/src/ghostpdl/base/sa85d.c

Source
/* Copyright (C) 2001-2026 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 the license contained in the file LICENSE in this distribution.

   Refer to licensing information at http://www.artifex.com or contact
   Artifex Software, Inc.,  39 Mesa Street, Suite 108A, San Francisco,
   CA 94129, USA, for further information.
*/


/* 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;

            if ((int)(wlimit - q) < ccount - 1) {
                status = 1;
                p--;
                break;
            }

            rlimit = pr->limit;   /* Here we use the real "limit" */
            /* Handle odd bytes. */
            if (p == rlimit) {
                if (!last)
                  p--;
                else if (ss->pdf_rules)
                  goto finish;
                else
                    status = ERRC;
                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 <= rlimit) && (p[i] == 13 || p[i] == 10))
                i++;
            if (p + i <= rlimit && 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;
                }
            }
          finish:
            if (p + i <= rlimit)
                p += i;    /* advance to the '>' */
            else
                p = rlimit; /* Can happen if the '>' is missing */
            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
};

Coverage Report

Created: 2026-09-14 07:34

Line	Count	Source
1		/* Copyright (C) 2001-2026 Artifex Software, Inc.
2		All Rights Reserved.
3
4		This software is provided AS-IS with no warranty, either express or
5		implied.
6
7		This software is distributed under license and may not be copied,
8		modified or distributed except as expressly authorized under the terms
9		of the license contained in the file LICENSE in this distribution.
10
11		Refer to licensing information at http://www.artifex.com or contact
12		Artifex Software, Inc., 39 Mesa Street, Suite 108A, San Francisco,
13		CA 94129, USA, for further information.
14		*/
15
16
17		/* ASCII85Decode filter */
18		#include "std.h"
19		#include "strimpl.h"
20		#include "sa85d.h"
21		#include "scanchar.h"
22
23		/* ------ ASCII85Decode ------ */
24
25		private_st_A85D_state();
26
27		/* Initialize the state */
28		static int
29		s_A85D_init(stream_state * st)
30	804	{
31	804	stream_A85D_state const ss = (stream_A85D_state ) st;
32
33	804	s_A85D_init_inline(ss);
34	804	return 0;
35	804	}
36
37		/* Process a buffer */
38		static int a85d_finish(int, ulong, stream_cursor_write *);
39		static int
40		s_A85D_process(stream_state * st, stream_cursor_read * pr,
41		stream_cursor_write * pw, bool last)
42	534k	{
43	534k	stream_A85D_state const ss = (stream_A85D_state ) st;
44	534k	register const byte *p = pr->ptr;
45	534k	register byte *q = pw->ptr;
46		/* stop processing early unless the target is empty (last == true) */
47		/* to make sure we consume the EOD marker correctly. The EOD marker */
48		/* might be as many as 6 characters after the last valid data char */
49		/* D <cr> <lf> '~' <cr> <lf> '>' where 'D' is a data character. */
50	534k	const byte rlimit = pr->limit - (last ? 0 : 7); / max EOD len + 1 */
51	534k	const byte *r = max(p, rlimit);
52	534k	byte *wlimit = pw->limit;
53	534k	int ccount = ss->odd;
54	534k	ulong word = ss->word;
55	534k	int status = 0;
56
57		/* scan to make sure that an EOD isn't fully contained in the */
58		/* last part of the buffer (between rlimit and pr->limit). */
59	1.20M	while (r < pr->limit) {
60	674k	if (*++r == '~')
61	3.57M	while (r < pr->limit)
62	3.05M	if (*++r == '>') {
63		/* we have both characters of a complete EOD. */
64	407	rlimit = pr->limit; /* go ahead and process everything */
65	407	r = rlimit; /* break out of the while loops */
66	407	break;
67	407	}
68	674k	}
69	5.98M	while (p < rlimit) {
70	5.97M	int ch = *++p;
71	5.97M	uint ccode = ch - '!';
72
73	5.97M	if (ccode < 85) { /* catches ch < '!' as well */
74	5.42M	if (ccount == 4) {
75		/*
76		* We've completed a 32-bit group. Make sure we have
77		* room for it in the output.
78		*/
79	1.08M	if (wlimit - q < 4) {
80	1.26k	p--;
81	1.26k	status = 1;
82	1.26k	break;
83	1.26k	}
84		/* Check for overflow condition, throw ioerror if so */
85	1.08M	if (word >= 0x03030303 && ccode > 0) {
86	183	status = ERRC;
87	183	break;
88	183	}
89	1.08M	word = word * 85 + ccode;
90	1.08M	q[1] = (byte) (word >> 24);
91	1.08M	q[2] = (byte) (word >> 16);
92	1.08M	q[3] = (byte) ((uint) word >> 8);
93	1.08M	q[4] = (byte) word;
94	1.08M	q += 4;
95	1.08M	word = 0;
96	1.08M	ccount = 0;
97	4.34M	} else {
98	4.34M	word = word * 85 + ccode;
99	4.34M	++ccount;
100	4.34M	}
101	5.42M	} else if (ch == 'z' && ccount == 0) {
102	497	if (wlimit - q < 4) {
103	0	p--;
104	0	status = 1;
105	0	break;
106	0	}
107	497	q[1] = q[2] = q[3] = q[4] = 0,
108	497	q += 4;
109	549k	} else if (scan_char_decoder[ch] == ctype_space)
110	549k	DO_NOTHING;
111	528k	else if (ch == '~') {
112	527k	int i = 1;
113
114	527k	if ((int)(wlimit - q) < ccount - 1) {
115	114	status = 1;
116	114	p--;
117	114	break;
118	114	}
119
120	527k	rlimit = pr->limit; /* Here we use the real "limit" */
121		/* Handle odd bytes. */
122	527k	if (p == rlimit) {
123	17	if (!last)
124	0	p--;
125	17	else if (ss->pdf_rules)
126	0	goto finish;
127	17	else
128	17	status = ERRC;
129	17	break;
130	17	}
131
132		/* According to PLRM 3rd, if the A85 filter encounters '~',
133		* the next character must be '>'.
134		* And any other characters should raise an ioerror.
135		* But Adobe Acrobat allows CR/LF between ~ and >.
136		* So we allow CR/LF between them. */
137		/* PDF further relaxes the requirements and accepts bare '~'.
138		*/
139	529k	while ((p + i <= rlimit) && (p[i] == 13 \|\| p[i] == 10))
140	1.43k	i++;
141	527k	if (p + i <= rlimit && p[i] != '>') {
142	527k	if (ss->pdf_rules) {
143	2	if (p[i] == 13 \|\| p[i] == 10) {
144	0	if (!last)
145	0	break;
146	2	} else {
147	2	p--;
148	2	}
149	527k	} else {
150	527k	if (p + i == rlimit) {
151	11	if (last)
152	11	status = ERRC;
153	0	else
154	0	p--; /* we'll see the '~' after filling the buffer */
155	11	}
156	527k	break;
157	527k	}
158	527k	}
159	343	finish:
160	343	if (p + i <= rlimit)
161	307	p += i; /* advance to the '>' */
162	36	else
163	36	p = rlimit; /* Can happen if the '>' is missing */
164	343	pw->ptr = q;
165	343	status = a85d_finish(ccount, word, pw);
166	343	q = pw->ptr;
167	343	break;
168	527k	} else { /* syntax error or exception */
169	439	status = ERRC;
170	439	break;
171	439	}
172	5.97M	}
173	534k	pw->ptr = q;
174	534k	if (status == 0 && last) {
175	497	if ((int)(wlimit - q) < ccount - 1)
176	0	status = 1;
177	497	else if (ss->require_eod)
178	490	status = ERRC;
179	7	else
180	7	status = a85d_finish(ccount, word, pw);
181	497	}
182	534k	pr->ptr = p;
183	534k	ss->odd = ccount;
184	534k	ss->word = word;
185	534k	return status;
186	534k	}
187		/* Handle the end of input data. */
188		static int
189		a85d_finish(int ccount, ulong word, stream_cursor_write * pw)
190	350	{
191		/* Assume there is enough room in the output buffer! */
192	350	byte *q = pw->ptr;
193	350	int status = EOFC;
194
195	350	switch (ccount) {
196	67	case 0:
197	67	break;
198	20	case 1: /* syntax error */
199	20	status = ERRC;
200	20	break;
201	67	case 2: /* 1 odd byte */
202	67	word = word * (85L * 85 * 85) + 85L * 85 * 85 - 1L;
203	67	goto o1;
204	112	case 3: /* 2 odd bytes */
205	112	word = word * (85L * 85) + 85L * 85L - 1L;
206	112	goto o2;
207	84	case 4: /* 3 odd bytes */
208	84	word = word * 85L + 84L;
209	84	q[3] = (byte) (word >> 8);
210	196	o2: q[2] = (byte) (word >> 16);
211	263	o1: q[1] = (byte) (word >> 24);
212	263	q += ccount - 1;
213	263	pw->ptr = q;
214	350	}
215	350	return status;
216	350	}
217
218		/* Stream template */
219		const stream_template s_A85D_template = {
220		&st_A85D_state, s_A85D_init, s_A85D_process, 2, 4
221		};