I would like to display an image of my choice instead of the grey icon above my username on the login screen:
No doubt there is a setting for this somewhere, but I have failed to find it so far.
How do I change user images in the SDDM greeter?
Useful for testing -- you can use this command to preview the login screen without logging out.
System: Ubuntu MATE 18.04.
I have a collection of files (in multiple directories in a directory structure) and I want to move many of them to another directory. I can tell which files I want to move by their names (they are recipe titles), but there is no other way to distinguish between files I do want to move and files I don't want to move; I need to make a decision for every file by examining the name myself.
So, I am running a command once for each file I want to move, like
mv Snacks/OkaraCrackers NeedsTesting/
mv Snacks/SunflowerBrittle NeedsTesting/
mv Treats/ChocolateChilliFudge NeedsTesting/
mv Treats/PecanBlondies NeedsTesting/
Even with tab completion this is toooo much typing. What I really want to do is something like
shopt -s globstar
mv --prompt-me-for-every-file ** NeedsTesting/
So I can just press y or n for each file.
How can I move a group of files and be prompted for confirmation for every file?
I want to know how many regular files have the extension .c in a large complex directory structure, and also how many directories these files are spread across. The output I want is just those two numbers.
I've seen this question about how to get the number of files, but I need to know the number of directories the files are in too.
. or - and have spaces or newlines..c, and symlinks to directories. I don't want symlinks to be followed or counted, or I at least want to know if and when they are being counted..c file in it.I hastily wrote some commands in the (Bash) shell to count them myself, but I don't think the result is accurate...
shopt -s dotglob
shopt -s globstar
mkdir out
for d in **/; do
find "$d" -maxdepth 1 -type f -name "*.c" >> out/$(basename "$d")
done
ls -1Aq out | wc -l
cat out/* | wc -l
This outputs complaints about ambiguous redirects, misses files in the current directory, and trips up on special characters (for example, redirected find output prints newlines in filenames) and writes a whole bunch of empty files (oops).
How can I reliably enumerate my .c files and their containing directories?
In case it helps, here are some commands to create a test structure with bad names and symlinks:
mkdir -p cfiles/{1..3}/{a..b} && cd cfiles
mkdir space\ d
touch -- i.c -.c bad\ .c 'terrible
.c' not-c .hidden.c
for d in space\ d 1 2 2/{a..b} 3/b; do cp -t "$d" -- *.c; done
ln -s 2 dirlink
ln -s 3/b/i.c filelink.c
In the resulting structure, 7 directories contain .c files, and 29 regular files end with .c (if dotglob is off when the commands are run) (if I've miscounted, please let me know). These are the numbers I want.
Please feel free not to use this particular test.
N.B.: Answers in any shell or other language will be tested & appreciated by me. If I have to install new packages, no problem. If you know a GUI solution, I encourage you to share (but I might not go so far as to install a whole DE to test it) :) I use Ubuntu MATE 17.10.
From time to time, Ubuntu users encounter errors from apt or apt-get, of the form:
The package some-package needs to be reinstalled, but I can't find an archive for it.I want to know what causes this error so that I can reproduce it and try to find safe ways of fixing it, even when the broken package is important and has many dependencies. This bit of the APT source code might give a bit of a clue, but I don't know how packages get the status that causes the error.
How can I produce this error for testing on my (Ubuntu MATE 17.10) system?
This is prompted by this recent Ask Ubuntu question and my long-standing discomfort with closing questions about this error against Apt/Synaptic needs to reinstall package but can't find the archive for it, whose answer uses dpkg --force-all to remove the offending package. Anonymous feedback suggests this solution has worked for many users, but I am not confident that using --force-all won't cause later problems, or that it's a good idea to use this method to remove important packages like APT.
I noticed that there is a binary executable /bin/echo on my Ubuntu MATE 17.04 system.
I thought, that's odd, because
$ type echo
echo is a shell builtinCursory testing suggests that /bin/echo does the same sort of thing as the Bash builtin echo:
$ /bin/echo foo
foo
$ /bin/echo $USER
zanna
So, why is there another version of echo separate from the Bash program, and why or when would I want to use it?
This question is prompted by a short script I found in a Linux magazine. As evidence that I didn't make this up, here's a picture of it:
I would like to write a letter to the editor of this publication about what's wrong with this and how to write it better.
The script attempts to capture jpeg files into a variable, so that something (compression using lepton) can be done with them.
for jpeg in `echo "$(file $(find ./ ) |
grep JPEG | cut -f 1 -d ':')"`
do
/path/to/command "$jpeg"
...
Apparently in this instance we can't trust the files to be named with a .jpg extension so we can't catch them with something like
for f in *.JPG *.jpg *.JPEG *.jpeg ; do ...
because the writer has used file to check their type, but if the filenames can't be trusted to have a sensible extension, then I don't see how we can trust them not to be -rf * or (; \ $!| or have newlines or whatever else.
How can I sanely capture files into a variable by type with for or while, or perhaps avoid doing so by using find with -exec, or some other method?
Bonus for insights into and demonstrations of what's wrong with the code in the picture.
I've tagged this question with [bash] since it's about a bash script, but if you feel like answering with a way to do this that doesn't use bash, then please feel free to do that.
I would like to write a script or function to tell me how many days from now until a given date in the future. What I'm struggling to work out is how to process the given date and compare it with the current date... I'm imagining something like
read -p "enter the date in the format YYYY-MM-DD "
And then I'm assuming I have a string that's meaningless to the shell and I have to do some evaluations like...?? (This is just an example; I guess bc would be needed)
i=$(($(date +%Y)-${REPLY%%-*}))
j=$(($(date +%m)-${REPLY:5:2}))
k=$(($(date +%d)-${REPLY##*-}))
And then I don't know what to do with those numbers...??
if $i > 1 then assign l=$((i*365)) and else what?? # what about leap years?
Using $j somehow assign m # confused before I've started
Using $k somehow assign n # just as bad
echo $((l+m+n))
I'm surely making it too hard for myself; probably there's a text processing tool that understands dates and can compare them.
How can I do this?
I want to carry out some action (say chown) on all the hidden files in a directory.
I know that this .* is not a good idea because it will also find the current . and parent .. directories (I know that rm will fail to operate on . and .. but other commands, including chown and chmod, will happily take effect)
But all my hidden files have different names!
How should I glob for all hidden files while excluding . and .. ?
I'm working with the .csv output of this SE data query which looks like this (only with 5022 entries):
"{
""id"": 281952,
""title"": ""Flash 11.2 No Longer Supported by Google Play""
}"
"{
""id"": 281993,
""title"": ""Netbeans won't open in Ubuntu""
}"
(And it has line ^M endings between [number], and ""title""). I need it to look like this:
281952,Flash 11.2 No Longer Supported by Google Play
281993,Netbeans won't open in Ubuntu
I fixed this in a certain text editor which shall remain nameless quite easily, but I wanted to make a script so that I don't have to do it again every time the query is refreshed & so others can use it. I used sed...
This series of commands works perfectly (although it may well be inefficient; it is just a trial-and-error solution):
# Print the ^M and remove them, write to a new file:
cat -v QueryR* | sed 's/\^M//' > QueryNew
# remove all the other junk:
sed -i 's/{//' QueryNew
sed -i 's/}//' QueryNew
sed -i 's/""//g' QueryNew
sed -i 's/^"//' QueryNew
sed -i '/,/{N;/\n.*title:\s/{s/,\n.*title:\s/,\ /}}' QueryNew
sed -i 's/^\s\+//' QueryNew
sed -i '/^\s*$/d' QueryNew
sed -i 's/^id:\ //' QueryNew
sed -i 's/,\ /,/' QueryNew
sed -i 's/\\//g' QueryNew
So, why doesn't this? Only the ^M and {} get removed, and everything else is still there.
#!/bin/bash
cat -v QueryR* | sed 's/\^M//' > QueryNew
sed -i '{
s/{//
s/}//
s/""//g
s/^"//
/,/{N;/\n.*title:\s/{s/,\n.*title:\s/,\ /}}
s/^\s\+//
/^\s*$/d
s/^id:\ //
s/,\ /,/
s/\\//g
}' QueryNew
I'm sure my mistake is really obvious...
Apparently I cannot count. I think there are three files in /media
$ tree /media
/media
├── foo
├── onex
└── zanna
3 directories, 0 files
However, ls -l finds 12.
$ ls -l /media
total 12
drwxr-xr-x 2 root root 4096 Jul 31 20:57 foo
drwxrwxr-x 2 root root 4096 Jun 26 06:36 onex
drwxr-x---+ 2 root root 4096 Aug 7 21:17 zanna
And, if I do ls -la I get only . and .. in addition to the above, but the count is total 20
What's the explanation?
I have an alias defined in my .bashrc
alias l.='ls -d .* --color=auto'
It's very useful :) but it doesn't work via ssh:
$ ssh localhost l.
bash: l.: command not found
Why is that?
I use Character Map (charmap) but I find it annoying to search through all the characters for the ones I want, because I only use about 30 of them regularly.
What I want to do is add characters that I use often to use to a custom "script" that would ideally appear in the character map sidebar as shown, as for example "my charset"
However, it would be perfectly fine for this favourite character set to be available from some other interface, (it doesn't have to be graphical) as long as it displays the characters so I don't have to try to remember all the unicode numbers :)
I wrote an answer advising the use of sudo -i nautilus. I know that gksu is deprecated, and I was under the impression that sudo -i is how we should run graphical applications as root without killing kittens, but someone commented 'pkexec is the new gksu'. My understanding is that pkexec is not the same as using any form of sudo:
When to use pkexec vs gksu/gksudo?
and requires setting up, especially to use with graphical applications:
When I do sudo -i everything always works, so why should I go to the trouble of configuring pkexec? I can see that for writing a script that launches something as root I would need to do it:
Can I use pkexec in a Python Script or .desktop file?
But otherwise, what do I gain from using pkexec? When should I use these commands? What is currently the best practice for opening graphical applications as root?
(I already read the... man pages, but I am feeling no wiser.)
I always shut down my Ubuntu MATE 16.04 system in exactly the same way: I keep hitting alt+F4 until I get to the very last terminal window and all other applications have been exited, and then I type poweroff
Despite the consistency of my procedure, the time taken by the system to fully shut down (ie the amount of time I'm watching the splash screen) varies a lot, from about 3 seconds to about 25 seconds. Why does this happen, or how can I find out?
I know I can read the logs, but I would love some tips on how to analyse for this
I have no indicator LEDs on my keyboard to show whether Caps-Lock is on (I don't have Num-Lock and Scroll-Lock keys so the best option would be to have an indicator only for Caps Lock, but three indicators would be better than none)
I would really like to see the lock-keys status in the indicator panel so I can tell at a glance whether I'm typing my passwords correctly
I'm using Ubuntu MATE 16.04. I've seen this question but it relates to Unity DE.
Using MATE desktop I get two icons for Telegram in the Applications > Internet menu:
clicking either of these icons launches the application (and if I click either one while telegram is running, it is unminimised and/or gets focus)
I installed Telegram Desktop by this method:
sudo add-apt-repository ppa:atareao/telegram
sudo apt-get update
sudo apt-get install telegram
It works perfectly (& I love it) but I only want one icon... How can I make one of them disappear?
I know how to pronounce Ubuntu but I don't know how to pronounce Xubuntu (which I'm currently enjoying). Should it be
(Similarly 'loo'-buntu or 'el'-ubuntu, 'koo'-buntu or 'kay'-ubuntu?)
Maybe this is just a matter of opinion (unlike for the real preexisting word ubuntu), but I want to get it right if there is a right way.
I use Character Map regularly alongside LibreOffice writer to write IPA (international phonetic alphabet) notation. All the (30ish) characters I need are in the 'Latin' script list, but there are many characters in this list that I do not need, and consequently it is a long process to find the characters I want, and making IPA notation is very time consuming. How can I streamline writing IPA notation in LibreOffice?
I am using Ubuntu 16.04
I mark exams for Pearson online using a Java applet called e-PEN on this site. The 'minimum system requirements' state that only Vista or later Windows, and Internet Explorer 9 or later, are supported.
I am using Ubuntu 16.04 and I believe that the correct interpretation of 'minimum system requirements' is that these are the conditions in which I can expect assistance when I contact the ePEN helpdesk.
Bonus section: I have already answered my own question (see below) using Oracle Java 8 BUT I would prefer to use only FOSS. I have uninstalled Oracle Java and made great progress with OpenJDK8 and IcedTea8. After I wrote an elaborate java security policy (basically granting each permission I got an error about in a slow line by line process) and copied the swt.jar into the right directory (as in the Oracle fix), the applet does START with all features. However, I invariably get an InterruptedException thrown within a few seconds:
Caused by: java.lang.InterruptedException
at java.lang.Object.wait(Native Method)
at java.lang.Object.wait(Object.java:502)
at
sun.applet.PluginAppletViewer.requestPluginCookieInfo(PluginAppletViewer.java:1266)
(or the same with .requestPluginCookieInfo) which causes the applet to halt.
When running firefox in a terminal, I also see these two errors every time:
java.io.FileNotFoundException: https://path/to/applet/parent-url/xbean.jar
java.io.FileNotFoundException: https://path/to/applet/parent-url/tiny_mce_3.5.10.zip.pack.gz
If I try to go to the real URLs or wget them, I get a 404 Not Found error, so I guess those resources just aren't there. I tried downloading an xbean.jar which I put into the same directory as the swt library jar, but this had no effect. I also get a NoClassDef error and a Connection Refused error further down (after the fatal Interrupt error has already been thrown).
All the info I find online relating to these errors is on how to stop them happening from the other side, as the developer. I'm tempted to give up, but still hoping someone will have a clue :)
I needed to regenerate initrd. Doing this:
update-initramfs -u -k all
threw up errors for a kernel I compiled and then deleted some time ago (because update-initramfs could not find the /lib/modules directory I had deleted of course).
I thought I had completely removed the kernel by deleting all its entries in /boot and directory in /lib/modules & running sudo update-grub (as in this answer), but apparently there are traces of it somewhere that I should probably clean up.
I read this and dug around in /usr/share/initramfs-tools but I couldn't figure it out.
In man update-initramfs I see that:
The use of "all" for the version string specifies update-initramfs to execute the chosen action for all kernel versions, that are already known to update-initramfs.
But how does it know them? Where does update-initramfs look for kernel versions?
Note: many months later... I added the answer to this question to the tag wiki for initramfs