LinuxQuestions.org
Share your knowledge at the LQ Wiki.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Newbie
User Name
Password
Linux - Newbie This Linux forum is for members that are new to Linux.
Just starting out and have a question? If it is not in the man pages or the how-to's this is the place!

Notices


Reply
  Search this Thread
Old 09-06-2007, 01:51 PM   #1
WingnutOne
Member
 
Registered: Sep 2007
Location: Kansas City
Distribution: Mixed, mostly RH / Fedora
Posts: 76

Rep: Reputation: 16
Question How to pull sequential lines from a file?


I have a text file that contains an inventory list of backup tapes stored in an automated tape storage bank. The file contains an entry for each storage location and each entry is 16 lines long.

I need to separate out each 16-line entry so that I can then manipulate the contents.

What function(s) would be best for this?
(The header line for each entry reads “Slot Address 1234”, “Slot Address 1235”, etc.)

Any advice would be appreciated.
 
Old 09-06-2007, 02:13 PM   #2
AlucardZero
Senior Member
 
Registered: May 2006
Location: USA
Distribution: Debian
Posts: 4,817

Rep: Reputation: 610Reputation: 610Reputation: 610Reputation: 610Reputation: 610Reputation: 610
In what framework? Are you thinking of writing a program to do this, or do you want a command-line?
 
Old 09-06-2007, 02:13 PM   #3
visaris
Member
 
Registered: Dec 2004
Distribution: gentoo
Posts: 190

Rep: Reputation: 30
I need to separate out each 16-line entry so that I can then manipulate the contents.

I have to say I'm confused. Can't you manipulate the entries in a normal text editor? Do you want these entries to be in separate files? Do you need them in memory/variables so you can use them for an app/project? If so, what language do you want? Bash, Java, C++, C, Perl, Python, etc?

I've said this a few times today, but I guess I need to say it again. I'm really sorry if I'm coming across as being mean spirited, but you haven't really given us enough information to help you. As written, we have no idea what exactly you want, what form/language you want an answer in, or what exactly you are even trying to do.

Please make another post with LOTS! HEAPS! MOUNTAINS! of detail, and then we can direct you to a proper solution. Thanks!
 
Old 09-06-2007, 03:31 PM   #4
trashbird1240
Member
 
Registered: Sep 2006
Location: Durham, NC
Distribution: Slackware, Ubuntu (yes, both)
Posts: 463

Rep: Reputation: 31
I would suggest grep with the --context option. You can also use --before-context and --after-context to give you exactly the number of lines you need.

For instance

Code:
grep --after-context=16  -e "“Slot Address 1234”, “Slot Address 1235”" filename > results
Should give you what you want.

Even AWK would be overkill in this case; Perl or C would be over-massacre.

Joel
 
Old 09-06-2007, 04:00 PM   #5
WingnutOne
Member
 
Registered: Sep 2007
Location: Kansas City
Distribution: Mixed, mostly RH / Fedora
Posts: 76

Original Poster
Rep: Reputation: 16
Details coming up!!!

LOL! This'll be a refreshing change. On most of the other boards I've tried posting to, I've tried giving lots of details and gotten no replies at all because no one wants to take the time to read & process it all!

1. I'm running Red Hat Enterprise 4 on kernel 2.6.9-55.EL (required for the drivers that operate the tape unit, an IBM TS3100).

2. I'm using Bash, and need to write this into a Bash script.

3. I originally said the source was a "file" to try to keep it simple.
The actual source of this data is the output of an inventory check command (which is part of the software package that came with the TS3100). I'll probably pipe the output directly from the inventory command, but could dump it into a file and manipulate it from there if required.

4. I need to generate this inventory output and have the script dig through it on a daily basis.

5. The ultimate purpose of this is to compare the data on the "Slot Address" line (the numbers 4097 & 4098 below) with the tape that's actually in each storage slot and make sure that all of the tapes are where they're supposed to be. The tape's ID number in the first example below is 000102L3 (near the end of the 11th line down).
The second example is of a Slot that was empty when the inventory check was run.
(The tapes can be moved manually; otherwise I wouldn't bother.)


Here are two sample items from the output of the inventory check command:

Slot Address 4097
Slot State ..................... Normal
ASC/ASCQ ....................... 0000
Media Present .................. Yes
Robot Access Allowed ........... Yes
Source Element Address ......... 4097
Media Inverted ................. No

Volume Tag, Length 36

0 1 2 3 4 5 6 7 8 9 A B C D E F 0123456789ABCDEF
0000 - 3030 3031 3032 4C33 2020 2020 2020 2020 [000102L3 ]
0010 - 2020 2020 2020 2020 2020 2020 2020 2020 [ ]
0020 - 0000 0000 [.... ]


Slot Address 4098
Slot State ..................... Normal
ASC/ASCQ ....................... 0000
Media Present .................. No
Robot Access Allowed ........... Yes
Source Element Address Valid ... No
Media Inverted ................. No

Volume Tag, Length 36

0 1 2 3 4 5 6 7 8 9 A B C D E F 0123456789ABCDEF
0000 - 0000 0000 0000 0000 0000 0000 0000 0000 [................]
0010 - 0000 0000 0000 0000 0000 0000 0000 0000 [................]
0020 - 0000 0000 [.... ]

-----END OF EXAMPLE------


6. I have a script already made which cross-references the names of the tapes with the slot where they're supposed to be stored.

Hope that's not too much info. If you need more specifice, let me know.
 
Old 09-06-2007, 04:53 PM   #6
AlucardZero
Senior Member
 
Registered: May 2006
Location: USA
Distribution: Debian
Posts: 4,817

Rep: Reputation: 610Reputation: 610Reputation: 610Reputation: 610Reputation: 610Reputation: 610
in that case, as mentioned by trashbird, perhaps something like:
Code:
grep --after-context=16  -e "^Slot Address"
?
 
Old 09-06-2007, 05:00 PM   #7
WingnutOne
Member
 
Registered: Sep 2007
Location: Kansas City
Distribution: Mixed, mostly RH / Fedora
Posts: 76

Original Poster
Rep: Reputation: 16
Quote:
Originally Posted by trashbird1240 View Post
I would suggest grep with the --context option. You can also use --before-context and --after-context to give you exactly the number of lines you need.

For instance

Code:
grep --after-context=16  -e "“Slot Address 1234”, “Slot Address 1235”" filename > results
Should give you what you want.

Even AWK would be overkill in this case; Perl or C would be over-massacre.

Joel

I entered your command line as written but didn't quite get the result I expected. The output included every line in the file from the first instance of "Slot Address" to the end.

I wonder if I might not be reading/writing some of the quote marks wrong? (I'm fairly new at this so that's entirely possible.) The inner double quotes around “Slot Address 1234” look different from the outer set, but I don't know where to find the set that's different than "standard".

Any ideas here?

Thanks again,
Doug
 
Old 09-06-2007, 08:55 PM   #8
choogendyk
Senior Member
 
Registered: Aug 2007
Location: Massachusetts, USA
Distribution: Solaris 9 & 10, Mac OS X, Ubuntu Server
Posts: 1,191

Rep: Reputation: 105Reputation: 105
That's a pretty heavy duty tape unit to be using bundled software that can't even give you reports on what's up. Are you backing up a number of computers?

I realize it's not the question you asked, but I might be inclined to dump the bundled software and use a decent open source backup scheduler like amanda (http://www.linuxquestions.org/bookmarks/tags/amanda). It not only might manage your backups better, but the built in reporting capabilities would make it easy to do things like inventorying your tapes. You've got what, 24 slots and something like LTO 3 or 4? It seems kind of silly to have that kind of hardware and be messing around trying to get the information out of it that you want.
 
Old 09-06-2007, 10:27 PM   #9
syg00
LQ Veteran
 
Registered: Aug 2003
Location: Australia
Distribution: Lots ...
Posts: 15,174

Rep: Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945
I'm sure the awk-ophytes could come up with something, but I used perl. Output was:
Code:
4097    000102L3
4098    ................
 
Old 09-07-2007, 08:44 AM   #10
WingnutOne
Member
 
Registered: Sep 2007
Location: Kansas City
Distribution: Mixed, mostly RH / Fedora
Posts: 76

Original Poster
Rep: Reputation: 16
Quote:
Originally Posted by choogendyk View Post
That's a pretty heavy duty tape unit to be using bundled software that can't even give you reports on what's up. Are you backing up a number of computers?

I realize it's not the question you asked, but I might be inclined to dump the bundled software and use a decent open source backup scheduler like amanda (http://www.linuxquestions.org/bookmarks/tags/amanda). It not only might manage your backups better, but the built in reporting capabilities would make it easy to do things like inventorying your tapes. You've got what, 24 slots and something like LTO 3 or 4? It seems kind of silly to have that kind of hardware and be messing around trying to get the information out of it that you want.

At this point, I'm backing up about a dozen remotely mounted iSCSI drives, but may add more in the future. Since my backup machine has no other purpose than to make backups of other servers' drives and all of the drives are mounted to the same directory, I'm using a simple tar command to get them all at once.
From the /mnt/backups/ dir:
Code:
tar -czvf /dev/IBMtape0 .
As for Amanda, I looked at it before I started this but, given the simplicity of the actual backup command, I discarded the idea because it seemed like overkill. I didn't know it could inventory the tapes though, so I'll take another look at it.

Thanks again!
 
Old 09-07-2007, 10:15 AM   #11
trashbird1240
Member
 
Registered: Sep 2006
Location: Durham, NC
Distribution: Slackware, Ubuntu (yes, both)
Posts: 463

Rep: Reputation: 31
Quote:
Originally Posted by WingnutOne View Post
I entered your command line as written but didn't quite get the result I expected. The output included every line in the file from the first instance of "Slot Address" to the end.

I wonder if I might not be reading/writing some of the quote marks wrong? (I'm fairly new at this so that's entirely possible.) The inner double quotes around “Slot Address 1234” look different from the outer set, but I don't know where to find the set that's different than "standard".

Any ideas here?

Thanks again,
Doug
Hi, I'm sorry I didn't make it clear that it was an example. My intention was to point you in the right direction, however you'll have to figure out the appropriate regular expression to use. Alucadzero had the right idea. If you know how long the entries are and a unique expression on the first line, then you have all the info you need.

My best advice is:

Code:
man grep
and get your hands on Classic Shell Scripting which is a great guidebook form beginning to end. It covers regular expressions for grep in great detail.

Joel
 
Old 09-07-2007, 10:37 AM   #12
syg00
LQ Veteran
 
Registered: Aug 2003
Location: Australia
Distribution: Lots ...
Posts: 15,174

Rep: Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945Reputation: 1945
Try something like this - might be easier to use your current setup than install something new. Shouldn't be too hard to organise the two lists so they can be compared
Code:
#! /usr/bin/perl
use warnings ;
# Parse slot address followed by tape number
# presumes input data is well-formed.
my $pattn1 = '(^Slot Add.*\s)(\d+$)' ;
my $pattn2 = '(^0000\s.*\[)(.*)\]$' ;
my $five_dots = '^\.{5}' ;
my ($slot, $tape_no) ;
while (<>) {
  if (/$pattn1/) { 
    $slot = $2 ;
  } elsif (/$pattn2/) {
    $tape_no = $2 ;
#   $tape_no = "Empty" if ($tape_no =~ $five_dots) ;
    print "$slot\t$tape_no\n" ;    
    }
}
Just feed your data into it as STDIN. The commented line is if you want a better label than a pile of dots ...
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
gnuplot, pull X data from file, specify Y data at cmd line? hedpe Programming 5 03-15-2007 11:32 PM
PHP Pull File Lines Into Table windisch Programming 7 11-02-2006 05:06 PM
Random file lines directed to a new file. In script an error. In command line no err leventis Programming 1 09-28-2006 07:16 AM
How can I sort the lines in a file? windhair Linux - Software 2 11-17-2005 08:37 AM
using DD to pull records out of a data file fhinkle Linux - Newbie 6 02-24-2005 04:20 PM


All times are GMT -5. The time now is 01:26 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration