10
0
mirror of https://github.com/LCPQ/EMSL_Basis_Set_Exchange_Local synced 2024-10-31 19:23:42 +01:00

Version 0.1

This commit is contained in:
Thomas Applencourt 2014-11-12 14:30:26 +01:00
parent 974e6ac451
commit 53aaf0cb71
4 changed files with 847 additions and 0 deletions

66
EMSL_api.py Executable file
View File

@ -0,0 +1,66 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""EMSL Api.
Usage:
EMSL_api.py get_basis <db_path>
EMSL_api.py get_list_elements <db_path> <basis_name>
EMSL_api.py get_basis_data <db_path> <basis_name> <elts>...
EMSL_api.py create_db <db_path> <format> <contraction>
EMSL_api.py (-h | --help)
EMSL_api.py --version
Options:
-h --help Show this screen.
--version Show version.
"""
import sys
sys.path.append('./src/')
from docopt import docopt
from EMS_utility import EMSL_dump
from EMS_utility import EMSL_local
if __name__ == '__main__':
arguments = docopt(__doc__, version='EMSL Api 0.1')
if arguments["get_basis"]:
db_path = arguments["<db_path>"]
e = EMSL_local(db_path=db_path)
l = e.get_list_basis_available()
for i in l:
print i
elif arguments["get_list_elements"]:
db_path = arguments["<db_path>"]
basis_name = arguments["<basis_name>"]
e = EMSL_local(db_path=db_path)
l = e.get_list_element_available(basis_name)
for i in l:
print i
elif arguments["get_basis_data"]:
db_path = arguments["<db_path>"]
basis_name = arguments["<basis_name>"]
elts = arguments["<elts>"]
e = EMSL_local(db_path=db_path)
l = e.get_basis(basis_name, elts)
for i in l:
print i
elif arguments["create_db"]:
db_path = arguments["<db_path>"]
format = arguments["<format>"]
contraction = arguments["<contraction>"]
print "go"
e = EMSL_dump(db_path=db_path, format=format, contraction=contraction)
e.new_db()

BIN
db/EMSL.db Normal file

Binary file not shown.

199
src/EMS_utility.py Executable file
View File

@ -0,0 +1,199 @@
# -*- coding: utf-8 -*-
import sqlite3
import re
import sys
class EMSL_dump:
def __init__(self, db_path=None, format="GAMESS-US", contraction="True"):
self.db_path = db_path
self.format = format
self.contraction = contraction
import requests
self.requests = requests
def set_db_path(self, path):
"""Define the database path"""
self.db_path = path
def dwl_basis_list_raw(self):
print "Dwl the basis list info",
sys.stdout.flush()
"""Download the source code of the iframe who contains the list of the basis set available"""
url = "https://bse.pnl.gov/bse/portal/user/anon/js_peid/11535052407933/panel/Main/template/content"
page = self.requests.get(url).text
print "Done"
return page
def bl_raw_to_array(self, data_raw):
"""Parse the raw html to create a basis set array whith all the info:
url, name,description"""
d = []
for line in data_raw.split('\n'):
if "new basisSet(" in line:
b = line.find("(")
e = line.find(");")
s = line[b + 1:e]
tup = eval(s)
url = tup[0]
name = tup[1]
junkers = re.compile('[[" \]]')
elts = junkers.sub('', tup[3]).split(',')
des = tup[-1]
if "-ecp" in url.lower():
continue
d.append([name, url, des, elts])
d_sort = sorted(d, key=lambda x: x[0])
return d_sort
def create_url(self, url, name, elts):
"""Create the adequate url to get the basis data"""
elts_string = " ".join(elts)
path = "https://bse.pnl.gov:443/bse/portal/user/anon/js_peid/11535052407933/action/portlets.BasisSetAction/template/courier_content/panel/Main/"
path += "/eventSubmit_doDownload/true"
path += "?bsurl=" + url
path += "&bsname=" + name
path += "&elts=" + elts_string
path += "&format=" + self.format
path += "&minimize=" + self.contraction
return path
def basis_data_row_to_array(self, data, name, des, elts):
"""Parse the basis data raw html to get a nice tuple"""
d = []
b = data.find("$DATA")
e = data.find("$END")
if (b == -1 or data.find("$DATA$END") != -1):
print data
raise StandardError("WARNING not DATA")
else:
data = data[b + 5:e].split('\n\n')
for (elt, data_elt) in zip(elts, data):
d.append((name, des, elt, data_elt))
return d
def create_sql(self, list_basis_array):
"""Create the sql from the list of basis available data"""
conn = sqlite3.connect(self.db_path)
c = conn.cursor()
# Create table
c.execute('''CREATE TABLE all_value
(name text, description text, elt text, data text)''')
for i, [name, url, des, elts] in enumerate(list_basis_array):
print i, [name, url, des, elts]
url = self.create_url(url, name, elts)
basis_raw = self.requests.get(url).text
try:
basis_data = self.basis_data_row_to_array(
basis_raw, name, des, elts)
c.executemany(
"INSERT INTO all_value VALUES (?,?,?,?)", basis_data)
conn.commit()
print "Done"
except:
pass
conn.close()
def new_db(self):
"""Create new_db from scratch"""
_data = self.dwl_basis_list_raw()
array_basis = self.bl_raw_to_array(_data)
del _data
self.create_sql(array_basis)
class EMSL_local:
def __init__(self, db_path=None):
self.db_path = db_path
def get_list_basis_available(self):
conn = sqlite3.connect(self.db_path)
c = conn.cursor()
c.execute("SELECT DISTINCT name from all_value")
data = c.fetchall()
data = [i[0] for i in data]
conn.close()
return data
def get_list_element_available(self, basis_name):
conn = sqlite3.connect(self.db_path)
c = conn.cursor()
c.execute("SELECT DISTINCT elt from all_value WHERE name=:name_us",
{"name_us": basis_name})
data = c.fetchall()
data = [str(i[0]) for i in data]
conn.close()
return data
def get_basis(self, basis_name, elts):
conn = sqlite3.connect(self.db_path)
c = conn.cursor()
d = []
for elt in elts:
c.execute("SELECT DISTINCT data from all_value WHERE name=:name_cur AND elt=:elt_cur",
{"name_cur": basis_name,
"elt_cur": elt})
data = c.fetchone()
d.append(data[0])
conn.close()
return d
if __name__ == "__main__":
e = EMSL_local(db_path="EMSL.db")
l = e.get_list_basis_available()
for i in l:
print i
l = e.get_list_element_available("pc-0")
print l
l = e.get_basis("cc-pVTZ", ["H", "He"])
for i in l:
print i

582
src/docopt.py Normal file
View File

@ -0,0 +1,582 @@
"""Pythonic command-line interface parser that will make you smile.
* http://docopt.org
* Repository and issue-tracker: https://github.com/docopt/docopt
* Licensed under terms of MIT license (see LICENSE-MIT)
* Copyright (c) 2013 Vladimir Keleshev, vladimir@keleshev.com
"""
import sys
import re
__all__ = ['docopt']
__version__ = '0.6.1'
class DocoptLanguageError(Exception):
"""Error in construction of usage-message by developer."""
class DocoptExit(SystemExit):
"""Exit in case user invoked program with incorrect arguments."""
usage = ''
def __init__(self, message=''):
SystemExit.__init__(self, (message + '\n' + self.usage).strip())
class Pattern(object):
def __eq__(self, other):
return repr(self) == repr(other)
def __hash__(self):
return hash(repr(self))
def fix(self):
self.fix_identities()
self.fix_repeating_arguments()
return self
def fix_identities(self, uniq=None):
"""Make pattern-tree tips point to same object if they are equal."""
if not hasattr(self, 'children'):
return self
uniq = list(set(self.flat())) if uniq is None else uniq
for i, child in enumerate(self.children):
if not hasattr(child, 'children'):
assert child in uniq
self.children[i] = uniq[uniq.index(child)]
else:
child.fix_identities(uniq)
def fix_repeating_arguments(self):
"""Fix elements that should accumulate/increment values."""
either = [list(child.children) for child in transform(self).children]
for case in either:
for e in [child for child in case if case.count(child) > 1]:
if type(e) is Argument or type(e) is Option and e.argcount:
if e.value is None:
e.value = []
elif type(e.value) is not list:
e.value = e.value.split()
if type(e) is Command or type(e) is Option and e.argcount == 0:
e.value = 0
return self
def transform(pattern):
"""Expand pattern into an (almost) equivalent one, but with single Either.
Example: ((-a | -b) (-c | -d)) => (-a -c | -a -d | -b -c | -b -d)
Quirks: [-a] => (-a), (-a...) => (-a -a)
"""
result = []
groups = [[pattern]]
while groups:
children = groups.pop(0)
parents = [Required, Optional, OptionsShortcut, Either, OneOrMore]
if any(t in map(type, children) for t in parents):
child = [c for c in children if type(c) in parents][0]
children.remove(child)
if type(child) is Either:
for c in child.children:
groups.append([c] + children)
elif type(child) is OneOrMore:
groups.append(child.children * 2 + children)
else:
groups.append(child.children + children)
else:
result.append(children)
return Either(*[Required(*e) for e in result])
class LeafPattern(Pattern):
"""Leaf/terminal node of a pattern tree."""
def __init__(self, name, value=None):
self.name, self.value = name, value
def __repr__(self):
return '%s(%r, %r)' % (self.__class__.__name__, self.name, self.value)
def flat(self, *types):
return [self] if not types or type(self) in types else []
def match(self, left, collected=None):
collected = [] if collected is None else collected
pos, match = self.single_match(left)
if match is None:
return False, left, collected
left_ = left[:pos] + left[pos + 1:]
same_name = [a for a in collected if a.name == self.name]
if type(self.value) in (int, list):
if type(self.value) is int:
increment = 1
else:
increment = ([match.value] if type(match.value) is str
else match.value)
if not same_name:
match.value = increment
return True, left_, collected + [match]
same_name[0].value += increment
return True, left_, collected
return True, left_, collected + [match]
class BranchPattern(Pattern):
"""Branch/inner node of a pattern tree."""
def __init__(self, *children):
self.children = list(children)
def __repr__(self):
return '%s(%s)' % (self.__class__.__name__,
', '.join(repr(a) for a in self.children))
def flat(self, *types):
if type(self) in types:
return [self]
return sum([child.flat(*types) for child in self.children], [])
class Argument(LeafPattern):
def single_match(self, left):
for n, pattern in enumerate(left):
if type(pattern) is Argument:
return n, Argument(self.name, pattern.value)
return None, None
@classmethod
def parse(class_, source):
name = re.findall('(<\S*?>)', source)[0]
value = re.findall('\[default: (.*)\]', source, flags=re.I)
return class_(name, value[0] if value else None)
class Command(Argument):
def __init__(self, name, value=False):
self.name, self.value = name, value
def single_match(self, left):
for n, pattern in enumerate(left):
if type(pattern) is Argument:
if pattern.value == self.name:
return n, Command(self.name, True)
else:
break
return None, None
class Option(LeafPattern):
def __init__(self, short=None, long=None, argcount=0, value=False):
assert argcount in (0, 1)
self.short, self.long, self.argcount = short, long, argcount
self.value = None if value is False and argcount else value
@classmethod
def parse(class_, option_description):
short, long, argcount, value = None, None, 0, False
options, _, description = option_description.strip().partition(' ')
options = options.replace(',', ' ').replace('=', ' ')
for s in options.split():
if s.startswith('--'):
long = s
elif s.startswith('-'):
short = s
else:
argcount = 1
if argcount:
matched = re.findall('\[default: (.*)\]', description, flags=re.I)
value = matched[0] if matched else None
return class_(short, long, argcount, value)
def single_match(self, left):
for n, pattern in enumerate(left):
if self.name == pattern.name:
return n, pattern
return None, None
@property
def name(self):
return self.long or self.short
def __repr__(self):
return 'Option(%r, %r, %r, %r)' % (self.short, self.long,
self.argcount, self.value)
class Required(BranchPattern):
def match(self, left, collected=None):
collected = [] if collected is None else collected
l = left
c = collected
for pattern in self.children:
matched, l, c = pattern.match(l, c)
if not matched:
return False, left, collected
return True, l, c
class Optional(BranchPattern):
def match(self, left, collected=None):
collected = [] if collected is None else collected
for pattern in self.children:
m, left, collected = pattern.match(left, collected)
return True, left, collected
class OptionsShortcut(Optional):
"""Marker/placeholder for [options] shortcut."""
class OneOrMore(BranchPattern):
def match(self, left, collected=None):
assert len(self.children) == 1
collected = [] if collected is None else collected
l = left
c = collected
l_ = None
matched = True
times = 0
while matched:
# could it be that something didn't match but changed l or c?
matched, l, c = self.children[0].match(l, c)
times += 1 if matched else 0
if l_ == l:
break
l_ = l
if times >= 1:
return True, l, c
return False, left, collected
class Either(BranchPattern):
def match(self, left, collected=None):
collected = [] if collected is None else collected
outcomes = []
for pattern in self.children:
matched, _, _ = outcome = pattern.match(left, collected)
if matched:
outcomes.append(outcome)
if outcomes:
return min(outcomes, key=lambda outcome: len(outcome[1]))
return False, left, collected
class Tokens(list):
def __init__(self, source, error=DocoptExit):
self += source.split() if hasattr(source, 'split') else source
self.error = error
@staticmethod
def from_pattern(source):
source = re.sub(r'([\[\]\(\)\|]|\.\.\.)', r' \1 ', source)
source = [s for s in re.split('\s+|(\S*<.*?>)', source) if s]
return Tokens(source, error=DocoptLanguageError)
def move(self):
return self.pop(0) if len(self) else None
def current(self):
return self[0] if len(self) else None
def parse_long(tokens, options):
"""long ::= '--' chars [ ( ' ' | '=' ) chars ] ;"""
long, eq, value = tokens.move().partition('=')
assert long.startswith('--')
value = None if eq == value == '' else value
similar = [o for o in options if o.long == long]
if tokens.error is DocoptExit and similar == []: # if no exact match
similar = [o for o in options if o.long and o.long.startswith(long)]
if len(similar) > 1: # might be simply specified ambiguously 2+ times?
raise tokens.error('%s is not a unique prefix: %s?' %
(long, ', '.join(o.long for o in similar)))
elif len(similar) < 1:
argcount = 1 if eq == '=' else 0
o = Option(None, long, argcount)
options.append(o)
if tokens.error is DocoptExit:
o = Option(None, long, argcount, value if argcount else True)
else:
o = Option(similar[0].short, similar[0].long,
similar[0].argcount, similar[0].value)
if o.argcount == 0:
if value is not None:
raise tokens.error('%s must not have an argument' % o.long)
else:
if value is None:
if tokens.current() in [None, '--']:
raise tokens.error('%s requires argument' % o.long)
value = tokens.move()
if tokens.error is DocoptExit:
o.value = value if value is not None else True
return [o]
def parse_shorts(tokens, options):
"""shorts ::= '-' ( chars )* [ [ ' ' ] chars ] ;"""
token = tokens.move()
assert token.startswith('-') and not token.startswith('--')
left = token.lstrip('-')
parsed = []
while left != '':
short, left = '-' + left[0], left[1:]
similar = [o for o in options if o.short == short]
if len(similar) > 1:
raise tokens.error('%s is specified ambiguously %d times' %
(short, len(similar)))
elif len(similar) < 1:
o = Option(short, None, 0)
options.append(o)
if tokens.error is DocoptExit:
o = Option(short, None, 0, True)
else: # why copying is necessary here?
o = Option(short, similar[0].long,
similar[0].argcount, similar[0].value)
value = None
if o.argcount != 0:
if left == '':
if tokens.current() in [None, '--']:
raise tokens.error('%s requires argument' % short)
value = tokens.move()
else:
value = left
left = ''
if tokens.error is DocoptExit:
o.value = value if value is not None else True
parsed.append(o)
return parsed
def parse_pattern(source, options):
tokens = Tokens.from_pattern(source)
result = parse_expr(tokens, options)
if tokens.current() is not None:
raise tokens.error('unexpected ending: %r' % ' '.join(tokens))
return Required(*result)
def parse_expr(tokens, options):
"""expr ::= seq ( '|' seq )* ;"""
seq = parse_seq(tokens, options)
if tokens.current() != '|':
return seq
result = [Required(*seq)] if len(seq) > 1 else seq
while tokens.current() == '|':
tokens.move()
seq = parse_seq(tokens, options)
result += [Required(*seq)] if len(seq) > 1 else seq
return [Either(*result)] if len(result) > 1 else result
def parse_seq(tokens, options):
"""seq ::= ( atom [ '...' ] )* ;"""
result = []
while tokens.current() not in [None, ']', ')', '|']:
atom = parse_atom(tokens, options)
if tokens.current() == '...':
atom = [OneOrMore(*atom)]
tokens.move()
result += atom
return result
def parse_atom(tokens, options):
"""atom ::= '(' expr ')' | '[' expr ']' | 'options'
| long | shorts | argument | command ;
"""
token = tokens.current()
result = []
if token in '([':
tokens.move()
matching, pattern = {'(': [')', Required], '[': [']', Optional]}[token]
result = pattern(*parse_expr(tokens, options))
if tokens.move() != matching:
raise tokens.error("unmatched '%s'" % token)
return [result]
elif token == 'options':
tokens.move()
return [OptionsShortcut()]
elif token.startswith('--') and token != '--':
return parse_long(tokens, options)
elif token.startswith('-') and token not in ('-', '--'):
return parse_shorts(tokens, options)
elif token.startswith('<') and token.endswith('>') or token.isupper():
return [Argument(tokens.move())]
else:
return [Command(tokens.move())]
def parse_argv(tokens, options, options_first=False):
"""Parse command-line argument vector.
If options_first:
argv ::= [ long | shorts ]* [ argument ]* [ '--' [ argument ]* ] ;
else:
argv ::= [ long | shorts | argument ]* [ '--' [ argument ]* ] ;
"""
parsed = []
while tokens.current() is not None:
if tokens.current() == '--':
return parsed + [Argument(None, v) for v in tokens]
elif tokens.current().startswith('--'):
parsed += parse_long(tokens, options)
elif tokens.current().startswith('-') and tokens.current() != '-':
parsed += parse_shorts(tokens, options)
elif options_first:
return parsed + [Argument(None, v) for v in tokens]
else:
parsed.append(Argument(None, tokens.move()))
return parsed
def parse_defaults(doc):
defaults = []
for s in parse_section('options:', doc):
# FIXME corner case "bla: options: --foo"
_, _, s = s.partition(':') # get rid of "options:"
split = re.split('\n[ \t]*(-\S+?)', '\n' + s)[1:]
split = [s1 + s2 for s1, s2 in zip(split[::2], split[1::2])]
options = [Option.parse(s) for s in split if s.startswith('-')]
defaults += options
return defaults
def parse_section(name, source):
pattern = re.compile('^([^\n]*' + name + '[^\n]*\n?(?:[ \t].*?(?:\n|$))*)',
re.IGNORECASE | re.MULTILINE)
return [s.strip() for s in pattern.findall(source)]
def formal_usage(section):
_, _, section = section.partition(':') # drop "usage:"
pu = section.split()
return '( ' + ' '.join(') | (' if s == pu[0] else s for s in pu[1:]) + ' )'
def extras(help, version, options, doc):
if help and any((o.name in ('-h', '--help')) and o.value for o in options):
print(doc.strip("\n"))
sys.exit()
if version and any(o.name == '--version' and o.value for o in options):
print(version)
sys.exit()
class Dict(dict):
def __repr__(self):
return '{%s}' % ',\n '.join('%r: %r' % i for i in sorted(self.items()))
def docopt(doc, argv=None, help=True, version=None, options_first=False):
"""Parse `argv` based on command-line interface described in `doc`.
`docopt` creates your command-line interface based on its
description that you pass as `doc`. Such description can contain
--options, <positional-argument>, commands, which could be
[optional], (required), (mutually | exclusive) or repeated...
Parameters
----------
doc : str
Description of your command-line interface.
argv : list of str, optional
Argument vector to be parsed. sys.argv[1:] is used if not
provided.
help : bool (default: True)
Set to False to disable automatic help on -h or --help
options.
version : any object
If passed, the object will be printed if --version is in
`argv`.
options_first : bool (default: False)
Set to True to require options precede positional arguments,
i.e. to forbid options and positional arguments intermix.
Returns
-------
args : dict
A dictionary, where keys are names of command-line elements
such as e.g. "--verbose" and "<path>", and values are the
parsed values of those elements.
Example
-------
>>> from docopt import docopt
>>> doc = '''
... Usage:
... my_program tcp <host> <port> [--timeout=<seconds>]
... my_program serial <port> [--baud=<n>] [--timeout=<seconds>]
... my_program (-h | --help | --version)
...
... Options:
... -h, --help Show this screen and exit.
... --baud=<n> Baudrate [default: 9600]
... '''
>>> argv = ['tcp', '127.0.0.1', '80', '--timeout', '30']
>>> docopt(doc, argv)
{'--baud': '9600',
'--help': False,
'--timeout': '30',
'--version': False,
'<host>': '127.0.0.1',
'<port>': '80',
'serial': False,
'tcp': True}
See also
--------
* For video introduction see http://docopt.org
* Full documentation is available in README.rst as well as online
at https://github.com/docopt/docopt#readme
"""
argv = sys.argv[1:] if argv is None else argv
usage_sections = parse_section('usage:', doc)
if len(usage_sections) == 0:
raise DocoptLanguageError('"usage:" (case-insensitive) not found.')
if len(usage_sections) > 1:
raise DocoptLanguageError('More than one "usage:" (case-insensitive).')
DocoptExit.usage = usage_sections[0]
options = parse_defaults(doc)
pattern = parse_pattern(formal_usage(DocoptExit.usage), options)
# [default] syntax for argument is disabled
# for a in pattern.flat(Argument):
# same_name = [d for d in arguments if d.name == a.name]
# if same_name:
# a.value = same_name[0].value
argv = parse_argv(Tokens(argv), list(options), options_first)
pattern_options = set(pattern.flat(Option))
for options_shortcut in pattern.flat(OptionsShortcut):
doc_options = parse_defaults(doc)
options_shortcut.children = list(set(doc_options) - pattern_options)
# if any_options:
# options_shortcut.children += [Option(o.short, o.long, o.argcount)
# for o in argv if type(o) is Option]
extras(help, version, argv, doc)
matched, left, collected = pattern.fix().match(argv)
if matched and left == []: # better error message if left?
return Dict((a.name, a.value) for a in (pattern.flat() + collected))
raise DocoptExit()