Just consider the shell/shell scripting as a higher level language in itself. Then nanocat is just another language binding, albeit with a different concept of data flow, different calling conventions, and different set of libraries available for data manipulation (grep/sed/cut/awk/head). Kevin On Fri, Aug 30, 2013 at 5:07 AM, Schmurfy <schmurfy@xxxxxxxxx> wrote: > what you need to compare is the c source vs the python source ;) > I was just pointing out that for debug/dev you have more control on a > scripted version than with a c utility you can't modify easily (parse json, > msgpack, ...). > > > > > On 29 August 2013 20:54, Paul Colomiets <paul@xxxxxxxxxxxxxx> wrote: > >> Hi Schmurfy, >> >> On Thu, Aug 29, 2013 at 4:53 PM, Schmurfy <schmurfy@xxxxxxxxx> wrote: >> > I really like the idea of such tool to help debugging although I think >> the >> > simplest way of actually doing it is using on the bindings available and >> > do it in higher level languages. >> > One of the advantages of using ruby/python/... is that you easily >> manipulate >> > the data. >> >> Compare: >> >> nanocat --sub --connect tcp://127.0.0.1:1234 >> >> with: >> >> python -c 'import nn; s=nn.socket(nn.AF_SP, nn.NN_SUB); >> s.connect("tcp://127.0.0.1:1245"); s.setsockopt(nn.SUBSCRIBE, b""); >> print(s.recv())' >> >> You may imagine loop for NN_REP. >> >> You may also try to google for zmqcat, to find out how many users of >> such an utility (and thats not counting zmqc, zc, pjutil, and other >> more rare names) >> >> -- >> Paul >> >> >