diff options
Diffstat (limited to 'shell_cmds/find')
| -rw-r--r-- | shell_cmds/find/extern.h | 124 | ||||
| -rw-r--r-- | shell_cmds/find/find.1 | 1078 | ||||
| -rw-r--r-- | shell_cmds/find/find.c | 323 | ||||
| -rw-r--r-- | shell_cmds/find/find.h | 145 | ||||
| -rw-r--r-- | shell_cmds/find/find.plist.part | 30 | ||||
| -rw-r--r-- | shell_cmds/find/function.c | 1772 | ||||
| -rw-r--r-- | shell_cmds/find/getdate.y | 961 | ||||
| -rw-r--r-- | shell_cmds/find/ls.c | 124 | ||||
| -rw-r--r-- | shell_cmds/find/main.c | 166 | ||||
| -rw-r--r-- | shell_cmds/find/misc.c | 106 | ||||
| -rw-r--r-- | shell_cmds/find/operator.c | 273 | ||||
| -rw-r--r-- | shell_cmds/find/option.c | 201 |
12 files changed, 5303 insertions, 0 deletions
diff --git a/shell_cmds/find/extern.h b/shell_cmds/find/extern.h new file mode 100644 index 0000000..f4282f7 --- /dev/null +++ b/shell_cmds/find/extern.h @@ -0,0 +1,124 @@ +/*- + * Copyright (c) 1991, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)extern.h 8.3 (Berkeley) 4/16/94 + * $FreeBSD: src/usr.bin/find/extern.h,v 1.26 2010/12/11 08:32:16 joel Exp $ + */ + +#include <sys/cdefs.h> + +void brace_subst(char *, char **, char *, int); +PLAN *find_create(char ***); +int find_execute(PLAN *, char **); +PLAN *find_formplan(char **); +PLAN *not_squish(PLAN *); +PLAN *or_squish(PLAN *); +PLAN *paren_squish(PLAN *); +time_t get_date(char *); +struct stat; +void printlong(char *, char *, struct stat *); +int queryuser(char **); +OPTION *lookup_option(const char *); +void finish_execplus(void); + +creat_f c_Xmin; +creat_f c_Xtime; +creat_f c_acl; +creat_f c_and; +creat_f c_delete; +creat_f c_depth; +creat_f c_empty; +creat_f c_exec; +creat_f c_flags; +creat_f c_follow; +creat_f c_fstype; +creat_f c_group; +creat_f c_inum; +creat_f c_links; +creat_f c_ls; +creat_f c_mXXdepth; +creat_f c_name; +creat_f c_newer; +creat_f c_nogroup; +creat_f c_nouser; +creat_f c_perm; +creat_f c_print; +creat_f c_regex; +creat_f c_samefile; +creat_f c_simple; +creat_f c_size; +creat_f c_type; +creat_f c_user; +creat_f c_xdev; + +exec_f f_Xmin; +exec_f f_Xtime; +exec_f f_acl; +exec_f f_always_true; +exec_f f_closeparen; +exec_f f_delete; +exec_f f_depth; +exec_f f_empty; +exec_f f_exec; +exec_f f_expr; +exec_f f_false; +exec_f f_flags; +exec_f f_fstype; +exec_f f_group; +exec_f f_inum; +exec_f f_links; +exec_f f_ls; +exec_f f_name; +exec_f f_newer; +exec_f f_nogroup; +exec_f f_not; +exec_f f_nouser; +exec_f f_openparen; +exec_f f_or; +exec_f f_path; +exec_f f_perm; +exec_f f_print; +exec_f f_print0; +exec_f f_prune; +exec_f f_quit; +exec_f f_regex; +exec_f f_size; +exec_f f_type; +exec_f f_user; +#ifdef __APPLE__ +exec_f f_xattr; +exec_f f_xattrname; +#endif /* __APPLE__ */ + +extern int ftsoptions, isdeprecated, isdepth, isoutput, issort, isxargs; +extern int mindepth, maxdepth; +extern int regexp_flags; +extern time_t now; +extern int dotfd; +extern FTS *tree; +extern int execplus_error; diff --git a/shell_cmds/find/find.1 b/shell_cmds/find/find.1 new file mode 100644 index 0000000..c6bb28f --- /dev/null +++ b/shell_cmds/find/find.1 @@ -0,0 +1,1078 @@ +.\" Copyright (c) 1990, 1993 +.\" The Regents of the University of California. All rights reserved. +.\" +.\" This code is derived from software contributed to Berkeley by +.\" the Institute of Electrical and Electronics Engineers, Inc. +.\" +.\" Redistribution and use in source and binary forms, with or without +.\" modification, are permitted provided that the following conditions +.\" are met: +.\" 1. Redistributions of source code must retain the above copyright +.\" notice, this list of conditions and the following disclaimer. +.\" 2. Redistributions in binary form must reproduce the above copyright +.\" notice, this list of conditions and the following disclaimer in the +.\" documentation and/or other materials provided with the distribution. +.\" 4. Neither the name of the University nor the names of its contributors +.\" may be used to endorse or promote products derived from this software +.\" without specific prior written permission. +.\" +.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND +.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE +.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS +.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) +.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT +.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY +.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF +.\" SUCH DAMAGE. +.\" +.\" @(#)find.1 8.7 (Berkeley) 5/9/95 +.\" $FreeBSD: src/usr.bin/find/find.1,v 1.91 2011/09/28 18:53:36 ed Exp $ +.\" +.Dd September 28, 2011 +.Dt FIND 1 +.Os +.Sh NAME +.Nm find +.Nd walk a file hierarchy +.Sh SYNOPSIS +.Nm +.Op Fl H | Fl L | Fl P +.Op Fl EXdsx +.Op Fl f Ar path +.Ar path ... +.Op Ar expression +.Nm +.Op Fl H | Fl L | Fl P +.Op Fl EXdsx +.Fl f Ar path +.Op Ar path ... +.Op Ar expression +.Sh DESCRIPTION +The +.Nm +utility recursively descends the directory tree for each +.Ar path +listed, evaluating an +.Ar expression +(composed of the +.Dq primaries +and +.Dq operands +listed below) in terms +of each file in the tree. +.Pp +The options are as follows: +.Bl -tag -width indent +.It Fl E +Interpret regular expressions followed by +.Ic -regex +and +.Ic -iregex +primaries as extended (modern) regular expressions rather than basic +regular expressions (BRE's). +The +.Xr re_format 7 +manual page fully describes both formats. +.It Fl H +Cause the file information and file type (see +.Xr stat 2 ) +returned for each symbolic link specified on the command line to be +those of the file referenced by the link, not the link itself. +If the referenced file does not exist, the file information and type will +be for the link itself. +File information of all symbolic links not on +the command line is that of the link itself. +.It Fl L +Cause the file information and file type (see +.Xr stat 2 ) +returned for each symbolic link to be those of the file referenced by the +link, not the link itself. +If the referenced file does not exist, the file information and type will +be for the link itself. +.Pp +This option is equivalent to the deprecated +.Ic -follow +primary. +.It Fl P +Cause the file information and file type (see +.Xr stat 2 ) +returned for each symbolic link to be those of the link itself. +This is the default. +.It Fl X +Permit +.Nm +to be safely used in conjunction with +.Xr xargs 1 . +If a file name contains any of the delimiting characters used by +.Xr xargs 1 , +a diagnostic message is displayed on standard error, and the file +is skipped. +The delimiting characters include single +.Pq Dq Li " ' " +and double +.Pq Dq Li " \*q " +quotes, backslash +.Pq Dq Li \e , +space, tab and newline characters. +.Pp +However, you may wish to consider the +.Fl print0 +primary in conjunction with +.Dq Nm xargs Fl 0 +as an effective alternative. +.It Fl d +Cause +.Nm +to perform a depth-first traversal, i.e., directories +are visited in post-order and all entries in a directory will be acted +on before the directory itself. +By default, +.Nm +visits directories in pre-order, i.e., before their contents. +Note, the default is +.Em not +a breadth-first traversal. +.Pp +This option is equivalent to the +.Ic -depth +primary of +.St -p1003.1-2001 . +The +.Fl d +option +can be useful when +.Nm +is used with +.Xr cpio 1 +to process files that are contained in directories with unusual permissions. +It ensures that you have write permission while you are placing files in a +directory, then sets the directory's permissions as the last thing. +.It Fl f +Specify a file hierarchy for +.Nm +to traverse. +File hierarchies may also be specified as the operands immediately +following the options. +.It Fl s +Cause +.Nm +to traverse the file hierarchies in lexicographical order, +i.e., alphabetical order within each directory. +Note: +.Ql find -s +and +.Ql "find | sort" +may give different results. +.It Fl x +Prevent +.Nm +from descending into directories that have a device number different +than that of the file from which the descent began. +.Pp +This option is equivalent to the deprecated +.Ic -xdev +primary. +.El +.Sh PRIMARIES +.Pp +All primaries which take a numeric argument allow the number to be +preceded by a plus sign +.Pq Dq Li + +or a minus sign +.Pq Dq Li - . +A preceding plus sign means +.Dq more than n , +a preceding minus sign means +.Dq less than n +and neither means +.Dq exactly n . +.Bl -tag -width indent +.It Ic -Bmin Ar n +True if the difference between the time of a file's inode creation +and the time +.Nm +was started, rounded up to the next full minute, is +.Ar n +minutes. +.It Ic -Bnewer Ar file +Same as +.Ic -newerBm . +.It Ic -Btime Ar n Ns Op Cm smhdw +If no units are specified, this primary evaluates to +true if the difference between the time of a file's inode creation +and the time +.Nm +was started, rounded up to the next full 24-hour period, is +.Ar n +24-hour periods. +.Pp +If units are specified, this primary evaluates to +true if the difference between the time of a file's inode creation +and the time +.Nm +was started is exactly +.Ar n +units. +Please refer to the +.Ic -atime +primary description for information on supported time units. +.It Ic -acl +May be used in conjunction with other primaries to locate +files with extended ACLs. +See +.Xr acl 3 +for more information. +.It Ic -amin Ar n +True if the difference between the file last access time and the time +.Nm +was started, rounded up to the next full minute, is +.Ar n +minutes. +.It Ic -anewer Ar file +Same as +.Ic -neweram . +.It Ic -atime Ar n Ns Op Cm smhdw +If no units are specified, this primary evaluates to +true if the difference between the file last access time and the time +.Nm +was started, rounded up to the next full 24-hour period, is +.Ar n +24-hour periods. +.Pp +If units are specified, this primary evaluates to +true if the difference between the file last access time and the time +.Nm +was started is exactly +.Ar n +units. +Possible time units are as follows: +.Pp +.Bl -tag -width indent -compact +.It Cm s +second +.It Cm m +minute (60 seconds) +.It Cm h +hour (60 minutes) +.It Cm d +day (24 hours) +.It Cm w +week (7 days) +.El +.Pp +Any number of units may be combined in one +.Ic -atime +argument, for example, +.Dq Li "-atime -1h30m" . +Units are probably only useful when used in conjunction with the +.Cm + +or +.Cm - +modifier. +.It Ic -cmin Ar n +True if the difference between the time of last change of file status +information and the time +.Nm +was started, rounded up to the next full minute, is +.Ar n +minutes. +.It Ic -cnewer Ar file +Same as +.Ic -newercm . +.It Ic -ctime Ar n Ns Op Cm smhdw +If no units are specified, this primary evaluates to +true if the difference between the time of last change of file status +information and the time +.Nm +was started, rounded up to the next full 24-hour period, is +.Ar n +24-hour periods. +.Pp +If units are specified, this primary evaluates to +true if the difference between the time of last change of file status +information and the time +.Nm +was started is exactly +.Ar n +units. +Please refer to the +.Ic -atime +primary description for information on supported time units. +.It Ic -d +Same as +.Ic depth . +GNU find implements this as a primary in mistaken emulation of +.Fx +.Xr find 1 . +.It Ic -delete +Delete found files and/or directories. +Always returns true. +This executes +from the current working directory as +.Nm +recurses down the tree. +It will not attempt to delete a filename with a +.Dq Pa / +character in its pathname relative to +.Dq Pa \&. +for security reasons. +Depth-first traversal processing is implied by this option. +Following symlinks is incompatible with this option. +.It Ic -depth +Always true; +same as the +.Fl d +option. +.It Ic -depth Ar n +True if the depth of the file relative to the starting point of the traversal +is +.Ar n . +.It Ic -empty +True if the current file or directory is empty. +.It Ic -exec Ar utility Oo Ar argument ... Oc Li \&; +True if the program named +.Ar utility +returns a zero value as its exit status. +Optional +.Ar arguments +may be passed to the utility. +The expression must be terminated by a semicolon +.Pq Dq Li \&; . +If you invoke +.Nm +from a shell you may need to quote the semicolon if the shell would +otherwise treat it as a control operator. +If the string +.Dq Li {} +appears anywhere in the utility name or the +arguments it is replaced by the pathname of the current file. +.Ar Utility +will be executed from the directory from which +.Nm +was executed. +.Ar Utility +and +.Ar arguments +are not subject to the further expansion of shell patterns +and constructs. +.It Ic -exec Ar utility Oo Ar argument ... Oc Li {} + +Same as +.Ic -exec , +except that +.Dq Li {} +is replaced with as many pathnames as possible for each invocation of +.Ar utility . +This behaviour is similar to that of +.Xr xargs 1 . +.It Ic -execdir Ar utility Oo Ar argument ... Oc Li \&; +The +.Ic -execdir +primary is identical to the +.Ic -exec +primary with the exception that +.Ar utility +will be executed from the directory that holds +the current file. +The filename substituted for +the string +.Dq Li {} +is not qualified. +.It Ic -execdir Ar utility Oo Ar argument ... Oc Li {} + +Same as +.Ic -execdir , +except that +.Dq Li {} +is replaced with as many pathnames as possible for each invocation of +.Ar utility . +This behaviour is similar to that of +.Xr xargs 1 . +.It Ic -flags Oo Cm - Ns | Ns Cm + Oc Ns Ar flags , Ns Ar notflags +The flags are specified using symbolic names (see +.Xr chflags 1 ) . +Those with the +.Qq Li no +prefix (except +.Qq Li nodump ) +are said to be +.Ar notflags . +Flags in +.Ar flags +are checked to be set, and flags in +.Ar notflags +are checked to be not set. +Note that this is different from +.Ic -perm , +which only allows the user to specify mode bits that are set. +.Pp +If flags are preceded by a dash +.Pq Dq Li - , +this primary evaluates to true +if at least all of the bits in +.Ar flags +and none of the bits in +.Ar notflags +are set in the file's flags bits. +If flags are preceded by a plus +.Pq Dq Li + , +this primary evaluates to true +if any of the bits in +.Ar flags +is set in the file's flags bits, +or any of the bits in +.Ar notflags +is not set in the file's flags bits. +Otherwise, +this primary evaluates to true +if the bits in +.Ar flags +exactly match the file's flags bits, +and none of the +.Ar flags +bits match those of +.Ar notflags . +.It Ic -fstype Ar type +True if the file is contained in a file system of type +.Ar type . +The +.Xr lsvfs 1 +command can be used to find out the types of file systems +that are available on the system. +In addition, there are two pseudo-types, +.Dq Li local +and +.Dq Li rdonly . +The former matches any file system physically mounted on the system where +the +.Nm +is being executed and the latter matches any file system which is +mounted read-only. +.It Ic -gid Ar gname +The same thing as +.Ar -group Ar gname +for compatibility with GNU find. +GNU find imposes a restriction that +.Ar gname +is numeric, while +.Xr find 1 +does not. +.It Ic -group Ar gname +True if the file belongs to the group +.Ar gname . +If +.Ar gname +is numeric and there is no such group name, then +.Ar gname +is treated as a group ID. +.It Ic -ignore_readdir_race +This option is for GNU find compatibility and is ignored. +.It Ic -ilname Ar pattern +Like +.Ic -lname , +but the match is case insensitive. +This is a GNU find extension. +.It Ic -iname Ar pattern +Like +.Ic -name , +but the match is case insensitive. +.It Ic -inum Ar n +True if the file has inode number +.Ar n . +.It Ic -ipath Ar pattern +Like +.Ic -path , +but the match is case insensitive. +.It Ic -iregex Ar pattern +Like +.Ic -regex , +but the match is case insensitive. +.It Ic -iwholename Ar pattern +The same thing as +.Ic -ipath , +for GNU find compatibility. +.It Ic -links Ar n +True if the file has +.Ar n +links. +.It Ic -lname Ar pattern +Like +.Ic -name , +but the contents of the symbolic link are matched instead of the file +name. +Note that this only matches broken symbolic links +if symbolic links are being followed. +This is a GNU find extension. +.It Ic -ls +This primary always evaluates to true. +The following information for the current file is written to standard output: +its inode number, size in 512-byte blocks, file permissions, number of hard +links, owner, group, size in bytes, last modification time, and pathname. +If the file is a block or character special file, the device number +will be displayed instead of the size in bytes. +If the file is a symbolic link, the pathname of the linked-to file will be +displayed preceded by +.Dq Li -> . +The format is identical to that produced by +.Bk -words +.Dq Nm ls Fl dgils . +.Ek +.It Ic -maxdepth Ar n +Always true; descend at most +.Ar n +directory levels below the command line arguments. +If any +.Ic -maxdepth +primary is specified, it applies to the entire expression even if it would +not normally be evaluated. +.Dq Ic -maxdepth Li 0 +limits the whole search to the command line arguments. +.It Ic -mindepth Ar n +Always true; do not apply any tests or actions at levels less than +.Ar n . +If any +.Ic -mindepth +primary is specified, it applies to the entire expression even if it would +not normally be evaluated. +.Dq Ic -mindepth Li 1 +processes all but the command line arguments. +.It Ic -mmin Ar n +True if the difference between the file last modification time and the time +.Nm +was started, rounded up to the next full minute, is +.Ar n +minutes. +.It Ic -mnewer Ar file +Same as +.Ic -newer . +.It Ic -mount +The same thing as +.Ic -xdev , +for GNU find compatibility. +.It Ic -mtime Ar n Ns Op Cm smhdw +If no units are specified, this primary evaluates to +true if the difference between the file last modification time and the time +.Nm +was started, rounded up to the next full 24-hour period, is +.Ar n +24-hour periods. +.Pp +If units are specified, this primary evaluates to +true if the difference between the file last modification time and the time +.Nm +was started is exactly +.Ar n +units. +Please refer to the +.Ic -atime +primary description for information on supported time units. +.It Ic -name Ar pattern +True if the last component of the pathname being examined matches +.Ar pattern . +Special shell pattern matching characters +.Dq ( Li \&[ , +.Dq Li \&] , +.Dq Li * , +and +.Dq Li \&? ) +may be used as part of +.Ar pattern . +These characters may be matched explicitly by escaping them with a +backslash +.Pq Dq Li \e . +.It Ic -newer Ar file +True if the current file has a more recent last modification time than +.Ar file . +.It Ic -newer Ns Ar X Ns Ar Y Ar file +True if the current file has a more recent last access time +.Pq Ar X Ns = Ns Cm a , +inode creation time +.Pq Ar X Ns = Ns Cm B , +change time +.Pq Ar X Ns = Ns Cm c , +or modification time +.Pq Ar X Ns = Ns Cm m +than the last access time +.Pq Ar Y Ns = Ns Cm a , +inode creation time +.Pq Ar Y Ns = Ns Cm B , +change time +.Pq Ar Y Ns = Ns Cm c , +or modification time +.Pq Ar Y Ns = Ns Cm m +of +.Ar file . +In addition, if +.Ar Y Ns = Ns Cm t , +then +.Ar file +is instead interpreted as a direct date specification of the form +understood by +.Xr cvs 1 . +Note that +.Ic -newermm +is equivalent to +.Ic -newer . +.It Ic -nogroup +True if the file belongs to an unknown group. +.It Ic -noignore_readdir_race +This option is for GNU find compatibility and is ignored. +.It Ic -noleaf +This option is for GNU find compatibility. +In GNU find it disables an optimization not relevant to +.Xr find 1 , +so it is ignored. +.It Ic -nouser +True if the file belongs to an unknown user. +.It Ic -ok Ar utility Oo Ar argument ... Oc Li \&; +The +.Ic -ok +primary is identical to the +.Ic -exec +primary with the exception that +.Nm +requests user affirmation for the execution of the +.Ar utility +by printing +a message to the terminal and reading a response. +If the response is not affirmative +.Ql ( y +in the +.Dq Li POSIX +locale), +the command is not executed and the +value of the +.Ic -ok +expression is false. +.It Ic -okdir Ar utility Oo Ar argument ... Oc Li \&; +The +.Ic -okdir +primary is identical to the +.Ic -execdir +primary with the same exception as described for the +.Ic -ok +primary. +.It Ic -path Ar pattern +True if the pathname being examined matches +.Ar pattern . +Special shell pattern matching characters +.Dq ( Li \&[ , +.Dq Li \&] , +.Dq Li * , +and +.Dq Li \&? ) +may be used as part of +.Ar pattern . +These characters may be matched explicitly by escaping them with a +backslash +.Pq Dq Li \e . +Slashes +.Pq Dq Li / +are treated as normal characters and do not have to be +matched explicitly. +.It Ic -perm Oo Cm - Ns | Ns Cm + Oc Ns Ar mode +The +.Ar mode +may be either symbolic (see +.Xr chmod 1 ) +or an octal number. +If the +.Ar mode +is symbolic, a starting value of zero is assumed and the +.Ar mode +sets or clears permissions without regard to the process' file mode +creation mask. +If the +.Ar mode +is octal, only bits 07777 +.Pq Dv S_ISUID | S_ISGID | S_ISTXT | S_IRWXU | S_IRWXG | S_IRWXO +of the file's mode bits participate +in the comparison. +If the +.Ar mode +is preceded by a dash +.Pq Dq Li - , +this primary evaluates to true +if at least all of the bits in the +.Ar mode +are set in the file's mode bits. +If the +.Ar mode +is preceded by a plus +.Pq Dq Li + , +this primary evaluates to true +if any of the bits in the +.Ar mode +are set in the file's mode bits. +Otherwise, this primary evaluates to true if +the bits in the +.Ar mode +exactly match the file's mode bits. +Note, the first character of a symbolic mode may not be a dash +.Pq Dq Li - . +.It Ic -print +This primary always evaluates to true. +It prints the pathname of the current file to standard output. +If none of +.\" 4772561 +.Ic -exec , -ls , -print , -print0 , +or +.Ic -ok +is specified, the given expression shall be effectively replaced by +.Cm \&( Ar "given expression" Cm \&) Ic -print . +.It Ic -print0 +This primary always evaluates to true. +It prints the pathname of the current file to standard output, followed by an +.Tn ASCII +.Dv NUL +character (character code 0). +.It Ic -prune +This primary always evaluates to true. +It causes +.Nm +to not descend into the current file. +Note, the +.Ic -prune +primary has no effect if the +.Fl d +option was specified. +.It Ic -regex Ar pattern +True if the whole path of the file matches +.Ar pattern +using regular expression. +To match a file named +.Dq Pa ./foo/xyzzy , +you can use the regular expression +.Dq Li ".*/[xyz]*" +or +.Dq Li ".*/foo/.*" , +but not +.Dq Li xyzzy +or +.Dq Li /foo/ . +.It Ic -samefile Ar name +True if the file is a hard link to +.Ar name . +If the command option +.Ic -L +is specified, it is also true if the file is a symbolic link and +points to +.Ar name . +.It Ic -size Ar n Ns Op Cm ckMGTP +True if the file's size, rounded up, in 512-byte blocks is +.Ar n . +If +.Ar n +is followed by a +.Cm c , +then the primary is true if the +file's size is +.Ar n +bytes (characters). +Similarly if +.Ar n +is followed by a scale indicator then the file's size is compared to +.Ar n +scaled as: +.Pp +.Bl -tag -width indent -compact +.It Cm k +kilobytes (1024 bytes) +.It Cm M +megabytes (1024 kilobytes) +.It Cm G +gigabytes (1024 megabytes) +.It Cm T +terabytes (1024 gigabytes) +.It Cm P +petabytes (1024 terabytes) +.El +.It Ic -type Ar t +True if the file is of the specified type. +Possible file types are as follows: +.Pp +.Bl -tag -width indent -compact +.It Cm b +block special +.It Cm c +character special +.It Cm d +directory +.It Cm f +regular file +.It Cm l +symbolic link +.It Cm p +FIFO +.It Cm s +socket +.El +.It Ic -uid Ar uname +The same thing as +.Ar -user Ar uname +for compatibility with GNU find. +GNU find imposes a restriction that +.Ar uname +is numeric, while +.Xr find 1 +does not. +.It Ic -user Ar uname +True if the file belongs to the user +.Ar uname . +If +.Ar uname +is numeric and there is no such user name, then +.Ar uname +is treated as a user ID. +.It Ic -wholename Ar pattern +The same thing as +.Ic -path , +for GNU find compatibility. +.It Ic -xattr +True if the file has any extended attributes. +.It Ic -xattrname Ar name +True if the file has an extended attribute with the specified +.Ar name . +.El +.Sh OPERATORS +The primaries may be combined using the following operators. +The operators are listed in order of decreasing precedence. +.Pp +.Bl -tag -width indent -compact +.It Cm \&( Ar expression Cm \&) +This evaluates to true if the parenthesized expression evaluates to +true. +.Pp +.It Cm \&! Ar expression +.It Cm -not Ar expression +This is the unary +.Tn NOT +operator. +It evaluates to true if the expression is false. +.Pp +.It Cm -false +Always false. +.It Cm -true +Always true. +.Pp +.It Ar expression Cm -and Ar expression +.It Ar expression expression +The +.Cm -and +operator is the logical +.Tn AND +operator. +As it is implied by the juxtaposition of two expressions it does not +have to be specified. +The expression evaluates to true if both expressions are true. +The second expression is not evaluated if the first expression is false. +.Pp +.It Ar expression Cm -or Ar expression +The +.Cm -or +operator is the logical +.Tn OR +operator. +The expression evaluates to true if either the first or the second expression +is true. +The second expression is not evaluated if the first expression is true. +.El +.Pp +All operands and primaries must be separate arguments to +.Nm . +Primaries which themselves take arguments expect each argument +to be a separate argument to +.Nm . +.Sh ENVIRONMENT +The +.Ev LANG , LC_ALL , LC_COLLATE , LC_CTYPE , LC_MESSAGES +and +.Ev LC_TIME +environment variables affect the execution of the +.Nm +utility as described in +.Xr environ 7 . +.Sh EXAMPLES +The following examples are shown as given to the shell: +.Bl -tag -width indent +.It Li "find / \e! -name \*q*.c\*q -print" +Print out a list of all the files whose names do not end in +.Pa .c . +.It Li "find / -newer ttt -user wnj -print" +Print out a list of all the files owned by user +.Dq wnj +that are newer +than the file +.Pa ttt . +.It Li "find / \e! \e( -newer ttt -user wnj \e) -print" +Print out a list of all the files which are not both newer than +.Pa ttt +and owned by +.Dq wnj . +.It Li "find / \e( -newer ttt -or -user wnj \e) -print" +Print out a list of all the files that are either owned by +.Dq wnj +or that are newer than +.Pa ttt . +.It Li "find / -newerct '1 minute ago' -print" +Print out a list of all the files whose inode change time is more +recent than the current time minus one minute. +.It Li "find / -type f -exec echo {} \e;" +Use the +.Xr echo 1 +command to print out a list of all the files. +.It Li "find -L /usr/ports/packages -type l -exec rm -- {} +" +Delete all broken symbolic links in +.Pa /usr/ports/packages . +.It Li "find /usr/src -name CVS -prune -o -depth +6 -print" +Find files and directories that are at least seven levels deep +in the working directory +.Pa /usr/src . +.It Li "find /usr/src -name CVS -prune -o -mindepth 7 -print" +Is not equivalent to the previous example, since +.Ic -prune +is not evaluated below level seven. +.El +.Sh COMPATIBILITY +The +.Ic -follow +primary is deprecated; the +.Fl L +option should be used instead. +See the +.Sx STANDARDS +section below for details. +.Sh SEE ALSO +.Xr chflags 1 , +.Xr chmod 1 , +.Xr cvs 1 , +.Xr locate 1 , +.Xr lsvfs 1 , +.Xr whereis 1 , +.Xr which 1 , +.Xr xargs 1 , +.Xr stat 2 , +.Xr acl 3 , +.Xr fts 3 , +.Xr getgrent 3 , +.Xr getpwent 3 , +.Xr strmode 3 , +.Xr re_format 7 , +.Xr symlink 7 +.Sh STANDARDS +The +.Nm +utility syntax is a superset of the syntax specified by the +.St -p1003.1-2001 +standard. +.Pp +All the single character options except +.Fl H +and +.Fl L +as well as +.Ic -amin , -anewer , -cmin , -cnewer , -delete , -empty , -fstype , +.Ic -iname , -inum , -iregex , -ls , -maxdepth , -mindepth , -mmin , +.Ic -path , -print0 , -regex +and all of the +.Ic -B* +birthtime related primaries are extensions to +.St -p1003.1-2001 . +.Pp +Historically, the +.Fl d , L +and +.Fl x +options were implemented using the primaries +.Ic -depth , -follow , +and +.Ic -xdev . +These primaries always evaluated to true. +As they were really global variables that took effect before the traversal +began, some legal expressions could have unexpected results. +An example is the expression +.Ic -print Cm -o Ic -depth . +As +.Ic -print +always evaluates to true, the standard order of evaluation +implies that +.Ic -depth +would never be evaluated. +This is not the case. +.Pp +The operator +.Cm -or +was implemented as +.Cm -o , +and the operator +.Cm -and +was implemented as +.Cm -a . +.Pp +Historic implementations of the +.Ic -exec +and +.Ic -ok +primaries did not replace the string +.Dq Li {} +in the utility name or the +utility arguments if it had preceding or following non-whitespace characters. +This version replaces it no matter where in the utility name or arguments +it appears. +.Pp +The +.Fl E +option was inspired by the equivalent +.Xr grep 1 +and +.Xr sed 1 +options. +.Sh HISTORY +A +.Nm +command appeared in +.At v1 . +.Sh BUGS +The special characters used by +.Nm +are also special characters to many shell programs. +In particular, the characters +.Dq Li * , +.Dq Li \&[ , +.Dq Li \&] , +.Dq Li \&? , +.Dq Li \&( , +.Dq Li \&) , +.Dq Li \&! , +.Dq Li \e +and +.Dq Li \&; +may have to be escaped from the shell. +.Pp +As there is no delimiter separating options and file names or file +names and the +.Ar expression , +it is difficult to specify files named +.Pa -xdev +or +.Pa \&! . +These problems are handled by the +.Fl f +option and the +.Xr getopt 3 +.Dq Fl Fl +construct. +.Pp +The +.Ic -delete +primary does not interact well with other options that cause the file system +tree traversal options to be changed. +.Pp +The +.Ic -mindepth +and +.Ic -maxdepth +primaries are actually global options (as documented above). +They should +probably be replaced by options which look like options. diff --git a/shell_cmds/find/find.c b/shell_cmds/find/find.c new file mode 100644 index 0000000..c59a75a --- /dev/null +++ b/shell_cmds/find/find.c @@ -0,0 +1,323 @@ +/*- + * Copyright (c) 1991, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Cimarron D. Taylor of the University of California, Berkeley. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#ifndef lint +#if 0 +static char sccsid[] = "@(#)find.c 8.5 (Berkeley) 8/5/94"; +#else +#endif +#endif /* not lint */ + +#include <sys/cdefs.h> +__FBSDID("$FreeBSD: src/usr.bin/find/find.c,v 1.23 2010/12/11 08:32:16 joel Exp $"); + +#include <sys/types.h> +#include <sys/stat.h> + +#include <err.h> +#include <errno.h> +#include <fts.h> +#include <regex.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#ifdef __APPLE__ +#include <get_compat.h> +#include <unistd.h> +#else +#define COMPAT_MODE(func, mode) 1 +#endif + +#include "find.h" + +#ifdef __APPLE__ +static int find_compare(const FTSENT **s1, const FTSENT **s2); +#else /* !__APPLE__ */ +static int find_compare(const FTSENT * const *s1, const FTSENT * const *s2); +#endif /* __APPLE__ */ + +/* + * find_compare -- + * tell fts_open() how to order the traversal of the hierarchy. + * This variant gives lexicographical order, i.e., alphabetical + * order within each directory. + */ +static int +#ifdef __APPLE__ +find_compare(const FTSENT **s1, const FTSENT **s2) +#else /* !__APPLE__ */ +find_compare(const FTSENT * const *s1, const FTSENT * const *s2) +#endif /* __APPLE__ */ +{ + + return (strcoll((*s1)->fts_name, (*s2)->fts_name)); +} + +/* + * find_formplan -- + * process the command line and create a "plan" corresponding to the + * command arguments. + */ +PLAN * +find_formplan(char *argv[]) +{ + PLAN *plan, *tail, *new; + + /* + * for each argument in the command line, determine what kind of node + * it is, create the appropriate node type and add the new plan node + * to the end of the existing plan. The resulting plan is a linked + * list of plan nodes. For example, the string: + * + * % find . -name foo -newer bar -print + * + * results in the plan: + * + * [-name foo]--> [-newer bar]--> [-print] + * + * in this diagram, `[-name foo]' represents the plan node generated + * by c_name() with an argument of foo and `-->' represents the + * plan->next pointer. + */ + for (plan = tail = NULL; *argv;) { + if (!(new = find_create(&argv))) + continue; + if (plan == NULL) + tail = plan = new; + else { + tail->next = new; + tail = new; + } + } + + /* + * if the user didn't specify one of -print, -ok or -exec, then -print + * is assumed so we bracket the current expression with parens, if + * necessary, and add a -print node on the end. + */ + if (!isoutput) { + OPTION *p; + char **argv1 = 0; + + if (plan == NULL) { + p = lookup_option("-print"); + new = (p->create)(p, &argv1); + tail = plan = new; + } else { + p = lookup_option("("); + new = (p->create)(p, &argv1); + new->next = plan; + plan = new; + p = lookup_option(")"); + new = (p->create)(p, &argv1); + tail->next = new; + tail = new; + p = lookup_option("-print"); + new = (p->create)(p, &argv1); + tail->next = new; + tail = new; + } + } + + /* + * the command line has been completely processed into a search plan + * except for the (, ), !, and -o operators. Rearrange the plan so + * that the portions of the plan which are affected by the operators + * are moved into operator nodes themselves. For example: + * + * [!]--> [-name foo]--> [-print] + * + * becomes + * + * [! [-name foo] ]--> [-print] + * + * and + * + * [(]--> [-depth]--> [-name foo]--> [)]--> [-print] + * + * becomes + * + * [expr [-depth]-->[-name foo] ]--> [-print] + * + * operators are handled in order of precedence. + */ + + plan = paren_squish(plan); /* ()'s */ + plan = not_squish(plan); /* !'s */ + plan = or_squish(plan); /* -o's */ + return (plan); +} + +/* addPath - helper function used to build a list of paths that were + * specified on the command line that we are allowed to search. + */ +static char **addPath(char **array, char *newPath) +{ + static int pathCounter = 0; + + if (newPath == NULL) { /* initialize array */ + if ((array = malloc(sizeof(char *))) == NULL) + err(2, "out of memory"); + array[0] = NULL; + } + else { + array = realloc(array, (++pathCounter + 1) * sizeof(char *)); + if (array == NULL) + err(2, "out of memory"); + else { + array[pathCounter - 1] = newPath; + array[pathCounter] = NULL; /* ensure array is null terminated */ + } + } + return (array); +} + +FTS *tree; /* pointer to top of FTS hierarchy */ + +/* + * find_execute -- + * take a search plan and an array of search paths and executes the plan + * over all FTSENT's returned for the given search paths. + */ +int +find_execute(PLAN *plan, char *paths[]) +{ + FTSENT *entry; + PLAN *p; + int rval; + char **myPaths; + int nonSearchableDirFound = 0; + int pathIndex; + struct stat statInfo; + + /* special-case directories specified on command line - explicitly examine + * mode bits, to ensure failure if the directory cannot be searched + * (whether or not it's empty). UNIX conformance... <sigh> + */ + + int strict_symlinks = (ftsoptions & (FTS_COMFOLLOW|FTS_LOGICAL)) + && COMPAT_MODE("bin/find", "unix2003"); + + myPaths = addPath(NULL, NULL); + for (pathIndex = 0; paths[pathIndex] != NULL; ++pathIndex) { + int stat_ret = stat(paths[pathIndex], &statInfo); + int stat_errno = errno; + if (strict_symlinks && stat_ret < 0) { + if (stat_errno == ELOOP) { + errx(1, "Symlink loop resolving %s", paths[pathIndex]); + } + } + + /* retrieve mode bits, and examine "searchable" bit of + directories, exempt root from POSIX conformance */ + if (COMPAT_MODE("bin/find", "unix2003") && getuid() + && stat_ret == 0 + && ((statInfo.st_mode & S_IFMT) == S_IFDIR)) { + if (access(paths[pathIndex], X_OK) == 0) { + myPaths = addPath(myPaths, paths[pathIndex]); + } else { + if (stat_errno != ENAMETOOLONG) { /* if name is too long, just let existing logic handle it */ + warnx("%s: Permission denied", paths[pathIndex]); + nonSearchableDirFound = 1; + } + } + } else { + /* not a directory, so add path to array */ + myPaths = addPath(myPaths, paths[pathIndex]); + } + } + if (myPaths[0] == NULL) { /* were any directories searchable? */ + free(myPaths); + return(nonSearchableDirFound); /* no... */ + } + + tree = fts_open(myPaths, ftsoptions, (issort ? find_compare : NULL)); + if (tree == NULL) + err(1, "ftsopen"); + + for (rval = nonSearchableDirFound; (entry = fts_read(tree)) != NULL;) { + if (maxdepth != -1 && entry->fts_level >= maxdepth) { + if (fts_set(tree, entry, FTS_SKIP)) + err(1, "%s", entry->fts_path); + } + + switch (entry->fts_info) { + case FTS_D: + if (isdepth) + continue; + break; + case FTS_DP: + if (!isdepth) + continue; + break; + case FTS_DNR: + case FTS_ERR: + case FTS_NS: + (void)fflush(stdout); + warnx("%s: %s", + entry->fts_path, strerror(entry->fts_errno)); + rval = 1; + continue; +#ifdef FTS_W + case FTS_W: + continue; +#endif /* FTS_W */ + } +#define BADCH " \t\n\\'\"" + if (isxargs && strpbrk(entry->fts_path, BADCH)) { + (void)fflush(stdout); + warnx("%s: illegal path", entry->fts_path); + rval = 1; + continue; + } + + if (mindepth != -1 && entry->fts_level < mindepth) + continue; + + /* + * Call all the functions in the execution plan until one is + * false or all have been executed. This is where we do all + * the work specified by the user on the command line. + */ + for (p = plan; p && (p->execute)(p, entry); p = p->next); + } + free (myPaths); + finish_execplus(); + if (execplus_error) { + exit(execplus_error); + } + if (errno) + err(1, "fts_read"); + fts_close(tree); + return (rval); +} diff --git a/shell_cmds/find/find.h b/shell_cmds/find/find.h new file mode 100644 index 0000000..8960a0b --- /dev/null +++ b/shell_cmds/find/find.h @@ -0,0 +1,145 @@ +/*- + * Copyright (c) 1990, 1993 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Cimarron D. Taylor of the University of California, Berkeley. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * @(#)find.h 8.1 (Berkeley) 6/6/93 + * $FreeBSD: src/usr.bin/find/find.h,v 1.21 2010/12/11 08:32:16 joel Exp $ + */ + +#include <regex.h> + +/* forward declarations */ +struct _plandata; +struct _option; + +/* execute function */ +typedef int exec_f(struct _plandata *, FTSENT *); +/* create function */ +typedef struct _plandata *creat_f(struct _option *, char ***); + +/* function modifiers */ +#define F_NEEDOK 0x00000001 /* -ok vs. -exec */ +#define F_EXECDIR 0x00000002 /* -execdir vs. -exec */ +#define F_TIME_A 0x00000004 /* one of -atime, -anewer, -newera* */ +#define F_TIME_C 0x00000008 /* one of -ctime, -cnewer, -newerc* */ +#define F_TIME2_A 0x00000010 /* one of -newer?a */ +#define F_TIME2_C 0x00000020 /* one of -newer?c */ +#define F_TIME2_T 0x00000040 /* one of -newer?t */ +#define F_MAXDEPTH F_TIME_A /* maxdepth vs. mindepth */ +#define F_DEPTH F_TIME_A /* -depth n vs. -d */ +/* command line function modifiers */ +#define F_EQUAL 0x00000000 /* [acm]min [acm]time inum links size */ +#define F_LESSTHAN 0x00000100 +#define F_GREATER 0x00000200 +#define F_ELG_MASK 0x00000300 +#define F_ATLEAST 0x00000400 /* flags perm */ +#define F_ANY 0x00000800 /* perm */ +#define F_MTMASK 0x00003000 +#define F_MTFLAG 0x00000000 /* fstype */ +#define F_MTTYPE 0x00001000 +#define F_MTUNKNOWN 0x00002000 +#define F_IGNCASE 0x00010000 /* iname ipath iregex */ +#define F_EXACTTIME F_IGNCASE /* -[acm]time units syntax */ +#define F_EXECPLUS 0x00020000 /* -exec ... {} + */ +#define F_TIME_B 0x00040000 /* one of -Btime, -Bnewer, -newerB* */ +#define F_TIME2_B 0x00080000 /* one of -newer?B */ +#define F_LINK 0x00100000 /* lname or ilname */ + +/* node definition */ +typedef struct _plandata { + struct _plandata *next; /* next node */ + exec_f *execute; /* node evaluation function */ + int flags; /* private flags */ + union { + gid_t _g_data; /* gid */ + ino_t _i_data; /* inode */ + mode_t _m_data; /* mode mask */ + struct { + u_long _f_flags; + u_long _f_notflags; + } fl; + nlink_t _l_data; /* link count */ + short _d_data; /* level depth (-1 to N) */ + off_t _o_data; /* file size */ + time_t _t_data; /* time value */ + uid_t _u_data; /* uid */ + short _mt_data; /* mount flags */ + struct _plandata *_p_data[2]; /* PLAN trees */ + struct _ex { + char **_e_argv; /* argv array */ + char **_e_orig; /* original strings */ + int *_e_len; /* allocated length */ + int _e_pbnum; /* base num. of args. used */ + int _e_ppos; /* number of arguments used */ + int _e_pnummax; /* max. number of arguments */ + int _e_psize; /* number of bytes of args. */ + int _e_pbsize; /* base num. of bytes of args */ + int _e_psizemax; /* max num. of bytes of args */ + struct _plandata *_e_next;/* next F_EXECPLUS in tree */ + } ex; + char *_a_data[2]; /* array of char pointers */ + char *_c_data; /* char pointer */ + regex_t *_re_data; /* regex */ + } p_un; +} PLAN; +#define a_data p_un._a_data +#define c_data p_un._c_data +#define d_data p_un._d_data +#define fl_flags p_un.fl._f_flags +#define fl_notflags p_un.fl._f_notflags +#define g_data p_un._g_data +#define i_data p_un._i_data +#define l_data p_un._l_data +#define m_data p_un._m_data +#define mt_data p_un._mt_data +#define o_data p_un._o_data +#define p_data p_un._p_data +#define t_data p_un._t_data +#define u_data p_un._u_data +#define re_data p_un._re_data +#define e_argv p_un.ex._e_argv +#define e_orig p_un.ex._e_orig +#define e_len p_un.ex._e_len +#define e_pbnum p_un.ex._e_pbnum +#define e_ppos p_un.ex._e_ppos +#define e_pnummax p_un.ex._e_pnummax +#define e_psize p_un.ex._e_psize +#define e_pbsize p_un.ex._e_pbsize +#define e_psizemax p_un.ex._e_psizemax +#define e_next p_un.ex._e_next + +typedef struct _option { + const char *name; /* option name */ + creat_f *create; /* create function */ + exec_f *execute; /* execute function */ + int flags; +} OPTION; + +#include "extern.h" diff --git a/shell_cmds/find/find.plist.part b/shell_cmds/find/find.plist.part new file mode 100644 index 0000000..0148f95 --- /dev/null +++ b/shell_cmds/find/find.plist.part @@ -0,0 +1,30 @@ + <dict> + <key>OpenSourceProject</key> + <string>find</string> + <key>OpenSourceVersion</key> + <string>2011-12-10</string> + <key>OpenSourceWebsiteURL</key> + <string>http://svnweb.freebsd.org/base/head/usr.bin/find/</string> + <key>OpenSourceSCM</key> + <string>svn co http://svn.freebsd.org/base/head/usr.bin/find/</string> + <key>OpenSourceImportDate</key> + <string>2012-01-06</string> + <key>OpenSourceModifications</key> + <array> + <string>Add -xattr and -xattrname options.</string> + <string>execplus conformance fixes.</string> + <string>Man page fix (4772561).</string> + <string>Adjust FTS usage due to API differences.</string> + <string>Avoid searching unsearchable directories (3849245).</string> + <string>Add missing call to fts_close (4608460).</string> + <string>Conformance fixes related to F_EXACTTIME.</string> + <string>Apple-specific implementation of -acl primary.</string> + <string>Conformance fix for f_name (6591280).</string> + <string>setmode conformance fix (3936046).</string> + <string>Use a saner MAXLOGNAME definition.</string> + <string>Workaround for lack of rpmatch function.</string> + <string>FreeBSD revisions 260336, 260355.</string> + </array> + <key>OpenSourceLicense</key> + <string>bsd</string> + </dict> diff --git a/shell_cmds/find/function.c b/shell_cmds/find/function.c new file mode 100644 index 0000000..a673a02 --- /dev/null +++ b/shell_cmds/find/function.c @@ -0,0 +1,1772 @@ +/*- + * Copyright (c) 1990, 1993 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Cimarron D. Taylor of the University of California, Berkeley. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#ifndef lint +#if 0 +static const char sccsid[] = "@(#)function.c 8.10 (Berkeley) 5/4/95"; +#endif +#endif /* not lint */ + +#include <sys/cdefs.h> +__FBSDID("$FreeBSD: src/usr.bin/find/function.c,v 1.71 2011/06/13 05:22:07 avatar Exp $"); + +#include <sys/param.h> +#include <sys/ucred.h> +#include <sys/stat.h> +#include <sys/types.h> +#include <sys/acl.h> +#include <sys/wait.h> +#include <sys/mount.h> + +#include <dirent.h> +#include <err.h> +#include <errno.h> +#include <fnmatch.h> +#include <fts.h> +#include <grp.h> +#include <limits.h> +#include <pwd.h> +#include <regex.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> +#include <ctype.h> + +#ifdef __APPLE__ +#include <sys/sysctl.h> +#include <sys/xattr.h> +#include <libgen.h> +#include <get_compat.h> +#else +#define COMPAT_MODE(func, mode) 1 +#endif + +#include "find.h" + +static PLAN *palloc(OPTION *); +static long long find_parsenum(PLAN *, const char *, char *, char *); +static long long find_parsetime(PLAN *, const char *, char *); +static char *nextarg(OPTION *, char ***); + +extern char **environ; + +static PLAN *lastexecplus = NULL; +int execplus_error; + +#define COMPARE(a, b) do { \ + switch (plan->flags & F_ELG_MASK) { \ + case F_EQUAL: \ + return (a == b); \ + case F_LESSTHAN: \ + return (a < b); \ + case F_GREATER: \ + return (a > b); \ + default: \ + abort(); \ + } \ +} while(0) + +static PLAN * +palloc(OPTION *option) +{ + PLAN *new; + + if ((new = malloc(sizeof(PLAN))) == NULL) + err(1, NULL); + new->execute = option->execute; + new->flags = option->flags; + new->next = NULL; + return new; +} + +/* + * find_parsenum -- + * Parse a string of the form [+-]# and return the value. + */ +static long long +find_parsenum(PLAN *plan, const char *option, char *vp, char *endch) +{ + long long value; + char *endchar, *str; /* Pointer to character ending conversion. */ + + /* Determine comparison from leading + or -. */ + str = vp; + switch (*str) { + case '+': + ++str; + plan->flags |= F_GREATER; + break; + case '-': + ++str; + plan->flags |= F_LESSTHAN; + break; + default: + plan->flags |= F_EQUAL; + break; + } + + /* + * Convert the string with strtoq(). Note, if strtoq() returns zero + * and endchar points to the beginning of the string we know we have + * a syntax error. + */ + value = strtoq(str, &endchar, 10); + if (value == 0 && endchar == str) + errx(1, "%s: %s: illegal numeric value", option, vp); + if (endchar[0] && endch == NULL) + errx(1, "%s: %s: illegal trailing character", option, vp); + if (endch) + *endch = endchar[0]; + return value; +} + +/* + * find_parsetime -- + * Parse a string of the form [+-]([0-9]+[smhdw]?)+ and return the value. + */ +static long long +find_parsetime(PLAN *plan, const char *option, char *vp) +{ + long long secs, value; + char *str, *unit; /* Pointer to character ending conversion. */ + + /* Determine comparison from leading + or -. */ + str = vp; + switch (*str) { + case '+': + ++str; + plan->flags |= F_GREATER; + break; + case '-': + ++str; + plan->flags |= F_LESSTHAN; + break; + default: + plan->flags |= F_EQUAL; + break; + } + + value = strtoq(str, &unit, 10); + if (value == 0 && unit == str) { + errx(1, "%s: %s: illegal time value", option, vp); + /* NOTREACHED */ + } + if (*unit == '\0') + return value; + + /* Units syntax. */ + secs = 0; + for (;;) { + switch(*unit) { + case 's': /* seconds */ + secs += value; + break; + case 'm': /* minutes */ + secs += value * 60; + break; + case 'h': /* hours */ + secs += value * 3600; + break; + case 'd': /* days */ + secs += value * 86400; + break; + case 'w': /* weeks */ + secs += value * 604800; + break; + default: + errx(1, "%s: %s: bad unit '%c'", option, vp, *unit); + /* NOTREACHED */ + } + str = unit + 1; + if (*str == '\0') /* EOS */ + break; + value = strtoq(str, &unit, 10); + if (value == 0 && unit == str) { + errx(1, "%s: %s: illegal time value", option, vp); + /* NOTREACHED */ + } + if (*unit == '\0') { + errx(1, "%s: %s: missing trailing unit", option, vp); + /* NOTREACHED */ + } + } + plan->flags |= F_EXACTTIME; + return secs; +} + +/* + * nextarg -- + * Check that another argument still exists, return a pointer to it, + * and increment the argument vector pointer. + */ +static char * +nextarg(OPTION *option, char ***argvp) +{ + char *arg; + + if ((arg = **argvp) == 0) + errx(1, "%s: requires additional arguments", option->name); + (*argvp)++; + return arg; +} /* nextarg() */ + +/* + * The value of n for the inode times (atime, birthtime, ctime, mtime) is a + * range, i.e. n matches from (n - 1) to n 24 hour periods. This interacts + * with -n, such that "-mtime -1" would be less than 0 days, which isn't what + * the user wanted. Correct so that -1 is "less than 1". + */ +#define TIME_CORRECT(p) \ + if (((p)->flags & F_ELG_MASK) == F_LESSTHAN) \ + ++((p)->t_data); + +/* + * -[acm]min n functions -- + * + * True if the difference between the + * file access time (-amin) + * file birth time (-Bmin) + * last change of file status information (-cmin) + * file modification time (-mmin) + * and the current time is n min periods. + */ +int +f_Xmin(PLAN *plan, FTSENT *entry) +{ + if (plan->flags & F_TIME_C) { + COMPARE((now - entry->fts_statp->st_ctime + + 60 - 1) / 60, plan->t_data); + } else if (plan->flags & F_TIME_A) { + COMPARE((now - entry->fts_statp->st_atime + + 60 - 1) / 60, plan->t_data); + } else if (plan->flags & F_TIME_B) { + COMPARE((now - entry->fts_statp->st_birthtime + + 60 - 1) / 60, plan->t_data); + } else { + COMPARE((now - entry->fts_statp->st_mtime + + 60 - 1) / 60, plan->t_data); + } +} + +PLAN * +c_Xmin(OPTION *option, char ***argvp) +{ + char *nmins; + PLAN *new; + + nmins = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + new->t_data = find_parsenum(new, option->name, nmins, NULL); + TIME_CORRECT(new); + return new; +} + +/* + * -[acm]time n functions -- + * + * True if the difference between the + * file access time (-atime) + * file birth time (-Btime) + * last change of file status information (-ctime) + * file modification time (-mtime) + * and the current time is n 24 hour periods. + */ + +int +f_Xtime(PLAN *plan, FTSENT *entry) +{ + time_t xtime; + + if (plan->flags & F_TIME_A) + xtime = entry->fts_statp->st_atime; + else if (plan->flags & F_TIME_B) + xtime = entry->fts_statp->st_birthtime; + else if (plan->flags & F_TIME_C) + xtime = entry->fts_statp->st_ctime; + else + xtime = entry->fts_statp->st_mtime; + + if (plan->flags & F_EXACTTIME) + COMPARE(now - xtime, plan->t_data); + else + COMPARE((now - xtime + (COMPAT_MODE("bin/find", "unix2003") ? 0 : 86400 - 1)) / 86400, plan->t_data); +} + +PLAN * +c_Xtime(OPTION *option, char ***argvp) +{ + char *value; + PLAN *new; + + value = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + new->t_data = find_parsetime(new, option->name, value); + if (!(new->flags & F_EXACTTIME) && !COMPAT_MODE("bin/find", "unix2003")) + TIME_CORRECT(new); + return new; +} + +/* + * -maxdepth/-mindepth n functions -- + * + * Does the same as -prune if the level of the current file is + * greater/less than the specified maximum/minimum depth. + * + * Note that -maxdepth and -mindepth are handled specially in + * find_execute() so their f_* functions are set to f_always_true(). + */ +PLAN * +c_mXXdepth(OPTION *option, char ***argvp) +{ + char *dstr; + PLAN *new; + + dstr = nextarg(option, argvp); + if (dstr[0] == '-') + /* all other errors handled by find_parsenum() */ + errx(1, "%s: %s: value must be positive", option->name, dstr); + + new = palloc(option); + if (option->flags & F_MAXDEPTH) + maxdepth = find_parsenum(new, option->name, dstr, NULL); + else + mindepth = find_parsenum(new, option->name, dstr, NULL); + return new; +} + +/* + * -acl function -- + * + * Show files with EXTENDED ACL attributes. + */ +#ifdef __APPLE__ +int +f_acl(PLAN *plan __unused, FTSENT *entry) +{ + acl_t facl; + int match; + acl_entry_t ae; + + match = 0; + if ((facl = acl_get_link_np(entry->fts_accpath, ACL_TYPE_EXTENDED)) != NULL) { + if (acl_get_entry(facl, ACL_FIRST_ENTRY, &ae) == 0) { + match = 1; + } + acl_free(facl); + } + return match; +} +#else /* !__APPLE__ */ +int +f_acl(PLAN *plan __unused, FTSENT *entry) +{ + acl_t facl; + acl_type_t acl_type; + int acl_supported = 0, ret, trivial; + + if (S_ISLNK(entry->fts_statp->st_mode)) + return 0; + ret = pathconf(entry->fts_accpath, _PC_ACL_NFS4); + if (ret > 0) { + acl_supported = 1; + acl_type = ACL_TYPE_NFS4; + } else if (ret < 0 && errno != EINVAL) { + warn("%s", entry->fts_accpath); + return (0); + } + if (acl_supported == 0) { + ret = pathconf(entry->fts_accpath, _PC_ACL_EXTENDED); + if (ret > 0) { + acl_supported = 1; + acl_type = ACL_TYPE_ACCESS; + } else if (ret < 0 && errno != EINVAL) { + warn("%s", entry->fts_accpath); + return (0); + } + } + if (acl_supported == 0) + return (0); + + facl = acl_get_file(entry->fts_accpath, acl_type); + if (facl == NULL) { + warn("%s", entry->fts_accpath); + return (0); + } + ret = acl_is_trivial_np(facl, &trivial); + acl_free(facl); + if (ret) { + warn("%s", entry->fts_accpath); + acl_free(facl); + return (0); + } + if (trivial) + return (0); + return (1); +} +#endif /* __APPLE__ */ + +PLAN * +c_acl(OPTION *option, char ***argvp __unused) +{ +#ifndef __APPLE__ + ftsoptions &= ~FTS_NOSTAT; +#endif /* !__APPLE__ */ + return (palloc(option)); +} + +#ifdef __APPLE__ +int +f_xattr(PLAN *plan __unused, FTSENT *entry) +{ + ssize_t xattr; + int match; + + match = 0; + xattr = listxattr(entry->fts_accpath, NULL, 0, XATTR_NOFOLLOW); + if (xattr > 0) { + match = 1; + } + return match; +} + +int +f_xattrname(PLAN *plan, FTSENT *entry) +{ + ssize_t xattr; + int match; + + match = 0; + xattr = getxattr(entry->fts_accpath, plan->c_data, NULL, 0, 0, XATTR_NOFOLLOW); + if (xattr > 0) { + match = 1; + } + return match; +} +#endif /* __APPLE__ */ + +/* + * -delete functions -- + * + * True always. Makes its best shot and continues on regardless. + */ +int +f_delete(PLAN *plan __unused, FTSENT *entry) +{ + /* ignore these from fts */ + if (strcmp(entry->fts_accpath, ".") == 0 || + strcmp(entry->fts_accpath, "..") == 0) + return 1; + + /* sanity check */ + if (isdepth == 0 || /* depth off */ + (ftsoptions & FTS_NOSTAT)) /* not stat()ing */ + errx(1, "-delete: insecure options got turned on"); + + if (!(ftsoptions & FTS_PHYSICAL) || /* physical off */ + (ftsoptions & FTS_LOGICAL)) /* or finally, logical on */ + errx(1, "-delete: forbidden when symlinks are followed"); + + /* Potentially unsafe - do not accept relative paths whatsoever */ + if (strchr(entry->fts_accpath, '/') != NULL) + errx(1, "-delete: %s: relative path potentially not safe", + entry->fts_accpath); + + /* Turn off user immutable bits if running as root */ + if ((entry->fts_statp->st_flags & (UF_APPEND|UF_IMMUTABLE)) && + !(entry->fts_statp->st_flags & (SF_APPEND|SF_IMMUTABLE)) && + geteuid() == 0) + lchflags(entry->fts_accpath, + entry->fts_statp->st_flags &= ~(UF_APPEND|UF_IMMUTABLE)); + + /* rmdir directories, unlink everything else */ + if (S_ISDIR(entry->fts_statp->st_mode)) { + if (rmdir(entry->fts_accpath) < 0 && errno != ENOTEMPTY) + warn("-delete: rmdir(%s)", entry->fts_path); + } else { + if (unlink(entry->fts_accpath) < 0) + warn("-delete: unlink(%s)", entry->fts_path); + } + + /* "succeed" */ + return 1; +} + +PLAN * +c_delete(OPTION *option, char ***argvp __unused) +{ + + ftsoptions &= ~FTS_NOSTAT; /* no optimise */ + isoutput = 1; /* possible output */ + isdepth = 1; /* -depth implied */ + + return palloc(option); +} + + +/* + * always_true -- + * + * Always true, used for -maxdepth, -mindepth, -xdev, -follow, and -true + */ +int +f_always_true(PLAN *plan __unused, FTSENT *entry __unused) +{ + return 1; +} + +/* + * -depth functions -- + * + * With argument: True if the file is at level n. + * Without argument: Always true, causes descent of the directory hierarchy + * to be done so that all entries in a directory are acted on before the + * directory itself. + */ +int +f_depth(PLAN *plan, FTSENT *entry) +{ + if (plan->flags & F_DEPTH) + COMPARE(entry->fts_level, plan->d_data); + else + return 1; +} + +PLAN * +c_depth(OPTION *option, char ***argvp) +{ + PLAN *new; + char *str; + + new = palloc(option); + + str = **argvp; + if (str && !(new->flags & F_DEPTH)) { + /* skip leading + or - */ + if (*str == '+' || *str == '-') + str++; + /* skip sign */ + if (*str == '+' || *str == '-') + str++; + if (isdigit(*str)) + new->flags |= F_DEPTH; + } + + if (new->flags & F_DEPTH) { /* -depth n */ + char *ndepth; + + ndepth = nextarg(option, argvp); + new->d_data = find_parsenum(new, option->name, ndepth, NULL); + } else { /* -d */ + isdepth = 1; + } + + return new; +} + +/* + * -empty functions -- + * + * True if the file or directory is empty + */ +int +f_empty(PLAN *plan __unused, FTSENT *entry) +{ + if (S_ISREG(entry->fts_statp->st_mode) && + entry->fts_statp->st_size == 0) + return 1; + if (S_ISDIR(entry->fts_statp->st_mode)) { + struct dirent *dp; + int empty; + DIR *dir; + + empty = 1; + dir = opendir(entry->fts_accpath); + if (dir == NULL) + return 0; + for (dp = readdir(dir); dp; dp = readdir(dir)) + if (dp->d_name[0] != '.' || + (dp->d_name[1] != '\0' && + (dp->d_name[1] != '.' || dp->d_name[2] != '\0'))) { + empty = 0; + break; + } + closedir(dir); + return empty; + } + return 0; +} + +PLAN * +c_empty(OPTION *option, char ***argvp __unused) +{ + ftsoptions &= ~FTS_NOSTAT; + + return palloc(option); +} + +/* + * [-exec | -execdir | -ok] utility [arg ... ] ; functions -- + * + * True if the executed utility returns a zero value as exit status. + * The end of the primary expression is delimited by a semicolon. If + * "{}" occurs anywhere, it gets replaced by the current pathname, + * or, in the case of -execdir, the current basename (filename + * without leading directory prefix). For -exec and -ok, + * the current directory for the execution of utility is the same as + * the current directory when the find utility was started, whereas + * for -execdir, it is the directory the file resides in. + * + * The primary -ok differs from -exec in that it requests affirmation + * of the user before executing the utility. + */ +int +f_exec(PLAN *plan, FTSENT *entry) +{ + int cnt; + pid_t pid; + int status; + char *file; + + if (entry == NULL && plan->flags & F_EXECPLUS) { + if (plan->e_ppos == plan->e_pbnum) + return (1); + plan->e_argv[plan->e_ppos] = NULL; + goto doexec; + } + + /* XXX - if file/dir ends in '/' this will not work -- can it? */ + if ((plan->flags & F_EXECDIR) && \ + (file = strrchr(entry->fts_path, '/'))) + file++; + else + file = entry->fts_path; + + if (plan->flags & F_EXECPLUS) { + if ((plan->e_argv[plan->e_ppos] = strdup(file)) == NULL) + err(1, NULL); + plan->e_len[plan->e_ppos] = strlen(file); + plan->e_psize += plan->e_len[plan->e_ppos]; + if (++plan->e_ppos < plan->e_pnummax && + plan->e_psize < plan->e_psizemax) + return (1); + plan->e_argv[plan->e_ppos] = NULL; + } else { + for (cnt = 0; plan->e_argv[cnt]; ++cnt) + if (plan->e_len[cnt]) + brace_subst(plan->e_orig[cnt], + &plan->e_argv[cnt], file, + plan->e_len[cnt]); + } + +doexec: if ((plan->flags & F_NEEDOK) && !queryuser(plan->e_argv)) + return 0; + + /* make sure find output is interspersed correctly with subprocesses */ + fflush(stdout); + fflush(stderr); + + switch (pid = fork()) { + case -1: + err(1, "fork"); + /* NOTREACHED */ + case 0: + /* change dir back from where we started */ + if (!(plan->flags & F_EXECDIR) && fchdir(dotfd)) { + warn("chdir"); + _exit(1); + } + execvp(plan->e_argv[0], plan->e_argv); + warn("%s", plan->e_argv[0]); + _exit(1); + } + if (plan->flags & F_EXECPLUS) { + while (--plan->e_ppos >= plan->e_pbnum) + free(plan->e_argv[plan->e_ppos]); + plan->e_ppos = plan->e_pbnum; + plan->e_psize = plan->e_pbsize; + } + pid = waitpid(pid, &status, 0); + if (plan->flags & F_EXECPLUS && WIFEXITED(status) && WEXITSTATUS(status) && !execplus_error) { + /* Test 140 (8907531, 10656525) */ + execplus_error = WEXITSTATUS(status); + } + return (pid != -1 && WIFEXITED(status) && !WEXITSTATUS(status)); +} + +/* + * c_exec, c_execdir, c_ok -- + * build three parallel arrays, one with pointers to the strings passed + * on the command line, one with (possibly duplicated) pointers to the + * argv array, and one with integer values that are lengths of the + * strings, but also flags meaning that the string has to be massaged. + */ +PLAN * +c_exec(OPTION *option, char ***argvp) +{ + PLAN *new; /* node returned */ + long argmax; + int cnt, i; + char **argv, **ap, **ep, *p; + + /* XXX - was in c_execdir, but seems unnecessary!? + ftsoptions &= ~FTS_NOSTAT; + */ + isoutput = 1; + + /* XXX - this is a change from the previous coding */ + new = palloc(option); + + for (ap = argv = *argvp;; ++ap) { + if (!*ap) + errx(1, + "%s: no terminating \";\" or \"+\"", option->name); + if (**ap == ';') + break; + if (**ap == '+' && ap != argv && strcmp(*(ap - 1), "{}") == 0) { + new->flags |= F_EXECPLUS; + break; + } + } + + if (ap == argv) + errx(1, "%s: no command specified", option->name); + + cnt = ap - *argvp + 1; + if (new->flags & F_EXECPLUS) { + new->e_ppos = new->e_pbnum = cnt - 2; + if ((argmax = sysconf(_SC_ARG_MAX)) == -1) { + warn("sysconf(_SC_ARG_MAX)"); + argmax = _POSIX_ARG_MAX; + } + argmax -= 1024; + for (ep = environ; *ep != NULL; ep++) + argmax -= strlen(*ep) + 1 + sizeof(*ep); + argmax -= 1 + sizeof(*ep); + new->e_pnummax = argmax / 16; + argmax -= sizeof(char *) * new->e_pnummax; + if (argmax <= 0) + errx(1, "no space for arguments"); + new->e_psizemax = argmax; + new->e_pbsize = 0; + cnt += new->e_pnummax + 1; + new->e_next = lastexecplus; + lastexecplus = new; + } + if ((new->e_argv = malloc(cnt * sizeof(char *))) == NULL) + err(1, NULL); + if ((new->e_orig = malloc(cnt * sizeof(char *))) == NULL) + err(1, NULL); + if ((new->e_len = malloc(cnt * sizeof(int))) == NULL) + err(1, NULL); + + for (argv = *argvp, cnt = 0; argv < ap; ++argv, ++cnt) { + new->e_orig[cnt] = *argv; + if (new->flags & F_EXECPLUS) + new->e_pbsize += strlen(*argv) + 1; + for (p = *argv; *p; ++p) + if (!(new->flags & F_EXECPLUS) && p[0] == '{' && + p[1] == '}') { + if ((new->e_argv[cnt] = + malloc(MAXPATHLEN)) == NULL) + err(1, NULL); + new->e_len[cnt] = MAXPATHLEN; + break; + } + if (!*p) { + new->e_argv[cnt] = *argv; + new->e_len[cnt] = 0; + } + } + if (new->flags & F_EXECPLUS) { + new->e_psize = new->e_pbsize; + cnt--; + for (i = 0; i < new->e_pnummax; i++) { + new->e_argv[cnt] = NULL; + new->e_len[cnt] = 0; + cnt++; + } + argv = ap; + goto done; + } + new->e_argv[cnt] = new->e_orig[cnt] = NULL; + +done: *argvp = argv + 1; + return new; +} + +/* Finish any pending -exec ... {} + functions. */ +void +finish_execplus(void) +{ + PLAN *p; + + p = lastexecplus; + while (p != NULL) { + (p->execute)(p, NULL); + p = p->e_next; + } +} + +int +f_flags(PLAN *plan, FTSENT *entry) +{ + u_long flags; + + flags = entry->fts_statp->st_flags; + if (plan->flags & F_ATLEAST) + return (flags | plan->fl_flags) == flags && + !(flags & plan->fl_notflags); + else if (plan->flags & F_ANY) + return (flags & plan->fl_flags) || + (flags | plan->fl_notflags) != flags; + else + return flags == plan->fl_flags && + !(plan->fl_flags & plan->fl_notflags); +} + +PLAN * +c_flags(OPTION *option, char ***argvp) +{ + char *flags_str; + PLAN *new; + u_long flags, notflags; + + flags_str = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + + if (*flags_str == '-') { + new->flags |= F_ATLEAST; + flags_str++; + } else if (*flags_str == '+') { + new->flags |= F_ANY; + flags_str++; + } + if (strtofflags(&flags_str, &flags, ¬flags) == 1) + errx(1, "%s: %s: illegal flags string", option->name, flags_str); + + new->fl_flags = flags; + new->fl_notflags = notflags; + return new; +} + +/* + * -follow functions -- + * + * Always true, causes symbolic links to be followed on a global + * basis. + */ +PLAN * +c_follow(OPTION *option, char ***argvp __unused) +{ + ftsoptions &= ~FTS_PHYSICAL; + ftsoptions |= FTS_LOGICAL; + + return palloc(option); +} + +/* + * -fstype functions -- + * + * True if the file is of a certain type. + */ +int +f_fstype(PLAN *plan, FTSENT *entry) +{ + static dev_t curdev; /* need a guaranteed illegal dev value */ + static int first = 1; + struct statfs sb; + static int val_flags; + static char fstype[sizeof(sb.f_fstypename)]; + char *p, save[2] = {0,0}; + + if ((plan->flags & F_MTMASK) == F_MTUNKNOWN) + return 0; + + /* Only check when we cross mount point. */ + if (first || curdev != entry->fts_statp->st_dev) { + curdev = entry->fts_statp->st_dev; + + /* + * Statfs follows symlinks; find wants the link's filesystem, + * not where it points. + */ + if (entry->fts_info == FTS_SL || + entry->fts_info == FTS_SLNONE) { + if ((p = strrchr(entry->fts_accpath, '/')) != NULL) + ++p; + else + p = entry->fts_accpath; + save[0] = p[0]; + p[0] = '.'; + save[1] = p[1]; + p[1] = '\0'; + } else + p = NULL; + + if (statfs(entry->fts_accpath, &sb)) + err(1, "%s", entry->fts_accpath); + + if (p) { + p[0] = save[0]; + p[1] = save[1]; + } + + first = 0; + + /* + * Further tests may need both of these values, so + * always copy both of them. + */ + val_flags = sb.f_flags; + strlcpy(fstype, sb.f_fstypename, sizeof(fstype)); + } + switch (plan->flags & F_MTMASK) { + case F_MTFLAG: + return val_flags & plan->mt_data; + case F_MTTYPE: + return (strncmp(fstype, plan->c_data, sizeof(fstype)) == 0); + default: + abort(); + } +} + +PLAN * +c_fstype(OPTION *option, char ***argvp) +{ + char *fsname; + PLAN *new; + + fsname = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + switch (*fsname) { + case 'l': + if (!strcmp(fsname, "local")) { + new->flags |= F_MTFLAG; + new->mt_data = MNT_LOCAL; + return new; + } + break; + case 'r': + if (!strcmp(fsname, "rdonly")) { + new->flags |= F_MTFLAG; + new->mt_data = MNT_RDONLY; + return new; + } + break; + } + + new->flags |= F_MTTYPE; + new->c_data = fsname; + return new; +} + +/* + * -group gname functions -- + * + * True if the file belongs to the group gname. If gname is numeric and + * an equivalent of the getgrnam() function does not return a valid group + * name, gname is taken as a group ID. + */ +int +f_group(PLAN *plan, FTSENT *entry) +{ + COMPARE(entry->fts_statp->st_gid, plan->g_data); +} + +PLAN * +c_group(OPTION *option, char ***argvp) +{ + char *gname; + PLAN *new; + struct group *g; + gid_t gid; + + gname = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + g = getgrnam(gname); + if (g == NULL) { + char* cp = gname; + if (gname[0] == '-' || gname[0] == '+') + gname++; + gid = atoi(gname); + if (gid == 0 && gname[0] != '0') + errx(1, "%s: %s: no such group", option->name, gname); + gid = find_parsenum(new, option->name, cp, NULL); + } else + gid = g->gr_gid; + + new->g_data = gid; + return new; +} + +/* + * -inum n functions -- + * + * True if the file has inode # n. + */ +int +f_inum(PLAN *plan, FTSENT *entry) +{ + COMPARE(entry->fts_statp->st_ino, plan->i_data); +} + +PLAN * +c_inum(OPTION *option, char ***argvp) +{ + char *inum_str; + PLAN *new; + + inum_str = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + new->i_data = find_parsenum(new, option->name, inum_str, NULL); + return new; +} + +/* + * -samefile FN + * + * True if the file has the same inode (eg hard link) FN + */ + +/* f_samefile is just f_inum */ +PLAN * +c_samefile(OPTION *option, char ***argvp) +{ + char *fn; + PLAN *new; + struct stat sb; + + fn = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + if (stat(fn, &sb)) + err(1, "%s", fn); + new->i_data = sb.st_ino; + return new; +} + +/* + * -links n functions -- + * + * True if the file has n links. + */ +int +f_links(PLAN *plan, FTSENT *entry) +{ + COMPARE(entry->fts_statp->st_nlink, plan->l_data); +} + +PLAN * +c_links(OPTION *option, char ***argvp) +{ + char *nlinks; + PLAN *new; + + nlinks = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + new->l_data = (nlink_t)find_parsenum(new, option->name, nlinks, NULL); + return new; +} + +/* + * -ls functions -- + * + * Always true - prints the current entry to stdout in "ls" format. + */ +int +f_ls(PLAN *plan __unused, FTSENT *entry) +{ + printlong(entry->fts_path, entry->fts_accpath, entry->fts_statp); + return 1; +} + +PLAN * +c_ls(OPTION *option, char ***argvp __unused) +{ + ftsoptions &= ~FTS_NOSTAT; + isoutput = 1; + + return palloc(option); +} + +/* + * -name functions -- + * + * True if the basename of the filename being examined + * matches pattern using Pattern Matching Notation S3.14 + */ +int +f_name(PLAN *plan, FTSENT *entry) +{ + char fn[PATH_MAX]; + const char *name; + ssize_t len; + + if (plan->flags & F_LINK) { + /* + * The below test both avoids obviously useless readlink() + * calls and ensures that symlinks with existent target do + * not match if symlinks are being followed. + * Assumption: fts will stat all symlinks that are to be + * followed and will return the stat information. + */ + if (entry->fts_info != FTS_NSOK && entry->fts_info != FTS_SL && + entry->fts_info != FTS_SLNONE) + return 0; + len = readlink(entry->fts_accpath, fn, sizeof(fn) - 1); + if (len == -1) + return 0; + fn[len] = '\0'; + name = fn; + } else if (entry->fts_namelen == 0) { + name = basename(entry->fts_path); + } else + name = entry->fts_name; + return !fnmatch(plan->c_data, name, + plan->flags & F_IGNCASE ? FNM_CASEFOLD : 0); +} + +PLAN * +c_name(OPTION *option, char ***argvp) +{ + char *pattern; + PLAN *new; + + pattern = nextarg(option, argvp); + new = palloc(option); + new->c_data = pattern; + return new; +} + +/* + * -newer file functions -- + * + * True if the current file has been modified more recently + * then the modification time of the file named by the pathname + * file. + */ +int +f_newer(PLAN *plan, FTSENT *entry) +{ + if (plan->flags & F_TIME_C) + return entry->fts_statp->st_ctime > plan->t_data; + else if (plan->flags & F_TIME_A) + return entry->fts_statp->st_atime > plan->t_data; + else if (plan->flags & F_TIME_B) + return entry->fts_statp->st_birthtime > plan->t_data; + else + return entry->fts_statp->st_mtime > plan->t_data; +} + +PLAN * +c_newer(OPTION *option, char ***argvp) +{ + char *fn_or_tspec; + PLAN *new; + struct stat sb; + + fn_or_tspec = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + /* compare against what */ + if (option->flags & F_TIME2_T) { + new->t_data = get_date(fn_or_tspec); + if (new->t_data == (time_t) -1) + errx(1, "Can't parse date/time: %s", fn_or_tspec); + } else { + if (stat(fn_or_tspec, &sb)) + err(1, "%s", fn_or_tspec); + if (option->flags & F_TIME2_C) + new->t_data = sb.st_ctime; + else if (option->flags & F_TIME2_A) + new->t_data = sb.st_atime; + else if (option->flags & F_TIME2_B) + new->t_data = sb.st_birthtime; + else + new->t_data = sb.st_mtime; + } + return new; +} + +/* + * -nogroup functions -- + * + * True if file belongs to a user ID for which the equivalent + * of the getgrnam() 9.2.1 [POSIX.1] function returns NULL. + */ +int +f_nogroup(PLAN *plan __unused, FTSENT *entry) +{ + return group_from_gid(entry->fts_statp->st_gid, 1) == NULL; +} + +PLAN * +c_nogroup(OPTION *option, char ***argvp __unused) +{ + ftsoptions &= ~FTS_NOSTAT; + + return palloc(option); +} + +/* + * -nouser functions -- + * + * True if file belongs to a user ID for which the equivalent + * of the getpwuid() 9.2.2 [POSIX.1] function returns NULL. + */ +int +f_nouser(PLAN *plan __unused, FTSENT *entry) +{ + return user_from_uid(entry->fts_statp->st_uid, 1) == NULL; +} + +PLAN * +c_nouser(OPTION *option, char ***argvp __unused) +{ + ftsoptions &= ~FTS_NOSTAT; + + return palloc(option); +} + +/* + * -path functions -- + * + * True if the path of the filename being examined + * matches pattern using Pattern Matching Notation S3.14 + */ +int +f_path(PLAN *plan, FTSENT *entry) +{ + return !fnmatch(plan->c_data, entry->fts_path, + plan->flags & F_IGNCASE ? FNM_CASEFOLD : 0); +} + +/* c_path is the same as c_name */ + +/* + * -perm functions -- + * + * The mode argument is used to represent file mode bits. If it starts + * with a leading digit, it's treated as an octal mode, otherwise as a + * symbolic mode. + */ +int +f_perm(PLAN *plan, FTSENT *entry) +{ + mode_t mode; + + mode = entry->fts_statp->st_mode & + (S_ISUID|S_ISGID|S_ISTXT|S_IRWXU|S_IRWXG|S_IRWXO); + if (plan->flags & F_ATLEAST) + return (plan->m_data | mode) == mode; + else if (plan->flags & F_ANY) + return (mode & plan->m_data); + else + return mode == plan->m_data; + /* NOTREACHED */ +} + +PLAN * +c_perm(OPTION *option, char ***argvp) +{ + char *perm; + PLAN *new; + mode_t *set; + + perm = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + + if (*perm == '-') { + new->flags |= F_ATLEAST; + ++perm; + } else if (*perm == '+') { + if ((set = setmode(perm + 1)) != NULL) { + new->flags |= F_ANY; + ++perm; + free(set); + } + } + + if ((set = setmode(perm)) == NULL) + errx(1, "%s: %s: illegal mode string", option->name, perm); + + new->m_data = getmode(set, 0); + free(set); + return new; +} + +/* + * -print functions -- + * + * Always true, causes the current pathname to be written to + * standard output. + */ +int +f_print(PLAN *plan __unused, FTSENT *entry) +{ + (void)puts(entry->fts_path); + return 1; +} + +PLAN * +c_print(OPTION *option, char ***argvp __unused) +{ + isoutput = 1; + + return palloc(option); +} + +/* + * -print0 functions -- + * + * Always true, causes the current pathname to be written to + * standard output followed by a NUL character + */ +int +f_print0(PLAN *plan __unused, FTSENT *entry) +{ + fputs(entry->fts_path, stdout); + fputc('\0', stdout); + return 1; +} + +/* c_print0 is the same as c_print */ + +/* + * -prune functions -- + * + * Prune a portion of the hierarchy. + */ +int +f_prune(PLAN *plan __unused, FTSENT *entry) +{ + if (fts_set(tree, entry, FTS_SKIP)) + err(1, "%s", entry->fts_path); + return 1; +} + +/* c_prune == c_simple */ + +/* + * -regex functions -- + * + * True if the whole path of the file matches pattern using + * regular expression. + */ +int +f_regex(PLAN *plan, FTSENT *entry) +{ + char *str; + int len; + regex_t *pre; + regmatch_t pmatch; + int errcode; + char errbuf[LINE_MAX]; + int matched; + + pre = plan->re_data; + str = entry->fts_path; + len = strlen(str); + matched = 0; + + pmatch.rm_so = 0; + pmatch.rm_eo = len; + + errcode = regexec(pre, str, 1, &pmatch, REG_STARTEND); + + if (errcode != 0 && errcode != REG_NOMATCH) { + regerror(errcode, pre, errbuf, sizeof errbuf); + errx(1, "%s: %s", + plan->flags & F_IGNCASE ? "-iregex" : "-regex", errbuf); + } + + if (errcode == 0 && pmatch.rm_so == 0 && pmatch.rm_eo == len) + matched = 1; + + return matched; +} + +PLAN * +c_regex(OPTION *option, char ***argvp) +{ + PLAN *new; + char *pattern; + regex_t *pre; + int errcode; + char errbuf[LINE_MAX]; + + if ((pre = malloc(sizeof(regex_t))) == NULL) + err(1, NULL); + + pattern = nextarg(option, argvp); + + if ((errcode = regcomp(pre, pattern, + regexp_flags | (option->flags & F_IGNCASE ? REG_ICASE : 0))) != 0) { + regerror(errcode, pre, errbuf, sizeof errbuf); + errx(1, "%s: %s: %s", + option->flags & F_IGNCASE ? "-iregex" : "-regex", + pattern, errbuf); + } + + new = palloc(option); + new->re_data = pre; + + return new; +} + +/* c_simple covers c_prune, c_openparen, c_closeparen, c_not, c_or, c_true, c_false */ + +PLAN * +c_simple(OPTION *option, char ***argvp __unused) +{ + return palloc(option); +} + +/* + * -size n[c] functions -- + * + * True if the file size in bytes, divided by an implementation defined + * value and rounded up to the next integer, is n. If n is followed by + * one of c k M G T P, the size is in bytes, kilobytes, + * megabytes, gigabytes, terabytes or petabytes respectively. + */ +#define FIND_SIZE 512 +static int divsize = 1; + +int +f_size(PLAN *plan, FTSENT *entry) +{ + off_t size; + + size = divsize ? (entry->fts_statp->st_size + FIND_SIZE - 1) / + FIND_SIZE : entry->fts_statp->st_size; + COMPARE(size, plan->o_data); +} + +PLAN * +c_size(OPTION *option, char ***argvp) +{ + char *size_str; + PLAN *new; + char endch; + off_t scale; + + size_str = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + endch = 'c'; + new->o_data = find_parsenum(new, option->name, size_str, &endch); + if (endch != '\0') { + divsize = 0; + + switch (endch) { + case 'c': /* characters */ + scale = 0x1LL; + break; + case 'k': /* kilobytes 1<<10 */ + scale = 0x400LL; + break; + case 'M': /* megabytes 1<<20 */ + scale = 0x100000LL; + break; + case 'G': /* gigabytes 1<<30 */ + scale = 0x40000000LL; + break; + case 'T': /* terabytes 1<<40 */ + scale = 0x1000000000LL; + break; + case 'P': /* petabytes 1<<50 */ + scale = 0x4000000000000LL; + break; + default: + errx(1, "%s: %s: illegal trailing character", + option->name, size_str); + break; + } + if (new->o_data > QUAD_MAX / scale) + errx(1, "%s: %s: value too large", + option->name, size_str); + new->o_data *= scale; + } + return new; +} + +/* + * -type c functions -- + * + * True if the type of the file is c, where c is b, c, d, p, f or w + * for block special file, character special file, directory, FIFO, + * regular file or whiteout respectively. + */ +int +f_type(PLAN *plan, FTSENT *entry) +{ + return (entry->fts_statp->st_mode & S_IFMT) == plan->m_data; +} + +PLAN * +c_type(OPTION *option, char ***argvp) +{ + char *typestring; + PLAN *new; + mode_t mask; + + typestring = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + switch (typestring[0]) { + case 'b': + mask = S_IFBLK; + break; + case 'c': + mask = S_IFCHR; + break; + case 'd': + mask = S_IFDIR; + break; + case 'f': + mask = S_IFREG; + break; + case 'l': + mask = S_IFLNK; + break; + case 'p': + mask = S_IFIFO; + break; + case 's': + mask = S_IFSOCK; + break; +#ifdef FTS_WHITEOUT + case 'w': + mask = S_IFWHT; + ftsoptions |= FTS_WHITEOUT; + break; +#endif /* FTS_WHITEOUT */ + default: + errx(1, "%s: %s: unknown type", option->name, typestring); + } + + new = palloc(option); + new->m_data = mask; + return new; +} + +/* + * -user uname functions -- + * + * True if the file belongs to the user uname. If uname is numeric and + * an equivalent of the getpwnam() S9.2.2 [POSIX.1] function does not + * return a valid user name, uname is taken as a user ID. + */ +int +f_user(PLAN *plan, FTSENT *entry) +{ + COMPARE(entry->fts_statp->st_uid, plan->u_data); +} + +PLAN * +c_user(OPTION *option, char ***argvp) +{ + char *username; + PLAN *new; + struct passwd *p; + uid_t uid; + + username = nextarg(option, argvp); + ftsoptions &= ~FTS_NOSTAT; + + new = palloc(option); + p = getpwnam(username); + if (p == NULL) { + char* cp = username; + if( username[0] == '-' || username[0] == '+' ) + username++; + uid = atoi(username); + if (uid == 0 && username[0] != '0') + errx(1, "%s: %s: no such user", option->name, username); + uid = find_parsenum(new, option->name, cp, NULL); + } else + uid = p->pw_uid; + + new->u_data = uid; + return new; +} + +/* + * -xdev functions -- + * + * Always true, causes find not to descend past directories that have a + * different device ID (st_dev, see stat() S5.6.2 [POSIX.1]) + */ +PLAN * +c_xdev(OPTION *option, char ***argvp __unused) +{ + ftsoptions |= FTS_XDEV; + + return palloc(option); +} + +/* + * ( expression ) functions -- + * + * True if expression is true. + */ +int +f_expr(PLAN *plan, FTSENT *entry) +{ + PLAN *p; + int state = 0; + + for (p = plan->p_data[0]; + p && (state = (p->execute)(p, entry)); p = p->next); + return state; +} + +/* + * f_openparen and f_closeparen nodes are temporary place markers. They are + * eliminated during phase 2 of find_formplan() --- the '(' node is converted + * to a f_expr node containing the expression and the ')' node is discarded. + * The functions themselves are only used as constants. + */ + +int +f_openparen(PLAN *plan __unused, FTSENT *entry __unused) +{ + abort(); +} + +int +f_closeparen(PLAN *plan __unused, FTSENT *entry __unused) +{ + abort(); +} + +/* c_openparen == c_simple */ +/* c_closeparen == c_simple */ + +/* + * AND operator. Since AND is implicit, no node is allocated. + */ +PLAN * +c_and(OPTION *option __unused, char ***argvp __unused) +{ + return NULL; +} + +/* + * ! expression functions -- + * + * Negation of a primary; the unary NOT operator. + */ +int +f_not(PLAN *plan, FTSENT *entry) +{ + PLAN *p; + int state = 0; + + for (p = plan->p_data[0]; + p && (state = (p->execute)(p, entry)); p = p->next); + return !state; +} + +/* c_not == c_simple */ + +/* + * expression -o expression functions -- + * + * Alternation of primaries; the OR operator. The second expression is + * not evaluated if the first expression is true. + */ +int +f_or(PLAN *plan, FTSENT *entry) +{ + PLAN *p; + int state = 0; + + for (p = plan->p_data[0]; + p && (state = (p->execute)(p, entry)); p = p->next); + + if (state) + return 1; + + for (p = plan->p_data[1]; + p && (state = (p->execute)(p, entry)); p = p->next); + return state; +} + +/* c_or == c_simple */ + +/* + * -false + * + * Always false. + */ +int +f_false(PLAN *plan __unused, FTSENT *entry __unused) +{ + return 0; +} + +/* c_false == c_simple */ + +/* + * -quit + * + * Exits the program + */ +int +f_quit(PLAN *plan __unused, FTSENT *entry __unused) +{ + exit(0); +} + +/* c_quit == c_simple */ diff --git a/shell_cmds/find/getdate.y b/shell_cmds/find/getdate.y new file mode 100644 index 0000000..5a97c2c --- /dev/null +++ b/shell_cmds/find/getdate.y @@ -0,0 +1,961 @@ +%{ +/* +** Originally written by Steven M. Bellovin <smb@research.att.com> while +** at the University of North Carolina at Chapel Hill. Later tweaked by +** a couple of people on Usenet. Completely overhauled by Rich $alz +** <rsalz@bbn.com> and Jim Berets <jberets@bbn.com> in August, 1990; +** +** This grammar has 10 shift/reduce conflicts. +** +** This code is in the public domain and has no copyright. +*/ +/* SUPPRESS 287 on yaccpar_sccsid *//* Unused static variable */ +/* SUPPRESS 288 on yyerrlab *//* Label unused */ + +#include <sys/cdefs.h> +__FBSDID("$FreeBSD: src/usr.bin/find/getdate.y,v 1.7 2010/02/09 21:24:41 ed Exp $"); + +#include <stdio.h> +#include <ctype.h> + +/* The code at the top of get_date which figures out the offset of the + current time zone checks various CPP symbols to see if special + tricks are need, but defaults to using the gettimeofday system call. + Include <sys/time.h> if that will be used. */ + +#if defined(vms) +# include <types.h> +#else /* defined(vms) */ +# include <sys/types.h> +# include <sys/time.h> +#endif /* !defined(vms) */ + +#if defined (__STDC__) || defined (USG) +#include <string.h> +#endif + +/* Some old versions of bison generate parsers that use bcopy. + That loses on systems that don't provide the function, so we have + to redefine it here. */ +#if !defined (HAVE_BCOPY) && defined (HAVE_MEMCPY) && !defined (bcopy) +#define bcopy(from, to, len) memcpy ((to), (from), (len)) +#endif + +#if defined (__STDC__) +#include <stdlib.h> +#endif + +/* NOTES on rebuilding getdate.c (particularly for inclusion in CVS + releases): + + We don't want to mess with all the portability hassles of alloca. + In particular, most (all?) versions of bison will use alloca in + their parser. If bison works on your system (e.g. it should work + with gcc), then go ahead and use it, but the more general solution + is to use byacc instead of bison, which should generate a portable + parser. I played with adding "#define alloca dont_use_alloca", to + give an error if the parser generator uses alloca (and thus detect + unportable getdate.c's), but that seems to cause as many problems + as it solves. */ + +#include <time.h> + +#define yyparse getdate_yyparse +#define yylex getdate_yylex +#define yyerror getdate_yyerror + +static int yyparse(void); +static int yylex(void); +static int yyerror(const char *); + +time_t get_date(char *); + +#define EPOCH 1970 +#define HOUR(x) ((time_t)(x) * 60) +#define SECSPERDAY (24L * 60L * 60L) + + +/* +** An entry in the lexical lookup table. +*/ +typedef struct _TABLE { + const char *name; + int type; + time_t value; +} TABLE; + + +/* +** Daylight-savings mode: on, off, or not yet known. +*/ +typedef enum _DSTMODE { + DSTon, DSToff, DSTmaybe +} DSTMODE; + +/* +** Meridian: am, pm, or 24-hour style. +*/ +typedef enum _MERIDIAN { + MERam, MERpm, MER24 +} MERIDIAN; + + +/* +** Global variables. We could get rid of most of these by using a good +** union as the yacc stack. (This routine was originally written before +** yacc had the %union construct.) Maybe someday; right now we only use +** the %union very rarely. +*/ +static char *yyInput; +static DSTMODE yyDSTmode; +static time_t yyDayOrdinal; +static time_t yyDayNumber; +static int yyHaveDate; +static int yyHaveDay; +static int yyHaveRel; +static int yyHaveTime; +static int yyHaveZone; +static time_t yyTimezone; +static time_t yyDay; +static time_t yyHour; +static time_t yyMinutes; +static time_t yyMonth; +static time_t yySeconds; +static time_t yyYear; +static MERIDIAN yyMeridian; +static time_t yyRelMonth; +static time_t yyRelSeconds; + +%} + +%union { + time_t Number; + enum _MERIDIAN Meridian; +} + +%token tAGO tDAY tDAYZONE tID tMERIDIAN tMINUTE_UNIT tMONTH tMONTH_UNIT +%token tSEC_UNIT tSNUMBER tUNUMBER tZONE tDST + +%type <Number> tDAY tDAYZONE tMINUTE_UNIT tMONTH tMONTH_UNIT +%type <Number> tSEC_UNIT tSNUMBER tUNUMBER tZONE +%type <Meridian> tMERIDIAN o_merid + +%% + +spec : /* NULL */ + | spec item + ; + +item : time { + yyHaveTime++; + } + | zone { + yyHaveZone++; + } + | date { + yyHaveDate++; + } + | day { + yyHaveDay++; + } + | rel { + yyHaveRel++; + } + | number + ; + +time : tUNUMBER tMERIDIAN { + yyHour = $1; + yyMinutes = 0; + yySeconds = 0; + yyMeridian = $2; + } + | tUNUMBER ':' tUNUMBER o_merid { + yyHour = $1; + yyMinutes = $3; + yySeconds = 0; + yyMeridian = $4; + } + | tUNUMBER ':' tUNUMBER tSNUMBER { + yyHour = $1; + yyMinutes = $3; + yyMeridian = MER24; + yyDSTmode = DSToff; + yyTimezone = - ($4 % 100 + ($4 / 100) * 60); + } + | tUNUMBER ':' tUNUMBER ':' tUNUMBER o_merid { + yyHour = $1; + yyMinutes = $3; + yySeconds = $5; + yyMeridian = $6; + } + | tUNUMBER ':' tUNUMBER ':' tUNUMBER tSNUMBER { + yyHour = $1; + yyMinutes = $3; + yySeconds = $5; + yyMeridian = MER24; + yyDSTmode = DSToff; + yyTimezone = - ($6 % 100 + ($6 / 100) * 60); + } + ; + +zone : tZONE { + yyTimezone = $1; + yyDSTmode = DSToff; + } + | tDAYZONE { + yyTimezone = $1; + yyDSTmode = DSTon; + } + | + tZONE tDST { + yyTimezone = $1; + yyDSTmode = DSTon; + } + ; + +day : tDAY { + yyDayOrdinal = 1; + yyDayNumber = $1; + } + | tDAY ',' { + yyDayOrdinal = 1; + yyDayNumber = $1; + } + | tUNUMBER tDAY { + yyDayOrdinal = $1; + yyDayNumber = $2; + } + ; + +date : tUNUMBER '/' tUNUMBER { + yyMonth = $1; + yyDay = $3; + } + | tUNUMBER '/' tUNUMBER '/' tUNUMBER { + if ($1 >= 100) { + yyYear = $1; + yyMonth = $3; + yyDay = $5; + } else { + yyMonth = $1; + yyDay = $3; + yyYear = $5; + } + } + | tUNUMBER tSNUMBER tSNUMBER { + /* ISO 8601 format. yyyy-mm-dd. */ + yyYear = $1; + yyMonth = -$2; + yyDay = -$3; + } + | tUNUMBER tMONTH tSNUMBER { + /* e.g. 17-JUN-1992. */ + yyDay = $1; + yyMonth = $2; + yyYear = -$3; + } + | tMONTH tUNUMBER { + yyMonth = $1; + yyDay = $2; + } + | tMONTH tUNUMBER ',' tUNUMBER { + yyMonth = $1; + yyDay = $2; + yyYear = $4; + } + | tUNUMBER tMONTH { + yyMonth = $2; + yyDay = $1; + } + | tUNUMBER tMONTH tUNUMBER { + yyMonth = $2; + yyDay = $1; + yyYear = $3; + } + ; + +rel : relunit tAGO { + yyRelSeconds = -yyRelSeconds; + yyRelMonth = -yyRelMonth; + } + | relunit + ; + +relunit : tUNUMBER tMINUTE_UNIT { + yyRelSeconds += $1 * $2 * 60L; + } + | tSNUMBER tMINUTE_UNIT { + yyRelSeconds += $1 * $2 * 60L; + } + | tMINUTE_UNIT { + yyRelSeconds += $1 * 60L; + } + | tSNUMBER tSEC_UNIT { + yyRelSeconds += $1; + } + | tUNUMBER tSEC_UNIT { + yyRelSeconds += $1; + } + | tSEC_UNIT { + yyRelSeconds++; + } + | tSNUMBER tMONTH_UNIT { + yyRelMonth += $1 * $2; + } + | tUNUMBER tMONTH_UNIT { + yyRelMonth += $1 * $2; + } + | tMONTH_UNIT { + yyRelMonth += $1; + } + ; + +number : tUNUMBER { + if (yyHaveTime && yyHaveDate && !yyHaveRel) + yyYear = $1; + else { + if($1>10000) { + yyHaveDate++; + yyDay= ($1)%100; + yyMonth= ($1/100)%100; + yyYear = $1/10000; + } + else { + yyHaveTime++; + if ($1 < 100) { + yyHour = $1; + yyMinutes = 0; + } + else { + yyHour = $1 / 100; + yyMinutes = $1 % 100; + } + yySeconds = 0; + yyMeridian = MER24; + } + } + } + ; + +o_merid : /* NULL */ { + $$ = MER24; + } + | tMERIDIAN { + $$ = $1; + } + ; + +%% + +/* Month and day table. */ +static TABLE const MonthDayTable[] = { + { "january", tMONTH, 1 }, + { "february", tMONTH, 2 }, + { "march", tMONTH, 3 }, + { "april", tMONTH, 4 }, + { "may", tMONTH, 5 }, + { "june", tMONTH, 6 }, + { "july", tMONTH, 7 }, + { "august", tMONTH, 8 }, + { "september", tMONTH, 9 }, + { "sept", tMONTH, 9 }, + { "october", tMONTH, 10 }, + { "november", tMONTH, 11 }, + { "december", tMONTH, 12 }, + { "sunday", tDAY, 0 }, + { "monday", tDAY, 1 }, + { "tuesday", tDAY, 2 }, + { "tues", tDAY, 2 }, + { "wednesday", tDAY, 3 }, + { "wednes", tDAY, 3 }, + { "thursday", tDAY, 4 }, + { "thur", tDAY, 4 }, + { "thurs", tDAY, 4 }, + { "friday", tDAY, 5 }, + { "saturday", tDAY, 6 }, + { NULL, 0, 0 } +}; + +/* Time units table. */ +static TABLE const UnitsTable[] = { + { "year", tMONTH_UNIT, 12 }, + { "month", tMONTH_UNIT, 1 }, + { "fortnight", tMINUTE_UNIT, 14 * 24 * 60 }, + { "week", tMINUTE_UNIT, 7 * 24 * 60 }, + { "day", tMINUTE_UNIT, 1 * 24 * 60 }, + { "hour", tMINUTE_UNIT, 60 }, + { "minute", tMINUTE_UNIT, 1 }, + { "min", tMINUTE_UNIT, 1 }, + { "second", tSEC_UNIT, 1 }, + { "sec", tSEC_UNIT, 1 }, + { NULL, 0, 0 } +}; + +/* Assorted relative-time words. */ +static TABLE const OtherTable[] = { + { "tomorrow", tMINUTE_UNIT, 1 * 24 * 60 }, + { "yesterday", tMINUTE_UNIT, -1 * 24 * 60 }, + { "today", tMINUTE_UNIT, 0 }, + { "now", tMINUTE_UNIT, 0 }, + { "last", tUNUMBER, -1 }, + { "this", tMINUTE_UNIT, 0 }, + { "next", tUNUMBER, 2 }, + { "first", tUNUMBER, 1 }, +/* { "second", tUNUMBER, 2 }, */ + { "third", tUNUMBER, 3 }, + { "fourth", tUNUMBER, 4 }, + { "fifth", tUNUMBER, 5 }, + { "sixth", tUNUMBER, 6 }, + { "seventh", tUNUMBER, 7 }, + { "eighth", tUNUMBER, 8 }, + { "ninth", tUNUMBER, 9 }, + { "tenth", tUNUMBER, 10 }, + { "eleventh", tUNUMBER, 11 }, + { "twelfth", tUNUMBER, 12 }, + { "ago", tAGO, 1 }, + { NULL, 0, 0 } +}; + +/* The timezone table. */ +/* Some of these are commented out because a time_t can't store a float. */ +static TABLE const TimezoneTable[] = { + { "gmt", tZONE, HOUR( 0) }, /* Greenwich Mean */ + { "ut", tZONE, HOUR( 0) }, /* Universal (Coordinated) */ + { "utc", tZONE, HOUR( 0) }, + { "wet", tZONE, HOUR( 0) }, /* Western European */ + { "bst", tDAYZONE, HOUR( 0) }, /* British Summer */ + { "wat", tZONE, HOUR( 1) }, /* West Africa */ + { "at", tZONE, HOUR( 2) }, /* Azores */ +#if 0 + /* For completeness. BST is also British Summer, and GST is + * also Guam Standard. */ + { "bst", tZONE, HOUR( 3) }, /* Brazil Standard */ + { "gst", tZONE, HOUR( 3) }, /* Greenland Standard */ +#endif +#if 0 + { "nft", tZONE, HOUR(3.5) }, /* Newfoundland */ + { "nst", tZONE, HOUR(3.5) }, /* Newfoundland Standard */ + { "ndt", tDAYZONE, HOUR(3.5) }, /* Newfoundland Daylight */ +#endif + { "ast", tZONE, HOUR( 4) }, /* Atlantic Standard */ + { "adt", tDAYZONE, HOUR( 4) }, /* Atlantic Daylight */ + { "est", tZONE, HOUR( 5) }, /* Eastern Standard */ + { "edt", tDAYZONE, HOUR( 5) }, /* Eastern Daylight */ + { "cst", tZONE, HOUR( 6) }, /* Central Standard */ + { "cdt", tDAYZONE, HOUR( 6) }, /* Central Daylight */ + { "mst", tZONE, HOUR( 7) }, /* Mountain Standard */ + { "mdt", tDAYZONE, HOUR( 7) }, /* Mountain Daylight */ + { "pst", tZONE, HOUR( 8) }, /* Pacific Standard */ + { "pdt", tDAYZONE, HOUR( 8) }, /* Pacific Daylight */ + { "yst", tZONE, HOUR( 9) }, /* Yukon Standard */ + { "ydt", tDAYZONE, HOUR( 9) }, /* Yukon Daylight */ + { "hst", tZONE, HOUR(10) }, /* Hawaii Standard */ + { "hdt", tDAYZONE, HOUR(10) }, /* Hawaii Daylight */ + { "cat", tZONE, HOUR(10) }, /* Central Alaska */ + { "ahst", tZONE, HOUR(10) }, /* Alaska-Hawaii Standard */ + { "nt", tZONE, HOUR(11) }, /* Nome */ + { "idlw", tZONE, HOUR(12) }, /* International Date Line West */ + { "cet", tZONE, -HOUR(1) }, /* Central European */ + { "met", tZONE, -HOUR(1) }, /* Middle European */ + { "mewt", tZONE, -HOUR(1) }, /* Middle European Winter */ + { "mest", tDAYZONE, -HOUR(1) }, /* Middle European Summer */ + { "swt", tZONE, -HOUR(1) }, /* Swedish Winter */ + { "sst", tDAYZONE, -HOUR(1) }, /* Swedish Summer */ + { "fwt", tZONE, -HOUR(1) }, /* French Winter */ + { "fst", tDAYZONE, -HOUR(1) }, /* French Summer */ + { "eet", tZONE, -HOUR(2) }, /* Eastern Europe, USSR Zone 1 */ + { "bt", tZONE, -HOUR(3) }, /* Baghdad, USSR Zone 2 */ +#if 0 + { "it", tZONE, -HOUR(3.5) },/* Iran */ +#endif + { "zp4", tZONE, -HOUR(4) }, /* USSR Zone 3 */ + { "zp5", tZONE, -HOUR(5) }, /* USSR Zone 4 */ +#if 0 + { "ist", tZONE, -HOUR(5.5) },/* Indian Standard */ +#endif + { "zp6", tZONE, -HOUR(6) }, /* USSR Zone 5 */ +#if 0 + /* For completeness. NST is also Newfoundland Stanard, and SST is + * also Swedish Summer. */ + { "nst", tZONE, -HOUR(6.5) },/* North Sumatra */ + { "sst", tZONE, -HOUR(7) }, /* South Sumatra, USSR Zone 6 */ +#endif /* 0 */ + { "wast", tZONE, -HOUR(7) }, /* West Australian Standard */ + { "wadt", tDAYZONE, -HOUR(7) }, /* West Australian Daylight */ +#if 0 + { "jt", tZONE, -HOUR(7.5) },/* Java (3pm in Cronusland!) */ +#endif + { "cct", tZONE, -HOUR(8) }, /* China Coast, USSR Zone 7 */ + { "jst", tZONE, -HOUR(9) }, /* Japan Standard, USSR Zone 8 */ +#if 0 + { "cast", tZONE, -HOUR(9.5) },/* Central Australian Standard */ + { "cadt", tDAYZONE, -HOUR(9.5) },/* Central Australian Daylight */ +#endif + { "east", tZONE, -HOUR(10) }, /* Eastern Australian Standard */ + { "eadt", tDAYZONE, -HOUR(10) }, /* Eastern Australian Daylight */ + { "gst", tZONE, -HOUR(10) }, /* Guam Standard, USSR Zone 9 */ + { "nzt", tZONE, -HOUR(12) }, /* New Zealand */ + { "nzst", tZONE, -HOUR(12) }, /* New Zealand Standard */ + { "nzdt", tDAYZONE, -HOUR(12) }, /* New Zealand Daylight */ + { "idle", tZONE, -HOUR(12) }, /* International Date Line East */ + { NULL, 0, 0 } +}; + +/* Military timezone table. */ +static TABLE const MilitaryTable[] = { + { "a", tZONE, HOUR( 1) }, + { "b", tZONE, HOUR( 2) }, + { "c", tZONE, HOUR( 3) }, + { "d", tZONE, HOUR( 4) }, + { "e", tZONE, HOUR( 5) }, + { "f", tZONE, HOUR( 6) }, + { "g", tZONE, HOUR( 7) }, + { "h", tZONE, HOUR( 8) }, + { "i", tZONE, HOUR( 9) }, + { "k", tZONE, HOUR( 10) }, + { "l", tZONE, HOUR( 11) }, + { "m", tZONE, HOUR( 12) }, + { "n", tZONE, HOUR(- 1) }, + { "o", tZONE, HOUR(- 2) }, + { "p", tZONE, HOUR(- 3) }, + { "q", tZONE, HOUR(- 4) }, + { "r", tZONE, HOUR(- 5) }, + { "s", tZONE, HOUR(- 6) }, + { "t", tZONE, HOUR(- 7) }, + { "u", tZONE, HOUR(- 8) }, + { "v", tZONE, HOUR(- 9) }, + { "w", tZONE, HOUR(-10) }, + { "x", tZONE, HOUR(-11) }, + { "y", tZONE, HOUR(-12) }, + { "z", tZONE, HOUR( 0) }, + { NULL, 0, 0 } +}; + + + + +/* ARGSUSED */ +static int +yyerror(const char *s __unused) +{ + return 0; +} + + +static time_t +ToSeconds(time_t Hours, time_t Minutes, time_t Seconds, MERIDIAN Meridian) +{ + if (Minutes < 0 || Minutes > 59 || Seconds < 0 || Seconds > 59) + return -1; + switch (Meridian) { + case MER24: + if (Hours < 0 || Hours > 23) + return -1; + return (Hours * 60L + Minutes) * 60L + Seconds; + case MERam: + if (Hours < 1 || Hours > 12) + return -1; + if (Hours == 12) + Hours = 0; + return (Hours * 60L + Minutes) * 60L + Seconds; + case MERpm: + if (Hours < 1 || Hours > 12) + return -1; + if (Hours == 12) + Hours = 0; + return ((Hours + 12) * 60L + Minutes) * 60L + Seconds; + default: + abort (); + } + /* NOTREACHED */ +} + + +/* Year is either + * A negative number, which means to use its absolute value (why?) + * A number from 0 to 99, which means a year from 1900 to 1999, or + * The actual year (>=100). */ +static time_t +Convert(time_t Month, time_t Day, time_t Year, + time_t Hours, time_t Minutes, time_t Seconds, + MERIDIAN Meridian, DSTMODE DSTmode) +{ + static int DaysInMonth[12] = { + 31, 0, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 + }; + time_t tod; + time_t Julian; + int i; + + if (Year < 0) + Year = -Year; + if (Year < 69) + Year += 2000; + else if (Year < 100) + Year += 1900; + DaysInMonth[1] = Year % 4 == 0 && (Year % 100 != 0 || Year % 400 == 0) + ? 29 : 28; + /* Checking for 2038 bogusly assumes that time_t is 32 bits. But + I'm too lazy to try to check for time_t overflow in another way. */ + if (Year < EPOCH || Year > 2038 + || Month < 1 || Month > 12 + /* Lint fluff: "conversion from long may lose accuracy" */ + || Day < 1 || Day > DaysInMonth[(int)--Month]) + return -1; + + for (Julian = Day - 1, i = 0; i < Month; i++) + Julian += DaysInMonth[i]; + for (i = EPOCH; i < Year; i++) + Julian += 365 + (i % 4 == 0); + Julian *= SECSPERDAY; + Julian += yyTimezone * 60L; + if ((tod = ToSeconds(Hours, Minutes, Seconds, Meridian)) < 0) + return -1; + Julian += tod; + if (DSTmode == DSTon + || (DSTmode == DSTmaybe && localtime(&Julian)->tm_isdst)) + Julian -= 60 * 60; + return Julian; +} + + +static time_t +DSTcorrect(time_t Start, time_t Future) +{ + time_t StartDay; + time_t FutureDay; + + StartDay = (localtime(&Start)->tm_hour + 1) % 24; + FutureDay = (localtime(&Future)->tm_hour + 1) % 24; + return (Future - Start) + (StartDay - FutureDay) * 60L * 60L; +} + + +static time_t +RelativeDate(time_t Start, time_t DayOrdinal, time_t DayNumber) +{ + struct tm *tm; + time_t now; + + now = Start; + tm = localtime(&now); + now += SECSPERDAY * ((DayNumber - tm->tm_wday + 7) % 7); + now += 7 * SECSPERDAY * (DayOrdinal <= 0 ? DayOrdinal : DayOrdinal - 1); + return DSTcorrect(Start, now); +} + + +static time_t +RelativeMonth(time_t Start, time_t RelMonth) +{ + struct tm *tm; + time_t Month; + time_t Year; + + if (RelMonth == 0) + return 0; + tm = localtime(&Start); + Month = 12 * (tm->tm_year + 1900) + tm->tm_mon + RelMonth; + Year = Month / 12; + Month = Month % 12 + 1; + return DSTcorrect(Start, + Convert(Month, (time_t)tm->tm_mday, Year, + (time_t)tm->tm_hour, (time_t)tm->tm_min, (time_t)tm->tm_sec, + MER24, DSTmaybe)); +} + + +static int +LookupWord(char *buff) +{ + char *p; + char *q; + const TABLE *tp; + int i; + int abbrev; + + /* Make it lowercase. */ + for (p = buff; *p; p++) + if (isupper(*p)) + *p = tolower(*p); + + if (strcmp(buff, "am") == 0 || strcmp(buff, "a.m.") == 0) { + yylval.Meridian = MERam; + return tMERIDIAN; + } + if (strcmp(buff, "pm") == 0 || strcmp(buff, "p.m.") == 0) { + yylval.Meridian = MERpm; + return tMERIDIAN; + } + + /* See if we have an abbreviation for a month. */ + if (strlen(buff) == 3) + abbrev = 1; + else if (strlen(buff) == 4 && buff[3] == '.') { + abbrev = 1; + buff[3] = '\0'; + } + else + abbrev = 0; + + for (tp = MonthDayTable; tp->name; tp++) { + if (abbrev) { + if (strncmp(buff, tp->name, 3) == 0) { + yylval.Number = tp->value; + return tp->type; + } + } + else if (strcmp(buff, tp->name) == 0) { + yylval.Number = tp->value; + return tp->type; + } + } + + for (tp = TimezoneTable; tp->name; tp++) + if (strcmp(buff, tp->name) == 0) { + yylval.Number = tp->value; + return tp->type; + } + + if (strcmp(buff, "dst") == 0) + return tDST; + + for (tp = UnitsTable; tp->name; tp++) + if (strcmp(buff, tp->name) == 0) { + yylval.Number = tp->value; + return tp->type; + } + + /* Strip off any plural and try the units table again. */ + i = strlen(buff) - 1; + if (buff[i] == 's') { + buff[i] = '\0'; + for (tp = UnitsTable; tp->name; tp++) + if (strcmp(buff, tp->name) == 0) { + yylval.Number = tp->value; + return tp->type; + } + buff[i] = 's'; /* Put back for "this" in OtherTable. */ + } + + for (tp = OtherTable; tp->name; tp++) + if (strcmp(buff, tp->name) == 0) { + yylval.Number = tp->value; + return tp->type; + } + + /* Military timezones. */ + if (buff[1] == '\0' && isalpha(*buff)) { + for (tp = MilitaryTable; tp->name; tp++) + if (strcmp(buff, tp->name) == 0) { + yylval.Number = tp->value; + return tp->type; + } + } + + /* Drop out any periods and try the timezone table again. */ + for (i = 0, p = q = buff; *q; q++) + if (*q != '.') + *p++ = *q; + else + i++; + *p = '\0'; + if (i) + for (tp = TimezoneTable; tp->name; tp++) + if (strcmp(buff, tp->name) == 0) { + yylval.Number = tp->value; + return tp->type; + } + + return tID; +} + + +static int +yylex(void) +{ + char c; + char *p; + char buff[20]; + int Count; + int sign; + + for ( ; ; ) { + while (isspace(*yyInput)) + yyInput++; + + if (isdigit(c = *yyInput) || c == '-' || c == '+') { + if (c == '-' || c == '+') { + sign = c == '-' ? -1 : 1; + if (!isdigit(*++yyInput)) + /* skip the '-' sign */ + continue; + } + else + sign = 0; + for (yylval.Number = 0; isdigit(c = *yyInput++); ) + yylval.Number = 10 * yylval.Number + c - '0'; + yyInput--; + if (sign < 0) + yylval.Number = -yylval.Number; + return sign ? tSNUMBER : tUNUMBER; + } + if (isalpha(c)) { + for (p = buff; isalpha(c = *yyInput++) || c == '.'; ) + if (p < &buff[sizeof buff - 1]) + *p++ = c; + *p = '\0'; + yyInput--; + return LookupWord(buff); + } + if (c != '(') + return *yyInput++; + Count = 0; + do { + c = *yyInput++; + if (c == '\0') + return c; + if (c == '(') + Count++; + else if (c == ')') + Count--; + } while (Count > 0); + } +} + +#define TM_YEAR_ORIGIN 1900 + +/* Yield A - B, measured in seconds. */ +static long +difftm (struct tm *a, struct tm *b) +{ + int ay = a->tm_year + (TM_YEAR_ORIGIN - 1); + int by = b->tm_year + (TM_YEAR_ORIGIN - 1); + int days = ( + /* difference in day of year */ + a->tm_yday - b->tm_yday + /* + intervening leap days */ + + ((ay >> 2) - (by >> 2)) + - (ay/100 - by/100) + + ((ay/100 >> 2) - (by/100 >> 2)) + /* + difference in years * 365 */ + + (long)(ay-by) * 365 + ); + return (60*(60*(24*days + (a->tm_hour - b->tm_hour)) + + (a->tm_min - b->tm_min)) + + (a->tm_sec - b->tm_sec)); +} + +time_t +get_date(char *p) +{ + struct tm *tm, *gmt_ptr, gmt; + int tzoff; + time_t Start; + time_t tod; + time_t nowtime; + + bzero (&gmt, sizeof(struct tm)); + yyInput = p; + + (void)time (&nowtime); + + gmt_ptr = gmtime (&nowtime); + if (gmt_ptr != NULL) + { + /* Make a copy, in case localtime modifies *tm (I think + that comment now applies to *gmt_ptr, but I am too + lazy to dig into how gmtime and locatime allocate the + structures they return pointers to). */ + gmt = *gmt_ptr; + } + + if (! (tm = localtime (&nowtime))) + return -1; + + if (gmt_ptr != NULL) + tzoff = difftm (&gmt, tm) / 60; + else + /* We are on a system like VMS, where the system clock is + in local time and the system has no concept of timezones. + Hopefully we can fake this out (for the case in which the + user specifies no timezone) by just saying the timezone + is zero. */ + tzoff = 0; + + if(tm->tm_isdst) + tzoff += 60; + + tm = localtime(&nowtime); + yyYear = tm->tm_year + 1900; + yyMonth = tm->tm_mon + 1; + yyDay = tm->tm_mday; + yyTimezone = tzoff; + yyDSTmode = DSTmaybe; + yyHour = 0; + yyMinutes = 0; + yySeconds = 0; + yyMeridian = MER24; + yyRelSeconds = 0; + yyRelMonth = 0; + yyHaveDate = 0; + yyHaveDay = 0; + yyHaveRel = 0; + yyHaveTime = 0; + yyHaveZone = 0; + + if (yyparse() + || yyHaveTime > 1 || yyHaveZone > 1 || yyHaveDate > 1 || yyHaveDay > 1) + return -1; + + if (yyHaveDate || yyHaveTime || yyHaveDay) { + Start = Convert(yyMonth, yyDay, yyYear, yyHour, yyMinutes, yySeconds, + yyMeridian, yyDSTmode); + if (Start < 0) + return -1; + } + else { + Start = nowtime; + if (!yyHaveRel) + Start -= ((tm->tm_hour * 60L + tm->tm_min) * 60L) + tm->tm_sec; + } + + Start += yyRelSeconds; + Start += RelativeMonth(Start, yyRelMonth); + + if (yyHaveDay && !yyHaveDate) { + tod = RelativeDate(Start, yyDayOrdinal, yyDayNumber); + Start += tod; + } + + /* Have to do *something* with a legitimate -1 so it's distinguishable + * from the error return value. (Alternately could set errno on error.) */ + return Start == -1 ? 0 : Start; +} + + +#if defined(TEST) + +/* ARGSUSED */ +int +main(int ac, char *av[]) +{ + char buff[128]; + time_t d; + + (void)printf("Enter date, or blank line to exit.\n\t> "); + (void)fflush(stdout); + while (gets(buff) && buff[0]) { + d = get_date(buff); + if (d == -1) + (void)printf("Bad format - couldn't convert.\n"); + else + (void)printf("%s", ctime(&d)); + (void)printf("\t> "); + (void)fflush(stdout); + } + exit(0); + /* NOTREACHED */ +} +#endif /* defined(TEST) */ diff --git a/shell_cmds/find/ls.c b/shell_cmds/find/ls.c new file mode 100644 index 0000000..f58d969 --- /dev/null +++ b/shell_cmds/find/ls.c @@ -0,0 +1,124 @@ +/* + * Copyright (c) 1989, 1993 + * The Regents of the University of California. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#ifndef lint +#if 0 +static char sccsid[] = "@(#)ls.c 8.1 (Berkeley) 6/6/93"; +#endif +#endif /* not lint */ + +#include <sys/cdefs.h> +__FBSDID("$FreeBSD: src/usr.bin/find/ls.c,v 1.23 2011/09/28 18:53:36 ed Exp $"); + +#include <sys/param.h> +#include <sys/stat.h> + +#include <err.h> +#include <errno.h> +#include <fts.h> +#include <grp.h> +#include <inttypes.h> +#include <langinfo.h> +#include <pwd.h> +#include <stdio.h> +#include <string.h> +#include <time.h> +#include <unistd.h> + +#ifdef __APPLE__ +#undef MAXLOGNAME +#define MAXLOGNAME 17 +#endif /* __APPLE__ */ + +#include "find.h" + +/* Derived from the print routines in the ls(1) source code. */ + +static void printlink(char *); +static void printtime(time_t); + +void +printlong(char *name, char *accpath, struct stat *sb) +{ + char modep[15]; + + (void)printf("%6lu %8"PRId64" ", (u_long) sb->st_ino, sb->st_blocks); + (void)strmode(sb->st_mode, modep); + (void)printf("%s %3u %-*s %-*s ", modep, sb->st_nlink, MAXLOGNAME - 1, + user_from_uid(sb->st_uid, 0), MAXLOGNAME - 1, + group_from_gid(sb->st_gid, 0)); + + if (S_ISCHR(sb->st_mode) || S_ISBLK(sb->st_mode)) + (void)printf("%#8jx ", (uintmax_t)sb->st_rdev); + else + (void)printf("%8"PRId64" ", sb->st_size); + printtime(sb->st_mtime); + (void)printf("%s", name); + if (S_ISLNK(sb->st_mode)) + printlink(accpath); + (void)putchar('\n'); +} + +static void +printtime(time_t ftime) +{ + char longstring[80]; + static time_t lnow; + const char *format; + static int d_first = -1; + + if (d_first < 0) + d_first = (*nl_langinfo(D_MD_ORDER) == 'd'); + if (lnow == 0) + lnow = time(NULL); + +#define SIXMONTHS ((365 / 2) * 86400) + if (ftime + SIXMONTHS > lnow && ftime < lnow + SIXMONTHS) + /* mmm dd hh:mm || dd mmm hh:mm */ + format = d_first ? "%e %b %R " : "%b %e %R "; + else + /* mmm dd yyyy || dd mmm yyyy */ + format = d_first ? "%e %b %Y " : "%b %e %Y "; + strftime(longstring, sizeof(longstring), format, localtime(&ftime)); + fputs(longstring, stdout); +} + +static void +printlink(char *name) +{ + int lnklen; + char path[MAXPATHLEN]; + + if ((lnklen = readlink(name, path, MAXPATHLEN - 1)) == -1) { + warn("%s", name); + return; + } + path[lnklen] = '\0'; + (void)printf(" -> %s", path); +} diff --git a/shell_cmds/find/main.c b/shell_cmds/find/main.c new file mode 100644 index 0000000..e95e648 --- /dev/null +++ b/shell_cmds/find/main.c @@ -0,0 +1,166 @@ +/*- + * Copyright (c) 1990, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Cimarron D. Taylor of the University of California, Berkeley. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#ifndef lint +static const char copyright[] = +"@(#) Copyright (c) 1990, 1993, 1994\n\ + The Regents of the University of California. All rights reserved.\n"; +#endif /* not lint */ + +#ifndef lint +#if 0 +static char sccsid[] = "@(#)main.c 8.4 (Berkeley) 5/4/95"; +#endif +#endif /* not lint */ + +#include <sys/cdefs.h> +__FBSDID("$FreeBSD: src/usr.bin/find/main.c,v 1.23 2011/12/10 18:11:06 ed Exp $"); + +#include <sys/types.h> +#include <sys/stat.h> + +#include <err.h> +#include <errno.h> +#include <fcntl.h> +#include <fts.h> +#include <locale.h> +#include <regex.h> +#include <stdio.h> +#include <stdlib.h> +#include <time.h> +#include <unistd.h> + +#include "find.h" + +time_t now; /* time find was run */ +int dotfd; /* starting directory */ +int ftsoptions; /* options for the ftsopen(3) call */ +int isdeprecated; /* using deprecated syntax */ +int isdepth; /* do directories on post-order visit */ +int isoutput; /* user specified output operator */ +int issort; /* do hierarchies in lexicographical order */ +int isxargs; /* don't permit xargs delimiting chars */ +int mindepth = -1, maxdepth = -1; /* minimum and maximum depth */ +int regexp_flags = REG_BASIC; /* use the "basic" regexp by default*/ + +static void usage(void); + +int +main(int argc, char *argv[]) +{ + char **p, **start; + int Hflag, Lflag, ch; + + (void)setlocale(LC_ALL, ""); + + (void)time(&now); /* initialize the time-of-day */ + + p = start = argv; + Hflag = Lflag = 0; + ftsoptions = FTS_NOSTAT | FTS_PHYSICAL; + while ((ch = getopt(argc, argv, "EHLPXdf:sx")) != -1) + switch (ch) { + case 'E': + regexp_flags |= REG_EXTENDED; + break; + case 'H': + Hflag = 1; + Lflag = 0; + break; + case 'L': + Lflag = 1; + Hflag = 0; + break; + case 'P': + Hflag = Lflag = 0; + break; + case 'X': + isxargs = 1; + break; + case 'd': + isdepth = 1; + break; + case 'f': + *p++ = optarg; + break; + case 's': + issort = 1; + break; + case 'x': + ftsoptions |= FTS_XDEV; + break; + case '?': + default: + usage(); + } + + argc -= optind; + argv += optind; + + if (Hflag) + ftsoptions |= FTS_COMFOLLOW; + if (Lflag) { + ftsoptions &= ~FTS_PHYSICAL; + ftsoptions |= FTS_LOGICAL; + } + + /* + * Find first option to delimit the file list. The first argument + * that starts with a -, or is a ! or a ( must be interpreted as a + * part of the find expression, according to POSIX .2. + */ + for (; *argv != NULL; *p++ = *argv++) { + if (argv[0][0] == '-') + break; + if ((argv[0][0] == '!' || argv[0][0] == '(') && + argv[0][1] == '\0') + break; + } + + if (p == start) + usage(); + *p = NULL; + + if ((dotfd = open(".", O_RDONLY, 0)) < 0) + err(1, "."); + + exit(find_execute(find_formplan(argv), start)); +} + +static void +usage(void) +{ + (void)fprintf(stderr, "%s\n%s\n", +"usage: find [-H | -L | -P] [-EXdsx] [-f path] path ... [expression]", +" find [-H | -L | -P] [-EXdsx] -f path [path ...] [expression]"); + exit(1); +} diff --git a/shell_cmds/find/misc.c b/shell_cmds/find/misc.c new file mode 100644 index 0000000..a165c5e --- /dev/null +++ b/shell_cmds/find/misc.c @@ -0,0 +1,106 @@ +/*- + * Copyright (c) 1990, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Cimarron D. Taylor of the University of California, Berkeley. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#ifndef lint +#if 0 +static char sccsid[] = "@(#)misc.c 8.2 (Berkeley) 4/1/94"; +#else +#endif +#endif /* not lint */ + +#include <sys/cdefs.h> +__FBSDID("$FreeBSD: src/usr.bin/find/misc.c,v 1.13 2010/12/11 08:32:16 joel Exp $"); + +#include <sys/types.h> +#include <sys/stat.h> + +#include <err.h> +#include <errno.h> +#include <fts.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "find.h" + +/* + * brace_subst -- + * Replace occurrences of {} in s1 with s2 and return the result string. + */ +void +brace_subst(char *orig, char **store, char *path, int len) +{ + int plen; + char ch, *p; + + plen = strlen(path); + for (p = *store; (ch = *orig) != '\0'; ++orig) + if (ch == '{' && orig[1] == '}') { + while ((p - *store) + plen > len) + if (!(*store = realloc(*store, len *= 2))) + err(1, NULL); + memmove(p, path, plen); + p += plen; + ++orig; + } else + *p++ = ch; + *p = '\0'; +} + +/* + * queryuser -- + * print a message to standard error and then read input from standard + * input. If the input is an affirmative response (according to the + * current locale) then 1 is returned. + */ +int +queryuser(char *argv[]) +{ + char *p, resp[256]; + + (void)fprintf(stderr, "\"%s", *argv); + while (*++argv) + (void)fprintf(stderr, " %s", *argv); + (void)fprintf(stderr, "\"? "); + (void)fflush(stderr); + + if (fgets(resp, sizeof(resp), stdin) == NULL) + *resp = '\0'; + if ((p = strchr(resp, '\n')) != NULL) + *p = '\0'; + else { + (void)fprintf(stderr, "\n"); + (void)fflush(stderr); + } + + return (rpmatch(resp) == 1); +} diff --git a/shell_cmds/find/operator.c b/shell_cmds/find/operator.c new file mode 100644 index 0000000..1aa851e --- /dev/null +++ b/shell_cmds/find/operator.c @@ -0,0 +1,273 @@ +/*- + * Copyright (c) 1990, 1993 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Cimarron D. Taylor of the University of California, Berkeley. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#ifndef lint +#if 0 +static char sccsid[] = "@(#)operator.c 8.1 (Berkeley) 6/6/93"; +#endif +#endif /* not lint */ + +#include <sys/cdefs.h> +__FBSDID("$FreeBSD: src/usr.bin/find/operator.c,v 1.17 2010/12/11 08:32:16 joel Exp $"); + +#include <sys/types.h> + +#include <err.h> +#include <fts.h> +#include <stdio.h> + +#include "find.h" + +static PLAN *yanknode(PLAN **); +static PLAN *yankexpr(PLAN **); + +/* + * yanknode -- + * destructively removes the top from the plan + */ +static PLAN * +yanknode(PLAN **planp) +{ + PLAN *node; /* top node removed from the plan */ + + if ((node = (*planp)) == NULL) + return (NULL); + (*planp) = (*planp)->next; + node->next = NULL; + return (node); +} + +/* + * yankexpr -- + * Removes one expression from the plan. This is used mainly by + * paren_squish. In comments below, an expression is either a + * simple node or a f_expr node containing a list of simple nodes. + */ +static PLAN * +yankexpr(PLAN **planp) +{ + PLAN *next; /* temp node holding subexpression results */ + PLAN *node; /* pointer to returned node or expression */ + PLAN *tail; /* pointer to tail of subplan */ + PLAN *subplan; /* pointer to head of ( ) expression */ + + /* first pull the top node from the plan */ + if ((node = yanknode(planp)) == NULL) + return (NULL); + + /* + * If the node is an '(' then we recursively slurp up expressions + * until we find its associated ')'. If it's a closing paren we + * just return it and unwind our recursion; all other nodes are + * complete expressions, so just return them. + */ + if (node->execute == f_openparen) + for (tail = subplan = NULL;;) { + if ((next = yankexpr(planp)) == NULL) + errx(1, "(: missing closing ')'"); + /* + * If we find a closing ')' we store the collected + * subplan in our '(' node and convert the node to + * a f_expr. The ')' we found is ignored. Otherwise, + * we just continue to add whatever we get to our + * subplan. + */ + if (next->execute == f_closeparen) { + if (subplan == NULL) + errx(1, "(): empty inner expression"); + node->p_data[0] = subplan; + node->execute = f_expr; + break; + } else { + if (subplan == NULL) + tail = subplan = next; + else { + tail->next = next; + tail = next; + } + tail->next = NULL; + } + } + return (node); +} + +/* + * paren_squish -- + * replaces "parenthesized" plans in our search plan with "expr" nodes. + */ +PLAN * +paren_squish(PLAN *plan) +{ + PLAN *expr; /* pointer to next expression */ + PLAN *tail; /* pointer to tail of result plan */ + PLAN *result; /* pointer to head of result plan */ + + result = tail = NULL; + + /* + * the basic idea is to have yankexpr do all our work and just + * collect its results together. + */ + while ((expr = yankexpr(&plan)) != NULL) { + /* + * if we find an unclaimed ')' it means there is a missing + * '(' someplace. + */ + if (expr->execute == f_closeparen) + errx(1, "): no beginning '('"); + + /* add the expression to our result plan */ + if (result == NULL) + tail = result = expr; + else { + tail->next = expr; + tail = expr; + } + tail->next = NULL; + } + return (result); +} + +/* + * not_squish -- + * compresses "!" expressions in our search plan. + */ +PLAN * +not_squish(PLAN *plan) +{ + PLAN *next; /* next node being processed */ + PLAN *node; /* temporary node used in f_not processing */ + PLAN *tail; /* pointer to tail of result plan */ + PLAN *result; /* pointer to head of result plan */ + + tail = result = NULL; + + while ((next = yanknode(&plan))) { + /* + * if we encounter a ( expression ) then look for nots in + * the expr subplan. + */ + if (next->execute == f_expr) + next->p_data[0] = not_squish(next->p_data[0]); + + /* + * if we encounter a not, then snag the next node and place + * it in the not's subplan. As an optimization we compress + * several not's to zero or one not. + */ + if (next->execute == f_not) { + int notlevel = 1; + + node = yanknode(&plan); + while (node != NULL && node->execute == f_not) { + ++notlevel; + node = yanknode(&plan); + } + if (node == NULL) + errx(1, "!: no following expression"); + if (node->execute == f_or) + errx(1, "!: nothing between ! and -o"); + /* + * If we encounter ! ( expr ) then look for nots in + * the expr subplan. + */ + if (node->execute == f_expr) + node->p_data[0] = not_squish(node->p_data[0]); + if (notlevel % 2 != 1) + next = node; + else + next->p_data[0] = node; + } + + /* add the node to our result plan */ + if (result == NULL) + tail = result = next; + else { + tail->next = next; + tail = next; + } + tail->next = NULL; + } + return (result); +} + +/* + * or_squish -- + * compresses -o expressions in our search plan. + */ +PLAN * +or_squish(PLAN *plan) +{ + PLAN *next; /* next node being processed */ + PLAN *tail; /* pointer to tail of result plan */ + PLAN *result; /* pointer to head of result plan */ + + tail = result = next = NULL; + + while ((next = yanknode(&plan)) != NULL) { + /* + * if we encounter a ( expression ) then look for or's in + * the expr subplan. + */ + if (next->execute == f_expr) + next->p_data[0] = or_squish(next->p_data[0]); + + /* if we encounter a not then look for or's in the subplan */ + if (next->execute == f_not) + next->p_data[0] = or_squish(next->p_data[0]); + + /* + * if we encounter an or, then place our collected plan in the + * or's first subplan and then recursively collect the + * remaining stuff into the second subplan and return the or. + */ + if (next->execute == f_or) { + if (result == NULL) + errx(1, "-o: no expression before -o"); + next->p_data[0] = result; + next->p_data[1] = or_squish(plan); + if (next->p_data[1] == NULL) + errx(1, "-o: no expression after -o"); + return (next); + } + + /* add the node to our result plan */ + if (result == NULL) + tail = result = next; + else { + tail->next = next; + tail = next; + } + tail->next = NULL; + } + return (result); +} diff --git a/shell_cmds/find/option.c b/shell_cmds/find/option.c new file mode 100644 index 0000000..5771da4 --- /dev/null +++ b/shell_cmds/find/option.c @@ -0,0 +1,201 @@ +/*- + * Copyright (c) 1990, 1993, 1994 + * The Regents of the University of California. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Cimarron D. Taylor of the University of California, Berkeley. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#ifndef lint +/* +static char sccsid[] = "@(#)option.c 8.2 (Berkeley) 4/16/94"; +*/ +#endif /* not lint */ + +#include <sys/cdefs.h> +__FBSDID("$FreeBSD: src/usr.bin/find/option.c,v 1.32 2011/05/27 22:14:49 jilles Exp $"); + +#include <sys/types.h> +#include <sys/stat.h> + +#include <err.h> +#include <fts.h> +#include <regex.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "find.h" + +static int typecompare(const void *, const void *); + +/* NB: the following table must be sorted lexically. */ +/* Options listed with C++ comments are in gnu find, but not our find */ +static OPTION const options[] = { + { "!", c_simple, f_not, 0 }, + { "(", c_simple, f_openparen, 0 }, + { ")", c_simple, f_closeparen, 0 }, + { "-Bmin", c_Xmin, f_Xmin, F_TIME_B }, + { "-Bnewer", c_newer, f_newer, F_TIME_B }, + { "-Btime", c_Xtime, f_Xtime, F_TIME_B }, + { "-a", c_and, NULL, 0 }, + { "-acl", c_acl, f_acl, 0 }, + { "-amin", c_Xmin, f_Xmin, F_TIME_A }, + { "-and", c_and, NULL, 0 }, + { "-anewer", c_newer, f_newer, F_TIME_A }, + { "-atime", c_Xtime, f_Xtime, F_TIME_A }, + { "-cmin", c_Xmin, f_Xmin, F_TIME_C }, + { "-cnewer", c_newer, f_newer, F_TIME_C }, + { "-ctime", c_Xtime, f_Xtime, F_TIME_C }, + { "-d", c_depth, f_depth, 0 }, +// -daystart + { "-delete", c_delete, f_delete, 0 }, + { "-depth", c_depth, f_depth, 0 }, + { "-empty", c_empty, f_empty, 0 }, + { "-exec", c_exec, f_exec, 0 }, + { "-execdir", c_exec, f_exec, F_EXECDIR }, + { "-false", c_simple, f_false, 0 }, + { "-flags", c_flags, f_flags, 0 }, +// -fls + { "-follow", c_follow, f_always_true, 0 }, +// -fprint +// -fprint0 +// -fprintf + { "-fstype", c_fstype, f_fstype, 0 }, + { "-gid", c_group, f_group, 0 }, + { "-group", c_group, f_group, 0 }, + { "-ignore_readdir_race",c_simple, f_always_true,0 }, + { "-ilname", c_name, f_name, F_LINK | F_IGNCASE }, + { "-iname", c_name, f_name, F_IGNCASE }, + { "-inum", c_inum, f_inum, 0 }, + { "-ipath", c_name, f_path, F_IGNCASE }, + { "-iregex", c_regex, f_regex, F_IGNCASE }, + { "-iwholename",c_name, f_path, F_IGNCASE }, + { "-links", c_links, f_links, 0 }, + { "-lname", c_name, f_name, F_LINK }, + { "-ls", c_ls, f_ls, 0 }, + { "-maxdepth", c_mXXdepth, f_always_true, F_MAXDEPTH }, + { "-mindepth", c_mXXdepth, f_always_true, 0 }, + { "-mmin", c_Xmin, f_Xmin, 0 }, + { "-mnewer", c_newer, f_newer, 0 }, + { "-mount", c_xdev, f_always_true, 0 }, + { "-mtime", c_Xtime, f_Xtime, 0 }, + { "-name", c_name, f_name, 0 }, + { "-newer", c_newer, f_newer, 0 }, + { "-newerBB", c_newer, f_newer, F_TIME_B | F_TIME2_B }, + { "-newerBa", c_newer, f_newer, F_TIME_B | F_TIME2_A }, + { "-newerBc", c_newer, f_newer, F_TIME_B | F_TIME2_C }, + { "-newerBm", c_newer, f_newer, F_TIME_B }, + { "-newerBt", c_newer, f_newer, F_TIME_B | F_TIME2_T }, + { "-neweraB", c_newer, f_newer, F_TIME_A | F_TIME2_B }, + { "-neweraa", c_newer, f_newer, F_TIME_A | F_TIME2_A }, + { "-newerac", c_newer, f_newer, F_TIME_A | F_TIME2_C }, + { "-neweram", c_newer, f_newer, F_TIME_A }, + { "-newerat", c_newer, f_newer, F_TIME_A | F_TIME2_T }, + { "-newercB", c_newer, f_newer, F_TIME_C | F_TIME2_B }, + { "-newerca", c_newer, f_newer, F_TIME_C | F_TIME2_A }, + { "-newercc", c_newer, f_newer, F_TIME_C | F_TIME2_C }, + { "-newercm", c_newer, f_newer, F_TIME_C }, + { "-newerct", c_newer, f_newer, F_TIME_C | F_TIME2_T }, + { "-newermB", c_newer, f_newer, F_TIME2_B }, + { "-newerma", c_newer, f_newer, F_TIME2_A }, + { "-newermc", c_newer, f_newer, F_TIME2_C }, + { "-newermm", c_newer, f_newer, 0 }, + { "-newermt", c_newer, f_newer, F_TIME2_T }, + { "-nogroup", c_nogroup, f_nogroup, 0 }, + { "-noignore_readdir_race",c_simple, f_always_true,0 }, + { "-noleaf", c_simple, f_always_true, 0 }, + { "-not", c_simple, f_not, 0 }, + { "-nouser", c_nouser, f_nouser, 0 }, + { "-o", c_simple, f_or, 0 }, + { "-ok", c_exec, f_exec, F_NEEDOK }, + { "-okdir", c_exec, f_exec, F_NEEDOK | F_EXECDIR }, + { "-or", c_simple, f_or, 0 }, + { "-path", c_name, f_path, 0 }, + { "-perm", c_perm, f_perm, 0 }, + { "-print", c_print, f_print, 0 }, + { "-print0", c_print, f_print0, 0 }, +// -printf + { "-prune", c_simple, f_prune, 0 }, + { "-quit", c_simple, f_quit, 0 }, + { "-regex", c_regex, f_regex, 0 }, + { "-samefile", c_samefile, f_inum, 0 }, + { "-size", c_size, f_size, 0 }, + { "-true", c_simple, f_always_true, 0 }, + { "-type", c_type, f_type, 0 }, + { "-uid", c_user, f_user, 0 }, + { "-user", c_user, f_user, 0 }, + { "-wholename", c_name, f_path, 0 }, +#ifdef __APPLE__ + { "-xattr", c_simple, f_xattr, 0 }, + { "-xattrname", c_name, f_xattrname, 0 }, +#endif /* __APPLE__ */ + { "-xdev", c_xdev, f_always_true, 0 }, +// -xtype +}; + +/* + * find_create -- + * create a node corresponding to a command line argument. + * + * TODO: + * add create/process function pointers to node, so we can skip + * this switch stuff. + */ +PLAN * +find_create(char ***argvp) +{ + OPTION *p; + PLAN *new; + char **argv; + + argv = *argvp; + + if ((p = lookup_option(*argv)) == NULL) + errx(1, "%s: unknown primary or operator", *argv); + ++argv; + + new = (p->create)(p, &argv); + *argvp = argv; + return (new); +} + +OPTION * +lookup_option(const char *name) +{ + OPTION tmp; + + tmp.name = name; + return ((OPTION *)bsearch(&tmp, options, + sizeof(options)/sizeof(OPTION), sizeof(OPTION), typecompare)); +} + +static int +typecompare(const void *a, const void *b) +{ + return (strcmp(((const OPTION *)a)->name, ((const OPTION *)b)->name)); +} |