def parse( path,
tablenames = None,
required = not REQUIRED,
keycolname = None,
tablename2key = lambda x: x,
stream_beginTable = None,
stream_row = None):
f = open(path, 'r')
if tablenames:
tablenames_found = dict([(tablename, False) for tablename in tablenames])
if not __is_datalib_file(f):
raise InvalidFileError(path)
version = common_functions.get_version(f.readline())
if version > CURRENT_VERSION:
raise InvalidFileError('invalid version (%s)' % version)
if version < 3:
schema = 'table'
colformat = 'fixed'
else:
schema = common_functions.get_equals_decl(f.readline(), 'schema')
colformat = common_functions.get_equals_decl(f.readline(), 'colformat')
assert( schema in ['table', 'single'] )
assert( colformat in ['fixed', 'none'] )
class TableResult:
def __init__(self):
self.tables = {}
def beginTable( self,
tablename,
colnames,
coltypes,
path,
table_index,
keycolname):
self.table = Table(tablename,
colnames,
coltypes,
path,
table_index,
keycolname = keycolname)
self.tables[tablename2key(tablename)] = self.table
def row(self, data):
row = self.table.createRow()
for col in self.table.columns():
col.set(row.index, data[col.index])
def retval(self):
return self.tables
class StreamResult:
def __init__(self):
pass
def beginTable(self,
tablename,
colnames,
coltypes,
path,
table_index,
keycolname):
if stream_beginTable:
stream_beginTable(tablename,
colnames,
coltypes,
path,
table_index,
keycolname)
# a private table used for a row context in callback
self.table = Table(tablename,
colnames,
coltypes,
path,
table_index,
keycolname = keycolname)
self.__row = self.table.createRow()
def row(self, data):
if stream_row:
for col in self.table.columns():
col.set(self.__row.index, data[col.index])
stream_row(self.__row)
def retval(self):
return None
if stream_beginTable == None and stream_row == None:
result = TableResult()
else:
result = StreamResult()
table_index = -1
if schema == 'single':
__seek_meta(f, 'L')
colnames = __parse_colnames( f )
coltypes = __parse_coltypes( f )
while True:
tablename = __seek_next_tag(f)
if not tablename: break
table_index += 1
if tablenames:
if not tablename in tablenames:
__seek_end_tag(f, tablename)
continue
else:
tablenames_found[tablename] = True
if schema == 'table':
colnames = __parse_colnames( f )
f.readline() # skip blank line
coltypes = __parse_coltypes( f )
f.readline() # skip blank line
# --- begin table
result.beginTable(tablename,
colnames,
coltypes,
path,
table_index,
keycolname)
found_end_tag = False
# --- parse data until we reach </name>
while True:
line = f.readline()
tag = __get_end_tag(line)
if tag:
assert(tag == tablename)
found_end_tag = True
break
data = line.split()
if len(data) == 0:
raise InvalidFileError("Missing end tag for %s" % tablename)
elif len(data) != len(colnames):
raise InvalidFileError("Missing data for %s" % tablename)
result.row(data)
if tablenames and required:
for name, found in tablenames_found.items():
if not found:
raise MissingTableError('Failed to find %s in %s' % (name, path))
return result.retval();
def parse(path, tablenames=None, required=not REQUIRED, keycolname=None):
f = open(path, 'r')
if tablenames:
tablenames_found = dict([(tablename, False)
for tablename in tablenames])
if not __is_datalib_file(f):
raise InvalidFileError(path)
version = common_functions.get_version(f.readline())
if version > CURRENT_VERSION:
raise InvalidFileError('invalid version (%s)' % version)
tables = {}
table_index = -1
while True:
tablename = __seek_next_tag(f)
if not tablename: break
table_index += 1
if tablenames:
if not tablename in tablenames:
__seek_end_tag(f, tablename)
continue
else:
tablenames_found[tablename] = True
# --- read column names
colnames = f.readline().split()
if len(colnames) == 0:
raise InvalidFileError('expecting column labels')
elif colnames[0] != COLUMN_LABEL_MARKER:
raise InvalidFileError('unexpected token (%s)' % colnames[0])
colnames.pop(0) # remove marker
f.readline() # skip blank line
# --- read column types
coltypes = f.readline().split()
if len(coltypes) == 0:
raise InvalidFileError('expecting column types')
elif coltypes[0] != COLUMN_TYPE_MARKER:
raise InvalidFileError('unexpected token (%s)' % coltypes[0])
coltypes.pop(0) # remove marker
f.readline() # skip blank line
# --- construct table object
table = Table(tablename,
colnames,
coltypes,
path,
table_index,
keycolname=keycolname)
tables[tablename] = table
found_end_tag = False
# --- parse data until we reach </name>
while True:
line = f.readline()
tag = __get_end_tag(line)
if tag:
assert (tag == tablename)
found_end_tag = True
break
data = line.split()
if len(data) == 0:
raise InvalidFileError("Missing end tag for %s" % tablename)
elif len(data) != len(colnames):
raise InvalidFileError("Missing data for %s" % tablename)
row = table.createRow()
for col in table.columns():
col.set(row.index, data[col.index])
if tablenames and required:
for name, found in tablenames_found.items():
if not found:
raise MissingTableError('Failed to find %s in %s' %
(name, path))
return tables
def parse(path, tablenames = None, required = not REQUIRED, keycolname = None):
f = open(path, 'r')
if tablenames:
tablenames_found = dict([(tablename, False) for tablename in tablenames])
if not __is_datalib_file(f):
raise InvalidFileError(path)
version = common_functions.get_version(f.readline())
if version > CURRENT_VERSION:
raise InvalidFileError('invalid version (%s)' % version)
tables = {}
table_index = -1
while True:
tablename = __seek_next_tag(f)
if not tablename: break
table_index += 1
if tablenames:
if not tablename in tablenames:
__seek_end_tag(f, tablename)
continue
else:
tablenames_found[tablename] = True
# --- read column names
colnames = f.readline().split()
if len(colnames) == 0:
raise InvalidFileError('expecting column labels')
elif colnames[0] != COLUMN_LABEL_MARKER:
raise InvalidFileError('unexpected token (%s)' % colnames[0])
colnames.pop(0) # remove marker
f.readline() # skip blank line
# --- read column types
coltypes = f.readline().split()
if len(coltypes) == 0:
raise InvalidFileError('expecting column types')
elif coltypes[0] != COLUMN_TYPE_MARKER:
raise InvalidFileError('unexpected token (%s)' % coltypes[0])
coltypes.pop(0) # remove marker
f.readline() # skip blank line
# --- construct table object
table = Table(tablename,
colnames,
coltypes,
path,
table_index,
keycolname = keycolname)
tables[tablename] = table
found_end_tag = False
# --- parse data until we reach </name>
while True:
line = f.readline()
tag = __get_end_tag(line)
if tag:
assert(tag == tablename)
found_end_tag = True
break
data = line.split()
if len(data) == 0:
raise InvalidFileError("Missing end tag for %s" % tablename)
elif len(data) != len(colnames):
raise InvalidFileError("Missing data for %s" % tablename)
row = table.createRow()
for col in table.columns():
col.set(row.index, data[col.index])
if tablenames and required:
for name, found in tablenames_found.items():
if not found:
raise MissingTableError('Failed to find %s in %s' % (name, path))
return tables
def parse( path,
tablenames = None,
required = not REQUIRED,
keycolname = None,
tablename2key = lambda x: x,
stream_beginTable = None,
stream_row = None):
f = open(path, 'r')
if tablenames:
tablenames_found = dict([(tablename, False) for tablename in tablenames])
if not __is_datalib_file(f):
raise InvalidFileError(path)
version = common_functions.get_version(f.readline())
if version > CURRENT_VERSION:
raise InvalidFileError('invalid version (%s)' % version)
if version < 3:
schema = 'table'
colformat = 'fixed'
else:
schema = common_functions.get_equals_decl(f.readline(), 'schema')
colformat = common_functions.get_equals_decl(f.readline(), 'colformat')
assert( schema in ['table', 'single'] )
assert( colformat in ['fixed', 'none'] )
class TableResult:
def __init__(self):
self.tables = {}
def beginTable( self,
tablename,
colnames,
coltypes,
path,
table_index,
keycolname):
self.table = Table(tablename,
colnames,
coltypes,
path,
table_index,
keycolname = keycolname)
self.tables[tablename2key(tablename)] = self.table
def row(self, data):
row = self.table.createRow()
for col in self.table.columns():
col.set(row.index, data[col.index])
def retval(self):
return self.tables
class StreamResult:
def __init__(self):
pass
def beginTable(self,
tablename,
colnames,
coltypes,
path,
table_index,
keycolname):
if stream_beginTable:
stream_beginTable(tablename,
colnames,
coltypes,
path,
table_index,
keycolname)
# a private table used for a row context in callback
self.table = Table(tablename,
colnames,
coltypes,
path,
table_index,
keycolname = keycolname)
self.__row = self.table.createRow()
def row(self, data):
if stream_row:
for col in self.table.columns():
col.set(self.__row.index, data[col.index])
stream_row(self.__row)
def retval(self):
return None
if stream_beginTable == None and stream_row == None:
result = TableResult()
else:
result = StreamResult()
table_index = -1
if schema == 'single':
__seek_meta(f, 'L')
colnames = __parse_colnames( f )
coltypes = __parse_coltypes( f )
while True:
tablename = __seek_next_tag(f)
if not tablename: break
table_index += 1
if tablenames:
if not tablename in tablenames:
__seek_end_tag(f, tablename)
continue
else:
tablenames_found[tablename] = True
if schema == 'table':
colnames = __parse_colnames( f )
f.readline() # skip blank line
coltypes = __parse_coltypes( f )
f.readline() # skip blank line
# --- begin table
result.beginTable(tablename,
colnames,
coltypes,
path,
table_index,
keycolname)
found_end_tag = False
# --- parse data until we reach </name>
while True:
line = f.readline()
tag = __get_end_tag(line)
if tag:
assert(tag == tablename)
found_end_tag = True
break
data = line.split()
if len(data) == 0:
raise InvalidFileError("Missing end tag for %s" % tablename)
elif len(data) != len(colnames):
raise InvalidFileError("Missing data for %s" % tablename)
result.row(data)
if tablenames and required:
for name, found in tablenames_found.items():
if not found:
raise MissingTableError('Failed to find %s in %s' % (name, path))
return result.retval();