Simplify build: nested makefiles considered harmful.

This commit is contained in:
Eric S. Raymond 2017-05-29 01:33:59 -04:00
parent 7e82c2f910
commit 5ae76995d2
10 changed files with 18 additions and 32 deletions

547
compile.c Normal file
View file

@ -0,0 +1,547 @@
#define LINESIZE 100
#define RTXSIZ 277
#define CLSMAX 12
#define LOCSIZ 185
#define LINSIZ 12500
#define TRNSIZ 5
#define TABSIZ 330
#define VRBSIZ 35
#define HNTSIZ 20
#define TRVSIZ 885
#include <stdio.h>
#include <stdlib.h>
#include <stdbool.h>
#include <unistd.h>
const char advent_to_ascii[] = {0, 32, 33, 34, 39, 40, 41, 42, 43, 44, 45, 46, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 37, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 0, 1, 2, 3, 4, 5, 6, 7, 8, 0, 0, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 35, 36, 38, 47, 58, 59, 60, 61, 62, 63, 64, 91, 92, 93, 94, 95, 96, 123, 124, 125, 126, 0};
/* Rendered from the now-gone MPINIT() function */
const char ascii_to_advent[] = {0, 74, 75, 76, 77, 78, 79, 80, 81, 82, 0, 0, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 0, 1, 2, 106, 107, 63, 108, 3, 4, 5, 6, 7, 8, 9, 10, 109, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 110, 111, 112, 113, 114, 115, 116, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 117, 118, 119, 120, 121, 122, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 123, 124, 125, 126, 83};
// Global variables for use in functions below that can gradually disappear as code is cleaned up
long LNLENG;
long LNPOSN;
signed char INLINE[LINESIZE+1];
long I;
long K;
long KK;
long L;
long LOC;
long NEWLOC;
long OBJ;
long OLDLOC;
long SECT;
long VERB;
// Storage for what comes out of the database
long LINUSE;
long TRVS;
long CLSSES;
long TRNVLS;
long TABNDX;
long HNTMAX;
long PTEXT[101];
long RTEXT[RTXSIZ + 1];
long CTEXT[CLSMAX + 1];
long OBJSND[101];
long OBJTXT[101];
long STEXT[LOCSIZ + 1];
long LTEXT[LOCSIZ + 1];
long COND[LOCSIZ + 1];
long KEY[LOCSIZ + 1];
long LOCSND[LOCSIZ + 1];
long LINES[LINSIZ + 1];
long CVAL[CLSMAX + 1];
long TTEXT[TRNSIZ + 1];
long TRNVAL[TRNSIZ + 1];
long TRAVEL[TRVSIZ + 1];
long KTAB[TABSIZ + 1];
long ATAB[TABSIZ + 1];
long PLAC[101];
long FIXD[101];
long ACTSPK[VRBSIZ + 1];
long HINTS[HNTSIZ + 1][5];
bool is_set(long, long);
long GETTXT(long, long, long);
void BUG(long);
void MAPLIN(FILE*);
long GETNUM(FILE*);
int read_database(FILE*);
void write_0d(FILE*, FILE*, long, char*);
void write_1d(FILE*, FILE*, long[], long, char*);
void write_hints(FILE*, FILE*, long[][5], long, long, char*);
void write_files(FILE*, FILE*);
bool is_set(long var, long position)
{
long mask = 1l << position;
bool result = (var & mask) == mask;
return(result);
}
long GETTXT(long SKIP,long ONEWRD, long UPPER) {
/* Take characters from an input line and pack them into 30-bit words.
* Skip says to skip leading blanks. ONEWRD says stop if we come to a
* blank. UPPER says to map all letters to uppercase. If we reach the
* end of the line, the word is filled up with blanks (which encode as 0's).
* If we're already at end of line when GETTXT is called, we return -1. */
long CHAR;
long GETTXT;
static long SPLITTING = -1;
if(LNPOSN != SPLITTING)
SPLITTING = -1;
GETTXT= -1;
L10:
if(LNPOSN > LNLENG)
return(GETTXT);
if((!SKIP) || INLINE[LNPOSN] != 0)
goto L11;
LNPOSN=LNPOSN+1;
goto L10;
L11:
GETTXT=0;
for (int I=1; I<=5; I++) {
GETTXT=GETTXT*64;
if(LNPOSN > LNLENG || (ONEWRD && INLINE[LNPOSN] == 0))
continue;
CHAR=INLINE[LNPOSN];
if(CHAR >= 63)
goto L12;
SPLITTING = -1;
if(UPPER && CHAR >= 37)
CHAR=CHAR-26;
GETTXT=GETTXT+CHAR;
goto L14;
L12:
if(SPLITTING == LNPOSN)
goto L13;
GETTXT=GETTXT+63;
SPLITTING = LNPOSN;
continue;
L13:
GETTXT=GETTXT+CHAR-63;
SPLITTING = -1;
L14:
LNPOSN=LNPOSN+1;
}
return(GETTXT);
}
void BUG(long NUM) {
/* The following conditions are currently considered fatal bugs. Numbers < 20
* are detected while reading the database; the others occur at "run time".
* 0 Message line > 70 characters
* 1 Null line in message
* 2 Too many words of messages
* 3 Too many travel options
* 4 Too many vocabulary words
* 5 Required vocabulary word not found
* 6 Too many RTEXT messages
* 7 Too many hints
* 8 Location has cond bit being set twice
* 9 Invalid section number in database
* 10 Too many locations
* 11 Too many class or turn messages
* 20 Special travel (500>L>300) exceeds goto list
* 21 Ran off end of vocabulary table
* 22 Vocabulary type (N/1000) not between 0 and 3
* 23 Intransitive action verb exceeds goto list
* 24 Transitive action verb exceeds goto list
* 25 Conditional travel entry with no alternative
* 26 Location has no travel entries
* 27 Hint number exceeds goto list
* 28 Invalid month returned by date function
* 29 Too many parameters given to SETPRM */
fprintf(stderr, "Fatal error %ld. See source code for interpretation.\n", NUM);
exit(EXIT_FAILURE);
}
void MAPLIN(FILE *OPENED) {
/* Read a line of input, from the specified input source,
* translate the chars to integers in the range 0-126 and store
* them in the common array "INLINE". Integer values are as follows:
* 0 = space [ASCII CODE 40 octal, 32 decimal]
* 1-2 = !" [ASCII 41-42 octal, 33-34 decimal]
* 3-10 = '()*+,-. [ASCII 47-56 octal, 39-46 decimal]
* 11-36 = upper-case letters
* 37-62 = lower-case letters
* 63 = percent (%) [ASCII 45 octal, 37 decimal]
* 64-73 = digits, 0 through 9
* Remaining characters can be translated any way that is convenient;
* The "TYPE" routine below is used to map them back to characters when
* necessary. The above mappings are required so that certain special
* characters are known to fit in 6 bits and/or can be easily spotted.
* Array elements beyond the end of the line should be filled with 0,
* and LNLENG should be set to the index of the last character.
*
* If the data file uses a character other than space (e.g., tab) to
* separate numbers, that character should also translate to 0.
*
* This procedure may use the map1,map2 arrays to maintain static data for
* the mapping. MAP2(1) is set to 0 when the program starts
* and is not changed thereafter unless the routines on this page choose
* to do so. */
do {
fgets(INLINE + 1, sizeof(INLINE) - 1, OPENED);
}
while (!feof(OPENED) && INLINE[1] == '#');
LNLENG = 0;
for (int i = 1; i <= sizeof(INLINE) && INLINE[i] != 0; ++i)
{
char val = INLINE[i] + 1;
INLINE[i] = ascii_to_advent[val];
if (INLINE[i] != 0)
LNLENG = i;
}
LNPOSN = 1;
}
long GETNUM(FILE *source) {
/* Obtain the next integer from an input line. If K>0, we first read a
* new input line from a file; if K<0, we read a line from the keyboard;
* if K=0 we use a line that has already been read (and perhaps partially
* scanned). If we're at the end of the line or encounter an illegal
* character (not a digit, hyphen, or blank), we return 0. */
long DIGIT, GETNUM, SIGN;
if(source != NULL) MAPLIN(source);
GETNUM = 0;
while (INLINE[LNPOSN] == 0)
{
if (LNPOSN > LNLENG) return(GETNUM);
++LNPOSN;
}
SIGN=1;
if(INLINE[LNPOSN] != 9) goto L32;
SIGN= -1;
L30: LNPOSN=LNPOSN+1;
L32: if(LNPOSN > LNLENG || INLINE[LNPOSN] == 0) goto L42;
DIGIT=INLINE[LNPOSN]-64;
if(DIGIT < 0 || DIGIT > 9) goto L40;
GETNUM=GETNUM*10+DIGIT;
goto L30;
L40: GETNUM=0;
L42: GETNUM=GETNUM*SIGN;
LNPOSN=LNPOSN+1;
return(GETNUM);
}
int read_database(FILE* database) {
/* Clear out the various text-pointer arrays. All text is stored in array
* lines; each line is preceded by a word pointing to the next pointer (i.e.
* the word following the end of the line). The pointer is negative if this is
* first line of a message. The text-pointer arrays contain indices of
* pointer-words in lines. STEXT(N) is short description of location N.
* LTEXT(N) is long description. PTEXT(N) points to message for PROP(N)=0.
* Successive prop messages are found by chasing pointers. RTEXT contains
* section 6's stuff. CTEXT(N) points to a player-class message. TTEXT is for
* section 14. We also clear COND (see description of section 9 for details). */
for (int I=1; I<=300; I++) {
if(I <= 100) PTEXT[I] = 0;
if(I <= RTXSIZ) RTEXT[I] = 0;
if(I <= CLSMAX) CTEXT[I] = 0;
if(I <= 100) OBJSND[I] = 0;
if(I <= 100) OBJTXT[I] = 0;
if(I > LOCSIZ) break;
STEXT[I] = 0;
LTEXT[I] = 0;
COND[I] = 0;
KEY[I] = 0;
LOCSND[I] = 0;
}
LINUSE = 1;
TRVS = 1;
CLSSES = 0;
TRNVLS = 0;
/* Start new data section. Sect is the section number. */
L1002: SECT=GETNUM(database);
OLDLOC= -1;
switch (SECT) {
case 0: return(0);
case 1: goto L1004;
case 2: goto L1004;
case 3: goto L1030;
case 4: goto L1040;
case 5: goto L1004;
case 6: goto L1004;
case 7: goto L1050;
case 8: goto L1060;
case 9: goto L1070;
case 10: goto L1004;
case 11: goto L1080;
case 12: break;
case 13: goto L1090;
case 14: goto L1004;
default: BUG(9);
}
/* Sections 1, 2, 5, 6, 10, 14. Read messages and set up pointers. */
L1004: KK=LINUSE;
L1005: LINUSE=KK;
LOC=GETNUM(database);
if(LNLENG >= LNPOSN+70)BUG(0);
if(LOC == -1) goto L1002;
if(LNLENG < LNPOSN)BUG(1);
L1006: KK=KK+1;
if(KK >= LINSIZ)BUG(2);
LINES[KK]=GETTXT(false,false,false);
if(LINES[KK] != -1) goto L1006;
LINES[LINUSE]=KK;
if(LOC == OLDLOC) goto L1005;
OLDLOC=LOC;
LINES[LINUSE]= -KK;
if(SECT == 14) goto L1014;
if(SECT == 10) goto L1012;
if(SECT == 6) goto L1011;
if(SECT == 5) goto L1010;
if(LOC > LOCSIZ)BUG(10);
if(SECT == 1) goto L1008;
STEXT[LOC]=LINUSE;
goto L1005;
L1008: LTEXT[LOC]=LINUSE;
goto L1005;
L1010: if(LOC > 0 && LOC <= 100)PTEXT[LOC]=LINUSE;
goto L1005;
L1011: if(LOC > RTXSIZ)BUG(6);
RTEXT[LOC]=LINUSE;
goto L1005;
L1012: CLSSES=CLSSES+1;
if(CLSSES > CLSMAX)BUG(11);
CTEXT[CLSSES]=LINUSE;
CVAL[CLSSES]=LOC;
goto L1005;
L1014: TRNVLS=TRNVLS+1;
if(TRNVLS > TRNSIZ)BUG(11);
TTEXT[TRNVLS]=LINUSE;
TRNVAL[TRNVLS]=LOC;
goto L1005;
/* The stuff for section 3 is encoded here. Each "from-location" gets a
* contiguous section of the "TRAVEL" array. Each entry in travel is
* NEWLOC*1000 + KEYWORD (from section 4, motion verbs), and is negated if
* this is the last entry for this location. KEY(N) is the index in travel
* of the first option at location N. */
L1030: LOC=GETNUM(database);
if(LOC == -1) goto L1002;
NEWLOC=GETNUM(NULL);
if(KEY[LOC] != 0) goto L1033;
KEY[LOC]=TRVS;
goto L1035;
L1033: TRVS--; TRAVEL[TRVS]= -TRAVEL[TRVS]; TRVS++;
L1035: L=GETNUM(NULL);
if(L == 0) goto L1039;
TRAVEL[TRVS]=NEWLOC*1000+L;
TRVS=TRVS+1;
if(TRVS == TRVSIZ)BUG(3);
goto L1035;
L1039: TRVS--; TRAVEL[TRVS]= -TRAVEL[TRVS]; TRVS++;
goto L1030;
/* Here we read in the vocabulary. KTAB(N) is the word number, ATAB(N) is
* the corresponding word. The -1 at the end of section 4 is left in KTAB
* as an end-marker. */
L1040:
for (TABNDX=1; TABNDX<=TABSIZ; TABNDX++) {
KTAB[TABNDX]=GETNUM(database);
if(KTAB[TABNDX] == -1) goto L1002;
ATAB[TABNDX]=GETTXT(true,true,true);
} /* end loop */
BUG(4);
/* Read in the initial locations for each object. Also the immovability info.
* plac contains initial locations of objects. FIXD is -1 for immovable
* objects (including the snake), or = second loc for two-placed objects. */
L1050: OBJ=GETNUM(database);
if(OBJ == -1) goto L1002;
PLAC[OBJ]=GETNUM(NULL);
FIXD[OBJ]=GETNUM(NULL);
goto L1050;
/* Read default message numbers for action verbs, store in ACTSPK. */
L1060: VERB=GETNUM(database);
if(VERB == -1) goto L1002;
ACTSPK[VERB]=GETNUM(NULL);
goto L1060;
/* Read info about available liquids and other conditions, store in COND. */
L1070: K=GETNUM(database);
if(K == -1) goto L1002;
L1071: LOC=GETNUM(NULL);
if(LOC == 0) goto L1070;
if(is_set(COND[LOC],K)) BUG(8);
COND[LOC]=COND[LOC] + (1l << K);
goto L1071;
/* Read data for hints. */
L1080: HNTMAX=0;
L1081: K=GETNUM(database);
if(K == -1) goto L1002;
if(K <= 0 || K > HNTSIZ)BUG(7);
for (int I=1; I<=4; I++) {
HINTS[K][I] =GETNUM(NULL);
} /* end loop */
HNTMAX=(HNTMAX>K ? HNTMAX : K);
goto L1081;
/* Read the sound/text info, store in OBJSND, OBJTXT, LOCSND. */
L1090: K=GETNUM(database);
if(K == -1) goto L1002;
KK=GETNUM(NULL);
I=GETNUM(NULL);
if(I == 0) goto L1092;
OBJSND[K]=(KK>0 ? KK : 0);
OBJTXT[K]=(I>0 ? I : 0);
goto L1090;
L1092: LOCSND[K]=KK;
goto L1090;
}
/* Finish constructing internal data format */
/* Having read in the database, certain things are now constructed. PROPS are
* set to zero. We finish setting up COND by checking for forced-motion travel
* entries. The PLAC and FIXD arrays are used to set up ATLOC(N) as the first
* object at location N, and LINK(OBJ) as the next object at the same location
* as OBJ. (OBJ>100 indicates that FIXED(OBJ-100)=LOC; LINK(OBJ) is still the
* correct link to use.) ABB is zeroed; it controls whether the abbreviated
* description is printed. Counts modulo 5 unless "LOOK" is used. */
void write_0d(FILE* c_file, FILE* header_file, long single, char* varname)
{
fprintf(c_file, "long %s = %ld;\n", varname, single);
fprintf(header_file, "extern long %s;\n", varname);
}
void write_1d(FILE* c_file, FILE* header_file, long array[], long dim, char* varname)
{
fprintf(c_file, "long %s[] = {\n", varname);
for (int i = 0; i < dim; ++i)
{
if (i % 10 == 0)
{
if (i > 0)
fprintf(c_file, "\n");
fprintf(c_file, " ");
}
fprintf(c_file, "%ld, ", array[i]);
}
fprintf(c_file, "\n};\n");
fprintf(header_file, "extern long %s[%ld];\n", varname, dim);
}
void write_hints(FILE* c_file, FILE* header_file, long matrix[][5], long dim1, long dim2, char* varname)
{
fprintf(c_file, "long %s[][%ld] = {\n", varname, dim2);
for (int i = 0; i < dim1; ++i)
{
fprintf(c_file, " {");
for (int j = 0; j < dim2; ++j)
{
fprintf(c_file, "%ld, ", matrix[i][j]);
}
fprintf(c_file, "},\n");
}
fprintf(c_file, "};\n");
fprintf(header_file, "extern long %s[%ld][%ld];\n", varname, dim1, dim2);
}
void write_files(FILE* c_file, FILE* header_file)
{
// preprocessor defines for the header
fprintf(header_file, "#define RTXSIZ 277\n");
fprintf(header_file, "#define CLSMAX 12\n");
fprintf(header_file, "#define LOCSIZ 185\n");
fprintf(header_file, "#define LINSIZ 12500\n");
fprintf(header_file, "#define TRNSIZ 5\n");
fprintf(header_file, "#define TABSIZ 330\n");
fprintf(header_file, "#define VRBSIZ 35\n");
fprintf(header_file, "#define HNTSIZ 20\n");
fprintf(header_file, "#define TRVSIZ 885\n");
fprintf(header_file, "\n");
// include the header in the C file
fprintf(c_file, "#include \"database.h\"\n");
fprintf(c_file, "\n");
// content variables
write_0d(c_file, header_file, LINUSE, "LINUSE");
write_0d(c_file, header_file, TRVS, "TRVS");
write_0d(c_file, header_file, CLSSES, "CLSSES");
write_0d(c_file, header_file, TRNVLS, "TRNVLS");
write_0d(c_file, header_file, TABNDX, "TABNDX");
write_0d(c_file, header_file, HNTMAX, "HNTMAX");
write_1d(c_file, header_file, PTEXT, 100 + 1, "PTEXT");
write_1d(c_file, header_file, RTEXT, RTXSIZ + 1, "RTEXT");
write_1d(c_file, header_file, CTEXT, CLSMAX + 1, "CTEXT");
write_1d(c_file, header_file, OBJSND, 100 + 1, "OBJSND");
write_1d(c_file, header_file, OBJTXT, 100 + 1, "OBJTXT");
write_1d(c_file, header_file, STEXT, LOCSIZ + 1, "STEXT");
write_1d(c_file, header_file, LTEXT, LOCSIZ + 1, "LTEXT");
write_1d(c_file, header_file, COND, LOCSIZ + 1, "COND");
write_1d(c_file, header_file, KEY, LOCSIZ + 1, "KEY");
write_1d(c_file, header_file, LOCSND, LOCSIZ + 1, "LOCSND");
write_1d(c_file, header_file, LINES, LINSIZ + 1, "LINES");
write_1d(c_file, header_file, CVAL, CLSMAX + 1, "CVAL");
write_1d(c_file, header_file, TTEXT, TRNSIZ + 1, "TTEXT");
write_1d(c_file, header_file, TRNVAL, TRNSIZ + 1, "TRNVAL");
write_1d(c_file, header_file, TRAVEL, TRVSIZ + 1, "TRAVEL");
write_1d(c_file, header_file, KTAB, TABSIZ + 1, "KTAB");
write_1d(c_file, header_file, ATAB, TABSIZ + 1, "ATAB");
write_1d(c_file, header_file, PLAC, 100 + 1, "PLAC");
write_1d(c_file, header_file, FIXD, 100 + 1, "FIXD");
write_1d(c_file, header_file, ACTSPK, VRBSIZ + 1, "ACTSPK");
write_hints(c_file, header_file, HINTS, HNTSIZ + 1, 5, "HINTS");
}
int main(int argc, char** argv)
{
argc = argc;
argv = argv;
FILE* database = fopen("adventure.text", "r");
read_database(database);
fclose(database);
FILE* c_file = fopen("database.c", "w");
FILE* header_file = fopen("database.h", "w");
write_files(c_file, header_file);
fclose(c_file);
fclose(header_file);
return(EXIT_SUCCESS);
}