I have a problem with the file meta source erroring out when trying to read a source file created with file meta sink.
I've attached a picture with the flowgraph, the upper, disabled portion was first used to generate the files, and then the lower portion was enabled to read. That however failed with:
Traceback (most recent call last):
File "/home/ec/Thesis/top_block.py", line 114, in <module>
File "/home/ec/Thesis/top_block.py", line 102, in main
tb = top_block_cls()
File "/home/ec/Thesis/top_block.py", line 67, in __init__
self.blocks_file_meta_source_0 = blocks.file_meta_source("/home/ec/Desktop/src_data.cf32", False, True, "/home/ec/Desktop/src_data.cf32.hdr")
File "/home/ec/gnuradio/lib/python2.7/dist-packages/gnuradio/blocks/blocks_swig0.py", line 520, in make
return _blocks_swig0.file_meta_source_make(*args, **kwargs)
RuntimeError: pmt::deserialize: malformed input stream: #f
When I disable the detached saving, the flowgraph will hang with:
thread[thread-per-block: <block file_meta_source (2)>]: pmt::deserialize: malformed input stream, tag value = : 116
I also got a notice about samp_rate not being available or somesuch when I tried it from a python script (this whole thing started with troubleshooting why I got different results in unittests and GRC...)
I've also attached the hdr file in question.
Reading an earlier recording from a USRP works fine, by the way. I don't know what's changed: I haven't done any updates, besides my own OOT, installed using the standard cmake mechanisms set up by gr_modtool, so I doubt that would ruin anything.
Thanks in advance,