/* Copyright (C) 2001-2006 Artifex Software, Inc.
   All Rights Reserved.

   This software is provided AS-IS with no warranty, either express or
   implied.

   This software is distributed under license and may not be copied, modified
   or distributed except as expressly authorized under the terms of that
   license.  Refer to licensing information at http://www.artifex.com/
   or contact Artifex Software, Inc.,  7 Mt. Lassen Drive - Suite A-134,
   San Rafael, CA  94903, U.S.A., +1(415)492-9861, for further information.
*/

/* $Id$ */
/* ASCII85Decode filter */
#include "std.h"
#include "strimpl.h"
#include "sa85d.h"
#include "scanchar.h"

/* ------ ASCII85Decode ------ */

private_st_A85D_state();

/* Initialize the state */
static int
s_A85D_init(stream_state * st)
{
    stream_A85D_state *const ss = (stream_A85D_state *) st;

    s_A85D_init_inline(ss);
    return 0;
}

/* Process a buffer */
static int a85d_finish(int, ulong, stream_cursor_write *);
static int
s_A85D_process(stream_state * st, stream_cursor_read * pr,
               stream_cursor_write * pw, bool last)
{
    stream_A85D_state *const ss = (stream_A85D_state *) st;
    register const byte *p = pr->ptr;
    register byte *q = pw->ptr;
    /* stop processing early unless the target is empty (last == true)  */
    /* to make sure we consume the EOD marker correctly. The EOD marker */
    /* might be as many as 6 characters after the last valid data char  */
    /* D <cr> <lf> '~' <cr> <lf> '>' where 'D' is a data character.     */
    const byte *rlimit = pr->limit - (last ? 0 : 7); /* max EOD len + 1 */
    const byte *r = max(p, rlimit);
    byte *wlimit = pw->limit;
    int ccount = ss->odd;
    ulong word = ss->word;
    int status = 0;

    /* scan to make sure that an EOD isn't fully contained in the */
    /* last part of the buffer (between rlimit and pr->limit).    */
    while (r < pr->limit) {
        if (*++r == '~')
            while (r < pr->limit)
                if (*++r == '>') {
                    /* we have both characters of a complete EOD. */
                    rlimit = pr->limit; /* go ahead and process everything */
                    r = rlimit;		/* break out of the while loops */
                    break;
                }
    }
    while (p < rlimit) {
        int ch = *++p;
        uint ccode = ch - '!';

        if (ccode < 85) {	/* catches ch < '!' as well */
            if (ccount == 4) {
                /*
                 * We've completed a 32-bit group.  Make sure we have
                 * room for it in the output.
                 */
                if (wlimit - q < 4) {
                    p--;
                    status = 1;
                    break;
                }
                /* Check for overflow condition, throw ioerror if so */
                if (word >= 0x03030303 && ccode > 0) {
                    status = ERRC;
                    break;
                }
                word = word * 85 + ccode;
                q[1] = (byte) (word >> 24);
                q[2] = (byte) (word >> 16);
                q[3] = (byte) ((uint) word >> 8);
                q[4] = (byte) word;
                q += 4;
                word = 0;
                ccount = 0;
            } else {
                word = word * 85 + ccode;
                ++ccount;
            }
        } else if (ch == 'z' && ccount == 0) {
            if (wlimit - q < 4) {
                p--;
                status = 1;
                break;
            }
            q[1] = q[2] = q[3] = q[4] = 0,
                q += 4;
        } else if (scan_char_decoder[ch] == ctype_space)
            DO_NOTHING;
        else if (ch == '~') {
            int i = 1;

            rlimit = pr->limit;		/* Here we use the real "limit" */
            /* Handle odd bytes. */
            if (p == rlimit) {
                if (last)
                    status = ERRC;
                else
                    p--;
                break;
            }
            if ((int)(wlimit - q) < ccount - 1) {
                status = 1;
                p--;
                break;
            }

            /* According to PLRM 3rd, if the A85 filter encounters '~',
             * the next character must be '>'.
             * And any other characters should raise an ioerror.
             * But Adobe Acrobat allows CR/LF between ~ and >.
             * So we allow CR/LF between them. */
            /* PDF further relaxes the requirements and accepts bare '~'.
             */
            while ((p[i] == 13 || p[i] == 10) && (p+i <= rlimit))
                i++;
            if (p[i] != '>') {
                if (ss->pdf_rules) {
                    if (p[i] == 13 || p[i] == 10) {
                        if (!last)
                            break;
                    } else {
                        p--;
                    }
                } else {
                    if (p+i == rlimit) {
                        if (last)
                            status = ERRC;
                        else
                            p--;	/* we'll see the '~' after filling the buffer */
                    }
                    break;
                }
            }
            p += i;		/* advance to the '>' */
            pw->ptr = q;
            status = a85d_finish(ccount, word, pw);
            q = pw->ptr;
            break;
        } else {		/* syntax error or exception */
            status = ERRC;
            break;
        }
    }
    pw->ptr = q;
    if (status == 0 && last) {
        if ((int)(wlimit - q) < ccount - 1)
            status = 1;
        else if (ss->require_eod)
            status = ERRC;
        else
            status = a85d_finish(ccount, word, pw);
    }
    pr->ptr = p;
    ss->odd = ccount;
    ss->word = word;
    return status;
}
/* Handle the end of input data. */
static int
a85d_finish(int ccount, ulong word, stream_cursor_write * pw)
{
    /* Assume there is enough room in the output buffer! */
    byte *q = pw->ptr;
    int status = EOFC;

    switch (ccount) {
        case 0:
            break;
        case 1:		/* syntax error */
            status = ERRC;
            break;
        case 2:		/* 1 odd byte */
            word = word * (85L * 85 * 85) + 85L * 85 * 85 - 1L;
            goto o1;
        case 3:		/* 2 odd bytes */
            word = word * (85L * 85) + 85L * 85L - 1L;
            goto o2;
        case 4:		/* 3 odd bytes */
            word = word * 85L + 84L;
            q[3] = (byte) (word >> 8);
o2:	    q[2] = (byte) (word >> 16);
o1:	    q[1] = (byte) (word >> 24);
            q += ccount - 1;
            pw->ptr = q;
    }
    return status;
}

/* Stream template */
const stream_template s_A85D_template = {
    &st_A85D_state, s_A85D_init, s_A85D_process, 2, 4
};