bash-20130727 remove leftover and stray files

This commit is contained in:
Chet Ramey
2013-08-05 09:56:54 -04:00
parent 50ee0e9709
commit 6bbd35c96a
25 changed files with 0 additions and 52789 deletions
-5114
View File
File diff suppressed because it is too large Load Diff
-82
View File
@@ -1,82 +0,0 @@
Starting bash with the `--posix' command-line option or executing
`set -o posix' while bash is running will cause bash to conform more
closely to the Posix.2 standard by changing the behavior to match that
specified by Posix.2 in areas where the bash default differs.
The following list is what's changed when `posix mode' is in effect:
1. When a command in the hash table no longer exists, bash will re-search
$PATH to find the new location. This is also available with
`shopt -s checkhash'.
2. The >& redirection does not redirect stdout and stderr.
3. The message printed by the job control code and builtins when a job
exits with a non-zero status is `Done(status)'.
4. Reserved words may not be aliased.
5. The Posix.2 PS1 and PS2 expansions of `!' -> history number and
`!!' -> `!' are enabled, and parameter expansion is performed on
the value regardless of the setting of the `promptvars' option.
6. Interactive comments are enabled by default. (Note that bash has
them on by default anyway.)
7. The Posix.2 startup files are executed ($ENV) rather than the normal
bash files.
8. Tilde expansion is only performed on assignments preceding a command
name, rather than on all assignment statements on the line.
9. The default history file is ~/.sh_history (default value of $HISTFILE).
10. The output of `kill -l' prints all the signal names on a single line,
separated by spaces.
11. Non-interactive shells exit if `file' in `. file' is not found.
12. Redirection operators do not perform pathname expansion on the word
in the redirection unless the shell is interactive
13. Function names must be valid shell identifiers. That is, they may not
contain characters other than letters, digits, and underscores, and
may not start with a digit. Declaring a function with an illegal name
causes a fatal syntax error in non-interactive shells.
14. Posix.2 `special' builtins are found before shell functions during command
lookup.
15. If a Posix.2 special builtin returns an error status, a non-interactive
shell exits. The fatal errors are those listed in the POSIX.2 standard,
and include things like passing incorrect options, redirection errors,
variable assignment errors for assignments preceding the command name,
and so on.
16. The environment passed to executed commands is not sorted. Neither is
the output of `set'. This is not strictly Posix.2 behavior, but sh
does it this way. Ksh does not. It's not necessary to sort the
environment; no program should rely on it being sorted.
17. If the `cd' builtin finds a directory to change to using $CDPATH, the
value it assigns to $PWD does not contain any symbolic links, as if
`cd -P' had been executed.
18. A non-interactive shell exits with an error status if a variable
assignment error occurs when no command name follows the assignment
statements. A variable assignment error occurs, for example, when
trying to assign a value to a read-only variable.
19. A non-interactive shell exits with an error status if the iteration
variable in a for statement or the selection variable in a select
statement is a read-only variable.
20. Process substitution is not available.
21. Assignment statements preceding POSIX.2 `special' builtins persist in
the shell environment after the builtin completes.
There is other Posix.2 behavior that bash does not implement. Specifically:
1. Assignment statements affect the execution environment of all builtins,
not just special ones.
-544
View File
@@ -1,544 +0,0 @@
This file is set.def, from which is created set.c.
It implements the "set" and "unset" builtins in Bash.
Copyright (C) 1987, 1989, 1991 Free Software Foundation, Inc.
This file is part of GNU Bash, the Bourne Again SHell.
Bash is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation; either version 1, or (at your option) any later
version.
Bash is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
for more details.
You should have received a copy of the GNU General Public License along
with Bash; see the file COPYING. If not, write to the Free Software
Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
$PRODUCES set.c
#include <stdio.h>
#include "../shell.h"
#include "../flags.h"
#include "bashgetopt.h"
extern int interactive;
extern int noclobber, posixly_correct;
#if defined (READLINE)
extern int rl_editing_mode, no_line_editing;
#endif /* READLINE */
$BUILTIN set
$FUNCTION set_builtin
$SHORT_DOC set [--abefhkmnptuvxldBCHP] [-o option] [arg ...]
-a Mark variables which are modified or created for export.
-b Notify of job termination immediately.
-e Exit immediately if a command exits with a non-zero status.
-f Disable file name generation (globbing).
-h Locate and remember function commands as functions are
defined. Function commands are normally looked up when
the function is executed.
-i Force the shell to be an "interactive" one. Interactive shells
always read `~/.bashrc' on startup.
-k All keyword arguments are placed in the environment for a
command, not just those that precede the command name.
-m Job control is enabled.
-n Read commands but do not execute them.
-o option-name
Set the variable corresponding to option-name:
allexport same as -a
braceexpand same as -B
#if defined (READLINE)
emacs use an emacs-style line editing interface
#endif /* READLINE */
errexit same as -e
histexpand same as -H
ignoreeof the shell will not exit upon reading EOF
interactive-comments
allow comments to appear in interactive commands
monitor same as -m
noclobber disallow redirection to existing files
noexec same as -n
noglob same as -f
nohash same as -d
notify save as -b
nounset same as -u
physical same as -P
posix change the behavior of bash where the default
operation differs from the 1003.2 standard to
match the standard
privileged same as -p
verbose same as -v
#if defined (READLINE)
vi use a vi-style line editing interface
#endif /* READLINE */
xtrace same as -x
-p Turned on whenever the real and effective user ids do not match.
Disables processing of the $ENV file and importing of shell
functions. Turning this option off causes the effective uid and
gid to be set to the real uid and gid.
-t Exit after reading and executing one command.
-u Treat unset variables as an error when substituting.
-v Print shell input lines as they are read.
-x Print commands and their arguments as they are executed.
-l Save and restore the binding of the NAME in a FOR command.
-d Disable the hashing of commands that are looked up for execution.
Normally, commands are remembered in a hash table, and once
found, do not have to be looked up again.
#if defined (BRACE_EXPANSION)
-B the shell will perform brace expansion
#endif /* BRACE_EXPANSION */
#if defined (BANG_HISTORY)
-H Enable ! style history substitution. This flag is on
by default.
#endif /* BANG_HISTORY */
-C If set, disallow existing regular files to be overwritten
by redirection of output.
-P If set, do not follow symbolic links when executing commands
such as cd which change the current directory.
Using + rather than - causes these flags to be turned off. The
flags can also be used upon invocation of the shell. The current
set of flags may be found in $-. The remaining n ARGs are positional
parameters and are assigned, in order, to $1, $2, .. $n. If no
ARGs are given, all shell variables are printed.
$END
/* An a-list used to match long options for set -o to the corresponding
option letter. */
struct {
char *name;
int letter;
} o_options[] = {
{ "allexport", 'a' },
#if defined (BRACE_EXPANSION)
{ "braceexpand",'B' },
#endif
{ "errexit", 'e' },
{ "histexpand", 'H' },
{ "monitor", 'm' },
{ "noexec", 'n' },
{ "noglob", 'f' },
{ "nohash", 'd' },
#if defined (JOB_CONTROL)
{ "notify", 'b' },
#endif /* JOB_CONTROL */
{"nounset", 'u' },
{"physical", 'P' },
{"privileged", 'p' },
{"verbose", 'v' },
{"xtrace", 'x' },
{(char *)NULL, 0},
};
#define MINUS_O_FORMAT "%-15s\t%s\n"
void
list_minus_o_opts ()
{
register int i;
char *on = "on", *off = "off";
printf (MINUS_O_FORMAT, "noclobber", (noclobber == 1) ? on : off);
if (find_variable ("ignoreeof") || find_variable ("IGNOREEOF"))
printf (MINUS_O_FORMAT, "ignoreeof", on);
else
printf (MINUS_O_FORMAT, "ignoreeof", off);
printf (MINUS_O_FORMAT, "interactive-comments",
interactive_comments ? on : off);
printf (MINUS_O_FORMAT, "posix", posixly_correct ? on : off);
#if defined (READLINE)
if (no_line_editing)
{
printf (MINUS_O_FORMAT, "emacs", off);
printf (MINUS_O_FORMAT, "vi", off);
}
else
{
/* Magic. This code `knows' how readline handles rl_editing_mode. */
printf (MINUS_O_FORMAT, "emacs", (rl_editing_mode == 1) ? on : off);
printf (MINUS_O_FORMAT, "vi", (rl_editing_mode == 0) ? on : off);
}
#endif /* READLINE */
for (i = 0; o_options[i].name; i++)
{
int *on_or_off, zero = 0;
on_or_off = find_flag (o_options[i].letter);
if (on_or_off == FLAG_UNKNOWN)
on_or_off = &zero;
printf (MINUS_O_FORMAT, o_options[i].name, (*on_or_off == 1) ? on : off);
}
}
set_minus_o_option (on_or_off, option_name)
int on_or_off;
char *option_name;
{
int option_char = -1;
if (STREQ (option_name, "noclobber"))
{
if (on_or_off == FLAG_ON)
bind_variable ("noclobber", "");
else
unbind_variable ("noclobber");
stupidly_hack_special_variables ("noclobber");
}
else if (STREQ (option_name, "ignoreeof"))
{
unbind_variable ("ignoreeof");
unbind_variable ("IGNOREEOF");
if (on_or_off == FLAG_ON)
bind_variable ("IGNOREEOF", "10");
stupidly_hack_special_variables ("IGNOREEOF");
}
#if defined (READLINE)
else if ((STREQ (option_name, "emacs")) || (STREQ (option_name, "vi")))
{
if (on_or_off == FLAG_ON)
{
rl_variable_bind ("editing-mode", option_name);
if (interactive)
with_input_from_stdin ();
no_line_editing = 0;
}
else
{
int isemacs = (rl_editing_mode == 1);
if ((isemacs && STREQ (option_name, "emacs")) ||
(!isemacs && STREQ (option_name, "vi")))
{
if (interactive)
with_input_from_stream (stdin, "stdin");
no_line_editing = 1;
}
else
builtin_error ("not in %s editing mode", option_name);
}
}
#endif /* READLINE */
else if (STREQ (option_name, "interactive-comments"))
interactive_comments = (on_or_off == FLAG_ON);
else if (STREQ (option_name, "posix"))
{
posixly_correct = (on_or_off == FLAG_ON);
unbind_variable ("POSIXLY_CORRECT");
unbind_variable ("POSIX_PEDANTIC");
if (on_or_off == FLAG_ON)
{
bind_variable ("POSIXLY_CORRECT", "");
stupidly_hack_special_variables ("POSIXLY_CORRECT");
}
}
else
{
register int i;
for (i = 0; o_options[i].name; i++)
{
if (STREQ (option_name, o_options[i].name))
{
option_char = o_options[i].letter;
break;
}
}
if (option_char == -1)
{
builtin_error ("%s: unknown option name", option_name);
return (EXECUTION_FAILURE);
}
if (change_flag (option_char, on_or_off) == FLAG_ERROR)
{
bad_option (option_name);
return (EXECUTION_FAILURE);
}
}
return (EXECUTION_SUCCESS);
}
/* Set some flags from the word values in the input list. If LIST is empty,
then print out the values of the variables instead. If LIST contains
non-flags, then set $1 - $9 to the successive words of LIST. */
set_builtin (list)
WORD_LIST *list;
{
int on_or_off, flag_name, force_assignment = 0;
if (!list)
{
SHELL_VAR **vars;
vars = all_shell_variables ();
if (vars)
{
print_var_list (vars);
free (vars);
}
vars = all_shell_functions ();
if (vars)
{
print_var_list (vars);
free (vars);
}
return (EXECUTION_SUCCESS);
}
/* Check validity of flag arguments. */
if (*list->word->word == '-' || *list->word->word == '+')
{
register char *arg;
WORD_LIST *save_list = list;
while (list && (arg = list->word->word))
{
char c;
if (arg[0] != '-' && arg[0] != '+')
break;
/* `-' or `--' signifies end of flag arguments. */
if (arg[0] == '-' &&
(!arg[1] || (arg[1] == '-' && !arg[2])))
break;
while (c = *++arg)
{
if (find_flag (c) == FLAG_UNKNOWN && c != 'o')
{
char s[2];
s[0] = c; s[1] = '\0';
bad_option (s);
if (c == '?')
builtin_usage ();
return (c == '?' ? EXECUTION_SUCCESS : EXECUTION_FAILURE);
}
}
list = list->next;
}
list = save_list;
}
/* Do the set command. While the list consists of words starting with
'-' or '+' treat them as flags, otherwise, start assigning them to
$1 ... $n. */
while (list)
{
char *string = list->word->word;
/* If the argument is `--' or `-' then signal the end of the list
and remember the remaining arguments. */
if (string[0] == '-' && (!string[1] || (string[1] == '-' && !string[2])))
{
list = list->next;
/* `set --' unsets the positional parameters. */
if (string[1] == '-')
force_assignment = 1;
/* Until told differently, the old shell behaviour of
`set - [arg ...]' being equivalent to `set +xv [arg ...]'
stands. Posix.2 says the behaviour is marked as obsolescent. */
else
{
change_flag ('x', '+');
change_flag ('v', '+');
}
break;
}
if ((on_or_off = *string) &&
(on_or_off == '-' || on_or_off == '+'))
{
int i = 1;
while (flag_name = string[i++])
{
if (flag_name == '?')
{
builtin_usage ();
return (EXECUTION_SUCCESS);
}
else if (flag_name == 'o') /* -+o option-name */
{
char *option_name;
WORD_LIST *opt;
opt = list->next;
if (!opt)
{
list_minus_o_opts ();
continue;
}
option_name = opt->word->word;
if (!option_name || !*option_name || (*option_name == '-'))
{
list_minus_o_opts ();
continue;
}
list = list->next; /* Skip over option name. */
if (set_minus_o_option (on_or_off, option_name) != EXECUTION_SUCCESS)
return (EXECUTION_FAILURE);
}
else
{
if (change_flag (flag_name, on_or_off) == FLAG_ERROR)
{
char opt[3];
opt[0] = on_or_off;
opt[1] = flag_name;
opt[2] = '\0';
bad_option (opt);
builtin_usage ();
return (EXECUTION_FAILURE);
}
}
}
}
else
{
break;
}
list = list->next;
}
/* Assigning $1 ... $n */
if (list || force_assignment)
remember_args (list, 1);
return (EXECUTION_SUCCESS);
}
$BUILTIN unset
$FUNCTION unset_builtin
$SHORT_DOC unset [-f] [-v] [name ...]
For each NAME, remove the corresponding variable or function. Given
the `-v', unset will only act on variables. Given the `-f' flag,
unset will only act on functions. With neither flag, unset first
tries to unset a variable, and if that fails, then tries to unset a
function. Some variables (such as PATH and IFS) cannot be unset; also
see readonly.
$END
#define NEXT_VARIABLE() any_failed++; list = list->next; continue;
unset_builtin (list)
WORD_LIST *list;
{
int unset_function, unset_variable, unset_array, opt, any_failed;
char *name;
unset_function = unset_variable = unset_array = any_failed = 0;
reset_internal_getopt ();
while ((opt = internal_getopt (list, "fv")) != -1)
{
switch (opt)
{
case 'f':
unset_function = 1;
break;
case 'v':
unset_variable = 1;
break;
default:
builtin_usage ();
return (EXECUTION_FAILURE);
}
}
list = loptend;
if (unset_function && unset_variable)
{
builtin_error ("cannot simultaneously unset a function and a variable");
return (EXECUTION_FAILURE);
}
while (list)
{
SHELL_VAR *var;
int tem;
#if defined (ARRAY_VARS)
char *t;
#endif
name = list->word->word;
#if defined (ARRAY_VARS)
if (!unset_function && valid_array_reference (name))
{
t = strchr (name, '[');
*t++ = '\0';
unset_array++;
}
#endif
var = unset_function ? find_function (name) : find_variable (name);
if (var && !unset_function && non_unsettable_p (var))
{
builtin_error ("%s: cannot unset", name);
NEXT_VARIABLE ();
}
/* Posix.2 says that unsetting readonly variables is an error. */
if (var && readonly_p (var))
{
builtin_error ("%s: cannot unset: readonly %s",
name, unset_function ? "function" : "variable");
NEXT_VARIABLE ();
}
/* Unless the -f option is supplied, the name refers to a variable. */
#if defined (ARRAY_VARS)
if (var && unset_array)
{
if (array_p (var) == 0)
{
builtin_error ("%s: not an array variable", name);
NEXT_VARIABLE ();
}
else
tem = unbind_array_element (var, t);
}
else
#endif /* ARRAY_VARS */
tem = makunbound (name, unset_function ? shell_functions : shell_variables);
/* This is what Posix.2 draft 11+ says. ``If neither -f nor -v
is specified, the name refers to a variable; if a variable by
that name does not exist, a function by that name, if any,
shall be unset.'' */
if ((tem == -1) && !unset_function && !unset_variable)
tem = makunbound (name, shell_functions);
if (tem == -1)
any_failed++;
else if (!unset_function)
stupidly_hack_special_variables (name);
list = list->next;
}
if (any_failed)
return (EXECUTION_FAILURE);
else
return (EXECUTION_SUCCESS);
}
-50
View File
@@ -1,50 +0,0 @@
/* unwind_prot.h - Macros and functions for hacking unwind protection. */
/* Copyright (C) 1993 Free Software Foundation, Inc.
This file is part of GNU Bash, the Bourne Again SHell.
Bash is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation; either version 2, or (at your option) any later
version.
Bash is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
for more details.
You should have received a copy of the GNU General Public License along
with Bash; see the file COPYING. If not, write to the Free Software
Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */
#if !defined (_UNWIND_PROT_H)
#define _UNWIND_PROT_H
/* Run a function without interrupts. */
extern void begin_unwind_frame ();
extern void discard_unwind_frame ();
extern void run_unwind_frame ();
extern void add_unwind_protect ();
extern void remove_unwind_protect ();
extern void run_unwind_protects ();
extern void unwind_protect_var ();
/* Define for people who like their code to look a certain way. */
#define end_unwind_frame()
/* How to protect an integer. */
#define unwind_protect_int(X) unwind_protect_var (&(X), (char *)(X), sizeof (int))
/* How to protect a pointer to a string. */
#define unwind_protect_string(X) \
unwind_protect_var ((int *)&(X), (X), sizeof (char *))
/* How to protect any old pointer. */
#define unwind_protect_pointer(X) unwind_protect_string (X)
/* How to protect the contents of a jmp_buf. */
#define unwind_protect_jmp_buf(X) \
unwind_protect_var ((int *)(X), (char *)(X), sizeof (procenv_t))
#endif /* _UNWIND_PROT_H */
-1250
View File
File diff suppressed because it is too large Load Diff
-391
View File
@@ -1,391 +0,0 @@
/* command.h -- The structures used internally to represent commands, and
the extern declarations of the functions used to create them. */
/* Copyright (C) 1993-2010 Free Software Foundation, Inc.
This file is part of GNU Bash, the Bourne Again SHell.
Bash is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Bash is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with Bash. If not, see <http://www.gnu.org/licenses/>.
*/
#if !defined (_COMMAND_H_)
#define _COMMAND_H_
#include "stdc.h"
/* Instructions describing what kind of thing to do for a redirection. */
enum r_instruction {
r_output_direction, r_input_direction, r_inputa_direction,
r_appending_to, r_reading_until, r_reading_string,
r_duplicating_input, r_duplicating_output, r_deblank_reading_until,
r_close_this, r_err_and_out, r_input_output, r_output_force,
r_duplicating_input_word, r_duplicating_output_word,
r_move_input, r_move_output, r_move_input_word, r_move_output_word,
r_append_err_and_out
};
/* Redirection flags; values for rflags */
#define REDIR_VARASSIGN 0x01
/* Redirection errors. */
#define AMBIGUOUS_REDIRECT -1
#define NOCLOBBER_REDIRECT -2
#define RESTRICTED_REDIRECT -3 /* can only happen in restricted shells. */
#define HEREDOC_REDIRECT -4 /* here-doc temp file can't be created */
#define BADVAR_REDIRECT -5 /* something wrong with {varname}redir */
#define CLOBBERING_REDIRECT(ri) \
(ri == r_output_direction || ri == r_err_and_out)
#define OUTPUT_REDIRECT(ri) \
(ri == r_output_direction || ri == r_input_output || ri == r_err_and_out || ri == r_append_err_and_out)
#define INPUT_REDIRECT(ri) \
(ri == r_input_direction || ri == r_inputa_direction || ri == r_input_output)
#define WRITE_REDIRECT(ri) \
(ri == r_output_direction || \
ri == r_input_output || \
ri == r_err_and_out || \
ri == r_appending_to || \
ri == r_append_err_and_out || \
ri == r_output_force)
/* redirection needs translation */
#define TRANSLATE_REDIRECT(ri) \
(ri == r_duplicating_input_word || ri == r_duplicating_output_word || \
ri == r_move_input_word || ri == r_move_output_word)
/* Command Types: */
enum command_type { cm_for, cm_case, cm_while, cm_if, cm_simple, cm_select,
cm_connection, cm_function_def, cm_until, cm_group,
cm_arith, cm_cond, cm_arith_for, cm_subshell, cm_coproc };
/* Possible values for the `flags' field of a WORD_DESC. */
#define W_HASDOLLAR 0x000001 /* Dollar sign present. */
#define W_QUOTED 0x000002 /* Some form of quote character is present. */
#define W_ASSIGNMENT 0x000004 /* This word is a variable assignment. */
#define W_SPLITSPACE 0x000008 /* Split this word on " " regardless of IFS */
#define W_NOSPLIT 0x000010 /* Do not perform word splitting on this word because ifs is empty string. */
#define W_NOGLOB 0x000020 /* Do not perform globbing on this word. */
#define W_NOSPLIT2 0x000040 /* Don't split word except for $@ expansion (using spaces) because context does not allow it. */
#define W_TILDEEXP 0x000080 /* Tilde expand this assignment word */
#define W_DOLLARAT 0x000100 /* $@ and its special handling */
#define W_DOLLARSTAR 0x000200 /* $* and its special handling */
#define W_NOCOMSUB 0x000400 /* Don't perform command substitution on this word */
#define W_ASSIGNRHS 0x000800 /* Word is rhs of an assignment statement */
#define W_NOTILDE 0x001000 /* Don't perform tilde expansion on this word */
#define W_ITILDE 0x002000 /* Internal flag for word expansion */
#define W_NOEXPAND 0x004000 /* Don't expand at all -- do quote removal */
#define W_COMPASSIGN 0x008000 /* Compound assignment */
#define W_ASSNBLTIN 0x010000 /* word is a builtin command that takes assignments */
#define W_ASSIGNARG 0x020000 /* word is assignment argument to command */
#define W_HASQUOTEDNULL 0x040000 /* word contains a quoted null character */
#define W_DQUOTE 0x080000 /* word should be treated as if double-quoted */
#define W_NOPROCSUB 0x100000 /* don't perform process substitution */
#define W_HASCTLESC 0x200000 /* word contains literal CTLESC characters */
#define W_ASSIGNASSOC 0x400000 /* word looks like associative array assignment */
#define W_ASSIGNARRAY 0x800000 /* word looks like a compound indexed array assignment */
#define W_ARRAYIND 0x1000000 /* word is an array index being expanded */
#define W_ASSNGLOBAL 0x2000000 /* word is a global assignment to declare (declare/typeset -g) */
#define W_NOBRACE 0x4000000 /* Don't perform brace expansion */
/* Possible values for subshell_environment */
#define SUBSHELL_ASYNC 0x01 /* subshell caused by `command &' */
#define SUBSHELL_PAREN 0x02 /* subshell caused by ( ... ) */
#define SUBSHELL_COMSUB 0x04 /* subshell caused by `command` or $(command) */
#define SUBSHELL_FORK 0x08 /* subshell caused by executing a disk command */
#define SUBSHELL_PIPE 0x10 /* subshell from a pipeline element */
#define SUBSHELL_PROCSUB 0x20 /* subshell caused by <(command) or >(command) */
#define SUBSHELL_COPROC 0x40 /* subshell from a coproc pipeline */
#define SUBSHELL_RESETTRAP 0x80 /* subshell needs to reset trap strings on first call to trap */
/* A structure which represents a word. */
typedef struct word_desc {
char *word; /* Zero terminated string. */
int flags; /* Flags associated with this word. */
} WORD_DESC;
/* A linked list of words. */
typedef struct word_list {
struct word_list *next;
WORD_DESC *word;
} WORD_LIST;
/* **************************************************************** */
/* */
/* Shell Command Structs */
/* */
/* **************************************************************** */
/* What a redirection descriptor looks like. If the redirection instruction
is ri_duplicating_input or ri_duplicating_output, use DEST, otherwise
use the file in FILENAME. Out-of-range descriptors are identified by a
negative DEST. */
typedef union {
int dest; /* Place to redirect REDIRECTOR to, or ... */
WORD_DESC *filename; /* filename to redirect to. */
} REDIRECTEE;
/* Structure describing a redirection. If REDIRECTOR is negative, the parser
(or translator in redir.c) encountered an out-of-range file descriptor. */
typedef struct redirect {
struct redirect *next; /* Next element, or NULL. */
REDIRECTEE redirector; /* Descriptor or varname to be redirected. */
int rflags; /* Private flags for this redirection */
int flags; /* Flag value for `open'. */
enum r_instruction instruction; /* What to do with the information. */
REDIRECTEE redirectee; /* File descriptor or filename */
char *here_doc_eof; /* The word that appeared in <<foo. */
} REDIRECT;
/* An element used in parsing. A single word or a single redirection.
This is an ephemeral construct. */
typedef struct element {
WORD_DESC *word;
REDIRECT *redirect;
} ELEMENT;
/* Possible values for command->flags. */
#define CMD_WANT_SUBSHELL 0x01 /* User wants a subshell: ( command ) */
#define CMD_FORCE_SUBSHELL 0x02 /* Shell needs to force a subshell. */
#define CMD_INVERT_RETURN 0x04 /* Invert the exit value. */
#define CMD_IGNORE_RETURN 0x08 /* Ignore the exit value. For set -e. */
#define CMD_NO_FUNCTIONS 0x10 /* Ignore functions during command lookup. */
#define CMD_INHIBIT_EXPANSION 0x20 /* Do not expand the command words. */
#define CMD_NO_FORK 0x40 /* Don't fork; just call execve */
#define CMD_TIME_PIPELINE 0x80 /* Time a pipeline */
#define CMD_TIME_POSIX 0x100 /* time -p; use POSIX.2 time output spec. */
#define CMD_AMPERSAND 0x200 /* command & */
#define CMD_STDIN_REDIR 0x400 /* async command needs implicit </dev/null */
#define CMD_COMMAND_BUILTIN 0x0800 /* command executed by `command' builtin */
#define CMD_COPROC_SUBSHELL 0x1000
#define CMD_LASTPIPE 0x2000
/* What a command looks like. */
typedef struct command {
enum command_type type; /* FOR CASE WHILE IF CONNECTION or SIMPLE. */
int flags; /* Flags controlling execution environment. */
int line; /* line number the command starts on */
REDIRECT *redirects; /* Special redirects for FOR CASE, etc. */
union {
struct for_com *For;
struct case_com *Case;
struct while_com *While;
struct if_com *If;
struct connection *Connection;
struct simple_com *Simple;
struct function_def *Function_def;
struct group_com *Group;
#if defined (SELECT_COMMAND)
struct select_com *Select;
#endif
#if defined (DPAREN_ARITHMETIC)
struct arith_com *Arith;
#endif
#if defined (COND_COMMAND)
struct cond_com *Cond;
#endif
#if defined (ARITH_FOR_COMMAND)
struct arith_for_com *ArithFor;
#endif
struct subshell_com *Subshell;
struct coproc_com *Coproc;
} value;
} COMMAND;
/* Structure used to represent the CONNECTION type. */
typedef struct connection {
int ignore; /* Unused; simplifies make_command (). */
COMMAND *first; /* Pointer to the first command. */
COMMAND *second; /* Pointer to the second command. */
int connector; /* What separates this command from others. */
} CONNECTION;
/* Structures used to represent the CASE command. */
/* Values for FLAGS word in a PATTERN_LIST */
#define CASEPAT_FALLTHROUGH 0x01
#define CASEPAT_TESTNEXT 0x02
/* Pattern/action structure for CASE_COM. */
typedef struct pattern_list {
struct pattern_list *next; /* Clause to try in case this one failed. */
WORD_LIST *patterns; /* Linked list of patterns to test. */
COMMAND *action; /* Thing to execute if a pattern matches. */
int flags;
} PATTERN_LIST;
/* The CASE command. */
typedef struct case_com {
int flags; /* See description of CMD flags. */
int line; /* line number the `case' keyword appears on */
WORD_DESC *word; /* The thing to test. */
PATTERN_LIST *clauses; /* The clauses to test against, or NULL. */
} CASE_COM;
/* FOR command. */
typedef struct for_com {
int flags; /* See description of CMD flags. */
int line; /* line number the `for' keyword appears on */
WORD_DESC *name; /* The variable name to get mapped over. */
WORD_LIST *map_list; /* The things to map over. This is never NULL. */
COMMAND *action; /* The action to execute.
During execution, NAME is bound to successive
members of MAP_LIST. */
} FOR_COM;
#if defined (ARITH_FOR_COMMAND)
typedef struct arith_for_com {
int flags;
int line; /* generally used for error messages */
WORD_LIST *init;
WORD_LIST *test;
WORD_LIST *step;
COMMAND *action;
} ARITH_FOR_COM;
#endif
#if defined (SELECT_COMMAND)
/* KSH SELECT command. */
typedef struct select_com {
int flags; /* See description of CMD flags. */
int line; /* line number the `select' keyword appears on */
WORD_DESC *name; /* The variable name to get mapped over. */
WORD_LIST *map_list; /* The things to map over. This is never NULL. */
COMMAND *action; /* The action to execute.
During execution, NAME is bound to the member of
MAP_LIST chosen by the user. */
} SELECT_COM;
#endif /* SELECT_COMMAND */
/* IF command. */
typedef struct if_com {
int flags; /* See description of CMD flags. */
COMMAND *test; /* Thing to test. */
COMMAND *true_case; /* What to do if the test returned non-zero. */
COMMAND *false_case; /* What to do if the test returned zero. */
} IF_COM;
/* WHILE command. */
typedef struct while_com {
int flags; /* See description of CMD flags. */
COMMAND *test; /* Thing to test. */
COMMAND *action; /* Thing to do while test is non-zero. */
} WHILE_COM;
#if defined (DPAREN_ARITHMETIC)
/* The arithmetic evaluation command, ((...)). Just a set of flags and
a WORD_LIST, of which the first element is the only one used, for the
time being. */
typedef struct arith_com {
int flags;
int line;
WORD_LIST *exp;
} ARITH_COM;
#endif /* DPAREN_ARITHMETIC */
/* The conditional command, [[...]]. This is a binary tree -- we slippped
a recursive-descent parser into the YACC grammar to parse it. */
#define COND_AND 1
#define COND_OR 2
#define COND_UNARY 3
#define COND_BINARY 4
#define COND_TERM 5
#define COND_EXPR 6
typedef struct cond_com {
int flags;
int line;
int type;
WORD_DESC *op;
struct cond_com *left, *right;
} COND_COM;
/* The "simple" command. Just a collection of words and redirects. */
typedef struct simple_com {
int flags; /* See description of CMD flags. */
int line; /* line number the command starts on */
WORD_LIST *words; /* The program name, the arguments,
variable assignments, etc. */
REDIRECT *redirects; /* Redirections to perform. */
} SIMPLE_COM;
/* The "function definition" command. */
typedef struct function_def {
int flags; /* See description of CMD flags. */
int line; /* Line number the function def starts on. */
WORD_DESC *name; /* The name of the function. */
COMMAND *command; /* The parsed execution tree. */
char *source_file; /* file in which function was defined, if any */
} FUNCTION_DEF;
/* A command that is `grouped' allows pipes and redirections to affect all
commands in the group. */
typedef struct group_com {
int ignore; /* See description of CMD flags. */
COMMAND *command;
} GROUP_COM;
typedef struct subshell_com {
int flags;
COMMAND *command;
} SUBSHELL_COM;
#define COPROC_RUNNING 0x01
#define COPROC_DEAD 0x02
typedef struct coproc {
char *c_name;
pid_t c_pid;
int c_rfd;
int c_wfd;
int c_rsave;
int c_wsave;
int c_flags;
int c_status;
int c_lock;
} Coproc;
typedef struct coproc_com {
int flags;
char *name;
COMMAND *command;
} COPROC_COM;
extern COMMAND *global_command;
extern Coproc sh_coproc;
/* Possible command errors */
#define CMDERR_DEFAULT 0
#define CMDERR_BADTYPE 1
#define CMDERR_BADCONN 2
#define CMDERR_BADJUMP 3
#define CMDERR_LAST 3
/* Forward declarations of functions declared in copy_cmd.c. */
extern FUNCTION_DEF *copy_function_def_contents __P((FUNCTION_DEF *, FUNCTION_DEF *));
extern FUNCTION_DEF *copy_function_def __P((FUNCTION_DEF *));
extern WORD_DESC *copy_word __P((WORD_DESC *));
extern WORD_LIST *copy_word_list __P((WORD_LIST *));
extern REDIRECT *copy_redirect __P((REDIRECT *));
extern REDIRECT *copy_redirects __P((REDIRECT *));
extern COMMAND *copy_command __P((COMMAND *));
#endif /* _COMMAND_H_ */
-42
View File
@@ -1,42 +0,0 @@
# This file is a shell script that caches the results of configure
# tests for CYGWIN32 so they don't need to be done when cross-compiling.
# AC_FUNC_GETPGRP should also define GETPGRP_VOID
ac_cv_func_getpgrp_void=${ac_cv_func_getpgrp_void='yes'}
# AC_FUNC_SETVBUF_REVERSED should not define anything else
ac_cv_func_setvbuf_reversed=${ac_cv_func_setvbuf_reversed='no'}
# on CYGWIN32, system calls do not restart
ac_cv_sys_restartable_syscalls=${ac_cv_sys_restartable_syscalls='no'}
bash_cv_sys_restartable_syscalls=${bash_cv_sys_restartable_syscalls='no'}
# these may be necessary, but they are currently commented out
#ac_cv_c_bigendian=${ac_cv_c_bigendian='no'}
ac_cv_sizeof_char_p=${ac_cv_sizeof_char_p='4'}
ac_cv_sizeof_int=${ac_cv_sizeof_int='4'}
ac_cv_sizeof_long=${ac_cv_sizeof_long='4'}
ac_cv_sizeof_double=${ac_cv_sizeof_double='8'}
bash_cv_dup2_broken=${bash_cv_dup2_broken='no'}
bash_cv_pgrp_pipe=${bash_cv_pgrp_pipe='no'}
bash_cv_type_rlimit=${bash_cv_type_rlimit='long'}
bash_cv_decl_under_sys_siglist=${bash_cv_decl_under_sys_siglist='no'}
bash_cv_under_sys_siglist=${bash_cv_under_sys_siglist='no'}
bash_cv_sys_siglist=${bash_cv_sys_siglist='no'}
bash_cv_opendir_not_robust=${bash_cv_opendir_not_robust='no'}
bash_cv_getenv_redef=${bash_cv_getenv_redef='yes'}
bash_cv_printf_declared=${bash_cv_printf_declared='yes'}
bash_cv_ulimit_maxfds=${bash_cv_ulimit_maxfds='no'}
bash_cv_getcwd_calls_popen=${bash_cv_getcwd_calls_popen='no'}
bash_cv_must_reinstall_sighandlers=${bash_cv_must_reinstall_sighandlers='no'}
bash_cv_job_control_missing=${bash_cv_job_control_missing='present'}
bash_cv_sys_named_pipes=${bash_cv_sys_named_pipes='missing'}
bash_cv_func_sigsetjmp=${bash_cv_func_sigsetjmp='missing'}
bash_cv_mail_dir=${bash_cv_mail_dir='unknown'}
bash_cv_func_strcoll_broken=${bash_cv_func_strcoll_broken='no'}
bash_cv_type_int32_t=${bash_cv_type_int32_t='int'}
bash_cv_type_u_int32_t=${bash_cv_type_u_int32_t='int'}
ac_cv_type_bits64_t=${ac_cv_type_bits64_t='no'}
# end of cross-build/cygwin32.cache
-1745
View File
File diff suppressed because it is too large Load Diff
-10310
View File
File diff suppressed because it is too large Load Diff
-8715
View File
File diff suppressed because it is too large Load Diff
-10
View File
@@ -1,10 +0,0 @@
@ignore
Copyright (C) 1988-2013 Free Software Foundation, Inc.
@end ignore
@set LASTCHANGE Mon Jul 8 19:31:30 EDT 2013
@set EDITION 4.3
@set VERSION 4.3
@set UPDATED 8 July 2013
@set UPDATED-MONTH July 2013
-238
View File
@@ -1,238 +0,0 @@
#
# Simple makefile for the sample loadable builtins
#
# Copyright (C) 1996 Free Software Foundation, Inc.
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111 USA.
# Include some boilerplate Gnu makefile definitions.
prefix = @prefix@
exec_prefix = @exec_prefix@
bindir = @bindir@
libdir = @libdir@
infodir = @infodir@
includedir = @includedir@
topdir = @top_srcdir@
BUILD_DIR = @BUILD_DIR@
srcdir = @srcdir@
VPATH = .:@srcdir@
@SET_MAKE@
CC = @CC@
RM = rm -f
SHELL = @MAKE_SHELL@
host_os = @host_os@
host_cpu = @host_cpu@
host_vendor = @host_vendor@
CFLAGS = @CFLAGS@
LOCAL_CFLAGS = @LOCAL_CFLAGS@
DEFS = @DEFS@
LOCAL_DEFS = @LOCAL_DEFS@
CPPFLAGS = @CPPFLAGS@
BASHINCDIR = ${topdir}/include
LIBBUILD = ${BUILD_DIR}/lib
INTL_LIBSRC = ${topdir}/lib/intl
INTL_BUILDDIR = ${LIBBUILD}/intl
INTL_INC = @INTL_INC@
LIBINTL_H = @LIBINTL_H@
CCFLAGS = $(DEFS) $(LOCAL_DEFS) $(LOCAL_CFLAGS) $(CFLAGS)
#
# These values are generated for configure by ${topdir}/support/shobj-conf.
# If your system is not supported by that script, but includes facilities for
# dynamic loading of shared objects, please update the script and send the
# changes to bash-maintainers@gnu.org.
#
SHOBJ_CC = @SHOBJ_CC@
SHOBJ_CFLAGS = @SHOBJ_CFLAGS@
SHOBJ_LD = @SHOBJ_LD@
SHOBJ_LDFLAGS = @SHOBJ_LDFLAGS@
SHOBJ_XLDFLAGS = @SHOBJ_XLDFLAGS@
SHOBJ_LIBS = @SHOBJ_LIBS@
SHOBJ_STATUS = @SHOBJ_STATUS@
INC = -I. -I.. -I$(topdir) -I$(topdir)/lib -I$(topdir)/builtins \
-I$(BASHINCDIR) -I$(BUILD_DIR) -I$(LIBBUILD) \
-I$(BUILD_DIR)/builtins $(INTL_INC)
.c.o:
$(SHOBJ_CC) $(SHOBJ_CFLAGS) $(CCFLAGS) $(INC) -c -o $@ $<
ALLPROG = print truefalse sleep pushd finfo logname basename dirname \
tty pathchk tee head mkdir rmdir printenv id whoami \
uname sync push ln unlink cut realpath getconf strftime
OTHERPROG = necho hello cat
all: $(SHOBJ_STATUS)
supported: $(ALLPROG)
others: $(OTHERPROG)
unsupported:
@echo "Your system (${host_os}) is not supported by the"
@echo "${topdir}/support/shobj-conf script."
@echo "If your operating system provides facilities for dynamic"
@echo "loading of shared objects using the dlopen(3) interface,"
@echo "please update the script and re-run configure.
@echo "Please send the changes you made to bash-maintainers@gnu.org"
@echo "for inclusion in future bash releases."
everything: supported others
print: print.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ print.o $(SHOBJ_LIBS)
necho: necho.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ necho.o $(SHOBJ_LIBS)
getconf: getconf.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ getconf.o $(SHOBJ_LIBS)
hello: hello.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ hello.o $(SHOBJ_LIBS)
truefalse: truefalse.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ truefalse.o $(SHOBJ_LIBS)
sleep: sleep.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ sleep.o $(SHOBJ_LIBS)
finfo: finfo.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ finfo.o $(SHOBJ_LIBS)
cat: cat.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ cat.o $(SHOBJ_LIBS)
logname: logname.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ logname.o $(SHOBJ_LIBS)
basename: basename.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ basename.o $(SHOBJ_LIBS)
dirname: dirname.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ dirname.o $(SHOBJ_LIBS)
tty: tty.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ tty.o $(SHOBJ_LIBS)
pathchk: pathchk.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ pathchk.o $(SHOBJ_LIBS)
tee: tee.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ tee.o $(SHOBJ_LIBS)
mkdir: mkdir.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ mkdir.o $(SHOBJ_LIBS)
rmdir: rmdir.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ rmdir.o $(SHOBJ_LIBS)
head: head.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ head.o $(SHOBJ_LIBS)
printenv: printenv.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ printenv.o $(SHOBJ_LIBS)
id: id.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ id.o $(SHOBJ_LIBS)
whoami: whoami.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ whoami.o $(SHOBJ_LIBS)
uname: uname.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ uname.o $(SHOBJ_LIBS)
sync: sync.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ sync.o $(SHOBJ_LIBS)
push: push.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ push.o $(SHOBJ_LIBS)
ln: ln.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ ln.o $(SHOBJ_LIBS)
unlink: unlink.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ unlink.o $(SHOBJ_LIBS)
cut: cut.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ cut.o $(SHOBJ_LIBS)
realpath: realpath.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ realpath.o $(SHOBJ_LIBS)
strftime: strftime.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ strftime.o $(SHOBJ_LIBS)
# pushd is a special case. We use the same source that the builtin version
# uses, with special compilation options.
#
pushd.c: ${topdir}/builtins/pushd.def
$(RM) $@
${BUILD_DIR}/builtins/mkbuiltins -D ${topdir}/builtins ${topdir}/builtins/pushd.def
pushd.o: pushd.c
$(RM) $@
$(SHOBJ_CC) -DHAVE_CONFIG_H -DPUSHD_AND_POPD -DLOADABLE_BUILTIN $(SHOBJ_CFLAGS) $(CFLAGS) $(CPPFLAGS) $(INC) -c -o $@ $<
pushd: pushd.o
$(SHOBJ_LD) $(SHOBJ_LDFLAGS) $(SHOBJ_XLDFLAGS) -o $@ pushd.o $(SHOBJ_LIBS)
clean:
$(RM) $(ALLPROG) $(OTHERPROG) *.o
-( cd perl && ${MAKE} ${MFLAGS} $@ )
mostlyclean: clean
-( cd perl && ${MAKE} ${MFLAGS} $@ )
distclean maintainer-clean: clean
$(RM) Makefile pushd.c
-( cd perl && ${MAKE} ${MFLAGS} $@ )
print.o: print.c
truefalse.o: truefalse.c
sleep.o: sleep.c
finfo.o: finfo.c
logname.o: logname.c
basename.o: basename.c
dirname.o: dirname.c
tty.o: tty.c
pathchk.o: pathchk.c
tee.o: tee.c
head.o: head.c
rmdir.o: rmdir.c
necho.o: necho.c
getconf.o: getconf.c
hello.o: hello.c
cat.o: cat.c
printenv.o: printenv.c
id.o: id.c
whoami.o: whoami.c
uname.o: uname.c
sync.o: sync.c
push.o: push.c
mkdir.o: mkdir.c
realpath.o: realpath.c
strftime.o: strftime.c
-5408
View File
File diff suppressed because it is too large Load Diff
-76
View File
@@ -1,76 +0,0 @@
# This makefile for Readline library documentation is in -*- text -*- mode.
# Emacs likes it that way.
RM = rm -f
MAKEINFO = makeinfo
TEXI2DVI = texi2dvi
TEXI2HTML = texi2html
QUIETPS = #set this to -q to shut up dvips
DVIPS = dvips -D 300 $(QUIETPS) -o $@ # tricky
INSTALL_DATA = cp
infodir = /usr/local/info
RLSRC = rlman.texinfo rluser.texinfo rltech.texinfo
HISTSRC = hist.texinfo hsuser.texinfo hstech.texinfo
DVIOBJ = readline.dvi history.dvi
INFOOBJ = readline.info history.info
PSOBJ = readline.ps history.ps
HTMLOBJ = readline.html history.html
all: info dvi html ps
nodvi: info html
readline.dvi: $(RLSRC)
$(TEXI2DVI) rlman.texinfo
mv rlman.dvi readline.dvi
readline.info: $(RLSRC)
$(MAKEINFO) --no-split -o $@ rlman.texinfo
history.dvi: ${HISTSRC}
$(TEXI2DVI) hist.texinfo
mv hist.dvi history.dvi
history.info: ${HISTSRC}
$(MAKEINFO) --no-split -o $@ hist.texinfo
readline.ps: readline.dvi
$(RM) $@
$(DVIPS) readline.dvi
history.ps: history.dvi
$(RM) $@
$(DVIPS) history.dvi
readline.html: ${RLSRC}
$(TEXI2HTML) rlman.texinfo
sed -e 's:rlman.html:readline.html:' -e 's:rlman_toc.html:readline_toc.html:' rlman.html > readline.html
sed -e 's:rlman.html:readline.html:' -e 's:rlman_toc.html:readline_toc.html:' rlman_toc.html > readline_toc.html
$(RM) rlman.html rlman_toc.html
history.html: ${HISTSRC}
$(TEXI2HTML) hist.texinfo
sed -e 's:hist.html:history.html:' -e 's:hist_toc.html:history_toc.html:' hist.html > history.html
sed -e 's:hist.html:history.html:' -e 's:hist_toc.html:history_toc.html:' hist_toc.html > history_toc.html
$(RM) hist.html hist_toc.html
info: $(INFOOBJ)
dvi: $(DVIOBJ)
ps: $(PSOBJ)
html: $(HTMLOBJ)
clean:
$(RM) *.aux *.cp *.fn *.ky *.log *.pg *.toc *.tp *.vr *.cps *.pgs \
*.fns *.kys *.tps *.vrs *.o core
distclean: clean
mostlyclean: clean
maintainer-clean: clean
$(RM) *.dvi *.info *.info-* *.ps *.html
install: info
${INSTALL_DATA} readline.info $(infodir)/readline.info
${INSTALL_DATA} history.info $(infodir)/history.info
-789
View File
@@ -1,789 +0,0 @@
/* isearch.c - incremental searching */
/* **************************************************************** */
/* */
/* I-Search and Searching */
/* */
/* **************************************************************** */
/* Copyright (C) 1987-2012 Free Software Foundation, Inc.
This file is part of the GNU Readline Library (Readline), a library
for reading lines of text with interactive input and history editing.
Readline is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Readline is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with Readline. If not, see <http://www.gnu.org/licenses/>.
*/
#define READLINE_LIBRARY
#if defined (HAVE_CONFIG_H)
# include <config.h>
#endif
#include <sys/types.h>
#include <stdio.h>
#if defined (HAVE_UNISTD_H)
# include <unistd.h>
#endif
#if defined (HAVE_STDLIB_H)
# include <stdlib.h>
#else
# include "ansi_stdlib.h"
#endif
#include "rldefs.h"
#include "rlmbutil.h"
#include "readline.h"
#include "history.h"
#include "rlprivate.h"
#include "xmalloc.h"
/* Variables exported to other files in the readline library. */
char *_rl_isearch_terminators = (char *)NULL;
_rl_search_cxt *_rl_iscxt = 0;
/* Variables imported from other files in the readline library. */
extern HIST_ENTRY *_rl_saved_line_for_history;
static int rl_search_history PARAMS((int, int));
static _rl_search_cxt *_rl_isearch_init PARAMS((int));
static void _rl_isearch_fini PARAMS((_rl_search_cxt *));
static int _rl_isearch_cleanup PARAMS((_rl_search_cxt *, int));
/* Last line found by the current incremental search, so we don't `find'
identical lines many times in a row. Now part of isearch context. */
/* static char *prev_line_found; */
/* Last search string and its length. */
static char *last_isearch_string;
static int last_isearch_string_len;
static char * const default_isearch_terminators = "\033\012";
_rl_search_cxt *
_rl_scxt_alloc (type, flags)
int type, flags;
{
_rl_search_cxt *cxt;
cxt = (_rl_search_cxt *)xmalloc (sizeof (_rl_search_cxt));
cxt->type = type;
cxt->sflags = flags;
cxt->search_string = 0;
cxt->search_string_size = cxt->search_string_index = 0;
cxt->lines = 0;
cxt->allocated_line = 0;
cxt->hlen = cxt->hindex = 0;
cxt->save_point = rl_point;
cxt->save_mark = rl_mark;
cxt->save_line = where_history ();
cxt->last_found_line = cxt->save_line;
cxt->prev_line_found = 0;
cxt->save_undo_list = 0;
cxt->keymap = _rl_keymap;
cxt->okeymap = _rl_keymap;
cxt->history_pos = 0;
cxt->direction = 0;
cxt->prevc = cxt->lastc = 0;
cxt->sline = 0;
cxt->sline_len = cxt->sline_index = 0;
cxt->search_terminators = 0;
return cxt;
}
void
_rl_scxt_dispose (cxt, flags)
_rl_search_cxt *cxt;
int flags;
{
FREE (cxt->search_string);
FREE (cxt->allocated_line);
FREE (cxt->lines);
xfree (cxt);
}
/* Search backwards through the history looking for a string which is typed
interactively. Start with the current line. */
int
rl_reverse_search_history (sign, key)
int sign, key;
{
return (rl_search_history (-sign, key));
}
/* Search forwards through the history looking for a string which is typed
interactively. Start with the current line. */
int
rl_forward_search_history (sign, key)
int sign, key;
{
return (rl_search_history (sign, key));
}
/* Display the current state of the search in the echo-area.
SEARCH_STRING contains the string that is being searched for,
DIRECTION is zero for forward, or non-zero for reverse,
WHERE is the history list number of the current line. If it is
-1, then this line is the starting one. */
static void
rl_display_search (search_string, flags, where)
char *search_string;
int flags, where;
{
char *message;
int msglen, searchlen;
searchlen = (search_string && *search_string) ? strlen (search_string) : 0;
message = (char *)xmalloc (searchlen + 64);
msglen = 0;
#if defined (NOTDEF)
if (where != -1)
{
sprintf (message, "[%d]", where + history_base);
msglen = strlen (message);
}
#endif /* NOTDEF */
message[msglen++] = '(';
if (flags & SF_FAILED)
{
strcpy (message + msglen, "failed ");
msglen += 7;
}
if (flags & SF_REVERSE)
{
strcpy (message + msglen, "reverse-");
msglen += 8;
}
strcpy (message + msglen, "i-search)`");
msglen += 10;
if (search_string)
{
strcpy (message + msglen, search_string);
msglen += searchlen;
}
strcpy (message + msglen, "': ");
rl_message ("%s", message);
xfree (message);
(*rl_redisplay_function) ();
}
static _rl_search_cxt *
_rl_isearch_init (direction)
int direction;
{
_rl_search_cxt *cxt;
register int i;
HIST_ENTRY **hlist;
cxt = _rl_scxt_alloc (RL_SEARCH_ISEARCH, 0);
if (direction < 0)
cxt->sflags |= SF_REVERSE;
cxt->search_terminators = _rl_isearch_terminators ? _rl_isearch_terminators
: default_isearch_terminators;
/* Create an arrary of pointers to the lines that we want to search. */
hlist = history_list ();
rl_maybe_replace_line ();
i = 0;
if (hlist)
for (i = 0; hlist[i]; i++);
/* Allocate space for this many lines, +1 for the current input line,
and remember those lines. */
cxt->lines = (char **)xmalloc ((1 + (cxt->hlen = i)) * sizeof (char *));
for (i = 0; i < cxt->hlen; i++)
cxt->lines[i] = hlist[i]->line;
if (_rl_saved_line_for_history)
cxt->lines[i] = _rl_saved_line_for_history->line;
else
{
/* Keep track of this so we can free it. */
cxt->allocated_line = (char *)xmalloc (1 + strlen (rl_line_buffer));
strcpy (cxt->allocated_line, &rl_line_buffer[0]);
cxt->lines[i] = cxt->allocated_line;
}
cxt->hlen++;
/* The line where we start the search. */
cxt->history_pos = cxt->save_line;
rl_save_prompt ();
/* Initialize search parameters. */
cxt->search_string = (char *)xmalloc (cxt->search_string_size = 128);
cxt->search_string[cxt->search_string_index = 0] = '\0';
/* Normalize DIRECTION into 1 or -1. */
cxt->direction = (direction >= 0) ? 1 : -1;
cxt->sline = rl_line_buffer;
cxt->sline_len = strlen (cxt->sline);
cxt->sline_index = rl_point;
_rl_iscxt = cxt; /* save globally */
return cxt;
}
static void
_rl_isearch_fini (cxt)
_rl_search_cxt *cxt;
{
/* First put back the original state. */
strcpy (rl_line_buffer, cxt->lines[cxt->save_line]);
rl_restore_prompt ();
/* Save the search string for possible later use. */
FREE (last_isearch_string);
last_isearch_string = cxt->search_string;
last_isearch_string_len = cxt->search_string_index;
cxt->search_string = 0;
if (cxt->last_found_line < cxt->save_line)
rl_get_previous_history (cxt->save_line - cxt->last_found_line, 0);
else
rl_get_next_history (cxt->last_found_line - cxt->save_line, 0);
/* If the string was not found, put point at the end of the last matching
line. If last_found_line == orig_line, we didn't find any matching
history lines at all, so put point back in its original position. */
if (cxt->sline_index < 0)
{
if (cxt->last_found_line == cxt->save_line)
cxt->sline_index = cxt->save_point;
else
cxt->sline_index = strlen (rl_line_buffer);
rl_mark = cxt->save_mark;
}
rl_point = cxt->sline_index;
/* Don't worry about where to put the mark here; rl_get_previous_history
and rl_get_next_history take care of it. */
rl_clear_message ();
}
int
_rl_search_getchar (cxt)
_rl_search_cxt *cxt;
{
int c;
/* Read a key and decide how to proceed. */
RL_SETSTATE(RL_STATE_MOREINPUT);
c = cxt->lastc = rl_read_key ();
RL_UNSETSTATE(RL_STATE_MOREINPUT);
#if defined (HANDLE_MULTIBYTE)
/* This ends up with C (and LASTC) being set to the last byte of the
multibyte character. In most cases c == lastc == mb[0] */
if (MB_CUR_MAX > 1 && rl_byte_oriented == 0)
c = cxt->lastc = _rl_read_mbstring (cxt->lastc, cxt->mb, MB_LEN_MAX);
#endif
RL_CHECK_SIGNALS ();
return c;
}
#define ENDSRCH_CHAR(c) \
((CTRL_CHAR (c) || META_CHAR (c) || (c) == RUBOUT) && ((c) != CTRL ('G')))
/* Process just-read character C according to isearch context CXT. Return
-1 if the caller should just free the context and return, 0 if we should
break out of the loop, and 1 if we should continue to read characters. */
int
_rl_isearch_dispatch (cxt, c)
_rl_search_cxt *cxt;
int c;
{
int n, wstart, wlen, limit, cval;
rl_command_func_t *f;
f = (rl_command_func_t *)NULL;
if (c < 0)
{
cxt->sflags |= SF_FAILED;
cxt->history_pos = cxt->last_found_line;
return -1;
}
/* If we are moving into a new keymap, modify cxt->keymap and go on.
This can be a problem if c == ESC and we want to terminate the
incremental search, so we check */
if (c >= 0 && cxt->keymap[c].type == ISKMAP && strchr (cxt->search_terminators, cxt->lastc) == 0)
{
/* _rl_keyseq_timeout specified in milliseconds; _rl_input_queued
takes microseconds, so multiply by 1000. If we don't get any
additional input and we this keymap shadows another function, process
that key as if it was all we read. */
if (_rl_keyseq_timeout > 0 &&
RL_ISSTATE (RL_STATE_CALLBACK) == 0 &&
RL_ISSTATE (RL_STATE_INPUTPENDING) == 0 &&
_rl_pushed_input_available () == 0 &&
((Keymap)(cxt->keymap[c].function))[ANYOTHERKEY].function &&
_rl_input_queued (_rl_keyseq_timeout*1000) == 0)
goto add_character;
cxt->okeymap = cxt->keymap;
cxt->keymap = FUNCTION_TO_KEYMAP (cxt->keymap, c);
cxt->sflags |= SF_CHGKMAP;
/* XXX - we should probably save this sequence, so we can do
something useful if this doesn't end up mapping to a command we
interpret here. Right now we just save the most recent character
that caused the index into a new keymap. */
cxt->prevc = c;
#if defined (HANDLE_MULTIBYTE)
if (MB_CUR_MAX > 1 && rl_byte_oriented == 0)
{
if (cxt->mb[1] == 0)
{
cxt->pmb[0] = c; /* XXX should be == cxt->mb[0] */
cxt->pmb[1] = '\0';
}
else
memcpy (cxt->pmb, cxt->mb, sizeof (cxt->pmb));
}
#endif
return 1;
}
add_character:
/* Translate the keys we do something with to opcodes. */
if (c >= 0 && cxt->keymap[c].type == ISFUNC)
{
f = cxt->keymap[c].function;
if (f == rl_reverse_search_history)
cxt->lastc = (cxt->sflags & SF_REVERSE) ? -1 : -2;
else if (f == rl_forward_search_history)
cxt->lastc = (cxt->sflags & SF_REVERSE) ? -2 : -1;
else if (f == rl_rubout)
cxt->lastc = -3;
else if (c == CTRL ('G') || f == rl_abort)
cxt->lastc = -4;
else if (c == CTRL ('W') || f == rl_unix_word_rubout) /* XXX */
cxt->lastc = -5;
else if (c == CTRL ('Y') || f == rl_yank) /* XXX */
cxt->lastc = -6;
}
/* If we changed the keymap earlier while translating a key sequence into
a command, restore it now that we've succeeded. */
if (cxt->sflags & SF_CHGKMAP)
{
cxt->keymap = cxt->okeymap;
cxt->sflags &= ~SF_CHGKMAP;
/* If we indexed into a new keymap, but didn't map to a command that
affects the search (lastc > 0), and the character that mapped to a
new keymap would have ended the search (ENDSRCH_CHAR(cxt->prevc)),
handle that now as if the previous char would have ended the search
and we would have read the current character. */
/* XXX - should we check cxt->mb? */
if (cxt->lastc > 0 && ENDSRCH_CHAR (cxt->prevc))
{
rl_stuff_char (cxt->lastc);
rl_execute_next (cxt->prevc);
/* XXX - do we insert everything in cxt->pmb? */
return (0);
}
/* Otherwise, if the current character is mapped to self-insert or
nothing (i.e., not an editing command), and the previous character
was a keymap index, then we need to insert both the previous
character and the current character into the search string. */
else if (cxt->lastc > 0 && cxt->prevc > 0 &&
cxt->keymap[cxt->prevc].type == ISKMAP &&
(f == 0 || f == rl_insert))
{
/* Make lastc be the next character read */
/* XXX - do we insert everything in cxt->mb? */
rl_execute_next (cxt->lastc);
/* Dispatch on the previous character (insert into search string) */
cxt->lastc = cxt->prevc;
#if defined (HANDLE_MULTIBYTE)
/* Have to overwrite cxt->mb here because dispatch uses it below */
if (MB_CUR_MAX > 1 && rl_byte_oriented == 0)
{
if (cxt->pmb[1] == 0)
{
cxt->mb[0] = cxt->lastc; /* == cxt->prevc */
cxt->mb[1] = '\0';
}
else
memcpy (cxt->mb, cxt->pmb, sizeof (cxt->mb));
}
#endif
cxt->prevc = 0;
}
else if (cxt->lastc > 0 && cxt->prevc > 0 && f && f != rl_insert)
{
rl_stuff_char (cxt->lastc);
rl_execute_next (cxt->prevc);
/* XXX - do we insert everything in cxt->pmb? */
return (0);
}
}
/* The characters in isearch_terminators (set from the user-settable
variable isearch-terminators) are used to terminate the search but
not subsequently execute the character as a command. The default
value is "\033\012" (ESC and C-J). */
if (cxt->lastc > 0 && strchr (cxt->search_terminators, cxt->lastc))
{
/* ESC still terminates the search, but if there is pending
input or if input arrives within 0.1 seconds (on systems
with select(2)) it is used as a prefix character
with rl_execute_next. WATCH OUT FOR THIS! This is intended
to allow the arrow keys to be used like ^F and ^B are used
to terminate the search and execute the movement command.
XXX - since _rl_input_available depends on the application-
settable keyboard timeout value, this could alternatively
use _rl_input_queued(100000) */
if (cxt->lastc == ESC && _rl_input_available ())
rl_execute_next (ESC);
return (0);
}
#if defined (HANDLE_MULTIBYTE)
if (MB_CUR_MAX > 1 && rl_byte_oriented == 0)
{
if (cxt->lastc >= 0 && (cxt->mb[0] && cxt->mb[1] == '\0') && ENDSRCH_CHAR (cxt->lastc))
{
/* This sets rl_pending_input to LASTC; it will be picked up the next
time rl_read_key is called. */
rl_execute_next (cxt->lastc);
return (0);
}
}
else
#endif
if (cxt->lastc >= 0 && ENDSRCH_CHAR (cxt->lastc))
{
/* This sets rl_pending_input to LASTC; it will be picked up the next
time rl_read_key is called. */
rl_execute_next (cxt->lastc);
return (0);
}
/* Now dispatch on the character. `Opcodes' affect the search string or
state. Other characters are added to the string. */
switch (cxt->lastc)
{
/* search again */
case -1:
if (cxt->search_string_index == 0)
{
if (last_isearch_string)
{
cxt->search_string_size = 64 + last_isearch_string_len;
cxt->search_string = (char *)xrealloc (cxt->search_string, cxt->search_string_size);
strcpy (cxt->search_string, last_isearch_string);
cxt->search_string_index = last_isearch_string_len;
rl_display_search (cxt->search_string, cxt->sflags, -1);
break;
}
return (1);
}
else if (cxt->sflags & SF_REVERSE)
cxt->sline_index--;
else if (cxt->sline_index != cxt->sline_len)
cxt->sline_index++;
else
rl_ding ();
break;
/* switch directions */
case -2:
cxt->direction = -cxt->direction;
if (cxt->direction < 0)
cxt->sflags |= SF_REVERSE;
else
cxt->sflags &= ~SF_REVERSE;
break;
/* delete character from search string. */
case -3: /* C-H, DEL */
/* This is tricky. To do this right, we need to keep a
stack of search positions for the current search, with
sentinels marking the beginning and end. But this will
do until we have a real isearch-undo. */
if (cxt->search_string_index == 0)
rl_ding ();
else
cxt->search_string[--cxt->search_string_index] = '\0';
break;
case -4: /* C-G, abort */
rl_replace_line (cxt->lines[cxt->save_line], 0);
rl_point = cxt->save_point;
rl_mark = cxt->save_mark;
rl_restore_prompt();
rl_clear_message ();
return -1;
case -5: /* C-W */
/* skip over portion of line we already matched and yank word */
wstart = rl_point + cxt->search_string_index;
if (wstart >= rl_end)
{
rl_ding ();
break;
}
/* if not in a word, move to one. */
cval = _rl_char_value (rl_line_buffer, wstart);
if (_rl_walphabetic (cval) == 0)
{
rl_ding ();
break;
}
n = MB_NEXTCHAR (rl_line_buffer, wstart, 1, MB_FIND_NONZERO);;
while (n < rl_end)
{
cval = _rl_char_value (rl_line_buffer, n);
if (_rl_walphabetic (cval) == 0)
break;
n = MB_NEXTCHAR (rl_line_buffer, n, 1, MB_FIND_NONZERO);;
}
wlen = n - wstart + 1;
if (cxt->search_string_index + wlen + 1 >= cxt->search_string_size)
{
cxt->search_string_size += wlen + 1;
cxt->search_string = (char *)xrealloc (cxt->search_string, cxt->search_string_size);
}
for (; wstart < n; wstart++)
cxt->search_string[cxt->search_string_index++] = rl_line_buffer[wstart];
cxt->search_string[cxt->search_string_index] = '\0';
break;
case -6: /* C-Y */
/* skip over portion of line we already matched and yank rest */
wstart = rl_point + cxt->search_string_index;
if (wstart >= rl_end)
{
rl_ding ();
break;
}
n = rl_end - wstart + 1;
if (cxt->search_string_index + n + 1 >= cxt->search_string_size)
{
cxt->search_string_size += n + 1;
cxt->search_string = (char *)xrealloc (cxt->search_string, cxt->search_string_size);
}
for (n = wstart; n < rl_end; n++)
cxt->search_string[cxt->search_string_index++] = rl_line_buffer[n];
cxt->search_string[cxt->search_string_index] = '\0';
break;
/* Add character to search string and continue search. */
default:
if (cxt->search_string_index + 2 >= cxt->search_string_size)
{
cxt->search_string_size += 128;
cxt->search_string = (char *)xrealloc (cxt->search_string, cxt->search_string_size);
}
#if defined (HANDLE_MULTIBYTE)
if (MB_CUR_MAX > 1 && rl_byte_oriented == 0)
{
int j, l;
if (cxt->mb[0] == 0 || cxt->mb[1] == 0)
cxt->search_string[cxt->search_string_index++] = cxt->mb[0];
else
for (j = 0, l = RL_STRLEN (cxt->mb); j < l; )
cxt->search_string[cxt->search_string_index++] = cxt->mb[j++];
}
else
#endif
cxt->search_string[cxt->search_string_index++] = cxt->lastc; /* XXX - was c instead of lastc */
cxt->search_string[cxt->search_string_index] = '\0';
break;
}
for (cxt->sflags &= ~(SF_FOUND|SF_FAILED);; )
{
limit = cxt->sline_len - cxt->search_string_index + 1;
/* Search the current line. */
while ((cxt->sflags & SF_REVERSE) ? (cxt->sline_index >= 0) : (cxt->sline_index < limit))
{
if (STREQN (cxt->search_string, cxt->sline + cxt->sline_index, cxt->search_string_index))
{
cxt->sflags |= SF_FOUND;
break;
}
else
cxt->sline_index += cxt->direction;
}
if (cxt->sflags & SF_FOUND)
break;
/* Move to the next line, but skip new copies of the line
we just found and lines shorter than the string we're
searching for. */
do
{
/* Move to the next line. */
cxt->history_pos += cxt->direction;
/* At limit for direction? */
if ((cxt->sflags & SF_REVERSE) ? (cxt->history_pos < 0) : (cxt->history_pos == cxt->hlen))
{
cxt->sflags |= SF_FAILED;
break;
}
/* We will need these later. */
cxt->sline = cxt->lines[cxt->history_pos];
cxt->sline_len = strlen (cxt->sline);
}
while ((cxt->prev_line_found && STREQ (cxt->prev_line_found, cxt->lines[cxt->history_pos])) ||
(cxt->search_string_index > cxt->sline_len));
if (cxt->sflags & SF_FAILED)
break;
/* Now set up the line for searching... */
cxt->sline_index = (cxt->sflags & SF_REVERSE) ? cxt->sline_len - cxt->search_string_index : 0;
}
if (cxt->sflags & SF_FAILED)
{
/* We cannot find the search string. Ding the bell. */
rl_ding ();
cxt->history_pos = cxt->last_found_line;
rl_display_search (cxt->search_string, cxt->sflags, (cxt->history_pos == cxt->save_line) ? -1 : cxt->history_pos);
return 1;
}
/* We have found the search string. Just display it. But don't
actually move there in the history list until the user accepts
the location. */
if (cxt->sflags & SF_FOUND)
{
cxt->prev_line_found = cxt->lines[cxt->history_pos];
rl_replace_line (cxt->lines[cxt->history_pos], 0);
rl_point = cxt->sline_index;
cxt->last_found_line = cxt->history_pos;
rl_display_search (cxt->search_string, cxt->sflags, (cxt->history_pos == cxt->save_line) ? -1 : cxt->history_pos);
}
return 1;
}
static int
_rl_isearch_cleanup (cxt, r)
_rl_search_cxt *cxt;
int r;
{
if (r >= 0)
_rl_isearch_fini (cxt);
_rl_scxt_dispose (cxt, 0);
_rl_iscxt = 0;
RL_UNSETSTATE(RL_STATE_ISEARCH);
return (r != 0);
}
/* Search through the history looking for an interactively typed string.
This is analogous to i-search. We start the search in the current line.
DIRECTION is which direction to search; >= 0 means forward, < 0 means
backwards. */
static int
rl_search_history (direction, invoking_key)
int direction, invoking_key;
{
_rl_search_cxt *cxt; /* local for now, but saved globally */
int c, r;
RL_SETSTATE(RL_STATE_ISEARCH);
cxt = _rl_isearch_init (direction);
rl_display_search (cxt->search_string, cxt->sflags, -1);
/* If we are using the callback interface, all we do is set up here and
return. The key is that we leave RL_STATE_ISEARCH set. */
if (RL_ISSTATE (RL_STATE_CALLBACK))
return (0);
r = -1;
for (;;)
{
c = _rl_search_getchar (cxt);
/* We might want to handle EOF here (c == 0) */
r = _rl_isearch_dispatch (cxt, cxt->lastc);
if (r <= 0)
break;
}
/* The searching is over. The user may have found the string that she
was looking for, or else she may have exited a failing search. If
LINE_INDEX is -1, then that shows that the string searched for was
not found. We use this to determine where to place rl_point. */
return (_rl_isearch_cleanup (cxt, r));
}
#if defined (READLINE_CALLBACKS)
/* Called from the callback functions when we are ready to read a key. The
callback functions know to call this because RL_ISSTATE(RL_STATE_ISEARCH).
If _rl_isearch_dispatch finishes searching, this function is responsible
for turning off RL_STATE_ISEARCH, which it does using _rl_isearch_cleanup. */
int
_rl_isearch_callback (cxt)
_rl_search_cxt *cxt;
{
int c, r;
c = _rl_search_getchar (cxt);
/* We might want to handle EOF here */
r = _rl_isearch_dispatch (cxt, cxt->lastc);
return (r <= 0) ? _rl_isearch_cleanup (cxt, r) : 0;
}
#endif
File diff suppressed because it is too large Load Diff
-265
View File
@@ -1,265 +0,0 @@
/* casemod.c -- functions to change case of strings */
/* Copyright (C) 2008,2009 Free Software Foundation, Inc.
This file is part of GNU Bash, the Bourne Again SHell.
Bash is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Bash is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with Bash. If not, see <http://www.gnu.org/licenses/>.
*/
#if defined (HAVE_CONFIG_H)
# include <config.h>
#endif
#if defined (HAVE_UNISTD_H)
# include <unistd.h>
#endif /* HAVE_UNISTD_H */
#include <stdc.h>
#include <bashansi.h>
#include <bashintl.h>
#include <bashtypes.h>
#include <stdio.h>
#include <ctype.h>
#include <xmalloc.h>
#include <shmbchar.h>
#include <shmbutil.h>
#include <chartypes.h>
#include <typemax.h>
#include <glob/strmatch.h>
#define _to_wupper(wc) (iswlower (wc) ? towupper (wc) : (wc))
#define _to_wlower(wc) (iswupper (wc) ? towlower (wc) : (wc))
#if !defined (HANDLE_MULTIBYTE)
# define cval(s, i) ((s)[(i)])
# define iswalnum(c) (isalnum(c))
# define TOGGLE(x) (ISUPPER (x) ? tolower (x) : (TOUPPER (x)))
#else
# define TOGGLE(x) (iswupper (x) ? towlower (x) : (_to_wupper(x)))
#endif
/* These must agree with the defines in externs.h */
#define CASE_NOOP 0x0000
#define CASE_LOWER 0x0001
#define CASE_UPPER 0x0002
#define CASE_CAPITALIZE 0x0004
#define CASE_UNCAP 0x0008
#define CASE_TOGGLE 0x0010
#define CASE_TOGGLEALL 0x0020
#define CASE_UPFIRST 0x0040
#define CASE_LOWFIRST 0x0080
#define CASE_USEWORDS 0x1000 /* modify behavior to act on words in passed string */
extern char *substring __P((char *, int, int));
#ifndef UCHAR_MAX
# define UCHAR_MAX TYPE_MAXIMUM(unsigned char)
#endif
#if defined (HANDLE_MULTIBYTE)
static wchar_t
cval (s, i)
char *s;
int i;
{
size_t tmp;
wchar_t wc;
int l;
mbstate_t mps;
if (MB_CUR_MAX == 1 || is_basic (s[i]))
return ((wchar_t)s[i]);
l = strlen (s);
if (i >= (l - 1))
return ((wchar_t)s[i]);
memset (&mps, 0, sizeof (mbstate_t));
tmp = mbrtowc (&wc, s + i, l - i, &mps);
if (MB_INVALIDCH (tmp) || MB_NULLWCH (tmp))
return ((wchar_t)s[i]);
return wc;
}
#endif
/* Modify the case of characters in STRING matching PAT based on the value of
FLAGS. If PAT is null, modify the case of each character */
char *
sh_modcase (string, pat, flags)
const char *string;
char *pat;
int flags;
{
int start, next, end;
int inword, c, nc, nop, match, usewords;
char *ret, *s;
wchar_t wc;
#if defined (HANDLE_MULTIBYTE)
wchar_t nwc;
char mb[MB_LEN_MAX+1];
int mlen;
size_t m;
mbstate_t state;
#endif
if (string == 0 || *string == 0)
{
ret = (char *)xmalloc (1);
ret[0] = '\0';
return ret;
}
#if defined (HANDLE_MULTIBYTE)
memset (&state, 0, sizeof (mbstate_t));
#endif
start = 0;
end = strlen (string);
ret = (char *)xmalloc (end + 1);
strcpy (ret, string);
/* See if we are supposed to split on alphanumerics and operate on each word */
usewords = (flags & CASE_USEWORDS);
flags &= ~CASE_USEWORDS;
inword = 0;
while (start < end)
{
wc = cval (ret, start);
if (iswalnum (wc) == 0)
{
inword = 0;
#if 0
ADVANCE_CHAR (ret, end, start);
continue;
#endif
}
if (pat)
{
next = start;
ADVANCE_CHAR (ret, end, next);
s = substring (ret, start, next);
match = strmatch (pat, s, FNM_EXTMATCH) != FNM_NOMATCH;
free (s);
if (match == 0)
{
start = next;
inword = 1;
continue;
}
}
/* XXX - for now, the toggling operators work on the individual
words in the string, breaking on alphanumerics. Should I
leave the capitalization operators to do that also? */
if (flags == CASE_CAPITALIZE)
{
if (usewords)
nop = inword ? CASE_LOWER : CASE_UPPER;
else
nop = (start > 0) ? CASE_LOWER : CASE_UPPER;
inword = 1;
}
else if (flags == CASE_UNCAP)
{
if (usewords)
nop = inword ? CASE_UPPER : CASE_LOWER;
else
nop = (start > 0) ? CASE_UPPER : CASE_LOWER;
inword = 1;
}
else if (flags == CASE_UPFIRST)
{
if (usewords)
nop = inword ? CASE_NOOP : CASE_UPPER;
else
nop = (start > 0) ? CASE_NOOP : CASE_UPPER;
inword = 1;
}
else if (flags == CASE_LOWFIRST)
{
if (usewords)
nop = inword ? CASE_NOOP : CASE_LOWER;
else
nop = (start > 0) ? CASE_NOOP : CASE_LOWER;
inword = 1;
}
else if (flags == CASE_TOGGLE)
{
nop = inword ? CASE_NOOP : CASE_TOGGLE;
inword = 1;
}
else
nop = flags;
/* Need to check UCHAR_MAX since wc may have already been converted to a
wide character by cval() */
if (MB_CUR_MAX == 1 || (wc <= UCHAR_MAX && is_basic ((int)wc)))
{
singlebyte:
switch (nop)
{
default:
case CASE_NOOP: nc = wc; break;
case CASE_UPPER: nc = TOUPPER (wc); break;
case CASE_LOWER: nc = TOLOWER (wc); break;
case CASE_TOGGLEALL:
case CASE_TOGGLE: nc = TOGGLE (wc); break;
}
ret[start] = nc;
}
#if defined (HANDLE_MULTIBYTE)
else
{
m = mbrtowc (&wc, string + start, end - start, &state);
if (MB_INVALIDCH (m))
{
wc = (wchar_t)string[start];
goto singlebyte;
}
else if (MB_NULLWCH (m))
wc = L'\0';
switch (nop)
{
default:
case CASE_NOOP: nwc = wc; break;
case CASE_UPPER: nwc = _to_wupper (wc); break;
case CASE_LOWER: nwc = _to_wlower (wc); break;
case CASE_TOGGLEALL:
case CASE_TOGGLE: nwc = TOGGLE (wc); break;
}
if (nwc != wc) /* just skip unchanged characters */
{
mlen = wcrtomb (mb, nwc, &state);
if (mlen > 0)
mb[mlen] = '\0';
/* Assume the same width */
strncpy (ret + start, mb, mlen);
}
}
#endif
/* This assumes that the upper and lower case versions are the same width. */
ADVANCE_CHAR (ret, end, start);
}
return ret;
}
-6165
View File
File diff suppressed because it is too large Load Diff
-9662
View File
File diff suppressed because it is too large Load Diff
-9
View File
@@ -1,9 +0,0 @@
BUILD_DIR=/usr/local/build/chet/bash/bash-current
THIS_SH=$BUILD_DIR/bash
PATH=$PATH:$BUILD_DIR
export THIS_SH PATH
rm -f /tmp/xx
/bin/sh "$@"
-398
View File
@@ -1,398 +0,0 @@
# this is needed so that the bad assignments (b[]=bcde, for example) do not
# cause fatal shell errors when in posix mode
set +o posix
set +a
# The calls to egrep -v are to filter out builtin array variables that are
# automatically set and possibly contain values that vary.
# first make sure we handle the basics
x=()
echo ${x[@]}
unset x
# this should be an error
test=(first & second)
echo $?
unset test
# make sure declare -a converts an existing variable to an array
unset a
a=abcde
declare -a a
echo ${a[0]}
unset a
a=abcde
a[2]=bdef
unset b
declare -a b[256]
unset c[2]
unset c[*]
a[1]=
_ENV=/bin/true
x=${_ENV[(_$-=0)+(_=1)-_${-%%*i*}]}
declare -r c[100]
echo ${a[0]} ${a[4]}
echo ${a[@]}
echo ${a[*]}
# this should print out values, too
declare -a | egrep -v '(BASH_VERSINFO|PIPESTATUS|GROUPS)'
unset a[7]
echo ${a[*]}
unset a[4]
echo ${a[*]}
echo ${a}
echo "${a}"
echo $a
unset a[0]
echo ${a}
echo ${a[@]}
a[5]="hello world"
echo ${a[5]}
echo ${#a[5]}
echo ${#a[@]}
a[4+5/2]="test expression"
declare a["7 + 8"]="test 2"
a[7 + 8]="test 2"
echo ${a[@]}
readonly a[5]
readonly a
# these two lines should output `declare' commands
readonly -a | egrep -v '(BASH_VERSINFO|PIPESTATUS|GROUPS)'
declare -ar | egrep -v '(BASH_VERSINFO|PIPESTATUS|GROUPS)'
# this line should output `readonly' commands, even for arrays
set -o posix
readonly -a | egrep -v '(BASH_VERSINFO|PIPESTATUS|GROUPS)'
set +o posix
declare -a d='([1]="" [2]="bdef" [5]="hello world" "test")'
d[9]="ninth element"
declare -a e[10]=test # this works in post-bash-2.05 versions
declare -a e[10]='(test)'
pass=/etc/passwd
declare -a f='("${d[@]}")'
b=([0]=this [1]=is [2]=a [3]=test [4]="$PS1" [5]=$pass)
echo ${b[@]:2:3}
declare -pa | egrep -v '(BASH_VERSINFO|PIPESTATUS|GROUPS)'
a[3]="this is a test"
b[]=bcde
b[*]=aaa
echo ${b[ ]}
c[-2]=4
echo ${c[-4]}
d[7]=(abdedfegeee)
d=([]=abcde [1]="test test" [*]=last [-65]=negative )
unset d[12]
unset e[*]
declare -a | egrep -v '(BASH_VERSINFO|PIPESTATUS|GROUPS)'
ps1='hello'
unset ps1[2]
unset ${ps1[2]}
declare +a ps1
declare +a c
# the prompt should not print when using a here doc
read -p "array test: " -a rv <<!
this is a test of read using arrays
!
echo ${rv[0]} ${rv[4]}
echo ${rv[@]}
# the variable should be converted to an array when `read -a' is done
vv=1
read -a vv <<!
this is a test of arrays
!
echo ${vv[0]} ${vv[3]}
echo ${vv[@]}
unset vv
declare -a | egrep -v '(BASH_VERSINFO|PIPESTATUS|GROUPS)'
export rv
#set
x[4]=bbb
x=abde
echo $x
echo ${x[0]}
echo ${x[4]}
echo efgh | ( read x[1] ; echo ${x[1]} )
echo wxyz | ( declare -a x ; read x ; echo $x ; echo ${x[0]} )
# Make sure that arrays can be used to save the positional paramters verbatim
set -- a 'b c' d 'e f g' h
ARGV=( [0]=$0 "$@" )
for z in "${ARGV[@]}"
do
echo "$z"
done
echo "$0"
for z in "$@"
do
echo "$z"
done
# do various pattern removal and length tests
XPATH=/bin:/usr/bin:/usr/ucb:/usr/local/bin:.:/sbin:/usr/sbin
xpath=( $( IFS=: ; echo $XPATH ) )
echo ${xpath[@]}
echo ${xpath[@]##*/}
echo ${xpath[0]##*/}
echo ${xpath[@]%%[!/]*}
echo ${xpath[0]%%[!/]*}
recho ${xpath##*/}
recho ${xpath%%[!/]*}
recho ${xpath[5]##*/}
recho ${xpath[5]%%[!/]*}
# let's try to make it a DOS-style path
zecho "${xpath[@]/\//\\}"
zecho "${xpath[@]//\//\\}"
zecho "${xpath[@]//[\/]/\\}"
# length of the first element of the array, since array without subscript
# is equivalent to referencing first element
echo ${#xpath} -- ${#xpath[0]}
# number of elements in the array
nelem=${#xpath[@]}
echo ${#xpath[@]} -- $nelem
# total length of all elements in the array, including space separators
xx="${xpath[*]}"
echo ${#xx}
# total length of all elements in the array
xx=$( IFS='' ; echo "${xpath[*]}" )
echo ${#xx}
unset xpath[nelem-1]
nelem=${#xpath[@]}
echo ${#xpath[@]} -- $nelem
# arrays and things that look like index assignments
array=(42 [1]=14 [2]=44)
array2=(grep [ 123 ] \*)
echo ${array[@]}
echo "${array2[@]}"
# arrays and implicit arithmetic evaluation
declare -i -a iarray
iarray=( 2+4 1+6 7+2 )
echo ${iarray[@]}
iarray[4]=4+1
echo ${iarray[@]}
# make sure assignment using the compound assignment syntax removes all
# of the old elements from the array value
barray=(old1 old2 old3 old4 old5)
barray=(new1 new2 new3)
echo "length = ${#barray[@]}"
echo "value = ${barray[*]}"
# make sure the array code behaves correctly with respect to unset variables
set -u
( echo ${#narray[4]} )
${THIS_SH} ./array1.sub
${THIS_SH} ./array2.sub
# some old bugs and ksh93 compatibility tests
${THIS_SH} ./array3.sub
# some compound assingment parsing problems that showed up in bash-3.1-release
${THIS_SH} ./array4.sub
set +u
cd /tmp
touch 1=bar
foo=([10]="bar")
echo ${foo[0]}
rm 1=bar
cd $OLDPWD
foo=(a b c d e f g)
echo ${foo[@]}
# quoted reserved words are ok
foo=(\for \case \if \then \else)
echo ${foo[@]}
# quoted metacharacters are ok
foo=( [1]='<>' [2]='<' [3]='>' [4]='!' )
echo ${foo[@]}
# numbers are just words when not in a redirection context
foo=( 12 14 16 18 20 )
echo ${foo[@]}
foo=( 4414758999202 )
echo ${foo[@]}
# this was a bug in all versions of bash 2.x up to and including bash-2.04
declare -a ddd=(aaa
bbb)
echo ${ddd[@]}
# errors until post-bash-2.05a; now reserved words are OK
foo=(a b c for case if then else)
foo=(for case if then else)
# errors
metas=( <> < > ! )
metas=( [1]=<> [2]=< [3]=> [4]=! )
# various expansions that didn't really work right until post-bash-2.04
foo='abc'
echo ${foo[0]} ${#foo[0]}
echo ${foo[1]} ${#foo[1]}
echo ${foo[@]} ${#foo[@]}
echo ${foo[*]} ${#foo[*]}
foo=''
echo ${foo[0]} ${#foo[0]}
echo ${foo[1]} ${#foo[1]}
echo ${foo[@]} ${#foo[@]}
echo ${foo[*]} ${#foo[*]}
# new expansions added after bash-2.05b
x[0]=zero
x[1]=one
x[4]=four
x[10]=ten
recho ${!x[@]}
recho "${!x[@]}"
recho ${!x[*]}
recho "${!x[*]}"
# sparse array tests for code fixed in bash-3.0
unset av
av[1]='one'
av[2]=''
av[3]=three
av[5]=five
av[7]=seven
echo include null element -- expect one
echo ${av[@]:1:2} # what happens when we include a null element?
echo include unset element -- expect three five
echo ${av[@]:3:2} # what happens when we include an unset element?
echo start at unset element -- expect five seven
echo ${av[@]:4:2} # what happens when we start at an unset element?
echo too many elements -- expect three five seven
echo ${av[@]:3:5} # how about too many elements?
echo positive offset - expect five seven
echo ${av[@]:5:2}
echo negative offset to unset element - expect seven
echo ${av[@]: -2:2}
echo positive offset 2 - expect seven
echo ${av[@]: 6:2}
echo negative offset 2 - expect seven
echo ${av[@]: -1:2}
echo out-of-range offset
echo ${av[@]:12}
# parsing problems and other inconsistencies not fixed until post bash-3.0
unset x
declare -a x=(')' $$)
[ ${x[1]} -eq $$ ] || echo bad
unset x
declare -a x=(a b c d e)
echo ${x[4]}
z=([1]=one [4]=four [7]=seven [10]=ten)
echo ${#z[@]}
echo ${!z[@]}
unset x
declare -a x=(a \'b c\')
echo "${x[1]}"
unset x
declare -a x=(a 'b c')
echo "${x[1]}"
unset x
declare -a x=($0)
[ "${x[@]}" = $0 ] || echo double expansion of \$0
declare -a x=(\$0)
echo "${x[@]}"
# tests for bash-3.1 problems
${THIS_SH} ./array5.sub
# tests for post-bash-3.2 problems, most fixed in bash-3.2 patches
${THIS_SH} ./array6.sub
${THIS_SH} ./array7.sub
${THIS_SH} ./array8.sub
${THIS_SH} ./array9.sub
${THIS_SH} ./array10.sub
${THIS_SH} ./array11.sub
${THIS_SH} ./array12.sub
${THIS_SH} ./array13.sub
${THIS_SH} ./array14.sub
-22
View File
@@ -1,22 +0,0 @@
# fixes for make_internal_declare not handling integer attribute for arrays
declare -ai -g foo=(1 2 xx 3)
echo "${foo[@]}"
declare -ai -g foo='(1 2 xx 3)'
echo "${foo[@]}"
unset foo
declare -ia -g foo=(1 2 xx 3)
echo "${foo[@]}"
declare -ia -g foo='(1 2 xx 3)'
echo "${foo[@]}"
unset foo
func()
{
declare -ai -g foo=(1 2 xx 3)
}
func
echo "${foo[@]}"
-50
View File
@@ -1,50 +0,0 @@
:; ./shx
sh:
<&$fd ok
nlbq Mon Aug 3 02:45:00 EDT 1992
bang geoff
quote 712824302
setbq defmsgid=<1992Aug3.024502.6176@host>
bgwait sleep done... wait 6187
bash:
<&$fd ok
nlbq Mon Aug 3 02:45:09 EDT 1992
bang geoff
quote 712824311
setbq defmsgid=<1992Aug3.024512.6212@host>
bgwait sleep done... wait 6223
ash:
<&$fd shx1: 4: Syntax error: Bad fd number
nlbq Mon Aug 3 02:45:19 EDT 1992
bang geoff
quote getdate: `"now"' not a valid date
setbq defmsgid=<1992Aug3.` echo 024521
bgwait sleep done... wait 6241
ksh:
<&$fd ok
nlbq ./shx: 6248 Memory fault - core dumped
bang geoff
quote getdate: `"now"' not a valid date
setbq defmsgid=<1992Aug3.024530.6257@host>
bgwait no such job: 6265
wait 6265
sleep done...
zsh:
<&$fd ok
nlbq Mon Aug 3 02:45:36 EDT 1992
bang shx3: event not found: /s/ [4]
quote 712824337
setbq defmsgid=<..6290@host>
bgwait shx7: unmatched " [9]
sleep done...
:;
-10
View File
@@ -1,10 +0,0 @@
#! /bin/sh
for cmd in sh bash ash ksh zsh
do
echo
echo $cmd:
for demo in shx?
do
$cmd $demo
done
done
-96
View File
@@ -1,96 +0,0 @@
# test the trap code
trap 'echo exiting' 0
trap 'echo aborting' 1 2 3 6 15
# make sure a user-specified subshell runs the exit trap, but does not
# inherit the exit trap from a parent shell
( trap 'echo subshell exit' 0; exit 0 )
( exit 0 )
trap
func()
{
trap 'echo ${FUNCNAME:-$0}[$LINENO] funcdebug' DEBUG
echo funcdebug line
}
trap 'echo [$LINENO] debug' DEBUG
echo debug line
trap
func
trap
trap 'echo ${FUNCNAME:-$0}[$LINENO] debug' DEBUG
func2()
{
echo func2debug line
}
declare -ft func2
func2
unset -f func2
trap '' DEBUG
trap
trap - debug
trap
trap - HUP
trap hup
trap '' INT
trap '' int
trap
# exit 0 in exit trap should set exit status
(
set -e
trap 'exit 0' EXIT
false
echo bad
)
echo $?
# hmmm...should this set the handling to SIG_IGN for children, too?
trap '' USR2
./trap1.sub
# test ERR trap
./trap2.sub
${THIS_SH} ./trap3.sub
${THIS_SH} ./trap4.sub
${THIS_SH} ./trap5.sub
#
# show that setting a trap on SIGCHLD is not disastrous.
#
set -o monitor
trap 'echo caught a child death' SIGCHLD
sleep 7 & sleep 6 & sleep 5 &
# this will only catch the first, since there's a trap on SIGCHLD
wait
trap -p SIGCHLD
# Now reset some of the signals the shell handles specially back to
# their default values (with or without the SIG prefix)
trap - SIGINT QUIT TERM
trap
trap - SIGCHLD
wait