X-Git-Url: http://git.marmaro.de/?p=mmh;a=blobdiff_plain;f=sbr%2Fm_getfld.c;h=1ec09a4e761eaf0ddeee4f71b82282054ee2e6e7;hp=8fd836430b4d6ef07ae9914236e95acf285e2aaa;hb=b6823062c9806496b39ff12bf1fbe43ae127c96c;hpb=9f9bfa91d88b3ca4cd9206c1f2b17c6d13ebcaba diff --git a/sbr/m_getfld.c b/sbr/m_getfld.c index 8fd8364..1ec09a4 100644 --- a/sbr/m_getfld.c +++ b/sbr/m_getfld.c @@ -7,55 +7,23 @@ */ #include -#include #include +#include +#include /* -** This module has a long and checkered history. First, it didn't burst -** maildrops correctly because it considered two CTRL-A:s in a row to be -** an inter-message delimiter. It really is four CTRL-A:s followed by a -** newline. Unfortunately, MMDF will convert this delimiter *inside* a -** message to a CTRL-B followed by three CTRL-A:s and a newline. This -** caused the old version of m_getfld() to declare eom prematurely. The -** fix was a lot slower than +** This module has a long and checkered history. ** -** c == '\001' && peekc (iob) == '\001' +** [ Here had been some history of delimiter problems in MMDF maildrops ... ] ** -** but it worked, and to increase generality, MBOX style maildrops could -** be parsed as well. Unfortunately the speed issue finally caught up with -** us since this routine is at the very heart of MH. +** Unfortunately the speed issue finally caught up with us since this +** routine is at the very heart of MH. To speed things up considerably, the +** routine Eom() was made an auxilary function called by the macro eom(). +** Unless we are bursting a maildrop, the eom() macro returns FALSE saying +** we aren't at the end of the message. ** -** To speed things up considerably, the routine Eom() was made an auxilary -** function called by the macro eom(). Unless we are bursting a maildrop, -** the eom() macro returns FALSE saying we aren't at the end of the -** message. +** [ ... and here had been some more of it. ] ** -** The next thing to do is to read the mts.conf file and initialize -** delimiter[] and delimlen accordingly... -** -** After mhl was made a built-in in msh, m_getfld() worked just fine -** (using m_unknown() at startup). Until one day: a message which was -** the result of a bursting was shown. Then, since the burst boundaries -** aren't CTRL-A:s, m_getfld() would blinding plunge on past the boundary. -** Very sad. The solution: introduce m_eomsbr(). This hook gets called -** after the end of each line (since testing for eom involves an fseek()). -** This worked fine, until one day: a message with no body portion arrived. -** Then the -** -** while (eom(c = getc(iob), iob)) -** continue; -** -** loop caused m_getfld() to return FMTERR. So, that logic was changed to -** check for (*eom_action) and act accordingly. -** -** [ Note by meillo 2011-10: -** as msh was removed from mmh, m_eomsbr() became irrelevant. ] -** -** This worked fine, until one day: someone didn't use four CTRL:A's as -** their delimiters. So, the bullet got bit and we read mts.h and -** continue to struggle on. It's not that bad though, since the only time -** the code gets executed is when inc (or msh) calls it, and both of these -** have already called mts_init(). ** ** ------------------------ ** (Written by Van Jacobson for the mh6 m_getfld, January, 1986): @@ -77,9 +45,8 @@ ** If you hack on this and slow it down, I, my children and my ** children's children will curse you. ** -** This routine gets used on three different types of files: normal, -** single msg files, "packed" unix or mmdf mailboxs (when used by inc) -** and packed, directoried bulletin board files (when used by msh). +** This routine gets used on two different types of files: normal, +** single msg files and "packed" unix mailboxs (when used by inc). ** The biggest impact of different file types is in "eom" testing. The ** code has been carefully organized to test for eom at appropriate ** times and at no other times (since the check is quite expensive). @@ -145,24 +112,20 @@ static int m_Eom(int, FILE *); static unsigned char *matchc(int, char *, int, char *); static unsigned char *locc(int, unsigned char *, unsigned char); -#define eom(c,iob) (msg_style != MS_DEFAULT && \ +#define eom(c,iob) (ismbox && \ (((c) == *msg_delim && m_Eom(c,iob)) ||\ (eom_action && (*eom_action)(c)))) static unsigned char **pat_map; /* -** defined in sbr/m_msgdef.c = 0 ** This is a disgusting hack for "inc" so it can know how many ** characters were stuffed in the buffer on the last call ** (see comments in uip/scansbr.c). */ -extern int msg_count; +int msg_count = 0; -/* -** defined in sbr/m_msgdef.c = MS_DEFAULT -*/ -extern int msg_style; +int ismbox = FALSE; /* ** The "full" delimiter string for a packed maildrop consists @@ -176,7 +139,8 @@ extern int msg_style; ** is used in m_Eom because the first character of the string ** has been read and matched before m_Eom is called. */ -extern char *msg_delim; /* defined in sbr/m_msgdef.c = "" */ +static char *msg_delim = ""; + static unsigned char *fdelim; static unsigned char *delimend; static int fdelimlen; @@ -192,14 +156,6 @@ static int (*eom_action)(int) = NULL; # define DEFINED__FILBUF_TO_SOMETHING_SPECIFIC #endif -#ifdef SCO_5_STDIO -# define _ptr __ptr -# define _cnt __cnt -# define _base __base -# define _filbuf(fp) ((fp)->__cnt = 0, __filbuf(fp)) -# define DEFINED__FILBUF_TO_SOMETHING_SPECIFIC -#endif - #ifndef DEFINED__FILBUF_TO_SOMETHING_SPECIFIC extern int _filbuf(FILE*); #endif @@ -209,8 +165,8 @@ int m_getfld(int state, unsigned char *name, unsigned char *buf, int bufsz, FILE *iob) { - register unsigned char *bp, *cp, *ep, *sp; - register int cnt, c, i, j; + unsigned char *bp, *cp, *ep, *sp; + int cnt, c, i, j; if ((c = getc(iob)) < 0) { msg_count = 0; @@ -231,8 +187,6 @@ m_getfld(int state, unsigned char *name, unsigned char *buf, } switch (state) { - case FLDEOF: - case BODYEOF: case FLD: if (c == '\n' || c == '-') { /* we hit the header/body separator */ @@ -240,7 +194,7 @@ m_getfld(int state, unsigned char *name, unsigned char *buf, ; if (c < 0 || (c = getc(iob)) < 0 || eom(c, iob)) { - if (! eom_action) { + if (!eom_action) { /* flush null messages */ while ((c = getc(iob)) >= 0 && eom(c, iob)) ; @@ -469,8 +423,8 @@ m_getfld(int state, unsigned char *name, unsigned char *buf, *cp++ = j = *(iob->_ptr + c); c = _filbuf(iob); #endif - if (c == EOF || - ((j == '\0' || j == '\n') && c != ' ' && c != '\t')) { + if (c == EOF || ((j == '\0' || j == '\n') + && c != ' ' && c != '\t')) { if (c != EOF) { #ifdef LINUX_STDIO --iob->_IO_read_ptr; @@ -509,7 +463,7 @@ m_getfld(int state, unsigned char *name, unsigned char *buf, cnt = ++iob->_cnt; #endif c = (cnt < i ? cnt : i); - if (msg_style != MS_DEFAULT && c > 1) { + if (ismbox && c > 1) { /* ** packed maildrop - only take up to the (possible) ** start of the next message. This "matchc" should @@ -608,7 +562,7 @@ m_getfld(int state, unsigned char *name, unsigned char *buf, break; default: - adios(NULL, "m_getfld() called with bogus state of %d", state); + adios(EX_SOFTWARE, NULL, "m_getfld() called with bogus state of %d", state); } finish: *cp = 0; @@ -617,58 +571,45 @@ finish: } -#ifdef RPATHS -static char unixbuf[BUFSIZ] = ""; -#endif /* RPATHS */ - void -m_unknown(FILE *iob) +thisisanmbox(FILE *iob) { - register int c; - register long pos; + int c; char text[10]; - register char *cp; - register char *delimstr; + char *cp; + char *delimstr; -/* -** Figure out what the message delimitter string is for this -** maildrop. (This used to be part of m_Eom but I didn't like -** the idea of an "if" statement that could only succeed on the -** first call to m_Eom getting executed on each call, i.e., at -** every newline in the message). -** -** If the first line of the maildrop is a Unix "From " line, we -** say the style is MBOX and eat the rest of the line. Otherwise -** we say the style is MMDF and look for the delimiter string -** specified when nmh was built (or from the mts.conf file). -*/ + c = getc(iob); + if (feof(iob)) { + return; + } + ungetc(c, iob); - msg_style = MS_UNKNOWN; + /* + ** Figure out what the message delimitter string is for this + ** maildrop. (This used to be part of m_Eom but I didn't like + ** the idea of an "if" statement that could only succeed on the + ** first call to m_Eom getting executed on each call, i.e., at + ** every newline in the message). + ** + ** If the first line of the maildrop is a Unix "From " line, we + ** say the style is MBOX and eat the rest of the line. Otherwise + ** abort. + */ - pos = ftell(iob); - if (fread(text, sizeof(*text), 5, iob) == 5 - && strncmp(text, "From ", 5) == 0) { - msg_style = MS_MBOX; - delimstr = "\nFrom "; -#ifndef RPATHS - while ((c = getc(iob)) != '\n' && c >= 0) - ; -#else /* RPATHS */ - cp = unixbuf; - while ((c = getc(iob)) != '\n' && cp - unixbuf < BUFSIZ - 1) - *cp++ = c; - *cp = 0; -#endif /* RPATHS */ - } else { - /* not a Unix style maildrop */ - fseek(iob, pos, SEEK_SET); - if (mmdlm2 == NULL || *mmdlm2 == 0) - mmdlm2 = "\001\001\001\001\n"; - delimstr = mmdlm2; - msg_style = MS_MMDF; + if (fread(text, sizeof(*text), 5, iob) != 5) { + adios(EX_IOERR, NULL, "Read error"); + } + if (strncmp(text, "From ", 5)!=0) { + adios(EX_USAGE, NULL, "No Unix style (mbox) maildrop."); + } + ismbox = TRUE; + delimstr = "\nFrom "; + while ((c = getc(iob)) != '\n' && c >= 0) { + continue; } c = strlen(delimstr); - fdelim = (unsigned char *) mh_xmalloc((size_t) (c + 3)); + fdelim = mh_xcalloc(c + 3, sizeof(char)); *fdelim++ = '\0'; *fdelim = '\n'; msg_delim = (char *)fdelim+1; @@ -678,25 +619,17 @@ m_unknown(FILE *iob) strcpy(msg_delim, delimstr); delimend = (unsigned char *)msg_delim + edelimlen; if (edelimlen <= 1) - adios(NULL, "maildrop delimiter must be at least 2 bytes"); + adios(EX_DATAERR, NULL, "maildrop delimiter must be at least 2 bytes"); /* ** build a Boyer-Moore end-position map for the matcher in m_getfld. ** N.B. - we don't match just the first char (since it's the newline ** separator) or the last char (since the matchc would have found it ** if it was a real delim). */ - pat_map = (unsigned char **) calloc(256, sizeof(unsigned char *)); + pat_map = mh_xcalloc(256, sizeof(unsigned char *)); for (cp = (char *) fdelim + 1; cp < (char *) delimend; cp++ ) pat_map[(unsigned char)*cp] = (unsigned char *) cp; - - if (msg_style == MS_MMDF) { - /* flush extra msg hdrs */ - while ((c = getc(iob)) >= 0 && eom(c, iob)) - ; - if (c >= 0) - ungetc(c, iob); - } } @@ -707,17 +640,14 @@ m_unknown(FILE *iob) static int m_Eom(int c, FILE *iob) { - register long pos = 0L; - register int i; + long pos = 0L; + int i; char text[10]; -#ifdef RPATHS - register char *cp; -#endif /* RPATHS */ pos = ftell(iob); - if ((i = fread(text, sizeof *text, edelimlen, iob)) != edelimlen - || (strncmp(text, (char *)edelim, edelimlen)!=0)) { - if (i == 0 && msg_style == MS_MBOX) + if ((i = fread(text, sizeof *text, edelimlen, iob)) != edelimlen || + (strncmp(text, (char *)edelim, edelimlen)!=0)) { + if (i == 0 && ismbox) /* ** the final newline in the (brain damaged) unix-format ** maildrop is part of the delimitter - delete it. @@ -729,92 +659,24 @@ m_Eom(int c, FILE *iob) return 0; } - if (msg_style == MS_MBOX) { -#ifndef RPATHS - while ((c = getc(iob)) != '\n') - if (c < 0) - break; -#else /* RPATHS */ - cp = unixbuf; - while ((c = getc(iob)) != '\n' && c >= 0 && cp - unixbuf < BUFSIZ - 1) - *cp++ = c; - *cp = 0; -#endif /* RPATHS */ - } - - return 1; -} - - -#ifdef RPATHS -/* -** Return the Return-Path and Delivery-Date -** header information. -** -** Currently, I'm assuming that the "From " line -** takes one of the following forms. -** -** From sender date remote from host (for UUCP delivery) -** From sender@host date (for sendmail delivery) -*/ - -int -get_returnpath(char *rp, int rplen, char *dd, int ddlen) -{ - char *ap, *bp, *cp, *dp; - - ap = unixbuf; - if (!(bp = cp = strchr(ap, ' '))) - return 0; - - /* - ** Check for "remote from" in envelope to see - ** if this message uses UUCP style addressing - */ - while ((cp = strchr(++cp, 'r'))) { - if (strncmp(cp, "remote from", 11) == 0) { - cp = strrchr(cp, ' '); - break; + if (ismbox) { + while ((c = getc(iob)) != '\n' && c >= 0) { + continue; } } - /* - ** Get the Return-Path information from - ** the "From " envelope. - */ - if (cp) { - /* return path for UUCP style addressing */ - dp = strchr(++cp, '\n'); - snprintf(rp, rplen, "%.*s!%.*s\n", (int)(dp - cp), cp, (int)(bp - ap), ap); - } else { - /* return path for standard domain addressing */ - snprintf(rp, rplen, "%.*s\n", (int)(bp - ap), ap); - } - - /* - ** advance over the spaces to get to - ** delivery date on envelope - */ - while (*bp == ' ') - bp++; - - /* Now get delivery date from envelope */ - snprintf(dd, ddlen, "%.*s\n", 24, bp); - - unixbuf[0] = 0; return 1; } -#endif /* RPATHS */ static unsigned char * matchc(int patln, char *pat, int strln, char *str) { - register char *es = str + strln - patln; - register char *sp; - register char *pp; - register char *ep = pat + patln; - register char pc = *pat++; + char *es = str + strln - patln; + char *sp; + char *pp; + char *ep = pat + patln; + char pc = *pat++; for(;;) { while (pc != *str++)