Update Differences-from-js_of_ocaml.md

[hhugo]

• Js_of_ocaml can generate non mangled variable names
• I don't see why js_of_ocaml syntax makes it difficult to integrate with existing IDEs and build system

[coveralls]

Coverage remained the same at 53.457% when pulling 86e0074 on hhugo:patch-1 into 91859db on bloomberg:master.

[bobzhang]

@hhugo thanks for clarification, I was still using js_of_ocaml 2.5

[hhugo]

Also (as a reminder as I see that you've commented on the PR) , js_of_ocaml can also compile individual units (cmo & cma) ocsigen/js_of_ocaml#313.

Note that this still requires some glue code to be really usable.

[bobzhang]

yes, but it is more like dynamic linking, it has its use cases, but it is not separate compilation

[hhugo]

The dynlink use case is not supported yet - the aim of it was also separate
compilation. Linking and cross module optimization are still missing but
should not be too hard to add.
On May 12, 2016 7:04 PM, "Hongbo Zhang" notifications@github.com wrote:

yes, but it is more like dynamic linking, it has its use cases, but it is
not separate compilation

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
#371 (comment)

[bobzhang]

@hhugo if you are moving in this direction, then js_of_ocaml is not non-intrusive any more(I saw recent changes in js_of_ocaml introduced dependency on compiler-libs).
Note that I am really not interested in making such comparison, people who enjoy using js_of_ocaml should continue using it, and I use it in some cases too. I made this page because people keep asking me what's the difference, I know this is not easy to make it objective. I am more than happy to see js_of_ocaml keep improving and hope to have some collaboration with you. If you feel this page is too opinionated, I can remove it.

[hhugo]

Js_of_ocaml always had some kind of deps on compiler-libs, mostly to interpret debug events. I don't see how it makes js_of_ocaml more intrusive.

I understand the purpose of bucklescript, I'm not saying that this page should not exists - there are relevant points. My comment about separate compilation was just to say that this point may not be relevant in the future. I'm not (yet :) ) arguing for any other changes in this page.

[bobzhang]

@hhugo it would be nice to see js_of_ocaml more aggressive in whole program optimization!