I installed msys2-x86_64-20161025.exe following these steps.
When I try autocompletion for programs on the PATH, e.g. by typing g and then pressing TAB key twice, Msys2 console shows me a lot of options (.exe , .dll , and others).
However, when I type ./ and then press TAB key twice, it shows me only subdirectories of the current directory, but not .sh or .bat files in the current directory.
ls -al command shows that Msys2 doesn't see "executable" file permission on the .bat and .sh files. chmod +x * doesn't seem to work on Windows.
So how to get autocomplete to work in Msys2 for current directory script files?
I am on Windows 10 x64. Let me know if some additional info about the environment is needed.
If your script contains a shebang line, msys2 will recognize the file as a script. That's has been my experience at least.
Particularly for Bash scripts the first line of each script can be #!/bin/bash, if Bash is installed in /bin/. See more on Shebang line here.
Related
I want to copy a folder ~/Projects/LocalProject onto my server //VM-Server/ServerProject.
I know that I can use GitBash:
cp -r directory-name-1 directory-name-2
But what I'm curious about is, can I create a script to do that by double clicking that script, or adding it as a command to my GitBash, cause I will need that alot?
--Edit--
Tried nothing, as I don't know how to do that. Yes there are hidden files, I don't want them to be copied. There shouldn't be newer files on the destination. I need to manually run it, I thought that's clear as I mentioned the option to have a executable script / or a terminal command.
Option 1: Batch file
You don't even need git-bash; you can make a batch file in any text editor, name it copy to server.bat, and type in cp C:\Users\<Your username>\Projects\LocalProject \\VM-Server\ServerProject.
You can also make a .sh file for use in bash. The command is the same, just make note that Windows uses \, while bash uses / for directory tree
Option 2: Alias
Open your bash_profile file (it's in your git bash install location).
Add a line at the end of the file that says alias copyToServer = 'cp ~/Projects/LocalProject //VM-Server/ServerProject'. Then close git-bash, reopen it and use the command by typing copyToServer as a bash command. (It doesn't need to be named copyToServer)
I just installed CD-HIT and followed these instructions:
Installation
Most CD-HIT programs were written in C++. Installing CD-HIT package is very simple:
download current CD-HIT at http://bioinformatics.org/cd-hit, for example cd-hit-2006-0215.tar.gz
unpack the file with ” tar xvf cd-hit-2006-0215.tar.gz --gunzip”
change dir by “cd cd-hit-2006”
compile the programs by “make”
you will have all cd-hit programs compiled
I followed these steps and indeed .exe files were made:
cdhit.c++ cdhit-est-2d.c++ clstr_select_rep.pl
cd-hit.exe cd-hit-est-2d.exe clstr_size_histogram.pl
cdhit.o cdhit-est-2d.o clstr_size_stat.pl
cdhit-2d.c++ cd-hit-para.pl clstr_sort_by.pl
cd-hit-2d.exe cdhit-utility.c++ clstr_sort_prot_by.pl
cdhit-2d.o cdhit-utility.h clstr_sql_tbl.pl
cd-hit-2d-para.pl cdhit-utility.o clstr_sql_tbl_sort.pl
cdhit-454.c++ ChangeLog clstr2tree.pl
cd-hit-454.exe clstr_cut.pl clstr2txt.pl
cdhit-454.o clstr_list.pl clstr2xml.pl
cd-hit-auxtools clstr_list_sort.pl doc
cdhit-common.c++ clstr_merge.pl FET.pl
cdhit-common.h clstr_merge_noorder.pl license.txt
cdhit-common.o clstr_quality_eval.pl make_multi_seq.pl
cdhit-div.c++ clstr_quality_eval_by_link.pl Makefile
cd-hit-div.exe clstr_reduce.pl plot_2d.pl
cdhit-div.o clstr_renumber.pl plot_len1.pl
cd-hit-div.pl clstr_rep.pl psi-cd-hit
cdhit-est.c++ clstr_reps_faa_rev.pl README
cd-hit-est.exe clstr_rev.pl usecases
cdhit-est.o clstr_select.pl
However when I typ: cd-hit-est in the cygwin command line, it will say: -bash: cd-hit.exe: command not found. I'm not able to figure out why this is happening. Even when I'm in the folder were the .exe files are located it still will give the same error.
When you try to run command without a path, bash tries to find the command name in each directory from the PATH environment variable.
Usually current directory "." is not in the PATH so you have to run command adding absolute or relative path before command name:
./cd-hit.exe
or
/absolute/path/to/your/cd-hit/program/cd-hit.exe
If you want to permanently run command cd-hit.exe from anywhere, just add the directory where cd-hit was compiled to any existing auto-loaded bash profile files:
.profile
.bashrc
.bash_profile
like
PATH=$PATH:/absolute/path/to/your/cd-hit/program
export PATH
I just made a .c file and compiled it with gcc in the terminal on OS X 10.8.2.
My syntax was gcc -o <filename> <sourcefile> and that was fine. Now I see I have an executable and file <filename> tells me as such, but I'm not sure how to actually run it beside double clicking the icon, /lame. When I try running the executable by just typing I get command not found, but I thought it was executable...?
Thank you.
EDIT: I"m so stupid, it was just open <filename>
Unix will only run commands if they are available on the system path, as you can view by the $PATH variable
echo $PATH
Executables located in directories that are not on the path cannot be run unless you specify their full location. So in your case, assuming the executable is in the current directory you are working with, then you can execute it as such
./my-exec
Where my-exec is the name of your program.
To run an executable in mac
1). Move to the path of the file:
cd/PATH_OF_THE_FILE
2). Run the following command to set the file's executable bit using the chmod command:
chmod +x ./NAME_OF_THE_FILE
3). Run the following command to execute the file:
./NAME_OF_THE_FILE
Once you have run these commands, going ahead you just have to run command 3, while in the files path.
I have a question I have been trying to fix for a while. I want to understand what's the difference between starting a script from the command line and making it executable and then running it from the Finder.
Because this is what I am experiencing;
I have a simple script called trash-files which contains this command:
trash ~/Downloads/*
When I run from the terminal it works as expected; however if I doubleclick the shell script in the finder I see this:
/Users/xx/Desktop/trash-files: line 1: trash: command not found
I hope anyone can tell me why this doesn't work as expected
trash is not a standard command in OS X. Is it something defined in your ~/.profile or a similar file? If so, these are not run for non-login shells, such as those created to run a script.
If you're using homebrew, you could run
brew install trash
which would install the necessary scripts to have the trash command work in the way you're expecting.
There is a folder in your home folder location called
.Trash
The "dot" in front of the folder name makes it hidden while searching for it in finder. You'll have to use Terminal to execute the following command:
cd ~/
ls -la
This will change the directory to the current logged in users home folder, then second command will list files and show hidden files. You can then run:
rm .Trash/*
This will remove everything inside the Trashcan on the dock.
So open TextEdit from the /Applications folder, go to "Format" and make it "Plain Text". Paste in the two lines below.
#!/bin/sh
rm ~/.Trash/*
Save the file as "emptyTrash.sh" (uncheck use txt extension). Save it to your Desktop or wherever you'd like. Then open Terminal, cd (change directory) to where the files is and run this command to make the script executable:
chmod +x emptyTrash.sh
Then you can execute the script by cd (changing directory) to path where the script is, and run:
./emptyTrash.sh
That's it.
Is it possible to run Matlab commands from Mac command line?
The matlab script is in the bin subdirectory of the MATLAB app bundle. On my machine, this means I can run it like so:
/Applications/MATLAB_R2012a_Student.app/bin/matlab
If you want this bin directory on your path (so that you can just run matlab, mex, etc), edit or create a new text file called .bash_profile in the top level of your home directory with the following line:
export PATH=/Applications/MATLAB_R2012a_Student.app/bin:$PATH
Replacing the "MATLAB_R2012a_Student" part with the name of your actual MATLAB app bundle. This will not come into effect for currently open terminals, but newly opened terminals should work properly.
You need the full path to the MATLAB executable, and you can use the -r option to run a command in the MATLAB that you start, as per the doc here.
As already mentioned above you need to first edit your .bash_profile file by adding the following line (replace 'MATLAB_R2020b' with your MATLAB version)
export PATH=/Applications/MATLAB_R2020b.app/bin:$PATH
Then after restarting the terminal you can open MATLAB by inserting the command
/Applications/MATLAB_R2020b.app/bin/matlab
You can also run your .m scripts by defining the working directory folder and the executable scripts and their paths. You just need to add more commands to the above.
/Applications/MATLAB_R2020b.app/bin/matlab -r "addpath(genpath('{Your working directory folder path}')); cd {Your working directory folder path}; {Your script name}; {Your other script name}; quit;"
Please find a more detailed description of MATLAB command line arguments from:
https://www.mathworks.com/help/matlab/ref/matlabmacos.html#d122e801165