Chapter 28
Notations

This chapter defines the notations used in this document.

28.1 Words

• code
  A piece of code (urbiscript, Java, C++ …).
• comment
  A comment in some programming language. For instance, /* hello /* world */ ! */ is a comment in urbiscript.
• environment-variable
  An environment variable name, e.g., PATH.
• ‘file-name’
  A file name.
• keyword
  A keyword in some programming language. For instance, watch is an urbiscript keyword.
• meta-variable
  Depending on the context, a variable name (i.e., an identifier in C++ or urbiscript), or a meta-variable name. A meta-variable denotes a place where some syntactic construct may be entered. For instance, in while (expression) statement, expression and statement do not denote two variable names, but two placeholders which can be filled with an arbitrary expression, and an arbitrary statement. For instance:

  while (!tasks.empty) { tasks.removeFront.process }.

• string
  A string in some programming language. For instance, "Hello, world!" is a string in urbiscript.

28.2 Frames

28.2.1 C++ Code

C++ source code is presented in frames as follows.

 
class Int 
{ 
public: 
  Foo(int v = 0) 
    : val_(v) 
  {} 
 
  void operator(int v) 
  { 
    std::swap(v, val_); 
    return v; 
  } 
 
  int operator() const 
  { 
    return val_; 
  } 
 
private: 
  int val_; 
};
       

28.2.2 Grammar Excerpts

The grammar fragments are written in EBNF (Extended Backus-Naur Form). The symbol ::= separates the left-hand symbol from the right-hand side part of the rule. Infix | denotes alternation, postfix-* 0-or-more repetition, postfix-+ 1-or-more repetition, and postfix-? denotes optional parts. Terminal symbols are written in double-quotes, and non-terminals in angle-brackets. Parentheses group.

The following frame defines the grammar syntax expressed in the same grammar syntax.

 
⟨grammar⟩ ::= ⟨rule⟩+ 
 
⟨rule⟩ ::= ⟨symbol⟩ "::=" ⟨rhs⟩ 
 
⟨rhs⟩ ::= ⟨rhs⟩∗ 
        ∣ ⟨rhs⟩ "|" ⟨rhs⟩ 
        ∣ ⟨rhs⟩ ("?" ∣ "*" ∣ "+") 
        ∣ "(" ⟨rhs⟩ ")" 
        ∣ ⟨symbol⟩ 
 
⟨symbol⟩ ::= "<" ⟨identifier⟩ ">" 
           ∣ ’"’ ⟨escaped-character⟩∗ ’"’ 
           ∣ "’" ⟨escaped-character⟩∗ "’"
       

Listing 29 provides the grammar of urbiscript.

28.2.3 Java Code

Java source code is presented in frames as follows.

 
import liburbi.main.*; 
public class Main 
{ 
    /// Load urbijava library. 
    static 
    { 
        System.loadLibrary("urbijava"); 
    } 
 
    public static void main(String argv[]) 
    { 
      // Does nothing for now. 
    } 
}
       

28.2.4 Shell Sessions

Interactive sessions with a (Unix) shell are represented as follows.

 
$ echo toto 
toto
       

The user entered ‘echo toto’, and the system answered ‘toto’. ‘$’ is the prompt: it starts the lines where the system invites the user to enter her commands.

28.2.5 urbiscript Sessions

Interactive sessions with Urbi are represented as follows.

 
echo("toto"); 
[00000001] *** toto
       

Contrary to shell interaction (see Section 28.2.4), there is no prompt that marks the user-entered lines (here echo("toto");, but, on the contrary, answers from the Urbi server start with a label that includes a timestamp (here ‘00000001’), and possibly a channel name, ‘output’ in the following example.

 
cout << "toto"; 
[00000002:output] "toto"
       

28.2.6 urbiscript Assertions

The following assertion frame:

 
true; 
1 < 2; 
1 + 2 * 3 == 7; 
"foobar"[0, 3] == "foo"; 
[1, 2, 3].map (function (a) { a * a }) == [1, 4, 9]; 
[ => ].empty;
       

denotes the following assertion-block (see Section 20.9) in an urbiscript-session frame:

 
assert 
{ 
  true; 
  1 < 2; 
  1 + 2 * 3 == 7; 
  "foobar"[0, 3] == "foo"; 
  [1, 2, 3].map (function (a) { a * a }) == [1, 4, 9]; 
  [ => ].empty; 
};