File: query.txt

package info (click to toggle)
rdkit 201203-3
  • links: PTS, VCS
  • area: main
  • in suites: wheezy
  • size: 37,840 kB
  • sloc: cpp: 93,902; python: 51,897; java: 5,192; ansic: 3,497; xml: 2,499; sql: 1,641; yacc: 1,518; lex: 1,076; makefile: 325; fortran: 183; sh: 153; cs: 51
file content (28 lines) | stat: -rwxr-xr-x 769 bytes parent folder | download | duplicates (2)
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
A query is a piece of topology, like a molecule, but nodes are
QueryAtoms and edges are QueryBonds.

QueryAtoms and QueryBonds are subclasses of QueryObjects

QueryObjects:

Different types exist:
  equality
  inequality (>,>=,<,<=)
  range (within a min,max range)
  set membership
  and
  or
  xor
all can be negated

can have child queries (not part of topology) to support (and, or, xor)

Include num_match and optional presence functionality (to allow "one
 or more", "maybe present"  etc.) so we can do wildcard searches:
  "does the molecule have two aldehydes separated by a chain of
   conjugated carbons"  (something like C(=O)[C=C]*C(=O))
  

QueryAtoms and QueryBonds have specific member data
(i.e. atomicNum, implicitValence, order, isAromatic, etc.).