#include #include #include #include #include /* * file - determine type of file */ #define LENDIAN(p) ((p)[0] | ((p)[1]<<8) | ((p)[2]<<16) | ((p)[3]<<24)) uchar buf[6001]; short cfreq[140]; short wfreq[50]; int nbuf; Dir* mbuf; int fd; char *fname; char *slash; int currfd; enum { Cword, Fword, Aword, Alword, Lword, I1, I2, I3, Clatin = 128, Cbinary, Cnull, Ceascii, Cutf, }; struct { char* word; int class; } dict[] = { "PATH", Lword, "TEXT", Aword, "adt", Alword, "aggr", Alword, "alef", Alword, "array", Lword, "block", Fword, "char", Cword, "common", Fword, "con", Lword, "data", Fword, "dimension", Fword, "double", Cword, "extern", Cword, "bio", I2, "float", Cword, "fn", Lword, "function", Fword, "h", I3, "implement", Lword, "import", Lword, "include", I1, "int", Cword, "integer", Fword, "iota", Lword, "libc", I2, "long", Cword, "module", Lword, "real", Fword, "ref", Lword, "register", Cword, "self", Lword, "short", Cword, "static", Cword, "stdio", I2, "struct", Cword, "subroutine", Fword, "u", I2, "void", Cword, }; /* codes for 'mode' field in language structure */ enum { Normal = 0, First, /* first entry for language spanning several ranges */ Multi, /* later entries " " " ... */ Shared, /* codes used in several languages */ }; struct { int mode; /* see enum above */ int count; int low; int high; char *name; } language[] = { Normal, 0, 0x00a0, 0x00ff, "Latin", Normal, 0, 0x0100, 0x01FF, "Extended Latin", Normal, 0, 0x0370, 0x03FF, "Greek", Normal, 0, 0x0400, 0x04FF, "Cyrillic", Normal, 0, 0x0530, 0x058F, "Armenian", Normal, 0, 0x0590, 0x05FF, "Hebrew", Normal, 0, 0x0600, 0x06FF, "Arabic", Normal, 0, 0x0900, 0x097F, "Devanagari", Normal, 0, 0x0980, 0x09FF, "Bengali", Normal, 0, 0x0A00, 0x0A7F, "Gurmukhi", Normal, 0, 0x0A80, 0x0AFF, "Gujarati", Normal, 0, 0x0B00, 0x0B7F, "Oriya", Normal, 0, 0x0B80, 0x0BFF, "Tamil", Normal, 0, 0x0C00, 0x0C7F, "Telugu", Normal, 0, 0x0C80, 0x0CFF, "Kannada", Normal, 0, 0x0D00, 0x0D7F, "Malayalam", Normal, 0, 0x0E00, 0x0E7F, "Thai", Normal, 0, 0x0E80, 0x0EFF, "Lao", Normal, 0, 0x1000, 0x105F, "Tibetan", Normal, 0, 0x10A0, 0x10FF, "Georgian", Normal, 0, 0x3040, 0x30FF, "Japanese", Normal, 0, 0x3100, 0x312F, "Chinese", First, 0, 0x3130, 0x318F, "Korean", Multi, 0, 0x3400, 0x3D2F, "Korean", Shared, 0, 0x4e00, 0x9fff, "CJK", Normal, 0, 0, 0, 0, /* terminal entry */ }; enum { Fascii, /* printable ascii */ Flatin, /* latin 1*/ Futf, /* UTF character set */ Fbinary, /* binary */ Feascii, /* ASCII with control chars */ Fnull, /* NULL in file */ } guess; void bump_utf_count(Rune); int cistrncmp(char*, char*, int); void filetype(int); int getfontnum(uchar*, uchar**); void hist(char*, char*); int isas(void); int isc(void); int iscint(void); int isenglish(void); int ishp(void); int ishtml(void); int isrfc822(void); int ismbox(void); int islimbo(void); int ismung(void); int isp9bit(void); int isp9font(void); int isrtf(void); int ismsdos(void); int iself(void); int istring(void); int isoffstr(void); int iff(void); int long0(void); int longoff(void); int istar(void); int isface(void); int isexec(void); int p9bitnum(uchar*); int p9subfont(uchar*); void print_utf(void); void type(char*, int); int utf_count(void); void wordfreq(void); int (*call[])(void) = { long0, /* recognizable by first 4 bytes */ istring, /* recognizable by first string */ iself, /* ELF (foreign) executable */ isexec, /* native executables */ iff, /* interchange file format (strings) */ longoff, /* recognizable by 4 bytes at some offset */ isoffstr, /* recognizable by string at some offset */ isrfc822, /* email file */ ismbox, /* mail box */ istar, /* recognizable by tar checksum */ ishtml, /* html keywords */ iscint, /* compiler/assembler intermediate */ islimbo, /* limbo source */ isc, /* c & alef compiler key words */ isas, /* assembler key words */ isp9font, /* plan 9 font */ isp9bit, /* plan 9 image (as from /dev/window) */ isrtf, /* rich text format */ ismsdos, /* msdos exe (virus file attachement) */ isface, /* ascii face file */ /* last resorts */ ismung, /* entropy compressed/encrypted */ isenglish, /* char frequency English */ 0 }; int mime; int cset = 1; char OCTET[] = "application/octet-stream"; char PLAIN[] = "text/plain"; void prtype(char *m, char *s) { char buf[80], *p; int add, n; if(mime){ if(cset) add = cistrncmp(m, "text/", 5) == 0 && strchr(m, ';') == nil; else{ add = 0; if((p = strchr(m, ';')) && p-m < sizeof buf){ snprint(buf, p-m, "%s", m); m = buf; } } for(;;){ if(cfreq[Cnull] + cfreq[Cbinary] > 0 ||(cfreq[Clatin] > 0 && cfreq[Cutf] > 0)){ add = 0; m = OCTET; break; } if((n = read(currfd, buf, sizeof buf)) <= 0) break; hist(buf, buf+n); } if(add && cfreq[Cutf] != 0) print("%s; charset=\"utf-8\"\n", m); else if(add && cfreq[Clatin]) print("%s; charset=\"latin1\"\n", m); else if(add) print("%s; charset=\"us-ascii\"\n", m); else print("%s\n", m); }else print("%s\n", s); } void main(int argc, char *argv[]) { int i, j, maxlen; char *cp; Rune r; ARGBEGIN{ case 'c': cset ^= 1; case 'm': mime = 1; break; default: fprint(2, "usage: file [-m] [file...]\n"); exits("usage"); }ARGEND; maxlen = 0; if(mime == 0 || argc > 1){ for(i = 0; i < argc; i++) { for (j = 0, cp = argv[i]; *cp; j++, cp += chartorune(&r, cp)) ; if(j > maxlen) maxlen = j; } } if (argc <= 0) { if(!mime) print ("stdin: "); filetype(0); } else { for(i = 0; i < argc; i++) type(argv[i], maxlen); } exits(0); } void type(char *file, int nlen) { Rune r; int i; char *p; if(nlen > 0){ slash = 0; for (i = 0, p = file; *p; i++) { if (*p == '/') /* find rightmost slash */ slash = p; p += chartorune(&r, p); /* count runes */ } print("%s:%*s",file, nlen-i+1, ""); } fname = file; if ((fd = open(file, OREAD)) < 0) { print("cannot open: %r\n"); return; } filetype(fd); close(fd); } /* * Unicode 4.0 4-byte runes. */ typedef int Rune1; enum { UTFmax1 = 4, }; int fullrune1(char *p, int n) { int c; if(n >= 1) { c = *(uchar*)p; if(c < 0x80) return 1; if(n >= 2 && c < 0xE0) return 1; if(n >= 3 && c < 0xF0) return 1; if(n >= 4) return 1; } return 0; } int chartorune1(Rune1 *rune, char *str) { int c, c1, c2, c3, n; Rune r; c = *(uchar*)str; if(c < 0xF0){ r = 0; n = chartorune(&r, str); *rune = r; return n; } c &= ~0xF0; c1 = *(uchar*)(str+1) & ~0x80; c2 = *(uchar*)(str+2) & ~0x80; c3 = *(uchar*)(str+3) & ~0x80; n = (c<<18) | (c1<<12) | (c2<<6) | c3; if(n < 0x10000 || n > 0x10FFFF){ *rune = Runeerror; return 1; } *rune = n; return 4; } void hist(char *buf, char *eob) { char x[5], *p; int n, f; Rune1 r; static char push[5]; static int npush; for(p = buf; p < eob; ) { if(npush){ memset(x, 0, sizeof x); memcpy(x, push, npush); memcpy(x + npush, p, sizeof x - 1 - npush); n = chartorune1(&r, x); if(n >= npush){ n -= npush; npush = 0; p += n; }else{ memmove(push, push + n, sizeof push - n); npush -= n; } }else{ if (!fullrune1(p, eob-p) && eob-p < UTFmax1) break; p += chartorune1(&r, p); } if (r == 0) f = Cnull; else if (r <= 0x7f) { if (!isprint(r) && !isspace(r)) f = Ceascii; /* ASCII control char */ else f = r; } else if (r == 0x80) { bump_utf_count(r); f = Cutf; } else if (r < 0xA0) f = Cbinary; /* Invalid Runes */ else if (r <= 0xff) f = Clatin; /* Latin 1 */ else { bump_utf_count(r); f = Cutf; /* UTF extension */ } cfreq[f]++; /* ASCII chars peg directly */ } npush = eob-p; memcpy(push, p, npush); } void filetype(int fd) { int i; char *m; free(mbuf); mbuf = dirfstat(fd); if(mbuf == nil){ print("cannot stat: %r\n"); return; } if(mbuf->mode & DMDIR) { prtype("text/directory", "directory"); return; } if(mbuf->type != 'M' && mbuf->type != '|') { m = smprint("special file #%c/%s", mbuf->type, mbuf->name); prtype(OCTET, m); free(m); return; } /* may be reading a pipe on standard input */ nbuf = readn(fd, buf, sizeof(buf)-1); if(nbuf < 0) { print("cannot read: %r\n"); return; } if(nbuf == 0) { print(PLAIN, "empty file"); return; } buf[nbuf] = 0; /* * build histogram table */ memset(cfreq, 0, sizeof(cfreq)); for (i = 0; language[i].name; i++) language[i].count = 0; hist((char*)buf, (char*)buf+nbuf); /* * gross classify */ if (cfreq[Cbinary]) guess = Fbinary; else if (cfreq[Cutf]) guess = Futf; else if (cfreq[Clatin]) guess = Flatin; else if (cfreq[Ceascii]) guess = Feascii; else if (cfreq[Cnull]) guess = Fbinary; else guess = Fascii; /* * lookup dictionary words */ memset(wfreq, 0, sizeof(wfreq)); if(guess == Fascii || guess == Flatin || guess == Futf) wordfreq(); /* * call individual classify routines */ currfd = fd; for(i=0; call[i]; i++) if((*call[i])()) return; /* * if all else fails, * print out gross classification */ if (nbuf < 100 && !mime) print("short "); if (guess == Fascii) prtype(PLAIN, "Ascii"); else if (guess == Feascii) prtype(PLAIN,"extended ascii"); else if (guess == Flatin) prtype(PLAIN,"latin ascii"); else if (guess == Futf && utf_count() < 4) print_utf(); else prtype(OCTET, "binary"); } void bump_utf_count(Rune r) { int low, high, mid; high = sizeof(language)/sizeof(language[0])-1; for (low = 0; low < high;) { mid = (low+high)/2; if (r >= language[mid].low) { if (r <= language[mid].high) { language[mid].count++; break; } else low = mid+1; } else high = mid; } } int utf_count(void) { int i, count; count = 0; for (i = 0; language[i].name; i++) if (language[i].count > 0) switch (language[i].mode) { case Normal: case First: count++; break; default: break; } return count; } int chkascii(void) { int i; for (i = 'a'; i < 'z'; i++) if (cfreq[i]) return 1; for (i = 'A'; i < 'Z'; i++) if (cfreq[i]) return 1; return 0; } int find_first(char *name) { int i; for (i = 0; language[i].name != 0; i++) if (language[i].mode == First && strcmp(language[i].name, name) == 0) return i; return -1; } void print_utf(void) { int i, printed, j; if(mime){ prtype(PLAIN, nil); return; } if (chkascii()) { printed = 1; print("Ascii"); } else printed = 0; for (i = 0; language[i].name; i++) if (language[i].count) { switch(language[i].mode) { case Multi: j = find_first(language[i].name); if (j < 0) break; if (language[j].count > 0) break; /* Fall through */ case Normal: case First: if (printed) print(" & "); else printed = 1; print("%s", language[i].name); break; case Shared: default: break; } } if(!printed) print("UTF"); print(" text\n"); } void wordfreq(void) { int low, high, mid, r; uchar *p, *p2, c; p = buf; for(;;) { while (p < buf+nbuf && !isalpha(*p)) p++; if (p >= buf+nbuf) return; p2 = p; while(p < buf+nbuf && isalpha(*p)) p++; c = *p; *p = 0; high = sizeof(dict)/sizeof(dict[0]); for(low = 0;low < high;) { mid = (low+high)/2; r = strcmp(dict[mid].word, (char*)p2); if(r == 0) { wfreq[dict[mid].class]++; break; } if(r < 0) low = mid+1; else high = mid; } *p++ = c; } } typedef struct Filemagic Filemagic; struct Filemagic { ulong x; ulong mask; char *desc; char *mime; }; /* * integers in this table must be as seen on a little-endian machine * when read from a file. */ Filemagic long0tab[] = { 0xF16DF16D, 0xFFFFFFFF, "pac1 audio file", OCTET, /* "pac1" */ 0x31636170, 0xFFFFFFFF, "pac3 audio file", OCTET, /* "pXc2 */ 0x32630070, 0xFFFF00FF, "pac4 audio file", OCTET, 0xBA010000, 0xFFFFFFFF, "mpeg system stream", OCTET, 0x30800CC0, 0xFFFFFFFF, "inferno .dis executable", OCTET, 0x04034B50, 0xFFFFFFFF, "zip archive", "application/zip", 070707, 0xFFFF, "cpio archive", OCTET, 0x2F7, 0xFFFF, "tex dvi", "application/dvi", 0xfaff, 0xfeff, "mp3 audio", "audio/mpeg", 0xfeff0000, 0xffffffff, "utf-32be", "text/plain; charset=utf-32be", 0xfffe, 0xffffffff, "utf-32le", "text/plain; charset=utf-32le", 0xfeff, 0xffff, "utf-16be", "text/plain; charset=utf-16be", 0xfffe, 0xffff, "utf-16le", "text/plain; charset=utf-16le", /* * venti & fossil magic numbers are stored big-endian on disk, * thus the numbers appear reversed in this table. */ 0xad4e5cd1, 0xFFFFFFFF, "venti arena", OCTET, }; int filemagic(Filemagic *tab, int ntab, ulong x) { int i; for(i=0; ioff, 0) == -1) return 0; if (readn(fd, buf, sizeof buf) != sizeof buf) continue; x = LENDIAN(buf); if((x&tp->mask) == tp->x){ prtype(tp->mime, tp->desc); return 1; } } return 0; } int longoff(void) { return fileoffmagic(longofftab, nelem(longofftab)); } int isexec(void) { Fhdr f; if(seek(fd, 0, 0) == -1) /* reposition to start of file */ return 0; if(crackhdr(fd, &f)) { prtype(OCTET, f.name); return 1; } return 0; } /* from tar.c */ enum { NAMSIZ = 100, TBLOCK = 512 }; union hblock { char dummy[TBLOCK]; struct header { char name[NAMSIZ]; char mode[8]; char uid[8]; char gid[8]; char size[12]; char mtime[12]; char chksum[8]; char linkflag; char linkname[NAMSIZ]; /* rest are defined by POSIX's ustar format; see p1003.2b */ char magic[6]; /* "ustar" */ char version[2]; char uname[32]; char gname[32]; char devmajor[8]; char devminor[8]; char prefix[155]; /* if non-null, path = prefix "/" name */ } dbuf; }; int checksum(union hblock *hp) { int i; char *cp; struct header *hdr = &hp->dbuf; for (cp = hdr->chksum; cp < &hdr->chksum[sizeof hdr->chksum]; cp++) *cp = ' '; i = 0; for (cp = hp->dummy; cp < &hp->dummy[TBLOCK]; cp++) i += *cp & 0xff; return i; } int istar(void) { int chksum; char tblock[TBLOCK]; union hblock *hp = (union hblock *)tblock; struct header *hdr = &hp->dbuf; if(seek(fd, 0, 0) == -1) /* reposition to start of file */ return 0; if (readn(fd, tblock, sizeof tblock) != sizeof tblock) return 0; chksum = strtol(hdr->chksum, 0, 8); if (hdr->name[0] != '\0' && checksum(hp) == chksum) { if (strcmp(hdr->magic, "ustar") == 0) prtype("application/x-ustar", "posix tar archive"); else prtype("application/x-tar", "tar archive"); return 1; } return 0; } /* * initial words to classify file */ struct FILE_STRING { char *key; char *filetype; int length; char *mime; } file_string[] = { "!\n__.SYMDEF", "archive random library", 16, "application/octet-stream", "!\n", "archive", 8, "application/octet-stream", "070707", "cpio archive - ascii header", 6, "application/octet-stream", "#!/bin/rc", "rc executable file", 9, PLAIN, "#!/bin/sh", "sh executable file", 9, PLAIN, "%!", "postscript", 2, "application/postscript", "\004%!", "postscript", 3, "application/postscript", "x T post", "troff output for post", 8, "application/troff", "x T Latin1", "troff output for Latin1", 10, "application/troff", "x T utf", "troff output for UTF", 7, "application/troff", "x T 202", "troff output for 202", 7, "application/troff", "x T aps", "troff output for aps", 7, "application/troff", "GIF", "GIF image", 3, "image/gif", "\0PC Research, Inc\0", "ghostscript fax file", 18, "application/ghostscript", "%PDF", "PDF", 4, "application/pdf", "\n", "HTML file", 7, "text/html", "\n", "HTML file", 7, "text/html", "\111\111\052\000", "tiff", 4, "image/tiff", "\115\115\000\052", "tiff", 4, "image/tiff", "\377\330\377\340", "jpeg", 4, "image/jpeg", "\377\330\377\341", "jpeg", 4, "image/jpeg", "\377\330\377\333", "jpeg", 4, "image/jpeg", "BM", "bmp", 2, "image/bmp", "\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1", "microsoft office document", 8, "application/octet-stream", "key; p++) { l = p->length; if(l == -1) l = strlen(p->key); if(nbuf >= l && memcmp(buf, p->key, l) == 0) { prtype(p->mime, p->filetype); return 1; } } if(strncmp((char*)buf, "TYPE=", 5) == 0) { /* td */ for(i = 5; i < nbuf; i++) if(buf[i] == '\n') break; s = (char*)buf + 5; m = smprint("%.*s picture", utfnlen(s, i-5), s); prtype(OCTET, m); free(m); return 1; } return 0; } struct offstr { ulong off; struct FILE_STRING; } offstrs[] = { 32*1024, "\001CD001\001", "ISO9660 CD image", 7, OCTET, 0, 0, 0, 0, 0 }; int isoffstr(void) { int n; char buf[256]; struct offstr *p; for(p = offstrs; p->key; p++) { if(seek(fd, p->off, 0) == -1) return 0; n = p->length; if (n > sizeof buf) n = sizeof buf; if (readn(fd, buf, n) != n) continue; if(memcmp(buf, p->key, n) == 0) { prtype(p->mime, p->filetype); return 1; } } return 0; } int iff(void) { if (strncmp((char*)buf, "FORM", 4) == 0 && strncmp((char*)buf+8, "AIFF", 4) == 0) { prtype("audio/x-aiff", "aiff audio"); return 1; } if (strncmp((char*)buf, "RIFF", 4) == 0) { if (strncmp((char*)buf+8, "WAVE", 4) == 0) prtype("audio/wave", "wave audio"); else if (strncmp((char*)buf+8, "AVI ", 4) == 0) prtype("video/avi", "avi video"); else prtype("application/octet-stream", "riff file"); return 1; } return 0; } char* html_string[] = { "title", "body", "head", "strong", "h1", "h2", "h3", "h4", "h5", "h6", "ul", "li", "dl", "br", "em", 0, }; int ishtml(void) { uchar *p, *q; int i, count; /* compare strings between '<' and '>' to html table */ count = 0; p = buf; for(;;) { while (p < buf+nbuf && *p != '<') p++; p++; if (p >= buf+nbuf) break; if(*p == '/') p++; q = p; while(p < buf+nbuf && *p != '>') p++; if (p >= buf+nbuf) break; for(i = 0; html_string[i]; i++) { if(cistrncmp(html_string[i], (char*)q, p-q) == 0) { if(count++ > 4) { prtype("text/html", "HTML file"); return 1; } break; } } p++; } return 0; } char* rfc822_string[] = { "from:", "date:", "to:", "subject:", "received:", "reply to:", "sender:", 0, }; int isrfc822(void) { char *p, *q, *r; int i, count; count = 0; p = (char*)buf; for(;;) { q = strchr(p, '\n'); if(q == nil) break; *q = 0; if(p == (char*)buf && strncmp(p, "From ", 5) == 0 && strstr(p, " remote from ")){ count++; *q = '\n'; p = q+1; continue; } *q = '\n'; if(*p != '\t' && *p != ' '){ r = strchr(p, ':'); if(r == 0 || r > q) break; for(i = 0; rfc822_string[i]; i++) { if(cistrncmp(p, rfc822_string[i], strlen(rfc822_string[i])) == 0){ count++; break; } } } p = q+1; } if(count >= 3){ prtype("message/rfc822", "email file"); return 1; } return 0; } int ismbox(void) { char *p, *q; p = (char*)buf; q = strchr(p, '\n'); if(q == nil) return 0; *q = 0; if(strncmp(p, "From ", 5) == 0 && strstr(p, " remote from ") == nil){ prtype(PLAIN, "mail box"); return 1; } *q = '\n'; return 0; } int iscint(void) { int type; char *name, *m; Biobuf b; if(seek(fd, 0, 0) == -1) return 0; if(Binit(&b, fd, OREAD) == Beof) return 0; type = objtype(&b, &name); if(type < 0) return 0; m = smprint("%s intermediate", name); prtype(OCTET, m); free(m); return 1; } int isc(void) { int n; n = wfreq[I1]; /* * includes */ if(n >= 2 && wfreq[I2] >= n && wfreq[I3] >= n && cfreq['.'] >= n) goto yes; if(n >= 1 && wfreq[Alword] >= n && wfreq[I3] >= n && cfreq['.'] >= n) goto yes; /* * declarations */ if(wfreq[Cword] >= 5 && cfreq[';'] >= 5) goto yes; /* * assignments */ if(cfreq[';'] >= 10 && cfreq['='] >= 10 && wfreq[Cword] >= 1) goto yes; return 0; yes: prtype(PLAIN, wfreq[Alword] > 0? "alef program": "c program"); return 1; } int islimbo(void) { /* * includes */ if(wfreq[Lword] < 4) return 0; prtype(PLAIN, "limbo program"); return 1; } int isas(void) { /* * includes */ if(wfreq[Aword] < 2) return 0; prtype(PLAIN, "as program"); return 1; } /* * low entropy means encrypted */ int ismung(void) { int i, bucket[8]; float cs; if(nbuf < 64) return 0; memset(bucket, 0, sizeof(bucket)); for(i=nbuf-64; i>5)&07] += 1; cs = 0.; for(i=0; i<8; i++) cs += (bucket[i]-8)*(bucket[i]-8); cs /= 8.; if(cs <= 24.322) { if(buf[0]==0x1f && buf[1]==0x9d) prtype(OCTET, "compressed"); else if(buf[0]==0x1f && buf[1]==0x8b) prtype(OCTET, "gzip compressed"); else if(buf[0]=='B' && buf[1]=='Z' && buf[2]=='h') prtype(OCTET, "bzip2 compressed"); else prtype(OCTET, "encrypted"); return 1; } return 0; } /* * english by punctuation and frequencies */ int isenglish(void) { int vow, comm, rare, badpun, punct; char *p; if(guess != Fascii && guess != Feascii) return 0; badpun = 0; punct = 0; for(p = (char *)buf; p < (char *)buf+nbuf-1; p++) switch(*p) { case '.': case ',': case ')': case '%': case ';': case ':': case '?': punct++; if(p[1] != ' ' && p[1] != '\n') badpun++; } if(badpun*5 > punct) return 0; if(cfreq['>']+cfreq['<']+cfreq['/'] > cfreq['e']) /* shell file test */ return 0; if(2*cfreq[';'] > cfreq['e']) return 0; vow = 0; for(p="AEIOU"; *p; p++) { vow += cfreq[*p]; vow += cfreq[tolower(*p)]; } comm = 0; for(p="ETAION"; *p; p++) { comm += cfreq[*p]; comm += cfreq[tolower(*p)]; } rare = 0; for(p="VJKQXZ"; *p; p++) { rare += cfreq[*p]; rare += cfreq[tolower(*p)]; } if(vow*5 >= nbuf-cfreq[' '] && comm >= 10*rare) { prtype(PLAIN, "English text"); return 1; } return 0; } /* * pick up a number with * syntax _*[0-9]+_ */ #define P9BITLEN 12 int p9bitnum(uchar *bp) { int n, c, len; len = P9BITLEN; while(*bp == ' ') { bp++; len--; if(len <= 0) return -1; } n = 0; while(len > 1) { c = *bp++; if(!isdigit(c)) return -1; n = n*10 + c-'0'; len--; } if(*bp != ' ') return -1; return n; } int depthof(char *s, int *newp) { char *es; int d; *newp = 0; es = s+12; while(s= 0) len = (hix+px-1)/px - lox/px; else{ /* make positive before divide */ t = (-lox)+px-1; t = (t/px)*px; len = (t+hix+px-1)/px; } }else len = (hix-lox)*dep/8; len *= hiy - loy; /* col length */ len += 5 * P9BITLEN; /* size of initial ascii */ /* * for compressed images, don't look any further. otherwise: * for image file, length is non-zero and must match calculation above. * for /dev/window and /dev/screen the length is always zero. * for subfont, the subfont header should follow immediately. */ if (cmpr) { m = smprint("Compressed %splan 9 image or subfont, depth %d", newlabel, dep); prtype(OCTET, m); free(m); return 1; } /* * mbuf->length == 0 probably indicates reading a pipe. * Ghostscript sometimes produces a little extra on the end. */ if (len != 0 && (mbuf->length == 0 || mbuf->length == len || mbuf->length > len && mbuf->length < len+P9BITLEN)) { m = smprint("%splan 9 image, depth %d", newlabel, dep); prtype(OCTET, m); free(m); return 1; } if (p9subfont(buf+len)) { m = smprint("%ssubfont file, depth %d", newlabel, dep); prtype(OCTET, m); free(m); return 1; } return 0; } int p9subfont(uchar *p) { int n, h, a; /* if image too big, assume it's a subfont */ if (p+3*P9BITLEN > buf+sizeof(buf)) return 1; n = p9bitnum(p + 0*P9BITLEN); /* char count */ if (n < 0) return 0; h = p9bitnum(p + 1*P9BITLEN); /* height */ if (h < 0) return 0; a = p9bitnum(p + 2*P9BITLEN); /* ascent */ if (a < 0) return 0; return 1; } #define WHITESPACE(c) ((c) == ' ' || (c) == '\t' || (c) == '\n') int isp9font(void) { uchar *cp, *p; int i, n; char pathname[1024]; cp = buf; if (!getfontnum(cp, &cp)) /* height */ return 0; if (!getfontnum(cp, &cp)) /* ascent */ return 0; for (i = 0; cp=(uchar*)strchr((char*)cp, '\n'); i++) { if (!getfontnum(cp, &cp)) /* min */ break; if (!getfontnum(cp, &cp)) /* max */ return 0; getfontnum(cp, &cp); /* optional offset */ while (WHITESPACE(*cp)) cp++; for (p = cp; *cp && !WHITESPACE(*cp); cp++) ; /* construct a path name, if needed */ n = 0; if (*p != '/' && slash) { n = slash-fname+1; if (n < sizeof(pathname)) memcpy(pathname, fname, n); else n = 0; } if (n+cp-p+4 < sizeof(pathname)) { memcpy(pathname+n, p, cp-p); n += cp-p; pathname[n] = 0; if (access(pathname, AEXIST) < 0) { strcpy(pathname+n, ".0"); if (access(pathname, AEXIST) < 0) return 0; } } } if (i) { prtype(PLAIN, "font file"); return 1; } return 0; } int getfontnum(uchar *cp, uchar **rp) { while (WHITESPACE(*cp)) /* extract ulong delimited by whitespace */ cp++; if (*cp < '0' || *cp > '9') return 0; strtoul((char *)cp, (char **)rp, 0); if (!WHITESPACE(**rp)) { *rp = cp; return 0; } return 1; } int isrtf(void) { if(strstr((char *)buf, "\\rtf1")){ prtype("application/rtf", "rich text format"); return 1; } return 0; } int ismsdos(void) { if (buf[0] == 0x4d && buf[1] == 0x5a){ prtype("application/x-msdownload", "MSDOS executable"); return 1; } return 0; } int iself(void) { static char *cpu[] = { /* NB: incomplete and arbitary list */ [1] "WE32100", [2] "SPARC", [3] "i386", [4] "M68000", [5] "M88000", [6] "i486", [7] "i860", [8] "R3000", [9] "S370", [10] "R4000", [15] "HP-PA", [18] "sparc v8+", [19] "i960", [20] "PPC-32", [21] "PPC-64", [40] "ARM", [41] "Alpha", [43] "sparc v9", [50] "IA-64", [62] "AMD64", [75] "VAX", }; static char *type[] = { [1] "relocatable object", [2] "executable", [3] "shared library", [4] "core dump", }; if (memcmp(buf, "\x7fELF", 4) == 0){ if (!mime){ int isdifend = 0; int n = (buf[19] << 8) | buf[18]; char *p = "unknown"; char *t = "unknown"; if (n > 0 && n < nelem(cpu) && cpu[n]) p = cpu[n]; else { /* try the other byte order */ isdifend = 1; n = (buf[18] << 8) | buf[19]; if (n > 0 && n < nelem(cpu) && cpu[n]) p = cpu[n]; } if(isdifend) n = (buf[16]<< 8) | buf[17]; else n = (buf[17]<< 8) | buf[16]; if(n>0 && n < nelem(type) && type[n]) t = type[n]; print("%s ELF %s\n", p, t); } else prtype("application/x-elf-executable", nil); return 1; } return 0; } int isface(void) { int i, j, ldepth, l; char *p, *m; ldepth = -1; for(j = 0; j < 3; j++){ for(p = (char*)buf, i=0; i<3; i++){ if(p[0] != '0' || p[1] != 'x') return 0; if(buf[2+8] == ',') l = 2; else if(buf[2+4] == ',') l = 1; else return 0; if(ldepth == -1) ldepth = l; if(l != ldepth) return 0; strtoul(p, &p, 16); if(*p++ != ',') return 0; while(*p == ' ' || *p == '\t') p++; } if (*p++ != '\n') return 0; } m = smprint("face image depth %d", ldepth); prtype("application/x-face", m); free(m); return 1; }