Databases
StackOverflow2013
Postsdbo
POHow should I pyparsing to generate lists of elements from Boolean expressions?
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POHow should I pyparsing to generate lists of elements from Boolean expressions?
    primarykey
    data
    text
    <p>I have a list of Boolean expressions that represent physical objects that can be combined to represent larger objects. They look something like this: ((A and B) or C). This object can be represented by a combination of A and B or by C alone. I would like to generate a list of lists of strings that can be used to create the object. In this case I want [[A,B], [C]]. </p> <p>Pyparsing looks pretty intriguing so I've decided to give it a shot for this problem. After a few failed attempts I've settled on adapting the fourFn.py example from the website. This is what I have so far:</p> <pre><code>from pyparsing import Literal, CaselessLiteral, Word, Combine, \ Group, Optional, ZeroOrMore, Forward, alphanums exprStack = [] def myAnd(op1, op2): if isinstance(op1, str): return([op1, op2]) else: return op1.append(op2) def myOr(op1, op2): if isinstance(op1, str): return([[op1], [op2]]) else: return op1.append([op2]) def pushFirst(strg, loc, toks): exprStack.append(toks[0]) bnf = None def BNF(): """ boolop :: 'and' | 'or' gene :: alphanum atom :: gene | '(' expr ')' """ global bnf if not bnf: element = Word(alphanums) andop = Literal( "and" ) orop = Literal( "or" ) lpar = Literal( "(" ).suppress() rpar = Literal( ")" ).suppress() boolop = andop | orop expr = Forward() atom = ((element | lpar + expr + rpar).setParseAction(pushFirst) | (lpar + expr.suppress() + rpar)) expr &lt;&lt; atom + ZeroOrMore((boolop + expr).setParseAction(pushFirst)) bnf = expr return bnf # map operator symbols to corresponding arithmetic operations fn = {"or": myOr, "and": myAnd} def evaluateStack( s ): op = s.pop() if op in fn: op2 = evaluateStack(s) op1 = evaluateStack(s) return fn[op](op1, op2) else: return op if __name__ == "__main__": def test(s, expVal): global exprStack exprStack = [] results = BNF().parseString(s) val = evaluateStack(exprStack[:]) if val == expVal: print s, "=", val, results, "=&gt;", exprStack else: print "!!! "+s, val, "!=", expVal, results, "=&gt;", exprStack test("((A and B) or C)", [['A','B'], ['C']]) test("(A and B) or C", [['A','B'], ['C']]) test("(A or B) and C", [['A', 'C'], ['B', 'C']]) test("A and B", ['A', 'B']) test("A or B", [['A'], ['B']]) </code></pre> <p>The first three tests fail here and only return the first element of each expression in parentheses. A will get pushed to the stack multiple times. It seems that the way I modified fourFn.py has broken my script's ability to handle these groups. Is there a better way to approach this problem?</p> <p><strong>edit</strong> After a cup of coffee I realized the problems I was having were pretty easy to solve. My new and and or functions are as follows:</p> <pre><code>def myAnd(op1, op2): if isinstance(op1, str) and isinstance(op2, str): newlist = [op1, op2] elif isinstance(op1, str): newlist = [op1] newlist.append(op2) elif isinstance(op2, str): newlist = op1 newlist.append(op2) else: newlist = [op1.append(item) for item in op2] return newlist def myOr(op1, op2): if isinstance(op1, str) and isinstance(op2, str): newlist = [[op1], [op2]] r elif isinstance(op1, str): newlist = [op1] newlist.append([op2]) elif isinstance(op2, str): newlist = [op1] newlist.append([op2]) else: newlist = [op1, [op2]] return newlist1 </code></pre> <p>And the parser is constructed as follows:</p> <pre><code>expr = Forward() atom = element.setParseAction(pushFirst) | (lpar + expr + rpar) expr &lt;&lt; atom + ZeroOrMore((boolop + expr).setParseAction(pushFirst)) </code></pre> <p>A new and more interesting question involves how to deal with a case like this (A or B) and C. The result should be [[A, C], [B, C]]. Is there a typical pyparsing way of dealing with this issue?</p>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. USChristopher Garay
    1. USChristopher Garay
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POHow should I pyparsing to generate lists of elements from Boolean expressions?
      1. This table or related slice is empty.
      1. VTUndeletion
    2. VO
      singulars
      1. POHow should I pyparsing to generate lists of elements from Boolean expressions?
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. POHow should I pyparsing to generate lists of elements from Boolean expressions?
      1. This table or related slice is empty.
      1. VTUpMod
    1. COIf I were you I'd just created a plain classic AST with operations as nodes and operands as leaves, and then traverse it to build the constructs I like. In your case, it would be just replacing 'and' nodes with lists of their operands.
      singulars
      1. POHow should I pyparsing to generate lists of elements from Boolean expressions?
      1. US9000
    2. COCheck out the SimpleBool.py example on the pyparsing wiki. It uses a pyparsing helper method `operatorPrecedence`, which does a lot of the atom/factor/term work for you. SimpleBool.py will give you back a structure of Bool objects, similar to the AST proposed by @9000, but with richer object APIs, so you can evaluate the results directly.
      singulars
      1. POHow should I pyparsing to generate lists of elements from Boolean expressions?
      1. USPaulMcG
    3. COThanks for the comments. I like the AST suggestion and will definitely take a look at the SimpleBool.py example. I did manage to get the approach above to work for my test cases.
      singulars
      1. POHow should I pyparsing to generate lists of elements from Boolean expressions?
      1. USChristopher Garay
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload