1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419
|
# begin.py
# Copyright (C) 2008 Bernhard Lohkamp, The University of York
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
print("===============================================================")
print("==================== Testing ==================================")
print("===============================================================")
import coot
import sys
import unittest, os
import inspect
global have_test_skip
global skipped_tests
have_test_skip = False
skipped_tests = []
if ('skip' in dir(unittest.TestCase)):
have_test_skip = True
else:
print("WARNING:: unittest skip not avaliable!!!!!!")
home = os.getenv('HOME')
if ((not home) and (os.name == 'nt')):
home = os.getenv('COOT_HOME')
if (not home):
# badness we dont have a home dir
print("ERROR:: Cannot find a HOME directory")
def get_unittest_data_dir():
d = os.getenv("COOT_TEST_DATA_DIR")
if (not d):
d = os.path.normpath(os.path.join(home, "data", "greg-data"))
return d
global unittest_data_dir
unittest_data_dir = get_unittest_data_dir() # too lazy to convert all
# occurences of unittest_data_dir
################
# some functions
################
def unittest_pdb(file_name):
ret = read_pdb(os.path.join(unittest_data_dir, file_name))
return ret
# use functions rather than variables which would have to be globals
def rnase_pdb():
return os.path.join(unittest_data_dir, "tutorial-modern.pdb")
def rnase_mtz():
return os.path.join(unittest_data_dir, "rnasa-1.8-all_refmac1.mtz")
def rotate_n_frames(n):
rotate_speed = 1
return int(rotate_speed * n)
# whatever this shall do
#def chains_in_order_qm(ls):
def bond_length(pos_1, pos_2):
def square(x):
return x*x
def sub(x, y):
return x-y
import math
ret = math.sqrt(sum(map(square, list(map(sub, pos_1, pos_2)))))
return ret
pos_diff = bond_length
def bond_length_from_atoms(atom_1, atom_2):
from types import ListType
if (type(atom_1) is not ListType):
print(" WARNING:: bond_length_from_atoms: atom_1 not a list:", atom_1)
return False
elif (type(atom_2) is not ListType):
print(" WARNING:: bond_length_from_atoms: atom_2 not a list:", atom_2)
return False
else:
return bond_length(atom_1[2],
atom_2[2])
def bond_length_within_tolerance_qm(atom_1, atom_2, ideal_length, tolerance):
# BL says:: this should (as in greg) probably throw an error, but we dont have the
# test object here, so cannot. Buh. Should pass self?!
if (not atom_1):
return False
if (not atom_2):
return False
b = bond_length_from_atoms(atom_1, atom_2)
return abs(b - ideal_length) < tolerance
def shelx_waters_all_good_occ_qm(test, imol_insulin_res):
chain_id = water_chain(imol_insulin_res)
n_residues = chain_n_residues(chain_id, imol_insulin_res)
serial_number = n_residues - 1
res_name = resname_from_serial_number(imol_insulin_res, chain_id, serial_number)
res_no = seqnum_from_serial_number (imol_insulin_res, chain_id, serial_number)
ins_code = insertion_code_from_serial_number(imol_insulin_res, chain_id, serial_number)
if (res_name == "HOH"):
atom_list = residue_info(imol_insulin_res, chain_id, res_no, ins_code)
for atom in atom_list:
occ = atom[1][0]
test.assertAlmostEqual(occ, 11.0, 1, " bad occupancy in SHELXL molecule %s" %atom)
# return restraints without the given bond restraints or
# False if no restraints given
def strip_bond_from_restraints(atom_pair, restraints):
import copy
restr_dict = copy.deepcopy(restraints)
if (restr_dict == []):
return []
else:
if not restr_dict:
return False
else:
# make a copy of restraints, so that we dont overwrite
# the original dictionary
bonds = restr_dict["_chem_comp_bond"]
if bonds:
for i, bond in enumerate(bonds):
atom1 = bond[0]
atom2 = bond[1]
if (atom1 in atom_pair and atom2 in atom_pair):
del restr_dict["_chem_comp_bond"][i]
return restr_dict
return restr_dict
# are the attributes of atom-1 the same as atom-2? (given we test for
# failUnless/IfAlmostEqual)
#
def atoms_match_qm(atom_1, atom_2):
from types import ListType, StringType, FloatType, IntType, BooleanType
if ((not atom_1) or (not atom_2)):
return False # no matching residues
if (len(atom_1) != len(atom_2)):
return False # comparing differnt list sizes -> not equal anyway
for i in range(len(atom_1)):
if (type(atom_1[i]) is ListType):
if (atoms_match_qm(atom_1[i], atom_2[i])):
pass
else:
return False
elif (type(atom_1[i]) is StringType):
if (atom_1[i] == atom_2[i]):
pass
else:
return False
elif (type(atom_1[i]) is FloatType):
if (abs(atom_1[i] - atom_2[i]) < 0.01):
pass
else:
return False
elif (type(atom_1[i]) is IntType):
if (atom_1[i] == atom_2[i]):
pass
else:
return False
elif (type(atom_1[i]) is BooleanType):
if (atom_1[i] == atom_2[i]):
pass
else:
return False
return True
# transform
# Just a (eg 3x3) (no vector) matrix
# The matrix does not have to be symmetric.
#
def transpose_mat(mat, defval=None):
if not mat:
return []
return list(map(lambda *row: [elem or defval for elem in row], *mat))
# What is the distance atom-1 to atom-2?
# return False on not able to calculate
def atom_distance(atom_1, atom_2):
def square(x):
return x*x
def sub(x, y):
return x-y
import math
ret = math.sqrt(sum(map(square, list(map(sub, atom_1[2], atom_2[2])))))
return ret
# a function from 04_cootaneering:
#
# note that things could go wrong if there is a mising EOL (not tested)
# not sure about this in python
#
def file2string(rnase_pir):
fin = open(rnase_pir, 'r')
s = fin.read()
fin.close()
return s
def atoms_have_correct_seg_id_qm(atoms, seg_id):
for atom in atoms:
atom_seg_id = atom[1][3]
if (not atom_seg_id == seg_id):
return False
return True
# return the residue name given a 4 element spec
#
def residue_name_from_spec(imol, spec):
if len(spec) == 4:
return residue_name(imol, *spec[1:4])
return False
# return residue specs of all residues that are type residue-type
#
def get_residues_in_molecule_of_type(imol, residue_type):
return [x for x in fit_protein_make_specs(imol, 'all-chains') if residue_name_from_spec(imol, x) == residue_type]
# This takes 4 member specs, return True or False
#
def spec_match_qm(spec_1, spec_2):
from types import StringType
from types import IntType
# first test if specs are of length 4.
#
if not (len(spec_1) == 4):
return False
if not (len(spec_2) == 4):
return False
chain_id_1 = spec_1[1]
chain_id_2 = spec_2[1]
res_no_1 = spec_1[2]
res_no_2 = spec_2[2]
ins_code_1 = spec_1[3]
ins_code_2 = spec_2[3]
if not (type(chain_id_1) is StringType):
return False
if not (chain_id_1 == chain_id_2):
return False
if not (type(res_no_1) is IntType):
return False
if not (res_no_1 == res_no_2):
return False
if not (type(ins_code_1) is StringType):
return False
if not (ins_code_1 == ins_code_2):
return False
return True
coot.set_console_display_commands_hilights(0, 0, 0)
#################################
# SOME MORE FUNCTIONS ONLY PYTHON
#################################
def skip_test(self, skip_query, skip_msg):
global have_test_skip
global skipped_tests
if have_test_skip:
self.skipIf(skip_query, skip_msg)
else:
if skip_query:
skipped_tests.append(self.shortDescription())
# no way so far to exit the test from here
# either work with return value or Exceptions
# former for now
return True
unittest.TestCase.skip_test = skip_test
###############################
# NOW DEFINE THE ACTUAL TESTS
###############################
def get_this_dir():
pass
test_file_list = ["01_pdb_mtz.py",
"02_shelx.py",
"03_ligand.py",
"04_cootaneer.py",
"05_rna_ghosts.py",
"06_ssm.py",
"07_ncs.py",
"08_utils.py",
"09_internal.py"]
# get directory of this file and execute tests found in this dir
fn = inspect.getfile(get_this_dir)
current_dir = os.path.dirname(fn)
for test_file in test_file_list:
load_file = os.path.join(current_dir, test_file)
load_file = os.path.normpath(load_file)
if (os.path.isfile(load_file)):
exec(compile(open(load_file, "rb").read(), load_file, 'exec'), globals())
test_list = [PdbMtzTestFunctions, ShelxTestFunctions,
LigandTestFunctions, CootaneerTestFunctions,
RnaGhostsTestFunctions, SsmTestFunctions,
NcsTestFunctions, UtilTestFunctions,
InternalTestFunctions]
suite = unittest.TestSuite()
for test in test_list:
suite.addTests(unittest.TestLoader().loadTestsFromTestCase(test))
# returns a list of all tests, in pairs containing the test class and description
# i.e. [[__main__.PdbMtzTestFunctions.test00_0, "Post Go To Atom no molecule"],
# [__main__.PdbMtzTestFunctions.test01_0, "short discr Close bad molecule"]]
def list_of_all_tests():
ret =[]
for test in suite:
ret.append([test, test.shortDescription(), test.id()])
return ret
global unittest_output
unittest_output = False
# class to write output of unittest into a 'memory file' (unittest_output)
# as well as to sys.stdout
class StreamIO:
def __init__(self, etxra, src=sys.stderr, dst=sys.stdout):
import io
global unittest_output
unittest_output = io.StringIO()
self.src = src
self.dst = dst
self.extra = unittest_output
def write(self, msg):
#self.src.write(msg)
self.extra.write(msg)
self.dst.write(msg)
def flush(self):
pass
class StreamIOnew:
def __init__(self, etxra, src=sys.stderr, dst=sys.stdout):
import io
import io
global unittest_output
unittest_output = io.StringIO()
# unittest_output = io.StringIO()
self.src = src
self.dst = dst
self.extra = unittest_output
def write(self, msg):
#msg = msg.decode()
self.extra.write(msg)
self.dst.write(msg)
def flush(self):
pass
# function to run one test
def run_one_test(no_of_test):
result = unittest.TextTestRunner(verbosity=2).run(list_of_all_tests()[no_of_test][0])
return result
# function to run one test set
def run_test_set(no_of_test_set):
print("BL DEBUG:: should run ", test_list[no_of_test_set])
if (no_of_test_set > 0):
read_imol_rnase_and_map()
set = unittest.TestSuite()
set.addTests(unittest.TestLoader().loadTestsFromTestCase(test_list[no_of_test_set]))
result = unittest.TextTestRunner(verbosity=2).run(set)
return result
# read imol_rnase and map if not there?!
def read_imol_rnase_and_map():
global imol_rnase
global imol_rnase_map
if imol_rnase < 0:
# not set up-> read, i.e. run test02 from 01
imol = read_pdb(rnase_pdb())
imol_rnase = imol
if imol_rnase_map < 0:
imol_map = make_and_draw_map(rnase_mtz(), "FWT","PHWT","",0,0)
set_imol_refinement_map(imol_map)
imol_rnase_map = imol_map
|