The guy who wrote RubyScript2Exe needs to seriously reconsid

I have a very simple Ruby script that does nothing more than compute
Dhrystones so there is a fair amount of looping going on in this
script as would be expected.

RubyScript2Exe seems to need to trace the actual execution of this
script before placing it and all of its code into an .EXE however this
process, if this is the process, should not be required since all Ruby
code that uses “requires” keywords would be adding in code, yes ?

Anyway RubyScript2Exe is a bit inefficient - not that it doesn’t meet
my expectations for the Ruby language which is also quite inefficient
but I mean really folks this is a bit too much for me.

I have a similar Python program that makes a Python script into
an .EXE and it runs quite fast as would be expected so why does
RubyScript2Exe have to be so blamin’ slow ?!?

Call me a Troll if you must be if we in the Ruby community don’t bring
to light those things that need to change nothing about Ruby will ever
get better meanwhile Ruby as a language remains useful only to those
who don’t care about performance.

On 9/25/07, Ruby M. [email protected] wrote:

Anyway RubyScript2Exe is a bit inefficient - not that it doesn’t meet
who don’t care about performance.

Yes, you are a troll. This does not belong here. It belongs, in a much
nicer
tone, and maybe with code and tests, in a message to the Ruby2Exe team.

Jason

On Wed, 26 Sep 2007 05:55:10 +0900, Ruby M. [email protected]
wrote:

RubyScript2Exe seems to need to trace the actual execution of this
script before placing it and all of its code into an .EXE however this
process, if this is the process, should not be required since all Ruby
code that uses “requires” keywords would be adding in code, yes ?

Unfortunately, simply scanning for calls to require is insufficient
since:

  1. require is not the only way to load scripts

  2. require may be called dynamically (c.f. Rails)

-mental

If you are worried about the processing actually taking a long time:

RUBYSCRIPT2EXE.is_compiling?

is your friend. Just run through one loop, to give RubyScript2Exe a
“taste”
of what you need.

/Shawn

Hey Maniac.

Please keep posting messages such as these.

They are the light in my dark day. Your tone and style is
great. They make me laugh. I’m looking forward to reading
further enlightment analyzed / gathered by you.

Thank you!

-Martin

On 9/25/07, M. Edward (Ed) Borasky [email protected] wrote:

Ruby M. wrote:

I have a very simple Ruby script that does nothing more than compute
Dhrystones so there is a fair amount of looping going on in this
script as would be expected.

Dhont Repeat Yourself Stones??

http://en.wikipedia.org/wiki/DMIPS

Apparently it’s a synthetic benchmark for scoring different
programming languages. I’d be interested to see how Ruby compares
with Fortran, PL/1, SAL, and Algol.

I bet we could make it fast by doing the Dhrystone loop in C or
assembly even. Put that into a loop {} construct and it would be
wicked fast!

Blessings,
TwP

Ruby M. wrote:

I have a very simple Ruby script that does nothing more than compute
Dhrystones so there is a fair amount of looping going on in this
script as would be expected.

Dhont Repeat Yourself Stones??

On Sep 25, 2:16 pm, MenTaLguY [email protected] wrote:

  1. require may be called dynamically (c.f. Rails)

-mental

Pardon my ignorance as to the brilliance of the guy who wrote this
thing but if the person who is running RubyScript2Exe knows there are
scripts being loaded using some technique other than “require” then it
would still not be necessary to run the script being saved into an EXE
from start to finish as long as some kind of config file was being
used to tell RubyScript2Exe which additional files to visit while
collecting up source files to pack into the EXE. ← This is how the
Python gizmo called py2EXE works.

I realize Ruby people like to leverage the power of Ruby but sometimes
it may be more efficient to leverage the power of a config file and
just cut to the chase. It would become pretty clear which specific
files needed to be included based on the feedback at runtime whenever
the .EXE was being tested.

As much as I would love to be using Ruby it is tools such as
RubyScript2Exe that tend to turn me off but only because I have to
realize the person who coded py2EXE did a much better job without
having to execute the code during the packing process.

Your posts are intriguing. I’d like to subscribe to your newsletter.

Tim P. wrote:

Apparently it’s a synthetic benchmark for scoring different

Sadly, I’m all too familiar with Dhrystones, Whetstones, Lawrence
Livermore Kernels, Linpack, SPEC and half a dozen other MIPS
(Meaningless Indicators of Processing Speed). A happy exception to the
uselessness of typical “small/tiny/synthetic/micro” benchmarks was John
Gustafson’s “Hierarchical INTegration” (HINT) benchmark. It’s tough to
find the source for it on the web, but if you Google for “Gustafson HINT
Benchmark” you may turn up a copy.

On Sep 25, 2007, at 7:40 PM, Ruby M. wrote:

Unfortunately, simply scanning for calls to require is insufficient
scripts being loaded using some technique other than “require” then it
the .EXE was being tested.

As much as I would love to be using Ruby it is tools such as
RubyScript2Exe that tend to turn me off but only because I have to
realize the person who coded py2EXE did a much better job without
having to execute the code during the packing process.

Well, if you have a better way to do it, or a better way for you,
then implement it! Then send it back out to the community.
If you don’t somebody will. (if it is really an improvement)

On Wed, 2007-09-26 at 09:45 +0900, Eric H. wrote:

Your posts are intriguing. I’d like to subscribe to your newsletter.

Amen.

Arlen Christian Mart C. wrote:

On Wed, 2007-09-26 at 09:45 +0900, Eric H. wrote:

Your posts are intriguing. I’d like to subscribe to your newsletter.

Amen.

I think I like this groups sense of humour :slight_smile:

If was was going to worry about the speed of any thing, I wouldn’t use a
language whose primary implementation is (< 1.9) interpreter. I’m
interested
to see where the changes in Ruby 1.9 lead to for end users, then concern
over
pure speed.

The only thing I have to complain about, is I’m used to people having a
run
time pre-installed (C/C++/Java).

Doesn’t RubyScript2EXE have a mailing list 8=)

TerryP.