obitools/obitools3
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 66 Wiki Activity

Cython: updated the URI decoding to handle outputs other than DMS

Browse Source
This commit is contained in:
Celine Mercier
2018-10-17 11:21:29 +02:00
parent 58589e04be
commit 135d3b6e67
1 changed files with 143 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

205
python/obitools3/uri/decode.pyx
View File

@ -12,6 +12,11 @@ from obitools3.parsers.ngsfilter import ngsfilterIterator
from obitools3.parsers.embl import emblIterator from obitools3.parsers.embl import emblIterator
from obitools3.parsers.universal import entryIteratorFactory from obitools3.parsers.universal import entryIteratorFactory
from obitools3.writers.fasta import FastaNucWriter
from obitools3.writers.fastq import FastqWriter
from obitools3.format.fasta import FastaFormat
from obitools3.format.fastq import FastqFormat
from obitools3.dms.obiseq import Nuc_Seq from obitools3.dms.obiseq import Nuc_Seq
from obitools3.apps.config import getConfiguration,logger from obitools3.apps.config import getConfiguration,logger
from obitools3.apps.temp import get_temp_dms from obitools3.apps.temp import get_temp_dms
@ -56,7 +61,9 @@ cdef open_dms(bytes path, bint create=False):
pos=pos+1 pos=pos+1
return None return None
def open_dms_element(DMS dms, bytes path,
def open_dms_element(DMS dms,
bytes path,
bint create=False, bint create=False,
type newviewtype=View): type newviewtype=View):
""" """
@ -139,12 +146,21 @@ def open_dms_element(DMS dms, bytes path,
if len(path_parts) > 4: if len(path_parts) > 4:
raise MalformedURIException('Malformed View URI') raise MalformedURIException('Malformed View URI')
return (dms,subsubpart) return (dms, subsubpart)
'''
#TODO discuss returned object. Return a dict? or some class instance?
Reads an URI and returns a tuple containing:
(1) The opened file or DMS, or the URI itself if nothing could be opened by the function
(2) The opened view or iterator on the opened file or writer
(3) The class of object returned or handled by (2)
(4) The original URI in bytes
'''
def open_uri(uri, def open_uri(uri,
bint input=True, bint input=True,
type newviewtype=View): type newviewtype=View):
cdef bytes urib = tobytes(uri) cdef bytes urib = tobytes(uri)
cdef bytes scheme cdef bytes scheme
cdef tuple dms cdef tuple dms
@ -171,17 +187,21 @@ def open_uri(uri,
error = None error = None
if scheme==b"" or scheme==b"dms" : if scheme==b"dms" or \
dms = open_dms(urip.path,create) (scheme==b"" and \
(((not input) and "outputformat" not in config["obi"]) or \
(input and "inputformat" not in config["obi"]))): # TODO maybe not best way
dms = open_dms(urip.path, create)
if dms is None and default_dms is not None: if dms is None and default_dms is not None:
dms=(default_dms, urip.path) dms=(default_dms, urip.path)
if dms is not None: if dms is not None:
try: try:
resource=open_dms_element(dms[0],dms[1], resource=open_dms_element(dms[0],
dms[1],
create, create,
newviewtype newviewtype)
)
scheme=b"dms" scheme=b"dms"
urip = ParseResultBytes(scheme=b"dms", urip = ParseResultBytes(scheme=b"dms",
@ -205,27 +225,37 @@ def open_uri(uri,
logger('Error','cannot open DMS: %s', uri) logger('Error','cannot open DMS: %s', uri)
raise FileNotFoundError('uri') raise FileNotFoundError('uri')
#if not urip.scheme: # TODO not sure what it was supposed to do but not working as intended if not urip.scheme:
# urib=b"file:"+urib urib=b"file:"+urib
try: if input:
file = uopen(urib) try:
logger('info','Opened file: %s', tostr(urib)) file = uopen(urip.path, mode='rb')
except Exception as e: # TODO discuss: if can't open file, return the character string itself logger('info','Opened file: %s', urip.path)
file = urib except Exception as e: # TODO discuss: if can't open file, return the character string itself
iseq = urib file = tobytes(uri)
objclass = bytes iseq = urib
objclass = bytes
else: # TODO update uopen to be able to write?
file = open(urip.path, 'wb')
if file is not None: if file is not None:
qualifiers=parse_qs(urip.query) qualifiers=parse_qs(urip.query)
if b'format' in qualifiers: if input and b'format' in qualifiers:
format = qualifiers[b'format'][0] format = qualifiers[b'format'][0]
else: else: # TODO discuss priorities
try: if urip.scheme:
format=config["obi"]["format"] format = urip.scheme
except KeyError: else:
format=None try:
if input:
formatkey = "inputformat"
else:
formatkey = "outputformat"
format=config["obi"][formatkey]
except KeyError:
format=None
if b'seqtype' in qualifiers: if b'seqtype' in qualifiers:
seqtype=qualifiers[b'seqtype'][0] seqtype=qualifiers[b'seqtype'][0]
@ -248,7 +278,6 @@ def open_uri(uri,
if skip < 0: if skip < 0:
raise MalformedURIException('Malformed skip argument in URI') raise MalformedURIException('Malformed skip argument in URI')
if b'only' in qualifiers: if b'only' in qualifiers:
only=int(qualifiers[b"only"][0]) only=int(qualifiers[b"only"][0])
else: else:
@ -259,7 +288,6 @@ def open_uri(uri,
if only is not None and only <= 0: if only is not None and only <= 0:
raise MalformedURIException('Malformed only argument in URI') raise MalformedURIException('Malformed only argument in URI')
if b"skiperror" in qualifiers: if b"skiperror" in qualifiers:
try: try:
skiperror=eval(qualifiers[b"skiperror"][0]) skiperror=eval(qualifiers[b"skiperror"][0])
@ -333,11 +361,26 @@ def open_uri(uri,
except KeyError: except KeyError:
dec=b"." dec=b"."
if b"printna" in qualifiers:
try:
printna=eval(qualifiers[b"printna"][0])
except Exception as e:
raise MalformedURIException("Malformed 'print NA' argument in URI")
else:
try:
printna=config["obi"]["printna"]
except KeyError:
printna=False
if b"nastring" in qualifiers: if b"nastring" in qualifiers:
nastring=tobytes(qualifiers[b"nastring"][0]) nastring=tobytes(qualifiers[b"nastring"][0])
else: else:
try: try:
nastring=tobytes(config["obi"]["nastring"]) if input:
nakey = "inputnastring"
else:
nakey = "outputnastring"
nastring=tobytes(config["obi"][nakey])
except KeyError: except KeyError:
nastring=b'NA' nastring=b'NA'
@ -377,60 +420,88 @@ def open_uri(uri,
if format is not None: if format is not None:
if seqtype==b"nuc": if seqtype==b"nuc":
objclass = Nuc_Seq objclass = Nuc_Seq # Nuc_Seq_Stored? TODO
if format==b"fasta": if format==b"fasta":
iseq = fastaNucIterator(file, if input:
iseq = fastaNucIterator(file,
skip=skip,
only=only)
else:
iseq = FastaNucWriter(FastaFormat(printNAKeys=printna, NAString=nastring),
file,
skip=skip,
only=only)
elif format==b"fastq":
if input:
iseq = fastqIterator(file,
skip=skip,
only=only,
offset=offset,
noquality=noquality)
else:
iseq = FastqWriter(FastqFormat(printNAKeys=printna, NAString=nastring),
file,
skip=skip,
only=only)
elif format==b"embl":
if input:
iseq = emblIterator(file,
skip=skip, skip=skip,
only=only) only=only)
elif format==b"fastq": else:
iseq = fastqIterator(file, raise NotImplementedError('Output sequence file format not implemented')
skip=skip,
only=only,
offset=offset,
noquality=noquality)
elif format==b"embl":
iseq = emblIterator(file,
skip=skip,
only=only)
else: else:
raise NotImplementedError('Sequence file format not implemented') raise NotImplementedError('Sequence file format not implemented')
elif seqtype==b"prot": elif seqtype==b"prot":
raise NotImplementedError() raise NotImplementedError()
elif format==b"tabular": elif format==b"tabular":
objclass = dict objclass = dict
iseq = tabIterator(file, if input:
header = header, iseq = tabIterator(file,
sep = sep, header = header,
dec = dec, sep = sep,
stripwhite = stripwhite, dec = dec,
blanklineskip = blanklineskip, stripwhite = stripwhite,
commentchar = commentchar, blanklineskip = blanklineskip,
skip = skip, commentchar = commentchar,
only = only) skip = skip,
only = only)
else:
raise NotImplementedError('Output sequence file format not implemented')
elif format==b"ngsfilter": elif format==b"ngsfilter":
objclass = dict objclass = dict
iseq = ngsfilterIterator(file, if input:
sep = sep, iseq = ngsfilterIterator(file,
dec = dec, sep = sep,
stripwhite = stripwhite, dec = dec,
blanklineskip = blanklineskip, stripwhite = stripwhite,
commentchar = commentchar, blanklineskip = blanklineskip,
skip = skip, commentchar = commentchar,
only = only) skip = skip,
only = only)
else:
raise NotImplementedError('Output sequence file format not implemented')
else: else:
iseq, objclass = entryIteratorFactory(file, if input:
skip, only, iseq, objclass = entryIteratorFactory(file,
seqtype, skip, only,
offset, seqtype,
noquality, offset,
skiperror, noquality,
header, skiperror,
sep, header,
dec, sep,
nastring, dec,
stripwhite, nastring,
blanklineskip, stripwhite,
commentchar) blanklineskip,
commentchar)
else: # default export is in fasta? or tab? TODO
objclass = Nuc_Seq # Nuc_Seq_Stored? TODO
iseq = FastaNucWriter(FastaFormat(printNAKeys=printna, NAString=nastring),
file,
skip=skip,
only=only)
#tmpdms = get_temp_dms() #tmpdms = get_temp_dms()