diff options
author | Norbert Preining <norbert@preining.info> | 2019-09-02 13:46:59 +0900 |
---|---|---|
committer | Norbert Preining <norbert@preining.info> | 2019-09-02 13:46:59 +0900 |
commit | e0c6872cf40896c7be36b11dcc744620f10adf1d (patch) | |
tree | 60335e10d2f4354b0674ec22d7b53f0f8abee672 /language/oriya/bin |
Initial commit
Diffstat (limited to 'language/oriya/bin')
-rw-r--r-- | language/oriya/bin/oriya.c | 1147 |
1 files changed, 1147 insertions, 0 deletions
diff --git a/language/oriya/bin/oriya.c b/language/oriya/bin/oriya.c new file mode 100644 index 0000000000..c526900e8d --- /dev/null +++ b/language/oriya/bin/oriya.c @@ -0,0 +1,1147 @@ +/***************************************************************************/ +/* */ +/* oriya.c v0.99 */ +/* */ +/* Source code for "Oriya-TeX" preprocessor. */ +/* Anshuman Pandey <apandey@u.washington.edu>, 2003/01/22 */ +/* */ +/* Based on Revision 1.1 1996/03/05 of skt.c preprocessor developed by */ +/* Charles Wikner <wikner@nacdh4.nac.ac.za> */ +/* */ +/***************************************************************************/ + +#include <stdio.h> +#include <ctype.h> +#include <string.h> + +/* DECLARE FUNCTIONS */ +void exit (int); +void search (void); +void write_outbuf(void); +void write_line (char *); +char * str_find (char *, char *); +void getline (void); +char * command (char *); +void error (char *, int); +void process (void); +void chrcat (char *, char); +void bncont (void); +void bnword (void); +void single (char); +void frontac (void); +void backac (void); +void sam_warning (void); +void samyoga (void); + +FILE *infile, *outfile, *fopen(); +char infilename[80]; +char outfilename[80]; + +#define TRUE 1 +#define FALSE 0 + +unsigned char bnline; /* flag TRUE if there is any Oriya on this line */ +unsigned char bnmode; /* flag TRUE while within {\orx } */ +unsigned char eof_flag; /* flag True when end of file detected */ +unsigned char ac_flag; /* flag TRUE while processing Oriya vowels */ +unsigned char roman_flag; /* flag TRUE if previous output was Roman string */ + +int nest_cnt; /* '{' increments, '}' decrements, while in \orx */ +int err_cnt; /* incremented by any error while in \orx */ +#define err_max 10 /* after err_max errors, program aborts */ +int line_cnt; /* line number of current input line */ + +char inbuf[133]; /* input file line buffer of text being processed */ +char *i_ptr; /* general pointer to input buffer */ +char outbuf[512]; /* output file line buffer of text processed */ +char *o_ptr; /* general pointer to output buffer */ + +unsigned char cont_end; /* flag TRUE when line ends with %-continuation */ +unsigned char cont_begin; /* flag TRUE when line begins after %-continuation */ +unsigned char hal_flag; /* flag TRUE when hal_type detected in syllable */ +unsigned char ac_char; /* storage for working vowel character */ +unsigned char pre_ra; /* storage/flag for 'r' at beginning of samyoga */ +char ac_hook; /* vowel diacritic code */ +char bnbuf[255]; /* storage for Oriya in internal code */ +char *s_ptr; /* general pointer to Oriya buffer */ +char *old_sptr; /* points to samyoga start; used by warning message */ +char work[80]; /* general scratchpad */ +char *w_ptr; /* general pointer to work buffer */ +char tmp[80]; /* temporary buffer for previous syllable */ +int ra; /* post_ra type to use with this character */ +int ya; /* post_ya type to use with this character */ +int va; /* post_va type to use with this character */ +int na; /* post_na type to use with this character */ +int ma; /* post_ma type to use with this character */ +int la; /* post_la type to use with this character */ +int post_ra; /* flag to append ra to samyoga */ +int post_ya; /* flag to append ya to samyoga */ +int post_va; /* flag to append va to samyoga */ +int post_na; /* flag to append na to samyoga */ +int post_ma; /* flag to append ma to samyoga */ +int post_la; /* flag to append la to samyoga */ +int hasanta; /* flag to add hasanta to samyoga (i.e. no vowel) */ +int hr_flag; /* flag indicates vowel picked up in samyoga (h.r) */ +int low; + + +/***************************************************************************/ +/* Function: main() */ +/***************************************************************************/ + +main(argc,argv) +int argc; +char *argv[]; +{ char *p; int k; + + /* Initialization */ + + bnmode = eof_flag = FALSE; + nest_cnt = err_cnt = 0; + line_cnt = 0; + i_ptr = inbuf; *i_ptr = '\0'; + s_ptr = bnbuf; *s_ptr = '\0'; + o_ptr = outbuf; *o_ptr = '\0'; + + /* handle command-line options */ + + k=0; + if (argc>1) strcpy(infilename,argv[1]); + if (strcmp(infilename,"-h")==0) + { k=1; + strcpy(infilename,""); + printf("Preprocessor for \"Oriya-TeX\" package v.099, 2003.01.22\n"); + printf("Anshuman Pandey <apandey@u.washington.edu>\n"); + printf("Syntax: oriya infile[.or] [outfile[.tex]]\n"); + exit(0); + } + + /* then get file names */ + switch(argc-k) + { case 3: strcpy(infilename,argv[1+k]); + strcpy(outfilename,argv[2+k]); + break; + case 2: strcpy(infilename,argv[1+k]); + strcpy(outfilename,""); + break; + default: strcpy(infilename,""); + while(strlen(infilename) == 0) + { printf("Input file: "); gets(infilename); } + printf("Output file: "); + gets(outfilename); + } + + if (strlen(outfilename) == 0) + { strcpy (outfilename,infilename); /* default output file name */ + p = strchr(outfilename,'.'); + if (p != 0) *p = '\0'; /* delete out file name extension */ + } + p = strchr(infilename,'.'); + if (p == 0) strcat(infilename,".or"); /* default input file extension */ + if ((infile=fopen(infilename,"r")) == NULL) + { printf("Cannot open file %s\n",infilename); exit(1); } + getline(); if (eof_flag) + { printf("Input file %s is empty.\n",infilename); exit(1); } + p = strchr(outfilename,'.'); + if (p == 0) + { if (inbuf[0] == '@') strcat(outfilename,".dn"); + else strcat(outfilename,".tex"); /* set default output file extension */ + } + if ((outfile=fopen(outfilename,"w")) == NULL) + { printf("Cannot open output file %s\n",outfilename); exit(1); } + + /* Normal main loop */ + + while(eof_flag == 0) + { while(!bnmode && !eof_flag) search(); /* search for \orx command */ + while( bnmode && !eof_flag) process(); /* process Oriya text */ + if (err_cnt >= err_max) + { printf("Too many (%d) errors, aborting program\n",err_cnt); break; } + } + if ((err_cnt < err_max) && (nest_cnt != 0)) + printf("Brace mismatch within \\orx = %d\n",nest_cnt); + fclose(infile); + fclose(outfile); + exit(1); + +} + + +/***************************************************************************/ +/* Function: search() */ +/* */ +/* Search inbuf for '{\orx', getting more input lines as necessary */ +/* until string found or end of file, copying input to output; if */ +/* the string is found but command not recognised, it is treated as */ +/* ordinary text; if valid command i_ptr points to first sanskrit */ +/* char after command, and sets bnmode TRUE. */ +/***************************************************************************/ + +void search(void) +{ +unsigned char c; +char *p,*q; + while (eof_flag == 0) + { p = str_find(i_ptr,"{\\orx"); + if (p == 0) + { if (bnline == TRUE) { strcat(outbuf,i_ptr); write_outbuf(); } + else { write_line(inbuf); o_ptr = outbuf; *o_ptr = '\0'; } + getline(); + continue; + } + q = i_ptr; i_ptr = p; + if ((p = command(p)) == 0) /* test command string \orx */ + { p = i_ptr; i_ptr = q; /* if bad \orx command */ + c = *++p; *p = '\0'; /* copy partial line, and search more */ + strcat(outbuf,i_ptr); *p = c; i_ptr = p; continue; + } + i_ptr = q; + nest_cnt++; c = *p; *p = '\0'; /* skip over '{\orx' */ + strcat(outbuf,i_ptr); /* append partial line to outbuf */ + *p = c; i_ptr = p; + bnmode = TRUE; bnline = TRUE; /* now comes the fun! */ + break; + } +} + + +/***************************************************************************/ +/* Function: write_outbuf() */ +/* */ +/* Write outbuf in 80 character lines */ +/***************************************************************************/ + +void write_outbuf(void) +{ +char c, d, e; + while(1) + { c = '\0'; + if (strlen(outbuf) < 81) { write_line(outbuf); break; } + for (o_ptr = outbuf + 78; o_ptr > outbuf + 50; o_ptr--) + { if (*o_ptr == ' ') break; } + if (*o_ptr != ' ') { for (o_ptr = outbuf+78; o_ptr > outbuf + 50; o_ptr--) + if ((*o_ptr=='\\') && (*(o_ptr-1)!='\\')) break; + if (o_ptr == outbuf+50) o_ptr = outbuf+78; + c = *o_ptr; *o_ptr++ = '%'; d = *o_ptr; + } + *o_ptr++ = '\n'; e = *o_ptr; *o_ptr = '\0'; + write_line(outbuf); + *o_ptr = e; + if (c!='\0') { *--o_ptr = d; *--o_ptr = c; } /* restore displaced chars */ + strcpy(outbuf,o_ptr); + } + o_ptr = outbuf; + *o_ptr = '\0'; +} + + +/***************************************************************************/ +/* Function: write_line() */ +/* */ +/* Write p-string to output device */ +/***************************************************************************/ + +void write_line(char *p) +{ + if (err_cnt == 0) fputs(p,outfile); +} + + +/***************************************************************************/ +/* Function: str_find() */ +/* */ +/* Find first occasion of string *str within *buf before '%' char; */ +/* return pointer first char of str within buf, else 0. */ +/***************************************************************************/ + +char * str_find(char *buf, char *str) +{ char *p, *x; + p = strstr(buf,str); + if (p == 0) return(0); + x = strchr(buf,'%'); + if ((x != 0) && (p > x)) return(0); + return(p); +} + + +/***************************************************************************/ +/* Function: getline() */ +/* */ +/* Get another line from input file; reset i_ptr, increments */ +/* line_cnt, and sets eof_flag if EOF. */ +/***************************************************************************/ + +void getline(void) +{ +char *p; + i_ptr = inbuf; + *i_ptr = '\0'; + line_cnt++; + if (fgets(inbuf,133,infile) == NULL) eof_flag = TRUE; + if (bnmode == FALSE) bnline = FALSE; +} + + +/***************************************************************************/ +/* Function: command() */ +/* */ +/* Check for valid \orx command; if invalid command, print error message */ +/***************************************************************************/ + +char * command(char *p) +{ p += 5; /* skip over '{\orx' */ + if (*p++ != ' ') p = 0; + if (p == 0) error("Unrecognised command string",7); + return(p); +} + + +/***************************************************************************/ +/* Function: error() */ +/* */ +/* Print out error message, including string *s and 'n' characters */ +/* of inbuf. */ +/***************************************************************************/ + +void error(char *s, int n) +{ char err_str[80]; int j; + if (++err_cnt <= err_max) + { if (n > 0) { for (j=0; j<n; j++) err_str[j] = *(i_ptr+j); + err_str[j] = '\0'; + } + if (n == 0) { strcpy(err_str,"oct("); + chrcat(err_str,'0' + (*i_ptr/64)); + chrcat(err_str,'0' + (*i_ptr/8)); + chrcat(err_str,'0' + (*i_ptr & 7)); + strcat(err_str,")"); + } + if (n < 0) { err_str[0] = '\0'; } + } + printf("Line %4d Error: %s %s\n",line_cnt,s,err_str); +} + + +/***************************************************************************/ +/* Function: process() */ +/* */ +/* Process input text within {\orx, converting to internal format in bnbuf */ +/***************************************************************************/ + +#define ISAC(c) (((strchr("aAiIuUxXeEoO",c) != 0) && c) ? TRUE : FALSE) + +/* wW removed from the definition of ISAC above (.R .L) */ + +void process(void) +{ int cap_flag, underscore; +unsigned char *i, c, d; +#define CF ac_flag=underscore=cap_flag=roman_flag=FALSE +#define CC CF; continue +#define CR ac_flag=underscore=cap_flag=FALSE; +#define CI i_ptr++; CC + + CF; + while(1) + { if (eof_flag) return; + if (err_cnt >= err_max) + { bnmode = FALSE; return; } + c = *i_ptr; d = *(i_ptr+1); +/* END OF LINE */ + if ((c == '\0') || (c == '\n')) + { bnword(); strcat (outbuf,i_ptr); write_outbuf(); getline(); CC; } +/* IMBEDDED ROMAN */ + if (strchr("!'()*+,-/:;=?[]`",c) || ((c == '.') && (*(i_ptr+1) == '.'))) + { if (c == '.') i_ptr++; + if (bnbuf[0]) { bnword(); } + while(1) + { chrcat(outbuf,c); c = *++i_ptr; + if (c == '.') + { if (*(i_ptr+1) != '.') break; + i_ptr++; continue; + } + if ((strchr("!'()*+,-/:;=?[]`",c) && c) == 0) break; + } + CR; continue; + } +/* ILLEGAL CHARS */ + if (strchr("_$fqwxzBCDEFGJKNOPQSTVWXYZ\177",c)) + { error("Illegal oriya character: ",1); CI; } + if (c>127) { error("Invalid character >80H: ",1); CI; } +/*?? Since we are now case sensitive (unlike skt), the list of */ +/*?? illegal chars has been increased (_ added, and & removed) */ +/* CONTROL CHARACTERS */ + if (c < ' ') + { error("Illegal control character: ",0); CI; } +/* IMBEDDED LATEX COMMAND STRINGS */ + if (c == '\\') + { if (d == '-') /* imbedded discretionary hyphen */ + { strcat(bnbuf,"!"); i_ptr++; CI; } + bnword(); + if (isalpha(d) == 0) + { chrcat(outbuf,c); chrcat(outbuf,*++i_ptr); CI; } + else + { while (1) + { chrcat(outbuf,c); c = *++i_ptr; if (isalpha(c) == 0) break; } + } + CC; + } +/* SPACE CHAR */ + if (c == ' ') + { bnword(); while(*++i_ptr == ' '); chrcat(outbuf,c); CC; + } +/*?? slight change here, since underscore is now an illegal character */ +/* COMMENT DELIMITER */ + if (c == '%') + { if (*(i_ptr+1) == '\n') bncont(); + else bnword(); + strcat(outbuf,i_ptr); write_outbuf(); getline(); CC; + } + +/* HASANTA */ + if (c == '&') { + c = '@'; + } + +/* BRACES */ + if (c == '{') { if (d == '}') { i_ptr++; CI; } /* for words like pra{}uga */ + else { nest_cnt++; bncont(); chrcat(outbuf,c); CI; } + } + if (c == '}') + { bnword(); chrcat(outbuf,c); + if (--nest_cnt == 0) + { bnmode = FALSE; + i_ptr++; return; + } + else CI; + } +/* UPPER CASE */ + if (isupper(c)) + { switch (c) + { case 'A': + case 'I': + case 'U': + case 'M': + case 'H': break; + case 'L': break; + case 'R': c = 'w'; break; + default: c = '*'; break; + } + if (c=='*') { error("Invalid upper case: ",1); CI; } + } +/*?? big change with that code: the upper case has a different *meaning* than */ +/*?? the lower case: fortunately, AIUMH are the same as the internal code :-) */ + +/* EQUATE VA WITH BA */ + if (c == 'v') { + c = 'b'; + } + +/* DOT_CHAR */ + if (c == '.') { switch(d) + { case 'y': c = 'Y'; break; + case 'd': c = 'q'; break; + case 'h': c = 'H'; break; + case 'l': c = 'X'; break; + case 'm': c = 'M'; break; + case 'n': c = 'N'; break; + case 'o': c = '%'; break; + case 'r': c = 'x'; break; + case 's': c = 'S'; break; + case 't': c = 'f'; break; + case 'a': c = '\\'; break; + case 'b': c = 'v'; break; + } + if (c=='.') { error("Invalid dot_character: ",2); CI; } + i_ptr++; d = *(i_ptr+1); + } + +/* NEXT CHAR IS H */ + if (d=='h') + { if (strchr("bcdfgjkptqw",c)) { c=toupper(c); i_ptr++; d=*(i_ptr+1); } + } + +/* The upper/lowercase stuff removed: a following 'h' converts a consonant */ +/* to its upper case internal code, e.g th --> T. Note that 'w' is added */ +/* to the list for R Rh */ + +/* QUOTE CHAR */ + if (c == '\"') { switch(d) + { case 'n': c = 'z'; break; + case 's': c = 'Z'; break; + } + if (c=='\"') { error("Invalid quote_character",2); CI; } + i_ptr++; d = *(i_ptr+1); + } + +/* TILDE CHAR */ + if (c == '~') { switch (d) + { case 'n': c = 'V'; break; + case 'm': c = '~'; break; + case 'r': c = 'R'; break; + default : c = '*'; break; + } + if (c=='*') + { error("Invalid use of tilde character: ",2); CI; } + i_ptr++; d = *(i_ptr+1); + } +/* TWO CHAR VOWELS */ + if ( strchr("aiu",c) && strchr("aiu",d) ) + { switch(c) + { case 'a': switch(d) + { case 'a': c = 'A'; break; + case 'i': c = 'E'; break; + case 'u': c = 'O'; break; + } break; + case 'i': if (d=='i') c = 'I'; break; + case 'u': if (d=='u') c = 'U'; break; + } + if (isupper(c)) { i_ptr++; d = *(i_ptr+1); } + } +/*?? all the upper/lowercase stuff removed */ +/* NOW CHAR SHOULD BE INTERNAL REPRESENTATION OF SANSKRIT CHAR */ + if ( ((c=='~' || c=='M') && !(ac_flag)) ) { + i_ptr -=2; error("No vowel before nasal: ",3); i_ptr +=2; CF; + } + + if (c=='H' && !(ac_flag)) { + i_ptr -=2; error("No vowel before visarga: ",3); i_ptr +=2; CF; + } + + chrcat(bnbuf,c); + CR; + if (ISAC(c)) ac_flag = TRUE; + i_ptr++; + } +} + +#undef CI +#undef CC +#undef CR +#undef CF + + +/***************************************************************************/ +/* Function: chrcat() */ +/* */ +/* Append character c to end of buffer s */ +/***************************************************************************/ + +void chrcat(char *s, char c) +{ char temp[] = " "; temp[0] = c; strcat(s,temp); +} + + +/***************************************************************************/ +/* Function: bncont() */ +/* */ +/* Similar to bnword() but used where input text line ends in '%' to */ +/* continue on next line. */ +/***************************************************************************/ + +void bncont(void) +{ + cont_end = TRUE; bnword(); + cont_end = FALSE; cont_begin = TRUE; +} + + +/***************************************************************************/ +/* Function: bnword() */ +/* */ +/* Convert contents of bnbuf to output string in outbuf */ +/***************************************************************************/ + +/* internal code for consonants */ +static char hal_chars[] = "BCDFGJKLNPQRSTVWYZbcdfghjklmnpqrstvwyz"; + +#define ISHAL(c) (((strchr(hal_chars,c) != 0) && c) ? TRUE : FALSE) + +#define CLRFLAGS ac_hook=post_ra=pre_ra=hasanta=hal_flag=post_ya=post_va=post_na=post_ma=post_la=0 + +#define CAT(w,x,z) \ +strcat(w,x); strcat(w,z) + +void bnword(void) +{ char c; + if (roman_flag && bnbuf[0]) { strcat(outbuf,"\\,"); roman_flag = FALSE; } + +/* A word is built up one syllable at a time: a syllable typically comprises */ +/* a consonant (or samyoga) followed by a vowel (with its nasalisation). */ +/* If there is no consonant, then a front-vowel is output; if there */ +/* is no vowel, then a viraama is appended to the consonant/samyoga. */ +/* One effect of this is that, if a consonant cluster is not fully resolved */ +/* into a single samyoga, it will be treated as two syllable: in particular, */ +/* the hook of the short-i will span one samyoga only. */ +/* */ +/* The `work' buffer is used as a scratchpad while building a syllable; on */ +/* completion it is stored in the `tmp' buffer before shipping to the output */ +/* buffer. This temporary storage while working on the next syllable, allows */ +/* changes to the back spacing of the previous syllable for more effiecient */ +/* output. */ +/* */ +/* `ra' is difficult: the first `r' of a consonant cluster is simply flagged */ +/* in `pre_ra', and similarly the final `r' in `post_ra', and then these are */ +/* dealt with when appending a vowel. */ + + CLRFLAGS; + s_ptr = bnbuf; c = *s_ptr; + if (c == '\0') return; + *tmp = '\0'; *work = '\0'; + while (1) + { CLRFLAGS; /* in particular, need to clear hal_flag for the likes of kara */ + c= *s_ptr++; + if (c == '\0') + { if (*tmp) { if (outbuf[0]=='\0' && tmp[0]=='[') strcat(outbuf,"{}"); + strcat(outbuf,tmp); + } + break; + } + if (ISAC(c)) + { ac_char = c; + frontac(); + if (*tmp) { if (outbuf[0]=='\0' && tmp[0]=='[') strcat(outbuf,"{}"); + strcat(outbuf,tmp); + } + strcpy(tmp,work); + *work = '\0'; cont_begin = 0; + continue; + } + if (strchr("0123456789\"!%|\\@~HM",c)) + { single(c); + if (*tmp) { if (outbuf[0]=='\0' && tmp[0]=='[') strcat(outbuf,"{}"); + strcat(outbuf,tmp); + } + strcpy(tmp,work); + *work = '\0'; cont_begin = 0; + continue; + } + if (c == 'r') { pre_ra = TRUE; c = *s_ptr; } + else s_ptr--; + old_sptr = s_ptr; /* save pointer to start of samyoga */ + if (ISHAL(c)) { hal_flag = TRUE; samyoga(); c = *s_ptr; } + ac_char = hasanta = 0; + if (!hr_flag) { if (ISAC(c)) { ac_char = *s_ptr++; } + else hasanta = TRUE; /* hr_flag = h.r parsed by samyoga */ + } + backac(); hr_flag = FALSE; + if (*tmp) { if (outbuf[0]=='\0' && tmp[0]=='[') strcat(outbuf,"{}"); + strcat(outbuf,tmp); + } + strcpy(tmp,work); + *work = '\0'; cont_begin = FALSE; + + } + strcat(outbuf,work); + s_ptr = bnbuf; *s_ptr = '\0'; + cont_begin = 0; +} + + +/***************************************************************************/ +/* Function: single() */ +/* */ +/* Output single (stand-alone) character to work buffer */ +/***************************************************************************/ + +void single(char c) +{ + switch(c) + { case '0': strcat(work,"0"); break; /* numerals */ + case '1': strcat(work,"1"); break; + case '2': strcat(work,"2"); break; + case '3': strcat(work,"3"); break; + case '4': strcat(work,"4"); break; + case '5': strcat(work,"5"); break; + case '6': strcat(work,"6"); break; + case '7': strcat(work,"7"); break; + case '8': strcat(work,"8"); break; + case '9': strcat(work,"9"); break; + case '!': strcat(tmp,"\\-"); break; /* discretionary hyphen */ + case '%': strcat(work,"{\\char21}"); break; /* OM */ + case '|': strcat(work,"."); break; /* dnari */ + case '\\': strcat(work,"{\\char253}"); break; /* avagraha */ + case '@': strcat(work,"+"); break; /* hasanta */ + case '~': strcat(work,"/"); break; /* candrabindu */ + case 'H': strcat(work,"{\\char250}"); break; /* visarga */ + case 'M': strcat(work,"M"); break; /* anusvara */ + + } +} + + +/***************************************************************************/ +/* Function: frontac() */ +/* */ +/* Process a front-vowel to workbuf */ +/***************************************************************************/ + +void frontac(void) +{ + CLRFLAGS; + switch(ac_char) + { case 'a': strcat(work,"a"); break; + case 'A': strcat(work,"aA"); break; + case 'i': strcat(work,"i"); break; + case 'I': strcat(work,"I"); break; + case 'u': strcat(work,"u"); break; + case 'U': strcat(work,"U"); break; + case 'x': strcat(work,"{\\char128}"); break; + case 'X': strcat(work,"{\\char132}"); break; + case 'e': strcat(work,"e"); break; + case 'E': strcat(work,"E"); break; + case 'o': strcat(work,"o"); break; + case 'O': strcat(work,"O"); break; + default : error("Lost in frontac()",-1); + } +} + + +/***************************************************************************/ +/* Function: sam_warning() */ +/* */ +/* Print a warning message that a hasanta will be used within a */ +/* samyoga. Also print input file line number, together with an */ +/* indication of the samyoga and where the viraama will be placed. */ +/***************************************************************************/ + +void sam_warning(void) +{ + char *p, msg[80]=""; + p = old_sptr; + if (pre_ra) + { strcat(msg,"r"); + if (p==s_ptr) strcat(msg,"-"); + } + while (ISHAL(*p)) + { switch (*p) + { case 'B': strcat(msg,"bh"); break; + case 'C': strcat(msg,"ch"); break; + case 'D': strcat(msg,"dh"); break; + case 'G': strcat(msg,"gh"); break; + case 'H': strcat(msg,".h"); break; + case 'J': strcat(msg,"jh"); break; + case 'K': strcat(msg,"kh"); break; + case 'P': strcat(msg,"ph"); break; + case 'T': strcat(msg,"th"); break; + case 'f': strcat(msg,".t"); break; + case 'F': strcat(msg,".th"); break; + case 'N': strcat(msg,".n"); break; + case 'q': strcat(msg,".d"); break; + case 'Q': strcat(msg,".dh"); break; + case 'S': strcat(msg,".s"); break; + case 'V': strcat(msg,"~n"); break; + case 'Y': strcat(msg,".y"); break; + case 'z': strcat(msg,"\"n"); break; + case 'Z': strcat(msg,"\"s"); break; + default: chrcat(msg,*p); break; + } + if (++p == s_ptr) strcat(msg,"-"); + } + if (ISAC(*p)) + { switch (*p) + { /* case 'w': strcat(msg,".l"); break; */ + case 'x': strcat(msg,".r"); break; + default: chrcat(msg,*p); break; + } + } + printf("Line %4d Warning: samyoga viraama: %s\n",line_cnt,msg); +} + + +/***************************************************************************/ +/* Function: backac() */ +/* */ +/* Handle vowel diacritics */ +/***************************************************************************/ + +void backac(void) +{ int j,k; char c, *p; + + if (pre_ra && (*work=='\0')) /* case r.r, r.R, r.l, r.L, ru, rU, ra */ + { c = toupper(ac_char); + /* if (c == 'X') {frontac(); } */ + + if (c == 'X') { + frontac(); + if ((ac_char == 'x') || (ac_char == 'X')) { + strcat(work,"{\\char147}"); + ac_char = 'a'; + } + } else { + strcat(work,"r"); + } + + pre_ra = FALSE; hal_flag = TRUE; + } + + if (post_ra) { strcat(work,"{\\char146}"); } /* ra-phala */ + if (post_ya) { strcat(work,"{\\char138}"); } /* ya-phala */ + if (post_va) { strcat(work,"{\\char164}"); } /* va-phala */ + if (post_na) { strcat(work,"{\\char240}"); } /* na-phala */ + if (post_ma) { strcat(work,"{\\char136}"); } /* ma-phala */ + if (post_la) { strcat(work,"{\\char155}"); } /* ma-phala */ + post_la = post_ma = post_na = post_va = post_ya = post_ra = 0; + + c = ac_char; + + if (pre_ra) { strcat(work,"{\\char147}"); } /* add repha */ + if (hasanta) { strcat(work,"+"); } /* add halanta */ + + if (ac_char == 'A') { strcat(work,"A");} /* add aa-dia */ + if (ac_char == 'i') { strcat(work,"{\\char91}");} /* add aa-dia */ + if (ac_char == 'I') { strcat(work,"X"); } /* add ii-dia */ + if (ac_char == 'u') { /* add u-dia */ + if (low) { + strcat(work,"{\\char216}"); + low = FALSE; + } else { + strcat(work,"{\\char93}"); + } + } + if (ac_char == 'U') { strcat(work,"{\\char90}");} /* add uu-dia */ + if (ac_char == 'x') { strcat(work,"{\\char129}");} /* add .r dia */ + if (ac_char == 'X') { strcat(work,"{\\char133}");} /* add .l dia */ + if (ac_char == 'e') { CAT(tmp,"<",""); } /* add e-dia */ + if (ac_char == 'E') { CAT(tmp,"<",""); strcat(work,">");} /* add ai-dia */ + if (ac_char == 'o') { CAT(tmp,"<",""); strcat(work,"A");} /* add o-dia */ + if (ac_char == 'O') { CAT(tmp,"<",""); strcat(work,"*");} /* add au-dia */ + +} + +/***************************************************************************/ +/* Function: samyoga() */ +/* */ +/* Work along bnbuf sequentially to build up a samyoga print */ +/* string in the work buffer and update the samyoga parameters. */ +/* */ +/* The method is quite unsophisticated, but its simplicity lends */ +/* clarity for later additions or changes, and for this reason */ +/* is done in Devanagari alphabetical order, but with longer */ +/* strings before shorter. */ +/* */ +/* Cr and Cy conjuncts are not defined in the individual cases for each */ +/* consonant. Rather these are handled in bulk by the program at the end */ +/* of the function. */ +/* */ +/* Macros are used to simplify reading the program --- believe it or not! */ +/* */ +/* Switch/case is used on the first letter, then the main LS macro tests: */ +/* (1) if the test string matches the input exactly, then */ +/* (2) bump input pointer to the character after string match */ +/* (3) use NC etc macro to break out of switch instruction */ +/***************************************************************************/ + + +#define LS(a,c,z) n=strlen(a); \ + if(strncmp(p,a,n)==0) { strcat(work,z); p+=n; c;} + +#define NX sam_flag = 'X'; break; +#define NR sam_flag = 'R'; break; +#define NC sam_flag = 'C'; break; + +#define IX p++; sam_flag = 'X'; break; +#define IR p++; sam_flag = 'R'; break; +#define IC p++; sam_flag = 'C'; break; + +/******************************************************************************/ + +void samyoga(void) +{ +char *p, sam_flag; int n; + sam_flag = 0; + p = s_ptr; + while (1) + { if (!ISHAL(*p)) { NX; } + switch (*p++) + { + + + /* k */ + case 'k': if(*p=='A') + {p+=1; strcat(work,"{\\char157}"); hr_flag = TRUE; NR;} + if(*p=='i') + {p+=1; strcat(work,"{\\char141}"); hr_flag = TRUE; NR;} + if(*p=='u') + {p+=1; strcat(work,"{\\char139}"); hr_flag = TRUE; NR;} + if(*p=='U') + {p+=1; strcat(work,"{\\char150}"); hr_flag = TRUE; NR;} + if(*p=='S' && *(p+1)=='N') + {p+=2; strcat(work,"{\\char225}");NR;} + LS("k", NR, "{\\char199}" ); + LS("f", NR, "{\\char27}" ); + LS("t", NR, "{\\char182}" ); + LS("T", NR, "k{\\char195}" ); + LS("r", NR, "{\\char176}" ); + LS("s", NR, "{\\char203}" ); + LS("S", NR, "x" ); + strcat(work,"k"); NR; + + /* kh */ + case 'K': strcat(work,"K"); NR; + + /* g */ + case 'g': LS("g", NR, "{\\char30}" ); + LS("D", NR, "{\\char31}" ); + strcat(work,"g"); NR; + + /* gh */ + case 'G': strcat(work,"G"); NR; + + /* "n */ + case 'z': if(*p=='k' && *(p+1)=='S') + {p+=2; strcat(work,"");NR;} + LS("k", NR, "{\\char148}"); + LS("K", NR, "{\\char159}"); + LS("g", NR, "{\\char160}"); + LS("G", NR, "{\\char202}"); + LS("z", NR, "{\\char10}"); + LS("t", NR, "f{\\char174}"); + LS("D", NR, "{\\char10}"); + strcat(work,"f"); NR; + + /* c */ + case 'c': if(*p=='A') + {p+=1; strcat(work,"{\\char137}"); hr_flag = TRUE; NR;} + LS("c", NR, "{\\char152}"); + LS("C", NR, "{\\char210}"); + strcat(work,"c"); NR; + + /* ch */ + case 'C': strcat(work,"C"); NR; + + /* j */ + case 'j': LS("j", NR, "{\\char153}" ); + LS("J", NR, "{\\char36}" ); + LS("V", NR, "{\\char179}" ); + strcat(work,"j"); NR; + + /* jh */ + case 'J': strcat(work,"J"); NR; + + /* ~n */ + case 'V': LS("c", NR, "{\\char172}" ); + LS("C", NR, "{\\char230}" ); + LS("j", NR, "{\\char189}" ); + LS("J", NR, "{\\char173}" ); + strcat(work,"F"); NR; + + /* .t */ + case 'f': LS("k", NR, "{\\char10}" ); + LS("f", NR, "{\\char161}" ); + LS("F", NR, "{\\char10}" ); + strcat(work,"q"); NR; + + /* .th */ + case 'F': strcat(work,"Q"); NR; + + /* .da */ + case 'q': if(*p=='i') + {p+=1; strcat(work,"{\\char14}"); hr_flag = TRUE; NR;} + LS("g", NR, "{\\char127}" ); + LS("q", NR, "{\\char236}" ); + strcat(work,"w"); NR; + + /* .dh */ + case 'Q': if(*p=='i') + {p+=1; strcat(work,"{\\char15}"); hr_flag = TRUE; NR;} + strcat(work,"W"); NR; + + /* .n */ + case 'N': LS("f", NR, "{\\char242}" ); + LS("F", NR, "{\\char220}" ); + LS("q", NR, "{\\char187}" ); + LS("Q", NR, "{\\char188}" ); + LS("N", NR, "{\\char186}" ); + LS("t", NR, "{\\char10}" ); + /* LS("m", NR, "\\oSp{G}" ); */ + strcat(work,"N"); NR; + + /* t */ + case 't': if(*p=='A') + {p+=1; strcat(work,"{\\char12}"); hr_flag = TRUE; NR;} + if(*p=='i') + {p+=1; strcat(work,"{\\char193}"); hr_flag = TRUE; NR;} + if(*p=='u') + {p+=1; strcat(work,"{\\char194}"); hr_flag = TRUE; NR;} + if(*p=='U') + {p+=1; strcat(work,"{\\char13}"); hr_flag = TRUE; NR;} + LS("k", NR, "k{\\char151}" ); + LS("t", NR, "{\\char154}" ); + LS("T", NR, "T{\\char151}" ); + LS("n", NR, "{\\char162}" ); + LS("p", NR, "{\\char28}" ); + LS("m", NR, "{\\char163}" ); + /* LS("y", NR, "y{\\char151}" ); */ + LS("r", NR, "{\\char177}" ); + LS("b", NR, "b{\\char151}" ); + LS("Z", NR, "z{\\char151}" ); + LS("s", NR, "{\\char241}" ); + strcat(work,"t"); NR; + + /* th */ + case 'T': strcat(work,"T"); NR; + + /* d */ + case 'd': if(*p=='i') + {p+=1; strcat(work,"{\\char192}"); hr_flag = TRUE; NR;} + LS("g", NR, "{\\char234}"); + LS("G", NR, "{\\char10}"); + LS("d", NR, "{\\char169}"); + LS("D", NR, "{\\char170}"); + LS("b", NR, "{\\char10}"); + LS("B", NR, "{\\char214}"); + strcat(work,"d"); NR; + + /* dh */ + case 'D': LS("y", NR, "{\\char178}" ); + strcat(work,"D"); NR; + + /* n */ + case 'n': if(*p=='i') + {p+=1; strcat(work,"{\\char22}"); hr_flag = TRUE; NR;} + if(*p=='t' && *(p+1)=='r') + {p+=2; strcat(work,"{\\char232}");NR;} + LS("f", NR, "{\\char10}" ); + LS("q", NR, "{\\char10}" ); + LS("t", NR, "{\\char226}" ); + LS("T", NR, "{\\char227}" ); + LS("d", NR, "{\\char165}" ); + LS("D", NR, "{\\char166}" ); + strcat(work,"n"); NR; + + /* p */ + case 'p': LS("t", NR, "{\\char181}" ); + LS("p", NR, "{\\char183}" ); + LS("s", NR, "{\\char197}" ); + strcat(work,"p"); NR; + + /* ph */ + case 'P': strcat(work,"P"); NR; + + /* b */ + case 'b': LS("j", NR, "{\\char34}" ); + LS("d", NR, "{\\char167}" ); + LS("D", NR, "{\\char29}" ); + LS("b", NR, "{\\char168}" ); + strcat(work,"b"); NR; + + /* bh */ + case 'B': LS("N", NR, "{\\char10}" ); + strcat(work,"B"); NR; + + /* m */ + case 'm': LS("N", NR, "\\oSp{B}" ); + /* LS("n", NR, "{\\char163}" ); */ + LS("p", NR, "{\\char190}" ); + LS("P", NR, "{\\char191}" ); + LS("B", NR, "m{\\char239}" ); + /* LS("m", NR, "{\\char215}" ); */ + strcat(work,"m"); NR; + + /* y */ + case 'y': strcat(work,"y"); NR; + + /* .y */ + case 'Y': strcat(work,"Y"); NR; + + /* r */ + case 'r': if(*p=='A') + {p+=1; strcat(work,"{\\char11}"); hr_flag = TRUE; NR;} + strcat(work,"r"); NR; + + /* l */ + case 'l': if(*p=='A') + {p+=1; strcat(work,"{\\char158}"); hr_flag = TRUE; NR;} + LS("k", NR, "{\\char219}" ); + LS("g", NR, "\\oSp{I}" ); + LS("f", NR, "{\\char10}" ); + LS("q", NR, "{\\char10}" ); + LS("p", NR, "{\\char221}" ); + LS("P", NR, "{\\char223}" ); + LS("l", NR, "{\\char237}" ); + LS("Z", NR, "{\\char10}" ); + LS("S", NR, "{\\char10}" ); + strcat(work,"l"); NR; + + /* v */ + case 'v': LS("v", NR, "{\\char168}" ); + strcat(work,"{\\char20}"); NR; + + /* "s */ + case 'Z': LS("c", NR, "{\\char205}" ); + LS("C", NR, "{\\char229}" ); + LS("f", NR, "{\\char204}" ); + LS("Z", NR, "{\\char10}" ); + strcat(work,"z"); NR; + + /* .s */ + case 'S': LS("k", NR, "{\\char200}" ); + LS("f", NR, "{\\char198}" ); + LS("F", NR, "{\\char207}" ); + LS("N", NR, "{\\char16}" ); + LS("p", NR, "{\\char184}" ); + LS("P", NR, "{\\char222}" ); + LS("S", NR, "{\\char10}" ); + strcat(work,"S"); NR; + + /* s */ + case 's': if(*p=='t' && *(p+1)=='u') + {p+=2; strcat(work,"{\\char32}"); hr_flag = TRUE; NR;} + if(*p=='t' && *(p+1)=='r') + {p+=2; strcat(work,"{\\char233}");NR;} + LS("k", NR, "{\\char201}" ); + LS("K", NR, "{\\char26}" ); + LS("f", NR, "{\\char10}" ); + LS("t", NR, "{\\char180}" ); + LS("T", NR, "{\\char228}" ); + LS("p", NR, "{\\char185}" ); + LS("P", NR, "{\\char224}" ); + LS("s", NR, "\\oSp{H}" ); + strcat(work,"s"); NR; + + /* h */ + case 'h': if(*p=='i') + {p+=1; strcat(work,"{\\char196}"); hr_flag = TRUE; NR;} + if(*p=='u') + {p+=1; strcat(work,"{\\char140}"); hr_flag = TRUE; NR;} + /* if(*p=='x') { strcat(work,""); hr_flag = TRUE; IX; } */ + LS("N", NR, "{\\char10}"); + LS("n", NR, "{\\char94}"); + LS("b", NR, "{\\char95}"); + strcat(work,"h"); NR; + + /* R */ + case 'w': LS("g", NR, "{\\char126}"); + strcat(work,"w{\\char144}"); NR; + + /* Rh */ + case 'W': strcat(work,"W{\\char144}"); NR; + + /* L */ + case 'L': if(*p=='A') + {p+=1; strcat(work,"{\\char208}"); hr_flag = TRUE; IX;} + if(*p=='i') + {p+=1; strcat(work,"{\\char149}"); hr_flag = TRUE; IX;} + strcat(work,"L"); NR; + + /* Assamese r */ + case 'R': strcat(work,""); NR; + + default: error("Lost in samyoga()",-1); NX; + } + + if (sam_flag == 'X') { s_ptr = p; break; } + if (sam_flag == 'R') { if ((*p=='r')) { post_ra = TRUE; p++; } + if ((*p=='y')) { post_ya = TRUE; p++; } + if ((*p=='v' || *p=='b')) { post_va = TRUE; p++; } + if ((*p=='n')) { post_na = TRUE; p++; } + if ((*p=='m')) { post_ma = TRUE; p++; } + if ((*p=='l')) { post_la = TRUE; p++; } + s_ptr = p; break; + } + if (!ISHAL(*p)) { s_ptr = p; break; } + } +} + +/***************************************************************************/ +/* samapta */ +/***************************************************************************/ |