There is a documentation about setting window title: https://www.gnu.org/software/screen/manual/html_node/Naming-Windows.html
I don't want to change .bashrc on all machines to make dynamic screen. However it will be good set hotkey in my .screenrc to set title using :title (C-a A) command with proper argument.
Maybe there is a solution to provide an output from uname -n to :titile command. Or something similar which automatically or semi-automatically sets window title to hostname.
Questions are:
is there a way to provide output of uname -n to :title command?
is there any other way to set window title to current hostname without changing .bashrc?
There are two approaches to solve it:
Print out sequence which gnu screen can parse for window titile ESC k my-titile ESC \. Screen will extract my-title. It can be done via:
printf '\ek%s\e\\' $(uname -n);
Run screen -X title my-titile to set a titile of current window.
Since I want to do it for all hosts which I sshed into, there is a trick to wrap actual ssh command into ssh() function using one of aforementioned approaches.
E.g.:
ssh() { printf '\ek%s\e\\' "$1"; command ssh "$#"; }
Thanks to #screen on freenode : )
ref.: https://www.gnu.org/software/screen/manual/html_node/Dynamic-Titles.html#Dynamic-Titles
Hello all and thanks for your time in advance;
I'm running into a slight issue.
I'm running a command and piping it into a variable so i can manipulate the output.
$variable = some command
this normally works fine but doesn't output what's happening to the screen, which is fine most of the time. However occasionally this command requires some user input (like yes or no or skip for example), and since it's not actually piping anything to the command window, it just sits there hanging instead of prompting the user. If you know what to expect you can hit y or n or s and it'll proceed normally.
Is there anyway to run the command so that the output is piped to a variable AND appears on screen? I've already tried:
($variable = some command)
I've also tried:
write-host ($variable = some command)
But neither work. Note that the command running isn't a native windows or shell command and I cannot just run it twice in a row.
To clarify (because i probably wasn't clear)
I've also tried :
$variable = some command : Out-host
and
$variable = some command | out-default
with all their parameters, But the "prompt" from the command (to write y, n, s) doesn't show up.
Being able to pass S automatically would also be acceptable.
Sounds like you need Tee-Object. Example:
some command | Tee-Object -Variable MyVariable
This should pass everything from the command down the pipe, as well as store all output from the command in the $MyVariable variable for you.
You need to give some specific example that doesn't work. I tried this and it works:
function test { $c = read-host "Say something"; $c }
$x = test
I still see "Say something". read-host does not output to standard output so your problem is surprising. Even this works:
read-host "Say something" *> out
=== EDIT ===
Since this is interaction with cmd.exe you have two options AFAIK. First, test command:
test.cmd
#echo off
set /p something="Say something: "
echo %something%
This doesn't work as you said: $x= ./test.cmd
To make it work:
a) Replace above command with: "Say something:"; $x= ./test.cmd. This is obviously not ideal in general scenario as you might not know in advance what the *.cmd will ask. But when you do know its very easy.
b) Try this:
Start-transcript test_out;
./test.cmd;
Stop-transcript;
gc .\test_out | sls 'test.cmd' -Context 0,1 | select -Last 1 | set y
rm test_out
$z = ($y -split "`n").Trim()
After this $z variable contains: Say something: something. This could be good general solution that you could convert to function:
$z = Get-CmdOutput test.cmd
Details of text parsing might be slightly different in general case - I assumed here that only 1 question is asked and answer is on the same line but in any case with some work you will be able to get everything cmd.exe script outputed in general case:
=== EDIT 2 ===
This might be a better general extraction:
$a = gi test_out; rm test_out
$z = $a | select -Index 14,($a.count-5)
$z
$variable = ""
some command | % {$variable += $_;"$_"}
This executes the command, and each line of output is both added to $variable and printed to the console.
As part of my startup script to set up my desktop, I initialize a screen with several windows. I do this by starting a daemon and sending it -X screen and -X stuff commands, finally reattaching with -r.
Unfortunately, the "create daemon and reattach" method makes all the windows I created turn backspace into a "kill whole line" action. If I create new windows within screen with C-c c, the new windows do not have this behavior. Is this a screen bug, or can I do something special to fix this behavior? I'm using xfce4 and ubuntu 12.10 if that matters
Repro with the following:
screen -S -dm
screen -r
Type several characters and press backspace.
I'm not sure if I'm having the exact same problem as you, as your repro steps didn't work for me, but I did have the same bad behavior in screen (backspace killing the whole line), and managed to fix it.
For me, somehow I repeatedly get into a state where the output of stty is this:
$ stty
speed 9600 baud;
lflags: echoe echok echoke echoctl
iflags: -ixany -imaxbel ignpar
oflags: tab3
cflags: cs8 -parenb -hupcl clocal
eol eol2 erase2 kill min
^# ^# ^# ^H 0
Two things to note here:
There is no erase, only erase2
kill is mapped to ^H
#2 explains my issue, though #1 needs fixing, too.
Normally, ^U is "kill line", but here it is ^H instead.
If I type Ctrl-V, <backspace>, my terminal outputs ^H. So due to that mapping above, that causes the kill (kill line) to happen.
This fixed it for me:
$ stty kill ^U
# now, backspace outputs a literal ^H to the screen, so...
$ stty erase ^H
Note that in order to input the ^H and ^U, you have to use the literal control characters. I do this on my terminal with Ctrl-V, <backspace> and Ctrl-V, Ctrl-U, respectively.
I hope it helps!
I have found a work-around to this issue.
screen -r {session_name} -p 0 -X stuff "stty $(stty -g)"
screen -r {session_name} -p 0 -X stuff $'\n'
screen -r {session_name} -p 0 -X width $COLUMNS $LINES
screen -r {session_name} -p 0 -X stuff $'clear\n'
This takes the current tty settings and "stuffs" them into the screen session on Window 0, the default window that screen create on first launch. Then it "stuff" a newline to simulate pressing enter.
The next two lines are just to tell screen that initial columns and lines should match the calling terminal's columns and lines, then it clears the session's screen so that when you attach to the screen session your prompt will be at the top left. I was having issues of when attaching to the screen session the prompt would be in the middle of the terminal. A minor annoyance, but I wanted it gone.
NOTE: If you call the last two lines from within a script, those environment variables are not set. You will need to replace with $(tput cols) and $(tput lines)
In the repository home page , i can see comments posted in recent activity at the bottom, bit it only shows 10 commnets.
i want to all the comments posted since beginning.
Is there any way
Comments of pull requests, issues and commits can be retrieved using bitbucket’s REST API.
However it seems that there is no way to list all of them at one place, so the only way to get them would be to query the API for each PR, issue or commit of the repository.
Note that this takes a long time, since bitbucket has seemingly set a limit to the number of accesses via API to repository data: I got Rate limit for this resource has been exceeded errors after retrieving around a thousand results, then I could retrieve about only one entry per second elapsed from the time of the last rate limit error.
Finding the API URL to the repository
The first step is to find the URL to the repo. For private repositories, it is necessary to get authenticated by providing username and password (using curl’s -u switch). The URL is of the form:
https://api.bitbucket.org/2.0/repositories/{repoOwnerName}/{repoName}
Running git remote -v from the local git repository should provide the missing values. Check the forged URL (below referred to as $url) by verifying that repository information is correctly retrieved as JSON data from it: curl -u username $url.
Fetching comments of commits
Comments of a commit can be accessed at $url/commit/{commitHash}/comments.
The resulting JSON data can be processed by a script. Beware that the results are paginated.
Below I simply extract the number of comments per commit. It is indicated by the value of the member size of the retrieved JSON object; I also request a partial response by adding the GET parameter fields=size.
My script getNComments.sh:
#!/bin/sh
pw=$1
id=$2
json=$(curl -s -u username:"$pw" \
https://api.bitbucket.org/2.0/repositories/{repoOwnerName}/{repoName}/commit/$id/comments'?fields=size')
printf '%s' "$json" | grep -q '"type": "error"' \
&& printf "ERROR $id\n" && exit 0
nComments=$(printf '%s' "$json" | grep -o '"size": [0-9]*' | cut -d' ' -f2)
: ${nComments:=EMPTY}
checkNumeric=$(printf '%s' "$nComments" | tr -dc 0-9)
[ "$nComments" != "$checkNumeric" ] \
&& printf >&2 "!ERROR! $id:\n%s\n" "$json" && exit 1
printf "$nComments $id\n"
To use it, taking into account the possibility for the error mentioned above:
A) Prepare input data. From the local repository, generate the list of commits as wanted (run git fetch -a prior to update the local git repo if needed); check out git help rev-list for how it can be customised.
git rev-list --all | sort > sorted-all.id
cp sorted-all.id remaining.id
B) Run the script. Note that the password is passed here as a parameter – so first assign it to a variable safely using stty -echo; IFS= read -r passwd; stty echo, in one line; also see security considerations below. The processing is parallelised onto 15 processes here, using the option -P.
< remaining.id xargs -P 15 -L 1 ./getNComments.sh "$passwd" > commits.temp
C) When the rate limit is reached, that is when getNComments.sh prints !ERROR!, then kill the above command (Ctrl-C), and execute these below to update the input and output files. Wait a while for the request limit to increase, then re-execute the above one command and repeat until all the data is processed (that is when wc -l remaining.id returns 0).
cat commits.temp >> commits.result
cut -d' ' -f2 commits.result | sort | comm -13 - sorted-all.id > remaining.id
D) Finally, you can get the commits which received comments with:
grep '^[1-9]' commits.result
Fetching comments of pull requests and issues
The procedure is the same as for fetching commits’ comments, but for the following two adjustments:
Edit the script to replace in the URL commit by pullrequests or by issues, as appropriate;
Let $n be the number of issues/PRs to search. The git rev-list command above becomes: seq 1 $n > sorted-all.id
The total number of PRs in the repository can be obtained with:
curl -su username $url/pullrequests'?state=&fields=size'
and, if the issue tracker is set up, the number of issues with:
curl -su username $url/issues'?fields=size'
Hopefully, the repository has few enough PRs and issues so that all data can be fetched in one go.
Viewing comments
They can be viewed normally via the web interface on their commit/PR/issue page at:
https://bitbucket.org/{repoOwnerName}/{repoName}/commits/{commitHash}
https://bitbucket.org/{repoOwnerName}/{repoName}/pull-requests/{prId}
https://bitbucket.org/{repoOwnerName}/{repoName}/issues/{issueId}
For example, to open all PRs with comments in firefox:
awk '/^[1-9]/{print "https://bitbucket.org/{repoOwnerName}/{repoName}/pull-requests/"$2}' PRs.result | xargs firefox
Security considerations
Arguments passed on the command line are visible to all users of the system, via ps ax (or /proc/$PID/cmdline). Therefore the bitbucket password will be exposed, which could be a concern if the system is shared by multiple users.
There are three commands getting the password from the command line: xargs, the script, and curl.
It appears that curl tries to hide the password by overwriting its memory, but it is not guaranteed to work, and even if it does, it leaves it visible for a (very short) time after the process starts. On my system, the parameters to curl are not hidden.
A better option could be to pass the sensitive information through environment variables. They should be visible only to the current user and root via ps axe (or /proc/$PID/environ); although it seems that there are systems that let all users access this information (do a ls -l /proc/*/environ to check the environment files’ permissions).
In the script simply replace the lines pw=$1 id=$2 with id=$1, then pass pw="$passwd" before xargs in the command line invocation. It will make the environment variable pw visible to xargs and all of its descendent processes, that is the script and its children (curl, grep, cut, etc), which may or may not read the variable. curl does not read the password from the environment, but if its password hiding trick mentioned above works then it might be good enough.
There are ways to avoid passing the password to curl via the command line, notably via standard input using the option -K -. In the script, replace curl -s -u username:"$pw" with printf -- '-s\n-u "%s"\n' "$authinfo" | curl -K - and define the variable authinfo to contain the data in the format username:password. Note that this method needs printf to be a shell built-in to be safe (check with type printf), otherwise the password will show up in its process arguments. If it is not a built-in, try with print or echo instead.
A simple alternative to an environment variable that will not appear in ps output in any case is via a file. Create a file with read/write permissions restricted to the current user (chmod 600), and edit it so that it contains username:password as its first line. In the script, replace pw=$1 with IFS= read -r authinfo < "$1", and edit it to use curl’s -K option as in the paragraph above. In the command line invocation replace $passwd with the filename.
The file approach has the drawback that the password will be written to disk (note that files in /proc are not on the disk). If this too is undesirable, it is possible to pass a named pipe instead of a regular file:
mkfifo pipe
chmod 600 pipe
# make sure printf is a builtin, or use an equivalent instead
(while :; do printf -- '%s\n' "username:$passwd"; done) > pipe&
pid=$!
exec 3<pipe
Then invoke the script passing pipe instead of the file. Finally, to clean up do:
kill $pid
exec 3<&-
This will ensure the authentication info is passed directly from the shell to the script (through the kernel), is not written to disk and is not exposed to other users via ps.
You can go to Commits and see the top line for each commit, you will need to click on each one to see further information.
If I find a way to see all without drilling into each commit, I will update this answer.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 9 years ago.
Improve this question
I've just introduced a friend to GNU Screen and they're having a hard time getting used to it. That makes me think about the essential things he needs to know about the excellent Screen utility, the same things that you'd think worthwhile to teach someone, a beginner, from the ground up. What are some analogies and handy tips for remembering binds, etc.?
It would be awesome.
I've been using Screen for over 10 years and probably use less than half the features. So it's definitely not necessary to learn all its features right away (and I wouldn't recommend trying). My day-to-day commands are:
^A ^W - window list, where am I
^A ^C - create new window
^A space - next window
^A p - previous window
^A ^A - switch to previous screen (toggle)
^A [0-9] - go to window [0-9]
^A esc - copy mode, which I use for scrollback
I think that's it. I sometimes use the split screen features, but certainly not daily. The other tip is if screen seems to have locked up because you hit some random key combination by accident, do both ^Q and ^A ^Q to try to unlock it.
I couldn't get used to screen until I found a way to set a 'status bar' at the bottom of the screen that shows what 'tab' or 'virtual screen' you're on and which other ones there are. Here is my setup:
[roel#roel ~]$ cat .screenrc
# Here comes the pain...
caption always "%{=b dw}:%{-b dw}:%{=b dk}[ %{-b dw}%{-b dg}$USER%{-b dw}#%{-b dg}%H %{=b dk}] [ %= %?%{-b dg}%-Lw%?%{+b dk}(%{+b dw}%n:%t%{+b dk})%?(%u)%?%{-b dw}%?%{-b dg}%+Lw%? %{=b dk}]%{-b dw}:%{+b dw}:"
backtick 2 5 5 $HOME/scripts/meminfo
hardstatus alwayslastline "%{+b dw}:%{-b dw}:%{+b dk}[%{-b dg} %0C:%s%a %{=b dk}]-[ %{-b dw}Load%{+b dk}:%{-b dg}%l %{+b dk}] [%{-b dg}%2`%{+b dk}] %=[ %{-b dg}%1`%{=b dk} ]%{-b dw}:%{+b dw}:%<"
sorendition "-b dw"
[roel#roel ~]$ cat ~/scripts/meminfo
#!/bin/sh
RAM=`cat /proc/meminfo | grep "MemFree" | awk -F" " '{print $2}'`
SWAP=`cat /proc/meminfo | grep "SwapFree" | awk -F" " '{print $2}'`
echo -n "${RAM}kb/ram ${SWAP}kb/swap"
[roel#roel ~]$
Ctrl+A ? - show the help screen!
If your friend is in the habit of pressing ^A to get to the beginning of the line in Bash, he/she is in for some surprises, since ^A is the screen command key binding. Usually I end up with a frozen screen, possibly because of some random key I pressed after ^A :-)
In those cases I try
^A s and ^A q block/unblock terminal scrolling
to fix that. To go to the beginning of a line inside screen, the key sequence is ^A a.
You can remap the escape key from Ctrl + A to be another key of your choice, so if you do use it for something else, e.g. to go to the beginning of the line in bash, you just need to add a line to your ~/.screenrc file. To make it ^b or ^B, use:
escape ^bB
From the command line, use names sessions to keep multiple sessions under control. I use one session per task, each with multiple tabs:
screen -ls # Lists your current screen sessions
screen -S <name> # Creates a new screen session called name
screen -r <name> # Connects to the named screen sessions
When using screen you only need a few commands:
^A c Create a new shell
^A [0-9] Switch shell
^A k Kill the current shell
^A d Disconnect from screen
^A ? Show the help
An excellent quick reference can be found here. It is worth bookmarking.
Some tips for those sorta familiar with screen, but who tend to not remember things they read in the man page:
To change the name of a screen window is very easy: ctrl+A shift+A.
Did you miss the last message from screen? ctrl+a ctrl+m will show it again for you.
If you want to run something (like tailing a file) and have screen tell you when there's a change, use ctrl+A shift+m on the target window. Warning: it will let you know if anything changes.
Want to select window 15 directly? Try these in your .screenrc file:
bind ! select 11
bind # select 12
bind \# select 13
bind $ select 14
bind % select 15
bind \^ select 16
bind & select 17
bind * select 18
bind ( select 19
bind ) select 10
That assigns ctrl+a shift+0 through 9 for windows 10 through 19.
Ctrl+A is the base command
Ctrl+A N = go to the ***N***ext screen
Ctrl+A P = go to the ***P***revious screen
Ctrl+A C = ***C***reate new screen
Ctrl+A D = ***D***etach your screen
http://www.debian-administration.org/articles/34
I wrote that a couple of years ago, but it is still a good introduction that gets a lot of positive feedback.
I "must" add this: add
bind s
to your .screenrc, if You - like me - used to use split windows, as C-a S splits the actual window, but C-a s freezes it. So I just disabled the freeze shortcut.
Ctrl+a is a special key.
Ctrl+a d - [d]etach, leave programs (irssi?) in background, go home.
Ctrl+a c [c]reate a new window
Ctrl+a 0-9 switch between windows by number
screen -r - get back to detached session
That covers 90% of use cases. Do not try to show all the functionality at the single time.
Not really essential not solely related to screen, but enabling 256 colors in my terminal, GNU Screen and Vim improved my screen experience big time (especially since I code in Vim about 8h a day - there are some great eye-friendly colorschemes).
The first modification I make to .screenrc is to change the escape command. Not unlike many of you, I do not like the default Ctrl-A sequence because of its interference with that fundamental functionality in almost every other context. In my .screenrc file, I add:
escape `e
That's backtick-e.
This enables me to use the backtick as the escape key (e.g. to create a new screen, I press backtick-c, detach is backtick-d, backtick-? is help, backtick-backtick is previous screen, etc.). The only way it interferes (and I had to break myself of the habit) is using backtick on the command line to capture execution output, or pasting anything that contains a backtick. For the former, I've modified my habit by using the BASH $(command) convention. For the latter, I usually just pop open another xterm or detach from screen then paste the content containing the backtick. Finally, if I wish to insert a literal backtick, I simply press backtick-e.
There is some interesting work being done on getting a good GNU screen setup happening by default in the next version of Ubuntu Server, which includes using the bottom of the screen to show all the windows as well as other useful machine details (like number of updates available and whether the machine needs a reboot). You can probably grab their .screenrc and customise it to your needs.
The most useful commands I have in my .screenrc are the following:
shelltitle "$ |bash" # Make screen assign window titles automatically
hardstatus alwayslastline "%w" # Show all window titles at bottom line of term
This way I always know what windows are open, and what is running in them at the moment, too.
I use the following for ssh:
#!/bin/sh
# scr - Runs a command in a fresh screen
#
# Get the current directory and the name of command
wd=`pwd`
cmd=$1
shift
# We can tell if we are running inside screen by looking
# for the STY environment variable. If it is not set we
# only need to run the command, but if it is set then
# we need to use screen.
if [ -z "$STY" ]; then
$cmd $*
else
# Screen needs to change directory so that
# relative file names are resolved correctly.
screen -X chdir $wd
# Ask screen to run the command
if [ $cmd == "ssh" ]; then
screen -X screen -t ""${1##*#}"" $cmd $*
else
screen -X screen -t "$cmd $*" $cmd $*
fi
fi
Then I set the following bash aliases:
vim() {
scr vim $*
}
man() {
scr man $*
}
info() {
scr info $*
}
watch() {
scr watch $*
}
ssh() {
scr ssh $*
}
It opens a new screen for the above aliases and iff using ssh, it renames the screen title with the ssh hostname.
I like to set up a screen session with descriptive names for the windows. ^a A will let you give a name to the current window and ^a " will give you a list of your windows.
When done, detach the screen with ^a d and re-attach with screen -R
I like to use screen -d -RR to automatically create/attach to a given screen. I created bash functions to make it easier...
function mkscreen
{
local add=n
if [ "$1" == '-a' ]; then
add=y
shift;
fi
local name=$1;
shift;
local command="$*";
if [ -z "$name" -o -z "$command" ]; then
echo 'Usage: mkscreen [ -a ] name command
-a Add to .bashrc.' 1>&2;
return 1;
fi
if [ $add == y ]; then
echo "mkscreen $name $command" >> $HOME/.bashrc;
fi
alias $name="/usr/bin/screen -d -RR -S $name $command";
return 0;
}
function rmscreen
{
local delete=n
if [ "$1" == '-d' ]; then
delete=y
shift;
fi
local name=$1;
if [ -z "$name" ]; then
echo 'Usage: rmscreen [ -d ] name
-d Delete from .bashrc.' 1>&2;
return 1;
fi
if [ $delete == y ]; then
sed -i -r "/^mkscreen $name .*/d" $HOME/.bashrc;
fi
unalias $name;
return 0;
}
They create an alias to /usr/bin/screen -d -RR -S $name $command. For example, I like to use irssi in a screen session, so in my .bashrc (beneath those functions), I have:
mkscreen irc /usr/bin/irssi
Then I can just type irc in a terminal to get into irssi. If the screen 'irc' doesn't exist yet then it is created and /usr/bin/irssi is run from it (which connects automatically, of course). If it's already running then I just reattach to it, forcibly detaching any other instance that is already attached to it. It's quite nice.
Another example is creating temporary screen aliases for perldocs as I come across them:
mkscreen perlipc perldoc perlipc
perlipc # Start reading the perldoc, ^A d to detach.
...
# Later, when I'm done reading it, or at least finished
# with the alias, I remove it.
rmscreen perlipc
The -a option (must be first argument) appends the screen alias to .bashrc (so it's persistent) and -d removes it (these can potentially be destructive, so use at own risk). xD
Append:
Another bash-ism that I find convenient when working a lot with screen:
alias sls='/usr/bin/screen -ls'
That way you can list your screens with a lot fewer keystrokes. I don't know if sls collides with any existing utilities, but it didn't at the time on my system so I went for it.
^A A switches back to the screen you just came from.
Ctrl + A is a great special character for Unix people, but if you're using screen to talk to OpenVMS, then not being able to ^A is going to make you bald prematurely.
In VMS, if you're editing a DCL command prior to execution from the history buffer, Insert mode is off (it has to be for a few reasons I won't get into here) ... to turn it on so you don't over-type your command rather than space things out, you have to hit `^A.