This package connects the bfabric system to the python and R world while providing a JSON and REST interface using Flask. The bfabricShiny R package is an extension and provides code snippets and sample implementation for a seamless R shiny bfabric integration. For more advanced users the bfabricPy package also provides a powerful query interface on the command-line though using the provided scripts.
The package can be installed like any other Python package, so if you are familiar you might not need to read this section. Currently, it's only available from GitHub.
The best way to install the package depends on your use case, i.e. whether you want to:
- Use the command line scripts
- Use the Python API
- Develop on the package
The command line scripts are currently included in all cases.
To use the command line scripts, it's recommended to install bfabricPy with pipx.
If you don't have pipx installed, refer to the pipx documentation for instructions.
You can execute a command using a specific version of bfabricPy with the pipx run command.
This command handles the dependencies of multiple concurrent installations:
pipx run --spec "git+https://github.com/fgcz/[email protected]" bfabric_read.py --helpTo install a specific version of bfabricPy on your system and make the command available without pipx run prefix, use the following command:
pipx install "git+https://github.com/fgcz/[email protected]"
bfabric_read.py --helpIf you're interested in using the Python API of bfabricPy, you have two options:
[project]
dependencies = [
"bfabricPy @ git+https://github.com/fgcz/bfabricPy.git@main"
]pip install git+https://github.com/fgcz/bfabricPy.gitAs a bfabricPy developer: (i.e. an editable install)
pip install -e ".[dev]"
Create a file as follows: (note: the password is not your login password, but the web service password)
# ~/.bfabricpy.yml
GENERAL:
default_config: PRODUCTION
PRODUCTION:
login: yourBfabricLogin
password: yourBfabricWebPassword
base_url: https://fgcz-bfabric.uzh.ch/bfabric
You can also include an additional config for the TEST instance
TEST:
login: yourBfabricLogin
password: yourBfabricWebPassword
base_url: https://fgcz-bfabric-test.uzh.ch/bfabric
bfabric_read.py storage
bfabric_read.py application
Simple database query examples
bfabric_read.py user login cpanse
bfabric_read.py project id 3000
bfabric_read.py workunit id 199387
bfabric_read.py sample name autoQC4L
bfabric_read.py workunit status running
bfabric_read.py workunit status pending
bfabric_read.py workunit status failed
# list empty resources
bfabric_read.py resource filechecksum d41d8cd98f00b204e9800998ecf8427e
Using the Python API:
from bfabric import Bfabric
client = Bfabric.from.config()
user = B.read(endpoint = 'user', obj={'login': 'cpanse'})
resource = B.read(endpoint = 'resource', obj={'id': 550327 })
bfabric_save_workunit_attribute.py 199387 status available
import json
rv = client.save('workunit', {'id': 254063, 'status': 'available'})
print(json.dumps(rv.to_list_dict(), indent=2))
Find empty resource files in bfabric
bfabric_read.py resource filechecksum `md5sum < /dev/null | cut -c-32` \
| cat -n \
| tail
Please be advised that integration tests will write to the TEST instance configured in your ~/.bfabricpy.yml config file.
Run unit tests:
python3 -m unittest discover -s "bfabric/tests/unit"
Run integration tests (see note above):
python3 -m unittest discover -s "bfabric/tests/integration"
externaljobid-45939_executableid-15312.bash listing:
#!/bin/bash
#
# $HeadURL: http://fgcz-svn.uzh.ch/repos/scripts/trunk/linux/bfabric/apps/python/README.md $
# $Id: README.md 2535 2016-10-24 08:49:17Z cpanse $
# Christian Panse <[email protected]> 2007-2015
# Grid Engine Parameters
#$ -q PRX@fgcz-c-071
#$ -e /home/bfabric/sgeworker/logs/workunitid-134923_resourceid-203236.err
#$ -o /home/bfabric/sgeworker/logs/workunitid-134923_resourceid-203236.out
set -e
set -o pipefail
export EXTERNALJOBID=45938
export RESSOURCEID_OUTPUT=203238
export RESSOURCEID_STDOUT_STDERR="203237 203238"
export OUTPUT="[email protected]:/srv/www/htdocs//p1000/bfabric/Proteomics/gerneric_yaml/2015/2015-09/2015-09-02//workunit_134923//203236.zip"
# job configuration set by B-Fabrics wrapper_creator executable
_OUTPUT=`echo $OUTPUT | cut -d"," -f1`
test $? -eq 0 && _OUTPUTHOST=`echo $_OUTPUT | cut -d":" -f1`
test $? -eq 0 && _OUTPUTPATH=`echo $_OUTPUT | cut -d":" -f2`
test $? -eq 0 && _OUTPUTPATH=`dirname $_OUTPUTPATH`
test $? -eq 0 && ssh $_OUTPUTHOST "mkdir -p $_OUTPUTPATH"
if [ $? -eq 1 ];
then
echo "writting to output url failed!";
exit 1;
fi
cat > /tmp/yaml_config.$$ <<EOF
application:
input:
mascot_dat:
- [email protected]//usr/local/mascot/:/data/20150807/F221967.dat
- [email protected]//usr/local/mascot/:/data/20150807/F221973.dat
output:
- [email protected]:/srv/www/htdocs//p1000/bfabric/Proteomics/gerneric_yaml/2015/2015-09/2015-09-02//workunit_134923//203236.zip
parameters:
gelcms: 'true'
mudpit: 'false'
qmodel: None
xtandem: 'false'
protocol: scp
job_configuration:
executable: /usr/local/fgcz/proteomics/bin/fgcz_scaffold.bash
external_job_id: 45938
input:
mascot_dat:
- 201919
- 201918
output:
protocol: scp
resource_id: 203238
ssh_args: -o StrictHostKeyChecking=no -c arcfour -2 -l bfabric -x
stderr:
protocol: file
resource_id: 203237
url: /home/bfabric/sgeworker/logs/workunitid-134923_resourceid-203236.err
stdout:
protocol: file
resource_id: 203238
url: /home/bfabric/sgeworker/logs/workunitid-134923_resourceid-203236.out
workunit_id: 134923
EOF
# debug / host statistics
hostname
uptime
echo $0
pwd
# run the application
test -f /tmp/yaml_config.$$ && /usr/local/fgcz/proteomics/bin/fgcz_scaffold.bash /tmp/yaml_config.$$
if [ $? -eq 0 ];
then
/home/bfabric/.python/fgcz_bfabric_setResourceStatus_available.py $RESSOURCEID_OUTPUT
/home/bfabric/.python/fgcz_bfabric_setExternalJobStatus_done.py $EXTERNALJOBID
else
echo "application failed"
/home/bfabric/.python/fgcz_bfabric_setResourceStatus_available.py $RESSOURCEID_STDOUT_STDERR $RESSOURCEID;
exit 1;
fi
# Should be available also as zero byte files
/home/bfabric/.python/fgcz_bfabric_setResourceStatus_available.py $RESSOURCEID_STDOUT_STDERR
exit 0#!/bin/bash
query(){
url=$1 \
&& curl \
${url} \
-v \
--header "Content-Type: text/xml;charset=UTF-8" \
--header "SOAPAction: read" \
-d '
<soapenv:Envelope xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/" xmlns:end="http://endpoint.webservice.component.bfabric.org/">
<soapenv:Header/>
<soapenv:Body>
<end:read>
<parameters>
<login>XXX</login>
<password>XXX</password>
<query>
<id>482</id>
</query>
</parameters>
</end:read>
</soapenv:Body>
</soapenv:Envelope>'
}
for url in https://fgcz-bfabric.uzh.ch/bfabric/user?wsdl https://fgcz-bfabric-test.uzh.ch/bfabric/user?wsdl;
do
echo
echo "==== ${url} === "
query ${url}
done
echo $?
remove accidentally inserted mgf files
bfabric_read.py importresource \
| grep mgf$ \
| awk '{print $1}' \
| tee /tmp/$$.log \
| while read i;
do
bfabric_delete.py importresource $i ;
done
# by CT,CP
# not implemented yet 2022-10-19 ,
rv = B.save_object(endpoint = 'mail',
obj={'subject': "TEST",
'recipientemail': '[email protected]',
'message': "TEST; ignore that email",
'parentId': 482,
'parentClassName': 'user'})
# shown as mail for user id 482
- bfabric documentation
- FAQ
- wsdl4BFabric wiki page
- WSDL Interface to B-Fabric endpoints
on macOSX
cd /Applications/Python 3.12 && ./Install\ Certificates.command
X.Y.Z
X is used for major changes, that contain breaking changes
Y should be the current bfabric release
Z is increased for feature releases, that should not break the API
Panse, Christian, Trachsel, Christian and Türker, Can. "Bridging data management platforms and visualization tools to enable ad-hoc and smart analytics in life sciences" Journal of Integrative Bioinformatics, 2022, pp. 20220031. doi: 10.1515/jib-2022-0031.