From 99a385752f595bceb2fa8fd797f3f7aec0f61ea6 Mon Sep 17 00:00:00 2001
From: able <abl3theabove@gmail.com>
Date: Thu, 7 Mar 2024 11:04:12 -0600
Subject: [PATCH] continued modularization

---
 Makefile       |   2 +-
 src/einfo.h    |   6 ++
 src/hbas.c     | 197 +------------------------------------------------
 src/register.c |   2 +-
 src/token.c    | 187 ++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 197 insertions(+), 197 deletions(-)
 create mode 100644 src/einfo.h
 create mode 100644 src/token.c

diff --git a/Makefile b/Makefile
index ab7f06b..5b7b14a 100644
--- a/Makefile
+++ b/Makefile
@@ -5,7 +5,7 @@ CFLAGS = -Wall -Wextra -Wpedantic -std=c17 -O3
 .PHONY: clean
 
 hbas: src/hbas.c
-	${CC} ${CFLAGS} ${CFLAGS_EXTRA} hbas.c -o hbas
+	${CC} ${CFLAGS} ${CFLAGS_EXTRA} src/hbas.c -o hbas
 
 example: hbas example.S
 	./hbas < example.S > example
diff --git a/src/einfo.h b/src/einfo.h
new file mode 100644
index 0000000..d6de8b3
--- /dev/null
+++ b/src/einfo.h
@@ -0,0 +1,6 @@
+typedef struct EInfo_s
+{
+    Token token;
+    size_t line;
+    size_t line_start;
+} EInfo;
diff --git a/src/hbas.c b/src/hbas.c
index bd8ae6e..4fd08a6 100644
--- a/src/hbas.c
+++ b/src/hbas.c
@@ -30,6 +30,8 @@ SOFTWARE.
 #include "instructions.c"
 #include "hash.c"
 #include "register.c"
+#include "token.c"
+#include "einfo.h"
 
 void hd(char *data, size_t len)
 {
@@ -152,194 +154,6 @@ int slurp(FILE *fd, ByteVec *out)
     return err;
 }
 
-typedef enum TokenKind_e
-{
-    TokInvalid = '!',
-    TokEOF = '$',
-    TokIdent = 'A',
-    TokNeg = '-',
-    TokNumber = '0',
-    TokBadNumber = '9',
-    TokComma = ',',
-    TokDot = '.',
-    TokColon = ':',
-    TokComment = ';',
-    TokNewline = 'n',
-} TokenKind;
-typedef struct Token_s
-{
-    TokenKind kind;
-    size_t start;
-    size_t len;
-    uint64_t num;
-} Token;
-
-Token token_ident(char *input, size_t len, size_t pos)
-{
-    size_t start = pos;
-    while (pos < len)
-    {
-        char chr = input[pos];
-        char chru = chr & ~0x20;
-        int good = chr == '_' || (chr >= '0' && chr <= '9') || (chru >= 'A' && chru <= 'Z');
-        if (!good)
-        {
-            break;
-        }
-        pos += 1;
-    }
-    return (Token){TokIdent, start, pos - start, 0};
-}
-
-Token token_number(char *input, size_t len, size_t pos)
-{
-    char *ptr = &input[pos];
-    char next = '\0';
-    size_t start = pos;
-    size_t digits = 0;
-    uint64_t base = 10;
-    uint64_t rv = 0;
-    uint64_t pre_overflow;
-    AsmError bad_num = ErrOk;
-
-    if (pos + 1 < len)
-    {
-        next = ptr[1] & ~0x20;
-    }
-
-    if (input[pos] == '0')
-    {
-        if (next == 'X')
-        {
-            base = 16;
-            pos += 2;
-        }
-        else if (next == 'D')
-        {
-            base = 10;
-            pos += 2;
-        }
-        else if (next == 'O')
-        {
-            base = 8;
-            pos += 2;
-        }
-        else if (next == 'B')
-        {
-            base = 2;
-            pos += 2;
-        }
-    }
-    pre_overflow = (~(size_t)0) / base;
-    // valid: "0x_0", "0_"
-    // invalid: "0x_"
-    while (pos < len)
-    {
-        uint64_t digit;
-        uint64_t next;
-        char chr = input[pos];
-        char chru = chr & ~0x20;
-        if (chr == '_')
-        {
-            pos += 1;
-            continue;
-        }
-        digit = (uint64_t)chr - (uint64_t)'0';
-        if (digit >= 10)
-        {
-            digit = (uint64_t)chru - (uint64_t)('A' - 10);
-        }
-        if (digit >= base)
-        {
-            if (chr >= '0' && chr <= '9')
-            {
-                bad_num = ErrBadNumDigit;
-            }
-            else if (chru >= 'A' && chru <= 'Z')
-            {
-                bad_num = ErrBadNumDigit;
-            }
-            break;
-        }
-
-        pos += 1;
-        digits += 1;
-
-        next = rv * base + digit;
-        if (rv > pre_overflow || next < rv)
-        {
-            bad_num = ErrBadNumOverflow;
-            break;
-        }
-        rv = next;
-    }
-
-    if (digits == 0)
-    {
-        bad_num = ErrBadNumNoDigit;
-    }
-
-    if (bad_num)
-    {
-        return (Token){TokBadNumber, start, pos - start, bad_num};
-    }
-    else
-    {
-        return (Token){TokNumber, start, pos - start, rv};
-    }
-}
-
-Token token(char *input, size_t len, size_t pos)
-{
-    char chr, chru;
-    char *ptr = &input[pos];
-    while (pos < len && (input[pos] == ' ' || input[pos] == '\t'))
-    {
-        pos += 1;
-    }
-    if (pos == len)
-    {
-        return (Token){TokEOF, pos, 0, 0};
-    }
-    ptr = &input[pos];
-    chr = *ptr;
-    if (chr == ',' || chr == '-' || chr == '.' || chr == ':')
-    {
-        return (Token){(TokenKind)chr, pos, 1, 0};
-    }
-    if (chr == '\n')
-    {
-        return (Token){TokNewline, pos, 1, 0};
-    }
-    if (chr == '\r')
-    {
-        if (pos + 1 < len && ptr[1] == '\n')
-        {
-            return (Token){TokNewline, pos, 2, 0};
-        }
-        return (Token){TokNewline, pos, 1, 0};
-    }
-    if (chr == ';')
-    {
-        size_t clen = 1;
-        while (pos + clen < len && ptr[clen] != '\n' && ptr[clen] != '\r')
-        {
-            clen += 1;
-        }
-        return (Token){TokComment, pos, clen, 0};
-    }
-    if (chr >= '0' && chr <= '9')
-    {
-        return token_number(input, len, pos);
-    }
-    chru = chr & ~0x20;
-    if (chr == '_' || (chru >= 'A' && chru <= 'Z'))
-    {
-        return token_ident(input, len, pos);
-    }
-    return (Token){TokInvalid, pos, 1, 0};
-}
-
 typedef struct Hole_s
 {
     size_t location;
@@ -538,13 +352,6 @@ AsmError assemble_instr(
     return 0;
 }
 
-typedef struct EInfo_s
-{
-    Token token;
-    size_t line;
-    size_t line_start;
-} EInfo;
-
 AsmError assemble(InstHt ht, char *input, size_t len, ByteVec *out, EInfo *einfo)
 {
     ByteVec rv = {malloc(MIN_SIZE), MIN_SIZE, 0};
diff --git a/src/register.c b/src/register.c
index ee8637c..0083a6f 100644
--- a/src/register.c
+++ b/src/register.c
@@ -27,4 +27,4 @@ int parse_register(char *name, size_t len)
         return 256; // Register number too large
     }
     return (int)rv;
-}
\ No newline at end of file
+}
diff --git a/src/token.c b/src/token.c
new file mode 100644
index 0000000..dec4919
--- /dev/null
+++ b/src/token.c
@@ -0,0 +1,187 @@
+typedef enum TokenKind_e
+{
+    TokInvalid = '!',
+    TokEOF = '$',
+    TokIdent = 'A',
+    TokNeg = '-',
+    TokNumber = '0',
+    TokBadNumber = '9',
+    TokComma = ',',
+    TokDot = '.',
+    TokColon = ':',
+    TokComment = ';',
+    TokNewline = 'n',
+} TokenKind;
+typedef struct Token_s
+{
+    TokenKind kind;
+    size_t start;
+    size_t len;
+    uint64_t num;
+} Token;
+
+Token token_ident(char *input, size_t len, size_t pos)
+{
+    size_t start = pos;
+    while (pos < len)
+    {
+        char chr = input[pos];
+        char chru = chr & ~0x20;
+        int good = chr == '_' || (chr >= '0' && chr <= '9') || (chru >= 'A' && chru <= 'Z');
+        if (!good)
+        {
+            break;
+        }
+        pos += 1;
+    }
+    return (Token){TokIdent, start, pos - start, 0};
+}
+
+Token token_number(char *input, size_t len, size_t pos)
+{
+    char *ptr = &input[pos];
+    char next = '\0';
+    size_t start = pos;
+    size_t digits = 0;
+    uint64_t base = 10;
+    uint64_t rv = 0;
+    uint64_t pre_overflow;
+    AsmError bad_num = ErrOk;
+
+    if (pos + 1 < len)
+    {
+        next = ptr[1] & ~0x20;
+    }
+
+    if (input[pos] == '0')
+    {
+        if (next == 'X')
+        {
+            base = 16;
+            pos += 2;
+        }
+        else if (next == 'D')
+        {
+            base = 10;
+            pos += 2;
+        }
+        else if (next == 'O')
+        {
+            base = 8;
+            pos += 2;
+        }
+        else if (next == 'B')
+        {
+            base = 2;
+            pos += 2;
+        }
+    }
+    pre_overflow = (~(size_t)0) / base;
+    // valid: "0x_0", "0_"
+    // invalid: "0x_"
+    while (pos < len)
+    {
+        uint64_t digit;
+        uint64_t next;
+        char chr = input[pos];
+        char chru = chr & ~0x20;
+        if (chr == '_')
+        {
+            pos += 1;
+            continue;
+        }
+        digit = (uint64_t)chr - (uint64_t)'0';
+        if (digit >= 10)
+        {
+            digit = (uint64_t)chru - (uint64_t)('A' - 10);
+        }
+        if (digit >= base)
+        {
+            if (chr >= '0' && chr <= '9')
+            {
+                bad_num = ErrBadNumDigit;
+            }
+            else if (chru >= 'A' && chru <= 'Z')
+            {
+                bad_num = ErrBadNumDigit;
+            }
+            break;
+        }
+
+        pos += 1;
+        digits += 1;
+
+        next = rv * base + digit;
+        if (rv > pre_overflow || next < rv)
+        {
+            bad_num = ErrBadNumOverflow;
+            break;
+        }
+        rv = next;
+    }
+
+    if (digits == 0)
+    {
+        bad_num = ErrBadNumNoDigit;
+    }
+
+    if (bad_num)
+    {
+        return (Token){TokBadNumber, start, pos - start, bad_num};
+    }
+    else
+    {
+        return (Token){TokNumber, start, pos - start, rv};
+    }
+}
+
+Token token(char *input, size_t len, size_t pos)
+{
+    char chr, chru;
+    char *ptr = &input[pos];
+    while (pos < len && (input[pos] == ' ' || input[pos] == '\t'))
+    {
+        pos += 1;
+    }
+    if (pos == len)
+    {
+        return (Token){TokEOF, pos, 0, 0};
+    }
+    ptr = &input[pos];
+    chr = *ptr;
+    if (chr == ',' || chr == '-' || chr == '.' || chr == ':')
+    {
+        return (Token){(TokenKind)chr, pos, 1, 0};
+    }
+    if (chr == '\n')
+    {
+        return (Token){TokNewline, pos, 1, 0};
+    }
+    if (chr == '\r')
+    {
+        if (pos + 1 < len && ptr[1] == '\n')
+        {
+            return (Token){TokNewline, pos, 2, 0};
+        }
+        return (Token){TokNewline, pos, 1, 0};
+    }
+    if (chr == ';')
+    {
+        size_t clen = 1;
+        while (pos + clen < len && ptr[clen] != '\n' && ptr[clen] != '\r')
+        {
+            clen += 1;
+        }
+        return (Token){TokComment, pos, clen, 0};
+    }
+    if (chr >= '0' && chr <= '9')
+    {
+        return token_number(input, len, pos);
+    }
+    chru = chr & ~0x20;
+    if (chr == '_' || (chru >= 'A' && chru <= 'Z'))
+    {
+        return token_ident(input, len, pos);
+    }
+    return (Token){TokInvalid, pos, 1, 0};
+}