def _getJyDictionary(self, obj):
ret = {}
found = java.util.HashMap()
original = obj
if hasattr(obj, '__class__') and obj.__class__ == java.lang.Class:
#get info about superclasses
classes = []
classes.append(obj)
c = obj.getSuperclass()
while c != None:
classes.append(c)
c = c.getSuperclass()
#get info about interfaces
interfs = []
for obj in classes:
interfs.extend(obj.getInterfaces())
classes.extend(interfs)
#now is the time when we actually get info on the declared methods and fields
for obj in classes:
declaredMethods = obj.getDeclaredMethods()
declaredFields = obj.getDeclaredFields()
for i in xrange(len(declaredMethods)):
name = declaredMethods[i].getName()
ret[name] = declaredMethods[i].toString()
found.put(name, 1)
for i in xrange(len(declaredFields)):
name = declaredFields[i].getName()
found.put(name, 1)
#if declaredFields[i].isAccessible():
declaredFields[i].setAccessible(True)
#ret[name] = declaredFields[i].get( declaredFields[i] )
try:
ret[name] = declaredFields[i].get(original)
except:
ret[name] = declaredFields[i].toString()
#this simple dir does not always get all the info, that's why we have the part before
#(e.g.: if we do a dir on String, some methods that are from other interfaces such as
#charAt don't appear)
try:
d = dir(original)
for name in d:
if found.get(name) is not 1:
ret[name] = getattr(original, name)
except:
#sometimes we're unable to do a dir
pass
return ret
def array_to_meta_xml(array, name, format):
type = array.dtype.kind
slice = name
l = len(array.shape)
# initial load, compute slice
if format == '%':
if l > 2:
slice += '[0]' * (l - 2)
for r in xrange(l - 2):
array = array[0]
if type == 'f':
format = '.5f'
elif type == 'i' or type == 'u':
format = 'd'
else:
format = 's'
else:
format = format.replace('%', '')
l = len(array.shape)
reslice = ""
if l > 2:
raise Exception("%s has more than 2 dimensions." % slice)
elif l == 1:
# special case with 1D arrays arr[i, :] - row, but arr[:, i] - column with equal shape and ndim
# http://stackoverflow.com/questions/16837946/numpy-a-2-rows-1-column-file-loadtxt-returns-1row-2-columns
# explanation: http://stackoverflow.com/questions/15165170/how-do-i-maintain-row-column-orientation-of-vectors-in-numpy?rq=1
# we use kind of a hack - get information about memory from C_CONTIGUOUS
is_row = array.flags['C_CONTIGUOUS']
if is_row:
rows = 1
cols = min(len(array), MAX_SLICE_SIZE)
if cols < len(array):
reslice = '[0:%s]' % (cols)
array = array[0:cols]
else:
cols = 1
rows = min(len(array), MAX_SLICE_SIZE)
if rows < len(array):
reslice = '[0:%s]' % (rows)
array = array[0:rows]
elif l == 2:
rows = min(array.shape[-2], MAX_SLICE_SIZE)
cols = min(array.shape[-1], MAX_SLICE_SIZE)
if cols < array.shape[-1] or rows < array.shape[-2]:
reslice = '[0:%s, 0:%s]' % (rows, cols)
array = array[0:rows, 0:cols]
# avoid slice duplication
if not slice.endswith(reslice):
slice += reslice
bounds = (0, 0)
if type in "biufc":
bounds = (array.min(), array.max())
xml = '<array slice=\"%s\" rows=\"%s\" cols=\"%s\" format=\"%s\" type=\"%s\" max=\"%s\" min=\"%s\"/>' % \
(slice, rows, cols, format, type, bounds[1], bounds[0])
return array, xml, rows, cols, format
def get_dictionary(self, obj):
ret = {}
for i in xrange(len(obj)):
ret[ i ] = obj[i]
ret['__len__'] = len(obj)
return ret
def array_to_xml(array, roffset, coffset, rows, cols, format):
xml = ""
rows = min(rows, MAXIMUM_ARRAY_SIZE)
cols = min(cols, MAXIMUM_ARRAY_SIZE)
# there is no obvious rule for slicing (at least 5 choices)
if len(array) == 1 and (rows > 1 or cols > 1):
array = array[0]
if array.size > len(array):
array = array[roffset:, coffset:]
rows = min(rows, len(array))
cols = min(cols, len(array[0]))
if len(array) == 1:
array = array[0]
elif array.size == len(array):
if roffset == 0 and rows == 1:
array = array[coffset:]
cols = min(cols, len(array))
elif coffset == 0 and cols == 1:
array = array[roffset:]
rows = min(rows, len(array))
xml += "<arraydata rows=\"%s\" cols=\"%s\"/>" % (rows, cols)
for row in xrange(rows):
xml += "<row index=\"%s\"/>" % to_string(row)
for col in xrange(cols):
value = array
if rows == 1 or cols == 1:
if rows == 1 and cols == 1:
value = array[0]
else:
if rows == 1:
dim = col
else:
dim = row
value = array[dim]
if "ndarray" in str(type(value)):
value = value[0]
else:
value = array[row][col]
value = format % value
xml += var_to_xml(value, '')
return xml
def array_to_xml(array, roffset, coffset, rows, cols, format):
xml = ""
rows = min(rows, MAXIMUM_ARRAY_SIZE)
cols = min(cols, MAXIMUM_ARRAY_SIZE)
# there is no obvious rule for slicing (at least 5 choices)
if len(array) == 1 and (rows > 1 or cols > 1):
array = array[0]
if array.size > len(array):
array = array[roffset:, coffset:]
rows = min(rows, len(array))
cols = min(cols, len(array[0]))
if len(array) == 1:
array = array[0]
elif array.size == len(array):
if roffset == 0 and rows == 1:
array = array[coffset:]
cols = min(cols, len(array))
elif coffset == 0 and cols == 1:
array = array[roffset:]
rows = min(rows, len(array))
xml += "<arraydata rows=\"%s\" cols=\"%s\"/>" % (rows, cols)
for row in xrange(rows):
xml += "<row index=\"%s\"/>" % to_string(row)
for col in xrange(cols):
value = array
if rows == 1 or cols == 1:
if rows == 1 and cols == 1:
value = array[0]
else:
if rows == 1:
dim = col
else:
dim = row
value = array[dim]
if "ndarray" in str(type(value)):
value = value[0]
else:
value = array[row][col]
value = format % value
xml += var_to_xml(value, '')
return xml
def set_args(self, frame, recursive=False):
self.args = []
code = frame.f_code
locals = frame.f_locals
for i in xrange(0, code.co_argcount):
name = code.co_varnames[i]
class_name = get_type_of_value(locals[name], recursive=recursive)
self.add_arg(name, class_name)
def set_args(self, frame, recursive=False):
self.args = []
code = frame.f_code
locals = frame.f_locals
for i in xrange(0, code.co_argcount):
name = code.co_varnames[i]
class_name = get_type_of_value(locals[name], recursive=recursive)
self.add_arg(name, class_name)
def get_dictionary(self, obj):
ret = {}
declaredFields = obj.__class__.getDeclaredFields()
for i in xrange(len(declaredFields)):
name = declaredFields[i].getName()
try:
declaredFields[i].setAccessible(True)
ret[name] = declaredFields[i].get(obj)
except:
traceback.print_exc()
return ret
def get_dictionary(self, obj):
ret = {}
declaredFields = obj.__class__.getDeclaredFields()
for i in xrange(len(declaredFields)):
name = declaredFields[i].getName()
try:
declaredFields[i].setAccessible(True)
ret[name] = declaredFields[i].get(obj)
except:
traceback.print_exc()
return ret
def do_find(f, mod):
import linecache
if inspect.ismodule(mod):
return f, 0, 0
lines = linecache.getlines(f)
if inspect.isclass(mod):
name = mod.__name__
pat = re.compile(r'^\s*class\s*' + name + r'\b')
for i in xrange(len(lines)):
if pat.match(lines[i]):
return f, i, 0
return f, 0, 0
if inspect.ismethod(mod):
mod = mod.im_func
if inspect.isfunction(mod):
try:
mod = mod.func_code
except AttributeError:
mod = mod.__code__ #python 3k
if inspect.istraceback(mod):
mod = mod.tb_frame
if inspect.isframe(mod):
mod = mod.f_code
if inspect.iscode(mod):
if not hasattr(mod, 'co_filename'):
return None, 0, 0
if not hasattr(mod, 'co_firstlineno'):
return mod.co_filename, 0, 0
lnum = mod.co_firstlineno
pat = re.compile(r'^(\s*def\s)|(.*(?<!\w)lambda(:|\s))|^(\s*@)')
while lnum > 0:
if pat.match(lines[lnum]):
break
lnum -= 1
return f, lnum, 0
raise RuntimeError('Do not know about: ' + f + ' ' + str(mod))
def do_find(f, mod):
import linecache
if inspect.ismodule(mod):
return f, 0, 0
lines = linecache.getlines(f)
if inspect.isclass(mod):
name = mod.__name__
pat = re.compile(r'^\s*class\s*' + name + r'\b')
for i in xrange(len(lines)):
if pat.match(lines[i]):
return f, i, 0
return f, 0, 0
if inspect.ismethod(mod):
mod = mod.im_func
if inspect.isfunction(mod):
try:
mod = mod.func_code
except AttributeError:
mod = mod.__code__ #python 3k
if inspect.istraceback(mod):
mod = mod.tb_frame
if inspect.isframe(mod):
mod = mod.f_code
if inspect.iscode(mod):
if not hasattr(mod, 'co_filename'):
return None, 0, 0
if not hasattr(mod, 'co_firstlineno'):
return mod.co_filename, 0, 0
lnum = mod.co_firstlineno
pat = re.compile(r'^(\s*def\s)|(.*(?<!\w)lambda(:|\s))|^(\s*@)')
while lnum > 0:
if pat.match(lines[lnum]):
break
lnum -= 1
return f, lnum, 0
raise RuntimeError('Do not know about: ' + f + ' ' + str(mod))
def create_signature(self, frame, with_args=True):
try:
code = frame.f_code
locals = frame.f_locals
filename, modulename, funcname = self.file_module_function_of(
frame)
res = Signature(filename, funcname)
if with_args:
for i in xrange(0, code.co_argcount):
name = code.co_varnames[i]
class_name = get_type_of_value(locals[name])
res.add_arg(name, class_name)
return res
except:
import traceback
traceback.print_exc()
def create_signature(self, frame):
try:
code = frame.f_code
locals = frame.f_locals
filename, modulename, funcname = self.file_module_function_of(frame)
res = Signature(filename, funcname)
for i in xrange(0, code.co_argcount):
name = code.co_varnames[i]
tp = type(locals[name])
class_name = tp.__name__
if class_name == 'instance': # old-style classes
tp = locals[name].__class__
class_name = tp.__name__
if hasattr(tp, '__module__') and tp.__module__ and tp.__module__ not in self._ignore_module_name:
class_name = "%s.%s"%(tp.__module__, class_name)
res.add_arg(name, class_name)
return res
except:
import traceback
traceback.print_exc()
def getVariable(thread_id, frame_id, scope, attrs):
"""
returns the value of a variable
:scope: can be BY_ID, EXPRESSION, GLOBAL, LOCAL, FRAME
BY_ID means we'll traverse the list of all objects alive to get the object.
:attrs: after reaching the proper scope, we have to get the attributes until we find
the proper location (i.e.: obj\tattr1\tattr2).
:note: when BY_ID is used, the frame_id is considered the id of the object to find and
not the frame (as we don't care about the frame in this case).
"""
if scope == 'BY_ID':
if thread_id != get_thread_id(threading.currentThread()):
raise VariableError("getVariable: must execute on same thread")
try:
import gc
objects = gc.get_objects()
except:
pass # Not all python variants have it.
else:
frame_id = int(frame_id)
for var in objects:
if id(var) == frame_id:
if attrs is not None:
attrList = attrs.split('\t')
for k in attrList:
_type, _typeName, resolver = get_type(var)
var = resolver.resolve(var, k)
return var
# If it didn't return previously, we coudn't find it by id (i.e.: alrceady garbage collected).
sys.stderr.write('Unable to find object with id: %s\n' % (frame_id, ))
return None
frame = find_frame(thread_id, frame_id)
if frame is None:
return {}
if attrs is not None:
attrList = attrs.split('\t')
else:
attrList = []
for attr in attrList:
attr.replace("@_@TAB_CHAR@_@", '\t')
if scope == 'EXPRESSION':
for count in xrange(len(attrList)):
if count == 0:
# An Expression can be in any scope (globals/locals), therefore it needs to evaluated as an expression
var = evaluate_expression(thread_id, frame_id, attrList[count],
False)
else:
_type, _typeName, resolver = get_type(var)
var = resolver.resolve(var, attrList[count])
else:
if scope == "GLOBAL":
var = frame.f_globals
del attrList[
0] # globals are special, and they get a single dummy unused attribute
else:
# in a frame access both locals and globals as Python does
var = {}
var.update(frame.f_globals)
var.update(frame.f_locals)
for k in attrList:
_type, _typeName, resolver = get_type(var)
var = resolver.resolve(var, k)
return var
def getVariable(thread_id, frame_id, scope, attrs):
"""
returns the value of a variable
:scope: can be BY_ID, EXPRESSION, GLOBAL, LOCAL, FRAME
BY_ID means we'll traverse the list of all objects alive to get the object.
:attrs: after reaching the proper scope, we have to get the attributes until we find
the proper location (i.e.: obj\tattr1\tattr2)
:note: when BY_ID is used, the frame_id is considered the id of the object to find and
not the frame (as we don't care about the frame in this case).
"""
if scope == 'BY_ID':
if thread_id != get_thread_id(threading.currentThread()):
raise VariableError("getVariable: must execute on same thread")
try:
import gc
objects = gc.get_objects()
except:
pass # Not all python variants have it.
else:
frame_id = int(frame_id)
for var in objects:
if id(var) == frame_id:
if attrs is not None:
attrList = attrs.split('\t')
for k in attrList:
_type, _typeName, resolver = get_type(var)
var = resolver.resolve(var, k)
return var
# If it didn't return previously, we coudn't find it by id (i.e.: alrceady garbage collected).
sys.stderr.write('Unable to find object with id: %s\n' % (frame_id,))
return None
frame = find_frame(thread_id, frame_id)
if frame is None:
return {}
if attrs is not None:
attrList = attrs.split('\t')
else:
attrList = []
for attr in attrList:
attr.replace("@_@TAB_CHAR@_@", '\t')
if scope == 'EXPRESSION':
for count in xrange(len(attrList)):
if count == 0:
# An Expression can be in any scope (globals/locals), therefore it needs to evaluated as an expression
var = evaluate_expression(thread_id, frame_id, attrList[count], False)
else:
_type, _typeName, resolver = get_type(var)
var = resolver.resolve(var, attrList[count])
else:
if scope == "GLOBAL":
var = frame.f_globals
del attrList[0] # globals are special, and they get a single dummy unused attribute
else:
# in a frame access both locals and globals as Python does
var = {}
var.update(frame.f_globals)
var.update(frame.f_locals)
for k in attrList:
_type, _typeName, resolver = get_type(var)
var = resolver.resolve(var, k)
return var
def dataframe_to_xml(df, name, roffset, coffset, rows, cols, format):
"""
:type df: pandas.core.frame.DataFrame
:type name: str
:type coffset: int
:type roffset: int
:type rows: int
:type cols: int
:type format: str
"""
num_rows = min(df.shape[0], MAX_SLICE_SIZE)
num_cols = min(df.shape[1], MAX_SLICE_SIZE)
if (num_rows, num_cols) != df.shape:
df = df.iloc[0:num_rows, 0:num_cols]
slice = '.iloc[0:%s, 0:%s]' % (num_rows, num_cols)
else:
slice = ''
slice = name + slice
xml = '<array slice=\"%s\" rows=\"%s\" cols=\"%s\" format=\"\" type=\"\" max=\"0\" min=\"0\"/>\n' % \
(slice, num_rows, num_cols)
if (rows, cols) == (-1, -1):
rows, cols = num_rows, num_cols
rows = min(rows, MAXIMUM_ARRAY_SIZE)
cols = min(min(cols, MAXIMUM_ARRAY_SIZE), num_cols)
# need to precompute column bounds here before slicing!
col_bounds = [None] * cols
for col in xrange(cols):
dtype = df.dtypes.iloc[coffset + col].kind
if dtype in "biufc":
cvalues = df.iloc[:, coffset + col]
bounds = (cvalues.min(), cvalues.max())
else:
bounds = (0, 0)
col_bounds[col] = bounds
df = df.iloc[roffset:roffset + rows, coffset:coffset + cols]
rows, cols = df.shape
xml += "<headerdata rows=\"%s\" cols=\"%s\">\n" % (rows, cols)
format = format.replace('%', '')
col_formats = []
get_label = lambda label: str(label) if not isinstance(
label, tuple) else '/'.join(map(str, label))
for col in xrange(cols):
dtype = df.dtypes.iloc[col].kind
if dtype == 'f' and format:
fmt = format
elif dtype == 'f':
fmt = '.5f'
elif dtype == 'i' or dtype == 'u':
fmt = 'd'
else:
fmt = 's'
col_formats.append('%' + fmt)
bounds = col_bounds[col]
xml += '<colheader index=\"%s\" label=\"%s\" type=\"%s\" format=\"%s\" max=\"%s\" min=\"%s\" />\n' % \
(str(col), get_label(df.axes[1].values[col]), dtype, fmt, bounds[1], bounds[0])
for row, label in enumerate(iter(df.axes[0])):
xml += "<rowheader index=\"%s\" label = \"%s\"/>\n" % \
(str(row), get_label(label))
xml += "</headerdata>\n"
xml += "<arraydata rows=\"%s\" cols=\"%s\"/>\n" % (rows, cols)
for row in xrange(rows):
xml += "<row index=\"%s\"/>\n" % str(row)
for col in xrange(cols):
value = df.iat[row, col]
value = col_formats[col] % value
xml += var_to_xml(value, '')
return xml
def dataframe_to_xml(df, name, roffset, coffset, rows, cols, format):
"""
:type df: pandas.core.frame.DataFrame
:type name: str
:type coffset: int
:type roffset: int
:type rows: int
:type cols: int
:type format: str
"""
num_rows = min(df.shape[0], MAX_SLICE_SIZE)
num_cols = min(df.shape[1], MAX_SLICE_SIZE)
if (num_rows, num_cols) != df.shape:
df = df.iloc[0:num_rows, 0: num_cols]
slice = '.iloc[0:%s, 0:%s]' % (num_rows, num_cols)
else:
slice = ''
slice = name + slice
xml = '<array slice=\"%s\" rows=\"%s\" cols=\"%s\" format=\"\" type=\"\" max=\"0\" min=\"0\"/>\n' % \
(slice, num_rows, num_cols)
if (rows, cols) == (-1, -1):
rows, cols = num_rows, num_cols
rows = min(rows, MAXIMUM_ARRAY_SIZE)
cols = min(min(cols, MAXIMUM_ARRAY_SIZE), num_cols)
# need to precompute column bounds here before slicing!
col_bounds = [None] * cols
for col in xrange(cols):
dtype = df.dtypes.iloc[coffset + col].kind
if dtype in "biufc":
cvalues = df.iloc[:, coffset + col]
bounds = (cvalues.min(), cvalues.max())
else:
bounds = (0, 0)
col_bounds[col] = bounds
df = df.iloc[roffset: roffset + rows, coffset: coffset + cols]
rows, cols = df.shape
xml += "<headerdata rows=\"%s\" cols=\"%s\">\n" % (rows, cols)
format = format.replace('%', '')
col_formats = []
get_label = lambda label: str(label) if not isinstance(label, tuple) else '/'.join(map(str, label))
for col in xrange(cols):
dtype = df.dtypes.iloc[col].kind
if dtype == 'f' and format:
fmt = format
elif dtype == 'f':
fmt = '.5f'
elif dtype == 'i' or dtype == 'u':
fmt= 'd'
else:
fmt= 's'
col_formats.append('%' + fmt)
bounds = col_bounds[col]
xml += '<colheader index=\"%s\" label=\"%s\" type=\"%s\" format=\"%s\" max=\"%s\" min=\"%s\" />\n' % \
(str(col), get_label(df.axes[1].values[col]), dtype, fmt, bounds[1], bounds[0])
for row, label in enumerate(iter(df.axes[0])):
xml += "<rowheader index=\"%s\" label = \"%s\"/>\n" % \
(str(row), get_label(label))
xml += "</headerdata>\n"
xml += "<arraydata rows=\"%s\" cols=\"%s\"/>\n" % (rows, cols)
for row in xrange(rows):
xml += "<row index=\"%s\"/>\n" % str(row)
for col in xrange(cols):
value = df.iat[row, col]
value = col_formats[col] % value
xml += var_to_xml(value, '')
return xml
