/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: 2025-11-16 07:40

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	1.21k	{
31	1.21k	stream_A85D_state const ss = (stream_A85D_state ) st;
32
33	1.21k	s_A85D_init_inline(ss);
34	1.21k	return 0;
35	1.21k	}
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	625k	{
43	625k	stream_A85D_state const ss = (stream_A85D_state ) st;
44	625k	register const byte *p = pr->ptr;
45	625k	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	625k	const byte rlimit = pr->limit - (last ? 0 : 7); / max EOD len + 1 */
51	625k	const byte *r = max(p, rlimit);
52	625k	byte *wlimit = pw->limit;
53	625k	int ccount = ss->odd;
54	625k	ulong word = ss->word;
55	625k	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	1.41M	while (r < pr->limit) {
60	784k	if (*++r == '~')
61	4.18M	while (r < pr->limit)
62	3.57M	if (*++r == '>') {
63		/* we have both characters of a complete EOD. */
64	553	rlimit = pr->limit; /* go ahead and process everything */
65	553	r = rlimit; /* break out of the while loops */
66	553	break;
67	553	}
68	784k	}
69	6.96M	while (p < rlimit) {
70	6.95M	int ch = *++p;
71	6.95M	uint ccode = ch - '!';
72
73	6.95M	if (ccode < 85) { /* catches ch < '!' as well */
74	6.31M	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	1.26M	if (wlimit - q < 4) {
80	1.35k	p--;
81	1.35k	status = 1;
82	1.35k	break;
83	1.35k	}
84		/* Check for overflow condition, throw ioerror if so */
85	1.26M	if (word >= 0x03030303 && ccode > 0) {
86	237	status = ERRC;
87	237	break;
88	237	}
89	1.26M	word = word * 85 + ccode;
90	1.26M	q[1] = (byte) (word >> 24);
91	1.26M	q[2] = (byte) (word >> 16);
92	1.26M	q[3] = (byte) ((uint) word >> 8);
93	1.26M	q[4] = (byte) word;
94	1.26M	q += 4;
95	1.26M	word = 0;
96	1.26M	ccount = 0;
97	5.04M	} else {
98	5.04M	word = word * 85 + ccode;
99	5.04M	++ccount;
100	5.04M	}
101	6.31M	} else if (ch == 'z' && ccount == 0) {
102	507	if (wlimit - q < 4) {
103	0	p--;
104	0	status = 1;
105	0	break;
106	0	}
107	507	q[1] = q[2] = q[3] = q[4] = 0,
108	507	q += 4;
109	647k	} else if (scan_char_decoder[ch] == ctype_space)
110	647k	DO_NOTHING;
111	618k	else if (ch == '~') {
112	618k	int i = 1;
113
114	618k	rlimit = pr->limit; /* Here we use the real "limit" */
115		/* Handle odd bytes. */
116	618k	if (p == rlimit) {
117	19	if (!last)
118	0	p--;
119	19	else if (ss->pdf_rules)
120	0	goto finish;
121	19	else
122	19	status = ERRC;
123	19	break;
124	19	}
125	618k	if ((int)(wlimit - q) < ccount - 1) {
126	134	status = 1;
127	134	p--;
128	134	break;
129	134	}
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	620k	while ((p + i <= rlimit) && (p[i] == 13 \|\| p[i] == 10))
139	2.87k	i++;
140	618k	if (p + i <= rlimit && p[i] != '>') {
141	617k	if (ss->pdf_rules) {
142	2	if (p[i] == 13 \|\| p[i] == 10) {
143	0	if (!last)
144	0	break;
145	2	} else {
146	2	p--;
147	2	}
148	617k	} else {
149	617k	if (p + i == rlimit) {
150	19	if (last)
151	19	status = ERRC;
152	0	else
153	0	p--; /* we'll see the '~' after filling the buffer */
154	19	}
155	617k	break;
156	617k	}
157	617k	}
158	622	finish:
159	622	if (p + i <= rlimit)
160	577	p += i; /* advance to the '>' */
161	45	else
162	45	p = rlimit; /* Can happen if the '>' is missing */
163	622	pw->ptr = q;
164	622	status = a85d_finish(ccount, word, pw);
165	622	q = pw->ptr;
166	622	break;
167	618k	} else { /* syntax error or exception */
168	606	status = ERRC;
169	606	break;
170	606	}
171	6.95M	}
172	625k	pw->ptr = q;
173	625k	if (status == 0 && last) {
174	595	if ((int)(wlimit - q) < ccount - 1)
175	0	status = 1;
176	595	else if (ss->require_eod)
177	584	status = ERRC;
178	11	else
179	11	status = a85d_finish(ccount, word, pw);
180	595	}
181	625k	pr->ptr = p;
182	625k	ss->odd = ccount;
183	625k	ss->word = word;
184	625k	return status;
185	625k	}
186		/* Handle the end of input data. */
187		static int
188		a85d_finish(int ccount, ulong word, stream_cursor_write * pw)
189	633	{
190		/* Assume there is enough room in the output buffer! */
191	633	byte *q = pw->ptr;
192	633	int status = EOFC;
193
194	633	switch (ccount) {
195	131	case 0:
196	131	break;
197	25	case 1: /* syntax error */
198	25	status = ERRC;
199	25	break;
200	106	case 2: /* 1 odd byte */
201	106	word = word * (85L * 85 * 85) + 85L * 85 * 85 - 1L;
202	106	goto o1;
203	204	case 3: /* 2 odd bytes */
204	204	word = word * (85L * 85) + 85L * 85L - 1L;
205	204	goto o2;
206	167	case 4: /* 3 odd bytes */
207	167	word = word * 85L + 84L;
208	167	q[3] = (byte) (word >> 8);
209	371	o2: q[2] = (byte) (word >> 16);
210	477	o1: q[1] = (byte) (word >> 24);
211	477	q += ccount - 1;
212	477	pw->ptr = q;
213	633	}
214	633	return status;
215	633	}
216
217		/* Stream template */
218		const stream_template s_A85D_template = {
219		&st_A85D_state, s_A85D_init, s_A85D_process, 2, 4
220		};