Module Name: src
Committed By: kre
Date: Fri Oct 11 09:02:10 UTC 2024
Modified Files:
src/bin/sh: miscbltin.c sh.1
Log Message:
Add -b and -nMAX options to the read builtin.
As requested on (perhaps more than one) mailing list, this adds
a -n MAX option, to allow the amount of data read by the read
builtin to be limited to MAX bytes (in case the record delimiter
doesn't appear in the input for a long time). There is
currently an upper bound of 8MiB on the value of MAX.
Also add a -b option, which allows for buffered input (with
some usage caveats) rather than 1 byte at a time.
Neither option exists in SMALL shells.
Note that the proposed -z option got deleted ... I couldn't
find a rational way to explain what the final state would be
if a \0 on input generated an error, so rather than have
things ambiguous, better just not to have the option, and
simply keep ignoring input \0's as always.
See the (updated) sh(1) man page for more details.
No pullups planned (new feature, only for new releases).
To generate a diff of this commit:
cvs rdiff -u -r1.54 -r1.55 src/bin/sh/miscbltin.c
cvs rdiff -u -r1.265 -r1.266 src/bin/sh/sh.1
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
Modified files:
Index: src/bin/sh/miscbltin.c
diff -u src/bin/sh/miscbltin.c:1.54 src/bin/sh/miscbltin.c:1.55
--- src/bin/sh/miscbltin.c:1.54 Thu Oct 5 20:33:31 2023
+++ src/bin/sh/miscbltin.c Fri Oct 11 09:02:10 2024
@@ -1,4 +1,4 @@
-/* $NetBSD: miscbltin.c,v 1.54 2023/10/05 20:33:31 kre Exp $ */
+/* $NetBSD: miscbltin.c,v 1.55 2024/10/11 09:02:10 kre Exp $ */
/*-
* Copyright (c) 1991, 1993
@@ -37,7 +37,7 @@
#if 0
static char sccsid[] = "@(#)miscbltin.c 8.4 (Berkeley) 5/4/95";
#else
-__RCSID("$NetBSD: miscbltin.c,v 1.54 2023/10/05 20:33:31 kre Exp $");
+__RCSID("$NetBSD: miscbltin.c,v 1.55 2024/10/11 09:02:10 kre Exp $");
#endif
#endif /* not lint */
@@ -45,30 +45,33 @@ __RCSID("$NetBSD: miscbltin.c,v 1.54 202
* Miscellaneous builtins.
*/
-#include <sys/types.h> /* quad_t */
#include <sys/param.h> /* BSD4_4 */
+#include <sys/resource.h>
#include <sys/stat.h>
#include <sys/time.h>
-#include <sys/resource.h>
-#include <unistd.h>
-#include <stdlib.h>
+#include <sys/types.h> /* quad_t */
+
#include <ctype.h>
#include <errno.h>
+#include <limits.h>
+#include <stdlib.h>
+#ifndef SMALL
+#include <termios.h>
+#endif
+#include <unistd.h>
#include "shell.h"
#include "options.h"
#include "var.h"
+#include "input.h" /* for whichprompt */
#include "output.h"
+#include "parser.h" /* for getprompt() */
#include "memalloc.h"
#include "error.h"
#include "builtins.h"
#include "mystring.h"
#include "redir.h" /* for user_fd_limit */
-#undef rflag
-
-
-
/*
* The read builtin.
* Backslashes escape the next char unless -r is specified.
@@ -86,13 +89,119 @@ __RCSID("$NetBSD: miscbltin.c,v 1.54 202
* ':b c:' x='', y='b c:'
*/
+#ifndef SMALL
+static int b_flag;
+
+static int
+setrawmode(int fd, int on, int end, struct termios *t)
+{
+ struct termios n;
+
+ if (on) {
+ if (tcgetattr(fd, t) != 0)
+ return 0;
+ n = *t;
+ if (on == 1 && b_flag) {
+ n.c_cc[VEOL] = end;
+ } else {
+ cfmakeraw(&n);
+ n.c_iflag |= ICRNL;
+ n.c_oflag = t->c_oflag;
+ n.c_lflag |= ECHO | ISIG;
+ }
+ if (tcsetattr(fd, TCSADRAIN | TCSASOFT, &n) == 0)
+ return 1;
+ } else
+ (void)tcsetattr(fd, TCSADRAIN | TCSASOFT, t);
+ return 0;
+}
+
+static int
+is_a_pipe(int fd)
+{
+ if (lseek(fd, 0, SEEK_CUR) == -1 && errno == ESPIPE) {
+ errno = 0;
+ return 1;
+ }
+ return 0;
+}
+
+#define READ_BUFFER_SIZE 512
+
+static int
+next_read_char(int fd, size_t max)
+{
+ static char buffer[READ_BUFFER_SIZE];
+ static int pos = 0, len = 0;
+
+ if (max == 0) {
+ pos = len = 0;
+ return -1;
+ }
+ if (max == (size_t)-1) {
+ /*
+ * If possible, and necessary, rewind the file
+ * so unprocessed data can be read again next time
+ *
+ * If that fails, never mind (-b allows that to happen)
+ */
+ if (b_flag && pos < len)
+ (void)lseek(fd, (off_t)(pos - len), SEEK_CUR);
+ return -1;
+ }
+
+ if (b_flag == 0) {
+ char c;
+
+ (void) max;
+ if (read(fd, &c, 1) != 1)
+ return -1;
+ return (c & 0xFF);
+ }
+
+ if (pos >= len) {
+ pos = 0;
+ if (max > sizeof buffer)
+ max = sizeof buffer;
+ len = read(fd, buffer, max);
+ if (len <= 0)
+ return -1;
+ }
+
+ return buffer[pos++] & 0xFF;
+}
+
+#define READ_OPTS "bd:n:p:r"
+
+#else
+
+static inline int
+next_read_char(int fd, size_t max)
+{
+ char c;
+
+ if (max == 0 || max == (size_t)-1)
+ return 0;
+
+ if (read(fd, &c, 1) != 1)
+ return -1;
+ return (c & 0xFF);
+}
+
+#define n_flag 0
+#define maxlen 0
+
+#define READ_OPTS "d:p:r"
+
+#endif
+
int
readcmd(int argc, char **argv)
{
char **ap;
- char c;
+ int c;
char end;
- int rflag;
+ int r_flag;
char *prompt;
const char *ifs;
char *p;
@@ -101,14 +210,27 @@ readcmd(int argc, char **argv)
int i;
int is_ifs;
int saveall = 0;
+ int read_tty = 0;
ptrdiff_t wordlen = 0;
char *newifs = NULL;
struct stackmark mk;
+#ifndef SMALL
+ struct termios ttystate;
+ int n_flag, maxlen;
+ int setraw = 0;
+
+ b_flag = 0;
+ n_flag = 0;
+ maxlen = READ_BUFFER_SIZE - 1;
+#endif
+
end = '\n'; /* record delimiter */
- rflag = 0;
+ r_flag = 0;
prompt = NULL;
- while ((i = nextopt("d:p:r")) != '\0') {
+ whichprompt = 2; /* for continuation lines */
+
+ while ((i = nextopt(READ_OPTS)) != '\0') {
switch (i) {
case 'd':
end = *optionarg; /* even if '\0' */
@@ -117,39 +239,80 @@ readcmd(int argc, char **argv)
prompt = optionarg;
break;
case 'r':
- rflag = 1;
+ r_flag = 1;
break;
+#ifndef SMALL
+ case 'n':
+ maxlen = number(optionarg);
+ if (maxlen > (INT_MAX >> 8) + 1) /* sanity */
+ error("-n %s too large", optionarg);
+ n_flag = 1;
+ break;
+ case 'b':
+ if (!is_a_pipe(0))
+ b_flag = 1;
+ break;
+#endif
}
}
if (*(ap = argptr) == NULL)
error("variable name required\n"
- "Usage: read [-r] [-p prompt] var...");
+#ifdef SMALL
+ "Usage: read [-r] [-d C] [-p prompt] var...");
+#else
+ "Usage: read [-br] [-d C] [-n len] [-p prompt] var...");
- if (prompt && isatty(0)) {
- out2str(prompt);
- flushall();
+ (void)next_read_char(0, 0); /* make sure the buffer is empty */
+#endif
+
+ if (isatty(0)) {
+ read_tty = 1;
+ if (prompt) {
+ out2str(prompt);
+ flushall();
+ }
+#ifndef SMALL
+ b_flag = 1; /* always buffer reads from ttys */
+
+ if (n_flag || end != '\n')
+ setraw = setrawmode(0, 1 + n_flag, end, &ttystate);
+#endif
}
+/*
if ((ifs = bltinlookup("IFS", 1)) == NULL)
ifs = " \t\n";
+*/
+ ifs = ifsval();
setstackmark(&mk);
status = 0;
startword = 2;
STARTSTACKSTR(p);
- for (;;) {
- if (read(0, &c, 1) != 1) {
+
+#ifdef SMALL
+ for ( ; ; ) {
+#else
+ for ( ; !n_flag || --maxlen >= 0 ; ) {
+#endif
+ if ((c = next_read_char(0, maxlen + 1)) < 0) {
status = 1;
break;
}
- if (c == '\\' && c != end && !rflag) {
- if (read(0, &c, 1) != 1) {
+ if (c == '\\' && c != end && !r_flag) {
+#ifndef SMALL
+ if (n_flag && --maxlen < 0)
+ break;
+#endif
+ if ((c = next_read_char(0, maxlen + 1)) < 0) {
status = 1;
break;
}
if (c != '\n') /* \ \n is always just removed */
goto wdch;
+ if (read_tty)
+ out2str(getprompt(NULL));
continue;
}
if (c == end)
@@ -179,7 +342,7 @@ readcmd(int argc, char **argv)
if (is_ifs == 0) {
wdch:;
- if (c == '\0') /* always ignore attempts to input \0 */
+ if (c == '\0') /* always ignore attempts to input \0 */
continue;
/* append this character to the current variable */
startword = 0;
@@ -223,6 +386,13 @@ readcmd(int argc, char **argv)
}
STACKSTRNUL(p);
+#ifndef SMALL
+ (void)next_read_char(0, (size_t)-1); /* attempt to seek back */
+ if (setraw)
+ setrawmode(0, 0, end, &ttystate);
+#endif
+
+
/* Remove trailing IFS chars */
for (; stackblock() + wordlen <= --p; *p = 0) {
if (!strchr(ifs, *p))
Index: src/bin/sh/sh.1
diff -u src/bin/sh/sh.1:1.265 src/bin/sh/sh.1:1.266
--- src/bin/sh/sh.1:1.265 Wed Oct 9 13:43:32 2024
+++ src/bin/sh/sh.1 Fri Oct 11 09:02:10 2024
@@ -1,4 +1,4 @@
-.\" $NetBSD: sh.1,v 1.265 2024/10/09 13:43:32 kre Exp $
+.\" $NetBSD: sh.1,v 1.266 2024/10/11 09:02:10 kre Exp $
.\" Copyright (c) 1991, 1993
.\" The Regents of the University of California. All rights reserved.
.\"
@@ -31,7 +31,7 @@
.\"
.\" @(#)sh.1 8.6 (Berkeley) 5/4/95
.\"
-.Dd October 6, 2024
+.Dd October 11, 2024
.Dt SH 1
.\" everything except c o and s (keep them ordered)
.ds flags abCEeFfhIiLlmnpquVvXx
@@ -3304,7 +3304,7 @@ if both the
and
.Fl q
options are given, then an error message will be printed about
-.Ar command Ns No s
+.Ar command Ns s
unable to be found, but the exit status will remain 0.
This is not considered useful.
.\"
@@ -3663,24 +3663,183 @@ the program will use
and the built-in uses a separately cached value.
.\"
.Pp
-.It Ic read Oo Fl d Ar delim Oc Oo Fl p Ar prompt Oc Oo Fl r Oc Ar variable Op Ar ...
+.It Ic read Oo Fl br Oc Oo Fl d Ar delim Oc Oo Fl n Ar max Oc Oo Fl p Ar prompt Oc Ar variable Op Ar ...
+.Pp
The
+.Ic read
+command reads a record (by default one line) from its standard input,
+splits that record as if by field splitting, and assigns the results
+to the named
+.Ar variable
+arguments, as detailed below.
+.Pp
+The options are:
+.Bl -tag -width ".Fl ppromptM"
+.It Fl b
+Do buffered reads, rather than reading one byte at a time.
+This option is ignored if reading from a pipe.
+Use of this option might result in reading more bytes from
+standard input than the
+.Ic read
+utility actually processes,
+causing some data from standard input to be unavailable
+to any subsequent utility that expects to obtain them,
+though this will be avoided wherever possible.
+.It Fl d Ar delim
+End the read when the first byte of
+.Ar delim
+is obtained from standard input.
+The default end delimiter is
+.Aq newline
+.Pq Sq \en .
+Specifying
+.Qq
+as
+.Ar delim
+causes the nul character
+.Pq Sq \e0
+to be the end delimiter.
+If the delimiter is set to
+.Aq backslash
+.Pq Sq \&\e
+then neither line continuation nor escaping are available.
+The input will end when the first
+.Aq backslash
+is read.
+.It Fl n Ar max
+.Ic read
+will read no more than
+.Ar max
+bytes from standard input.
+The default is unlimited.
+If the end
+.Ar delim
+has not been encountered within
+.Ar max
+bytes, and EOF has not been reached,
+.Ic read
+will act as if the delimiter immediately followed
+the
+.Ar max Ns 'th
+byte, without attempting to obtain it.
+However, even if the
+.Fl r
+option is not given and the final byte actually read
+were the escape character (not itself escaped),
+no more bytes will be read, and that escape
+character would simply be removed as described below.
+.It Fl p Ar prompt
+If the standard input is a terminal, then
.Ar prompt
-is printed on standard error if the
+is written to standard error before the read commences.
+If more lines of data are required in that case, the
+normal
+.Ev PS2
+prompt is written as each subsequent line is to be obtained.
+.It Fl r
+Reduced processing of the input.
+No escape characters are recognised,
+and line continuation is not performed.
+See below.
+.El
+.Pp
+If the read is from a terminal device,
+and the
.Fl p
-option is specified and the standard input is a terminal.
-Then a record, terminated by the
-first character of
+option was given,
+.Ar prompt
+is printed on standard error.
+Then a record, terminated by the first character of
.Ar delim
if the
.Fl d
-option was given, or a newline character otherwise,
+option was given, or a
+.Aq newline
+.Pq Sq \en
+character otherwise,
+but no longer than
+.Ar max
+bytes if the
+.Fl n
+option was given,
is read from the standard input.
-The ending delimiter is deleted from the
-record which is then split as described in the field splitting section of the
+If the
+.Fl b
+option is not given, no data from standard
+input beyond the end delimiter, or the
+.Ar max
+bytes that may be read,
+whichever occurs first,
+are obtained.
+Any nul characters
+.Pq Sq \e0
+encountered in the input stream,
+other than when being the delimiter if
+.Fl d Qq
+was used,
+are simply ignored, even if escaped (in which case the
+preceding
+.Sq \e
+is also ignored) though these ignored characters are
+included in the byte count for the purposes of the
+.Fl n
+option.
+.Pp
+If the
+.Fl r
+option was not given,
+and the two character sequence
+.Sq \&\e
+.Sq \en
+is encountered,
+those two characters are simply deleted,
+and provided that
+.Ar max
+bytes have not yet been obtained,
+and the end delimiter has yet to be encountered,
+more input is obtained,
+with the first character of the following line
+placed in the input where the deleted
+.Sq \e
+had been.
+This allows logical lines longer than the maximum line
+length permitted for text files to be processed.
+The two removed characters are still counted
+for the purposes of the
+.Ar max
+input limit.
+.Pp
+If the
+.Fl r
+flag was not given, the
+.Aq backslash
+character
+.Pq Sq \e
+character is then treated as an escape character,
+the character following it, if not
+.Sq \e0 ,
+is always treated as a normal,
+insignificant, data character,
+and is never treated as the end delimiter nor as
+an IFS character for field splitting.
+.Pp
+After field splitting has completed,
+but before data has been
+assigned to any variables,
+all escape characters are removed.
+Note that the two character sequence
+.Sq \&\e
+.Sq \&\e
+can be used to enter the escape character as data,
+the first acts as the escape character, the second
+becomes just a normal data character.
+.Pp
+The ending delimiter, if encountered, and not escaped,
+is deleted from the record which is then split as described
+in the field splitting section of the
.Sx Word Expansions
section above.
-The pieces are assigned to the
+The pieces (fields) are assigned to the
.Ar variable Ns s
in order.
If there are more pieces than variables,
@@ -3691,50 +3850,31 @@ that separated them) are all assigned to
.Ar variable .
If there are more variables than pieces,
the remaining variables are assigned the null string.
+.Pp
The
.Ic read
-built-in will indicate success unless EOF, or a read error,
-is encountered on input, in
-which case failure is returned.
-.Pp
-By default, unless the
-.Fl r
-option is specified, the backslash
-.Pq Ql \e
-acts as an escape character,
-causing the following character,
-when that character is the escape character, or end delimiter character,
-to be treated literally when reading the record.
-This is the only form of quoting that applies.
-If an unescaped backslash is followed by a newline,
-the backslash and the newline will be deleted,
-and replaced by the contents from the following line,
-which is processed as if it had been part of the original line.
-This includes reading yet more input if necessary,
-until a line is read that contains or ends with an unescaped
-copy of the delimiter character.
-If the end delimiter (when it is not a newline) is escaped,
-it is treated as a normal character, and
+built-in utility will indicate success (exit status 0)
+unless EOF, or a read error,
+is encountered on input, before encountering the delimiter,
+or having read
+.Ar max
+bytes.
+In this case any data previously read will be assigned to
+the variables, as if the delimiter had been located at that
+point, and
.Ic read
-continues looking for an unescaped end delimiter character.
-No other escape sequences are meaningful, the escape character
-is simply ignored.
-This happens as the record is read,
-before field splitting occurs.
-When
-.Fl r
-is used,
-no escaping occurs,
-no line joining happens,
-any input backslash is simply an input character.
+will exit with status 1.
+If there is a usage error (unknown option, etc)
+no variables will be altered,
+.Ic read
+will issue a diagnostic to standard error,
+and exit with a status greater than 1.
.Pp
-Note that if
-.Ar delim
-is given as an empty string, the nul character
-.Pq Ql \e0
-is used as the delimiter.
-Other than this use, any nul characters in the input
-stream are silently deleted.
+The
+.Fl b
+and
+.Fl n
+options are not available in SMALL shells.
.\"
.Pp
.It Ic readonly Ar name Ns Oo =value Oc ...
@@ -5009,7 +5149,8 @@ This is normally set to
.Aq tab ,
and
.Aq newline .
-.Sx White Space Splitting
+See the
+.Sx Field Splitting
section for more details.
.It Ev LANG
The string used to specify localization information that allows users
