def patchLibs(libdir):
"""Make sure that libraries can by dynamically loaded. This is a brute force approach"""
saveDir = os.getcwd()
# fix libVitaCFilters.dylib
dir = os.path.join(
libdir,
"python%s/site-packages/vitamind/analyticsLib/pipelineElements" %
pythonVersion)
os.chdir(dir)
# Fix up library references in libVitaCFilters.dylib
lib = "libVitaCFilters.dylib"
utils.runCommand([
"install_name_tool", "-change",
"/tmp/external.buildaccount/lib/libcv.1.dylib", "./libcv.1.dylib", lib
])
utils.runCommand([
"install_name_tool", "-change",
"/tmp/external.buildaccount/lib/libcxcore.1.dylib",
"./libcxcore.1.dylib", lib
])
for lib in ["libcv.1.dylib", "libcxcore.1.dylib"]:
try:
os.remove(lib)
except:
pass
os.symlink("../../../opencv/%s" % lib, lib)
os.chdir(saveDir)
def zipFasta(self):
"""
Compress the fasta file
"""
utils.log("zipping {} ...".format(self.fastaFileName))
cmd = "bgzip -f {}".format(self.fastaFileName)
utils.runCommand(cmd)
def patchLibs(libdir):
"""Make sure that libraries can by dynamically loaded. This is a brute force approach"""
saveDir = os.getcwd()
# fix libVitaCFilters.dylib
dir = os.path.join(
libdir,
"python%s/site-packages/vitamind/analyticsLib/pipelineElements" %
pythonVersion)
os.chdir(dir)
# Fix up library references in libVitaCFilters.dylib
lib = "libVitaCFilters.dylib"
utils.runCommand(["install_name_tool", "-change", "/tmp/external.buildaccount/lib/libcv.1.dylib", "./libcv.1.dylib", lib])
utils.runCommand(["install_name_tool", "-change", "/tmp/external.buildaccount/lib/libcxcore.1.dylib", "./libcxcore.1.dylib", lib])
for lib in ["libcv.1.dylib", "libcxcore.1.dylib"]:
try:
os.remove(lib)
except:
pass
os.symlink("../../../opencv/%s" % lib, lib)
os.chdir(saveDir)
def _downloadFasta(self, chromosome):
accession = self.accessions[chromosome]
fileName = '{}.fa'.format(chromosome)
minPos = 0
if self.excludeReferenceMin:
minPos = self.chromMinMax.getMinPos(chromosome)
maxPos = self.chromMinMax.getMaxPos(chromosome)
with open(fileName, "w") as outFasta:
print(">{}".format(chromosome), file=outFasta)
sequence = _fetchSequence(accession, minPos, maxPos)
for line in sequence:
print(line, file=outFasta)
utils.log("Compressing {}".format(fileName))
utils.runCommand("bgzip -f {}".format(fileName))
compressedFileName = fileName + '.gz'
utils.log("Indexing {}".format(compressedFileName))
utils.runCommand("samtools faidx {}".format(compressedFileName))
# Assemble the metadata.
metadata = {
"md5checksum": getReferenceChecksum(compressedFileName),
"sourceUri": None,
"ncbiTaxonId": 9606,
"isDerived": False,
"sourceDivergence": None,
"sourceAccessions": [accession + ".subset"],
}
metadataFilename = "{}.json".format(chromosome)
dumpDictToFileAsJson(metadata, metadataFilename)
def _downloadFasta(self, chromosome):
accession = self.accessions[chromosome]
fileName = '{}.fa'.format(chromosome)
minPos = 0
if self.excludeReferenceMin:
minPos = self.chromMinMax.getMinPos(chromosome)
maxPos = self.chromMinMax.getMaxPos(chromosome)
with open(fileName, "w") as outFasta:
print(">{}".format(chromosome), file=outFasta)
sequence = _fetchSequence(accession, minPos, maxPos)
for line in sequence:
print(line, file=outFasta)
utils.log("Compressing {}".format(fileName))
utils.runCommand("bgzip -f {}".format(fileName))
compressedFileName = fileName + '.gz'
utils.log("Indexing {}".format(compressedFileName))
utils.runCommand("samtools faidx {}".format(compressedFileName))
# Assemble the metadata.
metadata = {
"md5checksum": getReferenceChecksum(compressedFileName),
"sourceUri": None,
"ncbiTaxonId": 9606,
"isDerived": False,
"sourceDivergence": None,
"sourceAccessions": [accession + ".subset"],
}
metadataFilename = "{}.json".format(chromosome)
dumpDictToFileAsJson(metadata, metadataFilename)
def postProcess(self, videos, videoFile):
if len(videos) is 0:
return
# Merge all videos
configPath = os.path.join(OutputPath.DATA_OUTPUT_PATH, 'video.txt')
with open(configPath, 'w') as fp:
for video in videos:
fp.write('file \'{}\'\n'.format(video))
fp.write('file \'{}\'\n'.format(self.separatorPath))
fp.write('file \'{}\'\n'.format(self.separatorPath))
videoPath = os.path.join(OutputPath.DATA_OUTPUT_PATH, 'all.mp4')
print('Merge all to', videoPath, 'from', configPath)
cmd = 'ffmpeg -y -f concat -safe 0 -i {} -c copy {}'.format(configPath, videoPath)
runCommand(cmd)
# Add logo
self.videoPath = videoFile
if self.logo:
print('Add logo to', self.videoPath)
cmd = 'ffmpeg -y -i {} -i {} -max_muxing_queue_size 10240 -filter_complex "overlay=10:10" {}'.format(videoPath,
self.logo, self.videoPath)
else:
print('Rename', videoPath, 'to', self.videoPath)
cmd = 'mv {} {}'.format(videoPath, self.videoPath)
runCommand(cmd)
def align2sam(command, reference, fastq_file, sai_fastq_file, fastq_metadata, output_dir):
"""
Convert alignments to SAM format. Turn bwa sai alignments into a sam file.
It uses bwa samse commandline.
"""
(path, name, ext) = splitPath(sai_fastq_file)
if ext != '.sai':
sys.exit('align2Sam: alignment file %s does not have .sai extension' % sai_fastq_file)
sam_file = os.path.join(output_dir, os.path.splitext(os.path.basename(fastq_file))[0]) + '.sam'
sample = fastq_metadata[os.path.basename(fastq_file)]['sample']
run_id = fastq_metadata[os.path.basename(fastq_file)]['run_id']
lane = fastq_metadata[os.path.basename(fastq_file)]['lane']
identifier = fastq_metadata[os.path.basename(fastq_file)]['identifier']
readgroup_metadata = {'PL': 'ILLUMINA', 'SM': sample,
'LB': '%s_%s_%s_Lane%s' % (identifier, sample, run_id, lane),
'ID': '%s_%s_%s_Lane%s' % (identifier, sample, run_id, lane) }
metadata_str = make_metadata_string(readgroup_metadata)
command = command % {'out': sam_file, 'ref': reference, 'align': sai_fastq_file,
'seq': fastq_file, 'meta': metadata_str}
runCommand('bwa samse alignment from fastq: %s' % sample, command)
return sam_file
def createSlider(self):
self.imagePath = os.path.join(self.path, 'image.mp4')
if self.imageCount is 0:
print('Create slider to', self.imagePath, 'from', self.background)
# TODO: Use background as image
cmd = 'ffmpeg -y -loop 1 -i {} -c:v libx264 -t {:.2f} -pix_fmt yuv420p {}'.format(self.background,
self.length, self.imagePath)
runCommand(cmd)
return
duration = self.length / self.imageCount
videoMaker = None
for index in range(self.imageCount):
imagePath = os.path.join(self.path, '{}.jpg'.format(index))
if not os.path.exists(imagePath):
continue
imageVideoPath = '{}.mp4'.format(imagePath)
VideoKit.createLoopVideo(imageVideoPath, imagePath, duration)
videoMaker = VideoKit.appendVideo(imageVideoPath, videoMaker)
videoMaker.merge(self.imagePath)
def fastqc(command, sequences, fastq_metadata, output_dir):
'''
Run FastQC on each fastq file.
'''
for fastq_file in sequences:
command = command % {'outdir': output_dir, 'seq': fastq_file}
runCommand('Checking fastq quality', command)
def align_with_mem(command, threads, reference, fastq_file, pair_file, fastq_metadata, output_dir):
'''
Perform alignment on two paired-end fastq files to a reference genome to produce a sam file.
'''
(path, name, ext) = splitPath(fastq_file)
(pathP, nameP, extP) = splitPath(pair_file)
if ext != '.fastq' or extP != '.fastq':
sys.exit('align: one of the fastq file %s or %s does not have .fastq extension' % (fastq_file, pair_file))
sam_file = os.path.join(output_dir, os.path.splitext(os.path.basename(fastq_file))[0]) + '.sam'
sample = fastq_metadata[os.path.basename(fastq_file)]['sample']
run_id = fastq_metadata[os.path.basename(fastq_file)]['run_id']
lane = fastq_metadata[os.path.basename(fastq_file)]['lane']
identifier = fastq_metadata[os.path.basename(fastq_file)]['identifier']
readgroup_metadata = {'PL': 'ILLUMINA', 'SM': sample,
'LB': '%s_%s_%s_Lane%s' % (identifier, sample, run_id, lane),
'ID': '%s_%s_%s_Lane%s' % (identifier, sample, run_id, lane) }
metadata_str = make_metadata_string(readgroup_metadata)
command = command % {'threads': threads, 'meta': metadata_str, 'ref': reference,
'seq': fastq_file , 'pair': pair_file, 'out': sam_file}
runCommand('bwa mem alignment from fastq: %s' % sample, command)
return sam_file
def zipFasta(self):
"""
Compress the fasta file
"""
utils.log("zipping {} ...".format(self.fastaFileName))
cmd = "bgzip -f {}".format(self.fastaFileName)
utils.runCommand(cmd)
def create_executable(trunk_dir, script_dir, work_dir, target):
# Go to target dir and verify there is no executable yet
os.chdir(work_dir)
if os.path.exists('VisionToolkit.exe'):
os.remove('VisionToolkit.exe')
nsis = os.path.join(trunk_dir,
'external/win32/lib/buildtools/NSIS/makensis.exe')
# Copy the NSIS script to work_dir because that's where it will execute
shutil.copy(os.path.join(script_dir, 'vision_toolkit.nsi'),
'vision_toolkit.nsi')
assert os.path.isfile(nsis)
# Build the NSIS command line
cmd = [nsis, 'vision_toolkit.nsi']
#print ' '.join(cmd)
# Launch NSIS and verify that the final executable has been created
#subprocess.call(cmd)
import utils
import logging
# log level was earlier set to info. We want all output from this command
logging.getLogger('').setLevel(logging.DEBUG)
utils.runCommand(cmd)
assert os.path.isfile('VisionToolkit.exe')
# Rename to target name
try:
shutil.move('VisionToolkit.exe', target)
except Exception, e:
print e
raise
def build_win32(srcdir, builddir, installdir, assertions, customerRelease):
log.info("build_win32: srcdir = '%s'", srcdir)
log.info("build_win32: builddir = '%s'", builddir)
log.info("build_win32: installdir = '%s'", installdir)
# deprecated
os.environ["NTA"] = installdir
# These are what I would like to use. Currently only used by the test project
os.environ["NTAX_INSTALL_DIR"] = installdir
os.environ["NTAX_BUILD_DIR"] = builddir
log.debug("build_win32: srcdir: '%s'", srcdir)
log.debug("build_win32: installdir: '%s'", installdir)
log.debug("build_win32: builddir: '%s'", builddir)
# how to quote "Release|Win32" so that it doesn't cause an error?
# command = ["vcbuild", "/logcommands", "/showenv", "/time", os.path.join(srcdir, "trunk.sln")]
utils.changeDir(srcdir)
if customerRelease:
import glob
solutionFiles = glob.glob("*.sln")
if len(solutionFiles) == 0:
raise Exception("Unable to find any solution files in customer source release")
elif len(solutionFiles) > 1:
raise Exception("More than one solution file found in customer source release: %s" % solutionFiles)
command = 'vcbuild /logcommands /showenv /time %s "Release|Win32"' % solutionFiles[0]
else:
command = 'vcbuild /logcommands /showenv /time trunk.sln "Release|Win32"'
utils.runCommand(command)
postbuild_win32(srcdir, installdir)
def extract(self, runLogDir):
if os.path.exists(self.extract_dir):
shutil.rmtree(self.extract_dir)
os.makedirs(self.extract_dir)
utils.runCommand(
'tar xvf %s -C %s --strip-components=1 > %s/extract%s.log.txt 2>&1'
% (self.tgz, self.extract_dir, runLogDir, self.name))
def run(configfile, name):
OutputPath.init(configFile)
thread = ThreadWritableObject(configFile, name)
thread.start()
sys.stdout = thread
sys.errout = thread # XXX: Actually, it does NOT work
try:
db = Database(configFile, 'specials')
db.initialize()
evaluation = Evaluation(configFile, db)
evaluation.updateOverdue()
path = OutputPath.getSharePath()
sharePath = getProperty(configFile, 'output-share-file')
cmd = '/bin/rm -f {1} && /bin/ln -s {0} {1}'.format(path, sharePath)
runCommand(cmd)
data = evaluation.output()
with open(path, 'w') as fp:
fp.write(reprDict(data))
except KeyboardInterrupt:
pass
except Exception, e:
print 'Error occurs at', datetime.now().strftime('%Y-%m-%d %H:%M:%S')
traceback.print_exc(file=sys.stdout)
def createMovie(encoder='ffmpeg'):
"""Create a movie from a saved sequence of images.
encoder is one of: 'ffmpeg, mencoder, convert'
"""
if not multisave:
pf.warning('You need to start multisave mode first!')
return
names,format,quality,window,border,hotkey,autosave,rootcrop = multisave
glob = names.glob()
## if glob.split('.')[-1] != 'jpg':
## pf.warning("Currently you need to save in 'jpg' format to create movies")
## return
if encoder == 'convert':
cmd = "convert -delay 1 -colors 256 %s output.gif" % names.glob()
elif encoder == 'mencoder':
cmd = "mencoder -ovc lavc -fps 5 -o output.avi %s" % names.glob()
elif encoder == 'mencoder1':
cmd = "mencoder \"mf://%s\" -mf fps=10 -o output1.avi -ovc lavc -lavcopts vcodec=msmpeg4v2:vbitrate=800" % names.glob()
else:
cmd = "ffmpeg -qscale 1 -r 1 -i %s output.mp4" % names.glob()
pf.debug(cmd)
utils.runCommand(cmd)
def createVrtForAOI(fvrt, year, area):
verbose = CONFIG.get('verbose', False)
# temp file to write doqqs into
fvrtin = fvrt + '.in'
# temp file for intermeadiate vrt
fvrtvrt = fvrt + '.vrt'
# get a list of doqqs the intersect our area of interest
# and write them to a temp file
files = getDoqqsForArea(year, area)
fh = open(fvrtin, 'wb')
for f in files:
fh.write(f + "\n")
fh.close()
# create the intermeadiate vrt file
cmd = ['gdalbuildvrt', '-input_file_list', fvrtin, fvrtvrt]
runCommand(cmd, verbose)
# create the final vrt file with appropriate mask band defined
cmd = [
'gdal_translate', '-b', '1', '-b', '2', '-b', '3', '-b', '5', '-mask',
'4', '-of', 'VRT', fvrtvrt, fvrt
]
runCommand(cmd, verbose)
def create_executable(trunk_dir, script_dir, work_dir, target):
# Go to target dir and verify there is no executable yet
os.chdir(work_dir)
if os.path.exists('VisionDemo.exe'):
os.remove('VisionDemo.exe')
nsis = os.path.join(trunk_dir,
'external/win32/lib/buildtools/NSIS/makensis.exe')
# Copy the NSIS script to work_dir because that's where it will execute
shutil.copy(os.path.join(script_dir, 'demo.nsi'), 'demo.nsi')
assert os.path.isfile(nsis)
# Build the NSIS command line
cmd = [nsis, 'demo.nsi']
#print ' '.join(cmd)
# Launch NSIS and verify that the final executable has been created
#subprocess.call(cmd)
import utils
import logging
# log level was earlier set to info. We want all output from this command
logging.getLogger('').setLevel(logging.DEBUG)
utils.runCommand(cmd)
assert os.path.isfile('VisionDemo.exe')
# Rename to target name
try:
shutil.move('VisionDemo.exe', target)
except Exception, e:
print e
print
def setLogLevel(self, level):
"""Sets the module log level."""
if level != Defaults.vdoLogLevel:
runCommand(string.split("echo" + level + " > /sys/" + self._name +
"/log_level"),
shell=True,
noThrow=True)
def execute(self, args):
#pylint: disable=R0201
conf = Configuration(self.confFile, readonly=False)
if not args.name:
args.all = True
try:
print(_("VDO status:"))
print(_(" Node: ")
+ runCommand(['uname', '-n'], noThrow=True, strip=True))
print(_(" Date: ")
+ runCommand(['date', '--rfc-3339=seconds'], noThrow=True,
strip=True))
if os.getuid() != 0:
print(_(" Note: not running as root,"
+ " some status may be unavailable"))
print(os.linesep.join(VDOKernelModuleService().status("")))
print(os.linesep.join(conf.status("")))
print(_("VDOs: "))
for vdo in self.getVdoServices(args, conf):
print(os.linesep.join(vdo.status(" ")))
sys.stdout.flush()
sys.stderr.flush()
except IOError as ex:
self.log.debug("exception ignored: {0}".format(ex))
def runPlayer(path):
#runCommand("export DISPLAY=:0;sudo -u dima vlc -f \"%s\" &" % path)
#runCommand("sudo -u dima totem --fullscreen --display=:0 \"%s\"" % path)
#runCommand("export DISPLAY=:0; sudo -u dima mplayer --fs \"%s\"" % path)
#runCommand(r'"c:\Program Files (x86)\vlc-2.2.0\vlc.exe" -f "%s"' % path)
runCommand(
r'"c:\Program Files (x86)\K-Lite Codec Pack\MPC-HC64\mpc-hc64.exe" /fullscreen "%s"' % path)
def cleanNoPysvn(dir, doCleanup=True):
if doCleanup == False:
utils.runCommand("svn status --no-ignore %s" % dir)
else:
# Delete everythiung svn doesn't know about *except* for the top level directory, since svn can
# reports the top level directory as "!" (missing") if a sub-directory is missing.
# Use the xml format to avoid problems with spaces in filenames
utils.runCommand("svn status --no-ignore --xml %s | grep -A1 entry | grep path= | awk -F= '{print $2}' | sed 's/>//' | grep -v \\\"%s\\\" | xargs rm -rvf" % (dir, dir), logOutputOnlyOnFailure=False)
def merge(dstVideoPath, srcVideoPath, srcAudioPath):
print('Merge', srcVideoPath, 'and', srcAudioPath, 'to', dstVideoPath)
cmd = 'ffmpeg -y -i {} -i {} -c copy -map \'0:v:0\' -map \'1:a:0\' {}'.format(
srcVideoPath, srcAudioPath, dstVideoPath)
runCommand(cmd)
def indexFasta(self):
"""
Create index on the fasta file
"""
zipFileName = "{}.gz".format(self.fastaFileName)
utils.log("indexing {} ...".format(zipFileName))
cmd = "samtools faidx {}".format(zipFileName)
utils.runCommand(cmd)
def indexFasta(self):
"""
Create index on the fasta file
"""
zipFileName = "{}.gz".format(self.fastaFileName)
utils.log("indexing {} ...".format(zipFileName))
cmd = "samtools faidx {}".format(zipFileName)
utils.runCommand(cmd)
def compressSplits(splitFileNames):
compressedFileNames = []
for splitFileName in splitFileNames:
utils.log("Compressing {}".format(splitFileName))
cmd = "bgzip {}".format(splitFileName)
utils.runCommand(cmd)
compressedFileName = "{}.gz".format(splitFileName)
compressedFileNames.append(compressedFileName)
return compressedFileNames
def compressSplits(splitFileNames):
compressedFileNames = []
for splitFileName in splitFileNames:
utils.log("Compressing {}".format(splitFileName))
cmd = "bgzip {}".format(splitFileName)
utils.runCommand(cmd)
compressedFileName = "{}.gz".format(splitFileName)
compressedFileNames.append(compressedFileName)
return compressedFileNames
def decompressFasta(args):
fastaFileName = args.fastaFile
filename, extension = os.path.splitext(fastaFileName)
if extension == '.gz':
utils.log("Decompressing {}".format(fastaFileName))
cmd = "gunzip {}".format(fastaFileName)
utils.runCommand(cmd)
fastaFileName = filename
return fastaFileName
def decompressFasta(args):
fastaFileName = args.fastaFile
filename, extension = os.path.splitext(fastaFileName)
if extension == '.gz':
utils.log("Decompressing {}".format(fastaFileName))
cmd = "gunzip {}".format(fastaFileName)
utils.runCommand(cmd)
fastaFileName = filename
return fastaFileName
def ZipDir(zipFile, directory):
if HOST_OS == 'win':
cmd = os.path.normpath(os.path.join(
os.path.dirname(__file__),
'../../../third_party/lzma_sdk/Executable/7za.exe'))
options = ['a', '-r', '-tzip']
else:
cmd = 'zip'
options = ['-yr']
utils.runCommand([cmd] + options + [zipFile, directory])
def applyRules(rule):
print 'applying rule %s' % rule
if runCommand(rule):
print 'command <%s> execute failed' % rule
return False
rule = 'iptables-save > /etc/sysconfig/iptables'
if runCommand(rule):
print 'command <%s> execute failed' % rule
return False
return True
def run_protocol(self):
#print("data",self.alignerList[0])
#if not os.path.exists("indexes"):
# os.mkdir("indexes")
#os.chdir("indexes")
#for genome in self.inputfile:
for aligner in self.alignersList:
#if os.path.exists(dir):
# continue
#os.mkdir(dir)
#os.chdir(dir)
if aligner.upper() == "HISAT":
objhisat = hisat(inputfile=self.inputfile,
workingpath="aligners/HISAT")
if (self.parameters[0]['build_index']):
runCommand(
objhisat.build_command
) #go to the specific directory and run the build command
print("Building index file completed.")
deleteFile(os.path.join(
"/workingpath/",
"abc")) #delete the inputfile to save the space
#print("if 1",aligner.upper())
#print(cmd)
elif aligner.upper() == "HISAT2":
objhisat2 = hisat2(inputfile=self.inputfile,
workingpath="aligners/HISAT2")
if (self.parameters[1]['build_index']):
runCommand(objhisat2.build_command)
print("Building index file completed.")
deleteFile(os.path.join(
"/workingpath/",
"abc")) #delete the inputfile to save the space
elif aligner == "BOWTIE":
cmd = "bowtie-build ../data/%s %s" % (self.inputfile,
self.inputfile)
elif aligner == "STAR":
cmd = "../../aligners/bin/STAR --runMode genomeGenerate --genomeDir . --genomeFastaFiles ../../data/%s.fa" % (
self.inputfile)
elif aligner == "GSNAP":
cmd = "../../aligners/bin/gmap_build -B ../../aligners/bin -D . -d %s ../../data/%s.fa" % (
self.inputfile, self.inputfile)
else:
print("Aligner:", aligner, " does not exist.")
assert False
def install_vitamind(target_dir):
"""Run the install_vitamind.py script
Installs ffmpeg, scipy and vitamind obfuscated pipeline
to the target dir
"""
print 'install_vitamind() into', target_dir
save_dir = os.getcwd()
os.chdir(os.path.join(trunk_dir, 'external/src/python_modules/vitamind'))
utils.runCommand(['python', 'install_vitamind.py', target_dir])
os.chdir(save_dir)
def installVitamind(trunkDir, installDir):
"""Run the installVitamind.py script
Installs ffmpeg, scipy and vitamind obfuscated pipeline
to the install dir
"""
print 'installVitamind()'
saveDir = os.getcwd()
os.chdir(os.path.join(trunkDir, 'external/src/python_modules/vitamind'))
utils.runCommand(['python', 'install_vitamind.py', "--force", installDir])
os.chdir(saveDir)
def runTests(self):
testCommands = self.parseTestCommands()
for command in testCommands:
expandedCommand = self.expandCommand(command)
self.log('Running: "{0}"'.format(expandedCommand))
try:
utils.runCommand(expandedCommand)
except subprocess.CalledProcessError:
self.log('ERROR')
return
self.log('SUCCESS')
def install_vitamind(target_dir): """Run the install_vitamind.py script Installs ffmpeg, scipy and vitamind obfuscated pipeline to the target dir """ print 'install_vitamind() into', target_dir save_dir = os.getcwd() os.chdir(os.path.join(trunk_dir, 'external/src/python_modules/vitamind')) utils.runCommand(['python', 'install_vitamind.py', target_dir]) os.chdir(save_dir)
def installVitamind(trunkDir, installDir): """Run the installVitamind.py script Installs ffmpeg, scipy and vitamind obfuscated pipeline to the install dir """ print 'installVitamind()' saveDir = os.getcwd() os.chdir(os.path.join(trunkDir, 'external/src/python_modules/vitamind')) utils.runCommand(['python', 'install_vitamind.py', "--force", installDir]) os.chdir(saveDir)
def createLoopVideo(dstVideoPath, srcImagePath, videoLength):
print('Create video to', dstVideoPath, 'from', srcImagePath,
'with length', videoLength)
cmd = 'ffmpeg -y -loop 1 -i {} -c:v libx264 -t {:.2f} -pix_fmt yuv420p {}'.format(
srcImagePath, videoLength, dstVideoPath)
runCommand(cmd)
return dstVideoPath
def running(self, wait=True):
"""Returns True if the module is loaded and DM target is available."""
retries = 20 if wait else 1
try:
runCommand(string.split("lsmod | grep -q '" + self._name + "'"),
shell=True, retries=retries)
runCommand(string.split("dmsetup targets | grep -q "
+ Defaults.vdoTargetName),
shell=True, retries=retries)
return True
except CommandError:
return False
def _fsyncDirectory(self):
"""Open and issue an fsync on the directory containing the config file.
"""
dirname = os.path.dirname(self.filepath)
if Command.noRunMode():
runCommand(['fsync', dirname])
return
fd = os.open(dirname, os.O_RDONLY)
try:
os.fsync(fd)
finally:
os.close(fd)
def read_gambit_neutral(fn):
"""Read a triangular surface mesh in Gambit neutral format.
The .neu file nodes are numbered from 1!
Returns a nodes,elems tuple.
"""
runCommand("%s/external/gambit-neu '%s'" % (GD.cfg['pyformexdir'],fn))
nodesf = changeExt(fn,'.nodes')
elemsf = changeExt(fn,'.elems')
nodes = fromfile(nodesf,sep=' ',dtype=Float).reshape((-1,3))
elems = fromfile(elemsf,sep=' ',dtype=int32).reshape((-1,3))
return nodes, elems-1
def annotate(command, annovar_dir, annovar_file, output_dir):
'''
Annotate vcf using Annovar variant caller.
'''
(path, name, ext) = splitPath(annovar_file)
if ext != '.avinput':
sys.exit('Annotating vcf: vcf file %s does not have .avinput extension' % annovar_file)
out_prefix = os.path.join(output_dir, name)
command = command % {'out': out_prefix, 'annovarfile': annovar_file, 'annovardir': annovar_dir}
runCommand('Annotating with Annovar', command)
return out_prefix
def convert2annovar(command, annovar_dir, vcf, output_dir):
'''
Convert vcf file to Annovar variant caller .annovar
'''
(path, name, ext) = splitPath(vcf)
if ext != '.vcf':
sys.exit('Converting to .annovar: vcf file %s does not have .vcf extension' % vcf)
out_prefix = os.path.join(output_dir, name.split('.')[0])
command = command % {'out': out_prefix, 'vcf': vcf, 'annovardir': annovar_dir}
runCommand('Coverting to .annovar format', command)
return '.'.join([out_prefix, name.split('.')[0], 'avinput'])
def make_reference_database(command, algorithm, reference):
'''
Create reference database
'''
ref_database = reference + '.bwt'
if os.path.exists(ref_database):
print('Reference database already exists: using %s' % ref_database)
else:
command = command % {'algorithm': algorithm, 'prefix': reference, 'seq': reference + '.fasta'}
runCommand('Creating Reference Database', command)
return ref_database
def align(command, threads, reference, sequence, fastq_metadata, output_dir):
'''
Align sequence reads to the reference genome. This is the bwa's first stage, bwa aln.
'''
(path, name, ext) = splitPath(sequence)
if ext != '.fastq':
sys.exit('align: sequence file %s does not have .fastq extension' % sequence)
alignment_file = os.path.join(output_dir, name + '.sai')
command = command % {'out': alignment_file, 'threads': int(threads), 'ref': reference,
'seq': sequence, 'encodingflag': ''}
runCommand('Running Alignment', command)
return alignment_file
def read_gambit_neutral(fn):
"""Read a triangular surface mesh in Gambit neutral format.
The .neu file nodes are numbered from 1!
Returns a nodes,elems tuple.
"""
scr = os.path.join(pf.cfg['bindir'], 'gambit-neu ')
utils.runCommand("%s '%s'" % (scr, fn))
nodesf = utils.changeExt(fn, '.nodes')
elemsf = utils.changeExt(fn, '.elems')
nodes = fromfile(nodesf, sep=' ', dtype=Float).reshape((-1, 3))
elems = fromfile(elemsf, sep=' ', dtype=int32).reshape((-1, 3))
return nodes, elems - 1
def summarize(command, annovar_dir, annovar_file, ver1000g, veresp, verdbsnp, genetype, buildver, output_dir):
'''
Summarize information with Annovar.
'''
(path, name, ext) = splitPath(annovar_file)
if ext != '.avinput':
sys.exit('Summarizing annotations: vcf file %s does not have .avinput extension' % annovar_file)
out_prefix = os.path.join(output_dir, name)
command = command % {'out': out_prefix, 'annovarfile': annovar_file, 'annovardir': annovar_dir,
'ver1000g': ver1000g, 'veresp': veresp, 'verdbsnp': verdbsnp, 'genetype': genetype, 'buildver': buildver}
runCommand('Summarizing with Annovar', command)
return out_prefix
def read_gambit_neutral(fn):
"""Read a triangular surface mesh in Gambit neutral format.
The .neu file nodes are numbered from 1!
Returns a nodes,elems tuple.
"""
scr = os.path.join(pf.cfg['bindir'],'gambit-neu ')
utils.runCommand("%s '%s'" % (scr,fn))
nodesf = utils.changeExt(fn,'.nodes')
elemsf = utils.changeExt(fn,'.elems')
nodes = fromfile(nodesf,sep=' ',dtype=Float).reshape((-1,3))
elems = fromfile(elemsf,sep=' ',dtype=int32).reshape((-1,3))
return nodes, elems-1
def stl_to_off(stlname,offname=None,sanitize=True):
"""Transform an .stl file to .off format."""
if not offname:
offname = changeExt(stlname,'.off')
if sanitize:
options = ''
else:
# admesh always wants to perform some actions on the STL. The -c flag
# to suppress all actions makes admesh hang. Therefore we include the
# action -d (fix normal directions) as the default.
options = '-d'
runCommand("admesh %s --write-off '%s' '%s'" % (options,offname,stlname))
return offname
def read_ascii_large(fn,dtype=Float):
"""Read an ascii .stl file into an [n,3,3] float array.
This is an alternative for read_ascii, which is a lot faster on large
STL models.
It requires the 'awk' command though, so is probably only useful on
Linux/UNIX. It works by first transforming the input file to a
.nodes file and then reading it through numpy's fromfile() function.
"""
tmp = '%s.nodes' % fn
runCommand("awk '/^[ ]*vertex[ ]+/{print $2,$3,$4}' '%s' | d2u > '%s'" % (fn,tmp))
nodes = fromfile(tmp,sep=' ',dtype=dtype).reshape((-1,3,3))
return nodes
def execute(self, args):
#pylint: disable=R0201
conf = Configuration(self.confFile, readonly=False)
if not args.name:
args.all = True
try:
# To be consistent with previous output we must present each section as
# its own rather than organizing them into one structure to dump.
# Also, we gather all the info before printing it out to avoid
# interspersing command info when run in verbose mode.
values = {}
vdoStatus = {_("VDO status"): values}
values[_("Node")] = runCommand(['uname', '-n'],
noThrow=True,
strip=True)
values[_("Date")] = runCommand(['date', '--rfc-3339=seconds'],
noThrow=True,
strip=True)
if os.getuid() != 0:
values[_("Note")] = _("Not running as root," +
" some status may be unavailable")
kernelStatus = {
_("Kernel module"): VDOKernelModuleService().status()
}
confStatus = {_("Configuration"): conf.status()}
vdos = {}
perVdoStatus = {_("VDOs"): vdos}
for vdo in self.getVdoServices(args, conf):
import time
start_time = time.time()
try:
vdos[vdo.getName()] = vdo.status()
except VDOServiceError as ex:
vdos[vdo.getName()] = str(ex)
print "time use: {}'s".format(time.time() - start_time)
# YAML adds a newline at the end. To maintain consistency with the
# previous output we need to eliminate that.
# print(yaml.dump(vdoStatus, default_flow_style = False)[:-1])
# print(yaml.dump(kernelStatus, default_flow_style = False)[:-1])
# print(yaml.dump(confStatus, default_flow_style = False)[:-1])
# print(yaml.dump(perVdoStatus, default_flow_style = False)[:-1])
sys.stdout.flush()
sys.stderr.flush()
except IOError as ex:
self.log.debug("exception ignored: {0}".format(ex))
def filter_snps(command, command_options, gatk_dir, reference, vcf, filter_expression, output_dir):
'''
Use GATK VariantFiltration to filter raw SNP calls.
'''
(path, name, ext) = splitPath(vcf)
command_options = command_options[1]
if ext != '.vcf':
sys.exit('filtering SNPs: vcf file %s does not have .vcf extension' % vcf)
out_vcf = os.path.join(output_dir, name + '.filtered.vcf')
command = command % {'jvmoptions': command_options, 'out': out_vcf,
'vcf': vcf, 'gatkdir': gatk_dir, 'ref': reference + '.fasta', 'expression': filter_expression}
runCommand('Calling snps', command)
return out_vcf
def base_qual_recal_tabulate(command, command_options, gatk_dir, reference, recal_file, alignment, output_dir):
'''
GATK TableRecalibration: recalibrate base quality scores using the output of CountCovariates.
'''
(path, name, ext) = splitPath(alignment)
command_options = command_options[1]
if ext != '.bam':
sys.exit('table recalibration: alignment file %s does not have .bam extension' % alignment)
recal_bam = os.path.join(output_dir, name + '.recal.bam')
command = command % {'jvmoptions': command_options, 'out': recal_bam, 'recalfile': recal_file,
'bam': alignment, 'gatkdir': gatk_dir, 'ref': reference + '.fasta'}
runCommand('recalibrate base quality scores', command)
return recal_bam
def read_gambit_neutral_hex(fn):
"""Read an hexahedral mesh in Gambit neutral format.
The .neu file nodes are numbered from 1!
Returns a nodes,elems tuple.
"""
scr = os.path.join(pf.cfg['bindir'], 'gambit-neu-hex ')
pf.message("%s '%s'" % (scr, fn))
utils.runCommand("%s '%s'" % (scr, fn))
nodesf = utils.changeExt(fn, '.nodes')
elemsf = utils.changeExt(fn, '.elems')
nodes = fromfile(nodesf, sep=' ', dtype=Float).reshape((-1, 3))
elems = fromfile(fn_e, sep=' ', dtype=int32).reshape((-1, 8))
elems = elems[:, (0, 1, 3, 2, 4, 5, 7, 6)]
return nodes, elems - 1
def voiceOut(self, stringContent):
result = self.aipSpeech.synthesis(stringContent, 'zh', 1, {
'vol': 15,
'per': 0
})
if not isinstance(result, dict):
path = OutputPath.getDataPath('voice', 'wav')
with open(path, 'wb') as fp:
fp.write(result)
cmd = 'mplayer {}'.format(path)
runCommand(cmd)
def remesh(self, edgelen=None):
"""Remesh a TriSurface.
edgelen is the suggested edge length
"""
if edgelen is None:
self.getElemEdges()
E = Mesh(self.coords, self.edges, eltype='line2')
edgelen = E.lengths().mean()
tmp = utils.tempFile(suffix='.stl').name
tmp1 = utils.tempFile(suffix='.stl').name
pf.message("Writing temp file %s" % tmp)
self.write(tmp, 'stl')
pf.message("Remeshing using VMTK (edge length = %s)" % edgelen)
cmd = "vmtk vmtksurfaceremeshing -ifile %s -ofile %s -edgelength %s" % (
tmp, tmp1, edgelen)
sta, out = utils.runCommand(cmd)
os.remove(tmp)
if sta:
pf.message("An error occurred during the remeshing.")
pf.message(out)
return None
S = TriSurface.read(tmp1)
os.remove(tmp1)
return S