[rfc] [icedtea-web] actualization of readme

[Omair Majid]

* Jiri Vanek <jvanek at redhat.com> [2015-04-27 06:50]:
> Only minor accurateness. I would vote for removal it all and keeping just
> link to web... (but no.. this minimalist file, if accurate, save it purpose)

Some comments in-line below:

>  NetX provides a drop-in replacement for javaws (Java Web Start). Since
> -upstream NetX is dormant, we will be hosting and modifying the sources
> -in the IcedTea-Web repository, particularly in the netx/net/sourceforge/jnlp
> -directory.
> +upstream NetX is dead, and we based implementation on the netx/net/sourceforge/jnlp
> +package, whole implementation is still in this package, NetX is kept referenced
> +but whole icedtea-web as comlex, have now only few lines of code shared with
> +original implementation.

Can I suggest rewriting this as:

NetX provides a drop-in replacement for javaws (Java Web Start). Since
upstream NetX is dead, we forked the code into the
netx/net/sourceforge/jnlp package. However, as IcedTea-Web has evolved,
we only share a few lines of code with the original implementation.

>  IcedTea's NetX currently supports verification of signed jars, trusted
>  certificate storing, system certificate store checking, and provides
> -the services specified by the jnlp API.
> +the services specified by the jnlp API. 
> +In addition It also provides full desktop integration, offline run, own appletviwer,

s/It/it/. Also, can you combine the two paragraphs? Or maybe add a
one-line gap to make it explicit that these are two paragraphs?

> +many extended security features, own policy editor and much more.

This entire paragraph would be easier to read if it was a list.

> +* rhino (enables support for using proxy auto config files and javascript communication)

What does javascript communication here mean?

> -The following locations are checked for a JDK:
> -    * /usr/lib/jvm/java-openjdk
> -    * /usr/lib/jvm/icedtea6
> -    * /usr/lib/jvm/java-6-openjdk
> -    * /usr/lib/jvm/openjdk
> -    * /usr/lib/jvm/java-icedtea
> -    * /usr/lib/jvm/java-gcj
> -    * /usr/lib/jvm/gcj-jdk
> -    * /usr/lib/jvm/cacao
> -in the order given above.  
> +Check IT_CHECK_FOR_JDK in acinclude.m4 to see the list of default checked jdks.

I think it's expected that end users (not necessarily developers) may
read this. So please consider removing references to code from the
README.

> +To configure jdk used for build and runtime, 
> +* --with-jdk-home and --with-jre-home are your best friends.
> +if you are using jdk without jre, it may force you to use  both those switches
> +to point to same location. 

Since this is probably the two most important flags, please put this
first, not last.

> -By default, the following paths are checked for rhino:
> -  - /usr/share/java/js.jar
> -  - /usr/share/rhino-1.6/lib/js.jar
> -  - /usr/share/java/rhino.jar

Why remove the paths?

> +Grep acinclude.m4 to see default searched paths.

Maybe put the actual contents here.

> +
> +JUnit and Jacoco Support
> +========================
>  
>  JUnit is needed for running some tests. It has no run-time impact.
>  
> @@ -153,11 +154,17 @@
>  printed out, followed by <TestClass>.<testMethod>. This is also the format used
>  by JTreg.
>  
> +Jacoco is used in similar way and can be used to measure code-coverage both for
> +junit tests and for reproducer. For plugin fake binary is created. and used.

s/created. and/created and/

>  A set of automated tests is supplied for IcedTea-Web. They can be run by using
> -'make check'. Currently, this only tests a few parts of IcedTea-Web.
> +'make check' or "make run-netx-dist-tests"

Consider rewriting this:

> +Whether make check is tunning PAC tests and unittests, make run-netx-dist-tests
> +is running reproducers suite. It expects you to have ITW installed (if you wont it
> +to your personal space, use --prefix to specify final path)

as:

'make check' runs a set of unit tests are that always supposed to pass.
'make run-next-dist-tets' runs the reproducer test suite which requires
you to install IcedTea-Web first.

Thanks,
Omair

-- 
PGP Key: 66484681 (http://pgp.mit.edu/)
Fingerprint = F072 555B 0A17 3957 4E95  0056 F286 F14F 6648 4681