Using Scite 4.1.3 on Windows 10.
I tried the following Lua script:
function TestFile()
mycontent = "Hello World"
mytmpfilename = os.tmpname()
os.execute("echo aaaa > " .. mytmpfilename) -- create file explicitly;
mytmpfile = io.popen(mytmpfilename, "w") -- w+ crashes Scite in Windows!
mytmpfile:write(mycontent)
mytmpfile:close()
print("readall " .. mytmpfilename .. ": " .. io.popen(mytmpfilename, "r"):read("*a"))
end
If I run this, I get printed:
readall C:\Users\ME\AppData\Local\Temp\s8qk.m:
... which means Lua could not even read this file?! And also, this stupid Windows Explorer prompt shows up:
At end, the content of the C:\Users\ME\AppData\Local\Temp\s8qk.m is still just aaaa.
So obviously, mytmpfile:write part fails silently, and nothing new is written in the file - the only thing that wrote to the file is the echo aaaa > ... executed by cmd.exe via os.execute.
So my question is - how can I write a file with Lua in Scite on Windows? Preferably, without having that stupid "How do you want to open this file?" prompt show up?
Eh, I think I got it ...
See, the OP example uses io.popen - and, if we look at https://man7.org/linux/man-pages/man3/popen.3.html it says:
popen, pclose - pipe stream to or from a process
(emphasis mine).
So, basically, if under Windows I try to do io.popen(filename), then apparently tries to find the process that would be the default "opener" for that file type ... and also therefore the prompt I'm being shown in the OP (and therefore, I could never read or write the files accessed -- or rather, not accessed -- in that way).
However, Programming in Lua : 21.2 – The Complete I/O Model actually uses io.open (notice, no p for process); and then the files seem to open for read/write fine.
So, corrected example from OP should be:
function TestFile()
mycontent = "Hello World"
mytmpfilename = os.tmpname()
-- os.execute("echo aaaa > " .. mytmpfilename) -- create file explicitly; -- no need anymore, with io.open
mytmpfile = io.open(mytmpfilename, "w+") -- w+ crashes Scite in Windows, but only if using io.popen; is fine with io.open!
mytmpfile:write(mycontent)
mytmpfile:close()
print("readall " .. mytmpfilename .. ": " .. io.open(mytmpfilename, "r"):read("*a"))
end
Related
I'm new to ruby and this issue is bugging me for a while . Whenever i use gets to take user input , my gets statement is executed right after i run the file .I'm using git Bash to run my file.rb file ,
puts "some unnecessary text"
puts "Hello world"
puts "now you should input something"
x = gets.chomp
puts 36
puts "your input is " + x + " right?"
the program should print the first 3 lines before waiting for an input but it waits for the input right after i run it
$ruby file.rb
|
it waits for eternity unless I press enter . If i write something,
$ ruby file.rb
myInput
some unnecessary text
Hello world
now you should input something
36
your input is myInput right?
it runs okay . So I'm forced to write my input at the beginning .
It's not much of a problem right now but it'll cause a lot if headaches when i write bigger and more complex code . Any solutions ?
ps: It seems the problem only occurs with git Bash (windows) . Powershell works just fine .
It seems that the standard output is buffered.
Try to put at the beginning of the file (first two lines) old_sync = $stdout.sync $stdout.sync = true and a the end of the file (last line) $stdout.sync = old_sync.
The call to the IO#sync= method set the sync mode to true. This cause that all output is immediately flushed, at the end of the script we restore its value to its original old value, see Ruby documentation for details.
In summary:
old_sync = $stdout.sync # cache old value
$stdout.sync = true # set mode to true
# your scripting staff
$stdout.sync = old_sync # restore old value
If this trick works at least you know the reason for the weird behaviour. You can find some explanation also in this SO post.
The idea of the following was to use Bash's select from Python, e.g. use Bash select to get the input from the user, communicate with the Bash script to get the user selections and use it afterwords in the Python code. Please tell me if it at least possible.
Have the following simple Bash script:
#!/bin/bash -x
function select_target {
target_list=("Target1" "Target2" "Target3")
PS3="Select Target: "
select target in "${target_list[#]}"; do
break
done
echo $target
}
select_target
it works standalone
Now I tried to call it from Python like this:
import tempfile
import subprocess
select_target_sh_func = """
#!/bin/bash
function select_target {
target_list=(%s)
PS3="Select Target: "
select target in "${target_list[#]}"; do
break
done
echo $target
}
select_target
"""
target_list = ["Target1", "Target2", "Target3"]
with tempfile.NamedTemporaryFile() as temp:
temp.write(select_target_sh_func % ' '.join(map(lambda s : '\"%s\"' % str(s),target_list)))
subprocess.call(['chmod', '0777', temp.name])
sh_proc = subprocess.Popen(["bash", temp.name], stdout=subprocess.PIPE)
(output, err) = sh_proc.communicate()
exit_code = sh_proc.wait()
print output
It does nothing. No output, no selection.
I'm using High Sierra MacOS, PyCharm and Python 2.7.
PS
After some reading and experimenting ended up with the following:
with tempfile.NamedTemporaryFile() as temp:
temp.write(select_target_sh_func % ' '.join(map(lambda s : '\"%s\"' % str(s),target_list)))
temp.flush()
# bash: /var/folders/jm/4j4mq_w52bx2l5qwg4gt44580000gn/T/tmp00laDV: Permission denied
subprocess.call(['chmod', '0500', temp.name])
sh_proc = subprocess.Popen(["bash", "-c", temp.name], stdout=subprocess.PIPE)
(output, err) = sh_proc.communicate()
exit_code = sh_proc.wait()
print output
It behaves as I expected it would, the user is able to select the 'target' by just typing the number. My mistake was that I forgot to flush.
PPS
The solution works for MacOS X High Sierra, sadly it does not for Debian Jessie complaining the following:
bash: /tmp/tmpdTv4hp: Text file busy
I believe it is because `with tempfile.NamedTemporaryFile' keeps the temp file open and this somehow prevents Bash from working with it. This renders the whole idea useless.
Python is sitting between your terminal or console and the (noninteractive!) Bash process you are starting. Furthermore, you are failing to direct the standard output pipe anywhere, so subprocess.communicate() actually cannot capture standard error (and if it could, you would not be able to see the script's menu).
Running an interactive process programmatically is a nontrivial scenario; you'll want to look at pexpect or just implement your own select command in Python - I suspect this is going to turn out to be the easiest solution (trivially so if you can find an existing library).
I have a file called backup.lua, which the program should write to every so often in order to backup its status, in case of a failure.
The problem is that the program writes the backup.lua file completely fine first-time round, but any other times it refuses to write to the file.
I tried removing the file while the program was still open but Windows told me that the file was in use by 'CrysisWarsDedicatedServer.exe', which is the program. I have told the host Lua function to close the backup.lua file, so why isn't it letting me modify the file at will after it has been closed?
I can't find anything on the internet (Google actually tried to correct my search) and the secondary programmer on the project doesn't know either.
So I'm wondering if any of you folks know what we are doing wrong here?
Host function code:
function ServerBackup(todo)
local write, read;
if todo=="write" then
write = true;
else
read = true;
end
if (write) then
local source = io.open(Root().."Mods/Infinity/System/Read/backup.lua", "w");
System.Log(TeamInstantAction:GetTeamScore(2).." for 2, and for 1: "..TeamInstantAction:GetTeamScore(1))
System.LogAlways("[System] Backing up serverdata to file 'backup.lua'");
source:write("--[[ The server is dependent on this file; editing it will lead to serious problems.If there is a problem with this file, please re-write it by accessing the backup system ingame.--]]");
source:write("Backup = {};Backup.Time = '"..os.date("%H:%M").."';Backup.Date = '"..os.date("%d/%m/%Y").."';");
source:write(XFormat("TeamInstantAction:SetTeamScore(2, %d);TeamInstantAction:SetTeamScore(1, %d);TeamInstantAction:UpdateScores();",TeamInstantAction:GetTeamScore(2), TeamInstantAction:GetTeamScore(1) ));
source:close();
for i,player in pairs(g_gameRules.game:GetPlayers() or {}) do
if (IsModerator(player)) then
CMPlayer(player, "[!backup] Completed server backup.");
end
end
end
--local source = io.open(Root().."Mods/Infinity/System/Read/backup.lua", "r"); Can the file be open here and by the Lua scriptloader too?
if (read) then
System.LogAlways("[System] Restoring serverdata from file 'backup.lua'");
--source:close();
Backup = {};
Script.LoadScript(Root().."Mods/Infinity/System/Read/backup.lua");
if not Backup or #Backup < 1 then
System.LogAlways("[System] Error restoring serverdata from file 'backup.lua'");
end
end
end
Thanks all :).
Edit:
Although the file is now written to the disk fine, the system fails to read the dumped file.
So, now the problem is that the "LoadScript" function isn't doing what you expect:
Because I'm psychic, i have divined that you're writing a Crysis plugin, and are attempting to use it's LoadScript API call.
(Please don't assume everyone here would guess this, or be bothered to look for it. It's vital information that must form part of your questions)
The script you're writing attempts to set Backup - but your script, as written - does not separate lines with newline characters. As the first line is a comment, the entire script will be ignored.
Basicallty the script you've written looks like this, which is all treated as a comment.
--[[ comment ]]--Backup="Hello!"
You need to write a "\n" after the comment (and, I'd recommend in other places too) to make it like this. In fact, you don't really need block comments at all.
-- comment
Backup="Hello!"
I want to be able to read a currently open file. The test.rb is sending its output to test.log which I want to be able to read and ultimately send via email.
I am running this using cron:
*/5 * * * /tmp/test.rb > /tmp/log/test.log 2>&1
I have something like this in test.rb:
#!/usr/bin/ruby
def read_file(file_name)
file = File.open(file_name, "r")
data = file.read
file.close
return data
end
puts "Start"
puts read_file("/tmp/log/test.log")
puts "End"
When I run this code, it only gives me this output:
Start
End
I would expect the output to be something like this:
Start
Start (from the reading of the test.log since it should have the word start already)
End
Ok, you're trying to do several things at once, and I suspect you didn't systematically test before moving from one step to the next.
First we're going to clean up your code:
def read_file(file_name)
file = File.open(file_name, "r")
data = file.read
file.close
return data
end
puts "Start"
puts read_file("/tmp/log/test.log")
puts "End"
can be replaced with:
puts "Start"
puts File.read("./test.log")
puts "End"
It's plain and simple; There's no need for a method or anything complicated... yet.
Note that for ease of testing I'm working with a file in the current directory. To put some content in it I'll simply do:
echo "foo" > ./test.log
Running the test code gives me...
Greg:Desktop greg$ ruby test.rb
Start
foo
End
so I know the code is reading and printing correctly.
Now we can test what would go into the crontab, before we deal with its madness:
Greg:Desktop greg$ ruby test.rb > ./test.log
Greg:Desktop greg$
Hmm. No output. Something is broken with that. We knew there was content in the file previously, so what happened?
Greg:Desktop greg$ cat ./test.log
Start
End
Cat'ing the file shows it has the "Start" and "End" output of the code, but the part that should have been read and output is now missing.
What happening is that the shell truncated "test.log" just before it passed control to Ruby, which then opened and executed the code, which opened the now empty file to print it. In other words, you're asking the shell to truncate (empty) it just before you read it.
The fix is to read from a different file than you're going to write to, if you're trying to do something with the contents of it. If you're not trying to do something with its contents then there's no point in reading it with Ruby just to write it to a different file: We have cp and/or mv to do those things for us witout Ruby being involved. So, this makes more sense if we're going to do something with the contents:
ruby test.rb > ./test.log.out
I'll reset the file contents using echo "foo" > ./test.log, and cat'ing it showed 'foo', so I'm ready to try the redirection test again:
Greg:Desktop greg$ ruby test.rb > ./test.log.out
Greg:Desktop greg$ cat test.log.out
Start
foo
End
That time it worked. Trying it again has the same result, so I won't show the results here.
If you're going to email the file you could add that code at this point. Replacing the puts in the puts File.read('./test.log') line with an assignment to a variable will store the file's content:
contents = File.read('./test.log')
Then you can use contents as the body of a email. (And, rather than use Ruby for all of this I'd probably do it using mail or mailx or pipe it directly to sendmail, using the command-line and shell, but that's your call.)
At this point things are in a good position to add the command to crontab, using the same command as used on the command-line. Because it's running in cron, and errors can happen that we'd want to know about, we'd add the 2>&1 redirect to capture STDERR also, just as you did before. Just remember that you can NOT write to the same file you're going to read from or you'll have an empty file to read.
That's enough to get your app working.
class FileLineRead
File.open("file_line_read.txt") do |file|
file.each do |line|
phone_number = line.gsub(/\n/,'')
user = User.find_by_phone_number(line)
user.destroy unless user.nil?
end
end
end
open file
read line
DB Select
DB Update
In the cron job you have already opened and cleared test.log (via redirection) before you have read it in the Ruby script.
Why not do both the read and write in Ruby?
It may be a permissions issue or the file may not exist.
f = File.open("test","r")
puts f.read()
f.close()
The above will read the file test. If the file exists in the current directory
The problem is, as I can see, already solved by Slomojo. I'll only add:
to read and print a text file in Ruby, just:
puts File.read("/tmp/log/test.log")
This is related to: How do I read the results of a system() call in C++?
I am trying to do the exact the same thing only that my program needs to pass 'multiple parameters with spaces' to the command. I need the command line output and the exit code from the process.
Example: An example with Textpad. The application I'm really using prints stuff on stdout.
string command1 = "\"C:\Program Files\TextPad 5\Textpad.exe\" C:\readme0.txt";
string command2 = "\"C:\Program Files\TextPad 5\Textpad.exe\" \"C:\read me2.txt\"";
cout << system(command1.c_str()) << endl;
cout << system(command1.c_str()) << endl;
Output:
0
'C:\Program' is not recognized as an internal or external command,
operable program or batch file.
1
The first call to system passes and second one fails with the error above. _popen in Windows works similarly on Windows so no help there. I can easily do this on Linux as I can escape spaces in the parameters without having to use quotes.
An alternative is to write a huge chunk of non-cross-platform code as listed here:
http://msdn.microsoft.com/en-us/library/ms682499(VS.85).aspx
But in case I want to avoid that, are there any alternatives to system() and _popen() on Windows?
Thanks!
The lowest level Windows API function used by _popen() and system() is CreateProcess().
However CreateProcess() is not that simple to use - especially when you want to get the process'es output or write to the process'es input.
CreateProcess() will definitely work with file names that contain space characters (as long as they are written in quotation marks the way you did that).
The following solves the spaces in the path problem. Catching the output of the command is much more difficult, however:
#include <string>
#include <cstdlib>
using namespace std;
int main() {
string cmd = "\"c:\\program files\\notepad++\\notepad++.exe\"";
system( cmd.c_str() );
return 0;
}
A bunch of utility libraries have taken that chunk of non-portable code and wrapped it up with a portable interface. For an example, see Qt's QProcess.
I do this (note - this is VB.NET code), so I can write the output of the command to my log file (it's wrapped in a RunCommand() method):
Try
Dim myprocess As New Process()
myprocess.StartInfo.FileName = "C:\Program Files\TextPad 5\Textpad.exe"
myprocess.StartInfo.RedirectStandardOutput = True
myprocess.StartInfo.UseShellExecute = False
' inArgs are the arguments on the command line to the program
myprocess.StartInfo.WindowStyle = ProcessWindowStyle.Hidden
myprocess.StartInfo.Arguments = "C:\readme0.txt"
' the dir to set as default when the program runs
Then myprocess.StartInfo.WorkingDirectory = "C:\Program Files\TextPad 5\"
myprocess.Start()
' grab a reader to the standard output of the program
procReader = myprocess.StandardOutput()
' read all the output from the process
While (Not procReader.EndOfStream)
procLine = procReader.ReadLine()
' write the output to my log
writeNotes(procLine)
End While
procReader.Close()
Catch ex As Exception
' Write the error to my log
writeErrors("Couldn't execute command "C:\Program Files\TextPad 5\Textpad.exe", ex)
End Try
I think ShellExecute() is what you're looking for.
Never use system() in Windows !
Just redirect i/o handles.