Closed magomimmo closed 10 years ago
The same limitation exists for the same reason. In this case, the output of the "app" compilation needs to be loaded by the browser (instead of e.g. phantomjs
in the testing case). Perhaps there's a reliable way in this case to locate and preload the goog
bits before loading the compilation output — it's somewhere in out
, I think? — but I wouldn't want to recommend depending on anything in out
. :simple
(or :whitespace
, whatever) just works.
thanks Chas. The limitation applies to piggieback too, isn't it?
IIRC, :none
can be used in REPL environments, but I honestly don't recall.
Is all of this interest in :none
just re: TDD cycle time?
On Jan 26, 2014, at 1:37 PM, Chas Emerick notifications@github.com wrote:
IIRC, :none can be used in REPL environments, but I honestly don't recall.
ok. I'll try myself next week. Is all of this interest in :none just re: TDD cycle time?
mostly yes. I just read you comment on clojurescript.test. Comparing your numbers with mine it seems that my cljs-start template does something wrong, otherwise I wouldn't know how to justify the difference itself, even if I'm on an old iMac. — Reply to this email directly or view it on GitHub.
Hi Chas, I'm not able to make austin sample to work with the
:none
compilation. I did not find any information about austin non supporting it in the browser connected REPL scenario. It seems to me that austin has the same limitations of clojurescript.test about:none
option. Is it true?Have you any clue about it?
Thanks so much.