Ruby 1.9.3 RC1 is out

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,
Ruby 1.9.3 RC1 has been released. This is the first release candidate
of Ruby 1.9.3, the next stable release of Ruby.

Ruby 1.9.3 is a release for better implementation. It is mostly
backward-compatible with Ruby 1.9.2. But also includes some new
features, e.g. io/console.

You can read the list of changes since Ruby 1.9.2 in NEWS [1]. The RC1
is not so different from Ruby 1.9.3 preview1, but it includes many bug
fixes. See ChangeLog [2] for more details.

*1: http://svn.ruby-lang.org/repos/ruby/tags/v1_9_3_rc1/NEWS
*2: http://svn.ruby-lang.org/repos/ruby/tags/v1_9_3_rc1/ChangeLog

I will release Ruby 1.9.3 within this two weeks unless any serious
problem is reported. If you have any trouble with Ruby 1.9.3, please
let us know.

== Locations


Regards,
Yuki Yugui S. [email protected]

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk59ju4ACgkQOXzH5JLb/AU58wCgjjouy6Bt21KxpDcD1DSvWzXH
sHwAn3eeN1000hHjmVVwPOGVbbaWOGWx
=iiT+
-----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Am 24.09.2011 10:03, schrieb Yuki S. (Yugui):

I will release Ruby 1.9.3 within this two weeks unless any serious
problem is reported. If you have any trouble with Ruby 1.9.3,
please let us know.

It doesn’t compile (and yes, I did check the checksum of the download):

===============================================
√ quintus@hades => ~/Downloads/ruby-1.9.3-rc1
$ ./configure --prefix=/opt/rubies/ruby-1.9.3-rc1 --enable-shared

  • –program-suffix=-rc1 --disable-install-capi
    […]
    checking for if make is GNU make… yes
    checking for nroff… /usr/bin/nroff
    .ext/include/x86_64-linux/ruby/config.h updated
    ruby library version = 1.9.1
    configure: creating ./config.status
    config.status: creating Makefile
    config.status: creating ruby-1.9.pc
    √ quintus@hades => ~/Downloads/ruby-1.9.3-rc1
    $ make
    CC = gcc
    LD = ld
    LDSHARED = gcc -shared
    CFLAGS = -O3 -ggdb -Wextra -Wno-unused-parameter -Wno-parentheses
  • -Wno-long-long -Wno-missing-field-initializers -Werror=pointer-arith
  • -Werror=write-strings -Werror=declaration-after-statement
  • -Werror=implicit-function-declaration -fPIC
    XCFLAGS = -include ruby/config.h -include ruby/missing.h
  • -fvisibility=hidden -DRUBY_EXPORT
    CPPFLAGS = -I. -I.ext/include/x86_64-linux -I./include -I.
    DLDFLAGS = -Wl,-soname,libruby.so.1.9
    SOLIBS = -lpthread -lrt -ldl -lcrypt -lm
    compiling main.c
    compiling dmydln.c
    compiling dmyencoding.c
    compiling version.c
    compiling dmyversion.c
    compiling miniprelude.c
    compiling array.c
    compiling bignum.c
    bignum.c:17:19: fatal error:
    /usr/lib/gcc/x86_64-unknown-linux-gnu/4.6.1/include/float.h:
    Input/output error
    compilation terminated.
    make: *** [bignum.o] Error 1
    ===============================================

I’m using Arch Linux, here’s some more information:

===============================================
√ quintus@hades => ~/Downloads/ruby-1.9.3-rc1
$ gcc --version
gcc (GCC) 4.6.1 20110819 (prerelease)
Copyright (C) 2011 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is
NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE.
√ quintus@hades => ~/Downloads/ruby-1.9.3-rc1
$ uname -a
Linux hades 3.0-ARCH #1 SMP PREEMPT Tue Aug 30 08:53:25 CEST 2011
x86_64 AMD Athlon™ II X4 640 Processor AuthenticAMD GNU/Linux

Vale,
Marvin
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJOfbLkAAoJELh1XLHFkqhav5wH/0xFDQ6m0nYORbpdXO8uNO7F
+SHIRhIJReRzsKvKNwCZM1Oqpf6iyDco7+zKWqwIae/xXjYYsv45+00JGMmxLrkN
jxNB4E8m2zl6wb5CxOj2tRj/kezyWzMSfu2EyQVXd3SbCEZLksApsp30bEEdd8qt
goSmEeV/XrMGPfH3LOO98rXoitclc9ClhdNLhYekWw6o4XvNb78gcgDCUsdmoSYg
wEAcsrijr/q/9AZ6LTEBFUYBV/C3jM0lw/7GcjtzKwkbQF046OkmenChm939Obk7
4ECbYpAiT2uKzuCFB6+blNVwNXoIbMOc/ruiB0Livtj4qQxDF5SMS28jwBGAxLY=
=orbF
-----END PGP SIGNATURE-----

It just compiled. Arch Linux, i686.

$ cat /proc/version
Linux version 3.0-ARCH (tobias@T-POWA-LX) (gcc version 4.6.1 20110819
(prerelease) (GCC) ) #1 SMP PREEMPT Tue Aug 30 07:32:23 UTC 2011

$ gcc --version
gcc (GCC) 4.6.1 20110819 (prerelease)
Copyright © 2011 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is
NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Am 24.09.2011 13:01, schrieb Chris W.:

the disk or try a different kernel.
You’re indeed right, thanks!. I ran fsck over my disk and it reported
a frightening range of errors… Currently I’m doing a complete smart
test on the drive, hopefully I don’t need a new one…

However, it still doesn’t compile. I now get beyond the float.h thing
and it crashes at marshal.c:

=====================================================================
compiling marshal.c
marshal.c: In function ‘w_float’:
marshal.c:358:27: error: ‘DECIMAL_MANT’ undeclared (first use in this
function)
marshal.c:358:27: note: each undeclared identifier is reported only
once for each function it appears in
make: *** [marshal.o] Error 1

That shouldn’t be an issue with my hard drive, I just tried to compile
the previous version Ruby 1.9.3-preview1 and that one compiled just
fine.

Vale,
Marvin
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJOfdJ0AAoJELh1XLHFkqhac4AH/i7DQrZGgUfnyo8Ifa7GTt93
WgiJzyPbjwdCnLsuWXnKGepSdu8SqrP5nnKl9z3q4L9paY6PLgnTfjmGeLFcv6Da
LWin3/eBqNVDcjwtcg+8DTcJzMB4ngaRCHhBmKRx1umfLCpKo+7anQHohRsEgapG
RAcD7trkcsK07yxvdJ5wEJOIg8EBrZtddy+qzJjWD0bidCm0pCok1meNMLw51XEG
0o2nRx8LnaC+I9SigWHoW1hzP07wMPvGaMCPViSZvDk6nL97bGVplCJcy4XUQWCD
pfVbb4eiAuyiv/ncy0mFhLvonFbIaMHmRoqMQ7Ty3vBqkoOIBgm8azq0CS0oEJs=
=xM1c
-----END PGP SIGNATURE-----

Hello,

2011/9/24 Quintus [email protected]:

However, it still doesn’t compile. I now get beyond the float.h thing
and it crashes at marshal.c:

I bet that your float.h is still broken.
I’m afraid you’ll just have to reinstall your C compiler.
I recommend you to reinstall os :frowning:

It doesn’t compile (and yes, I did check the checksum of the download):

compiling array.c
compiling bignum.c
bignum.c:17:19: fatal error:
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.6.1/include/float.h:
Input/output error

Sounds like a hardware / kernel driver issue actually. Run fsck on the
disk or try a different kernel.

Regards,
Chris W.
Twitter: http://www.twitter.com/cwgem

[Are there / will there be before final 1.9.3 release] any binary
packages for Windows?

– Matma R.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Am 24.09.2011 15:31, schrieb Yusuke E.:

Hello,

2011/9/24 Quintus [email protected]:

However, it still doesn’t compile. I now get beyond the float.h
thing and it crashes at marshal.c:

I bet that your float.h is still broken. I’m afraid you’ll just
have to reinstall your C compiler. I recommend you to reinstall os
:frowning:

Hm… That’s gonna be the usual amount of work, with properly
installing and configuring anything. However, I wanted to do this
earlier already for cleanup. Thank you, Yusuke-san, I think I’ll do
this in a week or two, and then directly use the final 1.9.3 ;-).

Valete,
Marvin

PS: smart says the hard drive is ok. And reinstalling GCC didn’t help.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJOffYrAAoJELh1XLHFkqhagJkH/3NrE2K+20daBhWUmEP8vJLl
ntg6Y29vn5tFhVH+gPtRJrKaHRD1O+J89uReHPcwVmzmlQOubSibJjmeH5fOpkRL
cz7wbVWfsW4YpYfLeU+HzmhQ+U109Mo/2DIgYKaRuUHfitc1Gj3IwTrAKwKN9vaw
QlUUno9Ph1/8+fQl5wCuXGNzpDf5vUByq56mPvngbadQzH7xUuubO6CXvKlg3rux
jrg29KugfXtdSpO4convaNiy6R7xfc4ySx9YaXWsRvsnzvkrjhWFBRdUDFyO/WWe
qa2nOzDV7KTQUO+XmE/ER/KGYDWb6MlhyFf6qck534MSAGb4R6bu3XLh03Xi++o=
=CbNl
-----END PGP SIGNATURE-----

I am not sure if I can reply to this or should I file an issue.

But I was using python lately, on a yaml file where ruby 1.9.3* chokes.

And python reports better error:

yaml.reader.ReaderError: ‘utf8’ codec can’t decode byte #xfc: invalid
start byte in “colours.yml”, position 1059

It even tells the position.

It would be great if Ruby could also show the position of the wrong
character!

On Sep 24, 2011, at 13:52 , Marc H. wrote:

I am not sure if I can reply to this or should I file an issue.

you should file an issue on redmine

2011/9/24 Bartosz Dziewoński [email protected]:

[Are there / will there be before final 1.9.3 release] any binary
packages for Windows?

AFAICT, yes. The RCs need Windows testing, too. :wink:


Phillip G.

gplus.to/phgaw | twitter.com/phgaw

A method of solution is perfect if we can forsee from the start,
and even prove, that following that method we shall attain our aim.
– Leibniz

Bartosz Dziewoński wrote in post #1023582:

[Are there / will there be before final 1.9.3 release] any binary
packages for Windows?

Yes, Luis is currently planning to release 1.9.3 RC1 RubyInstaller
binary archives which will be made available at
Downloads Keep an eye out for his ANN.

If you’re feeling adventurous and would like to build your own 1.9.3 (on
Windows without having to configure a build toolchain) from MRI source,
it’s fairly simple to do using the RubyInstaller build recipes. Read my
reply for one way to get started:

http://groups.google.com/group/rubyinstaller/browse_thread/thread/86c6e1e6c2b7ef64

As always, if you have problems check out our Troubleshooting (and
other) wiki pages

search our Google Group at http://groups.google.com/group/rubyinstaller
and if none of that helps, post a question (with details) and we’ll do
our best to help.

Jon


blog: http://jonforums.github.com/
twitter: @jonforums

Most people die of a sort of creeping common sense, and discover when it
is too late that the only things one never regrets are one’s mistakes.
- Oscar Wilde

Bartosz Dziewoński wrote in post #1024400:

2011/9/25 Jon F. [email protected]:

Bartosz Dziewoński wrote in post #1023582:

[Are there / will there be before final 1.9.3 release] any binary
packages for Windows?

Yes, Luis is currently planning to release 1.9.3 RC1 RubyInstaller
binary archives which will be made available at
Downloads Keep an eye out for his ANN.

Any progress? :slight_smile:

Will work on a binary release this weekend, it was a very busy week at
work.

You can still compile your own… :stuck_out_tongue:


Luis L.

I did in fact try compiling on my own using Jon’s instructions, but
after long downloads and loooong crunching the rake was “aborted”.
I’ll keep trying, of course (I have some suspicions as to why it
barfed), but a premade package would be greatly appreciated :slight_smile:

2011/9/30, Luis L. [email protected]:

2011/9/25 Jon F. [email protected]:

Bartosz Dziewoński wrote in post #1023582:

[Are there / will there be before final 1.9.3 release] any binary
packages for Windows?

Yes, Luis is currently planning to release 1.9.3 RC1 RubyInstaller
binary archives which will be made available at
Downloads Keep an eye out for his ANN.

Any progress? :slight_smile:

If you’re feeling adventurous and would like to build your own 1.9.3 (on
Windows without having to configure a build toolchain) from MRI source,
it’s fairly simple to do using the RubyInstaller build recipes. Read my
reply for one way to get started:

http://groups.google.com/group/rubyinstaller/browse_thread/thread/86c6e1e6c2b7ef64

Thanks, I’m going to try it.

– Matma R.

Bartosz Dziewoński wrote in post #1024428:

I did in fact try compiling on my own using Jon’s instructions, but
after long downloads and loooong crunching the rake was “aborted”.
I’ll keep trying, of course (I have some suspicions as to why it
barfed), but a premade package would be greatly appreciated :slight_smile:

2011/9/30, Luis L. [email protected]:

One potential barf area to check is whether there are 0 byte archives in
the ‘downloads’ sub-directory of your RubyInstaller repo clone. While
it’s caused by network problems, the current scripts don’t deal with
this failure as gracefully as we’d like.

The manual fix is to delete those 0 byte archives and repeat the build
until the the ‘downloads’ directory cache has a full set of
dependencies. Once you get a full set of artifacts in the ‘downloads’
cache, things work well for repeatedly building.

If this is your problem, tweak my previous instructions by doing:

1. manually delete 0-byte files in ‘downloads’ subdir

2. clean out the build sandbox…DON’T run rake distclean

as that will also clean files from ‘downloads’

rake clean
ruby19 local=C:\ruby-src nogems=1

I’ll repeat Luis’ feedback…we need details on what you’re doing and
your results. Your situation may require other steps than that the
general use case that I listed.

Jon

(Should I move to RubyInstaller list with this?)

Yesterday’s problem was probably my fault (I messed up paths), that’s
why I didn’t provide any logs.

I tried again today, this time correctly ;), and it still doesn’t work.

Here’s the log: http://pastebin.com/ugmNRwMs This error is repeatable

  • the same thing happens if I try running again. (Sorry for the format
  • I copied it from the console.)

I also had some troubles with OpenSSL or something and a timeout - but
when I ran the command again it worked. Sorry, I didn’t save the log.

2011/10/1 Jon F. [email protected]:

One potential barf area to check is whether there are 0 byte archives in
the ‘downloads’ sub-directory of your RubyInstaller repo clone. While
it’s caused by network problems, the current scripts don’t deal with
this failure as gracefully as we’d like.

I checked, that’s not it.

– Matma R.

Bartosz Dziewoński wrote in post #1024581:

(Should I move to RubyInstaller list with this?)

Yesterday’s problem was probably my fault (I messed up paths), that’s
why I didn’t provide any logs.

I tried again today, this time correctly ;), and it still doesn’t work.

Here’s the log: http://pastebin.com/ugmNRwMs This error is repeatable

  • the same thing happens if I try running again. (Sorry for the format
  • I copied it from the console.)

I also had some troubles with OpenSSL or something and a timeout - but
when I ran the command again it worked. Sorry, I didn’t save the log.

You haven’t provided details of your starting point (version of Ruby,
Git and Windows)

I’m not able to reproduce the error you display in your pastie.

Will be better if you bring this conversation to RubyInstaller group so
other s might be able to assist you better:

http://groups.google.com/group/rubyinstaller/


Luis L.

Bartosz Dziewoński wrote in post #1024428:

I did in fact try compiling on my own using Jon’s instructions, but
after long downloads and loooong crunching the rake was “aborted”.
I’ll keep trying, of course (I have some suspicions as to why it
barfed), but a premade package would be greatly appreciated :slight_smile:

If you don’t include the output of the “barf”, there is little we can do
to help.

There are thousands of reasons rake could have aborted execution,
ranging from the version of Ruby you’re using to Rake, download issues
or specific details of your machine.

Without these details, we are clueless, same as you.

Feel free to post to RubyInstaller list if you need assistance on
getting the environment to compile:

http://groups.google.com/group/rubyinstaller/

Perhaps there is something on Jon’s instructions that is missing.


Luis L.