Hi Nathan, My guess is that the problem is a bit more tricky.
The XML block definition works fine as long as the message port is not present. I can use nports as shown in stream_to_streams. It also works fine as long as the nports statement is not used and a message port is defined. This means copy and paste the stream port definition for every port. Assign a port name manually etc. Nevertheless the combination using nports and message ports in one XML definition fails. I hope that clarifies my problem description. Johannes On 12.10.2013 13:43, West, Nathan wrote: > Re: #1, if the block's xml is malformed then GRC will basically show > nothing (that grey flowgraph page you describe). Are you defining $ports > in the make node? Compare to a similar block to make sure everything is > defined properly. > https://github.com/gnuradio/gnuradio/blob/master/gr-blocks/grc/blocks_streams_to_stream.xml > > I'm not sure of a way to debug what section might be wrong, so compare > against other blocks with message ports too. > > > > On Sat, Oct 12, 2013 at 12:58 PM, Jared Clements > <jared.cleme...@gmail.com <mailto:jared.cleme...@gmail.com>> wrote: > > I'll confirm that #2 is either a bug or merely unexpected behavior. > I work around it by prototyping hier blocks as custom GRC blocks, > then using that to build an OOT module block that actually works. > Being unable to enter parameters at run time is severely limiting. > > I've not experienced #1, that functionality seems to work once I get > the python file and the GRC file matching. I was getting the same > errors for a while until I found the right idiom. > > Jared > > On Oct 12, 2013 11:18 AM, "Johannes Demel" <uf...@student.kit.edu > <mailto:uf...@student.kit.edu>> wrote: > > Hi list, > > I discovered 2 problems with GRC recently. > > 1. I have a custom block with a message port (with a fixed port > name) > and some stream ports which include a <nports>...</nports> > definition. > The whole thing works fine as long as I have a fixed number of > ports. > Each declared separately. But as soon as I use a nport statement GRC > won't display this block correctly [1]. Even worse if I enter values > which affect the number of ports, the whole flowgraph will > disappear. > Just a grey flowgraph page. > > 2. To make things easier for me (and to not create a well known > kind of > artwork) I use hier blocks. This works fine as long as I have > fixed size > vector ports. But adding a parameter block, say vlen, for vector > size to > dynamically change the hier block doesn't work. The python generator > does not generate the hier block python file with vlen as vector > size. > Instead it puts in the default value. A parameter block without > default > value results in an error. A hard coded vector size is not exactly > helpful in this case. > > I didn't have time to dig into it yet. Thus I thought I share my > experience with you. Maybe I am not the only one with this > problem and > someone already knows how to fix it. > > Happy hacking > Johannes > > [1] > <sink> > <name>msg</name> > <type>message</type> > <optional>1</optional> > </sink> > > <sink> > <name>in</name> > <type>complex</type> > <vlen>$vlen</vlen> > <nports>$ports</nports> > </sink> > > _______________________________________________ > Discuss-gnuradio mailing list > Discuss-gnuradio@gnu.org <mailto:Discuss-gnuradio@gnu.org> > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio > > > _______________________________________________ > Discuss-gnuradio mailing list > Discuss-gnuradio@gnu.org <mailto:Discuss-gnuradio@gnu.org> > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio > > _______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
