xref: /freebsd/bin/dd/conv.c (revision 2e3507c25e42292b45a5482e116d278f5515d04d)
1 /*-
2  * SPDX-License-Identifier: BSD-3-Clause
3  *
4  * Copyright (c) 1991, 1993, 1994
5  *	The Regents of the University of California.  All rights reserved.
6  *
7  * This code is derived from software contributed to Berkeley by
8  * Keith Muller of the University of California, San Diego and Lance
9  * Visser of Convex Computer Corporation.
10  *
11  * Redistribution and use in source and binary forms, with or without
12  * modification, are permitted provided that the following conditions
13  * are met:
14  * 1. Redistributions of source code must retain the above copyright
15  *    notice, this list of conditions and the following disclaimer.
16  * 2. Redistributions in binary form must reproduce the above copyright
17  *    notice, this list of conditions and the following disclaimer in the
18  *    documentation and/or other materials provided with the distribution.
19  * 3. Neither the name of the University nor the names of its contributors
20  *    may be used to endorse or promote products derived from this software
21  *    without specific prior written permission.
22  *
23  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
24  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
25  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
26  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
27  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
28  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
29  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
30  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
31  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
32  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33  * SUCH DAMAGE.
34  */
35 
36 #include <sys/param.h>
37 
38 #include <err.h>
39 #include <inttypes.h>
40 #include <string.h>
41 
42 #include "dd.h"
43 #include "extern.h"
44 
45 /*
46  * def --
47  * Copy input to output.  Input is buffered until reaches obs, and then
48  * output until less than obs remains.  Only a single buffer is used.
49  * Worst case buffer calculation is (ibs + obs - 1).
50  */
51 void
52 def(void)
53 {
54 	u_char *inp;
55 	const u_char *t;
56 	size_t cnt;
57 
58 	if ((t = ctab) != NULL)
59 		for (inp = in.dbp - (cnt = in.dbrcnt); cnt--; ++inp)
60 			*inp = t[*inp];
61 
62 	/* Make the output buffer look right. */
63 	out.dbp = in.dbp;
64 	out.dbcnt = in.dbcnt;
65 
66 	if (in.dbcnt >= out.dbsz) {
67 		/* If the output buffer is full, write it. */
68 		dd_out(0);
69 
70 		/*
71 		 * dd_out copies the leftover output to the beginning of
72 		 * the buffer and resets the output buffer.  Reset the
73 		 * input buffer to match it.
74 	 	 */
75 		in.dbp = out.dbp;
76 		in.dbcnt = out.dbcnt;
77 	}
78 }
79 
80 void
81 def_close(void)
82 {
83 	/* Just update the count, everything is already in the buffer. */
84 	if (in.dbcnt)
85 		out.dbcnt = in.dbcnt;
86 }
87 
88 /*
89  * Copy variable length newline terminated records with a max size cbsz
90  * bytes to output.  Records less than cbs are padded with spaces.
91  *
92  * max in buffer:  MAX(ibs, cbsz)
93  * max out buffer: obs + cbsz
94  */
95 void
96 block(void)
97 {
98 	u_char *inp, *outp;
99 	const u_char *t;
100 	size_t cnt, maxlen;
101 	static int intrunc;
102 	int ch;
103 
104 	/*
105 	 * Record truncation can cross block boundaries.  If currently in a
106 	 * truncation state, keep tossing characters until reach a newline.
107 	 * Start at the beginning of the buffer, as the input buffer is always
108 	 * left empty.
109 	 */
110 	if (intrunc) {
111 		for (inp = in.db, cnt = in.dbrcnt; cnt && *inp++ != '\n'; --cnt)
112 			;
113 		if (!cnt) {
114 			in.dbcnt = 0;
115 			in.dbp = in.db;
116 			return;
117 		}
118 		intrunc = 0;
119 		/* Adjust the input buffer numbers. */
120 		in.dbcnt = cnt - 1;
121 		in.dbp = inp + cnt - 1;
122 	}
123 
124 	/*
125 	 * Copy records (max cbsz size chunks) into the output buffer.  The
126 	 * translation is done as we copy into the output buffer.
127 	 */
128 	ch = 0;
129 	for (inp = in.dbp - in.dbcnt, outp = out.dbp; in.dbcnt;) {
130 		maxlen = MIN(cbsz, (size_t)in.dbcnt);
131 		if ((t = ctab) != NULL)
132 			for (cnt = 0; cnt < maxlen && (ch = *inp++) != '\n';
133 			    ++cnt)
134 				*outp++ = t[ch];
135 		else
136 			for (cnt = 0; cnt < maxlen && (ch = *inp++) != '\n';
137 			    ++cnt)
138 				*outp++ = ch;
139 		/*
140 		 * Check for short record without a newline.  Reassemble the
141 		 * input block.
142 		 */
143 		if (ch != '\n' && (size_t)in.dbcnt < cbsz) {
144 			(void)memmove(in.db, in.dbp - in.dbcnt, in.dbcnt);
145 			break;
146 		}
147 
148 		/* Adjust the input buffer numbers. */
149 		in.dbcnt -= cnt;
150 		if (ch == '\n')
151 			--in.dbcnt;
152 
153 		/* Pad short records with spaces. */
154 		if (cnt < cbsz)
155 			(void)memset(outp, ctab ? ctab[' '] : ' ', cbsz - cnt);
156 		else {
157 			/*
158 			 * If the next character wouldn't have ended the
159 			 * block, it's a truncation.
160 			 */
161 			if (!in.dbcnt || *inp != '\n')
162 				++st.trunc;
163 
164 			/* Toss characters to a newline. */
165 			for (; in.dbcnt && *inp++ != '\n'; --in.dbcnt)
166 				;
167 			if (!in.dbcnt)
168 				intrunc = 1;
169 			else
170 				--in.dbcnt;
171 		}
172 
173 		/* Adjust output buffer numbers. */
174 		out.dbp += cbsz;
175 		if ((out.dbcnt += cbsz) >= out.dbsz)
176 			dd_out(0);
177 		outp = out.dbp;
178 	}
179 	in.dbp = in.db + in.dbcnt;
180 }
181 
182 void
183 block_close(void)
184 {
185 	/*
186 	 * Copy any remaining data into the output buffer and pad to a record.
187 	 * Don't worry about truncation or translation, the input buffer is
188 	 * always empty when truncating, and no characters have been added for
189 	 * translation.  The bottom line is that anything left in the input
190 	 * buffer is a truncated record.  Anything left in the output buffer
191 	 * just wasn't big enough.
192 	 */
193 	if (in.dbcnt) {
194 		++st.trunc;
195 		(void)memmove(out.dbp, in.dbp - in.dbcnt, in.dbcnt);
196 		(void)memset(out.dbp + in.dbcnt, ctab ? ctab[' '] : ' ',
197 		    cbsz - in.dbcnt);
198 		out.dbcnt += cbsz;
199 	}
200 }
201 
202 /*
203  * Convert fixed length (cbsz) records to variable length.  Deletes any
204  * trailing blanks and appends a newline.
205  *
206  * max in buffer:  MAX(ibs, cbsz) + cbsz
207  * max out buffer: obs + cbsz
208  */
209 void
210 unblock(void)
211 {
212 	u_char *inp;
213 	const u_char *t;
214 	size_t cnt;
215 
216 	/* Translation and case conversion. */
217 	if ((t = ctab) != NULL)
218 		for (inp = in.dbp - (cnt = in.dbrcnt); cnt--; ++inp)
219 			*inp = t[*inp];
220 	/*
221 	 * Copy records (max cbsz size chunks) into the output buffer.  The
222 	 * translation has to already be done or we might not recognize the
223 	 * spaces.
224 	 */
225 	for (inp = in.db; (size_t)in.dbcnt >= cbsz; inp += cbsz, in.dbcnt -= cbsz) {
226 		for (t = inp + cbsz - 1; t >= inp && *t == ' '; --t)
227 			;
228 		if (t >= inp) {
229 			cnt = t - inp + 1;
230 			(void)memmove(out.dbp, inp, cnt);
231 			out.dbp += cnt;
232 			out.dbcnt += cnt;
233 		}
234 		*out.dbp++ = '\n';
235 		if (++out.dbcnt >= out.dbsz)
236 			dd_out(0);
237 	}
238 	if (in.dbcnt)
239 		(void)memmove(in.db, in.dbp - in.dbcnt, in.dbcnt);
240 	in.dbp = in.db + in.dbcnt;
241 }
242 
243 void
244 unblock_close(void)
245 {
246 	u_char *t;
247 	size_t cnt;
248 
249 	if (in.dbcnt) {
250 		warnx("%s: short input record", in.name);
251 		for (t = in.db + in.dbcnt - 1; t >= in.db && *t == ' '; --t)
252 			;
253 		if (t >= in.db) {
254 			cnt = t - in.db + 1;
255 			(void)memmove(out.dbp, in.db, cnt);
256 			out.dbp += cnt;
257 			out.dbcnt += cnt;
258 		}
259 		++out.dbcnt;
260 		*out.dbp++ = '\n';
261 	}
262 }
263