2.4. Alphabetical Summary of GNU/Linux Commands  | GNU/Linux programs generally accept --help and --version options. In the interest of brevity, the individual command descriptions omit listing those options. |
|
aspell [options] [files]
aspell is intended to be a drop-in replacement for ispell, but with more functionality. It thus accepts the same options; see ispell for more information. See also spell in the earlier section "Alphabetical Summary of Common Commands." URL: http://aspell.net/ and http://www.gnu.org/software/aspell/.
cdda2wav [options] [output.wav]
Convert Compact Disc Digital Audio (CDDA) to the WAV format. This process is often called "ripping" a CD-ROM, and is generally performed before using an encoder to convert the file to a compressed music format such as OGG or MP3. By default, cdda2wav reads data from the /dev/cdrom device and outputs one WAV file per track. Options Some of the following options use sectors as a unit of measurement. Each sector of data on a CD represents approximately 1/75 second of play time.
-a divider, --divider divider Set rate to 44,100 Hz/divider. To get a list of possible values, use the -R option.
-A drivename, --auxdevice drivename Specify a different drive for ioctl purposes.
-b n, --bits-per-sample n Set the quality of samples to n bits per sample per channel. Possible values are 8, 12, and 16.
-B, --bulk, --alltracks Copy each track into its own file.
-c channels, --channels channels Set recording channels. Use 1 for mono, 2 for stereo, or s for stereo with both channels stopped.
-C endianess, --cdrom-endianess endianess Set the endianess of the input samples to endianess. Possible values are little, big, or guess.
-d amount, --duration amount Set to a number followed by f for frames (sectors) or s for seconds. Set time to zero to record an entire track. For example, to copy two minutes, use 120s.
-D devicename, --device devicename, dev=devicename Specify the device. The device must be able to work with the -I (--interface) setting.
-e, --echo Copy audio data to a sound device rather than to a file.
-E endianess, --output-endianess endianess Set the endianess of the output samples to endianess. Possible values are little or big.
-F, --find-extremes Find extreme amplitudes in samples.
-g, --gui Format the output for parsing by GUI frontend programs.
-G, --find-mono Determine if input samples are in mono.
-h, --help Display version and option summary, and quit.
-H, --no-infofile Do not write an info file or a CDDB file.
-i n, --index n Set the start index to n when recording.
-I ifname, --interface ifname Specify the type of interface. For Linux systems, the most appropriate value is usually cooked_ioctl.
-J, --info-only Do not write data to a file; instead just write information about the disc.
-L mode, --cddb mode Use the CDDB ID to do a cddbp album and track title lookup. The mode parameter directs handling of multiple entries. Use 0 for interactive choice, or 1 which always takes the first entry. Additional variables may be provided for CDDB server name and port number:
cddbp-port=portnum | Contact the CDDB server on port number portnum. | cddbp-server=server | Use CDDB server server. |
-m, --mono Record in mono. Use -s to record in stereo.
-M count, --md5 count Calculate MD5 checksum for count bytes from the beginning of a track.
-n count, --sectors-per-request count Read count sectors in each request.
-N, --no-write Do not write data to a file, just read the tracks. For debugging.
-o n, --offset n Start recording n sectors before the beginning of the first track.
-O, --output-format Choose the output file format. Normal file options are wav, aiff, aifc, au, and sun. You can also use cdr and raw for headerless files dumped into recording devices.
-p n, --set-pitch n Adjust the pitch by n percent when copying data to an audio device.
-paranoia Use the paranoia library instead of cdda2wav's built-in routines for reading.
-P n, --set-overlap n Use n sectors of overlap for jitter correction.
-q, --quiet Quiet mode; the program will not send any data to the screen.
-Q, --silent-SCSI Do not print SCSI command errors. Mainly for use by GUI frontends.
-r n, --rate n Set the sample rate in samples per second. To get a list of possible values, use the -R option.
-R, --dump-rates Output a list of possible sample rates and dividers. This option is typically used alone.
-s, --stereo Record in stereo. Use -m to record in mono.
-scanbus Scan all SCSI busses for all SCSI devices and print the inquiry strings. Use dev=device to specify devices to scan. For example, dev=ATA: for IDE CD Writers using Linux IDE to SCSI emulation.
-S n, --speed-select n Specify the speed at which your system will read the CD-ROM. Set the value to the multiple of normal playback speed given as your CD-ROM drive speed (4, 16, 32, and so forth). Setting the speed lower than the maximum can prevent errors in some cases.
-t tracknumber, --track tracknumber Set start track and, optionally, end track. Separate the track numbers with the + character.
-T, --deemphasize Undo pre-emphasis in the input samples.
-v list, --verbose-level list Print information about the CD. list is a comma-separated list of one or more of the following options:
all | All information | catalog | The media catalog number (MCN) | disable | No information, but do print warnings | indices | Index offsets | sectors | Table of contents in start sector notation | summary | Summary of the recording parameters | titles | Table of contents with track titles (when available) | toc | Table of contents | trackid | All International Standard Recording Codes (ISRC) |
-V, --verbose-SCSI Log SCSI commands to the output. Mainly for debugging.
-w, --wait Wait for a signal before recording anything.
-x, --max Set recording quality (and amount of hard disk usage) to maximum.
Examples For most systems, you should be able to copy a complete CD to a single WAV file with the following command: cdda2wav
To copy a complete CD to a set of WAV files, one per track: cdda2wav -B
Scan for IDE CD Writers: $ cdda2wav -scanbus dev=ATA: scsibus1: 1,0,0 100) 'ASUS ' 'CRW-5224A ' '1.20' Removable CD-ROM 1,1,0 101) * ...
cdparanoia [options] span [outfile]
Similar to cdda2wav, cdparanoia reads Compact Disc audio files as WAV, AIFF, AIFF-C, or raw format files. It uses additional data-verification and sound-improvement algorithms to make the process more reliable, and is used by a number of graphical recording programs as a backend. Options
-a, --output-aifc Output in AIFF-C format.
-B, --batch Split the output into multiple files on track boundaries like cdda2wav. Filenames are prefixed with TRack#.
-c, --force-cdrom-little-endian Force cdparanoia to treat the drive as a little-endian device.
-C, --force-cdrom-big-endian Force cdparanoia to treat the drive as a big-endian device.
-d devicename, --force-cdrom-device devicename Specify a device name to use instead of the first readable CD-ROM available.
-e, --stderr-progress Send all progress messages to standard error instead of standard output; used by wrapper scripts.
-f, --output-aiff Output in AIFF format.
-g device, --force-generic-device device Use with -g to set the generic device separately from that of the CD-ROM device. Useful only on nonstandard SCSI setups.
-h, --help Display options and syntax.
-n count, --force-default-sectors count Do atomic reads of count sectors per read. Not generally useful.
-O count, --sample-offset count Shift sample positions by the given count. This shifts track boundaries for the whole disc. May cause read errors or even lockups on buggy hardware.
-p, --output-raw Output headerless raw data.
-q, --quiet Quiet mode.
-Q, --query Display CD-ROM table of contents and quit.
-r, --output-raw-little-endian Output raw data in little-endian byte order.
-R, --output-raw-big-endian Output raw data in big-endian byte order.
-s, --search-for-drive Search for a drive, even if /dev/cdrom exists.
-S n, --force-read-speed n Set the read speed to n on drives that support it. This is useful if you have a slow drive or are low on memory.
-t n, --toc-offset n Shift the entire disc LBA address by n. The amount is added to the beginning offsets in the TOC. -T is similar.
-T, --toc-bias Compensate for the behavior of some drives whereby the actual track beginning offsets are correctly reported in the TOC, but the beginning of track 1 index 1 is treated as sector 0 for reads. May cause read errors or even lockups on buggy hardware.
-v, --verbose Verbose mode.
-V, --version Print version information and quit.
-w, --output-wav Output in WAV format. This is the default.
-X, --abort-on-skip If a read fails and must be skipped, skip the entire track and delete any partially completed output file.
-Y, --disable-extra-paranoia Use data verification and correction only at read boundaries. Not recommended.
-z [ retries], --never-skip[= retries] If a read fails (for example, due to a scratch in the disc), try again and again. If you specify a number, cdparanoia will try that number of times. If you do not, cdparanoia will retry until it succeeds.
-Z, --disable-paranoia Disable data verification and correction. Causes cdparanoia to behave exactly as cdda2wav would.
Progress Symbols The output during operation includes both smiley faces and more standard progress symbols. They are: :-) | Operation proceeding normally. | :-| | Operation proceeding normally, but with jitter during reads. | :-/ | Read drift. | :-P | Unreported loss of streaming in atomic read operation | 8-| | Repeated read problems in the same place. | :-0 | SCSI/ATAPI transport error (hardware problem not related to the disc itself). | :-( | Scratch detected. | ;-( | Unable to correct problem. | 8-X | Unknown and uncorrectable error. | :^D | Finished. | Blank space | Blank space in the progress indicator means that no corrections were necessary. | - | Jitter correction was required. | + | Read errors. | ! | Errors even after correction; repeated read errors. | e | Corrected transport errors. | V | An uncorrected error or a skipped read. |
The span Argument The cdparanoia command takes exactly one argument, which describes how much of the CD to record. It uses numbers followed by bracketed times to designate track numbers and time within them. For example, the string 1[2:23]-2[5] indicates a recording from the 2-minute and 23-second mark of the first track up to the fifth second of the second track. The time format is demarcated by colons, hours:minutes:seconds:.sectors, with the last item, sectors, preceded by a decimal point (a sector is 1/75 of a second). It's best to put this argument within quotes. If you use the -B option, the span argument is not required.
cdrdao command [options] toc-file
Write all content specified in description file toc-file to a CD-R disk drive in one step. This is called disk-at-once (DAO) mode , as opposed to the more commonly used track-at-once (TAO) mode . DAO mode allows you to change the length of gaps between tracks and define data to be written in these gaps (like hidden bonus tracks or track intros). The toc-file can be created by hand or generated from an existing CD using cdrdao's read-toc command. A cue file, as generated by other audio programs, can be used instead of a TOC file. The file format for TOC files is discussed at length in the cdrdao manpage. URL: http://cdrdao.sourceforge.net/. (cdrdao doesn't come with Fedora Core 3, apparently because cdrecord can also do DAO recording.) Commands The first argument must be a command. Note that not all options are available for all commands.
blank Blank a CD-RW disc.
copy Copy the CD. If you use a single drive, you will be prompted to insert the CD-R after reading. An image file will be created unless you use the --on-the-fly flag and two CD drives.
discid Print out CDDB information for a CD.
disk-info Display information about the CD-R currently in the drive.
msinfo Display multisession information. Useful mostly for wrapper scripts.
read-cd Create a TOC file and read in the audio data on a CD.
read-cddb Check a CDDB server for data about the CD represented by a given TOC file, then write that data to the TOC file as CD-TEXT data.
read-test Check the validity of the audio files described in the TOC file.
read-toc Read from a CD and create a disk image and TOC file that will allow creation of duplicates.
scanbus Scan the system bus for devices.
show-data Print out the data that will be written to the CD-R. Useful for checking byte order.
show-toc Print a summary of the CD to be created.
simulate A dry run: do everything except write the CD.
toc-info Print a summary of the TOC file.
toc-size Print the total number of blocks for the TOC.
unlock Unlock the recorder after a failure. Run this command if you cannot eject the CD after using cdrdao.
write Write the CD.
Options
--blank-mode mode Set the blanking mode for a rewritable disc. The value for mode is either full or minimal.
--buffer-under-run-protection n Use 0 to disable buffer underrun protection, or 1 to enable it. The default is enabled.
--buffers n Set the number of seconds of data to be buffered. Default is 32; set to a higher number if your read source is unreliable or is slower than the CD-R.
--capacity minutes Set the capacity in minutes for --full-burn.
--cddb-directory localpath CDDB data that is fetched will be saved in the directory localpath.
--cddb-servers server,server Enter hosts for servers. Servers may include ports, paths, and proxies; you can list multiple servers separated by spaces or commas.
--cddb-timeout s Set the timeout for CDDB server connections to s seconds.
--datafile filename When used with the read-toc command, specifies the data file placed in the TOC file. When used with read-cd and copy, specifies the name of the image file created.
--device bus,id,logicalunit Set the SCSI address of the CD-R using the bus number, ID number, and logical unit number.
--driver driver-id: option-flags Force cdrdao to use the driver you choose with the driver options named, instead of the driver it autodetects.
--eject Eject the disc when done.
--fast-toc Do not extract the pre-gaps and the index marks.
--force Override warnings and perform the action anyway.
--full-burn Force burning to the outer edge of the disc.
-h Print a help summary for individual commands.
--keepimage Used only with the copy command. Keeps the image file created during the copy process.
--multi Record as a multisession disc.
-n Do not wait 10 seconds before writing the disc.
--on-the-fly Do not create an image file: pipe data directly from source to CD-R.
--overburn If you are using a disc with more storage space than cdrdao detects, use this option to keep writing even when cdrdao thinks you're out of space.
--paranoia-mode n Specifies n, from 0 to 3, for the amount of error correction in the CD read. 0 is none, 3 is full (see cdparanoia for information about error correction). Set error correction to a lower number to increase read speed. The default is 3.
--query-string Just print out the CDDB query.
--read-raw mode Set sub-channel reading mode. Possible values are rw or rw_raw.
--read-subchan Used only with the read-cd command. Write raw data to the image file.
--reload Allow the drive to be opened before writing without interrupting the process. Used with simulation runs.
--save Save current options to the settings file $HOME/.cdrdao.
--session n Used only with the read-toc and read-cd commands when working with multisession CDs. Specifies the number of the session to be processed.
--simulate Don't actually write data.
--source-device bus,id,logicalunit Used only with the copy command. Set the SCSI address of the source device.
--source-driver driver-id: option-flags Used only with the copy command. Set the source device driver and flags.
--speed value Set the write speed to value. The default is the highest available; use a lower value if higher values give poor results.
--swap Swap byte order for all samples.
--tao-source For reading or copying, indicates that the source CD was written in TAO mode.
--tao-source-adjust n Use n link blocks for TAO source CDs. The default is two.
--with-cddb Use CDDB to fetch information about the disc and save it as CD-TEXT data. Used with the copy, read-toc, and read-cd commands.
-write-speed-control n If n is 0, disable writing speed control by the drive. The default is 1, which enables writing speed control.
-v verbose-level Set the amount of information printed to the screen. 0, 1, and 2 are fine for most users; greater numbers are useful for debugging.
cdrecord [general-options] dev=device [track-options] track1 track2 ...
Record data or audio compact discs or DVDs. This program normally requires privileged user access, and has a large number of options and settings. A number of useful examples can be found in the manpage, which is quite extensive. General Options General options go directly after the cdrecord command name. Options affecting the track arguments are placed after the device argument and before the track arguments themselves. Options have two forms: traditional options that start with a hyphen, and "variable" assignments, of the form variable=value. Long named options start with only a single hyphen. The general options are:
-abort Attempt to send an abort sequence to the drive. May be needed if other software has left the drive in an unusable state. cdrecord -reset may be necessary as well.
-atip Display the ATIP (Absolute Time In Pregroove) information for a disc. Only some drives allow you to read this information.
blank= type Erase data from a CD-RW in one of the following ways:
all | Erase all information on the disc. May take a long time. | fast | Perform a quick erase of the disc, erasing only the PMA, TOC, and pregap. | help | Display a possible list of blanking methods. | session | Blank the last session. | TRack | Blank a track. | trtail | Blank the tail of a track only. | unclose | Unclose the last session. | unreserve | Unreserve a track previously marked as reserved. |
-checkdrive Check to see if there are valid drivers for the current drive. Returns 0 if the drive is valid.
cuefile= file Obtain all recording information from file, which is a CDRWIN-compliant CUE sheet file. This option disallows specifying individual track files, and also requires the use of -dao.
-dao, -sao Disk-at-once mode. Works only with MMC drives that support non-raw session-at-once modes.
debug= n, -d Set the debug level to an integer (greater numbers are more verbose), or use multiple -d flags as with the -v and -V flags.
defpregap= n Set the default pre-gap size for all tracks except the first to n. Useful only with TEAC drives for creating TAO disks without the two-second silence between tracks. This option may be removed in future versions.
dev= target Set the device used for writing CD or DVD media. See the section "The device Argument" later in this entry.
driver= name Lets you specify a driver for your system. Suggested for experts only. The special drivers cdr_simul and dvd_simul are used for simulation and profiling tests.
driveropts= optlist Specify a comma-separated list of driver options. To get a list of valid options, use driveropts=help and -checkdrive.
-dummy Perform a dry run, doing all the steps of recording with the laser turned off. This will let you know whether the process is going to work.
-eject Eject disc after recording. Some hardware may need to eject a disc after a dummy recording and before the actual recording.
-fix Close ("fixate") the session, preventing future multisession recordings and allowing the disc to be played in standard audio CD players (some can also play a disc that has not been closed).
-force Override errors if possible. May allow you to blank an otherwise broken CD-RW.
-format Format CD-RW/DVD-RW/DVD+RW media. Currently only implemented for DVD+RW media, which must be formatted before the first use. However, cdrecord detects such media and automatically formats it in this case. This option is thus mainly useful to reformat a DVD+RW disc.
fs= n Set the FIFO buffer size to n, in bytes. You may use k, m, s, or f to specify kilobytes, megabytes, or units of 2048 and 2352 bytes, respectively. The default is 4MB.
gracetime= n Set the grace time before writing to n seconds. A value less than two seconds is ignored.
-ignsize Ignores the known size of the medium. Debugging option, use with extreme care. Implies -overburn.
-immed Experimental feature that sets the SCSI IMMED flag for certain commands. Useful on some systems where the CD/DVD writer and ATAPI hard disc are on the same bus or on SCSI systems that don't use disconnect/reconnect. Use with caution.
-inq Do a drive inquiry, print the resulting information, and exit.
kdebug= n, kd= n Set the kernel's debug notification value to n during SCSI command execution. Works through the scg-driver.
-load Load media and exit. Works with tray-loading mechanisms only.
-lock Load media, lock door, and exit. Works with tray-loading mechanisms only. Possibly useful with the Kodak disc transporter.
mcn= n Set the Media Catalog Number to n.
minbuf= n Experimental feature. Sets the minimum drive buffer fill ratio to n, which is a number between 25 and 95 for 25% to 95% minimum drive buffer fill ratio.
-msinfo Get multisession information from the CD. Used only with multisession discs onto which you can still record more sessions.
-multi Set to record in multisession mode. Must be present on all sessions but the last one for a multisession disc.
-noclose Experimental feature: do not close the current track. Useful only in packet-writing mode.
-nofix Do not close the disc after writing.
-overburn Allow writing of more data than the official size of a medium. Not guaranteed to work on any specific drive.
-packet Experimental feature: use packet-writing mode.
pktsize= n Experimental feature: set the packet size to n. Forces fixed packet mode.
-prcap Print the drive capabilities for SCSI-3/MMC-compliant drives. Values marked kB use 1000 bytes, while values marked KB use 1024 bytes.
-raw, -raw96r Use RAW writing mode with 2352 byte sectors plus 96 bytes of raw P-W subchannel data. This results in a sector size of 2448 bytes. Useful for drives with bad firmware where TAO and SAO mode don't work. This option does require more CPU time, thus it may cause buffer underruns on slow CPUs. Note: for this to work, cdrecord must know the size each track in advance. See the manpage for more information.
-raw16 Similar to -raw96r, but with a 2352 byte sector and 16 bytes of P-Q subchannel data. Does not allow writing CD-Text or CD+Graphics. Don't use if -raw96r does work on your drive.
-raw96p Similar to -raw96r but not as widely supported. Don't use if -raw96r or -raw16 do work on your drive.
-reset Attempt to reset the SCSI bus. Does not work on all systems.
-s, -silent Silent mode. Do not print any SCSI error commands.
-scanbus Scan for SCSI devices. Use this to find out which drives you have and to get the correct numbers for the dev= option.
-setdropts Set the driver options list as provided by driveropts= and the dummy flag, and then exit. Useful to set parameters without burning or reading media.
speed= n Set the speed to n, a multiple of the audio speed. Normally, cdrecord will get this from the CDR_SPEED environment variable. If your drive has trouble with higher numbers, try 0 as the value.
-tao Use Track At Once (TAO) mode. Required for multi-session recording. This was the default writing mode in previous versions.
-text Write CD-Text information. The information comes from either .inf files or from a CUE sheet file. Use with the -useinfo or cuefile= options.
textfile= file Obtain CD-Text information from file, which must be in the binary file format defined in the Red Book. This is the best way to copy CD-Text data obtained from existing CDs.
timeout= n Set the timeout to n seconds. Defaults to 40.
-toc Display the table of contents for the CD currently in the drive. Works for CD-ROM as well as CD-R and CD-RW media.
ts= n Set the maximum transfer size for a single SCSI command to n. The syntax for n is the same as for the fs= option. The default transfer size is 63 kB.
-useinfo Use .inf files to override audio options set elsewhere.
-v Verbose mode. Use one v for each level of verbosity: -vv would be very verbose, and -vvv would be even more so.
-V As with the -v, a verbose mode counter. However, this applies only to SCSI transport messages. This will slow down the application.
-waiti Wait for input to become available on standard input before opening the SCSI driver. This is necessary for multi-session recording, where mkisofs has to read the old session from the current disc before writing the new session, and mkisofs won't be able to open the device if cdrecord has already opened it.
The device Argument The device argument is one of the more difficult parts of cdrecord. It consists of an optional transport specification, followed by a comma-separated list of integers representing the bus, target, and logical unit of the drive. The default transport is for a SCSI device. You can use cdrecord dev=help to see a list of supported transports. On some systems, you can use a device name followed by :@ to use that device; for example, on GNU/Linux, dev=/dev/scd0:@ for an external USB-connected DVD writer. For an IDE-connected CD writer, on GNU/Linux, you might use ATA:1,0,0. Track Options and Arguments Track options may be mixed with track arguments, and normally apply to the track immediately after them or to all tracks after them. The track arguments themselves should be the files that you will be writing to the CD or DVD. Options are:
-audio Write all tracks after this track in digital audio format (playable by standard CD players). If you do not use this flag or the -data flag, cdrecord assumes that .au and .wav files are to be recorded as raw audio and that all other files are data.
-cdi Write subsequent tracks in CDI format.
-copy For subsequent audio tracks, indicate in the TOC that the audio data has permission to be copied without limit.
-data Record subsequent tracks as CD-ROM data. If you do not use this flag or the -audio flag, all files except for those ending in .wav or .au are assumed to be data.
index= a,b,c Set the index list for the next track. The values should be increasing comma-separated integers, starting with index 1 and counting in sectors (75ths of a second). For example, you could set three indices in a track with index=0,750,7500 and they would occur at the beginning of the track, after 10 seconds, and after 100 seconds.
-isosize The size of the next track should match the size of the ISO-9660 filesystem. This is used when duplicating CDs or copying from raw-data filesystems.
isrc= n Set the International Standard Recording Number for the track following this argument.
-mode2 Write all subsequent tracks in CD-ROM mode 2 format. Data size is a multipe of 2336 bytes.
-nocopy For subsequent audio tracks, indicate in the TOC that the audio data has permission to be copied only once for personal use. This is the default.
-nopad Do not insert blank data between data tracks following this flag. This is the default behavior.
-nopreemp For subsequent audio tracks, indicate in the TOC that audio data was mastered with linear data. This is the default.
-noshorttrack Require subsequent tracks to be at least four seconds in length. See -shorttrack.
-pad Insert 15 sectors of blank data padding between data tracks. Applies to all subsequent tracks or until you use the -nopad argument, and is overridden by the padsize=n argument.
padsize= n Insert n sectors of blank data padding after the next track. Applies only to the track immediately after it.
-preemp For subsequent audio tracks, indicate in the TOC that audio data was sampled with 50/15 microsecond pre-emphasis.
pregap= n Set the pre-gap size for the next track to n. Useful only with TEAC drives for creating TAO disks without the two-second silence between tracks. This option may be removed in future versions.
-scms For subsequent audio tracks, indicate in the TOC that the audio data has no permission to be copied.
-shorttrack Allow subsequent tracks to be less than four seconds in length, violating the Red Book standard. Useful only in SAO or RAW mode. Does not work with all drives.
-swab Declare that your data is in byte-swapped (little-endian) byte order. This is not normally necessary.
tsize= n Set the size of the next track. Useful only if you are recording from a raw disk for which cdrecord cannot determine the file size. If you are recording from an ISO 9660 filesystem, use the -isosize flag instead.
-xa Write subsequent tracks in CD-ROM XA mode 2 form 1 format, with 2048-byte sectors.
-xa1 Write subsequent tracks in CD-ROM XA mode 2 form 1 format, with 2056-byte sectors.
-xa2 Write subsequent tracks in CD-ROM XA mode 2 form 2 format, with 2324-byte sectors.
-xamix Write subsequent tracks in a way that allows mixing XA mode 2 forms 1 and 2. See the manpage.
dir [options] [file]
List directory contents. dir is equivalent to the command ls -C -b (list files in columns, sorted vertically, special characters escaped) and it takes the same arguments as ls. This is an alternate invocation of the ls command and is provided for the convenience of those converting from Microsoft Windows and the DOS shell.
dircolors [options] [file]
Set the color options for ls by changing the LS_COLORS environment variable. If you specify a file, dircolors reads it to determine which colors to use. Otherwise, it uses a default set of colors. Options
-b, --sh, --bourne-shell Use the Bourne shell syntax when setting the LS_COLORS variable.
-c, --csh, --c-shell Use csh (C shell) syntax when setting the LS_COLORS variable.
-p, --print-database Display the default colors. You can copy this information into a file and change it to suit your preferences, and then run the program with the file as its argument to set the colors to your new values.
Example In your .profile, you might have this: eval 'dircolors'
dvdrecord options files ...
Record DVDs. In earlier systems, dvdrecord was a modified version of cdrecord. In Fedora Core 3, the DVD functionality is part of cdrecord, and dvdrecord is a shell script that prints an informational message to this effect and then does an exec of cdrecord. See also cdrecord.
gawk [options] 'script' [var=value ...] [files ...] gawk [options] -f scriptfile [var=value ...] [files ...]
The GNU Project's implementation of the awk programming language. This is the standard version of awk on GNU/Linux systems. For more information see awk in the earlier section "Alphabetical Summary of Common Commands," and Chapter 11.
gettextize [options] [directory]
Install GNU gettext infrastructure into a source package. This command copies files and directories into a source package so that a program can use GNU gettext for managing translations. The files are placed in directory if given, otherwise in the current directory. If the package already uses gettext, the infrastructure is upgraded to the current version. Full documentation for GNU gettext is available in its manual. See http://www.gnu.org/software/gettext/ for more information and a pointer to the documentation. Options
-c, --copy Copy files instead of making symbolic links. (Recommended.)
-f, --force Force creation of files, even if old ones exist. Useful for upgrading.
--intl Create and install the libintl subdirectory, which holds a private copy of the gettext library.
-n, --dry-run Print the changes that would be made, but don't actually do them.
--no-changelog Do not create or update ChangeLog files.
igawk gawk-options files ...
A shell script that allows the use of file inclusion with awk programs. Distributed with GNU Awk (gawk). For more information, see Chapter 11.
ispell [options] [files]
Compare the words of one or more named files with the system dictionary. Display unrecognized words at the top of the screen, accompanied by possible correct spellings, and allow editing via a series of commands. See also aspell and spell in the earlier section "Alphabetical Summary of Common Commands." URL: http://www.gnu.org/software/ispell. Options The -c, -D, and -e1, -e2, -e3, and -e4 options are specialized (for use by the munchlist helper program) and are not covered here.
-a Function in back-end mode, printing a one-line version identification and then one line of output for each input word. See the manpage.
-A Like -A, but read files named on lines beginning with &Include_File&. Includes may be nested up to five levels deep.
-b Back up original file in filename.bak.
-B Search for missing blanks (resulting in concatenated words) in addition to ordinary misspellings.
-C Do not produce error messages in response to concatenated words.
-d file Search file instead of the standard dictionary file.
-f outfile Write output to outfile instead of to standard output. Must be used with -a or -A.
-F program Use program to remove formatting markup.
-H Input is in SGML/HTML format.
-ksetname list Add the keywords in list to the predefined set of keywords setname. See the manpage for details.
-l Generate a list of misspelled words (batch mode).
-L number Show number lines of context.
-m Suggest different root/affix combinations.
-M List interactive commands at bottom of screen.
-n Expect nroff or troff input file.
-N Suppress printing of interactive commands.
-p file Search file instead of personal dictionary file.
-P Do not attempt to suggest more root/affix combinations.
-S Sort suggested replacements by likelihood that they are correct.
-t Expect TEX or LATEX input file.
-T type Expect all files to be formatted by type.
-v, -vv With plain -v, print version information and exit. With -vv, also print compilation options.
-V Use hat notation (^L) to display control characters, and M- to display characters with the high bit set.
-w chars Consider chars to be legal, in addition to a-z and A-Z.
-W n Never consider words that are n characters or fewer to be misspelled.
-x Do not back up the original file.
Interactive commands
? Display help screen.
space Accept the word in this instance.
number Replace with suggested word that corresponds to number.
! command Invoke shell and execute command in it. Prompt before exiting.
a Accept word as correctly spelled, but do not add it to personal dictionary.
i Accept word and add it (with any current capitalization) to personal dictionary.
l Search system dictionary for words.
q Exit without saving.
r Replace word.
u Accept word and add lowercase version of it to personal dictionary.
x Skip to the next file, saving changes.
^L Redraw screen.
^Z Suspend ispell.
ltrace [options] command [arguments]
Trace the dynamic library calls for command and arguments. ltrace can also trace and print the system calls. It is very similar to strace (see also strace). Options
-a n Align the return values in column n.
-c Count all calls and signals and create a summary report when the program has ended.
-C, --demangle Demangle C++ encoded names.
-d, --debug Debug mode. Print debugging information for ltrace on standard error.
-e [ keyword= ][! ]values Pass an expression to ltrace to limit the types of calls or signals that are traced or to change how they are displayed. See strace for the full list.
-f Trace forked processes.
-h, --help Print help and exit.
-i Print instruction pointer with each system call.
-l file, --library file Print only the symbols from library file. Up to 20 files may be specified.
-L Do not display library calls. Use together with -S.
-n count, --indent count Indent trace output by count spaces for each new nested function call.
-o filename, --output filename Write output to filename instead of standard error. If filename starts with the pipe symbol |, treat the rest of the name as a command to which output should be piped.
-p pid Attach to the given process ID and begin tracking. ltrace can track more than one process if more than one -p option is given. Type CTRL-C to end the trace.
-r Relative timestamp. Print time in microseconds between system calls.
-s n Print only the first n characters of a string. Default value is 32.
-S Display system calls and library calls.
-t Print time of day on each line of output.
-tt Print time of day with microseconds on each line of output.
-ttt Print timestamp on each line as number of seconds since the Epoch.
-T Print time spent in each system call.
-u username Run command as username. Needed when tracing setuid and setgid programs.
-V Print version and exit.
lynx [options] [path | URL]
lynx is a "text mode" browser, for use on ASCII terminals or terminal emulators. It is particularly valuable for taking a quick look at a web page when you don't want to wait for a graphical browser to start up on a heavily loaded system, and for use in scripts. It has an astonishing number of options. URL: http://lynx.isc.org/. Primary Options
- Read arguments from standard input. Useful for long command lines and to avoid having sensitive arguments being visible with ps.
-base For use with -source, prepend a request URL and BASE tag to text or HTML outputs.
-case Make string searching case-sensitive.
-color For use with the slang terminal library. Enable a default set of color control sequences that work on many terminals if the terminal description does not provide color control information.
-crawl When used with -traversal, output each page to a separate file. When used with -dump, output is formatted as for -traversal but sent to standard output.
-dump Send the formatted output to standard output. Useful for converting web pages to text files.
-editor=program Use program as the external editor.
-emacskeys Enable Emacs-style motion commands.
-ftp Disallow FTP access.
-justify Do text justification.
-source Like -dump, but outputs HTML source, not formatted text.
-telnet Do not allow telnet commands.
-term=termtype Specify that the terminal is of type termtype. Especially useful for remote connections.
-traversal Traverse all the HTTP links derived from the starting file or URL.
-use_mouse Enable mouse actions if the underlying library supports the mouse. The left mouse button traverses a link and the right button pops back. Clicking on the top and bottom lines scroll up and down. For the ncurses library, the middle button pops up a simple menu.
-vikeys Enable vi-style motion commands.
-width=count Use count columns for formatting dumps. The default is 80.
-with_backspaces For -dump and -crawl, place backspaces in the output (similar to the man command).
See the manpage for a description of the other options.
mac2unix [options] [-n infile outfile]
Convert files from Macintosh OS 9 format to Unix format. On GNU/Linux, the dos2unix and unix2dos commands accept the same options as mac2unix. See dos2unix in the earlier section "Alphabetical Summary of Common Commands" for the option list.
md5sum [options] [file ...] md5sum [options] --check [file]
Compute or check the MD5 algorithm checksum for one or more files. The checksum is computed using the algorithm in RFC 1321. Use the saved output of the program as input when checking. See also sha1sum. Options
-b, --binary Use binary mode to read files. This is the default on non-Unix systems.
-c, --check Check MD5 sums against the given list.
--status Do not print anything. Instead, use the exit code to indicate success.
-t, --text Read files in "text" mode. This is the default on Unix systems.
-w, --warn Warn about checksum lines that are not formatted correctly.
Examples $ md5sum bash-3.0.tar.gz > MD5SUM Compute and save checksum $ cat MD5SUM Show it 26c4d642e29b3533d8d754995bc277b3 bash-3.0.tar.gz $ md5sum --check < MD5SUM Verify it bash-3.0.tar.gz: OK
mount [options] [[device] directory]
System administration command. Mount a file structure. The file structure on device is mounted on directory. If no device is specified, mount looks for an entry in /etc/fstab to find what device is associated with the given directory. The directory, which must already exist and should be empty, becomes the name of the root of the newly mounted file structure. If mount is invoked with no arguments, it displays the name of each mounted device, the directory on which it is mounted, its filesystem type, and any mount options associated with the device. Options
-a Mount all filesystems listed in /etc/fstab. Use -t to limit this to all filesystems of a particular type.
-f Fake mount. Go through the motions of checking the device and directory, but do not actually mount the filesystem.
-F When used with -a, fork a new process to mount each filesystem.
-h Print help message, then exit.
-i For a filesystem type fs, don't run the helper program /sbin/mount.fs.
-l When reporting on mounted filesystems, show filesystem labels for filesystems that have them.
-L label Mount filesystem with the specified label.
-n Do not record the mount in /etc/mtab.
-o option Qualify the mount with a mount option. Many filesystem types have their own options. The following are common to most filesystems:
async Do input and output to the device asynchronously.
atime, noatime Update inode access time for each access. This is the default behavior. noatime does not update the access time.
auto, noauto Allow (do not allow) mounting with the -a option.
defaults Use all options' default values (async, auto, dev, exec, nouser, rw, suid).
dev, nodev The dev option allows the system to interpret any special devices that exist on the filesystem as device files. The nodev option disallows it; device files are ignored.
dirsync All directory updates on the filesystem should be done synchronously.
exec, noexec The exec option allows the system to execute binary files on the filesystem. The noexec option disallows it.
_netdev Filesystem is a network device requiring network access.
remount Expect the filesystem to have already been mounted, and remount it.
ro Allow read-only access to the filesystem.
rw Allow read/write access to the filesystem.
suid, nosuid Acknowledge (do not acknowledge) setuid and setgid bits.
user, nouser Allow (do not allow) unprivileged users to mount or unmount the filesystem. The defaults on such a filesystem will be nodev, noexec, and nosuid, unless otherwise specified.
users Allow any user to mount or unmount the filesystem. The defaults on such a filesystem will be nodev, noexec, and nosuid, unless otherwise specified.
-O option Limit systems mounted with -a by its filesystem options. (As used with -o.) Use a comma-separated list to specify more than one option, and prefix an option with no to exclude filesystems with that option. Options -t and -O are cumulative.
-p fd For an encrypted filesystem, read the passphrase from file descriptor number fd.
-r Mount filesystem read-only.
-s Where possible, ignore mount options specified by -o that are not supported by the filesystem.
-t type Specify the filesystem type. Possible values include adfs, affs, autofs, coda, cramfs, devpts, efs, ext2, ext3, hfs, hpfs, iso9660, jfs, msdos, ncpfs, nfs, nfs4 ntfs, proc, qnx4, ramfs, reiserfs, romfs, smbfs, sysv, tmpfs, udf, ufs, umsdos, usbfs, vfat, and xfs. The default type is iso9660. The type auto may also be used to have mount autodetect the filesystem. When used with -a, this option can limit the types mounted. Use a comma-separated list to specify more than one type to mount, and prefix a type with no to exclude that type.
-U uuid Mount filesystem with the specified uuid.
-v Display mount information verbosely.
-V Print version, then exit.
-w Mount filesystem read/write. This is the default.
Files
/etc/fstab List of filesystems to be mounted and options to use when mounting them.
/etc/mtab List of filesystems currently mounted and the options with which they were mounted.
/proc/partitions Used to find filesystems by label and UUID.
mutt [reading-options] mutt [sending-options] address ...
mutt is a screen-oriented Mail User Agent (MUA) program, for reading and sending mail. Its design is derived from that of several earlier MUA programs, including Berkeley Mail, ELM, and MUSH: hence the name. URL: http://www.mutt.org/. Options
-a file Attach file to the message using MIME.
-b address Send a blind carbon copy (BCC) to address.
-c address Send a carbon copy (CC) to address.
-e command Run command after reading the configuration files.
-f file Use file as the mailbox to read messages from.
-F muttrcfile Read muttrcfile for initialization, instead of $HOME/.muttrc.
-h Print an option summary and exit.
-H draftfile Read an initial message header and body from draftfile when sending a message.
-i file Include the contents of file into a message.
-m type Use mailbox type type. Possible values are mbox, MMDF, MH, and Maildir.
-n Do not read the sytsem configuration file /etc/Muttrc.
-p Continue a postponed message.
-R Process a mailbox in read-only mode.
-s text Use text as the subject of the message.
-v Print version and compile-time option information, and exit.
-x Emulate the mailx compose mode.
-y At startup, list all mailboxes specified with the mailboxes command.
-z When used with -f, do not start if the mailbox is empty.
-Z Process the first mailbox specified with the mailboxes command that contains new mail. If there are none, exit immediately.
ooffice [files]
The Open Office office productivity suite . A set of commands that provides compatibility with other widely used office productivity programs. URL: http://www.openoffice.org/. Tools
Calc A "feature-packed" spreadsheet program with built-in charting tools.
Database tools Tools for doing database work in spreadsheet-like form. The tools support dBASE databases for simple applications, or OODBC or JDBC compliant databases for "industrial strength" work.
Draw A program for producing illustrations.
Impress A program for creating multimedia presentations.
Math A component for use with Writer in creating equations and formulae, it may also be used standalone.
Writer A word processor for documents, reports, newsletters, brochures, etc.
pdksh [options] [arguments ...]
The Public Domain Korn Shell. pdksh is a fairly complete although not exact clone of the 1988 Korn shell. Full details are available in the online manpage and at the web site. For more information about the Korn shell, see Chapter 4. URL: http://web.cs.mun.ca/~michael/pdksh/.
ps [options]
Report on active processes. ps has three types of options. GNU long options start with two dashes, which are required. BSD options may be grouped and do not start with a dash, while Unix98 options may be grouped and require an initial dash. The meaning of the short options can vary depending on whether or not there is a dash. In options, list arguments should either be separated by commas or put in double quotes. In comparing the amount of output produced, note that e prints more than a and l prints more than f for each entry. Options
nums, p nums, -p nums, --pid=nums Include only specified processes, which are given in a space-delimited list.
--nums, --sid=nums Include only specified session IDs, which are given in a space-delimited list.
[-]a As a, list all processes on a terminal. As -a, list all processes on a terminal except session leaders.
[-]c As c, show the true command name. As -c, show different scheduler information with -l.
-C cmds Select by command name.
--cols=cols, --columns=cols, --width=cols Set the output width (the number of columns to display).
-d Select all processes except session leaders.
-e, -A Select all processes.
e Include environment information after the command.
[-]f, -F, --forest As -f, display full listing. As f or --forest, display "forest" family tree format, with ASCII art showing the relationships.
-g list, -G list, --group=groups, --Group=groups For -g, select by session leader if list contains numbers, or by group if it contains group names. For -G, select by the group IDs in list. --group selects by effective group and --Group selects by real group, where groups can be either group names or group IDs.
h, --no-headers Suppress header. If you select a BSD personality by setting the environment variable PS_PERSONALITY to bsd, then h prints a header on each page.
H, -H As H, display threads as if they were processes. As -H, display "forest" family tree format, without ASCII art.
--headers Repeat headers.
--info Print debugging information.
[-]j Jobs format. j prints more information than -j.
[-]l Produce a long listing. -l prints more information than l.
L, -L As L, print list of field specifiers that can be used for output formatting or for sorting. As -L, show threads, possibly with LWP and NLWP columns.
--lines=num, --rows=num Set the screen height to num lines. If --headers is also set, the headers repeat every num lines.
[-]m Show threads.
-M Add security data for SELinux.
n Print user IDs and WCHAN numerically.
-nfile, N file Specify the system map file for ps to use as a namelist file. The map file must correspond to the Linux kernel; e.g., /boot/System.map-2.6.9-1.667.
-N, --deselect Negate the selection.
[- ]o fields, --format= fields As -o, o, or --format, specify user-defined format with a list of fields to display.
[- ]O fields As -O, is like -o, but some common fields are predefined. As O, can be either the same as -O in specifying fields to display, or can specify single-letter fields for sorting. For sorting, each field specified as a key can optionally have a leading + (return to default sort direction on key) or - (reverse the default direction).
--ppid pids Include only processes whose parent process IDs are in pids.
r Show only processes that are currently running.
s Display signal format.
-s sessions Show processes belonging to the specified sessions.
S, --cumulative As S, sum up certain information, such as CPU usage from dead children. Otherwise, include some dead child process data in parent total.
--sort sort-spec Like O for sorting.
[- ]t[ ttys], --tty= ttys Display processes running on the specified terminals. As t, ttys may be missing, which specifies the current terminal.
T Display all processes on this terminal.
[- ]u [ users], --user= users As u with no argument, display user-oriented output. As -u or --users, display by effective user ID (and also support names), showing results for users. With no argument, -u displays results for the current user.
[- ]U users, --User= users As U, display processes for the specified users. As -U or --User, display processes for users by real user ID (and also support names).
v Display virtual memory format.
[-]V, --version Display version information and then exit.
[-]w Wide format. Don't truncate long lines.
x Display processes without an associated terminal.
X Use old Linux i386 register format.
-y Do not show flags; show rss instead of addr.
[-]Z, --context As Z, add security data for SELinux. As -Z, display the security context format, also for SELinux.
Sort keys c, cmd | Name of executable. | C, pcpu | CPU utilization. | f, flags | Flags. | g, pgrp | Group ID of process. | G, tpgid | Group ID of associated tty. | j, cutime | Cumulative user time. | J, cstime | Cumulative system time. | k, utime | User time. | m, min_flt | Number of minor page faults. | M, maj_flt | Number of major page faults. | n, cmin_flt | Total minor page faults. | N, cmaj_flt | Total major page faults. | o, session | Session ID. | p, pid | Process ID. | P, ppid | Parent's process ID. | r, rss | Resident set size. | R, resident | Resident pages. | s, size | Kilobytes of memory used. | S, share | Number of shared pages. | t, tty | Terminal. | T, start_time | Process's start time. | u, user | User's name. | U, uid | User ID. | v, vsize | Bytes of virtual memory used. | y, priority | Kernel's scheduling priority. |
Fields %CPU | Percent of CPU time used recently. | %MEM | Percent of memory used. | ADDR | Address of the process. | BLOCKED | Mask of blocked signals. | C | CPU utilization percentage. | CAUGHT | Mask of caught signals. | CLS, POL | Process scheduling class. | CMD, COMMAND | The command the process is running. | CP | Per-mill CPU usage. | EGID, GID | Effective group ID as a decimal integer. | EGROUP, GROUP | Effective group ID as a name if available, otherwise as a number. | EIP | Instruction pointer. | ELAPSED | Elapsed time since the start of the process. | ESP | Stack pointer. | EUID, UID | Effective user ID as a decimal integer. | EUSER, USER | Effective user ID as a name if available, otherwise as a number. | F | Process flags: | | | 001 Print alignment warning messages. | | | 002 Being created. | | | 004 Being shut down. | | | 010 ptrace(0) has been called. | | 020 TRacing system calls. | | | 040 Forked but didn't exec. | | | 100 Used superuser privileges. | | | 200 Dumped core. | | | 400 Killed by a signal. | FGID | Filesystem access group ID as a decimal integer. | FGROUP | Filesystem access group ID as a name if available, otherwise as a number. | FUID | Filesystem access user ID as a decimal integer. | FUSER | Filesystem access user ID as a name if available, otherwise as a number. | IGNORED | Mask of ignored signals. | LABEL | SELinux security label for Mandatory Access Control. | LWP, SPID, TID | Lightweight process (thread) ID. | NI | The nice value of the process. A higher number indicates less CPU priority. | NLWP, THCNT | Number of lightweight processes (threads) in the process. | PENDING | Mask of pending signals. | PGID, PGRP | Process group ID. | PID | Process ID. | PPID | Parent process ID. | PRI | Process's scheduling priority. A higher number indicates lower priority. | P, PSR | Processor that the process is assigned to. | RGID | Real group ID as a decimal integer. | RGROUP | Real group ID as a name if available, otherwise as a number. | RSS, RSZ | Resident set size (the amount of physical memory), in kilobytes. | RTPRIO | Realtime priority. | RUID | Real user ID as a decimal integer. | RUSER | Real user ID as a name if available, otherwise as a number. | S | One-character state display. | SCH | Process's scheduling policy. | SESS, SID | Process's session ID. | SGID, SVGID | Saved group ID as a decimal integer. | SGROUP | Saved group ID as a name if available, otherwise as a number. | SHARE | Shared memory. | SIZE | Size of virtual image. | STACKP | Address of the bottom (start) of the process's stack. | START | Process start time in HH:MM format. | STARTED | Process start time in HH:MM:SS format. | STAT | Status: | | | D Asleep and not interruptible. | | | N Positive nice value (third field). | | | R Runnable. | | | S Asleep. | | | T Stopped. | | | W No resident pages (second field). | | | Z Zombie. | SUID, SVUID | Saved user ID as a decimal integer. | SUSER | Saved user ID as a name if available, otherwise as a number. | SWAP | Amount of swap used, in kilobytes. | SZ | Approximate amount of swap needed to write out entire process. | TIME | Cumulative CPU time. | TPGID | Foreground process group ID for terminal. | TRS | Size of resident text. | TT, TTY | Associated terminal. | VSZ | Virtual memory size, in kilobytes. | WCHAN | Kernel function in which process resides. |
rename from to files ...
Rename files by replacing the first occurrence of from in each filename with to. Example Rename files that start with test so they start with mytest: rename test mytest test*
seq [options] [first [increment]] last
Print the numbers from first through last by increment. The default is to print one number per line to standard output. Both first and increment can be omitted and default to 1, but if first is omitted then increment must also be omitted. In other words, if only two numbers are specified, they are taken to be the first and last numbers. The numbers are treated as floating point. Options
-f format, --format=format Write the output using the specified printf floating-point format, which can be one of %e, %f, or %g (the default).
-s string, --separator=string Use string to separate numbers in the output. Default is newline.
-w, --equal-width Equalize the width of the numbers by padding with leading zeros. (Use -f for other types of padding.)
sha1sum [options] [file ...] sha1sum [options] --check [file]
Compute or check SHA1 160-bit checksums for one or more files. The checksum is computed using the algorithm in FIPS-180-1. Use the saved output of the program as input when checking. See also md5sum. Options
-b, --binary Use binary mode to read files. This is the default on non-Unix systems.
-c, --check Check SHA1 sums against the given list.
--status Do not print anything. Instead, use the exit code to indicate success.
-t, --text Read files in "text" mode. This is the default on Unix systems.
-w, --warn Warn about checksum lines that are not formatted correctly.
Examples $ sha1sum bash-3.0.tar.gz > SHA1SUM Compute and save checksum $ cat SHA1SUM Show it 3acf1ff4910d4bc863620c7533cbf4858370017b bash-3.0.tar.gz $ sha1sum --check < SHA1SUM Verify it bash-3.0.tar.gz: OK
shred [options] files
Overwrite a file to make the contents unrecoverable, and delete the file afterwards if requested. Options
- Shred standard output.
- num, --iterations=num Overwrite files num times (default is 25).
-f, --force Force permissions to allow writing to files.
-s num, --size=num Shred num bytes. num can be expressed with suffixes (e.g., K, M, or G).
-u, --remove Remove file after overwriting. shred does not remove the file unless this option is specified.
-v, --verbose Verbose mode.
-x, --exact Shred the exact file size; do not round up to the next full block.
-z, --zero On the final pass, overwrite with zeros to hide the shredding.
skill [signal] [options] processes snice [priority] [options] processes
Send a signal to processes or reset the priority. The default signal for skill is TERM, and the default priority for snice is +4 but can be in the range +20 (slowest) to -20 (fastest). The selection options -c, -p, -t, and -u are not required, but can be specified to insure that processes are interpreted correctly. Options
-c The next argument is a command.
-i Use interactive mode.
-l, -L List available signals.
-n Display the process ID, but take no other action.
-p The next argument is a process ID.
-t The next argument is a tty or pty.
-u The next argument is a username.
-v Verbose mode.
-w Enable warnings.
slocate [options] [search-string]
Security Enhanced version of GNU locate. This command searches a database listing every file on the system; it is intended as a faster replacement for find / -name pattern .... (See locate in the "Alphabetical Summary of Common Commands" section earlier in this chapter). This version stores file permissions and ownership, in order to prevent unauthorized access to files. URL: http://www.geekreview.org/slocate/. Options
-c Read /etc/updatedb.conf when updating the database.
-d file, --database=file Use file as the database to search.
-e dir1[, dir2...] Exclude the given directories from the database.
-f fstype1[, fstype2...] Exclude files on filesystems of the given types (e.g., NFS).
-h, --help Print an option summary and exit.
-i Ignore case when searching.
-l level Security level. If level is 0, no security checks are done, providing faster searches. The default is 1, which turns on security checking.
-n count Do not print more than count results.
-o file, --output=file Use file as the database to create.
-q Do not print error messages (quiet mode).
-r regex, --regexp=regex Use the Basic Regular Expression regex to search the database.
-u Create the database starting at /.
-U dir Create the database starting at dir.
-v, --verbose Display filenames when creating the database.
-V, --version Print version information and exit.
splint [options] files ...
Secure Programming Lint. A freely available version of the original Unix lint command, splint performs static checking of C programs. By adding annotations in the form of special comments to your source files, splint can perform many additional, stronger checks than would otherwise be possible. splint has a very large number of options. They are broken down here into separate sections based on task. splint allows the use of a leading + and a leading minus to indicate an option. In many cases they do the same thing; in others one enables a feature while the other disables it. URL: http://www.splint.org/. Initialization Options
-f file Read options from file instead of from $HOME/.splintrc.
-Idir Add dir to the list of directories searched for C include files. As with the C compiler, there is no space separating the -I from the directory name.
-nof Do not read either of the default option files ./.splintrc and $HOME/.splintrc.
-S dir Add dir to the list of directories searched for .lcl specification files.
-tmpdir directory Use directory for temporary files. The default is /tmp.
-systemdirs dirlist Set the list of system directories to search for include files. The default is /usr/include. To include multiple directories, separate them with a colon.
-systemdirerrors Do not report errors for files in system directories.
Preprocessor Options
-D macro-definition Define a macro. This option is passed on to the C preprocessor.
-U macroname Undefine a macro. This option is passed on to the C preprocessor.
Library Options
-1-lib Use the Unix version of the standard library.
-1-strict-lib Use the "strict" version of the Unix standard library.
-ansi-lib Use the ANSI standard library. This is the default.
-dump file Save splint's state in file. The default file extension is .lcd.
-load file Load state from file, created previously with -dump. See the online documentation for more information.
-nolib Do not load information about any library. This also prevents loading information about the standard C library.
-posix-lib Use the POSIX version of the standard library.
-posix-strict-lib Use the "strict" version of the POSIX standard library.
-strict-lib Use the "strict" version of the ANSI standard library.
Output Options Use a leading minus to disable these options, and a leading + to enable them. By default they are all disabled.
-limit count Do not report more than count similar errors consecutively. Instead, show a count of suppressed errors.
-quiet Do not print version information or the error count summary.
-showalluses Print a list of external identifiers and their uses, sorted by the number of times each one is used.
-showscan Print each filename as it's processed.
-showsummary Print a summary of reported and suppressed errors. The count of suppressed errors may not be completely correct.
-stats Print the number of lines processed and the time it took to check them.
-timedist Print a distribution showing where the checking time was spent.
-usestderr Send error messages to standard error instead of to standard output.
-whichlib Print the filename and creation info for the standard library.
Expected Errors Option
-expect count Expect count errors. Exit with a failure status if exactly that number of errors are not detected. Useful for use with make.
Message Format Options For most of these options, a leading + turns on the behavior, whereas a leading minus turns it off. You may use either one for the -limit and -linelen options.
-forcehints Similar to -hints, but provide hints for all errors in a class, not just the first one. Default is off.
-hints Give hints describing the error and how to suppress it for the first error in each error class. Default is on.
-linelen count Set the maximum line length for a message line to count. Default is 80.
-paren-file-format Use messages of the form file(line).
-showallconj Print all possible alternate types. See the online documentation for more information. Default is off.
-showcolumn Print the column number of the error. Default is on.
-showfunc Print the name of the macro or function containing the error. Function names are only printed once. Default is on.
Mode Options The mode flags enable a coarse-grain grouping of different classes of checking. For more detail, use splint -help modes. From weakest to strongest, the options are -weak, -standard, -checks, and -strict.
-checks Strict checking. It does the same checking as -standard, plus must modification checking, rep exposure, return alias, memory management, and complete interfaces.
-standard This is the default. It does the same checking as -weak, plus modifies checking, global alias checking, use all parameters, using released storage, ignored return values of any type, macro checking, unreachable code, infinite loops, and fall-through cases. Old style declarations are reported. The types bool, int, and char are treated as being distinct.
-strict "Absurdly strict checking." Does the same checking as -checks, plus modifications and global variables used in unspecified functions, strict standard library, and strict typing of C operators. The manpage states "A special reward will be presented to the first person to produce a real program that produces no errors with strict checking."
-weak Checking intended for unannotated C code. In particular, splint does not do modifies checking, macro checking, rep exposure, or clean interface checking. It is allowed to ignore int return values. Old style declarations are not reported. The types bool, int, char, and enum are treated as being the same.
strace [options] command [arguments]
Trace the system calls and signals for command and arguments. strace shows you how data is passed between the program and the system kernel. With no options, strace prints a line to standard error for each system call. It shows the call name, arguments given, return value, and any error messages generated. A signal is printed with both its signal symbol and a descriptive string. Options
-a n Align the return values in column n.
-c Count all calls and signals and create a summary report when the program has ended.
-d Debug mode. Print debugging information for strace on standard error.
-e [ keyword= ][! ]values Pass an expression to strace to limit the types of calls or signals that are traced or to change how they are displayed. If no keyword is given, trace is assumed. The values can be given as a comma-separated list. Preceding the list with an exclamation mark (!) negates the list. The special values all and none are valid, as are the values listed with the following keywords.
abbrev=names Abbreviate output from large structures for system calls listed in names.
read=descriptors Print all data read from the given file descriptors.
signal=symbols Trace the listed signal symbols (for example, signal=SIGIO,SIGHUP).
trace=sets sets may be a list of system call names or one of the following:
file | Calls that take a filename as an argument | ipc | Interprocess communication | network | Network-related | process | Process management | signal | Signal-related |
raw=names Print arguments for the given system calls in hexadecimal.
verbose=names Unabbreviate structures for the given system calls. Default is none.
write=descriptors Print all data written to the given file descriptors.
-E var[= val] With no val, remove var from the environment before running command. Otherwise, run command with environment variable var defined to val in the environment.
-f Trace forked processes.
-ff Write system calls for forked processes to separate files named filename.pid when using the -o option.
-h Print help and exit.
-i Print instruction pointer with each system call.
-o filename Write output to filename instead of standard error. If filename starts with the pipe symbol |, treat the rest of the name as a command to which output should be piped.
-O n Override strace's built-in timing estimates, and just subtract n microseconds from the timing of each system call to adjust for the time it takes to measure the call.
-p pid Attach to the given process ID and begin tracing. strace can trace more than one process if more than one -p option is given. Type CTRL-C to end the trace.
-q Quiet mode. Suppress attach and detach messages.
-r Relative timestamp. Print time in microseconds between system calls.
-s n Print only the first n characters of a string. Default value is 32.
-S value Sort output of -c option by the given value. value may be calls, name, time, or nothing. Default is time.
-t Print time of day on each line of output.
-tt Print time of day with microseconds on each line of output.
-ttt Print timestamp on each line as number of seconds since the Epoch.
-T Print time spent in each system call.
-u username Run command as username. Needed when tracing setuid and setgid programs.
-v Verbose. Do not abbreviate structure information.
-V Print version and exit.
-x Print all non-ASCII strings in hexadecimal.
-xx Print all strings in hexadecimal.
umount [options] [directory | special-device]
System administration command. Unmount a filesystem. umount announces to the system that the removable file structure previously mounted on the specified directory is to be removed. umount also accepts the special-device to indicate the filesystem to be unmounted; however, this usage is obsolete and will fail if the device is mounted on more than one directory. Any pending I/O for the filesystem is completed, and the file structure is flagged as clean. A busy filesystem (one with open files or with a directory that is some process's current directory) cannot be unmounted. Options
-a Unmount all filesystems that are listed in /etc/mtab.
-d If the unmounted device was a loop device, free the loop device too. See also losetup(8).
-f Force the unmount.
-h Print help message and exit.
-l Lazy unmount. Detach the filesystem from the hierarchy immediately, but don't clean up references until it is no longer busy. Requires kernel 2.4.11 or later.
-n Unmount, but do not record changes in /etc/mtab.
-O options Unmount only filesystems with the specified options in /etc/fstab. Specify multiple options as a comma-separated list. Add no as a prefix to an option to indicate filesystems that should not be unmounted.
-r If unmounting fails, try to remount read-only.
-t type Unmount only filesystems of type type. Multiple types can be specified as a comma-separated list, and any type can be prefixed with no to specify that filesystems of that type should not be unmounted.
-v Verbose mode.
-V Print version information and exit.
watch [options] command [cmd_options]
Run the specified command repeatedly (by default, every 2 seconds) and display the output so you can watch it change over time. The command and any options are passed to sh -c, so you may need to use quotes to get correct results. Options
-d, --differences[=cumulative] Highlight changes between iterations. If cumulative is specified, the highlighting remains on the screen throughout, giving a cumulative picture of the changes.
-h, --help Display help message and exit.
-n secs, --interval=secs Run the command every secs seconds.
-t, --no-title Do not print a header line.
-v, --version Print version information and exit.
wget [options] [URL ...]
wget retrieves files from the Internet, most often using FTP or HTTP. It is capable of following links embedded in retrieved files, making it possible to mirror entire web sites. It has a plethora of options, making it difficult to use easily. See also curl in the earlier section "Alphabetical Summary of Common Commands." URL: http://www.gnu.org/software/wget/. Primary Options
-a file, --append-output=file The same as -o, but output is appended to the file, instead of overwriting it.
-b, --background Start off in the background. If no log file is specified with -o, use wget-log.
-c, --continue Continue retrieving a file that was partially downloaded. Very useful if a previous download was interrupted. See the manpage for some version-dependent caveats.
-d, --debug Enable debugging output.
-h, --help Print a help message summarizing the options and exit.
--http-passwd=password Use password as the password when an HTTP server prompts for a user and password. See the manpage for a discussion of security issues.
--http-user=user Use user as the user when an HTTP server prompts for a user and password. See the manpage for a discussion of security issues.
-I dirlist, --include-directories=dirlist When downloading, follow the directories in the comma-separated list dirlist. Elements in dirlist may contain wildcards.
-l depth, --level=depth Set the maximum recursion level to depth. The default is five.
-m, --mirror Enable options needed for mirroring. Equivalent to -r -N -l inf --no-remove-listing.
--no-glob Turn off FTP globbing (wildcard expansion). Globbing is automatically enabled if a URL contains shell wildcard characters. URLs with such characters should be quoted to protect them from the shell.
-nv, --non-verbose More verbose than -q but less verbose than -v. Only errors and basic information are printed.
-o file, --output-file=file Log all messages to file, instead of to standard error.
-P prefix, --directory-prefix=prefix Use prefix for the directory prefix, i.e., the directory under which all retrieved files are saved. The default is . (dot), i.e., the current directory.
--passive-ftp Use passive FTP to retrieve files. Often needed for clients residing behind a firewall.
--progress=type Set the progress indicator. Valid values are dot and bar, the default is bar.
--proxy-passwd=password Use password as the password for authentication on a proxy server. See the manpage for a discussion of security issues.
--proxy-user=user Use user as the user for authentication on a proxy server. See the manpage for a discussion of security issues.
-q, --quiet Be quiet, printing no output.
-r, --recursive Enable recursive retrieval.
-t count, --tries=count Retry no more than count times. Use 0 or inf to retry infinitely. The default is to retry 20 times.
-v, --verbose Be verbose. This is the default.
-V, --version Print a version message and exit.
-X dirlist, --exclude-directories=dirlist When downloading, exclude the directories in the comma-separated list dirlist from consideration. Elements in dirlist may contain wildcards.
-Y val, --proxy=val Enable/disable proxy support. Use one of on or off for val.
See the manpage for a description of the other options.
xmlto [options] format file
Convert a DocBook/XML file into a formatted file. The format is typically something like pdf or html. The file should be a text file containing DocBook/XML markup. xmlto applies a stylesheet to convert the document to the appropriate format, possibly formatting it in the process. Available Formats dvi | TeX DVI file. | fo | XSL-FO formatting objects. | html | Hypertext Markup Language. | html-nochunks | HTML, in one large file. | htmlhelp | HTML Help files. | javahelp | Java Help files. | man | Unix-style manpages. | pdf | Adobe's Portable Document Format (PDF) files. | ps | PostScript files. | txt | Plain text files. | xhtml | XHTML (XML-derived HTML) files. | xhtml-nochunks | XHTML, in one large file. |
Options
--extensions Enable stylesheet extensions for the toolchain in use.
-m XSL-file Use the XSL in XSL-file to modify the stylesheet.
-o dir Place output files in the directory dir instead of in the current directory.
-p options Pass the options on to the post-processing program. Repeat -p to pass options on to the second stage post-processor. (This currently applies only to the ps format, which uses TeX to produce DVI and dvips to produce PostScript.)
--searchpath path Add the directories in the colon-separated path to the search path for files to include.
--skip-validation Skip the normal validation check.
-x stylesheet Use stylesheet instead of picking one automatically.
Examples Produce a PDF file from DocBook/XML input: xmlto pdf brochure.xml
Generate HTML in a separate directory: xmlto -o /share/webserver/brochure html brochure.xml
|
