*BSD中文用户社区

这几天版主谈到split的问题，我无意中find一下／src 目录，发现了openbsd的split的源代码，一点也不大，可以自己学习了。
／src ／bin 是 bin目录下命令的源代码，
／src ／sbin 是 sbin目录下的源代码，
src/usr／bin 是 ／usr／bin 目录下的命令源代码 （split 就在这里）
。。。

这是OpenBSD 4.8的split源代码，有兴趣的话，大家可以讨论讨论。(希望有编程爱好者和我逐条语句分析讨论一下，多好的教材呀，（C和UNIX） ）。
/* $OpenBSD: split.c,v 1.17 2009/10/27 23:59:43 deraadt Exp $ */
/* $NetBSD: split.c,v 1.5 1995/08/31 22:22:05 jtc Exp $ */

/*
* Copyright (c) 1987, 1993, 1994
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/

#include <sys/param.h>
#include <sys/types.h>

#include <ctype.h>
#include <err.h>
#include <fcntl.h>
#include <limits.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <regex.h>
#include <sysexits.h>

#define DEFLINE 1000 /* Default num lines per file. */

ssize_t bytecnt; /* Byte count to split on. */
long numlines; /* Line count to split on. */
int file_open; /* If a file open. */
int ifd = -1, ofd = -1; /* Input/output file descriptors. */
char bfr[MAXBSIZE]; /* I/O buffer. */
char fname[MAXPATHLEN]; /* File name prefix. */
regex_t rgx;
int pflag;
int sufflen = 2; /* File name suffix length. */

void newfile(void);
void split1(void);
void split2(void);
__dead void usage(void);

int
main(int argc, char *argv[])
{
int ch, scale;
char *ep, *p;
const char *errstr;

while ((ch = getopt(argc, argv, "0123456789a:b:l:p:-")) != -1)
switch (ch) {
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
/*
* Undocumented kludge: split was originally designed
* to take a number after a dash.
*/
if (numlines == 0) {
p = argv[optind - 1];
if (p[0] == '-' && p[1] == ch && !p[2])
numlines = strtol(++p, &ep, 10);
else
numlines =
strtol(argv[optind] + 1, &ep, 10);
if (numlines <= 0 || *ep)
errx(EX_USAGE,
"%s: illegal line count", optarg);
}
break;
case '-': /* Undocumented: historic stdin flag. */
if (ifd != -1)
usage();
ifd = 0;
break;
case 'a': /* suffix length. */
sufflen = strtonum(optarg, 1, NAME_MAX, &errstr);
if (errstr)
errx(EX_USAGE, "%s: %s", optarg, errstr);
break;
case 'b': /* Byte count. */
if ((bytecnt = strtol(optarg, &ep, 10)) <= 0 ||
(*ep != '\0' && *ep != 'k' && *ep != 'm'))
errx(EX_USAGE,
"%s: illegal byte count", optarg);
if (*ep == 'k')
scale = 1024;
else if (*ep == 'm')
scale = 1048576;
else
scale = 1;
if (bytecnt > SSIZE_MAX / scale)
errx(EX_USAGE, "%s: byte count too large",
optarg);
bytecnt *= scale;
break;
case 'p' : /* pattern matching. */
if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0)
errx(EX_USAGE, "%s: illegal regexp", optarg);
pflag = 1;
break;
case 'l': /* Line count. */
if (numlines != 0)
usage();
if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep)
errx(EX_USAGE,
"%s: illegal line count", optarg);
break;
default:
usage();
}
argv += optind;
argc -= optind;

if (*argv != NULL)
if (ifd == -1) { /* Input file. */
if ((ifd = open(*argv, O_RDONLY, 0)) < 0)
err(EX_NOINPUT, "%s", *argv);
++argv;
}
if (*argv != NULL) /* File name prefix. */
(void)strlcpy(fname, *argv++, sizeof(fname));
if (*argv != NULL)
usage();

if (strlen(fname) + sufflen >= sizeof(fname))
errx(EX_USAGE, "suffix is too long");
if (pflag && (numlines != 0 || bytecnt != 0))
usage();

if (numlines == 0)
numlines = DEFLINE;
else if (bytecnt != 0)
usage();

if (ifd == -1) /* Stdin by default. */
ifd = 0;

if (bytecnt) {
split1();
exit (0);
}
split2();
if (pflag)
regfree(&rgx);
exit(0);
}

/*
* split1 --
* Split the input by bytes.
*/
void
split1(void)
{
ssize_t bcnt, dist, len;
char *C;

for (bcnt = 0;;)
switch ((len = read(ifd, bfr, MAXBSIZE))) {
case 0:
exit(0);
case -1:
err(EX_IOERR, "read");
/* NOTREACHED */
default:
if (!file_open)
newfile();
if (bcnt + len >= bytecnt) {
dist = bytecnt - bcnt;
if (write(ofd, bfr, dist) != dist)
err(EX_IOERR, "write");
len -= dist;
for (C = bfr + dist; len >= bytecnt;
len -= bytecnt, C += bytecnt) {
newfile();
if (write(ofd, C, bytecnt) != bytecnt)
err(EX_IOERR, "write");
}
if (len != 0) {
newfile();
if (write(ofd, C, len) != len)
err(EX_IOERR, "write");
} else
file_open = 0;
bcnt = len;
} else {
bcnt += len;
if (write(ofd, bfr, len) != len)
err(EX_IOERR, "write");
}
}
}

/*
* split2 --
* Split the input by lines.
*/
void
split2(void)
{
long lcnt = 0;
FILE *infp;

/* Stick a stream on top of input file descriptor */
if ((infp = fdopen(ifd, "r")) == NULL)
err(EX_NOINPUT, "fdopen");

/* Process input one line at a time */
while (fgets(bfr, sizeof(bfr), infp) != NULL) {
const int len = strlen(bfr);

if (len == 0)
continue;

/* If line is too long to deal with, just write it out */
if (bfr[len - 1] != '\n')
goto writeit;

/* Check if we need to start a new file */
if (pflag) {
regmatch_t pmatch;

pmatch.rm_so = 0;
pmatch.rm_eo = len - 1;
if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0)
newfile();
} else if (lcnt++ == numlines) {
newfile();
lcnt = 1;
}

writeit:
/* Open output file if needed */
if (!file_open)
newfile();

/* Write out line */
if (write(ofd, bfr, len) != len)
err(EX_IOERR, "write");
}

/* EOF or error? */
if (ferror(infp))
err(EX_IOERR, "read");
else
exit(0);
}

/*
* newfile --
* Open a new output file.
*/
void
newfile(void)
{
static char *suffix, *sufftail;
char *sptr;

if (ofd == -1) {
ofd = fileno(stdout);
if (*fname == '\0') {
*fname = 'x'; /* no name specified, use 'x' */
memset(fname + 1, 'a', sufflen);
suffix = fname;
sufflen++; /* treat 'x' as part of suffix */
} else {
suffix = fname + strlen(fname);
memset(suffix, 'a', sufflen);
}
suffix[sufflen] = '\0';
sufftail = suffix + sufflen - 1;
} else {
for (sptr = sufftail; sptr >= suffix; sptr--) {
if (*sptr != 'z') {
(*sptr)++;
break;
} else
*sptr = 'a';
}
if (sptr < suffix)
errx(EX_DATAERR, "too many files");
}

if (!freopen(fname, "w", stdout))
err(EX_IOERR, "%s", fname);
file_open = 1;
}

__dead void
usage(void)
{
extern char *__progname;

(void)fprintf(stderr, "usage: %s [-a suffix_length]\n"
" [-b byte_count[k|m] | -l line_count | -p pattern] "
"[file [name]]\n", __progname);
exit(EX_USAGE);
}

给改了一下，别忘了贴代码时禁用表情，我没有动别的。
圣诞节快乐，大家别脑子绷太紧了，放松一下吧。

谢谢了。现在好多中国人（时尚人士），春节不过了，跟着外国人过宗教节日，无语。

开源就是好啊~

.............
int ch, scale;
...........
if (*ep == 'k')
scale = 1024;
else if (*ep == 'm')
scale = 1048576;
else
scale = 1; (我觉得这个写法不好，可以一开始 int scale＝1，还可以避免if (bytecnt > SSIZE_MAX / scale) 中的除零错误的潜在风险［如果忘 scale＝1］）。
/*
具有 ``静态'' 生存期的未初始化变量 (即, 在函数外声明的变量和有静态存储类型的变量) 可以确保初始值为零, 就像程序员键入了 ``=0'' 一样。因此, 这些变量如果是指针会被初始化为正确的空指针, 如果是浮点数会被初始化为 0.0 (或正确的类型, 参见第 5 章)。
具有 ``自动'' 生存期的变量 (即, 没有静态存储类型的局部变量) 如果没有显示地初始化, 则包含的是垃圾内容。对垃圾内容不能作任何有用的假设。
*/
。。。。。。。。。。。。。。。。。
还有这句，我觉得可以琢磨一下。我们知道，在一般计算机内，除操作比乘操作昂贵，如果直接用乘做判断语句，好像有溢出判断问题，多一步溢出判断，不知道，双判断加一个乘法操作比一判断加一除法加一乘法操作，谁更好。
if (bytecnt > SSIZE_MAX / scale)
errx(EX_USAGE, "%s: byte count too large",
optarg);
bytecnt *= scale;

这个split源程序，写的不是很漂亮呀，有的地方值得商榷呀，不知道，大家看了啥感觉，反正我是觉得应该重写了，毕竟太老了。看着看着都想，我都想重写了（我可是第一门计算机语言是pascal，感觉真的影响挺深得，对C的自由派写法很有意见，比如不喜欢初始化变量，比如不喜欢逻辑书写，喜欢就近书写变量，等等。对小程序，pascal的书写方式，非常容易阅读理解，大程序的话，影响的是软件架构，但是如果非要按pascal方式书写的话，就是古板了）

看源代码，真的可以发现很多平时不知道的知识。一个200多行的源程序，我看了好些天了，越看越长见识，到现在还在看，没吃透，看来，自己的水平太挖了。。。呵呵。总的来说，我还是比较欣赏 netbsd的源代码，书写的很好看，很有看C语言学习的感觉。OpenBSD的源代码，初看不太漂亮，细看会发现，人家这么写是有原因的，暂不赞不同，是你自己的事，但绝不作者是随意写的。
一条语句引发的安全思考:
(void)strlcpy(fname, *argv++, sizeof(fname));
//////////
引用来源：
http://www.1-100.org/other/19324.htm

高人呢，我看着头疼，暂时不想走火入魔。

回头真想学习一下，就是感觉可能入不了门~~~

用什么软件打开　/usr/bin/split ? 为何用vi 打开是乱码？

/usr/bin/split是二进制可执行文件，用vi当然看不了。不信，你用file 命令看看。
例如：
$ file /usr/bin/split
/usr/bin/split: ELF 64-bit LSB executable, x86-64, version 1, for OpenBSD, dynamically linked (uses shared libs), stripped
$ file /tmp/src/usr.bin/split/split.c
/tmp/src/usr.bin/split/split.c: ASCII C program text

真正的源程序，在 src/usr.bin/split/ 目录下， 如果你想看 ／usr／bin／split 的话，用objdump 命令，反正我是看不懂的，需要汇编和硬件知识。