def read_dump_file(dump_file, cfg, data_to_print, gofr_data, out_fieldnames,
write_mode, evb_dict):
with open(dump_file) as d:
# spaces here allow file name to line up with the "completed reading" print line
if cfg[PRINT_PROGRESS]:
print("{:>17}: {}".format('Reading', dump_file))
section = None
box = np.zeros((3, ))
box_counter = 1
atom_counter = 1
timesteps_read = 0
num_atoms = 0
timestep = None
for line in d:
line = line.strip()
if section is None:
section = find_dump_section_state(line)
if section is None and len(line) > 0:
raise InvalidDataError(
'Unexpected line in file {}: {}'.format(
dump_file, line))
elif section == SEC_TIMESTEP:
# Reset variables
section = None
dump_atom_data = []
try:
timestep = int(line)
except ValueError as e:
raise InvalidDataError(
"In attempting to read an integer timestep, encountered error: {}"
.format(e))
timesteps_read += 1
if timesteps_read > cfg[MAX_TIMESTEPS]:
print(
"Reached the maximum timesteps per dumpfile ({}). "
"To increase this number, set a larger value for {}. "
"Continuing program.".format(cfg[MAX_TIMESTEPS],
MAX_TIMESTEPS))
break
if timesteps_read % cfg[PRINT_TIMESTEPS] == 0:
if cfg[PER_FRAME_OUTPUT]:
print_per_frame(dump_file, cfg, data_to_print,
out_fieldnames, write_mode)
data_to_print = []
write_mode = 'a'
if cfg[GOFR_OUTPUT]:
print_gofr(cfg, gofr_data)
result = {
FILE_NAME: os.path.basename(dump_file),
TIMESTEP: timestep
}
if cfg[EVB_SUM_FILE] is not None:
if cfg[ALIGN_COL] == FILE_NAME:
align_val = os.path.splitext(
result[cfg[ALIGN_COL]])[0] + cfg[EVB_FILE_EXT]
else:
align_val = result[cfg[ALIGN_COL]]
if align_val in evb_dict:
step_dict = evb_dict[align_val]
for evb_header in cfg[EVB_SUM_HEADERS]:
result[evb_header] = step_dict[evb_header]
if cfg[CALC_CEC_DIST]:
result[CEC_XYZ] = np.asarray([
step_dict[CEC_X], step_dict[CEC_Y],
step_dict[CEC_Z]
])
else:
warning(
"Did not find '{}' value {} in the data read from: {}"
.format(cfg[ALIGN_COL], result[cfg[ALIGN_COL]],
cfg[EVB_SUM_FILE]))
elif section == SEC_NUM_ATOMS:
num_atoms = int(line)
section = None
elif section == SEC_BOX_SIZE:
split_line = line.split()
diff = float(split_line[1]) - float(split_line[0])
box[box_counter - 1] = diff
if box_counter == 3:
box_counter = 0
section = None
box_counter += 1
elif section == SEC_ATOMS:
split_line = line.split()
# If there is an incomplete line in a dump file, move on to the next file
if len(split_line) < 7:
break
atom_num = int(split_line[0])
mol_num = int(split_line[1])
atom_type = int(split_line[2])
charge = float(split_line[3])
x, y, z = map(float, split_line[4:7])
# Here, the atoms counting starts at 1. However, the template counted from zero
atom_struct = {
ATOM_NUM: atom_num,
MOL_NUM: mol_num,
ATOM_TYPE: atom_type,
CHARGE: charge,
XYZ_COORDS: [x, y, z],
}
dump_atom_data.append(atom_struct)
if atom_counter == num_atoms:
if len(cfg[ONLY_STEPS]
) == 0 or timestep in cfg[ONLY_STEPS]:
result.update(
process_atom_data(cfg, dump_atom_data, box,
timestep, gofr_data, result))
data_to_print.append(result)
atom_counter = 0
section = None
atom_counter += 1
if atom_counter == 1:
if cfg[PRINT_PROGRESS]:
print("Completed reading: {}".format(dump_file))
else:
warning(
"FYI: dump file {} step {} did not have the full list of atom numbers. "
"Continuing to next dump file.".format(dump_file, timestep))
def process_dump_file(cfg, dump_file, atom_num_dict, atom_type_dict,
mol_num_dict):
section = None
box = np.zeros((3, ))
counter = 1
num_atoms = 0
head_content = []
steps_count = 0
step_stop = cfg[MAX_STEPS] * cfg[OUT_FREQ]
timestep = None
with open(dump_file) as d:
d_out = create_out_fname(dump_file,
suffix='_reorder',
base_dir=cfg[OUT_BASE_DIR])
write_mode = 'w'
for line in d:
line = line.strip()
if section == SEC_ATOMS:
split_line = line.split()
# If there is an incomplete line in a dump file, move on to the next file
if len(split_line) < 7:
break
atom_num = int(split_line[0])
if atom_num in atom_num_dict:
atom_num = atom_num_dict[atom_num]
mol_num = int(split_line[1])
if mol_num in mol_num_dict:
mol_num = mol_num_dict[mol_num]
# Default RENUM_START_MOL is neg 1; if still less than zero, user did not specify renumbering
if 0 <= cfg[RENUM_START_MOL] <= mol_num:
mol_num += cfg[RENUM_SHIFT]
atom_type = int(split_line[2])
if atom_type in atom_type_dict:
atom_type = atom_type_dict[atom_type]
charge = float(split_line[3])
x, y, z = map(float, split_line[4:7])
atom_struct = [atom_num, mol_num, atom_type, charge, x, y, z]
atom_data.append(atom_struct)
if counter == num_atoms:
if len(atom_num_dict) > 0:
atom_data = sorted(atom_data, key=lambda atom: atom[0])
steps_count += 1
if steps_count % cfg[OUT_FREQ] == 0:
print_to_dump_file(head_content,
atom_data,
d_out,
mode=write_mode)
if write_mode == 'w':
write_mode = 'a'
if steps_count == step_stop:
print(
"Reached the maximum number of steps ({})".format(
cfg[MAX_STEPS]))
counter = 1
break
# reset for next timestep
head_content = []
counter = 0
section = None
counter += 1
else:
head_content.append(line)
if section is None:
section = find_dump_section_state(line)
if section is None:
raise InvalidDataError(
'Unexpected line in file {}: {}'.format(d, line))
elif section == SEC_TIMESTEP:
timestep = line
# Reset variables
atom_data = []
section = None
elif section == SEC_NUM_ATOMS:
num_atoms = int(line)
section = None
elif section == SEC_BOX_SIZE:
split_line = line.split()
diff = float(split_line[1]) - float(split_line[0])
box[counter - 1] = diff
if counter == 3:
counter = 0
section = None
counter += 1
if counter == 1:
print("Completed reading: {}".format(dump_file))
else:
warning(
"Dump file {} step {} did not have the full list of atom numbers. "
"Continuing program.".format(dump_file, timestep))
def process_dump_file(cfg, dump_file, atom_num_dict, atom_type_dict, mol_num_dict):
section = None
box = np.zeros((3,))
counter = 1
num_atoms = 0
head_content = []
steps_count = 0
step_stop = cfg[MAX_STEPS] * cfg[OUT_FREQ]
timestep = None
with open(dump_file) as d:
d_out = create_out_fname(dump_file, suffix='_reorder', base_dir=cfg[OUT_BASE_DIR])
write_mode = 'w'
for line in d:
line = line.strip()
if section == SEC_ATOMS:
split_line = line.split()
# If there is an incomplete line in a dump file, move on to the next file
if len(split_line) < 7:
break
atom_num = int(split_line[0])
if atom_num in atom_num_dict:
atom_num = atom_num_dict[atom_num]
mol_num = int(split_line[1])
if mol_num in mol_num_dict:
mol_num = mol_num_dict[mol_num]
# Default RENUM_START_MOL is neg 1; if still less than zero, user did not specify renumbering
if 0 <= cfg[RENUM_START_MOL] <= mol_num:
mol_num += cfg[RENUM_SHIFT]
atom_type = int(split_line[2])
if atom_type in atom_type_dict:
atom_type = atom_type_dict[atom_type]
charge = float(split_line[3])
x, y, z = map(float, split_line[4:7])
atom_struct = [atom_num, mol_num, atom_type, charge, x, y, z]
atom_data.append(atom_struct)
if counter == num_atoms:
if len(atom_num_dict) > 0:
atom_data = sorted(atom_data, key=lambda atom: atom[0])
steps_count += 1
if steps_count % cfg[OUT_FREQ] == 0:
print_to_dump_file(head_content, atom_data, d_out, mode=write_mode)
if write_mode == 'w':
write_mode = 'a'
if steps_count == step_stop:
print("Reached the maximum number of steps ({})".format(cfg[MAX_STEPS]))
counter = 1
break
# reset for next timestep
head_content = []
counter = 0
section = None
counter += 1
else:
head_content.append(line)
if section is None:
section = find_dump_section_state(line)
if section is None:
raise InvalidDataError('Unexpected line in file {}: {}'.format(d, line))
elif section == SEC_TIMESTEP:
timestep = line
# Reset variables
atom_data = []
section = None
elif section == SEC_NUM_ATOMS:
num_atoms = int(line)
section = None
elif section == SEC_BOX_SIZE:
split_line = line.split()
diff = float(split_line[1]) - float(split_line[0])
box[counter - 1] = diff
if counter == 3:
counter = 0
section = None
counter += 1
if counter == 1:
print("Completed reading: {}".format(dump_file))
else:
warning("Dump file {} step {} did not have the full list of atom numbers. "
"Continuing program.".format(dump_file, timestep))
def read_dump_file(dump_file, cfg, data_to_print, gofr_data, out_fieldnames, write_mode):
with open(dump_file) as d:
# spaces here allow file name to line up with the "completed reading" print line
print("{:>17}: {}".format("Reading", dump_file))
section = None
box = np.zeros((3,))
box_counter = 1
atom_counter = 1
timesteps_read = 0
num_atoms = 0
timestep = None
for line in d:
line = line.strip()
if section is None:
section = find_dump_section_state(line)
if section is None and len(line) > 0:
raise InvalidDataError("Unexpected line in file {}: {}".format(dump_file, line))
elif section == SEC_TIMESTEP:
# Reset variables
section = None
dump_atom_data = []
try:
timestep = int(line)
except ValueError as e:
raise InvalidDataError("In attempting to read an integer timestep, encountered error: {}".format(e))
timesteps_read += 1
if timesteps_read > cfg[MAX_TIMESTEPS]:
print(
"Reached the maximum timesteps per dumpfile ({}). "
"To increase this number, set a larger value for {}. "
"Continuing program.".format(cfg[MAX_TIMESTEPS], MAX_TIMESTEPS)
)
break
if timesteps_read % cfg[PRINT_TIMESTEPS] == 0:
if cfg[PER_FRAME_OUTPUT]:
print_per_frame(dump_file, cfg, data_to_print, out_fieldnames, write_mode)
data_to_print = []
write_mode = "a"
if cfg[GOFR_OUTPUT]:
print_gofr(cfg, gofr_data)
result = {TIMESTEP: timestep}
elif section == SEC_NUM_ATOMS:
num_atoms = int(line)
section = None
elif section == SEC_BOX_SIZE:
split_line = line.split()
diff = float(split_line[1]) - float(split_line[0])
box[box_counter - 1] = diff
if box_counter == 3:
box_counter = 0
section = None
box_counter += 1
elif section == SEC_ATOMS:
split_line = line.split()
# If there is an incomplete line in a dump file, move on to the next file
if len(split_line) < 7:
break
atom_num = int(split_line[0])
mol_num = int(split_line[1])
atom_type = int(split_line[2])
charge = float(split_line[3])
x, y, z = map(float, split_line[4:7])
# Here, the atoms counting starts at 1. However, the template counted from zero
atom_struct = {
ATOM_NUM: atom_num,
MOL_NUM: mol_num,
ATOM_TYPE: atom_type,
CHARGE: charge,
XYZ_COORDS: [x, y, z],
}
dump_atom_data.append(atom_struct)
if atom_counter == num_atoms:
result.update(process_atom_data(cfg, dump_atom_data, box, timestep, gofr_data))
data_to_print.append(result)
atom_counter = 0
section = None
atom_counter += 1
if atom_counter == 1:
print("Completed reading: {}".format(dump_file))
else:
warning(
"FYI: dump file {} step {} did not have the full list of atom numbers. "
"Continuing to next dump file.".format(dump_file, timestep)
)
def process_dump_file(cfg, data_tpl_content, dump_file):
section = None
box = np.zeros((3,))
counter = 1
atom_list_order = [PRE_RES, PROT_RES, POST_RES, HYD_MOL, WAT_MOL, POST_WAT]
dump_atom_data = []
atom_lists = {PRE_RES: [],
PROT_RES: [],
POST_RES: [],
HYD_MOL: [],
WAT_MOL: [],
POST_WAT: []
}
with open(dump_file) as d:
for line in d:
line = line.strip()
if section is None:
section = find_dump_section_state(line)
if section is None:
raise InvalidDataError('Unexpected line in file {}: {}'.format(dump_file, line))
elif section == SEC_TIMESTEP:
timestep = line
# Reset variables
water_dict = defaultdict(list)
dump_atom_data = []
excess_proton = None
hydronium = []
for a_list in atom_lists:
atom_lists[a_list] = []
section = None
elif section == SEC_NUM_ATOMS:
if data_tpl_content[NUM_ATOMS] != int(line):
raise InvalidDataError('At timestep {} in file {}, the listed number of atoms ({}) does '
'not equal the number of atoms in the template data file '
'({}).'.format(timestep, dump_file, line, data_tpl_content[NUM_ATOMS]))
section = None
elif section == SEC_BOX_SIZE:
split_line = line.split()
diff = float(split_line[1]) - float(split_line[0])
box[counter - 1] = diff
if counter == 3:
counter = 0
section = None
counter += 1
elif section == SEC_ATOMS:
split_line = line.split()
# If there is an incomplete line in a dump file, move on to the next file
if len(split_line) < 7:
continue
atom_num = int(split_line[0])
mol_num = int(split_line[1])
atom_type = int(split_line[2])
charge = float(split_line[3])
x, y, z = map(float, split_line[4:7])
description = ''
atom_struct = [atom_num, mol_num, atom_type, charge, x, y, z, description]
# Keep track of separate portions of the system to allow sorting and processing
if mol_num == cfg[PROT_RES_MOL_ID]:
if atom_type == cfg[PROT_H_TYPE] and atom_num not in cfg[PROT_H_IGNORE]:
excess_proton = atom_struct
else:
atom_lists[PROT_RES].append(atom_struct)
elif atom_type == cfg[H3O_O_TYPE] or atom_type == cfg[H3O_H_TYPE]:
hydronium.append(atom_struct)
elif atom_type == cfg[WAT_O_TYPE] or atom_type == cfg[WAT_H_TYPE]:
water_dict[mol_num].append(atom_struct)
# Save everything else in three chunks for recombining sections post-processing
elif len(atom_lists[PROT_RES]) == 0:
atom_lists[PRE_RES].append(atom_struct)
elif len(water_dict) == 0:
atom_lists[POST_RES].append(atom_struct)
else:
atom_lists[POST_WAT].append(atom_struct)
if counter == data_tpl_content[NUM_ATOMS]:
counter = 0
section = None
# Now that finished reading all atom lines...
# Check and process!
if len(water_dict) == 0:
raise InvalidDataError('Found no water molecules. Check that the input types {} = {} '
'and {} = {} are in the dump '
'file.'.format(WAT_O_TYPE, cfg[WAT_O_TYPE],
WAT_H_TYPE, cfg[WAT_H_TYPE]))
if excess_proton is None:
if len(hydronium) != 4:
raise InvalidDataError('Did not find an excess proton or one hydronium ion. Check dump '
'file and input types: {} = {}; {} = {}; {} = {}'
.format(PROT_H_TYPE, cfg[PROT_H_TYPE],
H3O_O_TYPE, cfg[H3O_O_TYPE],
H3O_H_TYPE, cfg[H3O_H_TYPE]))
else:
if len(hydronium) != 0:
raise InvalidDataError('Found an excess proton and a hydronium atoms. Check dump file '
'and input types: {} = {}; {} = {}; {} = {}'
.format(PROT_H_TYPE, cfg[PROT_H_TYPE],
H3O_O_TYPE, cfg[H3O_O_TYPE],
H3O_H_TYPE, cfg[H3O_H_TYPE]))
deprotonate(cfg, atom_lists[PROT_RES], excess_proton, hydronium,
water_dict, box, data_tpl_content)
# Ensure in correct order for printing
atom_lists[HYD_MOL] = assign_hyd_mol(cfg, hydronium)
atom_lists[WAT_MOL] = sort_wat_mols(cfg, water_dict)
for a_list in atom_list_order:
dump_atom_data += atom_lists[a_list]
# overwrite atom_num, mol_num, atom_type, charge, then description
for index in range(len(dump_atom_data)):
if dump_atom_data[index][3] == data_tpl_content[ATOMS_CONTENT][index][3] or \
dump_atom_data[index][0] in cfg[PROT_TYPE_IGNORE_ATOMS]:
dump_atom_data[index][0:4] = data_tpl_content[ATOMS_CONTENT][index][0:4]
dump_atom_data[index][7] = ' '.join(data_tpl_content[ATOMS_CONTENT][index][7:])
else:
raise InvalidDataError("In reading file: {}\n found atom index {} with charge {} which "
"does not match the charge in the data template ({}). \n"
"To ignore this mis-match, list "
"the atom's index number in the keyword '{}' in the ini file."
"".format(dump_file,
dump_atom_data[index][0], dump_atom_data[index][3],
data_tpl_content[ATOMS_CONTENT][index][3],
PROT_TYPE_IGNORE_ATOMS))
d_out = create_out_fname(dump_file, suffix='_' + str(timestep),
ext='.data', base_dir=cfg[OUT_BASE_DIR])
data_tpl_content[HEAD_CONTENT][0] = "Created by evbdump2data from {} " \
"timestep {}".format(dump_file, timestep)
list_to_file(data_tpl_content[HEAD_CONTENT] + dump_atom_data + data_tpl_content[TAIL_CONTENT],
d_out)
counter += 1
if counter == 1:
print("Completed reading dumpfile {}".format(dump_file))
else:
warning("Dump file {} step {} did not have the full list of atom numbers. "
"Continuing program.".format(dump_file, timestep))
def process_dump_file(cfg, data_tpl_content, dump_file):
section = None
box = np.zeros((3, ))
counter = 1
atom_list_order = [PRE_RES, PROT_RES, POST_RES, HYD_MOL, WAT_MOL, POST_WAT]
dump_atom_data = []
atom_lists = {
PRE_RES: [],
PROT_RES: [],
POST_RES: [],
HYD_MOL: [],
WAT_MOL: [],
POST_WAT: []
}
with open(dump_file) as d:
for line in d:
line = line.strip()
if section is None:
section = find_dump_section_state(line)
if section is None:
raise InvalidDataError(
'Unexpected line in file {}: {}'.format(
dump_file, line))
elif section == SEC_TIMESTEP:
timestep = line
# Reset variables
water_dict = defaultdict(list)
dump_atom_data = []
excess_proton = None
hydronium = []
for a_list in atom_lists:
atom_lists[a_list] = []
section = None
elif section == SEC_NUM_ATOMS:
if data_tpl_content[NUM_ATOMS] != int(line):
raise InvalidDataError(
'At timestep {} in file {}, the listed number of atoms ({}) does '
'not equal the number of atoms in the template data file '
'({}).'.format(timestep, dump_file, line,
data_tpl_content[NUM_ATOMS]))
section = None
elif section == SEC_BOX_SIZE:
split_line = line.split()
diff = float(split_line[1]) - float(split_line[0])
box[counter - 1] = diff
if counter == 3:
counter = 0
section = None
counter += 1
elif section == SEC_ATOMS:
split_line = line.split()
# If there is an incomplete line in a dump file, move on to the next file
if len(split_line) < 7:
continue
atom_num = int(split_line[0])
mol_num = int(split_line[1])
atom_type = int(split_line[2])
charge = float(split_line[3])
x, y, z = map(float, split_line[4:7])
description = ''
atom_struct = [
atom_num, mol_num, atom_type, charge, x, y, z, description
]
# Keep track of separate portions of the system to allow sorting and processing
if mol_num == cfg[PROT_RES_MOL_ID]:
if atom_type == cfg[PROT_H_TYPE] and atom_num not in cfg[
PROT_H_IGNORE]:
excess_proton = atom_struct
else:
atom_lists[PROT_RES].append(atom_struct)
elif atom_type == cfg[H3O_O_TYPE] or atom_type == cfg[
H3O_H_TYPE]:
hydronium.append(atom_struct)
elif atom_type == cfg[WAT_O_TYPE] or atom_type == cfg[
WAT_H_TYPE]:
water_dict[mol_num].append(atom_struct)
# Save everything else in three chunks for recombining sections post-processing
elif len(atom_lists[PROT_RES]) == 0:
atom_lists[PRE_RES].append(atom_struct)
elif len(water_dict) == 0:
atom_lists[POST_RES].append(atom_struct)
else:
atom_lists[POST_WAT].append(atom_struct)
if counter == data_tpl_content[NUM_ATOMS]:
counter = 0
section = None
# Now that finished reading all atom lines...
# Check and process!
if len(water_dict) == 0:
raise InvalidDataError(
'Found no water molecules. Check that the input types {} = {} '
'and {} = {} are in the dump '
'file.'.format(WAT_O_TYPE, cfg[WAT_O_TYPE],
WAT_H_TYPE, cfg[WAT_H_TYPE]))
if excess_proton is None:
if len(hydronium) != 4:
raise InvalidDataError(
'Did not find an excess proton or one hydronium ion. Check dump '
'file and input types: {} = {}; {} = {}; {} = {}'
.format(PROT_H_TYPE, cfg[PROT_H_TYPE],
H3O_O_TYPE, cfg[H3O_O_TYPE],
H3O_H_TYPE, cfg[H3O_H_TYPE]))
else:
if len(hydronium) != 0:
raise InvalidDataError(
'Found an excess proton and a hydronium atoms. Check dump file '
'and input types: {} = {}; {} = {}; {} = {}'.
format(PROT_H_TYPE, cfg[PROT_H_TYPE],
H3O_O_TYPE, cfg[H3O_O_TYPE], H3O_H_TYPE,
cfg[H3O_H_TYPE]))
deprotonate(cfg, atom_lists[PROT_RES], excess_proton,
hydronium, water_dict, box,
data_tpl_content)
# Ensure in correct order for printing
atom_lists[HYD_MOL] = assign_hyd_mol(cfg, hydronium)
atom_lists[WAT_MOL] = sort_wat_mols(cfg, water_dict)
for a_list in atom_list_order:
dump_atom_data += atom_lists[a_list]
# overwrite atom_num, mol_num, atom_type, charge, then description
for index in range(len(dump_atom_data)):
if dump_atom_data[index][3] == data_tpl_content[ATOMS_CONTENT][index][3] or \
dump_atom_data[index][0] in cfg[PROT_TYPE_IGNORE_ATOMS]:
dump_atom_data[index][0:4] = data_tpl_content[
ATOMS_CONTENT][index][0:4]
dump_atom_data[index][7] = ' '.join(
data_tpl_content[ATOMS_CONTENT][index][7:])
else:
raise InvalidDataError(
"In reading file: {}\n found atom index {} with charge {} which "
"does not match the charge in the data template ({}). \n"
"To ignore this mis-match, list "
"the atom's index number in the keyword '{}' in the ini file."
"".format(
dump_file, dump_atom_data[index][0],
dump_atom_data[index][3],
data_tpl_content[ATOMS_CONTENT][index][3],
PROT_TYPE_IGNORE_ATOMS))
d_out = create_out_fname(dump_file,
suffix='_' + str(timestep),
ext='.data',
base_dir=cfg[OUT_BASE_DIR])
data_tpl_content[HEAD_CONTENT][0] = "Created by evbdump2data from {} " \
"timestep {}".format(dump_file, timestep)
list_to_file(
data_tpl_content[HEAD_CONTENT] + dump_atom_data +
data_tpl_content[TAIL_CONTENT], d_out)
counter += 1
if counter == 1:
print("Completed reading dumpfile {}".format(dump_file))
else:
warning(
"Dump file {} step {} did not have the full list of atom numbers. "
"Continuing program.".format(dump_file, timestep))
