2017-03-30 UTC
chrisaldrich and hectorMcSpector joined the channel
# 01:42 hectorMcSpector Hi Folks
# 01:44 hectorMcSpector I would like some advice on how best to make assertions.
# 01:46 hectorMcSpector Right now, our test logic takes a few forms... 1) comparison of entire extracted statements, 2) comparison of statement constituents, and 3) comparison of the number of extracted statements.
# 01:47 hectorMcSpector I get the fact that the test suite provides both sample HTML snippets and corresponding JSON identifying the extracted Microformats.
# 01:48 hectorMcSpector Is the best way just to work through each test manually and derive the result type(s) and propert(ies) manually?
# 01:49 hectorMcSpector Any guidance would be very much appreciated.
[kevinmarks] joined the channel
# 02:31 KartikPrabhu hectorMcSpector: what do you mean by "derive the result type(s) and propert(ies) manually?"
hober joined the channel
# 02:32 KartikPrabhu I think the intention of the test suite is so that the HTML and the JSON to be compared are both given
# 02:32 hectorMcSpector Hi @KartikPrabhu
# 02:33 hectorMcSpector What I mean is that the format of the JSON results is not something which I can currently compare very easily given the output format Any23 currently writes to.
# 02:34 hectorMcSpector We write out statements
# 02:34 hectorMcSpector ?s ?p ?o
# 02:34 hectorMcSpector or quads
# 02:34 KartikPrabhu you could convert the JSON-like tree into your favourite format if you'd like
# 02:34 hectorMcSpector ... yeah. We don't. However I could just write a microformtas test writer which will.
# 02:35 hectorMcSpector I think that is the solution
tantek joined the channel
# 02:39 hectorMcSpector Thanks KartikPrabhu
[cleverdevil], nitot_ and barpthewire joined the channel
pniedzielski[m], martymcguire[m], [dgold], [kevinmarks], nitot, [cleverdevil], [eddie], KartikPrabhu, [keithjgrant], termie, tommorris_, Erkan_Yilmaz_, Guest76513, gRegorLove, tantek, [mko], [aaronpk] and [acegiak_net] joined the channel