d0p1/ack
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

Run files through clang-format before editing.

Browse source
This commit is contained in:
David Given 2018-09-02 11:16:06 +02:00
parent bed39e1f4d
commit 788f439a48
2 changed files with 1112 additions and 964 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

481
lang/cem/cemcom.ansi/LLlex.c
View file

@ -22,10 +22,10 @@
/* Data about the token yielded */ /* Data about the token yielded */
struct token dot, ahead, aside; struct token dot, ahead, aside;
int token_nmb = 0; /* number of the ahead token */ int token_nmb = 0; /* number of the ahead token */
int tk_nmb_at_last_syn_err = -5/*ERR_SHADOW*/; int tk_nmb_at_last_syn_err = -5 /*ERR_SHADOW*/;
/* token number at last syntax error */ /* token number at last syntax error */
int idfsize = IDFSIZE; int idfsize = IDFSIZE;
char sp_occurred[SP_TOTAL+1]; char sp_occurred[SP_TOTAL + 1];
#ifndef NOPP #ifndef NOPP
int ReplaceMacros = 1; /* replacing macros */ int ReplaceMacros = 1; /* replacing macros */
@ -74,18 +74,19 @@ PopLex()
} }
#endif /* NOPP */ #endif /* NOPP */
int int LLlex()
LLlex()
{ {
/* LLlex() plays the role of Lexical Analyzer for the C parser. /* LLlex() plays the role of Lexical Analyzer for the C parser.
The look-ahead and putting aside of tokens are taken into The look-ahead and putting aside of tokens are taken into
account. account.
*/ */
if (ASIDE) { /* a token is put aside */ if (ASIDE)
{ /* a token is put aside */
dot = aside; dot = aside;
ASIDE = 0; ASIDE = 0;
} }
else { /* read ahead and return the old one */ else
{ /* read ahead and return the old one */
#ifdef LINT #ifdef LINT
lint_comment_ahead(); lint_comment_ahead();
#endif /* LINT */ #endif /* LINT */
@ -104,13 +105,10 @@ LLlex()
return DOT; return DOT;
} }
char* string_token();
char *string_token();
arith char_constant(); arith char_constant();
int int GetToken(ptok) register struct token* ptok;
GetToken(ptok)
register struct token *ptok;
{ {
/* GetToken() is the actual token recognizer. It calls the /* GetToken() is the actual token recognizer. It calls the
control line interpreter if it encounters a "\n{w}*#" control line interpreter if it encounters a "\n{w}*#"
@ -122,7 +120,8 @@ GetToken(ptok)
token_nmb++; token_nmb++;
if (File_Inserted) { if (File_Inserted)
{
File_Inserted = 0; File_Inserted = 0;
goto firstline; goto firstline;
} }
@ -138,9 +137,10 @@ go_on: /* rescan, the following character has been read */
ptok->tk_file = FileName; ptok->tk_file = FileName;
ptok->tk_line = LineNumber; ptok->tk_line = LineNumber;
switch (class(ch)) { /* detect character class */ switch (class(ch))
{ /* detect character class */
case STNL: /* newline, vertical space or formfeed */ case STNL: /* newline, vertical space or formfeed */
firstline: firstline:
LineNumber++; /* also at vs and ff */ LineNumber++; /* also at vs and ff */
ptok->tk_file = FileName; ptok->tk_file = FileName;
ptok->tk_line = LineNumber; ptok->tk_line = LineNumber;
@ -150,25 +150,32 @@ firstline:
*/ */
return ptok->tk_symb = EOI; return ptok->tk_symb = EOI;
while ((ch = GetChar()), while ((ch = GetChar()), (ch == '#'
(ch == '#'
#ifndef NOPP #ifndef NOPP
|| ch == '/' || ch == '/'
#endif #endif
|| class(ch) == STSKIP)) { || class(ch) == STSKIP))
{
/* blanks are allowed before hashes */ /* blanks are allowed before hashes */
if (ch == '#') { if (ch == '#')
{
/* a control line follows */ /* a control line follows */
domacro(); domacro();
#ifndef NOPP #ifndef NOPP
if (File_Inserted) { if (File_Inserted)
{
File_Inserted = 0; File_Inserted = 0;
goto firstline; goto firstline;
} }
} else if (ch == '/') { }
if ((GetChar() == '*') && !InputLevel) { else if (ch == '/')
{
if ((GetChar() == '*') && !InputLevel)
{
skipcomment(); skipcomment();
} else { }
else
{
UnGetChar(); UnGetChar();
break; break;
} }
@ -184,11 +191,14 @@ firstline:
goto again; goto again;
case STGARB: /* garbage character */ case STGARB: /* garbage character */
#ifndef NOPP #ifndef NOPP
garbage: garbage:
#endif #endif
if (040 < ch && ch < 0177) { if (040 < ch && ch < 0177)
{
return ptok->tk_symb = ch; return ptok->tk_symb = ch;
} else { }
else
{
lexerror("garbage char \\%03o", ch); lexerror("garbage char \\%03o", ch);
} }
goto again; goto again;
@ -196,7 +206,8 @@ garbage:
return ptok->tk_symb = ch; return ptok->tk_symb = ch;
case STCOMP: /* maybe the start of a compound token */ case STCOMP: /* maybe the start of a compound token */
nch = GetChar(); /* character lookahead */ nch = GetChar(); /* character lookahead */
switch (ch) { switch (ch)
{
case '!': case '!':
if (nch == '=') if (nch == '=')
return ptok->tk_symb = NOTEQUAL; return ptok->tk_symb = NOTEQUAL;
@ -222,13 +233,14 @@ garbage:
return ptok->tk_symb = MINAB; return ptok->tk_symb = MINAB;
break; break;
case '<': case '<':
if (AccFileSpecifier) { if (AccFileSpecifier)
{
UnGetChar(); /* pushback nch */ UnGetChar(); /* pushback nch */
ptok->tk_bts = string_token("file specifier", ptok->tk_bts = string_token("file specifier", '>', &(ptok->tk_len));
'>', &(ptok->tk_len));
return ptok->tk_symb = FILESPECIFIER; return ptok->tk_symb = FILESPECIFIER;
} }
if (nch == '<') { if (nch == '<')
{
if ((nch = GetChar()) == '=') if ((nch = GetChar()) == '=')
return ptok->tk_symb = LEFTAB; return ptok->tk_symb = LEFTAB;
UnGetChar(); UnGetChar();
@ -244,7 +256,8 @@ garbage:
case '>': case '>':
if (nch == '=') if (nch == '=')
return ptok->tk_symb = GREATEREQ; return ptok->tk_symb = GREATEREQ;
if (nch == '>') { if (nch == '>')
{
if ((nch = GetChar()) == '=') if ((nch = GetChar()) == '=')
return ptok->tk_symb = RIGHTAB; return ptok->tk_symb = RIGHTAB;
UnGetChar(); UnGetChar();
@ -271,7 +284,8 @@ garbage:
break; break;
case '/': case '/':
#ifndef NOPP #ifndef NOPP
if (nch == '*' && !InputLevel) { if (nch == '*' && !InputLevel)
{
skipcomment(); skipcomment();
goto again; goto again;
} }
@ -295,12 +309,14 @@ garbage:
return ptok->tk_symb = STRING; return ptok->tk_symb = STRING;
case STELL: /* wide character constant/string prefix */ case STELL: /* wide character constant/string prefix */
nch = GetChar(); nch = GetChar();
if (nch == '"') { if (nch == '"')
ptok->tk_bts = string_token("wide character string", {
'"', &(ptok->tk_len)); ptok->tk_bts = string_token("wide character string", '"', &(ptok->tk_len));
ptok->tk_fund = WCHAR; /* string of wide characters */ ptok->tk_fund = WCHAR; /* string of wide characters */
return ptok->tk_symb = STRING; return ptok->tk_symb = STRING;
} else if (nch == '\'') { }
else if (nch == '\'')
{
ptok->tk_ival = char_constant("wide character"); ptok->tk_ival = char_constant("wide character");
ptok->tk_fund = INT; ptok->tk_fund = INT;
return ptok->tk_symb = INTEGER; return ptok->tk_symb = INTEGER;
@ -309,21 +325,25 @@ garbage:
/* fallthrough */ /* fallthrough */
case STIDF: case STIDF:
{ {
register char *tg = &buf[0]; register char* tg = &buf[0];
register int pos = -1; register int pos = -1;
register struct idf *idef; register struct idf* idef;
extern int idfsize; /* ??? */ extern int idfsize; /* ??? */
#ifndef NOPP #ifndef NOPP
int NoExpandNext = 0; int NoExpandNext = 0;
if (Unstacked) EnableMacros(); /* unstack macro's when allowed. */ if (Unstacked)
if (ch == NOEXPM) { EnableMacros(); /* unstack macro's when allowed. */
if (ch == NOEXPM)
{
NoExpandNext = 1; NoExpandNext = 1;
ch = GetChar(); ch = GetChar();
} }
#endif #endif
do { /* read the identifier */ do
if (++pos < idfsize) { { /* read the identifier */
if (++pos < idfsize)
{
*tg++ = ch; *tg++ = ch;
} }
ch = GetChar(); ch = GetChar();
@ -337,47 +357,52 @@ garbage:
idef->id_file = ptok->tk_file; idef->id_file = ptok->tk_file;
idef->id_line = ptok->tk_line; idef->id_line = ptok->tk_line;
#ifndef NOPP #ifndef NOPP
if (idef->id_macro && ReplaceMacros && !NoExpandNext) { if (idef->id_macro && ReplaceMacros && !NoExpandNext)
{
if (replace(idef)) if (replace(idef))
goto again; goto again;
} }
if (UnknownIdIsZero && idef->id_reserved != SIZEOF) { if (UnknownIdIsZero && idef->id_reserved != SIZEOF)
{
ptok->tk_ival = (arith)0; ptok->tk_ival = (arith)0;
ptok->tk_fund = INT; ptok->tk_fund = INT;
return ptok->tk_symb = INTEGER; return ptok->tk_symb = INTEGER;
} }
#endif /* NOPP */ #endif /* NOPP */
ptok->tk_symb = ( ptok->tk_symb
idef->id_reserved = (idef->id_reserved
? idef->id_reserved ? idef->id_reserved
: idef->id_def && idef->id_def->df_sc == TYPEDEF : idef->id_def && idef->id_def->df_sc == TYPEDEF ? TYPE_IDENTIFIER
? TYPE_IDENTIFIER : IDENTIFIER);
: IDENTIFIER
);
return IDENTIFIER; return IDENTIFIER;
} }
case STNUM: /* a numeric constant */ case STNUM: /* a numeric constant */
{ {
register int siz_left = NUMSIZE - 1; register int siz_left = NUMSIZE - 1;
register char *np = &buf[0]; register char* np = &buf[0];
int flags = 0; int flags = 0;
#define store(ch) if (--siz_left >= 0) \ #define store(ch) \
if (--siz_left >= 0) \
*np++ = ch; *np++ = ch;
if (ch == '.') { if (ch == '.')
{
/* An embarrasing ambiguity. We have either a /* An embarrasing ambiguity. We have either a
pp-number, a field operator, an ELLIPSIS or pp-number, a field operator, an ELLIPSIS or
an error (..). an error (..).
*/ */
ch = GetChar(); ch = GetChar();
if (!is_dig(ch)) { /* . or ... */ if (!is_dig(ch))
if (ch == '.') { { /* . or ... */
if (ch == '.')
{
if ((ch = GetChar()) == '.') if ((ch = GetChar()) == '.')
return ptok->tk_symb = ELLIPSIS; return ptok->tk_symb = ELLIPSIS;
UnGetChar(); /* not '.' */ UnGetChar(); /* not '.' */
ChPushBack('.'); /* sigh ... */ ChPushBack('.'); /* sigh ... */
} else }
else
UnGetChar(); /* not '.' */ UnGetChar(); /* not '.' */
return ptok->tk_symb = '.'; return ptok->tk_symb = '.';
} }
@ -387,31 +412,37 @@ garbage:
} }
store(ch); store(ch);
ch = GetChar(); ch = GetChar();
while(in_idf(ch) || ch == '.') { while (in_idf(ch) || ch == '.')
{
store(ch); store(ch);
if (ch == '.') flags |= FLG_DOTSEEN; if (ch == '.')
if (ch == 'e' || ch == 'E') { flags |= FLG_DOTSEEN;
if (ch == 'e' || ch == 'E')
{
flags |= FLG_ESEEN; flags |= FLG_ESEEN;
ch = GetChar(); ch = GetChar();
if (ch == '+' || ch == '-') { if (ch == '+' || ch == '-')
{
flags |= FLG_DOTSEEN; /* trick */ flags |= FLG_DOTSEEN; /* trick */
store(ch); store(ch);
ch = GetChar(); ch = GetChar();
} }
} else ch = GetChar(); }
else
ch = GetChar();
} }
store('\0'); store('\0');
UnGetChar(); UnGetChar();
np = &buf[0]; np = &buf[0];
ch = *np++; ch = *np++;
if (siz_left < 0) { if (siz_left < 0)
{
lexerror("number too long"); lexerror("number too long");
if ((flags & FLG_DOTSEEN) if ((flags & FLG_DOTSEEN)
|| (flags & FLG_ESEEN || (flags & FLG_ESEEN && !(ch == '0' && (*np == 'x' || *np == 'X'))))
&& !(ch == '0' {
&& (*np == 'x' || *np == 'X')))) { ptok->tk_fval = Salloc("0.0", (unsigned)4);
ptok->tk_fval = Salloc("0.0", (unsigned) 4);
ptok->tk_fund = DOUBLE; ptok->tk_fund = DOUBLE;
return ptok->tk_symb = FLOATING; return ptok->tk_symb = FLOATING;
} }
@ -421,8 +452,8 @@ garbage:
} }
/* Now, the pp-number must be converted into a token */ /* Now, the pp-number must be converted into a token */
if ((flags & FLG_DOTSEEN) if ((flags & FLG_DOTSEEN)
|| (flags & FLG_ESEEN || (flags & FLG_ESEEN && !(ch == '0' && (*np == 'x' || *np == 'X'))))
&& !(ch == '0' && (*np == 'x' || *np == 'X')))) { {
strflt2tok(&buf[0], ptok); strflt2tok(&buf[0], ptok);
return ptok->tk_symb = FLOATING; return ptok->tk_symb = FLOATING;
} }
@ -433,9 +464,11 @@ garbage:
return ptok->tk_symb = EOI; return ptok->tk_symb = EOI;
#ifndef NOPP #ifndef NOPP
case STMSPEC: case STMSPEC:
if (!InputLevel) goto garbage; if (!InputLevel)
if (ch == TOKSEP) goto again; goto garbage;
/* fallthrough shouldn't happen */ if (ch == TOKSEP)
goto again;
/* fallthrough shouldn't happen */
#endif #endif
default: /* this cannot happen */ default: /* this cannot happen */
crash("bad class for char 0%o", ch); crash("bad class for char 0%o", ch);
@ -444,8 +477,7 @@ garbage:
} }
#ifndef NOPP #ifndef NOPP
void void skipcomment()
skipcomment()
{ {
/* The last character read has been the '*' of '/_*'. The /* The last character read has been the '*' of '/_*'. The
characters, except NL and EOI, between '/_*' and the first characters, except NL and EOI, between '/_*' and the first
@ -462,46 +494,54 @@ skipcomment()
NoUnstack++; NoUnstack++;
c = GetChar(); c = GetChar();
#ifdef LINT #ifdef LINT
if (! lint_skip_comment) { if (!lint_skip_comment)
{
lint_start_comment(); lint_start_comment();
lint_comment_char(c); lint_comment_char(c);
} }
#endif /* LINT */ #endif /* LINT */
do { do
while (c != '*') { {
if (class(c) == STNL) { while (c != '*')
{
if (class(c) == STNL)
{
++LineNumber; ++LineNumber;
} else if (c == EOI) { }
else if (c == EOI)
{
NoUnstack--; NoUnstack--;
#ifdef LINT #ifdef LINT
if (! lint_skip_comment) lint_end_comment(); if (!lint_skip_comment)
lint_end_comment();
#endif /* LINT */ #endif /* LINT */
return; return;
} }
oldc = c; oldc = c;
c = GetChar(); c = GetChar();
#ifdef LINT #ifdef LINT
if (! lint_skip_comment) lint_comment_char(c); if (!lint_skip_comment)
lint_comment_char(c);
#endif /* LINT */ #endif /* LINT */
} /* last Character seen was '*' */ } /* last Character seen was '*' */
c = GetChar(); c = GetChar();
if ( c != '/' && oldc == '/') if (c != '/' && oldc == '/')
lexwarning("comment inside comment ?"); lexwarning("comment inside comment ?");
oldc = '*'; oldc = '*';
#ifdef LINT #ifdef LINT
if (! lint_skip_comment) lint_comment_char(c); if (!lint_skip_comment)
lint_comment_char(c);
#endif /* LINT */ #endif /* LINT */
} while (c != '/'); } while (c != '/');
#ifdef LINT #ifdef LINT
if (! lint_skip_comment) lint_end_comment(); if (!lint_skip_comment)
lint_end_comment();
#endif /* LINT */ #endif /* LINT */
NoUnstack--; NoUnstack--;
} }
#endif /* NOPP */ #endif /* NOPP */
arith arith char_constant(nm) char* nm;
char_constant(nm)
char *nm;
{ {
register arith val = 0; register arith val = 0;
register int ch; register int ch;
@ -511,15 +551,18 @@ char_constant(nm)
if (ch == '\'') if (ch == '\'')
lexerror("%s constant too short", nm); lexerror("%s constant too short", nm);
else else
while (ch != '\'') { while (ch != '\'')
if (ch == '\n') { {
if (ch == '\n')
{
lexerror("newline in %s constant", nm); lexerror("newline in %s constant", nm);
LineNumber++; LineNumber++;
break; break;
} }
if (ch == '\\') if (ch == '\\')
ch = quoted(GetChar()); ch = quoted(GetChar());
if (ch >= 128) ch -= 256; if (ch >= 128)
ch -= 256;
if (size < (int)int_size) if (size < (int)int_size)
val |= ch << 8 * size; val |= ch << 8 * size;
size++; size++;
@ -532,24 +575,25 @@ char_constant(nm)
return val; return val;
} }
char * char* string_token(nm, stop_char, plen) char* nm;
string_token(nm, stop_char, plen) int* plen;
char *nm;
int *plen;
{ {
register int ch; register int ch;
register int str_size; register int str_size;
register char *str = Malloc((unsigned) (str_size = ISTRSIZE)); register char* str = Malloc((unsigned)(str_size = ISTRSIZE));
register int pos = 0; register int pos = 0;
ch = GetChar(); ch = GetChar();
while (ch != stop_char) { while (ch != stop_char)
if (ch == '\n') { {
if (ch == '\n')
{
lexerror("newline in %s", nm); lexerror("newline in %s", nm);
LineNumber++; LineNumber++;
break; break;
} }
if (ch == EOI) { if (ch == EOI)
{
lexerror("end-of-file inside %s", nm); lexerror("end-of-file inside %s", nm);
break; break;
} }
@ -557,7 +601,7 @@ string_token(nm, stop_char, plen)
ch = quoted(GetChar()); ch = quoted(GetChar());
str[pos++] = ch; str[pos++] = ch;
if (pos == str_size) if (pos == str_size)
str = Realloc(str, (unsigned) (str_size += RSTRSIZE)); str = Realloc(str, (unsigned)(str_size += RSTRSIZE));
ch = GetChar(); ch = GetChar();
} }
str[pos++] = '\0'; /* for filenames etc. */ str[pos++] = '\0'; /* for filenames etc. */
@ -565,16 +609,16 @@ string_token(nm, stop_char, plen)
return str; return str;
} }
int int quoted(ch) register int ch;
quoted(ch)
register int ch;
{ {
/* quoted() replaces an escaped character sequence by the /* quoted() replaces an escaped character sequence by the
character meant. character meant.
*/ */
/* first char after backslash already in ch */ /* first char after backslash already in ch */
if (!is_oct(ch)) { /* a quoted char */ if (!is_oct(ch))
switch (ch) { { /* a quoted char */
switch (ch)
{
case 'n': case 'n':
ch = '\n'; ch = '\n';
break; break;
@ -601,7 +645,8 @@ quoted(ch)
register int hex = 0; register int hex = 0;
register int vch; register int vch;
for (;;) { for (;;)
{
ch = GetChar(); ch = GetChar();
if ((vch = hex_val(ch)) == -1) if ((vch = hex_val(ch)) == -1)
break; break;
@ -612,32 +657,27 @@ quoted(ch)
} }
} }
} }
else { /* a quoted octal */ else
{ /* a quoted octal */
register int oct = 0, cnt = 0; register int oct = 0, cnt = 0;
do { do
oct = oct*8 + (ch-'0'); {
oct = oct * 8 + (ch - '0');
ch = GetChar(); ch = GetChar();
} while (is_oct(ch) && ++cnt < 3); } while (is_oct(ch) && ++cnt < 3);
UnGetChar(); UnGetChar();
ch = oct; ch = oct;
} }
return ch&0377; return ch & 0377;
} }
int hex_val(ch) register int ch;
int
hex_val(ch)
register int ch;
{ {
return is_dig(ch) ? ch - '0' return is_dig(ch) ? ch - '0' : is_hex(ch) ? (ch - 'a' + 10) & 017 : -1;
: is_hex(ch) ? (ch - 'a' + 10) & 017
: -1;
} }
int GetChar()
int
GetChar()
{ {
/* The routines GetChar and trigraph parses the trigraph /* The routines GetChar and trigraph parses the trigraph
sequences and removes occurences of \\\n. sequences and removes occurences of \\\n.
@ -655,9 +695,11 @@ again:
ch = trigraph(); ch = trigraph();
/* \<newline> is removed from the input stream */ /* \<newline> is removed from the input stream */
if (ch == '\\') { if (ch == '\\')
{
LoadChar(ch); LoadChar(ch);
if (ch == '\n') { if (ch == '\n')
{
++LineNumber; ++LineNumber;
goto again; goto again;
} }
@ -665,113 +707,133 @@ again:
ch = '\\'; ch = '\\';
} }
#endif #endif
return(LexSave = ch); return (LexSave = ch);
} }
#ifndef NOPP #ifndef NOPP
int int trigraph()
trigraph()
{ {
register int ch; register int ch;
LoadChar(ch); LoadChar(ch);
if (ch == '?') { if (ch == '?')
{
LoadChar(ch); LoadChar(ch);
switch (ch) { /* its a trigraph */ switch (ch)
{ /* its a trigraph */
case '=': case '=':
ch = '#'; ch = '#';
return(ch); return (ch);
case '(': case '(':
ch = '['; ch = '[';
return(ch); return (ch);
case '/': case '/':
ch = '\\'; ch = '\\';
return(ch); return (ch);
case ')': case ')':
ch = ']'; ch = ']';
return(ch); return (ch);
case '\'': case '\'':
ch = '^'; ch = '^';
return(ch); return (ch);
case '<': case '<':
ch = '{'; ch = '{';
return(ch); return (ch);
case '!': case '!':
ch = '|'; ch = '|';
return(ch); return (ch);
case '>': case '>':
ch = '}'; ch = '}';
return(ch); return (ch);
case '-': case '-':
ch = '~'; ch = '~';
return(ch); return (ch);
} }
PushBack(); PushBack();
} }
PushBack(); PushBack();
return('?'); return ('?');
} }
#endif #endif
/* strflt2tok only checks the syntax of the floating-point number and /* strflt2tok only checks the syntax of the floating-point number and
* selects the right type for the number. * selects the right type for the number.
*/ */
strflt2tok(fltbuf, ptok) strflt2tok(fltbuf, ptok) char fltbuf[];
char fltbuf[]; struct token* ptok;
struct token *ptok;
{ {
register char *cp = fltbuf; register char* cp = fltbuf;
int malformed = 0; int malformed = 0;
while (is_dig(*cp)) cp++; while (is_dig(*cp))
if (*cp == '.') { cp++;
if (*cp == '.')
{
cp++;
while (is_dig(*cp))
cp++; cp++;
while (is_dig(*cp)) cp++;
} }
if (*cp == 'e' || *cp == 'E') { if (*cp == 'e' || *cp == 'E')
{
cp++; cp++;
if (*cp == '+' || *cp == '-') if (*cp == '+' || *cp == '-')
cp++; cp++;
if (!is_dig(*cp)) malformed++; if (!is_dig(*cp))
while (is_dig(*cp)) cp++; malformed++;
while (is_dig(*cp))
cp++;
} }
if (*cp == 'f' || *cp == 'F') { if (*cp == 'f' || *cp == 'F')
if (*(cp + 1)) malformed++; {
if (*(cp + 1))
malformed++;
*cp = '\0'; *cp = '\0';
ptok->tk_fund = FLOAT; ptok->tk_fund = FLOAT;
} else if (*cp == 'l' || *cp == 'L') { }
if (*(cp + 1)) malformed++; else if (*cp == 'l' || *cp == 'L')
{
if (*(cp + 1))
malformed++;
*cp = '\0'; *cp = '\0';
ptok->tk_fund = LNGDBL; ptok->tk_fund = LNGDBL;
} else { }
if (*cp) malformed++; else
{
if (*cp)
malformed++;
ptok->tk_fund = DOUBLE; ptok->tk_fund = DOUBLE;
} }
if (malformed) { if (malformed)
{
lexerror("malformed floating constant"); lexerror("malformed floating constant");
ptok->tk_fval = Salloc("0.0", (unsigned) 4); ptok->tk_fval = Salloc("0.0", (unsigned)4);
} else { }
ptok->tk_fval = Salloc(fltbuf, (unsigned) (cp - fltbuf + 1)); else
{
ptok->tk_fval = Salloc(fltbuf, (unsigned)(cp - fltbuf + 1));
} }
} }
strint2tok(intbuf, ptok) strint2tok(intbuf, ptok) char intbuf[];
char intbuf[]; struct token* ptok;
struct token *ptok;
{ {
register char *cp = intbuf; register char* cp = intbuf;
int base = 10; int base = 10;
arith val = 0, dig, ubound; arith val = 0, dig, ubound;
int uns_flg = 0, lng_flg = 0, malformed = 0, ovfl = 0; int uns_flg = 0, lng_flg = 0, malformed = 0, ovfl = 0;
int fund; int fund;
assert(*cp != '-'); assert(*cp != '-');
if (*cp == '0') { if (*cp == '0')
{
cp++; cp++;
if (*cp == 'x' || *cp == 'X') { if (*cp == 'x' || *cp == 'X')
{
cp++; cp++;
base = 16; base = 16;
} else base = 8; }
else
base = 8;
} }
/* The upperbound will be the same as when computed with /* The upperbound will be the same as when computed with
* max_unsigned_arith / base (since base is even). The problem here * max_unsigned_arith / base (since base is even). The problem here
@ -779,68 +841,99 @@ struct token *ptok;
*/ */
ubound = max_arith / (base / 2); ubound = max_arith / (base / 2);
while (is_hex(*cp)) { while (is_hex(*cp))
{
dig = hex_val(*cp); dig = hex_val(*cp);
if (dig >= base) { if (dig >= base)
{
malformed++; /* ignore */ malformed++; /* ignore */
} }
else { else
if (val < 0 || val > ubound) ovfl++; {
if (val < 0 || val > ubound)
ovfl++;
val *= base; val *= base;
if (val < 0 && val + dig >= 0) ovfl++; if (val < 0 && val + dig >= 0)
ovfl++;
val += dig; val += dig;
} }
cp++; cp++;
} }
while (*cp) { while (*cp)
if (*cp == 'l' || *cp == 'L') lng_flg++; {
else if (*cp == 'u' || *cp == 'U') uns_flg++; if (*cp == 'l' || *cp == 'L')
else break; lng_flg++;
else if (*cp == 'u' || *cp == 'U')
uns_flg++;
else
break;
cp++; cp++;
} }
if (*cp) { if (*cp)
{
malformed++; malformed++;
} }
if (malformed) { if (malformed)
lexerror("malformed %s integer constant", {
(base == 10 ? "decimal" lexerror(
: (base == 8 ? "octal" "malformed %s integer constant",
: "hexadecimal"))); (base == 10 ? "decimal" : (base == 8 ? "octal" : "hexadecimal")));
} else { }
else
{
if (lng_flg > 1) if (lng_flg > 1)
lexerror("only one long suffix allowed"); lexerror("only one long suffix allowed");
if (uns_flg > 1) if (uns_flg > 1)
lexerror("only one unsigned suffix allowed"); lexerror("only one unsigned suffix allowed");
} }
if (ovfl) { if (ovfl)
{
lexwarning("overflow in constant"); lexwarning("overflow in constant");
fund = ULONG; fund = ULONG;
} else if (!lng_flg && (val & full_mask[(int)int_size]) == val) { }
if (val >= 0 && val <= max_int) { else if (!lng_flg && (val & full_mask[(int)int_size]) == val)
{
if (val >= 0 && val <= max_int)
{
fund = INT; fund = INT;
} else if (int_size == long_size) { }
else if (int_size == long_size)
{
fund = UNSIGNED; fund = UNSIGNED;
} else if (base == 10 && !uns_flg) }
else if (base == 10 && !uns_flg)
fund = LONG; fund = LONG;
else fund = UNSIGNED; else
} else if((val & full_mask[(int)long_size]) == val) { fund = UNSIGNED;
if (val >= 0) fund = LONG; }
else fund = ULONG; else if ((val & full_mask[(int)long_size]) == val)
} else { /* sizeof(arith) is greater than long_size */ {
if (val >= 0)
fund = LONG;
else
fund = ULONG;
}
else
{ /* sizeof(arith) is greater than long_size */
assert(arith_size > long_size); assert(arith_size > long_size);
lexwarning("constant too large for target machine"); lexwarning("constant too large for target machine");
/* cut the size to prevent further complaints */ /* cut the size to prevent further complaints */
val &= full_mask[(int)long_size]; val &= full_mask[(int)long_size];
fund = ULONG; fund = ULONG;
} }
if (lng_flg) { if (lng_flg)
{
/* fund can't be INT */ /* fund can't be INT */
if (fund == UNSIGNED) fund = ULONG; if (fund == UNSIGNED)
fund = ULONG;
} }
if (uns_flg) { if (uns_flg)
if (fund == INT) fund = UNSIGNED; {
else if (fund == LONG) fund = ULONG; if (fund == INT)
fund = UNSIGNED;
else if (fund == LONG)
fund = ULONG;
} }
ptok->tk_fund = fund; ptok->tk_fund = fund;
ptok->tk_ival = val; ptok->tk_ival = val;

295
lang/cem/cpp.ansi/LLlex.c
View file

@ -30,23 +30,19 @@ int AccFileSpecifier = 0; /* return filespecifier <...> */
int LexSave = 0; /* last character read by GetChar */ int LexSave = 0; /* last character read by GetChar */
extern int InputLevel; /* # of current macro expansions */ extern int InputLevel; /* # of current macro expansions */
extern char *string_token(); extern char* string_token();
extern arith char_constant(); extern arith char_constant();
#define FLG_ESEEN 0x01 /* possibly a floating point number */ #define FLG_ESEEN 0x01 /* possibly a floating point number */
#define FLG_DOTSEEN 0x02 /* certainly a floating point number */ #define FLG_DOTSEEN 0x02 /* certainly a floating point number */
void skipcomment(); void skipcomment();
int int LLlex()
LLlex()
{ {
return (DOT != EOF) ? GetToken(&dot) : EOF; return (DOT != EOF) ? GetToken(&dot) : EOF;
} }
int GetToken(ptok) register struct token* ptok;
int
GetToken(ptok)
register struct token *ptok;
{ {
/* GetToken() is the actual token recognizer. It calls the /* GetToken() is the actual token recognizer. It calls the
control line interpreter if it encounters a "\n{w}*#" control line interpreter if it encounters a "\n{w}*#"
@ -63,14 +59,15 @@ again: /* rescan the input after an error or replacement */
fatal("non-ascii '\\%03o' read", ch & 0377); fatal("non-ascii '\\%03o' read", ch & 0377);
/* keep track of the place of the token in the file */ /* keep track of the place of the token in the file */
switch (class(ch)) { /* detect character class */ switch (class(ch))
{ /* detect character class */
case STNL: /* newline, vertical space or formfeed */ case STNL: /* newline, vertical space or formfeed */
LineNumber++; LineNumber++;
return ptok->tk_symb = EOF; return ptok->tk_symb = EOF;
case STSKIP: /* just skip the skip characters */ case STSKIP: /* just skip the skip characters */
goto again; goto again;
case STGARB: /* garbage character */ case STGARB: /* garbage character */
garbage: garbage:
if (040 < ch && ch < 0177) if (040 < ch && ch < 0177)
error("garbage char %c", ch); error("garbage char %c", ch);
else else
@ -80,7 +77,8 @@ garbage:
return ptok->tk_symb = ch; return ptok->tk_symb = ch;
case STCOMP: /* maybe the start of a compound token */ case STCOMP: /* maybe the start of a compound token */
nch = GetChar(); /* character lookahead */ nch = GetChar(); /* character lookahead */
switch (ch) { switch (ch)
{
case '!': case '!':
if (nch == '=') if (nch == '=')
return ptok->tk_symb = NOTEQUAL; return ptok->tk_symb = NOTEQUAL;
@ -110,17 +108,20 @@ garbage:
UnGetChar(); UnGetChar();
return ptok->tk_symb = ch; return ptok->tk_symb = ch;
case '<': case '<':
if (AccFileSpecifier) { if (AccFileSpecifier)
{
UnGetChar(); /* pushback nch */ UnGetChar(); /* pushback nch */
ptok->tk_str = ptok->tk_str = string_token("file specifier", '>');
string_token("file specifier", '>');
return ptok->tk_symb = FILESPECIFIER; return ptok->tk_symb = FILESPECIFIER;
} else if (nch == '<') { }
else if (nch == '<')
{
if ((nch = GetChar()) == '=') if ((nch = GetChar()) == '=')
return ptok->tk_symb = LEFTAB; return ptok->tk_symb = LEFTAB;
UnGetChar(); UnGetChar();
return ptok->tk_symb = LEFT; return ptok->tk_symb = LEFT;
} else if (nch == '=') }
else if (nch == '=')
return ptok->tk_symb = LESSEQ; return ptok->tk_symb = LESSEQ;
UnGetChar(); UnGetChar();
return ptok->tk_symb = ch; return ptok->tk_symb = ch;
@ -132,7 +133,8 @@ garbage:
case '>': case '>':
if (nch == '=') if (nch == '=')
return ptok->tk_symb = GREATEREQ; return ptok->tk_symb = GREATEREQ;
else if (nch == '>') { else if (nch == '>')
{
if ((nch = GetChar()) == '=') if ((nch = GetChar()) == '=')
return ptok->tk_symb = RIGHTAB; return ptok->tk_symb = RIGHTAB;
UnGetChar(); UnGetChar();
@ -163,7 +165,8 @@ garbage:
UnGetChar(); UnGetChar();
return ptok->tk_symb = ch; return ptok->tk_symb = ch;
case '/': case '/':
if (nch == '*' && !InputLevel) { if (nch == '*' && !InputLevel)
{
skipcomment(); skipcomment();
goto again; goto again;
} }
@ -183,11 +186,13 @@ garbage:
return ptok->tk_symb = STRING; return ptok->tk_symb = STRING;
case STELL: /* wide character constant/string prefix */ case STELL: /* wide character constant/string prefix */
nch = GetChar(); nch = GetChar();
if (nch == '"') { if (nch == '"')
ptok->tk_str = {
string_token("wide character string", '"'); ptok->tk_str = string_token("wide character string", '"');
return ptok->tk_symb = STRING; return ptok->tk_symb = STRING;
} else if (nch == '\'') { }
else if (nch == '\'')
{
ptok->tk_val = char_constant("wide character"); ptok->tk_val = char_constant("wide character");
return ptok->tk_symb = INTEGER; return ptok->tk_symb = INTEGER;
} }
@ -196,33 +201,57 @@ garbage:
case STIDF: case STIDF:
{ {
extern int idfsize; /* ??? */ extern int idfsize; /* ??? */
register char *tg = &buf[0]; register char* tg = &buf[0];
register char *maxpos = &buf[idfsize]; register char* maxpos = &buf[idfsize];
int NoExpandNext = 0; int NoExpandNext = 0;
#define tstmac(bx) if (!(bits[ch] & bx)) goto nomac #define tstmac(bx) \
if (!(bits[ch] & bx)) \
goto nomac
#define cpy *tg++ = ch #define cpy *tg++ = ch
#define load (ch = GetChar()); if (!in_idf(ch)) goto endidf #define load \
(ch = GetChar()); \
if (!in_idf(ch)) \
goto endidf
if (Unstacked) EnableMacros(); /* unstack macro's when allowed. */ if (Unstacked)
if (ch == NOEXPM) { EnableMacros(); /* unstack macro's when allowed. */
if (ch == NOEXPM)
{
NoExpandNext = 1; NoExpandNext = 1;
ch = GetChar(); ch = GetChar();
} }
#ifdef DOBITS #ifdef DOBITS
cpy; tstmac(bit0); load;
cpy; tstmac(bit1); load;
cpy; tstmac(bit2); load;
cpy; tstmac(bit3); load;
cpy; tstmac(bit4); load;
cpy; tstmac(bit5); load;
cpy; tstmac(bit6); load;
cpy; tstmac(bit7); load;
#endif
for(;;) {
if (tg < maxpos) {
cpy; cpy;
tstmac(bit0);
load;
cpy;
tstmac(bit1);
load;
cpy;
tstmac(bit2);
load;
cpy;
tstmac(bit3);
load;
cpy;
tstmac(bit4);
load;
cpy;
tstmac(bit5);
load;
cpy;
tstmac(bit6);
load;
cpy;
tstmac(bit7);
load;
#endif
for (;;)
{
if (tg < maxpos)
{
cpy;
} }
load; load;
} }
@ -230,10 +259,12 @@ garbage:
/*if (ch != EOI) UnGetChar();*/ /*if (ch != EOI) UnGetChar();*/
UnGetChar(); UnGetChar();
*tg++ = '\0'; /* mark the end of the identifier */ *tg++ = '\0'; /* mark the end of the identifier */
if (ReplaceMacros) { if (ReplaceMacros)
register struct idf *idef = findidf(buf); {
register struct idf* idef = findidf(buf);
if (idef && idef->id_macro && !NoExpandNext) { if (idef && idef->id_macro && !NoExpandNext)
{
if (replace(idef)) if (replace(idef))
goto again; goto again;
} }
@ -241,15 +272,18 @@ garbage:
nomac: /* buf can already be null-terminated. soit */ nomac: /* buf can already be null-terminated. soit */
ch = GetChar(); ch = GetChar();
while (in_idf(ch)) { while (in_idf(ch))
if (tg < maxpos) *tg++ = ch; {
if (tg < maxpos)
*tg++ = ch;
ch = GetChar(); ch = GetChar();
} }
UnGetChar(); UnGetChar();
*tg++ = '\0'; /* mark the end of the identifier */ *tg++ = '\0'; /* mark the end of the identifier */
NoExpandNext = 0; NoExpandNext = 0;
if (UnknownIdIsZero) { if (UnknownIdIsZero)
{
ptok->tk_val = (arith)0; ptok->tk_val = (arith)0;
return ptok->tk_symb = INTEGER; return ptok->tk_symb = INTEGER;
} }
@ -262,52 +296,65 @@ garbage:
register int base = 10, vch; register int base = 10, vch;
register arith val = 0; register arith val = 0;
int ovfl = 0; int ovfl = 0;
arith ubound = max_arith/(base/2); arith ubound = max_arith / (base / 2);
/* Since the preprocessor only knows integers and has /* Since the preprocessor only knows integers and has
* nothing to do with ellipsis we just return when the * nothing to do with ellipsis we just return when the
* pp-number starts with a '.' * pp-number starts with a '.'
*/ */
if (ch == '.') { if (ch == '.')
{
return ptok->tk_symb = ch; return ptok->tk_symb = ch;
} }
if (ch == '0') { if (ch == '0')
{
ch = GetChar(); ch = GetChar();
if (ch == 'x' || ch == 'X') { if (ch == 'x' || ch == 'X')
{
base = 16; base = 16;
ch = GetChar(); ch = GetChar();
} else { }
else
{
base = 8; base = 8;
} }
} }
while ((vch = val_in_base(ch, base)) >= 0) { while ((vch = val_in_base(ch, base)) >= 0)
if (val < 0 || val > ubound) ovfl++; {
if (val < 0 || val > ubound)
ovfl++;
val *= base; val *= base;
if (val < 0 && val + vch >= 0) ovfl++; if (val < 0 && val + vch >= 0)
ovfl++;
val += vch; val += vch;
ch = GetChar(); ch = GetChar();
} }
ptok->tk_unsigned = 0; ptok->tk_unsigned = 0;
if (ch == 'u' || ch == 'U') { if (ch == 'u' || ch == 'U')
{
ptok->tk_unsigned = 1; ptok->tk_unsigned = 1;
ch = GetChar(); ch = GetChar();
if (ch == 'l' || ch == 'L') { if (ch == 'l' || ch == 'L')
{
ch = GetChar(); ch = GetChar();
} }
} }
else if (ch == 'l' || ch == 'L') { else if (ch == 'l' || ch == 'L')
{
ch = GetChar(); ch = GetChar();
if (ch == 'u' || ch == 'U') { if (ch == 'u' || ch == 'U')
{
ptok->tk_unsigned = 1; ptok->tk_unsigned = 1;
ch = GetChar(); ch = GetChar();
} }
} }
if (ovfl) { if (ovfl)
{
warning("overflow in constant"); warning("overflow in constant");
ptok->tk_unsigned = 1; ptok->tk_unsigned = 1;
} }
else if (val < 0) { else if (val < 0)
{
/* give warning??? */ /* give warning??? */
ptok->tk_unsigned = 1; ptok->tk_unsigned = 1;
} }
@ -318,8 +365,10 @@ garbage:
case STEOI: /* end of text on source file */ case STEOI: /* end of text on source file */
return ptok->tk_symb = EOF; return ptok->tk_symb = EOF;
case STMSPEC: case STMSPEC:
if (!InputLevel) goto garbage; if (!InputLevel)
if (ch == TOKSEP) goto again; goto garbage;
if (ch == TOKSEP)
goto again;
/* fallthrough shouldn't happen */ /* fallthrough shouldn't happen */
default: /* this cannot happen */ default: /* this cannot happen */
crash("bad class for char 0%o", ch); crash("bad class for char 0%o", ch);
@ -327,8 +376,7 @@ garbage:
/*NOTREACHED*/ /*NOTREACHED*/
} }
void void skipcomment()
skipcomment()
{ {
/* The last character read has been the '*' of '/_*'. The /* The last character read has been the '*' of '/_*'. The
characters, except NL and EOI, between '/_*' and the first characters, except NL and EOI, between '/_*' and the first
@ -344,11 +392,16 @@ skipcomment()
NoUnstack++; NoUnstack++;
c = GetChar(); c = GetChar();
do { do
while (c != '*') { {
if (class(c) == STNL) { while (c != '*')
{
if (class(c) == STNL)
{
++LineNumber; ++LineNumber;
} else if (c == EOI) { }
else if (c == EOI)
{
NoUnstack--; NoUnstack--;
return; return;
} }
@ -359,9 +412,7 @@ skipcomment()
NoUnstack--; NoUnstack--;
} }
arith arith char_constant(nm) char* nm;
char_constant(nm)
char *nm;
{ {
register arith val = 0; register arith val = 0;
register int ch; register int ch;
@ -371,15 +422,18 @@ char_constant(nm)
if (ch == '\'') if (ch == '\'')
error("%s constant too short", nm); error("%s constant too short", nm);
else else
while (ch != '\'') { while (ch != '\'')
if (ch == '\n') { {
if (ch == '\n')
{
error("newline in %s constant", nm); error("newline in %s constant", nm);
LineNumber++; LineNumber++;
break; break;
} }
if (ch == '\\') if (ch == '\\')
ch = quoted(GetChar()); ch = quoted(GetChar());
if (ch >= 128) ch -= 256; if (ch >= 128)
ch -= 256;
if (size < sizeof(arith)) if (size < sizeof(arith))
val |= ch << (8 * size); val |= ch << (8 * size);
size++; size++;
@ -392,23 +446,24 @@ char_constant(nm)
return val; return val;
} }
char * char* string_token(nm, stop_char) char* nm;
string_token(nm, stop_char)
char *nm;
{ {
register int ch; register int ch;
register int str_size; register int str_size;
register char *str = Malloc((unsigned) (str_size = ISTRSIZE)); register char* str = Malloc((unsigned)(str_size = ISTRSIZE));
register int pos = 0; register int pos = 0;
ch = GetChar(); ch = GetChar();
while (ch != stop_char) { while (ch != stop_char)
if (ch == '\n') { {
if (ch == '\n')
{
error("newline in %s", nm); error("newline in %s", nm);
LineNumber++; LineNumber++;
break; break;
} }
if (ch == EOI) { if (ch == EOI)
{
error("end-of-file inside %s", nm); error("end-of-file inside %s", nm);
break; break;
} }
@ -424,16 +479,16 @@ string_token(nm, stop_char)
return str; return str;
} }
int int quoted(ch) register int ch;
quoted(ch)
register int ch;
{ {
/* quoted() replaces an escaped character sequence by the /* quoted() replaces an escaped character sequence by the
character meant. character meant.
*/ */
/* first char after backslash already in ch */ /* first char after backslash already in ch */
if (!is_oct(ch)) { /* a quoted char */ if (!is_oct(ch))
switch (ch) { { /* a quoted char */
switch (ch)
{
case 'n': case 'n':
ch = '\n'; ch = '\n';
break; break;
@ -460,7 +515,8 @@ quoted(ch)
register int hex = 0; register int hex = 0;
register int vch; register int vch;
for (;;) { for (;;)
{
ch = GetChar(); ch = GetChar();
if (vch = val_in_base(ch, 16), vch == -1) if (vch = val_in_base(ch, 16), vch == -1)
break; break;
@ -470,42 +526,39 @@ quoted(ch)
ch = hex; ch = hex;
} }
} }
} else { /* a quoted octal */ }
else
{ /* a quoted octal */
register int oct = 0, cnt = 0; register int oct = 0, cnt = 0;
do { do
oct = oct*8 + (ch-'0'); {
oct = oct * 8 + (ch - '0');
ch = GetChar(); ch = GetChar();
} while (is_oct(ch) && ++cnt < 3); } while (is_oct(ch) && ++cnt < 3);
UnGetChar(); UnGetChar();
ch = oct; ch = oct;
} }
return ch&0377; return ch & 0377;
} }
int val_in_base(ch, base) register int ch;
int
val_in_base(ch, base)
register int ch;
{ {
switch (base) { switch (base)
{
case 8: case 8:
return (is_dig(ch) && ch < '9') ? ch - '0' : -1; return (is_dig(ch) && ch < '9') ? ch - '0' : -1;
case 10: case 10:
return is_dig(ch) ? ch - '0' : -1; return is_dig(ch) ? ch - '0' : -1;
case 16: case 16:
return is_dig(ch) ? ch - '0' return is_dig(ch) ? ch - '0' : is_hex(ch) ? (ch - 'a' + 10) & 017 : -1;
: is_hex(ch) ? (ch - 'a' + 10) & 017
: -1;
default: default:
fatal("(val_in_base) illegal base value %d", base); fatal("(val_in_base) illegal base value %d", base);
/* NOTREACHED */ /* NOTREACHED */
} }
} }
int GetChar()
int
GetChar()
{ {
/* The routines GetChar and trigraph parses the trigraph /* The routines GetChar and trigraph parses the trigraph
sequences and removes occurences of \\\n. sequences and removes occurences of \\\n.
@ -520,58 +573,60 @@ again:
ch = trigraph(); ch = trigraph();
/* \\\n are removed from the input stream */ /* \\\n are removed from the input stream */
if (ch == '\\') { if (ch == '\\')
{
LoadChar(ch); LoadChar(ch);
if (ch == '\n') { if (ch == '\n')
{
++LineNumber; ++LineNumber;
goto again; goto again;
} }
PushBack(); PushBack();
ch = '\\'; ch = '\\';
} }
return(LexSave = ch); return (LexSave = ch);
} }
int trigraph()
int
trigraph()
{ {
register int ch; register int ch;
LoadChar(ch); LoadChar(ch);
if (ch == '?') { if (ch == '?')
{
LoadChar(ch); LoadChar(ch);
switch (ch) { /* its a trigraph */ switch (ch)
{ /* its a trigraph */
case '=': case '=':
ch = '#'; ch = '#';
return(ch); return (ch);
case '(': case '(':
ch = '['; ch = '[';
return(ch); return (ch);
case '/': case '/':
ch = '\\'; ch = '\\';
return(ch); return (ch);
case ')': case ')':
ch = ']'; ch = ']';
return(ch); return (ch);
case '\'': case '\'':
ch = '^'; ch = '^';
return(ch); return (ch);
case '<': case '<':
ch = '{'; ch = '{';
return(ch); return (ch);
case '!': case '!':
ch = '|'; ch = '|';
return(ch); return (ch);
case '>': case '>':
ch = '}'; ch = '}';
return(ch); return (ch);
case '-': case '-':
ch = '~'; ch = '~';
return(ch); return (ch);
} }
PushBack(); PushBack();
} }
PushBack(); PushBack();
return('?'); return ('?');
} }