This problem statement is quite wordy by itself and does not need a legend. You are given a regular expression and your task is to render its corresponding automaton as an ASCII art text drawing following the specification in the problem statement. Please, see examples.
A regular expression in this problem consists of uppercase letters from A to Z, special characters +, ?, *, and parenthesis that are used for grouping. An input to the problem is given by an <input> non-terminal of the following BNF grammar:
<input> ::= <expr>
<expr> ::= <term> | <term> `|' <expr>
<term> ::= <atom> | <atom><term> | <term><atom>
<atom> ::= <letter> | `(' <expr> `)' | <atom> `+' | <atom> `?' | <atom> `*'
<letter> ::= `A' | `B' | ... | `Z'
A regular expression is rendered as an ASCII art picture using the precise rules that are given below. They recursively define a unique representation for each regular expression as a rectangular box of characters with the specified number of rows and columns. Empty characters of the representation, including trailing ones, must be filled with spaces.
A <term> that consists of a sequence of n uppercase letters is rendered as a box of 3 rows and 4+n columns using + and - characters to render a border on the first and the last rows and columns as shown in the example. The production rule for the <term> non-terminal in the grammar is intentionally ambiguous. The longest possible sequence of adjacent <letter> non-terminals in the regular expression must be grouped into a <term> and rendered as a single box. For example, a <term> of `NERC' is rendered as the following 3×8 box:
+------+
+ NERC +
+------+
A <term> that consists of a sequence of <atom> non-terminals and <term> non-terminals with letters (as described above) is rendered by laying out the constituent boxes left-to-right, aligned vertically to the top, with 2 columns separating adjacent boxes. The height of the resulting box is equal to the maximum height of the constituent boxes. Each pair of adjacent boxes is joined by rendering -> characters on the 2nd row in the columns between them. For example, a <term> of `N(E)RC' (consisting of a sequence: <atom> of `A', <atom> of `(E)', and a letters-only <term> of `RC') is rendered as the following 3×20 box:
+---+ +---+ +----+
+ N +->+ E +->+ RC +
+---+ +---+ +----+
An <expr> that consists of a single <term> is rendered as its <term>.
An <expr> that consists of a `|`-separated sequence of <term> non-literals is rendered by laying out the corresponding <term> boxes top-to-bottom, aligned to the left, with a single row separating adjacent <term> boxes. The width of the resulting box is equal to the maximum width of the <term> boxes plus 6. There are 3 additional columns on the left, and 3 on the right. The first column and the last column use + and | characters to join the 2nd rows of all the <term> boxes from the top to the bottom one, with + placed on the 2nd row of each <term> box. The 2nd and the 3rd columns on the left and the 3rd-to-last and the 2nd-to-last columns on the right have -> characters on the 2nd rows of the corresponding <term> boxes. Additionally, shorter <term> boxes are connected on the right with extra - characters on their 2nd rows. For example, an <expr> of `C|ON|TEST' is rendered as the following 11×14 box:
+---+
+->+ C +---->+
| +---+ |
| |
| +----+ |
+->+ ON +--->+
| +----+ |
| |
| +------+ |
+->+ TEST +->+
+------+
An <atom> of `(' <expr> `)' is rendered as its <expr>.
An <atom> of <atom> `+' is rendered as a box of its source <atom> with 2 additional rows at the bottom and 6 additional columns (3 on the left and 3 on the right). The first and the last columns, starting with the 2nd row, and the last row are filled with the connecting characters as shown in the example.
+-> and ends with ->+ to connect to the 2nd row of the source <atom> box. +<- to represent a backwards edge in the automaton. For example, an <atom> of `A+' is rendered as the following 5×11 box.
+---+
+->+ A +->+
| +---+ |
| |
+<--------+
An <atom> of <atom> `?' is rendered as a box of its source <atom> with 3 additional rows at the top and 6 additional columns (3 on the left and 3 on the right). The first and the last columns (from the 2nd to the 5th row) and the 2nd row are filled with the connecting characters as shown in the example.
<atom> `?' is always empty (filled with spaces).->+ to represent an epsilon-edge in the corresponding automaton.+-> and ends with ->+ to connect to the 2nd row of the source <atom> box.For example, an <atom> of `B?' is rendered as the following 6×11 box.
+-------->+
| |
| +---+ |
+->+ B +->+
+---+
An <atom> of <atom> `*' is rendered as a box of its source <atom> with 5 additional rows (3 at the top and 2 at the bottom) and 6 additional columns (3 on the left and 3 on the right). The first and the last columns, with the 2nd and the last row, are filled with the connecting characters as shown in the example.
<atom> `*' is always empty (filled with spaces).->+ to represent an epsilon-edge in the corresponding automaton.+-> and ends with ->+ to connect to the 2nd row of the source <atom> box.+<- to represent a backwards edge in the automata. For example, an <atom> of `C*' is rendered as the following 8×11 box.
+-------->+
| |
| +---+ |
+->+ C +->+
| +---+ |
| |
+<--------+
An <input> is rendered as a box that has 6 more columns than the corresponding box of the <expr>, with 3 additional columns on the left, and 3 on the right. The 2nd row starts with S-> to represent the starting state of the automaton and ends with ->F to represent the final state of the automaton. See the example output.
The input consists of a single line that corresponds to the <input> non-terminal of the grammar given the problem statement and has at most 100 characters in length.
On the first line of the output, write two integers h and w --- the height and the width, correspondingly, of the h×w box that corresponds to the given <input>. On each of the next h lines, write w characters of the corresponding ASCII art rendering.