/src/ghostpdl/base/sa85d.c

Source
/* Copyright (C) 2001-2023 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;

            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;
            }
            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 <= 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: 2026-02-14 07:09

Line	Count	Source
1		/* Copyright (C) 2001-2023 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	686	{
31	686	stream_A85D_state const ss = (stream_A85D_state ) st;
32
33	686	s_A85D_init_inline(ss);
34	686	return 0;
35	686	}
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	729k	{
43	729k	stream_A85D_state const ss = (stream_A85D_state ) st;
44	729k	register const byte *p = pr->ptr;
45	729k	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	729k	const byte rlimit = pr->limit - (last ? 0 : 7); / max EOD len + 1 */
51	729k	const byte *r = max(p, rlimit);
52	729k	byte *wlimit = pw->limit;
53	729k	int ccount = ss->odd;
54	729k	ulong word = ss->word;
55	729k	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	1.63M	while (r < pr->limit) {
60	903k	if (*++r == '~')
61	4.90M	while (r < pr->limit)
62	4.18M	if (*++r == '>') {
63		/* we have both characters of a complete EOD. */
64	641	rlimit = pr->limit; /* go ahead and process everything */
65	641	r = rlimit; /* break out of the while loops */
66	641	break;
67	641	}
68	903k	}
69	7.22M	while (p < rlimit) {
70	7.21M	int ch = *++p;
71	7.21M	uint ccode = ch - '!';
72
73	7.21M	if (ccode < 85) { /* catches ch < '!' as well */
74	6.46M	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	1.29M	if (wlimit - q < 4) {
80	1.43k	p--;
81	1.43k	status = 1;
82	1.43k	break;
83	1.43k	}
84		/* Check for overflow condition, throw ioerror if so */
85	1.29M	if (word >= 0x03030303 && ccode > 0) {
86	184	status = ERRC;
87	184	break;
88	184	}
89	1.29M	word = word * 85 + ccode;
90	1.29M	q[1] = (byte) (word >> 24);
91	1.29M	q[2] = (byte) (word >> 16);
92	1.29M	q[3] = (byte) ((uint) word >> 8);
93	1.29M	q[4] = (byte) word;
94	1.29M	q += 4;
95	1.29M	word = 0;
96	1.29M	ccount = 0;
97	5.17M	} else {
98	5.17M	word = word * 85 + ccode;
99	5.17M	++ccount;
100	5.17M	}
101	6.46M	} else if (ch == 'z' && ccount == 0) {
102	696	if (wlimit - q < 4) {
103	0	p--;
104	0	status = 1;
105	0	break;
106	0	}
107	696	q[1] = q[2] = q[3] = q[4] = 0,
108	696	q += 4;
109	752k	} else if (scan_char_decoder[ch] == ctype_space)
110	752k	DO_NOTHING;
111	722k	else if (ch == '~') {
112	722k	int i = 1;
113
114	722k	rlimit = pr->limit; /* Here we use the real "limit" */
115		/* Handle odd bytes. */
116	722k	if (p == rlimit) {
117	22	if (!last)
118	0	p--;
119	22	else if (ss->pdf_rules)
120	0	goto finish;
121	22	else
122	22	status = ERRC;
123	22	break;
124	22	}
125	722k	if ((int)(wlimit - q) < ccount - 1) {
126	149	status = 1;
127	149	p--;
128	149	break;
129	149	}
130
131		/* According to PLRM 3rd, if the A85 filter encounters '~',
132		* the next character must be '>'.
133		* And any other characters should raise an ioerror.
134		* But Adobe Acrobat allows CR/LF between ~ and >.
135		* So we allow CR/LF between them. */
136		/* PDF further relaxes the requirements and accepts bare '~'.
137		*/
138	725k	while ((p + i <= rlimit) && (p[i] == 13 \|\| p[i] == 10))
139	3.00k	i++;
140	722k	if (p + i <= rlimit && p[i] != '>') {
141	721k	if (ss->pdf_rules) {
142	1	if (p[i] == 13 \|\| p[i] == 10) {
143	0	if (!last)
144	0	break;
145	1	} else {
146	1	p--;
147	1	}
148	721k	} else {
149	721k	if (p + i == rlimit) {
150	20	if (last)
151	20	status = ERRC;
152	0	else
153	0	p--; /* we'll see the '~' after filling the buffer */
154	20	}
155	721k	break;
156	721k	}
157	721k	}
158	260	finish:
159	260	if (p + i <= rlimit)
160	208	p += i; /* advance to the '>' */
161	52	else
162	52	p = rlimit; /* Can happen if the '>' is missing */
163	260	pw->ptr = q;
164	260	status = a85d_finish(ccount, word, pw);
165	260	q = pw->ptr;
166	260	break;
167	722k	} else { /* syntax error or exception */
168	507	status = ERRC;
169	507	break;
170	507	}
171	7.21M	}
172	729k	pw->ptr = q;
173	729k	if (status == 0 && last) {
174	690	if ((int)(wlimit - q) < ccount - 1)
175	0	status = 1;
176	690	else if (ss->require_eod)
177	682	status = ERRC;
178	8	else
179	8	status = a85d_finish(ccount, word, pw);
180	690	}
181	729k	pr->ptr = p;
182	729k	ss->odd = ccount;
183	729k	ss->word = word;
184	729k	return status;
185	729k	}
186		/* Handle the end of input data. */
187		static int
188		a85d_finish(int ccount, ulong word, stream_cursor_write * pw)
189	268	{
190		/* Assume there is enough room in the output buffer! */
191	268	byte *q = pw->ptr;
192	268	int status = EOFC;
193
194	268	switch (ccount) {
195	58	case 0:
196	58	break;
197	30	case 1: /* syntax error */
198	30	status = ERRC;
199	30	break;
200	59	case 2: /* 1 odd byte */
201	59	word = word * (85L * 85 * 85) + 85L * 85 * 85 - 1L;
202	59	goto o1;
203	65	case 3: /* 2 odd bytes */
204	65	word = word * (85L * 85) + 85L * 85L - 1L;
205	65	goto o2;
206	56	case 4: /* 3 odd bytes */
207	56	word = word * 85L + 84L;
208	56	q[3] = (byte) (word >> 8);
209	121	o2: q[2] = (byte) (word >> 16);
210	180	o1: q[1] = (byte) (word >> 24);
211	180	q += ccount - 1;
212	180	pw->ptr = q;
213	268	}
214	268	return status;
215	268	}
216
217		/* Stream template */
218		const stream_template s_A85D_template = {
219		&st_A85D_state, s_A85D_init, s_A85D_process, 2, 4
220		};