www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - How to get output of piped process?

reply Jedi <Return Lucifer.com> writes:
I an using pipeShell, I have redirected stdout, stderr, and stdin.

I am trying to read from the output and display it in my app. I 
have followed this code almost exactly except I use try wait and 
flush because the app is continuously updating the output. (it 
outputs a progress text on the same line and I'm trying to poll 
it to report to the user)


auto pipes = pipeProcess("my_application", Redirect.stdout | 
Redirect.stderr);
scope(exit) wait(pipes.pid);

// Store lines of output.
string[] output;
foreach (line; pipes.stdout.byLine) output ~= line.idup;

// Store lines of errors.
string[] errors;
foreach (line; pipes.stderr.byLine) errors ~= line.idup;


My code

auto p = pipeShell(`app.exe "`~f.name~`"`, Redirect.stdout | 
Redirect.stdin | Redirect.stderr);

		
			while(!tryWait(p.pid).terminated)
			{
				string[] output;
				foreach (line; p.stdout.byLine)
				{
					output ~= line.idup;
					writeln(line);
				}

				string[] errors;
				foreach (line; p.stderr.byLine)
				{
					errors ~= line.idup;
					writeln("Err:"~line);
				}
			}

wait(p.pid);

None of this works though. What is strange is that when I close 
out the debugger the app starts working(no console output but I 
able to see that it is doing something) but is very slow.

auto p = executeShell(`app.exe "`~f.name~`"`);

Does work, except I have no output or input. I have another app 
that I do the exact same code and I can get the output and parse 
it, but this is after the app terminates. I imagine the issue 
here is that I'm trying to get the output while the app is 
running.


I want to be able to get the output so I can reduce much of the 
clutter and give a progress report. I am ok with simply hooking 
up the in and out of the console of the app to mine just as if I 
ran app.exe directly.
Feb 16
next sibling parent reply Steven Schveighoffer <schveiguy gmail.com> writes:
On 2/17/21 1:58 AM, Jedi wrote:
 I an using pipeShell, I have redirected stdout, stderr, and stdin.
 
 I am trying to read from the output and display it in my app. I have 
 followed this code almost exactly except I use try wait and flush 
 because the app is continuously updating the output. (it outputs a 
 progress text on the same line and I'm trying to poll it to report to 
 the user)
 
 
 auto pipes = pipeProcess("my_application", Redirect.stdout | 
 Redirect.stderr);
 scope(exit) wait(pipes.pid);
 
 // Store lines of output.
 string[] output;
 foreach (line; pipes.stdout.byLine) output ~= line.idup;
 
 // Store lines of errors.
 string[] errors;
 foreach (line; pipes.stderr.byLine) errors ~= line.idup;
 
 
 My code
 
 auto p = pipeShell(`app.exe "`~f.name~`"`, Redirect.stdout | 
 Redirect.stdin | Redirect.stderr);
 
 
              while(!tryWait(p.pid).terminated)
              {
                  string[] output;
                  foreach (line; p.stdout.byLine)
You need to be careful here -- this will wait until stdout is *closed*.
                  {
                      output ~= line.idup;
                      writeln(line);
                  }
 
                  string[] errors;
                  foreach (line; p.stderr.byLine)
                  {
                      errors ~= line.idup;
                      writeln("Err:"~line);
                  }
Same thing here.
              }
 
 wait(p.pid);
 
 None of this works though. What is strange is that when I close out the 
 debugger the app starts working(no console output but I able to see that 
 it is doing something) but is very slow.
 
 auto p = executeShell(`app.exe "`~f.name~`"`);
 
 Does work, except I have no output or input. I have another app that I 
 do the exact same code and I can get the output and parse it, but this 
 is after the app terminates. I imagine the issue here is that I'm trying 
 to get the output while the app is running.
Without knowing the pattern of what your app is outputting, it's hard to tell what will happen. The most common problem with people dealing with piped output is not reading data off the pipe, which then makes the child process hang trying to write to the pipe, because the buffer is full. For instance, if your process outputs tons of stuff to stderr, you will hang, because you are waiting for stdout to be closed first before you read anything from stderr, the child process fills up stderr pipe, and is put to sleep waiting for it to be writable, never closing stdout. Unfortunately, std.process wraps all the pipes in File structs, so you have almost no good mechanisms to properly read the data. -Steve
Feb 17
parent reply Jedi <Return Lucifer.com> writes:
On Wednesday, 17 February 2021 at 14:36:58 UTC, Steven 
Schveighoffer wrote:
 On 2/17/21 1:58 AM, Jedi wrote:
 I an using pipeShell, I have redirected stdout, stderr, and 
 stdin.
 
 I am trying to read from the output and display it in my app. 
 I have followed this code almost exactly except I use try wait 
 and flush because the app is continuously updating the output. 
 (it outputs a progress text on the same line and I'm trying to 
 poll it to report to the user)
 
 
 auto pipes = pipeProcess("my_application", Redirect.stdout | 
 Redirect.stderr);
 scope(exit) wait(pipes.pid);
 
 // Store lines of output.
 string[] output;
 foreach (line; pipes.stdout.byLine) output ~= line.idup;
 
 // Store lines of errors.
 string[] errors;
 foreach (line; pipes.stderr.byLine) errors ~= line.idup;
 
 
 My code
 
 auto p = pipeShell(`app.exe "`~f.name~`"`, Redirect.stdout | 
 Redirect.stdin | Redirect.stderr);
 
 
              while(!tryWait(p.pid).terminated)
              {
                  string[] output;
                  foreach (line; p.stdout.byLine)
You need to be careful here -- this will wait until stdout is *closed*.
                  {
                      output ~= line.idup;
                      writeln(line);
                  }
 
                  string[] errors;
                  foreach (line; p.stderr.byLine)
                  {
                      errors ~= line.idup;
                      writeln("Err:"~line);
                  }
Same thing here.
              }
 
 wait(p.pid);
 
 None of this works though. What is strange is that when I 
 close out the debugger the app starts working(no console 
 output but I able to see that it is doing something) but is 
 very slow.
 
 auto p = executeShell(`app.exe "`~f.name~`"`);
 
 Does work, except I have no output or input. I have another 
 app that I do the exact same code and I can get the output and 
 parse it, but this is after the app terminates. I imagine the 
 issue here is that I'm trying to get the output while the app 
 is running.
Without knowing the pattern of what your app is outputting, it's hard to tell what will happen. The most common problem with people dealing with piped output is not reading data off the pipe, which then makes the child process hang trying to write to the pipe, because the buffer is full. For instance, if your process outputs tons of stuff to stderr, you will hang, because you are waiting for stdout to be closed first before you read anything from stderr, the child process fills up stderr pipe, and is put to sleep waiting for it to be writable, never closing stdout. Unfortunately, std.process wraps all the pipes in File structs, so you have almost no good mechanisms to properly read the data.
WTF?
 -Steve
Seriously, I can't simply get the output in real time? Come on, that is lame, Surely D can do better than that? How hard is it to get a buffer? Is there any hacks? How can one communicate with an app using std io if one can't actually communicate until the app is closed? It makes no sense. But note that even executeShell doesn't display the output of the app.exe so it is more than just pipeShell. The app just outputs text, just like almost every other text. One shouldn't have to know any pattern, that defeats the purpose. I should just be able to get the output of the app.exe, and also if the app is requesting input. This isn't rocket science but it seems someone wants to turn it in to it? When the app.exe is running it just prints stuff out, every once in a while it might ask for input(e.g., to overwrite the file if it exists, but I can get around that by checking in D)... but ultimately I just want to consolidate the output it gives so I need access to it BEFORE the app closes. The app.exe processes files, takes some time to do so so if I have to wait to display anything nothing will be displayed for a long time.
Feb 17
parent reply frame <frame86 live.com> writes:
On Thursday, 18 February 2021 at 06:04:13 UTC, Jedi wrote:

 Unfortunately, std.process wraps all the pipes in File 
 structs, so you have almost no good mechanisms to properly 
 read the data.
WTF?
 -Steve
I'm wonder about this message. You can always use readln() and eof() on such kind of streams. byLine() is just not the best option here.
Feb 18
parent reply Steven Schveighoffer <schveiguy gmail.com> writes:
On 2/18/21 4:40 AM, frame wrote:
 On Thursday, 18 February 2021 at 06:04:13 UTC, Jedi wrote:
 
 Unfortunately, std.process wraps all the pipes in File structs, so 
 you have almost no good mechanisms to properly read the data.
WTF?
It's just the way it is. Everything in Phobos is a C FILE * (wrapped in a File). You need to use alternative i/o systems to get the information. you can I believe get the file descriptor out of the File, which should help with better mechanisms. Try `fileno` and `windowsHandle`.
 
 I'm wonder about this message. You can always use readln() and eof() on 
 such kind of streams. byLine() is just not the best option here.
readln will block. eof doesn't tell you that there is no data in the pipe, it just says whether the pipe has been closed. What you need is non-blocking ways to check "does this pipe have more output for me?" If you read the implementation of execute, it gets around this by redirecting stderr to stdout (so you can wait on just one pipe). If you need them separated, then you need to do something more asynchronous. And Phobos does not wrap that, you have to use OS primitives. -Steve
Feb 18
next sibling parent frame <frame86 live.com> writes:
On Thursday, 18 February 2021 at 17:27:48 UTC, Steven 
Schveighoffer wrote:

 readln will block. eof doesn't tell you that there is no data 
 in the pipe, it just says whether the pipe has been closed.
Of course, I must have been thinking of another language - I should take a coffee before posting. Is there a d-library that can handle process/named pipes on Windows and Linux? I did not found any.
Feb 18
prev sibling parent reply Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Thursday, 18 February 2021 at 17:27:48 UTC, Steven 
Schveighoffer wrote:
 On 2/18/21 4:40 AM, frame wrote:
 [...]
It's just the way it is. Everything in Phobos is a C FILE * (wrapped in a File). You need to use alternative i/o systems to get the information. [...]
Does your iopipe handle... Pipes? 😀 Like someone wrote, can it? 🤔
Feb 19
next sibling parent reply kdevel <kdevel vogtner.de> writes:
On Friday, 19 February 2021 at 08:37:50 UTC, Imperatorn wrote:
 Does your iopipe handle... Pipes? 😀
BTW: What about SIGPIPE? In an experimental code I have this : fout.rawWrite (buf); fout.rawWrite ("\n"); writeln ("flushing"); fout.flush (); // (a) enforce (! fout.eof, "eof on write to child"); // (b) writeln ("reading from pipe"); : fout actually is the child's stdin. Sometimes between (a) and (b) a SIGPIPE occurs and terminates the process (exit code 141). That prevents the proper handling of eof. Why isn't SIGPIPE blocked or handled by default?
Feb 19
parent reply Steven Schveighoffer <schveiguy gmail.com> writes:
On 2/19/21 5:41 AM, kdevel wrote:
 On Friday, 19 February 2021 at 08:37:50 UTC, Imperatorn wrote:
 Does your iopipe handle... Pipes? 😀
BTW: What about SIGPIPE? In an experimental code I have this    :    fout.rawWrite (buf);    fout.rawWrite ("\n");    writeln ("flushing");    fout.flush ();                                    // (a)    enforce (! fout.eof, "eof on write to child");    // (b)    writeln ("reading from pipe");    : fout actually is the child's stdin. Sometimes between (a) and (b) a SIGPIPE occurs and terminates the process (exit code 141). That prevents the proper handling of eof. Why isn't SIGPIPE blocked or handled by default?
ignoring SIGPIPE is a process-wide thing, and so it's not appropriate for Phobos to make that decision for you. But it's trivial to ignore it. I've never been a fan of SIGPIPE. If you look around on the Internet, you'll find that most people agree that the reasoning for SIGPIPE is to fix poor programming (i.e. ignoring of error codes). But it doesn't give you any good way to handle it. A SIGPIPE can be due to any pipe being written, it doesn't tell you which one. In order to know which one caused it, well, you have to look at the error code of the call! The end result is -- it makes poor programming the standard. If you ignore SIGPIPE for a child process, then if that process' operation depends on SIGPIPE killing it, then you have screwed over that child process, or rather, exposed the lack of error checking in the child process. -Steve
Feb 19
parent reply kdevel <kdevel vogtner.de> writes:
On Friday, 19 February 2021 at 13:42:46 UTC, Steven Schveighoffer 
wrote:

[...]

 ignoring SIGPIPE is a process-wide thing, and so it's not 
 appropriate for Phobos to make that decision for you. But it's 
 trivial to ignore it.
Sure.
 I've never been a fan of SIGPIPE. If you look around on the 
 Internet, you'll find that most people agree that the reasoning 
 for SIGPIPE is to fix poor programming (i.e. ignoring of error 
 codes).
As application programmer I don't want to check any error codes. Thankfully I don't have to in D. There is a nice off-topic example from the linux kernel [1] what happens when people do not check return values [1].
 But it doesn't give you any good way to handle it. A SIGPIPE 
 can be due to any pipe being written, it doesn't tell you which 
 one. In order to know which one caused it, well, you have to 
 look at the error code of the call!
Fortunately the D runtime /does/ take care and it throws---if the signal is ignored beforehand. I filed issue 21649. [1] <https://lore.kernel.org/patchwork/patch/260181/> move RLIMIT_NPROC check from set_user() to do_execve_common()
Feb 19
parent reply Danny Arends <Danny.Arends gmail.com> writes:
On Friday, 19 February 2021 at 15:39:25 UTC, kdevel wrote:
 On Friday, 19 February 2021 at 13:42:46 UTC, Steven 
 Schveighoffer wrote:

 [...]

 [...]
Sure.
 [...]
As application programmer I don't want to check any error codes. Thankfully I don't have to in D. There is a nice off-topic example from the linux kernel [1] what happens when people do not check return values [1].
 [...]
Fortunately the D runtime /does/ take care and it throws---if the signal is ignored beforehand. I filed issue 21649. [1] <https://lore.kernel.org/patchwork/patch/260181/> move RLIMIT_NPROC check from set_user() to do_execve_common()
Perhaps a bit late, but this is how I deal with pipes and spawnShell. Read one byte at a time from stdout and stderr: https://github.com/DannyArends/DaNode/blob/master/danode/process.d Danny
Feb 22
next sibling parent reply frame <frame86 live.com> writes:
On Monday, 22 February 2021 at 13:23:40 UTC, Danny Arends wrote:

 https://github.com/DannyArends/DaNode/blob/master/danode/process.d

 Danny
This example shows how easy it is to implement a non-blocking stream. Phobos knows this for sockets but not for pipes?
Feb 22
parent reply Danny Arends <Danny.Arends gmail.com> writes:
On Monday, 22 February 2021 at 14:52:22 UTC, frame wrote:
 On Monday, 22 February 2021 at 13:23:40 UTC, Danny Arends wrote:

 https://github.com/DannyArends/DaNode/blob/master/danode/process.d

 Danny
This example shows how easy it is to implement a non-blocking stream. Phobos knows this for sockets but not for pipes?
Sockets seem to be more OS independent and are way more mature in Phobos. Pipes seem to have been added as an afterthought in std.process and std.stdio I had to add code for windows to deal with non-blocking/buffering pipes, Linux uses fcntl/fileno to enable non-blocking No idea why non-blocking pipes aren't in Phobos, but pipes should not be an afterthought but a first class citizen imho
Mar 03
parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Wednesday, 3 March 2021 at 20:43:54 UTC, Danny Arends wrote:
 On Monday, 22 February 2021 at 14:52:22 UTC, frame wrote:
 On Monday, 22 February 2021 at 13:23:40 UTC, Danny Arends 
 wrote:

 https://github.com/DannyArends/DaNode/blob/master/danode/process.d

 Danny
This example shows how easy it is to implement a non-blocking stream. Phobos knows this for sockets but not for pipes?
Sockets seem to be more OS independent and are way more mature in Phobos. Pipes seem to have been added as an afterthought in std.process and std.stdio I had to add code for windows to deal with non-blocking/buffering pipes, Linux uses fcntl/fileno to enable non-blocking No idea why non-blocking pipes aren't in Phobos, but pipes should not be an afterthought but a first class citizen imho
True, we should add this ☀️
Mar 06
prev sibling next sibling parent reply Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Monday, 22 February 2021 at 13:23:40 UTC, Danny Arends wrote:
 On Friday, 19 February 2021 at 15:39:25 UTC, kdevel wrote:
 [...]
Perhaps a bit late, but this is how I deal with pipes and spawnShell. Read one byte at a time from stdout and stderr: https://github.com/DannyArends/DaNode/blob/master/danode/process.d Danny
Interesting, do you have any benchmarks for DaNode?
Feb 23
parent Danny Arends <Danny.Arends gmail.com> writes:
On Tuesday, 23 February 2021 at 10:07:03 UTC, Imperatorn wrote:
 On Monday, 22 February 2021 at 13:23:40 UTC, Danny Arends wrote:
 On Friday, 19 February 2021 at 15:39:25 UTC, kdevel wrote:
 [...]
Perhaps a bit late, but this is how I deal with pipes and spawnShell. Read one byte at a time from stdout and stderr: https://github.com/DannyArends/DaNode/blob/master/danode/process.d Danny
Interesting, do you have any benchmarks for DaNode?
I used to run Apache bench on the code to make sure it was as fast as possible, for static files it is pretty performant since it buffers the files, and serves them directly from memory. The main overhead comes from the external process booting up, rdmd is nice since it only does the compile once then reuses the compiled binary for external scripts. PHP and such are always hit with the additional overhead. Feel free to run your own tests, the development branch has the latest version with some additional bug fixes not yet available in the master branch, and feedback is welcome and can be posted as an issue in Github
Mar 03
prev sibling parent reply kdevel <kdevel vogtner.de> writes:
On Monday, 22 February 2021 at 13:23:40 UTC, Danny Arends wrote:
 On Friday, 19 February 2021 at 15:39:25 UTC, kdevel wrote:
[...]
 Fortunately the D runtime /does/ take care and it throws---if 
 the signal
 is ignored beforehand. I filed issue 21649.
[...]
 Perhaps a bit late,
It's never too late.™ :-)
 but this is how I deal with pipes and spawnShell.
 Read one byte at a time from stdout and stderr:

 https://github.com/DannyArends/DaNode/blob/master/danode/process.d
Is this immune to SIGPIPE and is this design able to serve infinite streams? BTW: Why does run use spawnShell and not spawnProcess (would save one File object). If the design is not intended to serve infinite streams I would suggest to open two temporary files "out" and "err", delete them, and let the child process write stdout/stderr into those files. IFAICS this avoid threads, sleep, pipe and reading with fgetc.
Feb 25
parent Danny Arends <Danny.Arends gmail.com> writes:
On Thursday, 25 February 2021 at 15:28:25 UTC, kdevel wrote:
 On Monday, 22 February 2021 at 13:23:40 UTC, Danny Arends wrote:
 On Friday, 19 February 2021 at 15:39:25 UTC, kdevel wrote:
[...]
 Fortunately the D runtime /does/ take care and it throws---if 
 the signal
 is ignored beforehand. I filed issue 21649.
[...]
 Perhaps a bit late,
It's never too late.™ :-)
 but this is how I deal with pipes and spawnShell.
 Read one byte at a time from stdout and stderr:

 https://github.com/DannyArends/DaNode/blob/master/danode/process.d
Is this immune to SIGPIPE and is this design able to serve infinite streams?
No I have linked up a signal handler to just ignore sigpipe, the web server closes connections after not seeing a valid output from the script for 5min (e.g. no header)
 BTW: Why does run use spawnShell and not spawnProcess (would
 save one File object).
I tried different approaches, this one worked for me™ and I just went with it. I need the stdin (for the get/post/cookies) stdout (php/d/brainf*ck script output) and stderr for the error stream from he external script
 If the design is not intended to serve infinite streams I would
 suggest to open two temporary files "out" and "err", delete 
 them,
 and let the child process write stdout/stderr into those files.
I used to do that, but it generated a lot of temporary files, and I would need to parse in the files after the process is done to serve the output to the client. Using pipes is cleaner code wise, since I can just stream back the output to the client (e.g. in keepalive connections)
 IFAICS this avoid threads, sleep, pipe and reading with fgetc.
Mar 03
prev sibling parent Steven Schveighoffer <schveiguy gmail.com> writes:
On 2/19/21 3:37 AM, Imperatorn wrote:
 On Thursday, 18 February 2021 at 17:27:48 UTC, Steven Schveighoffer wrote:
 On 2/18/21 4:40 AM, frame wrote:
 [...]
It's just the way it is. Everything in Phobos is a C FILE * (wrapped in a File). You need to use alternative i/o systems to get the information. [...]
Does your iopipe handle... Pipes? 😀 Like someone wrote, can it? 🤔
Not yet. Important to remember that iopipe does not do actual i/o, it's a library to add *buffering* to i/o. The std.io (which is the intended source for iopipe) library does not support pipes at the moment. But I plan to add them. https://github.com/MartinNowak/io/issues/24 -Steve
Feb 19
prev sibling parent reply Jesse Phillips <Jesse.K.Phillips+D gmail.com> writes:
On Wednesday, 17 February 2021 at 06:58:55 UTC, Jedi wrote:
 I an using pipeShell, I have redirected stdout, stderr, and 
 stdin.

 I am trying to read from the output and display it in my app. I 
 have followed this code almost exactly except I use try wait 
 and flush because the app is continuously updating the output. 
 (it outputs a progress text on the same line and I'm trying to 
 poll it to report to the user)
I think this post is going to answer your need. https://dev.to/jessekphillips/piping-process-output-1cai I haven't read all the replies, so maybe you have it working and this will benefit someone else.
Mar 05
parent reply kdevel <kdevel vogtner.de> writes:
On Saturday, 6 March 2021 at 01:53:15 UTC, Jesse Phillips wrote:
[...]
 I think this post is going to answer your need.

 https://dev.to/jessekphillips/piping-process-output-1cai

 I haven't read all the replies, so maybe you have it working 
 and this will benefit someone else.
If I understand your code correctly you have a program "reverse" writing output to another program "check". I am wondering why you have two threads shoveling data between pipes. In the example below data are read from stdin by the first process and is piped to another one which is finally piped into the program and copied into a string appender. ```pipechain.d import std.stdio; import std.process; import std.conv; import std.array; import std.range; import std.algorithm; int main (string [] args) { auto p = pipe (); auto proc1 = spawnProcess (["cat"], stdin, p.writeEnd); auto q = pipe (); auto proc2 = spawnProcess (["cat"], p.readEnd, q.writeEnd); auto os = appender!string; q.readEnd.byChunk (4096).copy (os); auto res2 = wait (proc2); auto res1 = wait (proc1); stderr.writeln ("res1 = ", res1, ", res2 = ", res2); write (os[]); stderr.writeln ("fin"); return 0; } ``` AFAICS this is immune to SIGPIPE. Do I miss anything?
Mar 06
parent Jesse Phillips <Jesse.K.Phillips+D gmail.com> writes:
On Saturday, 6 March 2021 at 21:20:30 UTC, kdevel wrote:
 ```pipechain.d
 import std.stdio;
 import std.process;
 import std.conv;
 import std.array;
 import std.range;
 import std.algorithm;

 int main (string [] args)
 {
    auto p = pipe ();
    auto proc1 = spawnProcess (["cat"], stdin, p.writeEnd);
    auto q = pipe ();
    auto proc2 = spawnProcess (["cat"], p.readEnd, q.writeEnd);
    auto os = appender!string;
    q.readEnd.byChunk (4096).copy (os);
    auto res2 = wait (proc2);
    auto res1 = wait (proc1);
    stderr.writeln ("res1 = ", res1, ", res2 = ", res2);
    write (os[]);
    stderr.writeln ("fin");
    return 0;
 }
 ```

 AFAICS this is immune to SIGPIPE. Do I miss anything?
I probably missed that from the documentation and was trying to make it work with pipeProcess.
Mar 07