How will you proceed the following string:
[p('WHITE'),p('GREEN'),p('GREEN','RED'),p('GREEN','YELLOW'),p('GREEN','YELLOW','RED')]
to get a List of Nodes as suggested in the answers.
-
To get a list of which strings? The values between p(' and ')? Unique or non unique?brian_d– brian_d2010-11-29 22:29:31 +00:00Commented Nov 29, 2010 at 22:29
-
That doesn't look like a string. Is it enclosed in double quotes?Skilldrick– Skilldrick2010-11-29 22:29:52 +00:00Commented Nov 29, 2010 at 22:29
-
What is the expected output that you are looking for? Can you give the exact output you want for your example that you have given.Mark Byers– Mark Byers2010-11-29 22:30:59 +00:00Commented Nov 29, 2010 at 22:30
-
Why should this not be a string? It is a "sequence of characters", which is often called a "string" for short.Roland Illig– Roland Illig2010-11-29 22:31:19 +00:00Commented Nov 29, 2010 at 22:31
-
Yes it is enclosed in double quotes. And yes to have the values between p(' and') that are uniqueManuel Selva– Manuel Selva2010-11-29 22:32:11 +00:00Commented Nov 29, 2010 at 22:32
|
Show 3 more comments
1 Answer
It depends on how complex the expressions can get. I would definitely write a proper parser for it, since only then can I be confident to have done it properly. Plus, I will parse the string into an abstract syntax tree so that at the end I will automatically have a nicely modeled tree of objects:
class Node { ... }
class StringLiteral extends Node {
public String getValue() { ... }
}
class Term extends Node {
public String getName() { ... }
public int getArity() { ... }
public Node getArgument(int index) { ... }
}
2 Comments
Manuel Selva
Yes you are right. But now how will you parse the string to get these nodes ??
Roland Illig
The basic idea is to write a
Parser class that has one method per syntactical element that you are interested in. I would call them List<Node> parseArray(), Term parseTerm(), StringLiteral parseStringLiteral(), Node parseExpression(). The implementation of parseArray would look roughly like read('[']); while (next() != ']') { parseExpression(); } read(']');. The methods basically contain sequences of these operations or large switch statements, which decide what to parse next, depending on the next character from the input.