/*
Hunter Brazee
CS-210 Assignment 4
This program uses lex and built a scanner that satisfies the same requirements from homework 1
To use another file as input, it must be typed into the command line
*/
%{
/*
* This is part of the definitions section. It starts with %{ and ends with %}.
* Any text placed in this area will be copied verbatim into the lex.yy.c
* output file that lex generates, and will be placed near the top of that file.
*
* Typically this section is used to #include header files that will be needed
* in the generated scanner, but any legal C code can be placed here.
* For the assignment, no extra header files are needed, so this section is empty
*/
%}
/*
* This is also part of the definitions section. Anything after the %} above
* until the %% below is interpreted by lex as formal definitions. Things
* here won't be copied verbatim into the generated lex.yy.c output file,
* but will be analyzed by lex.
* The definitions for identifier, numeric literal, comments, strings, and character literal are defined
* Comments are indented to not start a new formal defintion in the leftmost column
*/
//Definiton for identifier, matches the first letter, which is upper or lower case, then followed by letter, digit
or underscore
ID [a-zA-Z][_a-zA-Z0-9]*
//Definiton for numeric literal, starts with digit 0-9, followed by another digit, a hexidecimal digit, or the special
characters _ . and #
NUMLIT [0-9][0-9a-fA-F_.#]*
//Definiton for comments, which start with /*, then matches the contents of the comment, then matches the */
for the end of the comment, including the start and end symbols
COMMENT \/\*[^*]*\*+([^*/][^*]*\*+)*\/
//Definiton for string,which starts with the first ", then matches the rest of the string as the lexeme, including
the " "
STRING \"[^\"]*\"
//Definiton for character literal, which starts with ', followed by the character, then the closing '
CHARLIT \'[a-zA-Z]*\'
%%
/*
* This section defines the rules which are analyzed by lex, and each has a pattern and a rule
* Patterns for keyword and operator are chained using | which allows for the chained to match those lexemes
* This section defines all of the patterns and rules for printing the specific lexemes
* Comments are indented in this secton because leftmost is the start of new pattern rule
*/
/* Rules for num literal, comment, string and char literal are analyzed*/
{NUMLIT} { printf("%s (numeric literal)\n", yytext); }
{COMMENT} { printf("%s (comment)\n", yytext); }
{STRING} { printf("%s (string)\n", yytext); }
{CHARLIT} { printf("%s (character literal)\n", yytext); }
/* List of specifc keywords to catch as keyword lexeme before identifier */
accessor |
and |
array |
bool |
case |
character |
constant |
else |
elsif |
end |
exit |
float |
func |
if |
ifc |
in |
integer |
is |
mutator |
natural |
null |
of |
or |
others |
out |
pkg |
positive |
proc |
ptr |
range |
subtype |
then |
type |
when |
while {
/* The action to be taken when one of the specified keywords is found */
printf("%s (keyword)\n", yytext );
}
/* Rule for identifier */
{ID} printf("%s (identifier)\n", yytext);
/*List of symbols that are operator lexemes */
"."|"<"|">"|"("|")"|"+"|"-"|"*"|"/"|"|"|"&"|";"|","|":"|"="|":="|".."|"<<"|">>"|"<>"|"<="|">="|"**"|"!="|"=>"|"["|"]"|"{"|"}"|"{:
"|"}:"|"$"|"@" printf("%s (operator)\n", yytext);
[ \t\n]+ /* discard whitespace */
/* Prints UNK for unknown lexeme */
. printf("%s (UNK)\n", yytext);
%%
/*
* This section is the user subroutines section. It starts after the %% above
* and continues to the end of the file. Anything appearing in the section is
* copied verbatim into the lex.yy.c file generated by lex, and will appear
* toward the bottom of that file.
*
* This section is where subroutines specific to a given scanner are placed.
* Any legal C code can be placed here.
*
*
* A main function is defined which guides the lexer, and takes the file name on the command line as the input
* Not entering a file on the command line will allow the user to type in thier own input,
* however this is only used for testing and not using the file as input, that must be done on the command line
*/
int main(int argc, char **argv)
{
if (argc > 1)
yyin = fopen(argv[1], "r"); //Uses the file name on the command line as input
else
yyin = stdin; //Used for testing the lexemes for output
yylex();
}
/*
* This defines yywrap, a special function used by lex to determine what should
* be done when the end of an input file is encountered.
*/
int yywrap()
{
return 1;
}
