Programming :: `find . -exec' Handling Backticks In Bash
May 15, 2011
i had a problem with the find command in bash (which i deem is close enough to a promming language, if not please move this thread :P). i tried to reduce the command to the problem. i want the backticks, or $() for that matter; to be evaluated by -exec of find, not by bash. is that a caveat of find?
Code:
$ find testd -exec echo `basename {}` ; #confused me
test
test/a
test/b
[code]...
edit: i found out whats causing this. `basename {}` gets evaluated by bash before find is invoked, returns {} and `find . -exec echo {} ;" is run. now my question is, how to escape this eveluation from happening before.
I have a script that connects to a windows server, downloads a file, appends to it and then re-uploads the updated file. I want to implement error handling. An email is to be generated indicating whether there was an error or not. This email should include all standard and error output as a body. The current script looks something like this:
Code: function Email_ServerSupport { for time in once; do echo "Subject: Billing - smb copy to accounting" $1 cat /tmp/smbx cat /tmp/smbxerr done | mail $EMAILADDR }
/usr/bin/cp /dev/null /tmp/smbx /usr/bin/cp /dev/null /tmp/smbxerr cd /tmp /usr/sfw/bin/smbclient $LOCATION -A $AUTHFILE >>/tmp/smbx <<EOF get $OUTFILE exit EOF cat $INFILE >> $OUTFILE /usr/sfw/bin/smbxclient $LOCATION -A $AUTHFILE >>/tmp/smbx <<EOF put $OUTFILE
The reason for the grep -v's is because, from my understanding, when using smbclient, ALL output goes to stdout, even errors. For this reason, I need to filter out lines including "domain" "putting file" or "getting file", all of which aren't errors. The problem is that even though the script seems to catch errors successfully now and then, the success email ends up blank (/tmp/smbx is somehow empty). I'm also worried it could miss possible errors I haven't tested. I'm thinking it has to do with the way "EOF" functions. Is there any way to capture output from the "put" and "get" commands? I can't simply redirect the output, can I?
I wrote a script that easily runs it in the same directory as it was run below: #for f in *.MTS do ffmpeg -i "$f" -acodec copy -vcodec libx264 -threads 2 -deinterlace -vpre slow -b 20000k -bt 3000k -refs 4 "${f%.MTS}.mp4" ; #done
I want to be able to use the find command so it will recurse through all the videos in my videos folder. Is there a painless way to do this. Here is the start of my find command but it doesn't work. Any help appreciated:
Within a bash script, I'm attempting to redirect file descriptors with exec, e.g. exec 3>&1 1>&2; however, I'd like to do something like exec $FD>&1 1>&2, which doesn't work because bash tries to execute the value of $FD. Various placements of eval fail, as well. Is there a way around this, or am I stuck hard-coding the descriptor?
I have a system setup script for my Slackware installations that pulls all packages and source files from another machine and sets everything up to be identical between machines. The script works as expected but make it entirely unattended. How do I make the bash script deal with automatically selecting "Yes" for, for example "Install x(Yes/No): " when prompted by a make file?
Is there a way to use exec, but if exec fails to go on with the script?
Example:
Code: #!/usr/bin/env bash exec startx echo "Starting of X failed"
If startx fails, the echo will be seen on the screen. I tried all kind of stuff, but guess it ain't of much use to post it here. I searched the web, but searching for "exec and bash" in one sentence does give results which are not what i am looking for.
I have a friend (really), who is the author of a Perl application that uses backticks to launch an external application. The process reads a largish set of data files, and printing largish amounts of data which is consumed by Perl. Very rarely, the external application crashes. When it does, the Perl script has been seen to hang, and the host CPU usage goes to 100%. The Perl script can be killed and the host recovers.
The question is whether this is predictable/expected behavior, and if it is, is there anything within Perl which can be used to defend against this behavior. I have already recommended the use of the open() form of reading from a child process, but I'm not sure whether that really accomplishes anything. My theory on the matter is that Perl ends up busy-waiting on the child's output, never receiving any EOF, and consuming all CPU. I cannot see how using open() changes that behavior. I am now interested in this problem so I can avoid causing similar behaviors in the future, and just because 'inquiring minds wish to know'.
EDIT: I suppose I should have added that this is on generic 32-bit Linux hosts; Perl 5.8.X
But, if there is : symbol which separate namespace from action problems coming:
symfony doct[TAB] will be completed to symfony doctrine:
But nothing happens if you want complete after : symbol. I've found out that for readline there is three words because it splits line with $COMP_WORDBREAKS
Code:
$ echo $COMP_WORDBREAKS "'><=;|&(:
I played with $COMP_WORDS array and tried every thought I had to make it work, but failed. What I should do to escape colon and make readline consider it as one word? Or there is way perhaps to workaround it?
trouble when I log in the fedora as a normal user(not root) today. When I fill the username and the password press enter key. Then pop up a little window locate at topleft corner. And have one line message : "/usr/bin/xterm : Could not exec /bin/bash : Permission Denied"hen I log in as root, it's ok. And when I open shell and type "su username" then the console print "su: /bin/bash: Permission denied"I have checked the perms of "/bin/bash" , it's 755. And I have tried all things which suggested from articles searched through Google. like change / or /root and other directory's permissions but failed
I'm in trouble when I log in the fedora as a normal user(not root) today. When I fill the username and the password press enter key. Then pop up a little window locate at top left corner. And have one line message : "/usr/bin/xterm : Could not exec /bin/bash : Permission Denied" Then I log in as root, it's ok. And when I open shell and type "su username" then the console print "su: /bin/bash: Permission denied" I have checked the perms of "/bin/bash" , it's 755. And I have tried all things which suggested from articles searched through Google. like change / or /root and other directory's permissions but failed.
I have a bunch of .caf files in a directory that I want to convert to ogg but I cannot figure out way to do it (except manually). I want to do: sox filename.caf filename.ogg
I am working on tracing the signal handling mechanism in linux kernel internallly. For that, i build the kernel. Now, i want to trace the signal handling mechanism in the old kernel. I got to about SYSLOG and PRINTK for this. But, how to use these tools exactly in tracing the handling of signals internally ?. Is, there any tool similar to backtrace to do that?. How the call flow is done internally ?
Im looking for assistance to create a script to find and replace files.Probably best if I give you the background Our server uses a specific application which stores user data, each user data account (a folder on the server) has a file called 'Profile.xml' this file gets updated and replaced about every 30 mins similar to the fashion logrotate works i.e. Profile.xml.1 Profile.xml.2 -> .10
What we experience is that if the application crashes unexpectedly while it is doing its user profile refresh task we end up with sometimes a few hundred Profile.xml files which end up 0kb(should be around 4kb) , and our server see's these as corrupted profiles and will not see them. Our fix is to go back thru and rename the Profile.xml.1 to be Profile.xml (or sometimes up to Profile.xml.5 to Profile.xml) We want a script we can manually run to automate this process The server tree is
I am trying to find a nightly backup if it was successfully copied over, rename it and curl, but it's always passing the check even if the file is older than specified. From the command line it does as it should. Example is here;
Code: find /backup -type f -mmin +4440 -exec echo "found" {} ; - nothing returned (good). Then I change the time
I am trying to automate an svnadmin dump command for a backup script, and I want to do something like this:
find /var/svn/* ( ! -name dir -prune ) -type d -exec svnadmin dump {} > {}.svn ;
This seems to work, in that it looks through each svn repository in /var/svn, and runs svnadmin dump on it.
However, the second {} in the exec command doesn't get substituted for the name of the directory being processed. It basically just results a single file named {}.svn.
I suspect that this is because the shell interprets > to end the find command, and it tries redirecting stdout from that command to the file named {}.svn.
I've been following the instructions to install Resin, and all seems to go well until I try to run the /bin/httpd.sh file. It simple says:
Code:
exec: 40: -jar: not found
I suspected this was something to do with my JAVA_HOME variable not being set, so I looked up how to do this and added it to my /etc/bash.bashrc file. Saying "echo $JAVA_HOME" does indeed return the correct path so I don't think this is the problem.
I'm just starting out on a project relating to web search, to be done in C++. Which library should I use to help with downloading web pages into memory so that I can process them? The big thing is I want to be able to download the pages into variables/structures without actually putting them onto the hard disk.I googled and saw libcurl, but I was confused by some of the examples and wondering if this was really what I wanted.
I wonder how error handling is done in C in real applications, by returning a error code when something goes wrong, or by using setjmp and longjmp, or something else?
write such script (bash script). I have some text file with name filename.txt I must check if this file contains string "test-string-first", I must cut from this file string which follows string "keyword-string:" and till first white-space and save it to some variable.
For example. File: PHP Code: PHP Code: Start 15022011 Eng 12-3-42 SN1232324422 11 test-string-first SN322211 securities HH keyword-string:123456321-net mark (11-22)
I'm working on a bash script that will go through a directory, find the sub-directories that have been created since the last time the script ran, count the results, and output that integer (will most likely be '1' or less per each instance run) to a file. Give the circumstances, my previous (and very limited) experience with bash is not sufficient for me to pull this off. since it probably has bearing, is that my mail server stores files that it flags as viruses in a folder. It creates a sub-directory for each virus that it quarantines .I want to count those subdirectories and graph them with MRTG. Hence the script. I'm going to post what I've got so far and the purpose of it, because I'm told I have a very odd and efficient way of doing scripting.
[Code]...
But then it dawned on me that it wouldn't work because I would have to not count the directories that have already been counted and count the ones that have not been counted. Given that the purpose of this is to generate a graph about every 5 minutes, using find won't work because, to my knowledge, that will only find things based on whole day values, I need it almost down to the minute.
This may be a basic bash array/string operation related question, but I couldn't find any direct answer. So here it goes:I have a lot of data sorted in various directories. All directories need same processing except for a special group of directories. I have a symbolic link of the script in discussion in each directory. I want the script to get the name of the current directory, check if that belongs to special group and do specific operations.So I get the name of the directory
Code: mm=`basename `pwd`` Now the the group of directories that needs something different to be done, contains these
I have created a pthread, and installed a signal handler inside that, same way as we do in main( ) function. The thread's signal handler is a separate function. Surprisingly, it is not working, that is the thread's signal handler is not able to catch signals. Here is the code:
During my read, "Code Complete",the author described the different ways of handling the ADT interface.He has wrote this passage:
Quote:
Option 2: Explicitly provide the data used by the ADT services. In this approach, you declare the data that the ADT uses within each routine that uses an ADT service. In other words, you create a Font data type that you pass to each of the ADT service routines. You must design the ADT service routines so that they use the Font data that's passed to them each time they're called. The client code doesn't need a font ID if you use this approach because it keeps track of the font data itself. (Even though the data is available directly from the Font data type, you should access it only with the ADT service routines.This is called keeping the structure "closed.") The advantage of this approach is that the ADT service routines don't have to look up font information based on a font ID. The disadvantage is that it exposes font data to the rest of the program, which increases the likelihood that client code will make use of the ADT's implementation details that should have remained hidden within the ADT.
The problem is that it is a bit subtle in terms of coding for me. Can anybody give more concrete example, in working codes, especially in Java?
To parse several XML documents I'm using libxml++. I do programming for linux angstrom-armv5te. Because compiling libxml++ for this distribution and platform was impossible, I simply included the library to my project, which workes fine. Unfortunately I'm not able not use the exception handing which effects that a simple error inside a documents makes my whole program crash. This should not happen.
Is there a way to check the documents validity before parsing or better active the exception handling for doing this.
I'm starting bash shell script and I'm looping without any solution.
I'm trying to find some files under a folder hierarchy and in case of errors moving these files to a destination folder under the same hierarchy recreating this hierarchy if not exists.
Finding all ._* files under /src and moving them to /dest recreating folder1 or the others which contains ._* files but without moving files which does not correspond to the pattern.
Code:
I tried find command and I'am getting all needed files
Code:
But I don't know how to use the output to get the parent folder of files which are found to
1- create folder with mkdir -p /dest/folder1 or /dest/folder1/folder4
2- move found files from /src/... to /dest/... with rm command
I'm working on a find command as this trying to do all in the same line but ... little lost
