Hello! -- Here's a how-to question for subprocess and threading
experts. What's the right way to obtain an output stream for stdout
from a subprocess, and read it lazily?
At the moment, there's the sh function, but this collects all the
output from the stdout and stderr streams into arrays, and returns
them wholesale when these end. This doesn't work well for processes
which generate very long streams of data.
One approach might be to modify sh to return the stdout stream, but I
think that won't tend to work, because the stderr stream must still be
collected (for most programs), and the process would also probably
have to be stopped at the end somehow.
Another approach might be to make a function which creates the Process
object -- say, open-subprocess, which returns a map {:out <stdout> :in
<stdin> :err <stderr>}. But something still must service both the
output and error streams. (sh handles this by interleaving character
reads from the stdout and stderr streams.)
A companion function close-subprocess would shut things down, and
return the result code. (Could a close "method" be made for this,
which would work with with-open?)
At this point I'm thinking of an open-subprocess which spawns some
kind of thread which could read either or both streams into an array.
Is there another possibility for doing this? What's the usual way to
handle this situation, if any?
Of course, I could use sh to pipe the stdout to a temporary file, but
that seems a cop-out.
Thanks in advance for any and all answers!
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
