- From: Vasil Rangelov <boen.robot@gmail.com>
- Date: Sun, 30 Sep 2007 19:40:22 +0300
- To: public-xml-processing-model-comments@w3.org
>An XML document goes in the input port and it's connected to stdin on the program that runs. > >I'm tempted to special case an empty binding to stdin. I'm confused. Don't you need the program to be already started in order to use the input stream? That's how most shells (actually... all that I've used) usually do it at least. And if that's the case here too, what will the output and error streams look like when the input steam is used in the middle of the program? How exactly is the input document used? Passed at once as a serialized document with the default serialization options? If so, what about programs that don't accept XML document as their input? Is the input's text contents supplied at once? Or is it each line of the text contents supplied for each time the application accepts input on the stdin stream? Possibly this could be adjustable with another option if so? This whole input thing starts to overcomplicate this step if you ask me, but if you and the WG can figure it out, what the heck... it would be a nice feature. >Yeah, maybe it'd be better to just make command the whole command line, in which case...maybe it should be called command-line :-) The option or the step? Or both ;-) >The way I coded it up, you get an empty c:result on the output port if that output produced ... no output. I suppose it might be possible to make it >return an empty sequence instead, but I'd rather not. Fair enough. Regards, Vasil Rangelov
Received on Sunday, 30 September 2007 16:40:31 UTC