Re: [lmi] [lmi-commits] master 066d739f 3/3: Update mortality tables for 'sample' product

[Vadim Zeitlin]

On Fri, 13 May 2022 20:13:15 -0400 (EDT) Greg Chicares 
<gchicares@sbcglobal.net> wrote:

GC> branch: master
GC> commit 066d739fab21dfa17c63b2dbaa533c5e0fd267e3
GC> Author: Gregory W. Chicares <gchicares@sbcglobal.net>
GC> Commit: Gregory W. Chicares <gchicares@sbcglobal.net>
GC> 
GC>     Update mortality tables for 'sample' product
GC>     
GC>     See:
GC>       https://lists.nongnu.org/archive/html/lmi/2022-05/msg00006.html
GC>       https://lists.nongnu.org/archive/html/lmi/2022-05/msg00015.html
GC> ---
GC>  install_miscellanea.make | 2 +-
GC>  1 file changed, 1 insertion(+), 1 deletion(-)
GC> 
GC> diff --git a/install_miscellanea.make b/install_miscellanea.make
GC> index 07ff44d1..a53e7941 100644
GC> --- a/install_miscellanea.make
GC> +++ b/install_miscellanea.make
GC> @@ -45,7 +45,7 @@ third_party_source_dir  := $(dest_dir)/src
GC>  
GC>  cgicc_archive    := cgicc-3.1.4.tar.bz2
GC>  jing_archive     := jing-20091111.zip
GC> -sample_archive   := lmi-data-20050618T1440Z.tar.bz2
GC> +sample_archive   := lmi-data-20220511T0107Z.tar.bz2
GC>  trang_archive    := trang-20091111.zip
GC>  
GC>  file_list := \

 I was surprised to see that this commit didn't update the sample archive
MD5 and, for a moment, suspected that the checksum verification code in the
makefile got broken without anyone noticing. But then I got quickly
reassure by the multiple failures in all the CI builds: the verification
does work and all the builds failed because of this with the errors like
this:

echo "e7f07133abfc3b9c2252dfa3b61191bc 
*/srv/cache_for_lmi/downloads/lmi-data-20220511T0107Z.tar.bz2" | md5sum --check
md5sum: WARNING: 1 computed checksum did NOT match
/srv/cache_for_lmi/downloads/lmi-data-20220511T0107Z.tar.bz2: FAILED
make: *** [install_miscellanea.make:211: lmi-data-20220511T0107Z.tar.bz2] Error 
1

and so there is no problem here -- except that the builds are broken. So
I've made this commit:

---------------------------------- >8 --------------------------------------
commit bbdb0dca3441c38557d940d0d1ea269e4438e685
Author: Vadim Zeitlin <vadim@tt-solutions.com>
Date:   2022-05-14 15:45:32 +0200

    Fix sample archive MD5 after updating the file contents

    This should have been part of 066d739fa (Update mortality tables for
    'sample' product, 2022-05-14).

diff --git a/install_miscellanea.make b/install_miscellanea.make
index a53e7941d..fcb8bd577 100644
--- a/install_miscellanea.make
+++ b/install_miscellanea.make
@@ -67,7 +67,7 @@ $(trang_archive)-url    := 
https://storage.googleapis.com/google-code-archive-do

 $(cgicc_archive)-md5    := 6cb5153fc9fa64b4e50c7962aa557bbe
 $(jing_archive)-md5     := 13eef193921409a1636377d1efbf9843
-$(sample_archive)-md5   := e7f07133abfc3b9c2252dfa3b61191bc
+$(sample_archive)-md5   := 7f37a83f8bb112d9077a541e91cbd6cf
 $(trang_archive)-md5    := 9d31799b948c350850eb9dd14e5b832d

 # Utilities 
####################################################################
---------------------------------- >8 --------------------------------------

to fix the problem, but this didn't work neither because of the following
error:

tar --extract --keep-old-files --bzip2 --directory=/opt/lmi/third_party/ad_hoc 
--file=/srv/cache_for_lmi/downloads/lmi-data-20220511T0107Z.tar.bz2
bzip2: (stdin) is not a bzip2 file.
tar: Child returned status 2
tar: Error is not recoverable: exiting now
make: [install_miscellanea.make:212: lmi-data-20220511T0107Z.tar.bz2] Error 2 
(ignored)

And, downloading the file, I see that it's indeed a plain tar and not a
bzip2 archive. I don't know if I have permissions to reupload a new file to
Savannah download area (and if I do, then I don't know how to do it), so
I've done this commit for testing only instead:
---------------------------------- >8 --------------------------------------
commit 5c3a3783c8bb8e083c05d710fa0729403b46b30c
Author: Vadim Zeitlin <vadim@tt-solutions.com>
Date:   2022-05-14 16:02:16 +0200

    Use bzip2 compressed sample files archive

    Update the file name, URL and MD5 again.

diff --git a/install_miscellanea.make b/install_miscellanea.make
index fcb8bd577..232f3caeb 100644
--- a/install_miscellanea.make
+++ b/install_miscellanea.make
@@ -45,7 +45,7 @@ third_party_source_dir  := $(dest_dir)/src

 cgicc_archive    := cgicc-3.1.4.tar.bz2
 jing_archive     := jing-20091111.zip
-sample_archive   := lmi-data-20220511T0107Z.tar.bz2
+sample_archive   := lmi-data-20220514T1554Z.tar.bz2
 trang_archive    := trang-20091111.zip

 file_list := \
@@ -62,12 +62,12 @@ sample:     stem = data

 $(cgicc_archive)-url    := ftp://ftp.gnu.org/pub/gnu/cgicc/$(cgicc_archive)
 $(jing_archive)-url     := 
https://storage.googleapis.com/google-code-archive-downloads/v2/code.google.com/jing-trang/$(jing_archive)
-$(sample_archive)-url   := 
https://download.savannah.gnu.org/releases/lmi/$(sample_archive)
+$(sample_archive)-url   := https://www.zeitlins.org/lmi/$(sample_archive)
 $(trang_archive)-url    := 
https://storage.googleapis.com/google-code-archive-downloads/v2/code.google.com/jing-trang/$(trang_archive)

 $(cgicc_archive)-md5    := 6cb5153fc9fa64b4e50c7962aa557bbe
 $(jing_archive)-md5     := 13eef193921409a1636377d1efbf9843
-$(sample_archive)-md5   := 7f37a83f8bb112d9077a541e91cbd6cf
+$(sample_archive)-md5   := 36e8418d191dcc7b167da4f86f892601
 $(trang_archive)-md5    := 9d31799b948c350850eb9dd14e5b832d

 # Utilities 
####################################################################
---------------------------------- >8 --------------------------------------

where the new archive is just a bzip2-compressed version of the previous
file and with this the CI builds do pass again.

 Could you please compress the sample files archive on Savannah and update
its MD5 on Savannah to make them pass for master as well or, alternatively,
give me permission to do it there?

 Thanks,
VZ

pgpaV1Qmd23B6.pgp
Description: PGP signature