Emacs flymake on Mac Yosemite - path

Flymake in emacs was properly working until update to Yosemite. But now it's complaining:
Flymake: Failed to launch syntax check process 'pychecker' with args (<filename>_flymake.py): Searching for program: no such file or directory, pychecker. Flymake will be switched OFF.
where <filename> is the name of the file in the opened buffer.
Here's my flymake config:
(add-to-list 'load-path "~/.emacs.d/")
;; Setup for Flymake code checking.
(require 'flymake)
(load-library "flymake-cursor")
;; Script that flymake uses to check code. This script must be
;; present in the system path.
(setq pycodechecker "pychecker")
(when (load "flymake" t)
(defun flymake-pycodecheck-init ()
(let* ((temp-file (flymake-init-create-temp-buffer-copy
'flymake-create-temp-inplace))
(local-file (file-relative-name
temp-file
(file-name-directory buffer-file-name))))
(list pycodechecker (list local-file))))
(add-to-list 'flymake-allowed-file-name-masks
'("\\.py\\'" flymake-pycodecheck-init)))
(add-hook 'python-mode-hook 'flymake-mode)
here's /usr/local/bin/pychecker:
#! /bin/sh
pyflakes "$1"
pep8 --repeat "$1" --max-line-length=80 --ignore=E123,E133,E226,E501
true
here's $PATH:
/Users/<user>/Envs/<venv>/bin:/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin:/opt/X11/bin:/usr/texbin
where <user> is my user name and <venv> is the currently active virtual-env.
pychecker works properly if run from the shell.
I start emacs from shell by typing emacsgui (alias emacsgui='open -a emacs'), usually with the venv activated. I also tried opening emacs without any venv activated but the problem still occur. What is the problem?

I solved by adding this to my .emacs file:
(defun set-exec-path-from-shell-PATH ()
(let ((path-from-shell (replace-regexp-in-string
"[ \t\n]*$"
""
(shell-command-to-string "$SHELL --login -i -c 'echo $PATH'"))))
(setenv "PATH" path-from-shell)
(setq exec-path (split-string path-from-shell path-separator))))
(when (and window-system (eq system-type 'darwin))
;; When started from Emacs.app or similar, ensure $PATH
;; is the same the user would see in Terminal.app
(set-exec-path-from-shell-PATH))

Related

Adaptive Ruby version when opening a terminal

This question is the opposite of this one (also asked here, here and here)
I have two versions of ruby installed
ubuntu:~/environment $ rvm list
ruby-2.6.6 [ x86_64 ]
=* ruby-3.0.2 [ x86_64 ]
and every time I open a terminal window, ruby-3.0.2 (the default) is set. The problem is for a couple of my older projects I have to use ruby-2.6.6, so every time I have to switch with
rvm use 2.6.6
Is there a way to automatically select ruby 2.6.6 when I open the terminal window of the specific projects? I have tried to override the default rvm version with the .ruby-version file (as suggested here) but it does not do the trick.
EDIT File /home/ubuntu/.rvm/scripts/cd contains the following
#!/usr/bin/env bash
# Source a .rvmrc file in a directory after changing to it, if it exists. To
# disable this feature, set rvm_project_rvmrc=0 in /etc/rvmrc or $HOME/.rvmrc
case "${rvm_project_rvmrc:-1}" in
1|cd)
# cloned from git#github.com:mpapis/bash_zsh_support.git
source "$rvm_scripts_path/extras/bash_zsh_support/chpwd/function.sh"
# not using default loading to support older Zsh
[[ -n "${ZSH_VERSION:-}" ]] &&
__rvm_version_compare "$ZSH_VERSION" -gt 4.3.4 ||
{
function cd() { __zsh_like_cd cd "$#" ; }
function popd() { __zsh_like_cd popd "$#" ; }
function pushd() { __zsh_like_cd pushd "$#" ; }
}
__rvm_after_cd()
{
\typeset rvm_hook
rvm_hook="after_cd"
if [[ -n "${rvm_scripts_path:-}" || -n "${rvm_path:-}" ]]
then source "${rvm_scripts_path:-$rvm_path/scripts}/hook"
fi
}
__rvm_cd_functions_set()
{
__rvm_do_with_env_before
if [[ -n "${rvm_current_rvmrc:-""}" && "$OLDPWD" == "$PWD" ]]
then rvm_current_rvmrc=""
fi
__rvm_project_rvmrc >&2 || true
__rvm_after_cd || true
__rvm_do_with_env_after
return 0
}
[[ " ${chpwd_functions[*]} " == *" __rvm_cd_functions_set "* ]] ||
chpwd_functions=( "${chpwd_functions[#]}" __rvm_cd_functions_set )
# This functionality is opt-in by setting rvm_cd_complete_flag=1 in ~/.rvmrc
# Generic bash cd completion seems to work great for most, so this is only
# for those that have some issues with that.
if (( ${rvm_cd_complete_flag:-0} == 1 ))
then
# If $CDPATH is set, bash should tab-complete based on directories in those paths,
# but with the cd function above, the built-in tab-complete ignores $CDPATH. This
# function returns that functionality.
_rvm_cd_complete ()
{
\typeset directory current matches item index sep
sep="${IFS}"
export IFS
IFS=$'\n'
COMPREPLY=()
current="${COMP_WORDS[COMP_CWORD]}"
if [[ -n "$CDPATH" && ${current:0:1} != "/" ]] ; then
index=0
# The change to IFS above means that the \command \tr below should replace ':'
# with a newline rather than a space. A space would be ignored, breaking
# TAB completion based on CDPATH again
for directory in $(printf "%b" "$CDPATH" | \command \tr -s ':' '\n') ; do
for item in $( compgen -d "$directory/$current" ) ; do
COMPREPLY[index++]=${item#$directory/}
done
done
else
COMPREPLY=( $(compgen -d ${current}) )
fi
IFS="${sep}";
}
complete -o bashdefault -o default -o filenames -o dirnames -o nospace -F _rvm_cd_complete cd
fi
;;
2|prompt)
if
[[ -n "${ZSH_VERSION:-}" ]]
then
precmd_functions+=(__rvm_do_with_env_before __rvm_project_rvmrc __rvm_do_with_env_after)
else
PROMPT_COMMAND="${PROMPT_COMMAND%% }"
PROMPT_COMMAND="${PROMPT_COMMAND%%;}"
PROMPT_COMMAND="${PROMPT_COMMAND:-}${PROMPT_COMMAND:+; }__rvm_do_with_env_before; __rvm_project_rvmrc; __rvm_do_with_env_after"
fi
;;
esac
You are probably using RVM as a shell script, and not as a shell function.
You can check like this in a typical shell (bash, zsh, ...) : execute: type rvm
If it displays rvm is /home/ying/.rvm/bin/rvm : you are using as a script (found in $PATH)
If it displays rvm is a function : you are using as a function (much better).
Check out: https://rvm.io/rvm/basics#post-install-configuration
If you are using as a script, and want to use as a function: you need to "source" the rvm function, it is located in <rvm main folder>/scripts/rvm, for instance if installed in $HOME:
source $HOME/.rvm/scripts/rvm
source /usr/local/rvm/scripts/rvm
Typically, at RVM installation time, it add the following line in the equivalent of .profile (depending on shell and if its global or user):
[[ -s "$HOME/.rvm/scripts/rvm" ]] && source "$HOME/.rvm/scripts/rvm" # Load RVM into a shell session *as a function*
RVM changes automatically the version as described here:
https://rvm.io/workflow/projects
For detection of .ruby-version, the following is required:
RVM must be a recent version that supports the feature
RVM must be loaded in the shell (typically by .profile, or equivalent) so that is is executed as a function
the shell must be compatible with this callback feature (bash and zsh are)
Here is what happens:
When you load rvm as a function, it registers callback in the shell
when you cd into the project, RVM callbacks (in shell) detect the file .ruby-version (or others) and automatically do the equivalent of rvm use.
For instance, I use zsh (on osx) which has preexec and precmd callbacks
and it detect the ruby version file and applies when I cd into it or a sub folder.
It works with bash too.
If you are curious or want to see why it does not work for you look at the file <rvm main dir>/scripts/cd
typically the shell variable chpwd_functions is set to __rvm_cd_functions_set
which is the function called after a cd by rvm

xonsh "which" equivalent - how to test if a (subprocess mode) command is available?

I would like to test (from xonsh) if a command is available or not. If I try this from the xonsh command prompt:
which bash
Then it works:
user#server ~ $ which bash
/usr/bin/bash
But it does not work from xonsh script:
#!/usr/bin/env xonsh
$RAISE_SUBPROC_ERROR = True
try:
which bash
print("bash is available")
except:
print("bash is not available")
Because it results in this error:
NameError: name 'which' is not defined
I understand that which is a shell builtin. E.g. it is not an executable file. But it is available at the xnosh command prompt. Then why it is not available inside an xonsh script? The ultimate question is this: how can I test (from an xonsh script) if a (subprocess mode) command is available or not?
import shutil
print(shutil.which('bash'))
While nagylzs' answer led me to the right solution, I found it inadequate.
shutil.which defaults to os.environ['PATH']. On my machine, the default os.environ['PATH'] doesn't contain the active PATH recognized by xonsh.
~ $ os.environ['PATH']
'/usr/bin:/bin:/usr/sbin:/sbin'
I found I needed to pass $PATH to reliably resolve 'which' in the xonsh environment.
~ $ $PATH[:2]
['/opt/google-cloud-sdk/bin', '/Users/jaraco/.local/bin']
~ $ import shutil
~ $ shutil.which('brew', path=os.pathsep.join($PATH))
'/opt/homebrew/bin/brew'
The latest version of xonsh includes a built-in which command. Unfortunately, the version included will emit an error on stdout if the target isn't found, a behavior that is not great for non-interactive use.
As mentioned in another answer, which exists in the current version of xonsh (0.13.4 as of 15/12/2022) so your script would work. However, it outputs its own error message so it's necessary to redirect stderr to get rid of it.
Also, unless you redirect its stdout as well (using all>), it migh be a good idea to capture its output so the final version would look like this:
#!/usr/bin/env xonsh
$RAISE_SUBPROC_ERROR = True
try:
bash = $(which bash err> /dev/null)
print(f"bash is available: {bash}")
except:
print("bash is not available")

How to set environment variable in Mac 10.14.6 Mojave with 'Application Support' in pathfile?

I'm having trouble setting an environment variable that has a pathfile containing a space ' ' character.
Before you ask, I've already tried enclosing the whole pathfile within double quotes, single quotes, no quotes but escaping with backspace.
Could it be the something to do with the encoding? The variable would be:
export A_MEDIA="/Users/polo/Library/Application Support/Anki2/me/collection.media"
once I source ~/.bash_profile, I try cd $A_MEDIA (with or without quoting the name of the variable). The response is:
-bash: cd: /Users/polo/Library/Application: No such file or directory
It's as if bash didn't know how to interpret that space between 'Application' and 'Support'. It thinks the path goes from a folder named Application to a folder named Support. It just doesn't see them as a single folder name. Any help? Please?
Works for me so I strongly suspect what you showed us in your problem statement is not what you're actually doing. I normally use fish so this shows me setting the env var before starting bash to show that it correctly inherits the var and also setting and using it inside bash:
12:21 macbook opencv3 ~ > set -x A_MEDIA $HOME/Library/Application\ Support/Dock/
12:22 macbook opencv3 ~ > bash
running .bashrc
bash-5.0$ cd $A_MEDIA
bash: cd: too many arguments
bash-5.0$ cd "$A_MEDIA"
bash-5.0$ pwd
/Users/krader/Library/Application Support/Dock
bash-5.0$ export B_MEDIA="$HOME/Library/Application Support/Gitter"
bash-5.0$ cd $B_MEDIA
bash: cd: too many arguments
bash-5.0$ cd "$B_MEDIA"
bash-5.0$ pwd
/Users/krader/Library/Application Support/Gitter
bash-5.0$ exit
Note that in a POSIX shell like bash you should almost always use double-quotes around a var expansion so that if it contains whitespace the expanded value is not split on that whitespace.

ejabber's erl file give error "segmentation fault: 11" while running on mac os 10.7.5

I am trying to start ejabber 16.05 server on mac os 10.7.5. while starting up, it gives error. on further investigation, I found that "erl" executor file shipped with ejabber is throwing "Segmentation Fault:11" while running independently. I firmly believe that resolving issue with "erl" file execution will solve server start up issue. can anyone please help. Below is the code from "erl" that is causing segmentation fault error
#!/bin/sh
ROOTDIR=/Applications/ejabberd-15.06
export ROOTDIR
BINDIR=$ROOTDIR/bin
export BINDIR
EMU=beam
export EMU
PROGNAME=$BINDIR/erl
export PROGNAME
PATH=$BINDIR:$PATH
export PATH
arch() {
case `uname -m` in
i[3456]86 ) echo x86 ;;
i86pc) echo x86 ;;
armv7*) echo armhf ;;
arm*l) echo armel ;;
* ) echo `uname -m | tr A-Z a-z` ;;
esac
}
os=`uname -s | tr A-Z a-z`
cpu=`arch`
ARCHDIR=${os}-${cpu}
export ARCHDIR
# Dynamic libraries
LD_LIBRARY_PATH=$ROOTDIR/lib/$ARCHDIR
export LD_LIBRARY_PATH
DYLD_LIBRARY_PATH=$LD_LIBRARY_PATH
export DYLD_LIBRARY_PATH
exec $BINDIR/erlexec ${1+"$#"}
I just see that you are using OSX 10.7. We do not test ejabberd binary installer on version of OS that are so old.
I do not even expect the binary installer to run on such an old release.
If you have to run ejabberd on that OSX version, you have to build it from source.
ejabberd source are available on official ejabberd Github.
Documentation for installation and build is here Installing ejabberd from source code.

Nim(rod) not installing to path properly?

I'm following the isntructions for installing Nim(rod) onto linux. I then followed this site which showed to do this:
$ echo 'export PATH=$PATH:$your_install_dir/bin' >> ~/.profile
$ source ~/.profile
$ nim
Typing nim in the terminal doesn't do anything.
I have a Nim folder in my home directory with all the files but can't use it.
I did echo export PATH=$PATH:$/home/bob/Nim/bin' >> ~/.profile
And the nim command still does nothing. Am I not doing it right?
The command pwd told me /home/bob/Nim when I was in the Nim folder.
I tried running nim with sh in Nim/bin and did ./nim -path:/home/bob/Nim/bin
and it said:
config/nim.cfg(45, 2) Hint: added path: '/home/bob/.babel/pkgs/' [Path]
config/nim.cfg(46, 2) Hint: added path: '/home/bob/.nimble/pkgs/' [Path]
Hint: used config file '/home/bob/Nim/config/nim.cfg' [Conf]
echo 'export PATH=$PATH:$/home/bob/Nim/bin' >> ~/.profile
You appended the location $/home/bob/Nim/bin which doesn't exist. You must remove the $.

Resources