def test_logging():
l = Logging("INFO")
l.name = "test"
l.info("test")
l.debug("test")
l.warning("test")
l.error("test")
l.critical("test")
for level in ['DEBUG', 'INFO', 'ERROR', 'WARNING', 'CRITICAL']:
l.level = level
assert l.level == level
l.level = True
l.level = False
for x in [10, 20, 30, 40, 50]:
l.level = x
try:
l.level = "WARN"
assert Fales
except:
assert True
# FIXME is this working ??wierd syntax in loggibg_tools.
import copy
copy.copy(l)
copy.deepcopy(l)
class RPackageManager(object):
"""Implements a R package manager from Python
So far you can install a package (from source, or CRAN, or biocLite)
::
pm = PackageManager()
[(x, pm.installed[x][2]) for x in pm.installed.keys()]
You can access to all information within a dataframe called **packages** where
indices are the name packages. Some aliases are provided as attributes (e.g., available,
installed)
"""
cran_repos = "http://cran.univ-lyon1.fr/"
def __init__(self, verbose=True):
self.session = RSession()
self.logging = Logging(verbose)
self.logging.info('Fetching package information')
self.update()
def _update(self):
# local import ?
import numpy
import pandas
# figure out the installed packages first
code = """rvar_packages = as.data.frame(installed.packages())"""
self.session.run(code)
s = self.session.rvar_packages
# FIXME. these 4 lines are needed as a hack related to pyper.
try:
s = s.replace("\n", "")
df = eval(s)
except:
df = s
df.set_index('Package', inplace=True)
self._packages = df.copy()
# Now, fetch was is possible to install from the default cran repo
code = """rvar_status=packageStatus(repos="%s/src/contrib")"""
code = code % self.cran_repos
self.session.run(code)
s = self.session.rvar_status
# FIXME.
try:
s = s.replace("\n", "")
res = eval(s)
except:
res = s
res['inst'].set_index('Package', inplace=True)
res['avail'].set_index('Package', inplace=True)
self._status = res
def update(self):
"""If you install/remove packages yourself elsewhere, you may need to
call this function to update the package manager"""
try:
#self.session.reconnect()
self._update()
except:
self.logging.warning("Could not update the packages. Call update() again")
def _compat_version(self, version):
return version.replace("-", "a")
def _get_installed(self):
# we do not buffer because packages may be removed manually or from R of
# using remove_packages method, ....
#self._package_status()
return self._status['inst']
installed = property(_get_installed, "returns list of packages installed as a dataframe")
def _get_available(self):
# we do not buffer because packages may be removed manually or from R of
# using remove_packages method, ....
#self._package_status()
return self._status['avail']
available = property(_get_available, "returns list of packages available as a dataframe")
def _get_packages(self):
# do not buffer since it may change in many places
return self._packages
packages = property(_get_packages)
def get_package_latest_version(self, package):
"""Get latest version available of a package"""
return self.available['Version'].ix[package]
def get_package_version(self, package):
"""Get version of an install package"""
if package not in self.installed.index:
self.logging.error("package {0} not installed".format(package))
return self.installed['Version'].ix[package]
def biocLite(self, package=None, suppressUpdates=True, verbose=False):
"""Installs one or more biocLite packages
:param package: a package name (string) or list of package names (list of
strings) that will be installed from BioConductor. If package is set
to None, all packages already installed will be updated.
"""
if isinstance(package, str):
if package not in self.installed.index:
biocLite(package, suppressUpdates, verbose=verbose)
elif isinstance(package, list):
for pkg in package:
self.logging.info("Installing %s" % pkg)
if self.is_installed(pkg) is False:
biocLite(pkg, suppressUpdates, verbose=verbose)
else: # trying other cases (e.g., None updates biocLite itself).
biocLite(package, suppressUpdates, verbose=verbose)
self.update()
def _isLocal(self, pkg):
if os.path.exists(pkg):
return True
else:
return False
def remove(self, package):
"""Remove a package (or list) from local repository"""
rcode ="""remove.packages("%s")"""
if isinstance(package, str):
package = [package]
for pkg in package:
if pkg in self.installed.index:
self.session(rcode % pkg)
else:
self.logging.warning("Package not found. Nothing to remove")
self.update()
def require(self, pkg, version):
"Check if a package with given version is available"
if pkg not in self.installed.index:
self.logging.info("Package %s not installed" % pkg)
return False
currentVersion = self.packageVersion(pkg)
if self._get_version(currentVersion) >= self._get_version(version):
return True
else:
return False
def _install_package(self, packageName, dependencies=True):
"""Installs one or more CRAN packages
.. todo:: check if it is already available to prevent renstallation ?
"""
repos = self.cran_repos
# if this is a source file we want to reset the repo
if isinstance(packageName, str):
packageName = [packageName]
for pkg in packageName:
if self.is_installed(pkg) is False:
self.logging.info("Package not found. Installing %s..." % pkg)
install_package(pkg, dependencies=dependencies,
repos=repos)
else:
self.logging.info("Package %s found. " % pkg)
install_package(pkg, dependencies=dependencies,
repos=repos)
self.update()
def install(self, pkg, require=None, update=True, reinstall=False):
"""install a package automatically scanning CRAN and biocLite repos
if require is not set and update is True, when a newest version of a package
is available, it is installed
"""
from easydev import to_list
pkgs = to_list(pkg)
for pkg in pkgs:
self._install(pkg, require=require, update=update, reinstall=reinstall)
def _install(self, pkg, require=None, update=update, reinstall=False):
# LOCAL file
if self._isLocal(pkg):
# if a local file, we do not want to jump to biocLite or CRAN. Let
# us install it directly. We cannot check version yet so we will
# overwrite what is already installed
self.logging.warning("Installing from source")
self._install_package(pkg)
return
# From CRAN
if self.is_installed(pkg):
currentVersion = self.get_package_version(pkg)
# if not provided, require should be the latest version
if require is None and update is True:
try:
require = self.get_package_latest_version(pkg)
except:
# a non-cran package (bioclite maybe)
pass
if require is None:
self.logging.info("%s already installed with version %s" % \
(pkg, currentVersion))
return
# if require is not none, is it the required version ?
if self._get_version(currentVersion) >= self._get_version(require) and reinstall is False:
self.logging.info("%s already installed with required version %s" \
% (pkg, currentVersion))
# if so, nothing to do
else:
# Try updating
self.logging.info("Updating")
self._install_package(pkg)
if require is None:
return
currentVersion = self.get_package_version(pkg)
if self._get_version(currentVersion) < self._get_version(require):
self.logging.warning("%s installed but current version (%s) does not fulfill your requirement" % \
(pkg, currentVersion))
elif pkg in self.available.index:
self._install_package(pkg)
else:
# maybe a biocLite package ?
# require is ignored. The latest will be installed
self.logging.info("Trying to find the package on bioconductor")
self.biocLite(pkg)
if require is None:
return
currentVersion = self.get_package_version(pkg)
if self._get_version(currentVersion) >= self._get_version(require):
self.logging.warning("%s installed but version is %s too small (even after update)" % \
(pkg, currentVersion, require))
def _get_version(self, version):
# some pacakge do not use the correct version convention
try:
return StrictVersion(version)
except:
try:
return StrictVersion(version.replace("-", "a"))
except:
# snowfall package example was 1.86-6.1
# This becomes 1.86a61 which is not great but not workaround
# for now
left, right = version.split("-")
version = left + "a" + right.replace('.', '')
return StrictVersion(version)
def is_installed(self, pkg_name):
if pkg_name in self.installed.index:
return True
else:
return False
class RPackageManager(object):
"""Implements a R package manager from Python
So far you can install a package (from source, or CRAN, or biocLite)
::
pm = PackageManager()
[(x, pm.installed[x][2]) for x in pm.installed.keys()]
You can access to all information within a dataframe called **packages** where
indices are the name packages. Some aliases are provided as attributes (e.g., available,
installed)
"""
cran_repos = "http://cran.univ-lyon1.fr/"
def __init__(self, verbose=True):
self.session = RSession()
self.logging = Logging(verbose)
self.logging.info('Fetching package information')
self.update()
def _update(self):
# local import ?
import numpy
import pandas
# figure out the installed packages first
code = """rvar_packages = as.data.frame(installed.packages())"""
self.session.run(code)
s = self.session.rvar_packages
# FIXME. these 4 lines are needed as a hack related to pyper.
try:
s = s.replace("\n", "")
df = eval(s)
except:
df = s
df.set_index('Package', inplace=True)
self._packages = df.copy()
# Now, fetch was is possible to install from the default cran repo
code = """rvar_status=packageStatus(repos="%s/src/contrib")"""
code = code % self.cran_repos
self.session.run(code)
s = self.session.rvar_status
# FIXME.
try:
s = s.replace("\n", "")
res = eval(s)
except:
res = s
res['inst'].set_index('Package', inplace=True)
res['avail'].set_index('Package', inplace=True)
self._status = res
def update(self):
"""If you install/remove packages yourself elsewhere, you may need to
call this function to update the package manager"""
try:
#self.session.reconnect()
self._update()
except:
self.logging.warning("Could not update the packages. Call update() again")
def _compat_version(self, version):
return version.replace("-", "a")
def _get_installed(self):
# we do not buffer because packages may be removed manually or from R of
# using remove_packages method, ....
#self._package_status()
return self._status['inst']
installed = property(_get_installed, "returns list of packages installed as a dataframe")
def _get_available(self):
# we do not buffer because packages may be removed manually or from R of
# using remove_packages method, ....
#self._package_status()
return self._status['avail']
available = property(_get_available, "returns list of packages available as a dataframe")
def _get_packages(self):
# do not buffer since it may change in many places
return self._packages
packages = property(_get_packages)
def get_package_latest_version(self, package):
"""Get latest version available of a package"""
return self.available['Version'].ix[package]
def get_package_version(self, package):
"""Get version of an install package"""
if package not in self.installed.index:
self.logging.error("package {0} not installed".format(package))
return self.installed['Version'].ix[package]
def biocLite(self, package=None, suppressUpdates=True, verbose=False):
"""Installs one or more biocLite packages
:param package: a package name (string) or list of package names (list of
strings) that will be installed from BioConductor. If package is set
to None, all packages already installed will be updated.
"""
if isinstance(package, str):
if package not in self.installed.index:
biocLite(package, suppressUpdates, verbose=verbose)
elif isinstance(package, list):
for pkg in package:
self.logging.info("Installing %s" % pkg)
if self.is_installed(pkg) is False:
biocLite(pkg, suppressUpdates, verbose=verbose)
else: # trying other cases (e.g., None updates biocLite itself).
biocLite(package, suppressUpdates, verbose=verbose)
self.update()
def _isLocal(self, pkg):
if os.path.exists(pkg):
return True
else:
return False
def remove(self, package):
"""Remove a package (or list) from local repository"""
rcode ="""remove.packages("%s")"""
if isinstance(package, str):
package = [package]
for pkg in package:
if pkg in self.installed.index:
self.session(rcode % pkg)
else:
self.logging.warning("Package not found. Nothing to remove")
self.update()
def require(self, pkg, version):
"Check if a package with given version is available"
if pkg not in self.installed.index:
self.logging.info("Package %s not installed" % pkg)
return False
currentVersion = self.packageVersion(pkg)
if self._get_version(currentVersion) >= self._get_version(version):
return True
else:
return False
def _install_package(self, packageName, dependencies=True):
"""Installs one or more CRAN packages
.. todo:: check if it is already available to prevent renstallation ?
"""
repos = self.cran_repos
# if this is a source file we want to reset the repo
if isinstance(packageName, str):
packageName = [packageName]
for pkg in packageName:
if self.is_installed(pkg) is False:
self.logging.info("Package not found. Installing %s..." % pkg)
install_package(pkg, dependencies=dependencies,
repos=repos)
else:
self.logging.info("Package %s found. " % pkg)
install_package(pkg, dependencies=dependencies,
repos=repos)
self.update()
def install(self, pkg, require=None, update=True, reinstall=False):
"""install a package automatically scanning CRAN and biocLite repos
if require is not set and update is True, when a newest version of a package
is available, it is installed
"""
from easydev import to_list
pkgs = to_list(pkg)
for pkg in pkgs:
self._install(pkg, require=require, update=update, reinstall=reinstall)
def _install(self, pkg, require=None, update=update, reinstall=False):
# LOCAL file
if self._isLocal(pkg):
# if a local file, we do not want to jump to biocLite or CRAN. Let
# us install it directly. We cannot check version yet so we will
# overwrite what is already installed
self.logging.warning("Installing from source")
self._install_package(pkg)
return
# From CRAN
if self.is_installed(pkg):
currentVersion = self.get_package_version(pkg)
# if not provided, require should be the latest version
if require is None and update is True:
try:
require = self.get_package_latest_version(pkg)
except:
# a non-cran package (bioclite maybe)
pass
if require is None:
self.logging.info("%s already installed with version %s" % \
(pkg, currentVersion))
return
# if require is not none, is it the required version ?
if self._get_version(currentVersion) >= self._get_version(require) and reinstall is False:
self.logging.info("%s already installed with required version %s" \
% (pkg, currentVersion))
# if so, nothing to do
else:
# Try updating
self.logging.info("Updating")
self._install_package(pkg)
if require is None:
return
currentVersion = self.get_package_version(pkg)
if self._get_version(currentVersion) < self._get_version(require):
self.logging.warning("%s installed but current version (%s) does not fulfill your requirement" % \
(pkg, currentVersion))
elif pkg in self.available.index:
self._install_package(pkg)
else:
# maybe a biocLite package ?
# require is ignored. The latest will be installed
self.logging.info("Trying to find the package on bioconductor")
self.biocLite(pkg)
if require is None:
return
currentVersion = self.get_package_version(pkg)
if self._get_version(currentVersion) >= self._get_version(require):
self.logging.warning("%s installed but version is %s too small (even after update)" % \
(pkg, currentVersion, require))
def _get_version(self, version):
# some pacakge do not use the correct version convention
try:
return StrictVersion(version)
except:
try:
return StrictVersion(version.replace("-", "a"))
except:
# snowfall package example was 1.86-6.1
# This becomes 1.86a61 which is not great but not workaround
# for now
left, right = version.split("-")
version = left + "a" + right.replace('.', '')
return StrictVersion(version)
def is_installed(self, pkg_name):
if pkg_name in self.installed.index:
return True
else:
return False
class Service(object):
"""Base class for WSDL and REST classes
.. seealso:: :class:`REST`, :class:`WSDLService`
"""
#: some useful response codes
response_codes = {
200: 'OK',
201: 'Created',
400: 'Bad Request. There is a problem with your input',
404: 'Not found. The resource you requests does not exist',
405: 'Method not allowed',
406: "Not Acceptable. Usually headers issue",
410: 'Gone. The resource you requested was removed.',
415: "Unsupported Media Type",
500: 'Internal server error. Most likely a temporary problem',
503:
'Service not available. The server is being updated, try again later'
}
def __init__(self, name, url=None, verbose=True, requests_per_sec=10):
""".. rubric:: Constructor
:param str name: a name for this service
:param str url: its URL
:param bool verbose: prints informative messages if True (default is
True)
:param requests_per_sec: maximum number of requests per seconds
are restricted to 3. You can change that value. If you reach the
limit, an error is raise. The reason for this limitation is
that some services (e.g.., NCBI) may black list you IP.
If you need or can do more (e.g., ChEMBL does not seem to have
restrictions), change the value. You can also have several instance
but again, if you send too many requests at the same, your future
requests may be retricted. Currently implemented for REST only
All instances have an attribute called :attr:`~Service.logging` that
is an instanceof the :mod:`logging` module. It can be used to print
information, warning, error messages::
self.logging.info("informative message")
self.logging.warning("warning message")
self.logging.error("error message")
The attribute :attr:`~Service.debugLevel` can be used to set the behaviour
of the logging messages. If the argument verbose is True, the debugLebel
is set to INFO. If verbose if False, the debugLevel is set to WARNING.
However, you can use the :attr:`debugLevel` attribute to change it to
one of DEBUG, INFO, WARNING, ERROR, CRITICAL. debugLevel=WARNING means
that only WARNING, ERROR and CRITICAL messages are shown.
"""
super(Service, self).__init__()
self.requests_per_sec = requests_per_sec
self.name = name
self.logging = Logging("bioservices:%s" % self.name, verbose)
self._url = url
try:
if self.url is not None:
urlopen(self.url)
except Exception as err:
self.logging.warning("The URL (%s) provided cannot be reached." %
self.url)
self._easyXMLConversion = True
# used by HGNC where some XML contains non-utf-8 characters !!
# should be able to fix it with requests once HGNC works again
#self._fixing_unicode = False
#self._fixing_encoding = "utf-8"
self.devtools = DevTools()
self.settings = BioServicesConfig()
self._last_call = 0
def _calls(self):
time_lapse = 1. / self.requests_per_sec
current_time = time.time()
dt = current_time - self._last_call
if self._last_call == 0:
self._last_call = current_time
return
else:
self._last_call = current_time
if dt > time_lapse:
return
else:
time.sleep(time_lapse - dt)
def _get_caching(self):
return self.settings.params['cache.on'][0]
def _set_caching(self, caching):
self.devtools.check_param_in_list(caching, [True, False])
self.settings.params['cache.on'][0] = caching
# reset the session, which will be automatically created if we
# access to the session attribute
self._session = None
CACHING = property(_get_caching, _set_caching)
def _get_url(self):
return self._url
def _set_url(self, url):
# something more clever here to check the URL e.g. starts with http
if url is not None:
url = url.rstrip("/")
self._url = url
url = property(_get_url, _set_url, doc="URL of this service")
def _get_easyXMLConversion(self):
return self._easyXMLConversion
def _set_easyXMLConversion(self, value):
if isinstance(value, bool) is False:
raise TypeError("value must be a boolean value (True/False)")
self._easyXMLConversion = value
easyXMLConversion = property(
_get_easyXMLConversion,
_set_easyXMLConversion,
doc=
"""If True, xml output from a request are converted to easyXML object (Default behaviour)."""
)
def easyXML(self, res):
"""Use this method to convert a XML document into an
:class:`~bioservices.xmltools.easyXML` object
The easyXML object provides utilities to ease access to the XML
tag/attributes.
Here is a simple example starting from the following XML
.. doctest::
>>> from bioservices import *
>>> doc = "<xml> <id>1</id> <id>2</id> </xml>"
>>> s = Service("name")
>>> res = s.easyXML(doc)
>>> res.findAll("id")
[<id>1</id>, <id>2</id>]
"""
from bioservices import xmltools
return xmltools.easyXML(res)
def __str__(self):
txt = "This is an instance of %s service" % self.name
return txt
def pubmed(self, Id):
"""Open a pubmed Id into a browser tab
:param Id: a valid pubmed Id in string or integer format.
The URL is a concatenation of the pubmed URL
http://www.ncbi.nlm.nih.gov/pubmed/ and the provided Id.
"""
url = "http://www.ncbi.nlm.nih.gov/pubmed/"
import webbrowser
webbrowser.open(url + str(Id))
def on_web(self, url):
"""Open a URL into a browser"""
import webbrowser
webbrowser.open(url)
def save_str_to_image(self, data, filename):
"""Save string object into a file converting into binary"""
with open(filename, 'wb') as f:
import binascii
try:
#python3
newres = binascii.a2b_base64(bytes(data, "utf-8"))
except:
newres = binascii.a2b_base64(data)
f.write(newres)
class Service(object):
"""Base class for WSDL and REST classes
.. seealso:: :class:`REST`, :class:`WSDLService`
"""
#: some useful response codes
response_codes = {
200: 'OK',
201: 'Created',
400: 'Bad Request. There is a problem with your input',
404: 'Not found. The resource you requests does not exist',
405: 'Method not allowed',
406: "Not Acceptable. Usually headers issue",
410: 'Gone. The resource you requested was removed.',
415: "Unsupported Media Type",
500: 'Internal server error. Most likely a temporary problem',
503: 'Service not available. The server is being updated, try again later'
}
def __init__(self, name, url=None, verbose=True, requests_per_sec=3):
""".. rubric:: Constructor
:param str name: a name for this service
:param str url: its URL
:param bool verbose: prints informative messages if True (default is
True)
:param requests_per_sec: maximum number of requests per seconds
are restricted to 3. You can change that value. If you reach the
limit, an error is raise. The reason for this limitation is
that some services (e.g.., NCBI) may black list you IP.
If you need or can do more (e.g., ChEMBL does not seem to have
restrictions), change the value. You can also have several instance
but again, if you send too many requests at the same, your future
requests may be retricted. Currently implemented for REST only
All instances have an attribute called :attr:`~Service.logging` that
is an instanceof the :mod:`logging` module. It can be used to print
information, warning, error messages::
self.logging.info("informative message")
self.logging.warning("warning message")
self.logging.error("error message")
The attribute :attr:`~Service.debugLevel` can be used to set the behaviour
of the logging messages. If the argument verbose is True, the debugLebel
is set to INFO. If verbose if False, the debugLevel is set to WARNING.
However, you can use the :attr:`debugLevel` attribute to change it to
one of DEBUG, INFO, WARNING, ERROR, CRITICAL. debugLevel=WARNING means
that only WARNING, ERROR and CRITICAL messages are shown.
"""
super(Service, self).__init__()
self.requests_per_sec = requests_per_sec
self.name = name
self.logging = Logging("bioservices:%s" % self.name, verbose)
self._url = url
try:
if self.url is not None:
urlopen(self.url)
except Exception as err:
self.logging.warning("The URL (%s) provided cannot be reached." % self.url)
self._easyXMLConversion = True
# used by HGNC where some XML contains non-utf-8 characters !!
# should be able to fix it with requests once HGNC works again
#self._fixing_unicode = False
#self._fixing_encoding = "utf-8"
self.devtools = DevTools()
self.settings = BioServicesConfig()
def _get_caching(self):
return self.settings.params['cache.on'][0]
def _set_caching(self, caching):
self.devtools.check_param_in_list(caching, [True, False])
self.settings.params['cache.on'][0] = caching
# reset the session, which will be automatically created if we
# access to the session attribute
self._session = None
CACHING = property(_get_caching, _set_caching)
def _get_url(self):
return self._url
def _set_url(self, url):
# something more clever here to check the URL e.g. starts with http
if url is not None:
url = url.rstrip("/")
self._url = url
url = property(_get_url, _set_url, doc="URL of this service")
def _get_easyXMLConversion(self):
return self._easyXMLConversion
def _set_easyXMLConversion(self, value):
if isinstance(value, bool) is False:
raise TypeError("value must be a boolean value (True/False)")
self._easyXMLConversion = value
easyXMLConversion = property(_get_easyXMLConversion,
_set_easyXMLConversion,
doc="""If True, xml output from a request are converted to easyXML object (Default behaviour).""")
def easyXML(self, res):
"""Use this method to convert a XML document into an
:class:`~bioservices.xmltools.easyXML` object
The easyXML object provides utilities to ease access to the XML
tag/attributes.
Here is a simple example starting from the following XML
.. doctest::
>>> from bioservices import *
>>> doc = "<xml> <id>1</id> <id>2</id> </xml>"
>>> s = Service("name")
>>> res = s.easyXML(doc)
>>> res.findAll("id")
[<id>1</id>, <id>2</id>]
"""
from bioservices import xmltools
return xmltools.easyXML(res)
def __str__(self):
txt = "This is an instance of %s service" % self.name
return txt
def pubmed(self, Id):
"""Open a pubmed Id into a browser tab
:param Id: a valid pubmed Id in string or integer format.
The URL is a concatenation of the pubmed URL
http://www.ncbi.nlm.nih.gov/pubmed/ and the provided Id.
"""
url = "http://www.ncbi.nlm.nih.gov/pubmed/"
import webbrowser
webbrowser.open(url + str(Id))
def on_web(self, url):
"""Open a URL into a browser"""
import webbrowser
webbrowser.open(url)
def save_str_to_image(self, data, filename):
"""Save string object into a file converting into binary"""
with open(filename,'wb') as f:
import binascii
try:
#python3
newres = binascii.a2b_base64(bytes(data, "utf-8"))
except:
newres = binascii.a2b_base64(data)
f.write(newres)
