def terms_not_in_darwin_cloud(checklist,
dwccloudfile,
encoding=None,
vetted=True,
casesensitive=False):
''' Get the list of distinct values in a checklist that are not in the Darwin Cloud
vocabulary. Verbatim values in the Darwin Cloud vocabulary should be lower-case and
stripped already, so that is what must be matched here. The Darwin Cloud vocabulary
should have the case-sensitive standard value.
parameters:
checklist - list of values to check against the target list (required)
dwccloudfile - the vocabulary file for the Darwin Cloud (required)
vetted - set to False if unvetted values should also be returned (default True)
encoding - a string designating the input file encoding (optional; default None)
(e.g., 'utf-8', 'mac_roman', 'latin_1', 'cp1252')
returns:
a sorted list of distinct new values not in the Darwin Cloud vocabulary
'''
functionname = 'terms_not_in_darwin_cloud()'
if checklist is None or len(checklist) == 0:
s = 'No checklist given in %s.' % functionname
logging.debug(s)
return None
dialect = csv_file_dialect(dwccloudfile)
# Try to determine the encoding of the inputfile.
if encoding is None or len(encoding.strip()) == 0:
encoding = csv_file_encoding(dwccloudfile)
# csv_file_encoding() always returns an encoding if there is an input file.
# No need to check if dwccloudfile is given and exists, vocab_dict_from_file() and
# vetted_vocab_dict_from_file() do that.
if vetted == True:
darwinclouddict = vetted_vocab_dict_from_file(dwccloudfile,
'fieldname',
dialect=dialect,
encoding=encoding)
else:
darwinclouddict = vocab_dict_from_file(dwccloudfile,
'fieldname',
dialect=dialect,
encoding=encoding)
dwcloudlist = []
for key, value in darwinclouddict.iteritems():
dwcloudlist.append(key)
if casesensitive == True:
return not_in_list(dwcloudlist, checklist)
lowerdwclist = []
for term in dwcloudlist:
lowerdwclist.append(ustripstr(term))
notfound = not_in_list(lowerdwclist, checklist, function=ustripstr)
return notfound
def text_file_field_stripper(options):
''' Filter a text file into a new file based on matching a list of fields to keep.
options - a dictionary of parameters
loglevel - level at which to log (e.g., DEBUG) (optional)
workspace - the directory in which the output will be written (optional)
inputfile - full path to the input file (required)
outputfile - name of the output file, without path (required)
separator - string that separates the values in termlist (e.g., '|')
(optional; default None)
encoding - string signifying the encoding of the input file. If known, it speeds
up processing a great deal. (optional; default None) (e.g., 'utf-8')
format - output file format (e.g., 'csv' or 'txt') (optional; default 'txt')
termlist - list of fields to extract from the input file (required)
returns a dictionary with information about the results
workspace - actual path to the directory where the outputfile was written
outputfile - actual full path to the output tsv file
success - True if process completed successfully, otherwise False
message - an explanation of the reason if success=False
artifacts - a dictionary of persistent objects created
'''
#print '%s options: %s' % (__version__, options)
setup_actor_logging(options)
logging.debug( 'Started %s' % __version__ )
logging.debug( 'options: %s' % options )
# Make a list for the response
returnvars = ['workspace', 'outputfile', 'success', 'message', 'artifacts']
### Standard outputs ###
success = False
message = None
# Make a dictionary for artifacts left behind
artifacts = {}
### Establish variables ###
workspace = './'
inputfile = None
outputfile = None
format = 'txt'
termlist = None
separator = None
encoding = None
### Required inputs ###
try:
workspace = options['workspace']
except:
pass
try:
inputfile = options['inputfile']
except:
pass
if inputfile is None or len(inputfile)==0:
message = 'No input file given. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
if os.path.isfile(inputfile) == False:
message = 'Input file %s not found. %s' % (inputfile, __version__)
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
try:
termlist = options['termlist']
except:
pass
if termlist is None or len(termlist)==0:
message = 'No termlist given. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message: %s' % message)
return response(returnvars, returnvals)
try:
separator = options['separator']
except:
pass
try:
encoding = options['encoding']
except:
pass
if separator is None or len(separator.strip())==0:
theterms = [termlist]
else:
theterms = termlist.split(separator)
# Determine the file dialect
inputdialect = csv_file_dialect(inputfile)
# Determine the file encoding
if encoding is None or len(encoding.strip()) == 0:
encoding = csv_file_encoding(inputfile)
# csv_file_encoding() always returns an encoding if there is an input file.
# No need to check.
# If the termname is not in the header of the inputfile, nothing to do.
header = read_header(inputfile, dialect=inputdialect, encoding=encoding)
# Make a clean version of the input header
cleaninputheader = clean_header(header)
try:
format = options['format']
except:
pass
try:
outputfile = options['outputfile']
except:
pass
if outputfile is None or len(outputfile)==0:
message = 'No output file given. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
outputfile = '%s/%s' % (workspace.rstrip('/'), outputfile)
# Prepare the outputfile
if format is None or format.lower()=='txt':
outputdialect = tsv_dialect()
else:
outputdialect = csv_dialect()
if separator is None or len(separator.strip())==0:
theterms = [termlist]
else:
theterms = termlist.split(separator)
# Make a clean version of the output header
cleanoutputheader = clean_header(theterms)
# Create the outputfile and write the new header to it
write_header(outputfile, cleanoutputheader, outputdialect)
# Check to see that the file was created
if os.path.isfile(outputfile) == False:
message = 'Outputfile %s was not created. %s' % (outputfile, __version__)
returnvals = [workspace, outputfile, success, message, artifacts]
return response(returnvars, returnvals)
# Open the outputfile to start writing matching rows
with open(outputfile, 'a') as outfile:
writer = csv.DictWriter(outfile, dialect=outputdialect, encoding='utf-8',
fieldnames=cleanoutputheader)
# Iterate through all rows in the input file
for row in read_csv_row(inputfile, dialect=inputdialect, encoding=encoding,
header=True, fieldnames=cleaninputheader):
newrow = extract_fields_from_row(row, cleanoutputheader)
writer.writerow(newrow)
success = True
s = 'stripped_file'
artifacts[s] = outputfile
# Prepare the response dictionary
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('Finishing %s' % __version__)
return response(returnvars, returnvals)
def term_standardizer_report(inputfile,
reportfile,
vocabfile,
key,
separator=None,
encoding=None,
format=None):
''' Write a file with substitutions from a vocabfile for fields in a key and appended
terms showing the original values.
parameters:
inputfile - full path to the input file (required)
reportfile - full path to the output file (required)
vocabfile - path to the vocabulary file (required)
key - field or separator-separated fields to set (required)
separator - string to use as the key and value separator (optional; default '|')
encoding - string signifying the encoding of the input file. If known, it speeds
up processing a great deal. (optional; default None) (e.g., 'utf-8')
format - string signifying the csv.dialect of the report file ('csv' or 'txt')
(optional; default: txt)
returns:
success - True if the report was written, else False
'''
functionname = 'term_standardizer_report()'
if reportfile is None or len(reportfile) == 0:
s = 'No reportfile name given in %s.' % functionname
logging.debug(s)
return False
if inputfile is None or len(inputfile) == 0:
s = 'No inputfile file given in %s.' % functionname
logging.debug(s)
return False
if os.path.isfile(inputfile) == False:
s = 'Inputfile file %s not found in %s.' % (inputfile, functionname)
logging.debug(s)
return False
# Determine the dialect of the input file
inputdialect = csv_file_dialect(inputfile)
# Determine the dialect of the input file
if encoding is None or len(encoding.strip()) == 0:
encoding = csv_file_encoding(inputfile)
# Read the header from the input file
inputheader = read_header(inputfile,
dialect=inputdialect,
encoding=encoding)
if inputheader is None:
s = 'Unable to read header from input file %s in %s.' % (inputfile,
functionname)
logging.debug(s)
return False
if key is None or len(key.strip()) == 0:
s = 'No key given in %s.' % functionname
logging.debug(s)
return False
# Make sure there is a separator for the next step
if separator is None or len(separator) == 0:
separator = '|'
# Make a list of the fields in the key by splitting it on the separator
fieldlist = key.split(separator)
# Assume none of the fields is in the file
headerhaskey = False
# Search the cleaned up header for any field from the key
cleanedinputheader = strip_list(inputheader)
for field in fieldlist:
if field in cleanedinputheader:
headerhaskey = True
break
if headerhaskey == False:
s = 'No field from %s found ' % fieldlist
s += 'in input file %s in %s.' % (inputfile, functionname)
logging.debug(s)
return False
if vocabfile is None or len(vocabfile) == 0:
logging.debug('No vocabulary file given in %s.') % functionname
return False
if os.path.isfile(vocabfile) == False:
s = 'Vocabulary file %s not found in %s.' % (vocabfile, functionname)
logging.debug(s)
return False
# Get the vocabulary dictionary, but convert all entries using ustripstr. Assume
# vocabulary file is encoded as utf-8.
vocabdict = vocab_dict_from_file(vocabfile, key, encoding='utf-8', \
separator=separator, function=ustripstr)
if len(vocabdict) == 0:
s = 'Vocabulary file %s ' % vocabfile
s += 'had zero recommendations in %s.' % functionname
logging.debug(s)
return False
if format is None or format.lower() == 'txt':
dialect = tsv_dialect()
else:
dialect = csv_dialect()
if format is None or format.lower() == 'txt':
outputdialect = tsv_dialect()
else:
outputdialect = csv_dialect()
# Create an output header that is the same as the input header with fields
# appended to hold the original values of the key fields
# Get the fields to add by splitting the key with the separator
outputheader = cleanedinputheader
for field in fieldlist:
if field in outputheader:
outputheader = outputheader + [field + '_orig']
else:
outputheader = outputheader + [field]
# Create the outputfile and write the new header to it
write_header(reportfile, outputheader, outputdialect)
# Check to see if the outputfile was created
if os.path.isfile(reportfile) == False:
s = 'reportfile: %s not created in %s.' % (reportfile, functionname)
logging.debug(s)
return False
# Open the outputfile to append rows having the added fields
with open(reportfile, 'a') as outfile:
writer = csv.DictWriter(outfile,
dialect=outputdialect,
encoding='utf-8',
fieldnames=outputheader)
# Iterate through all rows in the input file
for row in read_csv_row(inputfile,
dialect=inputdialect,
encoding=encoding,
header=True,
fieldnames=cleanedinputheader):
# Set the _orig values for every field in the field list that exists in
# the row
for field in fieldlist:
if field in row:
row[field + '_orig'] = row[field]
# Construct a composite field value for the row to match a key in the
# vocabulary file
rowkey = extract_values_from_row(row, fieldlist, separator)
# Get dictionary for recommended value for the ustripstr(rowkey)
newvaluedict = recommended_value(vocabdict, ustripstr(rowkey))
# Only make changes if there is a standardized value found
if newvaluedict is not None:
# ustripstr(rowkey) was found in the vocabulary
# Get the standard value
standard = newvaluedict['standard']
# Treat standard value that is None or only whitespace as ''
if standard is None or len(standard.strip()) == 0:
standard = ''
# Make a list of values given in standard
newvalues = standard.split(separator)
# Only make changes if the number of recommendation fields is the
# same as the number of fields in the key
if len(newvalues) == len(fieldlist):
i = 0
# Update or add new value to field in the fieldlist
for field in fieldlist:
row[field] = newvalues[i]
i += 1
writer.writerow(row)
s = 'Report written to %s in %s.' % (reportfile, functionname)
logging.debug(s)
return True
def uuid_term_appender(inputfile,
outputfile,
key,
guidtype=None,
encoding=None,
format=None):
''' Write a file adding a field populated by global unique identifiers (GUIDs) to the
fields in the input file.
parameters:
inputfile - full path to the input file (required)
outputfile - full path to the output file (required)
key - field or separator-separated fields to set (required)
guidtype - type of GUID to use to populate the key (optional; default 'uuid')
encoding - string signifying the encoding of the input file. If known, it speeds
up processing a great deal. (optional; default None) (e.g., 'utf-8')
format - string signifying the csv.dialect of the report file ('csv' or 'txt')
(optional; default: txt)
returns:
success - True if the report was written, else False
'''
functionname = 'uuid_term_appender()'
if outputfile is None or len(outputfile) == 0:
s = 'No outputfile name given in %s.' % functionname
logging.debug(s)
return False
if inputfile is None or len(inputfile) == 0:
s = 'No inputfile file given in %s.' % functionname
logging.debug(s)
return False
if os.path.isfile(inputfile) == False:
s = 'Inputfile file %s not found in %s.' % (inputfile, functionname)
logging.debug(s)
return False
# Determine the dialect of the input file
inputdialect = csv_file_dialect(inputfile)
# Determine the dialect of the input file
if encoding is None or len(encoding.strip()) == 0:
encoding = csv_file_encoding(inputfile)
# Read the header from the input file
inputheader = read_header(inputfile,
dialect=inputdialect,
encoding=encoding)
if inputheader is None:
s = 'Unable to read header from input file %s in %s.' % (inputfile,
functionname)
logging.debug(s)
return False
if key is None or len(key.strip()) == 0:
s = 'No key given in %s.' % functionname
logging.debug(s)
return False
# Abort if the key exists in the inputheader
if key in inputheader:
s = 'field %s ' % key
s += 'already exists in file %s ' % inputfile
s += 'in %s.' % functionname
logging.debug(s)
return False
if format is None or format.lower() == 'txt':
outputdialect = tsv_dialect()
else:
outputdialect = csv_dialect()
# Make an outputheader that is a copy of the inputheader plus the new field to hold
# GUID.
outputheader = inputheader + [key]
# Create the outputfile and write the new header to it
write_header(outputfile, outputheader, outputdialect)
# Check to see if the outputfile was created
if os.path.isfile(outputfile) == False:
s = 'outputfile: %s was not created in %s.' % (outputfile,
functionname)
logging.debug(s)
return False
# Open the outputfile to append rows with appended GUID field
with open(outputfile, 'a') as outfile:
writer = csv.DictWriter(outfile,
dialect=outputdialect,
encoding='utf-8',
fieldnames=outputheader)
# Iterate through all rows in the input file
for row in read_csv_row(inputfile,
dialect=inputdialect,
encoding=encoding,
header=True,
fieldnames=inputheader):
# Create a GUID based on the selected guidtype
guid = get_guid(guidtype)
# Set the value of the key field to a GUID
row[key] = guid
# Write the updated row to the outputfile
writer.writerow(row)
s = 'Output file written to %s in %s.' % (outputfile, functionname)
logging.debug(s)
return True
def term_setter_report(inputfile,
reportfile,
key,
constantvalues=None,
separator=None,
encoding=None,
format=None):
''' Write a file substituting constants for fields that already exist in an input file
and with added fields with constants for fields that do not already exist in an
inputfile. Field name matching is exact.
parameters:
inputfile - full path to the input file (required)
reportfile - full path to the output file (required)
key - field or separator-separated fields to set (required)
constantvalues - value or separator-separated values to set the field(s) to
(required)
separator - string to use as the key and value separator (optional; default '|')
encoding - string signifying the encoding of the input file. If known, it speeds
up processing a great deal. (optional; default None) (e.g., 'utf-8')
format - string signifying the csv.dialect of the report file ('csv' or 'txt')
(optional; default: txt)
returns:
success - True if the report was written, else False
'''
functionname = 'term_setter_report()'
if reportfile is None or len(reportfile) == 0:
s = 'No reportfile name given in %s.' % functionname
logging.debug(s)
return False
if inputfile is None or len(inputfile) == 0:
s = 'No inputfile file given in %s.' % functionname
logging.debug(s)
return False
if os.path.isfile(inputfile) == False:
s = 'Inputfile file %s not found in %s.' % (inputfile, functionname)
logging.debug(s)
return False
# Determine the dialect of the input file
inputdialect = csv_file_dialect(inputfile)
# Determine the dialect of the input file
if encoding is None or len(encoding.strip()) == 0:
encoding = csv_file_encoding(inputfile)
# Read the header from the input file
inputheader = read_header(inputfile,
dialect=inputdialect,
encoding=encoding)
if inputheader is None:
s = 'Unable to read header from input file %s in %s.' % (inputfile,
functionname)
logging.debug(s)
return False
if key is None or len(key.strip()) == 0:
s = 'No key given in %s.' % functionname
logging.debug(s)
return False
if constantvalues is None or len(constantvalues) == 0:
s = 'No constantvalues given in %s.' % functionname
logging.debug(s)
return False
# Make sure there is a separator for the next step
if separator is None or len(separator) == 0:
separator = '|'
# Get the fields to set by splitting the key with the separator
fields = key.split(separator)
# Get the values to set by splitting the constantvalues with the separator
addedvalues = constantvalues.split(separator)
# Abort if there is a mismatch in the lengths of the field and constants lists
if len(fields) != len(addedvalues):
s = 'length of field list: %s ' % key
s += 'does not match length of constants list: %s ' % constantvalues
s += 'in %s.' % functionname
logging.debug(s)
return False
if format is None or format.lower() == 'txt':
outputdialect = tsv_dialect()
else:
outputdialect = csv_dialect()
# Make an outputheader that is a copy of the inputheader
outputheader = inputheader
# Add to the output header fields that are not in the inputheader
for field in fields:
if field not in outputheader:
outputheader = outputheader + [field]
# Create the outputfile and write the new header to it
write_header(reportfile, outputheader, outputdialect)
# Check to see if the outputfile was created
if os.path.isfile(reportfile) == False:
s = 'reportfile: %s was not created in %s.' % (outputfile,
functionname)
logging.debug(s)
return False
# Open the outputfile to append rows with fields set to constant values
with open(reportfile, 'a') as outfile:
writer = csv.DictWriter(outfile,
dialect=outputdialect,
encoding='utf-8',
fieldnames=outputheader)
# Iterate through all rows in the input file
for row in read_csv_row(inputfile,
dialect=inputdialect,
encoding=encoding,
header=True,
fieldnames=inputheader):
# For every field in the key list
for i in range(0, len(fields)):
# Set the value of the ith field to the ith constant
row[fields[i]] = addedvalues[i]
# Write the updated row to the outputfile
writer.writerow(row)
s = 'Report written to %s in %s.' % (reportfile, functionname)
logging.debug(s)
return True
def darwinize_header(options):
''' Translate field names from input file to Darwin Core field names in outputfile
using a Darwin Cloud vocabulary lookup.
options - a dictionary of parameters
loglevel - level at which to log (e.g., DEBUG) (optional)
workspace - path to a directory for the outputfile (optional)
inputfile - full path to the input file (required)
dwccloudfile - full path to the vocabulary file containing the Darwin Cloud
terms (required)
outputfile - name of the output file, without path (required)
encoding - string signifying the encoding of the input file. If known, it speeds
up processing a great deal. (optional; default None) (e.g., 'utf-8')
format - output file format (e.g., 'csv' or 'txt') (optional; default 'txt')
namespace - prepend namespace to fields that were darwinized
(optional; default 'no') (e.g., 'y', 'n')
returns a dictionary with information about the results
outputfile - actual full path to the output file
success - True if process completed successfully, otherwise False
message - an explanation of the reason if success=False
'''
#print '%s options: %s' % (__version__, options)
setup_actor_logging(options)
logging.debug('Started %s' % __version__)
logging.debug('options: %s' % options)
# Make a list for the response
returnvars = ['workspace', 'outputfile', 'success', 'message', 'artifacts']
### Standard outputs ###
success = False
message = None
# Make a dictionary for artifacts left behind
artifacts = {}
### Establish variables ###
workspace = './'
inputfile = None
dwccloudfile = None
outputfile = None
encoding = None
namespace = 'n'
format = None
### Required inputs ###
try:
workspace = options['workspace']
except:
pass
try:
inputfile = options['inputfile']
except:
pass
if inputfile is None or len(inputfile) == 0:
message = 'No input file given. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
if os.path.isfile(inputfile) == False:
message = 'Input file %s not found. %s' % (inputfile, __version__)
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
try:
dwccloudfile = options['dwccloudfile']
except:
pass
if dwccloudfile is None or len(dwccloudfile) == 0:
message = 'No Darwin Cloud vocabulary file given. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
if os.path.isfile(dwccloudfile) == False:
message = 'Darwin Cloud vocabulary file not found. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
try:
outputfile = options['outputfile']
except:
pass
if outputfile is None or len(outputfile) == 0:
message = 'No output file given. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
outputfile = '%s/%s' % (workspace.rstrip('/'), outputfile)
try:
encoding = options['encoding']
except:
pass
if encoding is None or len(encoding.strip()) == 0:
encoding = csv_file_encoding(inputfile)
try:
namespace = options['namespace']
except:
pass
inputdialect = csv_file_dialect(inputfile)
try:
format = options['format']
except:
pass
if format is None or len(format) == 0:
outputdialect = inputdialect
elif format.lower() == 'csv':
outputdialect = csv_dialect()
else:
outputdialect = tsv_dialect()
header = read_header(inputfile, dialect=inputdialect, encoding=encoding)
dwcheader = darwinize_list(header, dwccloudfile, namespace)
if dwcheader is None:
message = 'Unable to create darwinized header. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
# Write the new header to the outputfile
if write_header(outputfile, dwcheader, dialect=outputdialect) == False:
message = 'Unable to write header to output file. %s' % __version__
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
# Read the rows of the input file, append them to the output file after the
# header with columns in the same order.
with open(outputfile, 'a') as outfile:
writer = csv.DictWriter(outfile,
dialect=outputdialect,
encoding='utf-8',
fieldnames=header)
for row in read_csv_row(inputfile, inputdialect, encoding):
writer.writerow(row)
#print 'row: %s' % row
success = True
artifacts['darwinized_header_file'] = outputfile
returnvals = [workspace, outputfile, success, message, artifacts]
logging.debug('Finishing %s' % __version__)
return response(returnvars, returnvals)
def text_file_aggregator(options):
''' Join the contents of files in a given path. Headers and encodings are not assumed
to be the same. Write a file containing the joined files with one header line.
options - a dictionary of parameters
loglevel - level at which to log (e.g., DEBUG) (optional)
workspace - path to a directory for the outputfile (optional)
inputpath - full path to the input file set. The file extension of the outputfile
will be the substring following the last '.' in the inputpath.
Example: ./workspace/thefiles.txt will produce an output file ending in
'.txt' (required)
outputfile - name of the output file, without path (optional)
format - output file format (e.g., 'csv' or 'txt') (optional; default 'txt')
returns a dictionary with information about the results
workspace - actual path to the directory where the outputfile was written
outputfile - actual full path to the output file
aggregaterowcount - the number of rows in the aggregated file, not counting header
success - True if process completed successfully, otherwise False
message - an explanation of the reason if success=False
artifacts - a dictionary of persistent objects created
'''
#print '%s options: %s' % (__version__, options)
setup_actor_logging(options)
logging.debug( 'Started %s' % __version__ )
logging.debug( 'options: %s' % options )
# Make a list for the response
returnvars = ['workspace', 'outputfile', 'aggregaterowcount', 'success', 'message',
'artifacts']
### Standard outputs ###
success = False
message = None
### Custom outputs ###
aggregaterowcount = None
# Make a dictionary for artifacts left behind
artifacts = {}
### Establish variables ###
workspace = './'
inputpath = None
outputfile = None
format = None
### Required inputs ###
try:
workspace = options['workspace']
except:
pass
try:
inputpath = options['inputpath']
except:
pass
if inputpath is None or len(inputpath)==0:
message = 'No input file given. %s' % __version__
returnvals = [workspace, outputfile, aggregaterowcount, success, message,
artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
try:
format = options['format']
except:
pass
if format is None or len(format)==0:
format = 'txt'
if format.lower() == 'txt':
dialect = tsv_dialect()
else:
dialect = csv_dialect()
try:
outputfile = options['outputfile']
except:
pass
if outputfile is None or len(outputfile)==0:
outputfile='aggregate_'+str(uuid.uuid1())+format
# Construct the output file path in the workspace
outputfile = '%s/%s' % (workspace.rstrip('/'), outputfile)
# Create the composite header. Let composite_header determine the dialects and
# encodings of the files to aggregate.
aggregateheader = composite_header(inputpath)
aggregaterowcount = 0
# Open a file to write the aggregated results in chosen format and utf-8.
with open(outputfile, 'w') as outfile:
writer = csv.DictWriter(outfile, dialect=dialect, encoding='utf-8',
fieldnames=aggregateheader, extrasaction='ignore')
writer.writeheader()
files = glob.glob(inputpath)
for file in files:
dialect = csv_file_dialect(file)
encoding = csv_file_encoding(file)
with open(file, 'rU') as inputfile:
reader = csv.DictReader(utf8_data_encoder(inputfile, encoding),
dialect=dialect, encoding=encoding)
for line in reader:
try:
writer.writerow(line)
aggregaterowcount += 1
except:
message = 'failed to write line:\n%s\n' % line
message += 'to file %s. %s' % (file, __version__)
returnvals = [workspace, outputfile, aggregaterowcount, success,
message, artifacts]
logging.debug('message:\n%s' % message)
return response(returnvars, returnvals)
success = True
artifacts['aggregated_file'] = outputfile
if aggregateheader is not None:
aggregateheader = list(aggregateheader)
returnvals = [workspace, outputfile, aggregaterowcount, success, message,
artifacts]
logging.debug('Finishing %s' % __version__)
return response(returnvars, returnvals)
def vocab_dict_from_file(vocabfile,
key,
separator=None,
dialect=None,
encoding=None,
function=None,
*args,
**kwargs):
''' Get a vocabulary as a dictionary from a file.
parameters:
vocabfile - path to the vocabulary file (required)
key - the field or separator-separated fieldnames that hold the distinct values
in the vocabulary file (required)
separator - string to use as the value separator in the string
(optional; default None)
dialect - a csv.dialect object with the attributes of the vocabulary file
(default None)
encoding - a string designating the input file encoding (optional; default None)
(e.g., 'utf-8', 'mac_roman', 'latin_1', 'cp1252')
function - function to call for each value to compare (default None)
args - unnamed parameters to function as tuple (optional)
kwargs - named parameters to function as dictionary (optional)
Example:
vocab_dict_from_file(v,k,function=ustripstr) would return all of the stripped,
uppercased keys and their values from the vocabfile v.
returns:
vocabdict - dictionary of complete vocabulary records
'''
functionname = 'vocab_dict_from_file()'
if key is None or len(key.strip()) == 0:
s = 'No key given in %s.' % functionname
logging.debug(s)
return None
if vocabfile is None or len(vocabfile) == 0:
s = 'No vocabulary file given in %s.' % functionname
logging.debug(s)
return None
if os.path.isfile(vocabfile) == False:
s = 'Vocabulary file %s not found in %s.' % (vocabfile, functionname)
logging.debug(s)
return None
if dialect is None:
dialect = vocab_dialect()
# Try to determine the encoding of the inputfile.
if encoding is None or len(encoding.strip()) == 0:
encoding = csv_file_encoding(vocabfile)
# csv_file_encoding() always returns an encoding if there is an input file.
# Set up the field names to match the standard vocabulary header
fieldnames = vocabheader(key, separator)
# Create a dictionary to hold the vocabulary
vocabdict = {}
# Iterate through all rows in the input file
for row in read_csv_row(vocabfile,
dialect,
encoding,
header=True,
fieldnames=fieldnames):
# Make a complete copy of the row
rowdict = copy.deepcopy(row)
value = row[key]
# Remove the key from the row copy
rowdict.pop(key)
newvalue = value
# If we are not supposed to apply a function to the key value
if function is not None:
newvalue = function(value, *args, **kwargs)
vocabdict[newvalue] = rowdict
return vocabdict
def term_token_count_from_file(inputfile,
termname,
dialect=None,
encoding=None):
''' Make a dictionary of tokens for a given term in a file along with the number of
times each occurs.
parameters:
inputfile - full path to the input file (required)
termname - term for which to count rows (required)
dialect - csv.dialect object with the attributes of the input files, which must
all have the same dialect if dialect is given, otherwise it will be detected
(default None)
encoding - a string designating the input file encoding (optional; default None)
(e.g., 'utf-8', 'mac_roman', 'latin_1', 'cp1252')
returns:
tokens - a dictionary containing the tokens and their counts
'''
functionname = 'term_token_count_from_file()'
if inputfile is None or len(inputfile) == 0:
s = 'No input file given in %s.' % functionname
logging.debug(s)
return 0
if os.path.isfile(inputfile) == False:
s = 'File %s not found in %s.' % (inputfile, functionname)
logging.debug(s)
return 0
if termname is None or len(termname) == 0:
s = 'No term name given in %s.' % functionname
logging.debug(s)
return 0
# Determine the dialect of the input file
if dialect is None:
dialect = csv_file_dialect(inputfile)
# csv_file_dialect() always returns a dialect if there is an input file.
# No need to check.
# Determine the encoding of the input file
if encoding is None or len(encoding.strip()) == 0:
encoding = csv_file_encoding(inputfile)
# csv_file_encoding() always returns an encoding if there is an input file.
# No need to check.
inputheader = read_header(inputfile, dialect=dialect, encoding=encoding)
if termname not in inputheader:
s = 'Term %s not found in file %s ' % (termname, inputfile)
s += 'in %s.' % functionname
logging.debug(s)
return None
rowcount = 0
tokencount = 0
populatedrowcount = 0
tokens = {'tokenlist': {}}
for row in read_csv_row(inputfile, dialect, encoding):
try:
value = row[termname]
except:
pass
if value is not None and len(value.strip()) > 0:
rowdict = {}
wordlist = re.sub("[^\w]", " ", value).split()
for token in wordlist:
if token in rowdict:
rowdict[token][
'totalcount'] = rowdict[token]['totalcount'] + 1
else:
rowdict[token] = {}
rowdict[token]['rowcount'] = 1
rowdict[token]['totalcount'] = 1
populatedrowcount += 1
for key, value in rowdict.iteritems():
tokenlist = tokens['tokenlist']
if key in tokenlist:
tokenlist[key]['rowcount'] = \
tokenlist[key]['rowcount'] + value['rowcount']
tokenlist[key]['totalcount'] = \
tokenlist[key]['totalcount'] + value['totalcount']
else:
tokenlist[key] = {}
tokenlist[key]['rowcount'] = value['rowcount']
tokenlist[key]['totalcount'] = value['totalcount']
rowcount += 1
tokencount += len(wordlist)
tokens['rowcount'] = rowcount
tokens['tokencount'] = tokencount
tokens['input'] = inputfile
tokens['term'] = termname
return tokens