How am i able to execute UNIX commands on my PC Command prompt? Note i do not have cygwin installed, although i was going to before i discovered this.
This is a development machine so i have a lot installed on it like ruby, python, git, github, node and so on.
What does this mean? can i use this without cygwin?
Here is a list of programs installed on my PC program list
How am I able to execute UNIX commands on my PC Command prompt?
You can use the where command in a cmd shell to find out the exact location of your Unix commands, for example:
where ls
This assumes, of course, that ls is located somewhere in your current PATH.
The location returned will show you in which directory your Unix commands are installed and may be enough for you to determine how they were installed.
The where command is roughly equivalent to the Unix which command.
By default, the search is done in the current directory and in the
PATH.
Syntax
WHERE [/r Dir] [/q] [/f] [/t] Pattern ...
WHERE [/q] [/f] [/t] [$ENV:Pattern
Source where
Further Reading
An A-Z Index of the Windows CMD command line - An excellent reference for all things Windows cmd line related.
where - Locate and display files in a directory tree.
Running Unix commands in windows can be done by having a tool like Cygwin which has those commands.
You can also get many of those commands compiled for windows and then run them using the command with the full path or only the command if the executable is in a path known by adding the paths to the executable files in Windows by :
1) Running in the terminal: PATH %PATH%;C:\<new_path>
2) Creating command aliases like: doskey np=C:\<new_path>\new_command.exe $*. $* is used to be able to transmit parameters
Related
I want to execute a Linux binary from a QT application running on W10.
In QT we have QProcess to launch additional processes. However, since my binary is for Linux, I've thought of two possible approaches:
Running the binary in a container (i.e.: Docker, Kubernetes, Singularity...).
Executing the binary through WSL (Ubuntu) bash.
In any case, the QT application should initiate the process (the container or the bash) and, in turn, this process should launch my binary.
I've been searching on the web and I could not find something related, what makes me think that it will be difficult. For this reason, I am posting the question in order to know the viability of the proposed approaches.
EDITED
It looks like the WSL is easier, the problem is that the user has to have install ed it. Apart from requiring the sudo password when installing new software via apt-get.
The binary that I have to execute only exists for Linux, and let's say that cross-compiling is dismissed because of its complexity. Furthermore, this application needs CGAL, BOOST, MPI, among other pieces of software.
If you want to go with WSL, you can just run wsl myLinuxProgram --options.
Using WSL is the easiest way I believe as the current directory (PWD), is the current one i.e. the same as the PWD of your Qt app.
You can read Microsoft documenation for more info: https://learn.microsoft.com/en-us/windows/wsl/interop
If your linux binary depends on a lots of things, I really suggest you use docker for windows. Then, you have chance to pre-build an own docker image which put all dependency software also the linux binary you need to run in it.
Of course, to let your customer to use it, you should put it to dockerhub, register an account for yourself.
Then, the solution is simple: let the QT application to call docker run to setup a container base on your own image, execute it, and also let the linux binary to write the log or others to the bind mount volume among linux container & windows. After it run, the QT application fetch the linux binary output from this shared folder.
Finally, I give a minimal workable example for your reference:
Suppose the shared folder between windows & linux container is: C:\\abc\\log_share, it will mapped to linux container as /tmp folder. Of course you need to allow volume share by right click the docker icon in windows tray area & choose settings, like described here
Simplify the windows application as bat file, and simplfy the docker image as ubuntu, you should use your own prebuilt docker image with all dependency in it:
win_app.bat:
ECHO OFF
::New a shared folder with linux container
RD /s/q C:\\abc\\log_share > NUL 2>&1
MKDIR C:\\abc\\log_share
::From windows call docker to execute linux command like 'echo'
echo "Start to run linux binary in docker container..."
docker run -it -v C:\\abc\\log_share:/tmp ubuntu:16.04 bash -c "echo 'helloworld' > /tmp/linux_log_here.txt"
::In windows, get the log from shared bind mount from linux
echo "Linux binary run finish, print the log generated by the container..."
type C:\\abc\\log_share\linux_log_here.txt
Simplify the linux binary just as echo command in linux, the output things should be all write to shared directory:
echo 'helloworld' > /tmp/linux_log_here.txt
Now, execute the bat file with command win_app.bat:
C:\abc>win_app.bat
C:\abc>ECHO OFF
"Start to run linux binary in docker container..."
"Linux binary run finish, print the log generated by the container..."
helloworld
You can see the windows application already could fetch things(here is helloworld) which generated by linux binary from docker container.
I am trying to execute git commands with a public SSH key using the system() function in R (on Windows 7 64 bit). When I try to execute a git command I get the following error:
ssh_askpass: exec(rpostback-askpass): No such file or directory
Host key verification failed.
fatal: Could not read from remote repository.
Please make sure you have the correct access rights
and the repository exists.
I can run the exact same command on the command prompt and it works. I can also run the exact same command on the git bash shell and it works. I am not sure why cmd can't find the key when run from R using the system() command. Is there a way to specify the key location for the command prompt used by the system() command?
Update:
An example call that won't work:
system("git push --set-upstream git#gitlab.mygit.com:MyGroup/testGitlab.git master")
also
system("git clone git#gitlab.mygit.com:MyGroup/gitLabTest.git")
Both of these calls get the same error above.
Update 2: It appears that the ssh keys are not visible from R using list.files()
> list.files("C:/Users/Me/Documents/.ssh/", all.files = TRUE)
[1] "." ".."
I have verified that there are multiple ssh keys in this directory which are visible from the git bash shell using ls -a.
As it turns out, on my Windows system (7, 64 bit), when creating SSH keys with the git bash shell they were saved to "C:\Users\MyAcount\.ssh". However, the cmd.exe called by system() in R was looking in "C:\Users\MyAccount\Documents\.ssh".
For some reason, when using the cmd prompt called directly from the OS it looks in the same .ssh directory as the bash shell. Go Figure. Windows, I don't think I will ever understand you. Hope this helps someone out there.
How does R's built-in system() function know where to look to invoke some arbitrary OS command specified by the command argument? For example, if I homebrew install some_command_line_program, how does R's system() function know where it is located when I type:
cmd <- "some_complicated_code_from_some_command_line_program"
system(cmd, wait=FALSE)
In other words, how is R smart enough to know where to look without any user input? If I compile from source via Github (instead of homebrew install), would system() also recognize the command?
What system does depends on your OS, you've not told us (although you've given us some clues...).
On unix-alike systems, it executes it as a command in a bash shell, which searches for the first match in the directories on the $PATH environment variable. You can see what that is in R:
> Sys.getenv("PATH")
[1] "/usr/local/heroku/bin:/usr/local/heroku/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/nobackup/rowlings/NEST4B"
In Windows, it does something else.
You can get a full path to whatever it might run with Sys.which, which uses the systems' which command on unixes and fakes it on Windows. Read the help for more.
If you compile something from source then it will be found if the file that runs the command (a shell script, an executable, a #!-script in any language) is placed in a folder in your $PATH. You can create a folder yourself, say /home/user/bin, put your executables in there, add that to your $PATH, and (possibly after logging out an in again, or restarting R, or just firing up a new shell...) then R will find it.
I'm trying to run a R script through a .bat file. When I run myself the commands line by line it works. But when I try to run the .bat file, it doesn't works.
This is the .bat file
cd "C:\Program Files\R\R-3.1.2\bin"
R CMD BATCH "C:\Users\Administrator\Downloads\testa_vps.R"
This is the R script
setwd('C:\Users\Administrator\Documents')
file.create('mycsv.csv')
I'm not an expert with Windows and generally try to stick to Unix-like systems for things like this, but I have found that using programs non-interactively (e.g. via .bat files) is usually less error-prone when you add the appropriate directories to your (user) PATH variable, rather than cding into the directory and calling the executable from within the .bat file. For example, among other things, my user PATH variable contains C:\PROGRA~1\R\R-3.0\bin\; - the directory that contains both R.exe and Rscript.exe - (where PROGRA~1 is an alias for Program Files which you can use in an unquoted file path, since there are no spaces in the name).
After you do this, you can check that your PATH modification was successful by typing Rscript in a new Command Prompt - it should print out usage information for Rscript rather than the typical xxx is not recognized as an internal or external command... error message.
In the directory C:\Users\russe_000\Desktop\Tempfiles, I created test_r_script.r, which contains
library(methods)
setwd("C:\Users\russe_000\Desktop\Tempfiles")
file.create("mycsv.csv")
and test_r.bat, which contains
Rscript --vanilla --no-save "C:\Users\russe_000\Desktop\Tempfiles\test_r_script.r"
Clicking on the Windows Batch File test_r ran the process successfully and produced mycsv.csv in the correct folder.
Before running test_r.bat:
After running test_r.bat:
I've never worked with a Windows server, but I don't see why the process would be fundamentally different than on a personal computer; you just may need your sysadmin to modify the PATH variable if you don't have sufficient privileges to alter environment variables.
As already suggested by #nrussel in the comments you should use RScript.exe for this.
Create a file launcher.bat with the following content:
cd C:\Users\Administrator\Documents
Rscript testa_vps.R
In addition, add C:\Program Files\R\R-[your R version]\bin\x64; or C:\Program Files\R\R-[your R version]\bin\i386to the System PATH variable in the Environment Variables menu depending if you run R on a 64-bit or 32-bit system.
I just tested the approach above successfully on a Windows Server 2008 64-bit system and mycsv.csv got created as expected.
EDIT
One important point I forgot to mention is the following: You need to specify the path in your R file in the setwd() call using \\ instead of \.
setwd('C:\\Users\\Administrator\\Documents')
Here is a screenshot of the successful run on the Windows 2008 server:
Note: I added cmd /k to the .bat file so that the cmd window stays open after clicking on the file.
I just downloaded the Sqlite3 command line shell for windows (http://www.sqlite.org/download.html). However when I try to run the executable in Cygwin it does not load and just hangs.
The Sqlite3 shell itself works as if I browse to it in Windows explorer it runs or if I open it in cmd it runs. How do I get it to work with Cygwin?
Cygwin currently doesn't work well with interactive native programs. It has to do with the way the Cygwin terminal code works, in order to provide expected POSIX semantics to Cygwin programs.
Therefore, your best bet is to use Cygwin's own SQLite package.
If you absolutely must use the native sqlite3.exe, you'll have use cmd.exe to run sqlite3.