a4_0006
をテンプレートにして作成
[
トップ
] [
新規
|
一覧
|
単語検索
|
最終更新
|
ヘルプ
]
開始行:
* acl4のページ0006
-(by [[K]], 2026.02.02)
** (1)
#if (a_Version >= 1)
#define Token0 a_Token0
#define Token0Table a_Token0Table
#define Token0Table_ini a_Token0Table_ini
#define Token0_get a_Token0_get
#define Token1_get a_Token1_get
#endif
a_class(a_Token0Table) {
uint32_t *op[4];
unsigned char chrTyp[256];
// 0: 空白類.
// 1: 1文字記号.
// 2: 一般記号文字.
// 3: アルファベット/数字.
};
a_class(a_Token0) {
a_Token0Table *tbl;
const char *s, *s1;
unsigned char cTyp;
uint32_t c;
intptr_t len;
};
a_static int a_Token0Table_iniSub(uint32_t *op, const ch...
{
int i = 0;
for (;;) {
while (*s == ' ') s++;
if (*s == '\0') break;
if (s[1] <= ' ') { op[i++] = s[0]; ...
if (s[2] <= ' ') { op[i++] = s[0] | s[1] << 8; ...
if (s[3] <= ' ') { op[i++] = s[0] | s[1] << 8 | ...
if (s[4] <= ' ') { op[i++] = s[0] | s[1] << 8 | ...
break;
}
op[i] = 0;
return i;
}
a_static void a_Token0Table_ini(a_Token0Table *w)
{
static uint32_t op4[64], op3[64], op2[64], op1[64];
a_Token0Table_iniSub(op4, "");
a_Token0Table_iniSub(op3, ">>= <<= ...");
a_Token0Table_iniSub(op2, "== != <= >= << >> ++ -- +...
a_Token0Table_iniSub(op1, "= < > + - * / % ^ & | ! ~...
int c;
for (c = 0; c < 256; c++) {
unsigned char ct = 0;
if (c != 0) {
if (strchr(" \t\r\n", c) != NULL) ct = 0;
if (strchr("(){},;[]\"\'", c) != NULL) ct = 1;
if (strchr("!#%&-=^~:+*<>./?", c) != NULL) c...
if ('0' <= c && c <= '9') ct = 3;
if ('A' <= c && c <= 'Z') ct = 3;
if ('a' <= c && c <= 'z') ct = 3;
if (strchr("_$@`", c) != NULL) ct = 3;
}
w->chrTyp[c] = ct;
}
w->op[0] = op4;
w->op[1] = op3;
w->op[2] = op2;
w->op[3] = op1;
}
a_static char *a_Token0_get(a_Token0 *w)
{
const char *s = w->s, *s1 = w->s1;
unsigned char ct, *chrTyp = w->tbl->chrTyp;
intptr_t len = 0, i; w->c = 0;
for (;;) {
if (s >= s1) goto fin;
ct = chrTyp[*(const unsigned char *) s];
if (ct == 0) { s++; continue; }
break;
}
if (ct == 1) {
len = 1;
w->c = *s;
goto fin;
}
if (ct == 2) {
unsigned char ch0 = *s, ch1 = 0, ch2 = 0, ch3 = 0;
if (s + 3 < s1) {
ch1 = s[1]; ch2 = s[2]; ch3 = s[3];
} else {
if (s + 1 < s1) ch1 = s[1];
if (s + 2 < s1) ch2 = s[2];
}
uint32_t op = ch0 | ch1 << 8 | ch2 << 16 | ch3 <...
opTbl = w->tbl->op[0]; len = 4; w->c = op;
for (i = 0; opTbl[i] != 0; i++) {
if (opTbl[i] == op) goto fin;
}
opTbl = w->tbl->op[1]; len = 3; w->c = op &= 0x0...
for (i = 0; opTbl[i] != 0; i++) {
if (opTbl[i] == op) goto fin;
}
opTbl = w->tbl->op[2]; len = 2; w->c = op &= 0x0...
for (i = 0; opTbl[i] != 0; i++) {
if (opTbl[i] == op) goto fin;
}
opTbl = w->tbl->op[3]; len = 1; w->c = op &= 0x0...
for (i = 0; opTbl[i] != 0; i++) {
if (opTbl[i] == op) goto fin;
}
ct = 255; len = 0; w->c = 0;
}
if (ct == 3) {
w->c = *s;
for (len = 1;;) {
if (s + len >= s1) goto fin;
if (chrTyp[((unsigned char *) s)[len]] == ct...
break;
}
}
fin:
w->cTyp = ct;
w->len = len;
w->s = s + len;
return (char *) s;
}
a_static char *a_Token1_get(a_Token0 *w)
{
for (;;) {
const char *s = a_Token0_get(w), *s1 = w->s1, *t;
uint32_t c = w->c;
if (s >= s1) return (char *) s;
if (c == ('/' | '/' << 8)) {
t = memchr(s, '\n', s1 - s);
if (t != NULL) { w->s = t; continue; }
w->s = s1; w->len = 0; return (char *) s1;
}
if (c == 0x22 || c == 0x27) {
// 改行もしくはエスケープされていないcがでる...
t = s + 1;
for (;;) {
if (t >= s1) break;
if (*(unsigned char *) t == c) { t++; br...
if (*t == '\n') break;
if (*t != '\\') { t++; continue; }
t++;
if (t >= s1) break;
if (*t == '\n') break;
t++;
}
w->len = t - s;
w->s = t;
return (char *) s;
}
if ('0' <= c && c <= '9') {
t = s + w->len;
for (;;) {
if (t >= s1) break;
c = *t;
if (c == '.') { t++; continue; }
if ('0' <= c && c <= '9') { t++; continu...
if ('A' <= c && c <= 'Z') { t++; continu...
if ('a' <= c && c <= 'z') { t++; continu...
if (c == '_') { t++; continue; }
if ((c == '+' || c == '-') && t[-1] == '...
break;
}
w->len = t - s;
w->s = t;
return (char *) s;
}
return (char *) s;
}
}
** (2) Token1_get
-与えられた文字列から先頭のトークンを切り出す関数です。連...
-C言語にほぼ準拠していますが、[[K]]の好みで余計な演算子も...
-このバージョンでは /* ～ */ 形式のコメントに対応していま...
-プログラムの仕組みとしては Token0 と Token1 の二層構造に...
-t0006a.cの実行結果を見れば、大体どんな感じかわかると思い...
** (3) サンプルプログラム: t0006a.c
-acl4ライブラリは以下のようにして利用します。
-[1]まずa4_0001～a4_0006のプログラムをつなげて"acl4.c"と...
-[2]次にacl4を使ったプログラムを書きます。
-[3]コンパイル時に、"acl4.c"のあるパスをインクルードパス...
#define a_Version 1
#include <acl4.c>
int main(int argc, const char **argv)
{
VecChr vc[1]; VecChr_ini(vc);
VecChr_readFileAll_errChk(vc, argv[1]);
VecChr_eraseCr(vc);
Token0Table t0t[1]; Token0Table_ini(t0t);
Token0 t0[1]; t0->tbl = t0t;
t0->s = vc->p; t0->s1 = vc->p + vc->n;
int i = 1;
for (;;) {
char *t = Token1_get(t0);
if (t0->len == 0 || i > 20) break;
printf("%6d: %.*s\n", i++, (int) t0->len, t);
}
VecChr_din(vc);
a_malloc_debugList(_arg);
return 0;
}
>t0006a t0006a.c
1: #
2: define
3: a_Version
4: 1
5: #
6: include
7: <
8: acl4
9: .
10: c
11: >
12: int
13: main
14: (
15: int
16: argc
17: ,
18: const
19: char
20: *
t0006a.c(22): malloc_debugList()
** (99) 更新履歴
-2026.02.02(月) 初版
終了行:
* acl4のページ0006
-(by [[K]], 2026.02.02)
** (1)
#if (a_Version >= 1)
#define Token0 a_Token0
#define Token0Table a_Token0Table
#define Token0Table_ini a_Token0Table_ini
#define Token0_get a_Token0_get
#define Token1_get a_Token1_get
#endif
a_class(a_Token0Table) {
uint32_t *op[4];
unsigned char chrTyp[256];
// 0: 空白類.
// 1: 1文字記号.
// 2: 一般記号文字.
// 3: アルファベット/数字.
};
a_class(a_Token0) {
a_Token0Table *tbl;
const char *s, *s1;
unsigned char cTyp;
uint32_t c;
intptr_t len;
};
a_static int a_Token0Table_iniSub(uint32_t *op, const ch...
{
int i = 0;
for (;;) {
while (*s == ' ') s++;
if (*s == '\0') break;
if (s[1] <= ' ') { op[i++] = s[0]; ...
if (s[2] <= ' ') { op[i++] = s[0] | s[1] << 8; ...
if (s[3] <= ' ') { op[i++] = s[0] | s[1] << 8 | ...
if (s[4] <= ' ') { op[i++] = s[0] | s[1] << 8 | ...
break;
}
op[i] = 0;
return i;
}
a_static void a_Token0Table_ini(a_Token0Table *w)
{
static uint32_t op4[64], op3[64], op2[64], op1[64];
a_Token0Table_iniSub(op4, "");
a_Token0Table_iniSub(op3, ">>= <<= ...");
a_Token0Table_iniSub(op2, "== != <= >= << >> ++ -- +...
a_Token0Table_iniSub(op1, "= < > + - * / % ^ & | ! ~...
int c;
for (c = 0; c < 256; c++) {
unsigned char ct = 0;
if (c != 0) {
if (strchr(" \t\r\n", c) != NULL) ct = 0;
if (strchr("(){},;[]\"\'", c) != NULL) ct = 1;
if (strchr("!#%&-=^~:+*<>./?", c) != NULL) c...
if ('0' <= c && c <= '9') ct = 3;
if ('A' <= c && c <= 'Z') ct = 3;
if ('a' <= c && c <= 'z') ct = 3;
if (strchr("_$@`", c) != NULL) ct = 3;
}
w->chrTyp[c] = ct;
}
w->op[0] = op4;
w->op[1] = op3;
w->op[2] = op2;
w->op[3] = op1;
}
a_static char *a_Token0_get(a_Token0 *w)
{
const char *s = w->s, *s1 = w->s1;
unsigned char ct, *chrTyp = w->tbl->chrTyp;
intptr_t len = 0, i; w->c = 0;
for (;;) {
if (s >= s1) goto fin;
ct = chrTyp[*(const unsigned char *) s];
if (ct == 0) { s++; continue; }
break;
}
if (ct == 1) {
len = 1;
w->c = *s;
goto fin;
}
if (ct == 2) {
unsigned char ch0 = *s, ch1 = 0, ch2 = 0, ch3 = 0;
if (s + 3 < s1) {
ch1 = s[1]; ch2 = s[2]; ch3 = s[3];
} else {
if (s + 1 < s1) ch1 = s[1];
if (s + 2 < s1) ch2 = s[2];
}
uint32_t op = ch0 | ch1 << 8 | ch2 << 16 | ch3 <...
opTbl = w->tbl->op[0]; len = 4; w->c = op;
for (i = 0; opTbl[i] != 0; i++) {
if (opTbl[i] == op) goto fin;
}
opTbl = w->tbl->op[1]; len = 3; w->c = op &= 0x0...
for (i = 0; opTbl[i] != 0; i++) {
if (opTbl[i] == op) goto fin;
}
opTbl = w->tbl->op[2]; len = 2; w->c = op &= 0x0...
for (i = 0; opTbl[i] != 0; i++) {
if (opTbl[i] == op) goto fin;
}
opTbl = w->tbl->op[3]; len = 1; w->c = op &= 0x0...
for (i = 0; opTbl[i] != 0; i++) {
if (opTbl[i] == op) goto fin;
}
ct = 255; len = 0; w->c = 0;
}
if (ct == 3) {
w->c = *s;
for (len = 1;;) {
if (s + len >= s1) goto fin;
if (chrTyp[((unsigned char *) s)[len]] == ct...
break;
}
}
fin:
w->cTyp = ct;
w->len = len;
w->s = s + len;
return (char *) s;
}
a_static char *a_Token1_get(a_Token0 *w)
{
for (;;) {
const char *s = a_Token0_get(w), *s1 = w->s1, *t;
uint32_t c = w->c;
if (s >= s1) return (char *) s;
if (c == ('/' | '/' << 8)) {
t = memchr(s, '\n', s1 - s);
if (t != NULL) { w->s = t; continue; }
w->s = s1; w->len = 0; return (char *) s1;
}
if (c == 0x22 || c == 0x27) {
// 改行もしくはエスケープされていないcがでる...
t = s + 1;
for (;;) {
if (t >= s1) break;
if (*(unsigned char *) t == c) { t++; br...
if (*t == '\n') break;
if (*t != '\\') { t++; continue; }
t++;
if (t >= s1) break;
if (*t == '\n') break;
t++;
}
w->len = t - s;
w->s = t;
return (char *) s;
}
if ('0' <= c && c <= '9') {
t = s + w->len;
for (;;) {
if (t >= s1) break;
c = *t;
if (c == '.') { t++; continue; }
if ('0' <= c && c <= '9') { t++; continu...
if ('A' <= c && c <= 'Z') { t++; continu...
if ('a' <= c && c <= 'z') { t++; continu...
if (c == '_') { t++; continue; }
if ((c == '+' || c == '-') && t[-1] == '...
break;
}
w->len = t - s;
w->s = t;
return (char *) s;
}
return (char *) s;
}
}
** (2) Token1_get
-与えられた文字列から先頭のトークンを切り出す関数です。連...
-C言語にほぼ準拠していますが、[[K]]の好みで余計な演算子も...
-このバージョンでは /* ～ */ 形式のコメントに対応していま...
-プログラムの仕組みとしては Token0 と Token1 の二層構造に...
-t0006a.cの実行結果を見れば、大体どんな感じかわかると思い...
** (3) サンプルプログラム: t0006a.c
-acl4ライブラリは以下のようにして利用します。
-[1]まずa4_0001～a4_0006のプログラムをつなげて"acl4.c"と...
-[2]次にacl4を使ったプログラムを書きます。
-[3]コンパイル時に、"acl4.c"のあるパスをインクルードパス...
#define a_Version 1
#include <acl4.c>
int main(int argc, const char **argv)
{
VecChr vc[1]; VecChr_ini(vc);
VecChr_readFileAll_errChk(vc, argv[1]);
VecChr_eraseCr(vc);
Token0Table t0t[1]; Token0Table_ini(t0t);
Token0 t0[1]; t0->tbl = t0t;
t0->s = vc->p; t0->s1 = vc->p + vc->n;
int i = 1;
for (;;) {
char *t = Token1_get(t0);
if (t0->len == 0 || i > 20) break;
printf("%6d: %.*s\n", i++, (int) t0->len, t);
}
VecChr_din(vc);
a_malloc_debugList(_arg);
return 0;
}
>t0006a t0006a.c
1: #
2: define
3: a_Version
4: 1
5: #
6: include
7: <
8: acl4
9: .
10: c
11: >
12: int
13: main
14: (
15: int
16: argc
17: ,
18: const
19: char
20: *
t0006a.c(22): malloc_debugList()
** (99) 更新履歴
-2026.02.02(月) 初版
ページ名:
![[PukiWiki]](image/pukiwiki.png "[PukiWiki]")
