = []
+ for entry in rawobjs:
+ objs.append(objtype(self, entry))
+ self.add_element(indexname, objs)
+
+ # Quick lookup table for searching layerId and branchID combos
+ @property
+ def layerBranches_layerId_branchId(self):
+ def createCache(self):
+ cache = {}
+ for layerbranchid in self.layerBranches:
+ layerbranch = self.layerBranches[layerbranchid]
+ cache["%s:%s" % (layerbranch.layer_id, layerbranch.branch_id)] = layerbranch
+ return cache
+
+ if self.isLocked():
+ cache = getattr(self, '_layerBranches_layerId_branchId', None)
+ else:
+ cache = None
+
+ if not cache:
+ cache = createCache(self)
+
+ if self.isLocked():
+ super().__setattr__('_layerBranches_layerId_branchId', cache)
+
+ return cache
+
+ # Quick lookup table for finding all dependencies of a layerBranch
+ @property
+ def layerDependencies_layerBranchId(self):
+ def createCache(self):
+ cache = {}
+ # This ensures empty lists for all branchids
+ for layerbranchid in self.layerBranches:
+ cache[layerbranchid] = []
+
+ for layerdependencyid in self.layerDependencies:
+ layerdependency = self.layerDependencies[layerdependencyid]
+ cache[layerdependency.layerbranch_id].append(layerdependency)
+ return cache
+
+ if self.isLocked():
+ cache = getattr(self, '_layerDependencies_layerBranchId', None)
+ else:
+ cache = None
+
+ if not cache:
+ cache = createCache(self)
+
+ if self.isLocked():
+ super().__setattr__('_layerDependencies_layerBranchId', cache)
+
+ return cache
+
+ # Quick lookup table for finding all instances of a vcs_url
+ @property
+ def layerBranches_vcsUrl(self):
+ def createCache(self):
+ cache = {}
+ for layerbranchid in self.layerBranches:
+ layerbranch = self.layerBranches[layerbranchid]
+ if layerbranch.layer.vcs_url not in cache:
+ cache[layerbranch.layer.vcs_url] = [layerbranch]
+ else:
+ cache[layerbranch.layer.vcs_url].append(layerbranch)
+ return cache
+
+ if self.isLocked():
+ cache = getattr(self, '_layerBranches_vcsUrl', None)
+ else:
+ cache = None
+
+ if not cache:
+ cache = createCache(self)
+
+ if self.isLocked():
+ super().__setattr__('_layerBranches_vcsUrl', cache)
+
+ return cache
+
+
+ def find_vcs_url(self, vcs_url, branches=None):
+ ''''Return the first layerBranch with the given vcs_url
+
+ If a list of branches has not been specified, we will iterate on
+ all branches until the first vcs_url is found.'''
+
+ if not self.__bool__():
+ return None
+
+ for layerbranch in self.layerBranches_vcsUrl:
+ if branches and layerbranch.branch.name not in branches:
+ continue
+
+ return layerbranch
+
+ return None
+
+
+ def find_collection(self, collection, version=None, branches=None):
+ '''Return the first layerBranch with the given collection name
+
+ If a list of branches has not been specified, we will iterate on
+ all branches until the first collection is found.'''
+
+ if not self.__bool__():
+ return None
+
+ for layerbranchid in self.layerBranches:
+ layerbranch = self.layerBranches[layerbranchid]
+ if branches and layerbranch.branch.name not in branches:
+ continue
+
+ if layerbranch.collection == collection and \
+ (version is None or version == layerbranch.version):
+ return layerbranch
+
+ return None
+
+
+ def find_layerbranch(self, name, branches=None):
+ '''Return the first layerbranch whose layer name matches
+
+ If a list of branches has not been specified, we will iterate on
+ all branches until the first layer with that name is found.'''
+
+ if not self.__bool__():
+ return None
+
+ for layerbranchid in self.layerBranches:
+ layerbranch = self.layerBranches[layerbranchid]
+ if branches and layerbranch.branch.name not in branches:
+ continue
+
+ if layerbranch.layer.name == name:
+ return layerbranch
+
+ return None
+
+ def find_dependencies(self, names=None, branches=None, layerBranches=None, ignores=None):
+ '''Return a tuple of all dependencies and valid items for the list of (layer) names
+
+ The dependency scanning happens depth-first. The returned
+ dependencies should be in the best order to define bblayers.
+
+ names - list of layer names (searching layerItems)
+ branches - when specified (with names) only this list of branches are evaluated
+
+ layerBranches - list of layerBranches to resolve dependencies
+
+ ignores - list of layer names to ignore
+
+ return: (dependencies, invalid)
+
+ dependencies[LayerItem.name] = [ LayerBranch, LayerDependency1, LayerDependency2, ... ]
+ invalid = [ LayerItem.name1, LayerItem.name2, ... ]'''
+
+ invalid = []
+
+ # Convert name/branch to layerBranches
+ if layerbranches is None:
+ layerbranches = []
+
+ for name in names:
+ if ignores and name in ignores:
+ continue
+
+ layerbranch = self.find_layerbranch(name, branches)
+ if not layerbranch:
+ invalid.append(name)
+ else:
+ layerbranches.append(layerbranch)
+
+ for layerbranch in layerbranches:
+ if layerbranch.index != self:
+ raise LayerIndexException("Can not resolve dependencies across indexes with this class function!")
+
+ def _resolve_dependencies(layerbranches, ignores, dependencies, invalid):
+ for layerbranch in layerbranches:
+ if ignores and layerBranch.layer.name in ignores:
+ continue
+
+ for layerdependency in layerbranch.index.layerDependencies_layerBranchId[layerBranch.id]:
+ deplayerbranch = layerDependency.dependency_layerBranch
+
+ if ignores and deplayerbranch.layer.name in ignores:
+ continue
+
+ # New dependency, we need to resolve it now... depth-first
+ if deplayerbranch.layer.name not in dependencies:
+ (dependencies, invalid) = _resolve_dependencies([deplayerbranch], ignores, dependencies, invalid)
+
+ if deplayerbranch.layer.name not in dependencies:
+ dependencies[deplayerbranch.layer.name] = [deplayerbranch, layerdependency]
+ else:
+ if layerdependency not in dependencies[deplayerbranch.layer.name]:
+ dependencies[deplayerbranch.layer.name].append(layerdependency)
+
+ return (dependencies, invalid)
+
+ # OK, resolve this one...
+ dependencies = OrderedDict()
+ (dependencies, invalid) = _resolve_dependencies(layerbranches, ignores, dependencies, invalid)
+
+ # Is this item already in the list, if not add it
+ for layerbranch in layerbranches:
+ if layerbranch.layer.name not in dependencies:
+ dependencies[layerbranch.layer.name] = [layerbranch]
+
+ return (dependencies, invalid)
+
+
+# Define a basic LayerIndexItemObj. This object forms the basis for all other
+# objects. The raw Layer Index data is stored in the _data element, but we
+# do not want users to access data directly. So wrap this and protect it
+# from direct manipulation.
+#
+# It is up to the insantiators of the objects to fill them out, and once done
+# lock the objects to prevent further accidently manipulation.
+#
+# Using the getattr, setattr and properties we can access and manipulate
+# the data within the data element.
+class LayerIndexItemObj():
+ def __init__(self, index, data=None, lock=False):
+ if data is None:
+ data = {}
+
+ if type(data) != type(dict()):
+ raise TypeError('data (%s) is not a dict' % type(data))
+
+ super().__setattr__('_lock', lock)
+ super().__setattr__('index', index)
+ super().__setattr__('_data', data)
+
+ def __eq__(self, other):
+ if self.__class__ != other.__class__:
+ return False
+ res=(self._data == other._data)
+ return res
+
+ def __bool__(self):
+ return bool(self._data)
+
+ def __getattr__(self, name):
+ # These are internal to THIS class, and not part of data
+ if name == "index" or name.startswith('_'):
+ return super().__getattribute__(name)
+
+ if name not in self._data:
+ raise AttributeError('%s not in datastore' % name)
+
+ return self._data[name]
+
+ def _setattr(self, name, value, prop=True):
+ '''__setattr__ like function, but with control over property object behavior'''
+ if self.isLocked():
+ raise TypeError("Can not set attribute '%s': Object data is locked" % name)
+
+ if name.startswith('_'):
+ super().__setattr__(name, value)
+ return
+
+ # Since __setattr__ runs before properties, we need to check if
+ # there is a setter property and then execute it
+ # ... or return self._data[name]
+ propertyobj = getattr(self.__class__, name, None)
+ if prop and isinstance(propertyobj, property):
+ if propertyobj.fset:
+ propertyobj.fset(self, value)
+ else:
+ raise AttributeError('Attribute %s is readonly, and may not be set' % name)
+ else:
+ self._data[name] = value
+
+ def __setattr__(self, name, value):
+ self._setattr(name, value, prop=True)
+
+ def _delattr(self, name, prop=True):
+ # Since __delattr__ runs before properties, we need to check if
+ # there is a deleter property and then execute it
+ # ... or we pop it ourselves..
+ propertyobj = getattr(self.__class__, name, None)
+ if prop and isinstance(propertyobj, property):
+ if propertyobj.fdel:
+ propertyobj.fdel(self)
+ else:
+ raise AttributeError('Attribute %s is readonly, and may not be deleted' % name)
+ else:
+ self._data.pop(name)
+
+ def __delattr__(self, name):
+ self._delattr(name, prop=True)
+
+ def lockData(self):
+ '''Lock data object (make it readonly)'''
+ super().__setattr__("_lock", True)
+
+ def unlockData(self):
+ '''unlock data object (make it readonly)'''
+ super().__setattr__("_lock", False)
+
+ def isLocked(self):
+ '''Is this object locked (readonly)?'''
+ return self._lock
+
+# Branch object
+class Branch(LayerIndexItemObj):
+ def define_data(self, id, name, bitbake_branch,
+ short_description=None, sort_priority=1,
+ updates_enabled=True, updated=None,
+ update_environment=None):
+ self.id = id
+ self.name = name
+ self.bitbake_branch = bitbake_branch
+ self.short_description = short_description or name
+ self.sort_priority = sort_priority
+ self.updates_enabled = updates_enabled
+ self.updated = updated or datetime.datetime.today().isoformat()
+ self.update_environment = update_environment
+
+ @property
+ def name(self):
+ return self.__getattr__('name')
+
+ @name.setter
+ def name(self, value):
+ self._data['name'] = value
+
+ if self.bitbake_branch == value:
+ self.bitbake_branch = ""
+
+ @name.deleter
+ def name(self):
+ self._delattr('name', prop=False)
+
+ @property
+ def bitbake_branch(self):
+ try:
+ return self.__getattr__('bitbake_branch')
+ except AttributeError:
+ return self.name
+
+ @bitbake_branch.setter
+ def bitbake_branch(self, value):
+ if self.name == value:
+ self._data['bitbake_branch'] = ""
+ else:
+ self._data['bitbake_branch'] = value
+
+ @bitbake_branch.deleter
+ def bitbake_branch(self):
+ self._delattr('bitbake_branch', prop=False)
+
+
+class LayerItem(LayerIndexItemObj):
+ def define_data(self, id, name, status='P',
+ layer_type='A', summary=None,
+ description=None,
+ vcs_url=None, vcs_web_url=None,
+ vcs_web_tree_base_url=None,
+ vcs_web_file_base_url=None,
+ usage_url=None,
+ mailing_list_url=None,
+ index_preference=1,
+ classic=False,
+ updated=None):
+ self.id = id
+ self.name = name
+ self.status = status
+ self.layer_type = layer_type
+ self.summary = summary or name
+ self.description = description or summary or name
+ self.vcs_url = vcs_url
+ self.vcs_web_url = vcs_web_url
+ self.vcs_web_tree_base_url = vcs_web_tree_base_url
+ self.vcs_web_file_base_url = vcs_web_file_base_url
+ self.index_preference = index_preference
+ self.classic = classic
+ self.updated = updated or datetime.datetime.today().isoformat()
+
+
+class LayerBranch(LayerIndexItemObj):
+ def define_data(self, id, collection, version, layer, branch,
+ vcs_subdir="", vcs_last_fetch=None,
+ vcs_last_rev=None, vcs_last_commit=None,
+ actual_branch="",
+ updated=None):
+ self.id = id
+ self.collection = collection
+ self.version = version
+ if isinstance(layer, LayerItem):
+ self.layer = layer
+ else:
+ self.layer_id = layer
+
+ if isinstance(branch, Branch):
+ self.branch = branch
+ else:
+ self.branch_id = branch
+
+ self.vcs_subdir = vcs_subdir
+ self.vcs_last_fetch = vcs_last_fetch
+ self.vcs_last_rev = vcs_last_rev
+ self.vcs_last_commit = vcs_last_commit
+ self.actual_branch = actual_branch
+ self.updated = updated or datetime.datetime.today().isoformat()
+
+ # This is a little odd, the _data attribute is 'layer', but it's really
+ # referring to the layer id.. so lets adjust this to make it useful
+ @property
+ def layer_id(self):
+ return self.__getattr__('layer')
+
+ @layer_id.setter
+ def layer_id(self, value):
+ self._setattr('layer', value, prop=False)
+
+ @layer_id.deleter
+ def layer_id(self):
+ self._delattr('layer', prop=False)
+
+ @property
+ def layer(self):
+ try:
+ return self.index.layerItems[self.layer_id]
+ except KeyError:
+ raise AttributeError('Unable to find layerItems in index to map layer_id %s' % self.layer_id)
+ except IndexError:
+ raise AttributeError('Unable to find layer_id %s in index layerItems' % self.layer_id)
+
+ @layer.setter
+ def layer(self, value):
+ if not isinstance(value, LayerItem):
+ raise TypeError('value is not a LayerItem')
+ if self.index != value.index:
+ raise AttributeError('Object and value do not share the same index and thus key set.')
+ self.layer_id = value.id
+
+ @layer.deleter
+ def layer(self):
+ del self.layer_id
+
+ @property
+ def branch_id(self):
+ return self.__getattr__('branch')
+
+ @branch_id.setter
+ def branch_id(self, value):
+ self._setattr('branch', value, prop=False)
+
+ @branch_id.deleter
+ def branch_id(self):
+ self._delattr('branch', prop=False)
+
+ @property
+ def branch(self):
+ try:
+ logger.debug(1, "Get branch object from branches[%s]" % (self.branch_id))
+ return self.index.branches[self.branch_id]
+ except KeyError:
+ raise AttributeError('Unable to find branches in index to map branch_id %s' % self.branch_id)
+ except IndexError:
+ raise AttributeError('Unable to find branch_id %s in index branches' % self.branch_id)
+
+ @branch.setter
+ def branch(self, value):
+ if not isinstance(value, LayerItem):
+ raise TypeError('value is not a LayerItem')
+ if self.index != value.index:
+ raise AttributeError('Object and value do not share the same index and thus key set.')
+ self.branch_id = value.id
+
+ @branch.deleter
+ def branch(self):
+ del self.branch_id
+
+ @property
+ def actual_branch(self):
+ if self.__getattr__('actual_branch'):
+ return self.__getattr__('actual_branch')
+ else:
+ return self.branch.name
+
+ @actual_branch.setter
+ def actual_branch(self, value):
+ logger.debug(1, "Set actual_branch to %s .. name is %s" % (value, self.branch.name))
+ if value != self.branch.name:
+ self._setattr('actual_branch', value, prop=False)
+ else:
+ self._setattr('actual_branch', '', prop=False)
+
+ @actual_branch.deleter
+ def actual_branch(self):
+ self._delattr('actual_branch', prop=False)
+
+# Extend LayerIndexItemObj with common LayerBranch manipulations
+# All of the remaining LayerIndex objects refer to layerbranch, and it is
+# up to the user to follow that back through the LayerBranch object into
+# the layer object to get various attributes. So add an intermediate set
+# of attributes that can easily get us the layerbranch as well as layer.
+
+class LayerIndexItemObj_LayerBranch(LayerIndexItemObj):
+ @property
+ def layerbranch_id(self):
+ return self.__getattr__('layerbranch')
+
+ @layerbranch_id.setter
+ def layerbranch_id(self, value):
+ self._setattr('layerbranch', value, prop=False)
+
+ @layerbranch_id.deleter
+ def layerbranch_id(self):
+ self._delattr('layerbranch', prop=False)
+
+ @property
+ def layerbranch(self):
+ try:
+ return self.index.layerBranches[self.layerbranch_id]
+ except KeyError:
+ raise AttributeError('Unable to find layerBranches in index to map layerbranch_id %s' % self.layerbranch_id)
+ except IndexError:
+ raise AttributeError('Unable to find layerbranch_id %s in index branches' % self.layerbranch_id)
+
+ @layerbranch.setter
+ def layerbranch(self, value):
+ if not isinstance(value, LayerBranch):
+ raise TypeError('value (%s) is not a layerBranch' % type(value))
+ if self.index != value.index:
+ raise AttributeError('Object and value do not share the same index and thus key set.')
+ self.layerbranch_id = value.id
+
+ @layerbranch.deleter
+ def layerbranch(self):
+ del self.layerbranch_id
+
+ @property
+ def layer_id(self):
+ return self.layerbranch.layer_id
+
+ # Doesn't make sense to set or delete layer_id
+
+ @property
+ def layer(self):
+ return self.layerbranch.layer
+
+ # Doesn't make sense to set or delete layer
+
+
+class LayerDependency(LayerIndexItemObj_LayerBranch):
+ def define_data(self, id, layerbranch, dependency, required=True):
+ self.id = id
+ if isinstance(layerbranch, LayerBranch):
+ self.layerbranch = layerbranch
+ else:
+ self.layerbranch_id = layerbranch
+ if isinstance(dependency, LayerDependency):
+ self.dependency = dependency
+ else:
+ self.dependency_id = dependency
+ self.required = required
+
+ @property
+ def dependency_id(self):
+ return self.__getattr__('dependency')
+
+ @dependency_id.setter
+ def dependency_id(self, value):
+ self._setattr('dependency', value, prop=False)
+
+ @dependency_id.deleter
+ def dependency_id(self):
+ self._delattr('dependency', prop=False)
+
+ @property
+ def dependency(self):
+ try:
+ return self.index.layerItems[self.dependency_id]
+ except KeyError:
+ raise AttributeError('Unable to find layerItems in index to map layerbranch_id %s' % self.dependency_id)
+ except IndexError:
+ raise AttributeError('Unable to find dependency_id %s in index layerItems' % self.dependency_id)
+
+ @dependency.setter
+ def dependency(self, value):
+ if not isinstance(value, LayerDependency):
+ raise TypeError('value (%s) is not a dependency' % type(value))
+ if self.index != value.index:
+ raise AttributeError('Object and value do not share the same index and thus key set.')
+ self.dependency_id = value.id
+
+ @dependency.deleter
+ def dependency(self):
+ self._delattr('dependency', prop=False)
+
+ @property
+ def dependency_layerBranch(self):
+ layerid = self.dependency_id
+ branchid = self.layerbranch.branch_id
+
+ try:
+ return self.index.layerBranches_layerId_branchId["%s:%s" % (layerid, branchid)]
+ except IndexError:
+ # layerBranches_layerId_branchId -- but not layerId:branchId
+ raise AttributeError('Unable to find layerId:branchId %s:%s in index layerBranches_layerId_branchId' % (layerid, branchid))
+ except KeyError:
+ raise AttributeError('Unable to find layerId:branchId %s:%s in layerItems and layerBranches' % (layerid, branchid))
+
+ # dependency_layerBranch doesn't make sense to set or del
+
+
+class Recipe(LayerIndexItemObj_LayerBranch):
+ def define_data(self, id,
+ filename, filepath, pn, pv, layerbranch,
+ summary="", description="", section="", license="",
+ homepage="", bugtracker="", provides="", bbclassextend="",
+ inherits="", blacklisted="", updated=None):
+ self.id = id
+ self.filename = filename
+ self.filepath = filepath
+ self.pn = pn
+ self.pv = pv
+ self.summary = summary
+ self.description = description
+ self.section = section
+ self.license = license
+ self.homepage = homepage
+ self.bugtracker = bugtracker
+ self.provides = provides
+ self.bbclassextend = bbclassextend
+ self.inherits = inherits
+ self.updated = updated or datetime.datetime.today().isoformat()
+ self.blacklisted = blacklisted
+ if isinstance(layerbranch, LayerBranch):
+ self.layerbranch = layerbranch
+ else:
+ self.layerbranch_id = layerbranch
+
+ @property
+ def fullpath(self):
+ return os.path.join(self.filepath, self.filename)
+
+ # Set would need to understand how to split it
+ # del would we del both parts?
+
+ @property
+ def inherits(self):
+ if 'inherits' not in self._data:
+ # Older indexes may not have this, so emulate it
+ if '-image-' in self.pn:
+ return 'image'
+ return self.__getattr__('inherits')
+
+ @inherits.setter
+ def inherits(self, value):
+ return self._setattr('inherits', value, prop=False)
+
+ @inherits.deleter
+ def inherits(self):
+ return self._delattr('inherits', prop=False)
+
+
+class Machine(LayerIndexItemObj_LayerBranch):
+ def define_data(self, id,
+ name, description, layerbranch,
+ updated=None):
+ self.id = id
+ self.name = name
+ self.description = description
+ if isinstance(layerbranch, LayerBranch):
+ self.layerbranch = layerbranch
+ else:
+ self.layerbranch_id = layerbranch
+ self.updated = updated or datetime.datetime.today().isoformat()
+
+class Distro(LayerIndexItemObj_LayerBranch):
+ def define_data(self, id,
+ name, description, layerbranch,
+ updated=None):
+ self.id = id
+ self.name = name
+ self.description = description
+ if isinstance(layerbranch, LayerBranch):
+ self.layerbranch = layerbranch
+ else:
+ self.layerbranch_id = layerbranch
+ self.updated = updated or datetime.datetime.today().isoformat()
+
+# When performing certain actions, we may need to sort the data.
+# This will allow us to keep it consistent from run to run.
+def sort_entry(item):
+ newitem = item
+ try:
+ if type(newitem) == type(dict()):
+ newitem = OrderedDict(sorted(newitem.items(), key=lambda t: t[0]))
+ for index in newitem:
+ newitem[index] = sort_entry(newitem[index])
+ elif type(newitem) == type(list()):
+ newitem.sort(key=lambda obj: obj['id'])
+ for index, _ in enumerate(newitem):
+ newitem[index] = sort_entry(newitem[index])
+ except:
+ logger.error('Sort failed for item %s' % type(item))
+ pass
+
+ return newitem