Disallow keywords as identifier names

4 files changed, 78 insertions, 3 deletions

diff --git a/README b/README
index f30fdb1..624df36 100644
--- a/README
+++ b/README
@@ -19,11 +19,15 @@ Building the Oryx compiler is rather trivial.  The steps are as follows:
     the version of LLVM being utilized is 17.0.6.  The compiler may work
     with other versions, but it isn’t guaranteed.
 
-2.  Clone the compiler repository.
+2.  Install Gperf.  It should be available in your systems package
+    manager, and if not then you can easily find instructions online to
+    build from source.
+
+3.  Clone the compiler repository.
 
     $ git clone https://github.com/Mango0x45/oryx.git
 
-3.  Bootstrap and run the build script.
+4.  Bootstrap and run the build script.
 
     $ cc -o make make.c
     $ ./make  # See below for more details
diff --git a/make.c b/make.c
index 5d9212b..4692189 100644
--- a/make.c
+++ b/make.c
@@ -61,7 +61,7 @@ static void mkgmp(int);
 static bool tagvalid(const char *);
 static void chk_cpu_flags(void);
 static int globerr(const char *, int);
-static tjob cc, cc_test;
+static tjob cc, cc_test, gperf;
 
 static void
 usage(void)
@@ -158,7 +158,14 @@ main(int argc, char **argv)
 
 	glob_t g;
 
+	/* GNU Perf files */
+	assert(glob("src/*.gperf", 0, globerr, &g) == 0);
+	for (size_t i = 0; i < g.gl_pathc; i++)
+		tpenq(&tp, gperf, g.gl_pathv[i], NULL);
+	tpwait(&tp);
+
 	/* C files */
+	globfree(&g);
 	assert(glob("src/*.c", 0, globerr, &g) == 0);
 	for (size_t i = 0; i < g.gl_pathc; i++)
 		tpenq(&tp, cc, g.gl_pathv[i], NULL);
@@ -261,6 +268,24 @@ out:
 }
 
 void
+gperf(void *arg)
+{
+	struct strs cmd = {0};
+	char *dst = swpext(arg, "gen.c"), *src = arg;
+
+	if (!fflag && fmdnewer(dst, src))
+		goto out;
+
+	strspushl(&cmd, "gperf", src, "--output-file", dst);
+
+	cmdput(cmd);
+	cmdexec(cmd);
+	strsfree(&cmd);
+out:
+	free(dst);
+}
+
+void
 ld(void)
 {
 	glob_t g;
diff --git a/src/keywords.gperf b/src/keywords.gperf
new file mode 100644
index 0000000..7d81705
--- /dev/null
+++ b/src/keywords.gperf
@@ -0,0 +1,40 @@
+%compare-lengths
+%compare-strncmp
+%includes
+%readonly-tables
+
+%{
+#include <stdbool.h>
+
+#include "strview.h"
+
+static const char *in_word_set(const char *, size_t);
+%}
+
+%%
+pub
+static
+bool
+i8
+i16
+i32
+i64
+i128
+int
+u8
+u16
+u32
+u64
+u128
+uint
+f16
+f32
+f64
+f128
+rune
+%%
+bool
+iskeyword(strview_t sv)
+{
+	return in_word_set(sv.p, sv.len) != NULL;
+}
diff --git a/src/parser.c b/src/parser.c
index defe47d..9e748b5 100644
--- a/src/parser.c
+++ b/src/parser.c
@@ -47,6 +47,9 @@ static idx_t astalloc(ast_t *ast)
 static void astresz(ast_t *ast)
 	__attribute__((nonnull));
 
+/* Generated by Gperf */
+bool iskeyword(strview_t);
+
 /* TODO: Make thread-local? */
 static size_t toksidx;
 
@@ -177,6 +180,9 @@ parsedecl(ast_t *ast, aux_t *aux, lexemes_t toks, bool toplvl)
 		ast->lexemes[i] = toksidx;
 	}
 
+	if (iskeyword(toks.strs[toksidx]))
+		err("parser: Cannot use reserved word as identifier");
+
 	if (toks.kinds[toksidx++] != LEXIDENT)
 		err("parser: Expected identifier");
 	if (toks.kinds[toksidx++] != LEXCOLON)