perl.dbi.dev https://www.nntp.perl.org/group/perl.dbi.dev/ ... Copyright 1998-2018 perl.org Sun, 18 Feb 2018 23:57:51 +0000 ask@perl.org Re: DBD::mysql 4.046 released by Patrick M. Galbraith Yes, I will look at this later today.<br/><br/><br/>On 2/9/18 5:34 AM, H.Merijn Brand wrote:<br/>&gt; On Fri, 9 Feb 2018 05:20:25 -0500, &quot;Patrick M. Galbraith&quot;<br/>&gt; &lt;patg@patg.net&gt; wrote:<br/>&gt;<br/>&gt;&gt; Greetings!<br/>&gt;&gt;<br/>&gt;&gt; I&#39;m pleased to announce the release of DBD::mysql 4.046:<br/>&gt;&gt;<br/>&gt;&gt; * Use API function for reconnect; fixes compilation on MariaDB 10.2.6+.<br/>&gt;&gt; &nbsp; Fix provided by Pali.<br/>&gt;&gt; * Fixed broken link to MySQL download page, provided by Mohammad S Anwar.<br/>&gt;&gt; * Spelling fixes, provided by Ville Skytt&auml;.<br/>&gt; Could you please also address issue #232<br/>&gt; https://github.com/perl5-dbi/DBD-mysql/issues/232<br/>&gt;<br/>&gt; As DBD::mysql won&#39;t build at all on the affected systems<br/>&gt; Happy to help with questions you might have<br/>&gt;<br/>&gt;&gt; Many thanks to the Pali Roh&aacute;r, Ville Skytt&auml;, Michiel Beijen, and of<br/>&gt;&gt; course the community!<br/><br/>-- <br/>--<br/><br/>Patrick M. Galbraith<br/>Principal Platform Engineer, Oracle + Dyn<br/>http://patg.net<br/>patg@patg.net<br/>patrick.m.galbraith@oracle.com<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/02/msg8163.html Fri, 09 Feb 2018 12:24:31 +0000 Re: DBD::mysql 4.046 released by H.Merijn Brand On Fri, 9 Feb 2018 05:20:25 -0500, &quot;Patrick M. Galbraith&quot; <br/>&lt;patg@patg.net&gt; wrote: <br/> <br/>&gt; Greetings! <br/>&gt; <br/>&gt; I&#39;m pleased to announce the release of DBD::mysql 4.046: <br/>&gt; <br/>&gt; * Use API function for reconnect; fixes compilation on MariaDB 10.2.6+. <br/>&gt; &nbsp; Fix provided by Pali. <br/>&gt; * Fixed broken link to MySQL download page, provided by Mohammad S Anwar. <br/>&gt; * Spelling fixes, provided by Ville Skytt&auml;. <br/> <br/>Could you please also address issue #232 <br/>https://github.com/perl5-dbi/DBD-mysql/issues/232 <br/> <br/>As DBD::mysql won&#39;t build at all on the affected systems <br/>Happy to help with questions you might have <br/> <br/>&gt; Many thanks to the Pali Roh&aacute;r, Ville Skytt&auml;, Michiel Beijen, and of <br/>&gt; course the community! <br/> <br/>-- <br/>H.Merijn Brand http://tux.nl Perl Monger http://amsterdam.pm.org/ <br/>using perl5.00307 .. 5.27 porting perl5 on HP-UX, AIX, and openSUSE <br/>http://mirrors.develooper.com/hpux/ http://www.test-smoke.org/ <br/>http://qa.perl.org http://www.goldmark.org/jeff/stupid-disclaimers/ <br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/02/msg8162.html Fri, 09 Feb 2018 10:34:44 +0000 DBD::mysql 4.046 released by Patrick M. Galbraith Greetings!<br/><br/>I&#39;m pleased to announce the release of DBD::mysql 4.046:<br/><br/>* Use API function for reconnect; fixes compilation on MariaDB 10.2.6+.<br/> &nbsp; Fix provided by Pali.<br/>* Fixed broken link to MySQL download page, provided by Mohammad S Anwar.<br/>* Spelling fixes, provided by Ville Skytt&auml;.<br/><br/>Many thanks to the Pali Roh&aacute;r, Ville Skytt&auml;, Michiel Beijen, and of <br/>course the community!<br/><br/>Have a great day!<br/><br/>--<br/><br/>Patrick M. Galbraith<br/>Principal Platform Engineer, Oracle + Dyn<br/>http://patg.net<br/>patg@patg.net<br/>patrick.m.galbraith@oracle.com<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/02/msg8161.html Fri, 09 Feb 2018 10:20:49 +0000 Re: DBD::mysql 4.044 released! by pali On Wednesday 31 January 2018 10:46:58 Patrick M. Galbraith wrote:<br/>&gt; &quot;Should I create a ticket?&quot;<br/>&gt; <br/>&gt; Yes please.<br/>&gt; <br/>&gt; Thank you,<br/>&gt; <br/>&gt; Patrick<br/>&gt; <br/>&gt; On 1/25/18 3:44 AM, H.Merijn Brand wrote:<br/>&gt; &gt;On Tue, 23 Jan 2018 08:59:54 -0500, &quot;Patrick M. Galbraith&quot;<br/>&gt; &gt;&lt;patg@patg.net&gt; wrote:<br/>&gt; &gt;<br/>&gt; &gt;&gt;Dear Perl and MySQL community,<br/>&gt; &gt;&gt;<br/>&gt; &gt;&gt;I&#39;m quite pleased to announce the release of DBD::mysql 4.044!<br/>&gt; &gt;Hmm<br/>&gt; &gt;<br/>&gt; &gt;$ rpm -q openssl<br/>&gt; &gt;openssl-1.1.0g-1.1.noarch<br/>&gt; &gt;$ mysql --version<br/>&gt; &gt;mysql Ver 15.1 Distrib 10.2.12-MariaDB, for Linux (x86_64) using readline 5.1<br/>&gt; &gt;$ rpm -qa | grep -i mariadb | sort<br/>&gt; &gt;libmariadb3-3.0.2-4.2.x86_64<br/>&gt; &gt;libmariadb-devel-3.0.2-4.2.x86_64<br/>&gt; &gt;mariadb-10.2.12-1.1.x86_64<br/>&gt; &gt;mariadb-bench-10.2.12-1.1.x86_64<br/>&gt; &gt;mariadb-client-10.2.12-1.1.x86_64<br/>&gt; &gt;mariadb-connector-odbc-3.0.2-4.1.x86_64<br/>&gt; &gt;mariadb-errormessages-10.2.12-1.1.noarch<br/>&gt; &gt;mariadb-tools-10.2.12-1.1.x86_64<br/>&gt; &gt;<br/>&gt; &gt;$ perl -v | fgrep 5.2<br/>&gt; &gt;This is perl 5, version 26, subversion 0 (v5.26.0) built for x86_64-linux-thread-multi-ld<br/>&gt; &gt;$ perl -MV=DBI<br/>&gt; &gt;DBI<br/>&gt; &gt; /pro/lib/perl5/site_perl/5.26.0/x86_64-linux-thread-multi-ld/DBI.pm: 1.639<br/>&gt; &gt;<br/>&gt; &gt;Linux 4.14.14-1-default [openSUSE Tumbleweed 20180122] HP ZBook 15G3 Core(TM) i7-6820HQ CPU @ 2.70GHz/3411(8 cores) x86_64 15963 Mb<br/>&gt; &gt;<br/>&gt; &gt;Should I create a ticket?<br/>&gt; &gt;<br/>...<br/>&gt; &gt;dbdimp.c:2125:13: error: &lsquo;MYSQL {aka struct st_mysql}&rsquo; has no member named &lsquo;reconnect&rsquo;<br/>&gt; &gt; result-&gt;reconnect=0;<br/>&gt; &gt; ^~<br/><br/>Reported:<br/>https://rt.cpan.org/Ticket/Display.html?id=122065<br/>https://rt.cpan.org/Ticket/Display.html?id=122429<br/>https://rt.cpan.org/Ticket/Display.html?id=122431<br/>https://github.com/perl5-dbi/DBD-mysql/issues/139<br/>https://github.com/perl5-dbi/DBD-mysql/issues/169<br/>https://github.com/perl5-dbi/DBD-mysql/issues/218<br/>https://github.com/perl5-dbi/DBD-mysql/issues/219<br/>https://github.com/perl5-dbi/DBD-mysql/pull/133<br/>https://github.com/perl5-dbi/DBD-mysql/pull/224<br/><br/>And fixed:<br/>https://github.com/gooddata/DBD-MariaDB/commit/9cfed385af6c437e87e981022eef440cc2d99777<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8160.html Wed, 31 Jan 2018 17:12:47 +0000 Re: DBD::mysql 4.044 released! by Patrick M. Galbraith &quot;Should I create a ticket?&quot;<br/><br/>Yes please.<br/><br/>Thank you,<br/><br/>Patrick<br/><br/>On 1/25/18 3:44 AM, H.Merijn Brand wrote:<br/>&gt; On Tue, 23 Jan 2018 08:59:54 -0500, &quot;Patrick M. Galbraith&quot;<br/>&gt; &lt;patg@patg.net&gt; wrote:<br/>&gt;<br/>&gt;&gt; Dear Perl and MySQL community,<br/>&gt;&gt;<br/>&gt;&gt; I&#39;m quite pleased to announce the release of DBD::mysql 4.044!<br/>&gt; Hmm<br/>&gt;<br/>&gt; $ rpm -q openssl<br/>&gt; openssl-1.1.0g-1.1.noarch<br/>&gt; $ mysql --version<br/>&gt; mysql Ver 15.1 Distrib 10.2.12-MariaDB, for Linux (x86_64) using readline 5.1<br/>&gt; $ rpm -qa | grep -i mariadb | sort<br/>&gt; libmariadb3-3.0.2-4.2.x86_64<br/>&gt; libmariadb-devel-3.0.2-4.2.x86_64<br/>&gt; mariadb-10.2.12-1.1.x86_64<br/>&gt; mariadb-bench-10.2.12-1.1.x86_64<br/>&gt; mariadb-client-10.2.12-1.1.x86_64<br/>&gt; mariadb-connector-odbc-3.0.2-4.1.x86_64<br/>&gt; mariadb-errormessages-10.2.12-1.1.noarch<br/>&gt; mariadb-tools-10.2.12-1.1.x86_64<br/>&gt;<br/>&gt; $ perl -v | fgrep 5.2<br/>&gt; This is perl 5, version 26, subversion 0 (v5.26.0) built for x86_64-linux-thread-multi-ld<br/>&gt; $ perl -MV=DBI<br/>&gt; DBI<br/>&gt; /pro/lib/perl5/site_perl/5.26.0/x86_64-linux-thread-multi-ld/DBI.pm: 1.639<br/>&gt;<br/>&gt; Linux 4.14.14-1-default [openSUSE Tumbleweed 20180122] HP ZBook 15G3 Core(TM) i7-6820HQ CPU @ 2.70GHz/3411(8 cores) x86_64 15963 Mb<br/>&gt;<br/>&gt; Should I create a ticket?<br/>&gt;<br/>&gt; .cpan/build/DBD-mysql-4.044 509 &gt; make<br/>&gt; &quot;/data/pro/bin/perl5.26.0&quot; -MExtUtils::Command::MM -e &#39;cp_nonempty&#39; -- mysql.bs blib/arch/auto/DBD/mysql/mysql.bs 644<br/>&gt; cc -c -I/pro/lib/perl5/site_perl/5.26.0/x86_64-linux-thread-multi-ld/auto/DBI -I/usr/include/mysql -I/usr/include/mysql/mariadb -DDBD_MYSQL_WITH_SSL -DDBD_MYSQL_INSERT_ID_IS_GOOD -g -D_REENTRANT -D_GNU_SOURCE -fPIC -fwrapv -fno-strict-aliasing -pipe -fstack-protector-strong -I/pro/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -DVERSION=\&quot;4.044\&quot; -DXS_VERSION=\&quot;4.044\&quot; -fPIC &quot;-I/pro/lib/perl5/5.26.0/x86_64-linux-thread-multi-ld/CORE&quot; dbdimp.c<br/>&gt; dbdimp.c: In function &lsquo;mysql_dr_connect&rsquo;:<br/>&gt; dbdimp.c:1987:22: error: &lsquo;SSL_MODE_PREFERRED&rsquo; undeclared (first use in this function); did you mean &lsquo;SO_PEERCRED&rsquo;?<br/>&gt; ssl_mode = SSL_MODE_PREFERRED;<br/>&gt; ^~~~~~~~~~~~~~~~~~<br/>&gt; SO_PEERCRED<br/>&gt; dbdimp.c:1987:22: note: each undeclared identifier is reported only once for each function it appears in<br/>&gt; dbdimp.c:1989:19: error: &lsquo;SSL_MODE_VERIFY_IDENTITY&rsquo; undeclared (first use in this function); did you mean &lsquo;SSL_MODE_PREFERRED&rsquo;?<br/>&gt; ssl_mode = SSL_MODE_VERIFY_IDENTITY;<br/>&gt; ^~~~~~~~~~~~~~~~~~~~~~~~<br/>&gt; SSL_MODE_PREFERRED<br/>&gt; dbdimp.c:1991:19: error: &lsquo;SSL_MODE_VERIFY_CA&rsquo; undeclared (first use in this function); did you mean &lsquo;SSL_MODE_VERIFY_IDENTITY&rsquo;?<br/>&gt; ssl_mode = SSL_MODE_VERIFY_CA;<br/>&gt; ^~~~~~~~~~~~~~~~~~<br/>&gt; SSL_MODE_VERIFY_IDENTITY<br/>&gt; dbdimp.c:1993:19: error: &lsquo;SSL_MODE_REQUIRED&rsquo; undeclared (first use in this function); did you mean &lsquo;SSL_MODE_PREFERRED&rsquo;?<br/>&gt; ssl_mode = SSL_MODE_REQUIRED;<br/>&gt; ^~~~~~~~~~~~~~~~~<br/>&gt; SSL_MODE_PREFERRED<br/>&gt; dbdimp.c:1994:30: error: &lsquo;MYSQL_OPT_SSL_MODE&rsquo; undeclared (first use in this function); did you mean &lsquo;MYSQL_OPT_SSL_CRL&rsquo;?<br/>&gt; if (mysql_options(sock, MYSQL_OPT_SSL_MODE, &amp;ssl_mode) != 0) {<br/>&gt; ^~~~~~~~~~~~~~~~~~<br/>&gt; MYSQL_OPT_SSL_CRL<br/>&gt; dbdimp.c:2072:30: error: &lsquo;SSL_MODE_DISABLED&rsquo; undeclared (first use in this function); did you mean &lsquo;SSL_MODE_REQUIRED&rsquo;?<br/>&gt; unsigned int ssl_mode = SSL_MODE_DISABLED;<br/>&gt; ^~~~~~~~~~~~~~~~~<br/>&gt; SSL_MODE_REQUIRED<br/>&gt; dbdimp.c:2125:13: error: &lsquo;MYSQL {aka struct st_mysql}&rsquo; has no member named &lsquo;reconnect&rsquo;<br/>&gt; result-&gt;reconnect=0;<br/>&gt; ^~<br/>&gt; dbdimp.c: In function &lsquo;mysql_describe&rsquo;:<br/>&gt; dbdimp.c:3967:22: warning: assignment from incompatible pointer type [-Wincompatible-pointer-types]<br/>&gt; buffer-&gt;is_null= (my_bool*) &amp;(fbh-&gt;is_null);<br/>&gt; ^<br/>&gt; dbdimp.c:3968:20: warning: assignment from incompatible pointer type [-Wincompatible-pointer-types]<br/>&gt; buffer-&gt;error= (my_bool*) &amp;(fbh-&gt;error);<br/>&gt; ^<br/>&gt; make: *** [Makefile:353: dbdimp.o] Error 1<br/>&gt;<br/>&gt;&gt; This release in particular contains a re-application of an earlier SSL<br/>&gt;&gt; fix that was needed when we reverted a release last year.<br/>&gt;&gt;<br/>&gt;&gt; Per the Changelog:<br/>&gt;&gt;<br/>&gt;&gt; ---<br/>&gt;&gt;<br/>&gt;&gt; &nbsp;Reapply https://github.com/perl5-dbi/DBD-mysql/pull/114 &quot;Improve SSL<br/>&gt;&gt; settings, reflect changes for BACKRONYM and Riddle vulnerabilities,<br/>&gt;&gt; enforce SSL encryption when mysql_ssl=1 is set&quot;<br/>&gt;&gt; &nbsp; Thank you to Pali Rohar, Dani&euml;l van Eeden, Booking.com, et al:w<br/>&gt;&gt; * Fix parsing configure libs from mysql_config --libs output in Makefile.PL<br/>&gt;&gt; &nbsp; Libraries in mysql_config --libs output can be specified by library name<br/>&gt;&gt; &nbsp; with the -l prefix or by absolute path to library name without any<br/>&gt;&gt; prefix.<br/>&gt;&gt; &nbsp; Parameters must start with a hyphen, so treat all options without leading<br/>&gt;&gt; &nbsp; hyphen in mysql_config --libs output as libraries with full path.<br/>&gt;&gt; &nbsp; Partially fixes bug https://rt.cpan.org/Public/Bug/Display.html?id=100898<br/>&gt;&gt; &nbsp; Fix by Pali Roh&aacute;r.<br/>&gt;&gt; * Return INTs with ZEROFILL as strings. Reported by Knarf, fix by Pali<br/>&gt;&gt; Roh&aacute;r.<br/>&gt;&gt; &nbsp;&nbsp; https://rt.cpan.org/Public/Bug/Display.html?id=118977<br/>&gt;&gt; * Correct require on relative path for perl 5.26. Fix by Grinnz.<br/>&gt;&gt; &nbsp;&nbsp; https://github.com/perl5-dbi/DBD-mysql/pull/136<br/>&gt;&gt;<br/>&gt;&gt; This release has been made possible with the excellent work of Pali<br/>&gt;&gt; Roh&aacute;r, Dani&euml;l van Eeden (http://booking.com), Knarf, Grinnz, and my<br/>&gt;&gt; co-maintainer, Michiel Beijen.<br/>&gt;&gt;<br/>&gt;&gt; Also, thanks for Reggie Burnett (who recently set up a slack room for<br/>&gt;&gt; mysql), Matt Lord, Georg Richter, Paul Dubois, and other former MySQL<br/>&gt;&gt; colleagues who have been helpful in establishing communications about<br/>&gt;&gt; things that need to get done in the driver!<br/>&gt;&gt;<br/>&gt;&gt; I&#39;m particularly please that this release also coincides with my 50th<br/>&gt;&gt; birthday! I have been maintaining DBD::mysql since 2004 now, and look<br/>&gt;&gt; forward to future releases containing further improvements as discussed<br/>&gt;&gt; on the mailing lists and supporting new features in both MySQL and MariaDB.<br/>&gt;&gt;<br/>&gt;&gt; Thank&nbsp; you!<br/>&gt;&gt;<br/>&gt;&gt; Patrick<br/>&gt;&gt;<br/>&gt;&gt; --<br/>&gt;&gt;<br/>&gt;&gt; Patrick M. Galbraith<br/>&gt;&gt; Principal Platform Engineer, Oracle + Dyn<br/>&gt;&gt; http://patg.net<br/>&gt;&gt; patg@patg.net<br/>&gt;&gt; patrick.m.galbraith@oracle.com<br/>&gt;<br/><br/>-- <br/>--<br/><br/>Patrick M. Galbraith<br/>Principal Platform Engineer, Oracle + Dyn<br/>http://patg.net<br/>patg@patg.net<br/>patrick.m.galbraith@oracle.com<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8159.html Wed, 31 Jan 2018 15:47:11 +0000 [perl5-dbi/dbi] by timbunce Branch: refs/tags/1.640<br/> Home: https://github.com/perl5-dbi/dbi<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8158.html Mon, 29 Jan 2018 14:23:01 +0000 [perl5-dbi/dbi] 95bcd9: Update Changes. Bump version. by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: 95bcd9907875c39f76a21b8df8bd47424c5506ee<br/> https://github.com/perl5-dbi/dbi/commit/95bcd9907875c39f76a21b8df8bd47424c5506ee<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2018-01-28 (Sun, 28 Jan 2018)<br/><br/> Changed paths:<br/> M Changes<br/> M DBI.pm<br/><br/> Log Message:<br/> -----------<br/> Update Changes. Bump version.<br/><br/><br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8157.html Mon, 29 Jan 2018 14:23:00 +0000 [perl5-dbi/dbi] 8654af: Fix test t/91_store_warning.t by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: 8654af8fd0bf287b1429eb354e2102859ae74d41<br/> https://github.com/perl5-dbi/dbi/commit/8654af8fd0bf287b1429eb354e2102859ae74d41<br/> Author: Pali &lt;pali@cpan.org&gt;<br/> Date: 2018-01-27 (Sat, 27 Jan 2018)<br/><br/> Changed paths:<br/> M t/91_store_warning.t<br/><br/> Log Message:<br/> -----------<br/> Fix test t/91_store_warning.t<br/><br/>Use \x{263A} syntax instead of \N{U+263A} in qr//. It looks like \N does not work in qr// with Perl 5.10.0.<br/><br/><br/> Commit: fd08adff477b27a62c209f56341fff7d864bb1bc<br/> https://github.com/perl5-dbi/dbi/commit/fd08adff477b27a62c209f56341fff7d864bb1bc<br/> Author: Pali &lt;pali@cpan.org&gt;<br/> Date: 2018-01-27 (Sat, 27 Jan 2018)<br/><br/> Changed paths:<br/> M .travis.yml<br/><br/> Log Message:<br/> -----------<br/> Add Perl 5.10.0 and 5.8.1 to Travis testing<br/><br/><br/> Commit: 34ec603b6f08dbd2499fbe3a6c453caa3f091a06<br/> https://github.com/perl5-dbi/dbi/commit/34ec603b6f08dbd2499fbe3a6c453caa3f091a06<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2018-01-28 (Sun, 28 Jan 2018)<br/><br/> Changed paths:<br/> M .travis.yml<br/> M t/91_store_warning.t<br/><br/> Log Message:<br/> -----------<br/> Merge pull request #57 from pali/warning<br/><br/>Fix test t/91_store_warning.t for Perl 5.10.0<br/><br/><br/>Compare: https://github.com/perl5-dbi/dbi/compare/be4dc7873a33...34ec603b6f08 https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8156.html Mon, 29 Jan 2018 14:23:00 +0000 [perl5-dbi/dbi] 752baa: Add mariadb_ prefix for a new DBD::MariaDBdriver by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: 752baaf2b66270f30c064ee1eae6bd4ff2a81a78<br/> https://github.com/perl5-dbi/dbi/commit/752baaf2b66270f30c064ee1eae6bd4ff2a81a78<br/> Author: Pali &lt;pali@cpan.org&gt;<br/> Date: 2018-01-25 (Thu, 25 Jan 2018)<br/><br/> Changed paths:<br/> M DBI.pm<br/><br/> Log Message:<br/> -----------<br/> Add mariadb_ prefix for a new DBD::MariaDB driver<br/><br/>A new DBD::MariaDB driver is a fork of DBD::mysql and is currently<br/>developed at github: https://github.com/gooddata/DBD-MariaDB<br/><br/>Without registering DBI prefix, DBD::MariaDB itself cannot register own<br/>methods on $dbh or $sth handles and throw error:<br/><br/>method name prefix &#39;mariadb_&#39; is not associated with a registered driver<br/><br/><br/> Commit: 2e1b45c728cc92f470d971dcc2510a524c5d529b<br/> https://github.com/perl5-dbi/dbi/commit/2e1b45c728cc92f470d971dcc2510a524c5d529b<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2018-01-28 (Sun, 28 Jan 2018)<br/><br/> Changed paths:<br/> M DBI.pm<br/><br/> Log Message:<br/> -----------<br/> Merge pull request #56 from pali/master<br/><br/>Add mariadb_ prefix for a new DBD::MariaDB driver<br/><br/><br/>Compare: https://github.com/perl5-dbi/dbi/compare/34ec603b6f08...2e1b45c728cc https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8155.html Mon, 29 Jan 2018 14:22:42 +0000 Re: DBD::mysql 4.044 released! by H.Merijn Brand On Tue, 23 Jan 2018 08:59:54 -0500, &quot;Patrick M. Galbraith&quot; <br/>&lt;patg@patg.net&gt; wrote: <br/> <br/>&gt; Dear Perl and MySQL community, <br/>&gt; <br/>&gt; I&#39;m quite pleased to announce the release of DBD::mysql 4.044! <br/> <br/>Hmm <br/> <br/>$ rpm -q openssl <br/>openssl-1.1.0g-1.1.noarch <br/>$ mysql --version <br/>mysql Ver 15.1 Distrib 10.2.12-MariaDB, for Linux (x86_64) using readline 5.1 <br/>$ rpm -qa | grep -i mariadb | sort <br/>libmariadb3-3.0.2-4.2.x86_64 <br/>libmariadb-devel-3.0.2-4.2.x86_64 <br/>mariadb-10.2.12-1.1.x86_64 <br/>mariadb-bench-10.2.12-1.1.x86_64 <br/>mariadb-client-10.2.12-1.1.x86_64 <br/>mariadb-connector-odbc-3.0.2-4.1.x86_64 <br/>mariadb-errormessages-10.2.12-1.1.noarch <br/>mariadb-tools-10.2.12-1.1.x86_64 <br/> <br/>$ perl -v | fgrep 5.2 <br/>This is perl 5, version 26, subversion 0 (v5.26.0) built for x86_64-linux-thread-multi-ld <br/>$ perl -MV=DBI <br/>DBI <br/> /pro/lib/perl5/site_perl/5.26.0/x86_64-linux-thread-multi-ld/DBI.pm: 1.639 <br/> <br/>Linux 4.14.14-1-default [openSUSE Tumbleweed 20180122] HP ZBook 15G3 Core(TM) i7-6820HQ CPU @ 2.70GHz/3411(8 cores) x86_64 15963 Mb <br/> <br/>Should I create a ticket? <br/> <br/>.cpan/build/DBD-mysql-4.044 509 &gt; make <br/>&quot;/data/pro/bin/perl5.26.0&quot; -MExtUtils::Command::MM -e &#39;cp_nonempty&#39; -- mysql.bs blib/arch/auto/DBD/mysql/mysql.bs 644 <br/>cc -c -I/pro/lib/perl5/site_perl/5.26.0/x86_64-linux-thread-multi-ld/auto/DBI -I/usr/include/mysql -I/usr/include/mysql/mariadb -DDBD_MYSQL_WITH_SSL -DDBD_MYSQL_INSERT_ID_IS_GOOD -g -D_REENTRANT -D_GNU_SOURCE -fPIC -fwrapv -fno-strict-aliasing -pipe -fstack-protector-strong -I/pro/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -DVERSION=\&quot;4.044\&quot; -DXS_VERSION=\&quot;4.044\&quot; -fPIC &quot;-I/pro/lib/perl5/5.26.0/x86_64-linux-thread-multi-ld/CORE&quot; dbdimp.c <br/>dbdimp.c: In function &lsquo;mysql_dr_connect&rsquo;: <br/>dbdimp.c:1987:22: error: &lsquo;SSL_MODE_PREFERRED&rsquo; undeclared (first use in this function); did you mean &lsquo;SO_PEERCRED&rsquo;? <br/> ssl_mode = SSL_MODE_PREFERRED; <br/> ^~~~~~~~~~~~~~~~~~ <br/> SO_PEERCRED <br/>dbdimp.c:1987:22: note: each undeclared identifier is reported only once for each function it appears in <br/>dbdimp.c:1989:19: error: &lsquo;SSL_MODE_VERIFY_IDENTITY&rsquo; undeclared (first use in this function); did you mean &lsquo;SSL_MODE_PREFERRED&rsquo;? <br/> ssl_mode = SSL_MODE_VERIFY_IDENTITY; <br/> ^~~~~~~~~~~~~~~~~~~~~~~~ <br/> SSL_MODE_PREFERRED <br/>dbdimp.c:1991:19: error: &lsquo;SSL_MODE_VERIFY_CA&rsquo; undeclared (first use in this function); did you mean &lsquo;SSL_MODE_VERIFY_IDENTITY&rsquo;? <br/> ssl_mode = SSL_MODE_VERIFY_CA; <br/> ^~~~~~~~~~~~~~~~~~ <br/> SSL_MODE_VERIFY_IDENTITY <br/>dbdimp.c:1993:19: error: &lsquo;SSL_MODE_REQUIRED&rsquo; undeclared (first use in this function); did you mean &lsquo;SSL_MODE_PREFERRED&rsquo;? <br/> ssl_mode = SSL_MODE_REQUIRED; <br/> ^~~~~~~~~~~~~~~~~ <br/> SSL_MODE_PREFERRED <br/>dbdimp.c:1994:30: error: &lsquo;MYSQL_OPT_SSL_MODE&rsquo; undeclared (first use in this function); did you mean &lsquo;MYSQL_OPT_SSL_CRL&rsquo;? <br/> if (mysql_options(sock, MYSQL_OPT_SSL_MODE, &amp;ssl_mode) != 0) { <br/> ^~~~~~~~~~~~~~~~~~ <br/> MYSQL_OPT_SSL_CRL <br/>dbdimp.c:2072:30: error: &lsquo;SSL_MODE_DISABLED&rsquo; undeclared (first use in this function); did you mean &lsquo;SSL_MODE_REQUIRED&rsquo;? <br/> unsigned int ssl_mode = SSL_MODE_DISABLED; <br/> ^~~~~~~~~~~~~~~~~ <br/> SSL_MODE_REQUIRED <br/>dbdimp.c:2125:13: error: &lsquo;MYSQL {aka struct st_mysql}&rsquo; has no member named &lsquo;reconnect&rsquo; <br/> result-&gt;reconnect=0; <br/> ^~ <br/>dbdimp.c: In function &lsquo;mysql_describe&rsquo;: <br/>dbdimp.c:3967:22: warning: assignment from incompatible pointer type [-Wincompatible-pointer-types] <br/> buffer-&gt;is_null= (my_bool*) &amp;(fbh-&gt;is_null); <br/> ^ <br/>dbdimp.c:3968:20: warning: assignment from incompatible pointer type [-Wincompatible-pointer-types] <br/> buffer-&gt;error= (my_bool*) &amp;(fbh-&gt;error); <br/> ^ <br/>make: *** [Makefile:353: dbdimp.o] Error 1 <br/> <br/>&gt; This release in particular contains a re-application of an earlier SSL <br/>&gt; fix that was needed when we reverted a release last year. <br/>&gt; <br/>&gt; Per the Changelog: <br/>&gt; <br/>&gt; --- <br/>&gt; <br/>&gt; &nbsp;Reapply https://github.com/perl5-dbi/DBD-mysql/pull/114 &quot;Improve SSL <br/>&gt; settings, reflect changes for BACKRONYM and Riddle vulnerabilities, <br/>&gt; enforce SSL encryption when mysql_ssl=1 is set&quot; <br/>&gt; &nbsp; Thank you to Pali Rohar, Dani&euml;l van Eeden, Booking.com, et al:w <br/>&gt; * Fix parsing configure libs from mysql_config --libs output in Makefile.PL <br/>&gt; &nbsp; Libraries in mysql_config --libs output can be specified by library name <br/>&gt; &nbsp; with the -l prefix or by absolute path to library name without any <br/>&gt; prefix. <br/>&gt; &nbsp; Parameters must start with a hyphen, so treat all options without leading <br/>&gt; &nbsp; hyphen in mysql_config --libs output as libraries with full path. <br/>&gt; &nbsp; Partially fixes bug https://rt.cpan.org/Public/Bug/Display.html?id=100898 <br/>&gt; &nbsp; Fix by Pali Roh&aacute;r. <br/>&gt; * Return INTs with ZEROFILL as strings. Reported by Knarf, fix by Pali <br/>&gt; Roh&aacute;r. <br/>&gt; &nbsp;&nbsp; https://rt.cpan.org/Public/Bug/Display.html?id=118977 <br/>&gt; * Correct require on relative path for perl 5.26. Fix by Grinnz. <br/>&gt; &nbsp;&nbsp; https://github.com/perl5-dbi/DBD-mysql/pull/136 <br/>&gt; <br/>&gt; This release has been made possible with the excellent work of Pali <br/>&gt; Roh&aacute;r, Dani&euml;l van Eeden (http://booking.com), Knarf, Grinnz, and my <br/>&gt; co-maintainer, Michiel Beijen. <br/>&gt; <br/>&gt; Also, thanks for Reggie Burnett (who recently set up a slack room for <br/>&gt; mysql), Matt Lord, Georg Richter, Paul Dubois, and other former MySQL <br/>&gt; colleagues who have been helpful in establishing communications about <br/>&gt; things that need to get done in the driver! <br/>&gt; <br/>&gt; I&#39;m particularly please that this release also coincides with my 50th <br/>&gt; birthday! I have been maintaining DBD::mysql since 2004 now, and look <br/>&gt; forward to future releases containing further improvements as discussed <br/>&gt; on the mailing lists and supporting new features in both MySQL and MariaDB. <br/>&gt; <br/>&gt; Thank&nbsp; you! <br/>&gt; <br/>&gt; Patrick <br/>&gt; <br/>&gt; -- <br/>&gt; <br/>&gt; Patrick M. Galbraith <br/>&gt; Principal Platform Engineer, Oracle + Dyn <br/>&gt; http://patg.net <br/>&gt; patg@patg.net <br/>&gt; patrick.m.galbraith@oracle.com <br/> <br/> <br/>-- <br/>H.Merijn Brand http://tux.nl Perl Monger http://amsterdam.pm.org/ <br/>using perl5.00307 .. 5.27 porting perl5 on HP-UX, AIX, and openSUSE <br/>http://mirrors.develooper.com/hpux/ http://www.test-smoke.org/ <br/>http://qa.perl.org http://www.goldmark.org/jeff/stupid-disclaimers/ <br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8154.html Thu, 25 Jan 2018 08:44:36 +0000 Re: New DBI driver DBD::MariaDB by pali On Tuesday 23 January 2018 11:14:13 pali@cpan.org wrote:<br/>&gt; * Better compatibility with the MariaDB client library<br/><br/>Basic unit tests passed on Travis with last MySQL versions 5.5.58,<br/>5.6.38, 5.7.20, 8.0.3 and MariaDB versions 5.5.58, 10.0.33, 10.1.30,<br/>10.2.12, 10.3.3. Also with MariaDB C/Connector 2.3.4 and 3.0.2.<br/><br/>Therefore DBD::MariaDB is working with MariaDB 10.2 and 10.3 series.<br/><br/>Whole test matrix (about 100 passed configurations) is available at:<br/>https://travis-ci.org/gooddata/DBD-MariaDB<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8153.html Tue, 23 Jan 2018 15:21:25 +0000 DBD::mysql 4.044 released! by Patrick M. Galbraith Dear Perl and MySQL community,<br/><br/>I&#39;m quite pleased to announce the release of DBD::mysql 4.044!<br/><br/>This release in particular contains a re-application of an earlier SSL <br/>fix that was needed when we reverted a release last year.<br/><br/>Per the Changelog:<br/><br/>---<br/><br/> &nbsp;Reapply https://github.com/perl5-dbi/DBD-mysql/pull/114 &quot;Improve SSL <br/>settings, reflect changes for BACKRONYM and Riddle vulnerabilities, <br/>enforce SSL encryption when mysql_ssl=1 is set&quot;<br/> &nbsp; Thank you to Pali Rohar, Dani&euml;l van Eeden, Booking.com, et al:w<br/>* Fix parsing configure libs from mysql_config --libs output in Makefile.PL<br/> &nbsp; Libraries in mysql_config --libs output can be specified by library name<br/> &nbsp; with the -l prefix or by absolute path to library name without any <br/>prefix.<br/> &nbsp; Parameters must start with a hyphen, so treat all options without leading<br/> &nbsp; hyphen in mysql_config --libs output as libraries with full path.<br/> &nbsp; Partially fixes bug https://rt.cpan.org/Public/Bug/Display.html?id=100898<br/> &nbsp; Fix by Pali Roh&aacute;r.<br/>* Return INTs with ZEROFILL as strings. Reported by Knarf, fix by Pali <br/>Roh&aacute;r.<br/> &nbsp;&nbsp; https://rt.cpan.org/Public/Bug/Display.html?id=118977<br/>* Correct require on relative path for perl 5.26. Fix by Grinnz.<br/> &nbsp;&nbsp; https://github.com/perl5-dbi/DBD-mysql/pull/136<br/><br/>This release has been made possible with the excellent work of Pali <br/>Roh&aacute;r, Dani&euml;l van Eeden (http://booking.com), Knarf, Grinnz, and my <br/>co-maintainer, Michiel Beijen.<br/><br/>Also, thanks for Reggie Burnett (who recently set up a slack room for <br/>mysql), Matt Lord, Georg Richter, Paul Dubois, and other former MySQL <br/>colleagues who have been helpful in establishing communications about <br/>things that need to get done in the driver!<br/><br/>I&#39;m particularly please that this release also coincides with my 50th <br/>birthday! I have been maintaining DBD::mysql since 2004 now, and look <br/>forward to future releases containing further improvements as discussed <br/>on the mailing lists and supporting new features in both MySQL and MariaDB.<br/><br/>Thank&nbsp; you!<br/><br/>Patrick<br/><br/>--<br/><br/>Patrick M. Galbraith<br/>Principal Platform Engineer, Oracle + Dyn<br/>http://patg.net<br/>patg@patg.net<br/>patrick.m.galbraith@oracle.com<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8152.html Tue, 23 Jan 2018 14:00:50 +0000 New DBI driver DBD::MariaDB by pali Hello,<br/><br/>I would like to announce a new DBI driver for MariaDB and MySQL<br/>databases. As stated in the email [1] which I sent to this mailing list<br/>about a half year ago, we in the GoodData company decided to create a<br/>fork of the old DBD::mysql driver under the name DBD::MariaDB.<br/><br/>The source code of this new driver is now public, avaiable in our git<br/>repository on GitHub [2]. After we declare it stable we would like to<br/>upload it to CPAN, too.<br/><br/>We hope that this new DBD::MariaDB driver would be useful also for<br/>others and we are open for pull requests or other bug fixes from the<br/>community.<br/><br/>Major changes in DBD::MariaDB since DBD::mysql 4.043 are:<br/><br/>* Correct Unicode support according to the standard Perl Unicode model<br/>* Better compatibility with the MariaDB client library<br/>* Fixes for security issues, SSL/TLS (BACKRONYM and Riddle vulnerabilities)<br/>* Fixes for ZEROFILL columns and magic scalars (needed by Amavis)<br/>* Extended Travis-CI testing<br/><br/>[1] - https://www.nntp.perl.org/group/perl.dbi.dev/2017/08/msg8030.html<br/>[2] - https://github.com/gooddata/DBD-MariaDB<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8151.html Tue, 23 Jan 2018 10:14:25 +0000 Re: DBD::mysql and SSL/TLS by Michiel Beijen Thanks for pointing this out. We had applied these patches before but had <br/>to revert because of breakage created by other changes in the code. <br/> <br/>The difficulty is also that mariadb and MySQL use a different approach to <br/>handling TLS in the client libs. <br/> <br/>I&#39;ll make sure we&#39;ll apply this PR again and create a release ASAP. <br/> <br/>-- <br/>Michiel <br/> <br/>Op 14 jan. 2018 06:49 schreef &quot;Dani&Atilde;&laquo;l van Eeden via dbi-dev&quot; &lt; <br/>dbi-dev@perl.org&gt;: <br/> <br/>&gt; Hi, <br/>&gt; <br/>&gt; I have some serious concerns about the state of SSL/TLS in DBD::mysql. <br/>&gt; <br/>&gt; Issue 1: CVE-2017-10789 isn&#39;t fixed <br/>&gt; https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2017-10789 <br/>&gt; <br/>&gt; Issue 2: Using DBD::mysql with MariaDB 10.0 or higher or MySQL 8.0 or <br/>&gt; higher provides a false sense of security <br/>&gt; <br/>&gt; SSL_LAST_VERIFY_VERSION is set to 50799. <br/>&gt; Any version higher than that silently ignores mysql_ssl_verify_server_cert <br/>&gt; <br/>&gt; This can lead to unencrypted connections even with strict SSL settings. <br/>&gt; <br/>&gt; Issue 3: If SSL support is unavailable but ssl options are set then these <br/>&gt; options are silently ignored. <br/>&gt; <br/>&gt; issue 4: If compiled against MySQL 5.7 then SSL/TLS is used when <br/>&gt; available, but can&#39;t be disabled. (mysql_ssl=0 is ignored). <br/>&gt; <br/>&gt; This makes upgrading to 5.7 more difficult. And 5.7 is needed to get <br/>&gt; support for TLSv1.1 and TLSv1.2. <br/>&gt; <br/>&gt; There is a patch available for this: <br/>&gt; https://github.com/perl5-dbi/DBD-mysql/pull/114 <br/>&gt; <br/>&gt; <br/>&gt; -- <br/>&gt; Dani&Atilde;&laquo;l van Eeden <br/>&gt; Database Administrator <br/>&gt; <br/>&gt; Booking.com B.V. <br/>&gt; Vijzelstraat 66 <br/>&gt; &lt;https://maps.google.com/?q=Vijzelstraat+66&amp;entry=gmail&amp;source=g&gt;-80 <br/>&gt; Amsterdam 1017HL Netherlands <br/>&gt; Direct +31207033812 &lt;020%20703%203812&gt; <br/>&gt; [image: Booking.com] &lt;http://www.booking.com/&gt; <br/>&gt; The world&#39;s #1 accommodation site <br/>&gt; 43 languages, 187+ offices worldwide, 96,000+ global destinations, <br/>&gt; 1,200,000+ room nights booked every day <br/>&gt; No booking fees, best price always guaranteed <br/>&gt; Subsidiary of the Priceline Group (NASDAQ: PCLN) <br/>&gt; <br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8150.html Sun, 14 Jan 2018 07:38:48 +0000 DBD::mysql and SSL/TLS by =?UTF-8?Q?Dani=C3=ABl_van_Eeden?= via dbi-dev Hi, <br/> <br/>I have some serious concerns about the state of SSL/TLS in DBD::mysql. <br/> <br/>Issue 1: CVE-2017-10789 isn&#39;t fixed <br/>https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2017-10789 <br/> <br/>Issue 2: Using DBD::mysql with MariaDB 10.0 or higher or MySQL 8.0 or <br/>higher provides a false sense of security <br/> <br/>SSL_LAST_VERIFY_VERSION is set to 50799. <br/>Any version higher than that silently ignores mysql_ssl_verify_server_cert <br/> <br/>This can lead to unencrypted connections even with strict SSL settings. <br/> <br/>Issue 3: If SSL support is unavailable but ssl options are set then these <br/>options are silently ignored. <br/> <br/>issue 4: If compiled against MySQL 5.7 then SSL/TLS is used when available, <br/>but can&#39;t be disabled. (mysql_ssl=0 is ignored). <br/> <br/>This makes upgrading to 5.7 more difficult. And 5.7 is needed to get <br/>support for TLSv1.1 and TLSv1.2. <br/> <br/>There is a patch available for this: <br/>https://github.com/perl5-dbi/DBD-mysql/pull/114 <br/> <br/> <br/>-- <br/>Dani&Atilde;&laquo;l van Eeden <br/>Database Administrator <br/> <br/>Booking.com B.V. <br/>Vijzelstraat 66-80 Amsterdam 1017HL Netherlands <br/>Direct +31207033812 <br/>[image: Booking.com] &lt;http://www.booking.com/&gt; <br/>The world&#39;s #1 accommodation site <br/>43 languages, 187+ offices worldwide, 96,000+ global destinations, <br/>1,200,000+ room nights booked every day <br/>No booking fees, best price always guaranteed <br/>Subsidiary of the Priceline Group (NASDAQ: PCLN) <br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8149.html Sun, 14 Jan 2018 05:47:51 +0000 [perl5-dbi/dbi] be4dc7: Update Changes to record 1.639 release date by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: be4dc7873a33f3d4ef56812ed01a34587d71aa15<br/> https://github.com/perl5-dbi/dbi/commit/be4dc7873a33f3d4ef56812ed01a34587d71aa15<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2017-12-30 (Sat, 30 Dec 2017)<br/><br/> Changed paths:<br/> M Changes<br/><br/> Log Message:<br/> -----------<br/> Update Changes to record 1.639 release date<br/><br/><br/> https://www.nntp.perl.org/group/perl.dbi.dev/2018/01/msg8148.html Mon, 01 Jan 2018 16:19:21 +0000 [perl5-dbi/dbi] by timbunce Branch: refs/tags/1.639<br/> Home: https://github.com/perl5-dbi/dbi<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/12/msg8147.html Thu, 28 Dec 2017 15:04:43 +0000 [perl5-dbi/dbi] d497cd: Add new Changes in DBI 1.638 section withlatest c... by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: d497cd63c78f13fba0c20a6d23d40f9b99768e7b<br/> https://github.com/perl5-dbi/dbi/commit/d497cd63c78f13fba0c20a6d23d40f9b99768e7b<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2017-12-28 (Thu, 28 Dec 2017)<br/><br/> Changed paths:<br/> M Changes<br/><br/> Log Message:<br/> -----------<br/> Add new Changes in DBI 1.638 section with latest changes.<br/><br/><br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/12/msg8146.html Thu, 28 Dec 2017 15:04:41 +0000 [perl5-dbi/dbi] 5deedc: Implement warn_sv/croak_sv with UTF-8support also... by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: 5deedcfee79137c71a2183830f709da3bbfae77a<br/> https://github.com/perl5-dbi/dbi/commit/5deedcfee79137c71a2183830f709da3bbfae77a<br/> Author: Pali &lt;pali@cpan.org&gt;<br/> Date: 2017-08-14 (Mon, 14 Aug 2017)<br/><br/> Changed paths:<br/> M DBI.xs<br/> M t/91_store_warning.t<br/><br/> Log Message:<br/> -----------<br/> Implement warn_sv/croak_sv with UTF-8 support also for Perl prior to 5.14<br/><br/>Calling croak(NULL) would take exception from the $@ without loosing<br/>SVf_UTF8 flag with a bonus which support object references. So it is better<br/>equivalent for croak_sv() from Perl 5.14+.<br/><br/>Calling warn(&quot;%&quot; SVf, SVfARG(sv)) is what older Perl&#39;s pp_warn() does to<br/>pass scalar independently of SVf_UTF8 flag.<br/><br/>Fixes RT#102404 for Perl prior to 5.14.<br/><br/><br/> Commit: 654f47a4db12f6af9b6802415bc31888a985527b<br/> https://github.com/perl5-dbi/dbi/commit/654f47a4db12f6af9b6802415bc31888a985527b<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2017-12-28 (Thu, 28 Dec 2017)<br/><br/> Changed paths:<br/> M DBI.xs<br/> M t/91_store_warning.t<br/><br/> Log Message:<br/> -----------<br/> Merge pull request #53 from pali/master<br/><br/>Implement warn_sv/croak_sv with UTF-8 support also for Perl prior to 5.14<br/><br/><br/>Compare: https://github.com/perl5-dbi/dbi/compare/d497cd63c78f...654f47a4db12 https://www.nntp.perl.org/group/perl.dbi.dev/2017/12/msg8145.html Thu, 28 Dec 2017 15:04:41 +0000 [perl5-dbi/dbi] by timbunce Branch: refs/tags/1.638<br/> Home: https://github.com/perl5-dbi/dbi<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/12/msg8144.html Thu, 28 Dec 2017 15:04:38 +0000 [perl5-dbi/dbi] 7c5683: Update Changes. Add t/54_dbd_mem.t toMANIFEST by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: 7c5683771d5f8c384ce78b4d174ead45e98c1901<br/> https://github.com/perl5-dbi/dbi/commit/7c5683771d5f8c384ce78b4d174ead45e98c1901<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2017-12-28 (Thu, 28 Dec 2017)<br/><br/> Changed paths:<br/> M Changes<br/> M MANIFEST<br/><br/> Log Message:<br/> -----------<br/> Update Changes. Add t/54_dbd_mem.t to MANIFEST<br/><br/><br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/12/msg8143.html Thu, 28 Dec 2017 15:04:31 +0000 [perl5-dbi/dbi] 206032: Bump version to 1.639. Add lib/DBD/Mem.pm toMANIF... by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: 2060321973fca6ce324250879a89b4e013b5a952<br/> https://github.com/perl5-dbi/dbi/commit/2060321973fca6ce324250879a89b4e013b5a952<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2017-12-28 (Thu, 28 Dec 2017)<br/><br/> Changed paths:<br/> M DBI.pm<br/> M MANIFEST<br/><br/> Log Message:<br/> -----------<br/> Bump version to 1.639. Add lib/DBD/Mem.pm to MANIFEST.<br/><br/>Version 1.638 was tagged too soon, before I noticed lib/DBD/Mem.pm was missing.<br/><br/><br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/12/msg8142.html Thu, 28 Dec 2017 15:04:28 +0000 [perl5-dbi/dbi] 08dadf: DBD::Mem - add new pure perl driver DBD::Mem by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: 08dadfe404737721568c35bb9c9dfbbe107ebb4b<br/> https://github.com/perl5-dbi/dbi/commit/08dadfe404737721568c35bb9c9dfbbe107ebb4b<br/> Author: Jens Rehsack &lt;sno@netbsd.org&gt;<br/> Date: 2017-08-29 (Tue, 29 Aug 2017)<br/><br/> Changed paths:<br/> M DBI.pm<br/> A lib/DBD/Mem.pm<br/> A t/54_dbd_mem.t<br/><br/> Log Message:<br/> -----------<br/> DBD::Mem - add new pure perl driver DBD::Mem<br/><br/>Introduce new DBI::SQL::Nano based driver for advanced testing (e.g.<br/>combining DBD::AnyData2 with DBD::CSV, without relying on external<br/>drivers).<br/><br/>Signed-off-by: Jens Rehsack &lt;sno@netbsd.org&gt;<br/><br/><br/> Commit: dc48196b7ad22e321af1119bdbb6d1823c6e8de0<br/> https://github.com/perl5-dbi/dbi/commit/dc48196b7ad22e321af1119bdbb6d1823c6e8de0<br/> Author: Jens Rehsack &lt;sno@netbsd.org&gt;<br/> Date: 2017-08-29 (Tue, 29 Aug 2017)<br/><br/> Changed paths:<br/> M t/48dbi_dbd_sqlengine.t<br/><br/> Log Message:<br/> -----------<br/> t/48dbi_dbd_sqlengine.t: fix SQL syntax<br/><br/>Even if the SQL statements aren&#39;t executed, they should express what is meant.<br/><br/>Signed-off-by: Jens Rehsack &lt;sno@netbsd.org&gt;<br/><br/><br/> Commit: 1cc6ecc2f281697cd039b5d12d3320b7b9e9ad9c<br/> https://github.com/perl5-dbi/dbi/commit/1cc6ecc2f281697cd039b5d12d3320b7b9e9ad9c<br/> Author: Jens Rehsack &lt;sno@netbsd.org&gt;<br/> Date: 2017-09-05 (Tue, 05 Sep 2017)<br/><br/> Changed paths:<br/> M lib/DBD/Mem.pm<br/><br/> Log Message:<br/> -----------<br/> DBD::Mem: add basic documentation<br/><br/>Add some documentation to DBD::Mem explaining where it&#39;s here, what&#39;s the<br/>expected outcome, some basic use cases.<br/><br/>Let&#39;s see where it leads us to ...<br/><br/>Signed-off-by: Jens Rehsack &lt;sno@netbsd.org&gt;<br/><br/><br/> Commit: 2f28cd2ab5b19c6e1b891112e88a7b9da3f42482<br/> https://github.com/perl5-dbi/dbi/commit/2f28cd2ab5b19c6e1b891112e88a7b9da3f42482<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2017-12-28 (Thu, 28 Dec 2017)<br/><br/> Changed paths:<br/> M DBI.pm<br/> A lib/DBD/Mem.pm<br/> M t/48dbi_dbd_sqlengine.t<br/> A t/54_dbd_mem.t<br/><br/> Log Message:<br/> -----------<br/> Merge pull request #42 from rehsack/dbd_mem<br/><br/>Add DBD::Mem driver, a pure-perl in-memory driver using DBI::DBD::SqlEngine.<br/><br/><br/>Compare: https://github.com/perl5-dbi/dbi/compare/41e039cfa04c...2f28cd2ab5b1 https://www.nntp.perl.org/group/perl.dbi.dev/2017/12/msg8141.html Thu, 28 Dec 2017 15:04:26 +0000 [perl5-dbi/dbi] 88e038: Add missing semicolon in documentation by Tim Bunce Branch: refs/heads/master<br/> Home: https://github.com/perl5-dbi/dbi<br/> Commit: 88e03819620d2859c5ea219e5654dec035217922<br/> https://github.com/perl5-dbi/dbi/commit/88e03819620d2859c5ea219e5654dec035217922<br/> Author: Mark Fowler &lt;mark@twoshortplanks.com&gt;<br/> Date: 2017-08-28 (Mon, 28 Aug 2017)<br/><br/> Changed paths:<br/> M DBI.pm<br/><br/> Log Message:<br/> -----------<br/> Add missing semicolon in documentation<br/><br/><br/> Commit: 90d30958b93515a42a3b265d6bcae23584462764<br/> https://github.com/perl5-dbi/dbi/commit/90d30958b93515a42a3b265d6bcae23584462764<br/> Author: Tim Bunce &lt;Tim.Bunce@pobox.com&gt;<br/> Date: 2017-12-28 (Thu, 28 Dec 2017)<br/><br/> Changed paths:<br/> M DBI.pm<br/><br/> Log Message:<br/> -----------<br/> Merge pull request #55 from 2shortplanks/2shortplanks/missing-semi-colon-docs<br/><br/>Add missing semicolon in example in documentation<br/><br/><br/>Compare: https://github.com/perl5-dbi/dbi/compare/654f47a4db12...90d30958b935 https://www.nntp.perl.org/group/perl.dbi.dev/2017/12/msg8140.html Thu, 28 Dec 2017 15:04:26 +0000 Re: DBD::mysql next steps by Michiel Beijen Hi Pali!<br/><br/>On Thu, Nov 16, 2017 at 12:49 PM, &lt;pali@cpan.org&gt; wrote:<br/><br/>&gt; Hi! And what are you going to do with all commits which you reverted in<br/>&gt; 4.043 version? As I remember, months ago you wrote that you reapply<br/>&gt; fixes, but nothing happened. Instead you started merging conflicting<br/>&gt; new pull requests which other people opened on github. This just prevent<br/>&gt; any future reintroduction of fixes which were already done and was<br/>&gt; reverted. Or you going to invent wheel again? I&#39;m really surprised that<br/>&gt; you now want to throw out everything which I did, including security<br/>&gt; fixes for SSL/TLS or new versions of MariaDB/MySQL (with huge test<br/>&gt; setup). Otherwise I do not understand why you have not looked at<br/>&gt; reverted commit yet... Or are you expecting that new after one year I<br/>&gt; should start rebasing all my commits which I did on top of master and<br/>&gt; opening pull requests again and again? Sorry, but this really look that<br/>&gt; you are not interested in new supporting new MariaDB version and now is<br/>&gt; really a good time to create a fork of DBD::mysql which would contains<br/>&gt; support for new MariaDB, huge test coverage and security fixes for<br/>&gt; SSL/TLS.<br/><br/>As you might have seen I&#39;m adding back the commits that were added<br/>between 4.041 and 4.042 piece by piece to the master branch.<br/><br/>I&#39;m trying to prevent breaking stuff so I&#39;m moving slowly.<br/><br/>I&#39;m not expecting you to do any rebasing or sending new pull requests<br/>of code which we had already. I&#39;ll be applying the patches myself.<br/>Also, please note that the huge Travis setup and TLS fixes were really<br/>good and I absolutely am going to add this back to master.<br/><br/>That said, I merged https://github.com/perl5-dbi/DBD-mysql/pull/181<br/>which changes the bug tracker location to Github in the META.yml this<br/>morning; we don&#39;t need to merge everything exactly in order and<br/>accepting this pull request really does not make it impossible to<br/>later add the patches between 4.041 and 4.042 back to the code base.<br/><br/>And of course we *NEED* to support libmysqlclient of MySQL 8 and also<br/>MariaDB, I can&#39;t see us making a new release to CPAN that would *NOT*<br/>support this.<br/><br/>--<br/>Michiel<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8139.html Thu, 16 Nov 2017 12:02:53 +0000 Re: DBD::mysql next steps by pali On Friday 10 November 2017 10:13:55 Patrick M. Galbraith wrote:<br/>&gt; Greetings all!<br/>&gt; <br/>&gt; Michiel and I have been talking, weighing options of what course to take in<br/>&gt; dealing with moving forward-- with the goal of both offering both stability<br/>&gt; and the choice to have the latest functionality and bug fixes as well as<br/>&gt; give contributors the opportunity to be part of overall improvements to the<br/>&gt; driver.<br/>&gt; <br/>&gt; What we are going to do is:<br/>&gt; <br/>&gt; Add to the connection the ability to turn on proper UTF handling with<br/>&gt; &#39;mysql_enable_proper_unicode&#39;. This gives the user volition the option to knowingly<br/>&gt; toggle whether they want the new functionality and understand that<br/>&gt; data structures returned or accepted by the driver might be different<br/>&gt; than without this setting.<br/>&gt; <br/>&gt; The other options had their merits, but we think this will solve the issue<br/>&gt; while keeping the driver unified and prevent divergence.<br/>&gt; <br/>&gt; Thank you for your input over the last couple months-- we look forward to<br/>&gt; moving ahead!<br/>&gt; <br/>&gt; Patrick and Michiel<br/>&gt; <br/><br/>Hi! And what are you going to do with all commits which you reverted in<br/>4.043 version? As I remember, months ago you wrote that you reapply<br/>fixes, but nothing happened. Instead you started merging conflicting<br/>new pull requests which other people opened on github. This just prevent<br/>any future reintroduction of fixes which were already done and was<br/>reverted. Or you going to invent wheel again? I&#39;m really surprised that<br/>you now want to throw out everything which I did, including security<br/>fixes for SSL/TLS or new versions of MariaDB/MySQL (with huge test<br/>setup). Otherwise I do not understand why you have not looked at<br/>reverted commit yet... Or are you expecting that new after one year I<br/>should start rebasing all my commits which I did on top of master and<br/>opening pull requests again and again? Sorry, but this really look that<br/>you are not interested in new supporting new MariaDB version and now is<br/>really a good time to create a fork of DBD::mysql which would contains<br/>support for new MariaDB, huge test coverage and security fixes for<br/>SSL/TLS.<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8138.html Thu, 16 Nov 2017 11:49:42 +0000 Re: Bugtracker DBD::mysql by Michiel Beijen On Tue, Nov 14, 2017 at 5:40 PM, Dan Book &lt;grinnz@gmail.com&gt; wrote:<br/>&gt; I also prefer github but of course it&#39;s mainly up to what is easiest for the<br/>&gt; maintainers. Here is a script for migrating tickets if it&#39;s not what you<br/>&gt; already tried: https://metacpan.org/pod/RTx::ToGitHub<br/><br/>Actually, that is the script I used when I tried the last time around<br/>which did not work for me, I even send back a pull request for an<br/>installation issue I encountered.<br/>(https://github.com/houseabsolute/RTx-ToGitHub/pull/3)<br/><br/>This time I used dagoldens script:<br/>https://github.com/dagolden/zzz-rt-to-github - it mostly worked but<br/>there were some comments created by users who logged in on RT using<br/>OpenID and where the user field started with &#39;http&#39; and on which the<br/>RT API seemed to choke. This was the dirty patch I used:<br/><br/>diff --git a/rt-to-github.pl b/rt-to-github.pl<br/>index c6d20c8..06dca7e 100755<br/>--- a/rt-to-github.pl<br/>+++ b/rt-to-github.pl<br/>@@ -63,6 +63,10 @@ sub _dist_name {<br/><br/> sub _find_from {<br/> my ( $xact ) = @_;<br/>+ if ($xact-&gt;creator =~ /^http/) {<br/>+ # this is because the creator used OpenID<br/>+ return sprintf(&quot;From %s on %s&quot;, lc( $xact-&gt;creator), $xact-&gt;creat<br/>ed);<br/>+ }<br/><br/>So as of today, *the* bug tracker for DBD::mysql is on Github at<br/>https://github.com/perl5-dbi/DBD-mysql/issues and we no longer use RT!<br/><br/>--<br/>Michiel<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8137.html Wed, 15 Nov 2017 16:37:32 +0000 Re: Bugtracker DBD::mysql by Patrick Galbraith I agree on getting of RT to Github. Basically, it&#39;s 2017 and github is <br/>good at what it does.<br/><br/><br/>On 11/14/2017 11:40 AM, Dan Book wrote:<br/>&gt; On Tue, Nov 14, 2017 at 8:34 AM, Michiel Beijen <br/>&gt; &lt;michiel.beijen@gmail.com &lt;mailto:michiel.beijen@gmail.com&gt;&gt; wrote:<br/>&gt;<br/>&gt; On Tue, Nov 14, 2017 at 11:07 AM,&nbsp; &lt;pali@cpan.org<br/>&gt; &lt;mailto:pali@cpan.org&gt;&gt; wrote:<br/>&gt;<br/>&gt; &gt; Hi! As there are only few tickets on github, it would be easier to<br/>&gt; &gt; disable creating new tickets on github and those few which are not<br/>&gt; &gt; resolved yet either move on RT or (if github allows it; which I<br/>&gt; believe<br/>&gt; &gt; yet) continue to exists.<br/>&gt; &gt;<br/>&gt; &gt; And once any migration is ready, then it can be fully switched.<br/>&gt; &gt;<br/>&gt; &gt; Still one bugtracker is better then two, even if it is not on<br/>&gt; github.<br/>&gt;<br/>&gt; Yeah, I&#39;m sure this is *easier*. But I really hate the RT cpan bug<br/>&gt; tracker, so I&#39;d prefer to look again at migrating from RT &gt;&gt; Github -<br/>&gt; if that really would not work for some reason I&#39;ll consider migrating<br/>&gt; the other way around.<br/>&gt;<br/>&gt;<br/>&gt; I also prefer github but of course it&#39;s mainly up to what is easiest <br/>&gt; for the maintainers. Here is a script for migrating tickets if it&#39;s <br/>&gt; not what you already tried: https://metacpan.org/pod/RTx::ToGitHub<br/>&gt;<br/>&gt; -Dan<br/>&gt;<br/><br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8136.html Wed, 15 Nov 2017 15:05:44 +0000 Re: Bugtracker DBD::mysql by Dan Book On Tue, Nov 14, 2017 at 8:34 AM, Michiel Beijen &lt;michiel.beijen@gmail.com&gt;<br/>wrote:<br/><br/>&gt; On Tue, Nov 14, 2017 at 11:07 AM, &lt;pali@cpan.org&gt; wrote:<br/>&gt;<br/>&gt; &gt; Hi! As there are only few tickets on github, it would be easier to<br/>&gt; &gt; disable creating new tickets on github and those few which are not<br/>&gt; &gt; resolved yet either move on RT or (if github allows it; which I believe<br/>&gt; &gt; yet) continue to exists.<br/>&gt; &gt;<br/>&gt; &gt; And once any migration is ready, then it can be fully switched.<br/>&gt; &gt;<br/>&gt; &gt; Still one bugtracker is better then two, even if it is not on github.<br/>&gt;<br/>&gt; Yeah, I&#39;m sure this is *easier*. But I really hate the RT cpan bug<br/>&gt; tracker, so I&#39;d prefer to look again at migrating from RT &gt;&gt; Github -<br/>&gt; if that really would not work for some reason I&#39;ll consider migrating<br/>&gt; the other way around.<br/>&gt;<br/>&gt;<br/>I also prefer github but of course it&#39;s mainly up to what is easiest for<br/>the maintainers. Here is a script for migrating tickets if it&#39;s not what<br/>you already tried: https://metacpan.org/pod/RTx::ToGitHub<br/><br/>-Dan<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8135.html Tue, 14 Nov 2017 16:40:56 +0000 Re: Bugtracker DBD::mysql by Michiel Beijen On Tue, Nov 14, 2017 at 11:07 AM, &lt;pali@cpan.org&gt; wrote:<br/><br/>&gt; Hi! As there are only few tickets on github, it would be easier to<br/>&gt; disable creating new tickets on github and those few which are not<br/>&gt; resolved yet either move on RT or (if github allows it; which I believe<br/>&gt; yet) continue to exists.<br/>&gt;<br/>&gt; And once any migration is ready, then it can be fully switched.<br/>&gt;<br/>&gt; Still one bugtracker is better then two, even if it is not on github.<br/><br/>Yeah, I&#39;m sure this is *easier*. But I really hate the RT cpan bug<br/>tracker, so I&#39;d prefer to look again at migrating from RT &gt;&gt; Github -<br/>if that really would not work for some reason I&#39;ll consider migrating<br/>the other way around.<br/><br/>--<br/>Michiel<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8134.html Tue, 14 Nov 2017 13:35:31 +0000 Re: Bugtracker DBD::mysql by pali On Tuesday 14 November 2017 10:56:58 Michiel Beijen wrote:<br/>&gt; Hi Pali,<br/>&gt; <br/>&gt; On Mon, Nov 13, 2017 at 6:18 PM, &lt;pali@cpan.org&gt; wrote:<br/>&gt; &gt; And I would suggest to disable issue tracker on github as primary bug<br/>&gt; &gt; tracker (according to DBD::mysql documentation) is on RT and also<br/>&gt; &gt; probably all problems are reported there. The worst thing which can be<br/>&gt; &gt; is to have two independent bug trackers, which is current situation.<br/>&gt; <br/>&gt; You&#39;re right, it would be helpful and desirable to have only one bug tracker!<br/>&gt; <br/>&gt; I&#39;ve tried to migrate the existing RT tickets to Github before, but<br/>&gt; that did not work correctly because DBD::mysql uses a Github<br/>&gt; organization and the script I used at the time did not accept this.<br/>&gt; I&#39;ll put it higher on my to-do list again and look into getting that<br/>&gt; done.<br/><br/>Hi! As there are only few tickets on github, it would be easier to<br/>disable creating new tickets on github and those few which are not<br/>resolved yet either move on RT or (if github allows it; which I believe<br/>yet) continue to exists.<br/><br/>And once any migration is ready, then it can be fully switched.<br/><br/>Still one bugtracker is better then two, even if it is not on github.<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8133.html Tue, 14 Nov 2017 10:16:42 +0000 Bugtracker DBD::mysql by Michiel Beijen Hi Pali,<br/><br/>On Mon, Nov 13, 2017 at 6:18 PM, &lt;pali@cpan.org&gt; wrote:<br/>&gt; And I would suggest to disable issue tracker on github as primary bug<br/>&gt; tracker (according to DBD::mysql documentation) is on RT and also<br/>&gt; probably all problems are reported there. The worst thing which can be<br/>&gt; is to have two independent bug trackers, which is current situation.<br/><br/>You&#39;re right, it would be helpful and desirable to have only one bug tracker!<br/><br/>I&#39;ve tried to migrate the existing RT tickets to Github before, but<br/>that did not work correctly because DBD::mysql uses a Github<br/>organization and the script I used at the time did not accept this.<br/>I&#39;ll put it higher on my to-do list again and look into getting that<br/>done.<br/><br/>--<br/>Michiel<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8132.html Tue, 14 Nov 2017 09:57:29 +0000 Re: DBD::mysql next steps by pali And I would suggest to disable issue tracker on github as primary bug<br/>tracker (according to DBD::mysql documentation) is on RT and also<br/>probably all problems are reported there. The worst thing which can be<br/>is to have two independent bug trackers, which is current situation.<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8131.html Mon, 13 Nov 2017 17:18:11 +0000 Re: DBD::mysql next steps by pali On Friday 10 November 2017 10:13:55 Patrick M. Galbraith wrote:<br/>&gt; Greetings all!<br/>&gt; <br/>&gt; Michiel and I have been talking, weighing options of what course to take in<br/>&gt; dealing with moving forward-- with the goal of both offering both stability<br/>&gt; and the choice to have the latest functionality and bug fixes as well as<br/>&gt; give contributors the opportunity to be part of overall improvements to the<br/>&gt; driver.<br/>&gt; <br/>&gt; What we are going to do is:<br/>&gt; <br/>&gt; Add to the connection the ability to turn on proper UTF handling with<br/>&gt; &#39;mysql_enable_proper_unicode&#39;. This gives the user volition the option to knowingly<br/>&gt; toggle whether they want the new functionality and understand that<br/>&gt; data structures returned or accepted by the driver might be different<br/>&gt; than without this setting.<br/>&gt; <br/>&gt; The other options had their merits, but we think this will solve the issue<br/>&gt; while keeping the driver unified and prevent divergence.<br/>&gt; <br/>&gt; Thank you for your input over the last couple months-- we look forward to<br/>&gt; moving ahead!<br/>&gt; <br/>&gt; Patrick and Michiel<br/><br/>Great! Thank you very much, I&#39;m waiting for a new action which would<br/>bring DBD::mysql back to the active development.<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8130.html Sat, 11 Nov 2017 08:53:18 +0000 minor changes in a parser by Martin Kaspar &#137;PNG <br/>&#26;<br/>&#0;&#0;&#0; IHDR&#0;&#0;&#0;&#20;&#0;&#0;&#0;&#20;&#8;&#6;&#0;&#0;&#0;&#141;&#137;&#29; &#0;&#0;&#0;&#1;sRGB&#0;&reg;&Icirc;&#28;&eacute;&#0;&#0;&#0;&#4;gAMA&#0;&#0;&plusmn;&#143;&#11;&uuml;a&#5;&#0;&#0;&#0; pHYs&#0;&#0;&#14;&Atilde;&#0;&#0;&#14;&Atilde;&#1;&Ccedil;o&uml;d&#0;&#0;&#4;9IDAT8O}T[l&#148;E&#20;&thorn;&thorn;&ugrave;&yuml;&frac12;_&Uacute;-t[7&raquo;-&micro;Dk&yen;4&yen;&#141;&#17;&shy;&uml;&#11;5&#4;&#18;&#131;&#137;A &#137;&#26;#&Ntilde;&#4;&szlig;D&frac14;&lt;&#148;Jb$&#6;&#147;&gt;&#152;&#18;&#30; &Auml;h&cent;&ntilde;&#5;&#30;&#136;&lt;&#148;B+J&Ugrave;&#4;&#11;V&ETH;&Ograve;&#2;&#2;u&middot;-{q&middot;&Ucirc;&icirc;v&yuml;&Euml;xf&Uacute;&#154;%&#136;&#147;=&Ugrave;&ugrave;&Iuml;&Igrave;|s&aelig;|&szlig;9J&igrave;&sup1;V^&aring;v&Atilde;&auml;&#28;<br/>&#150;&AElig;&iquest;&#147;e&Ccedil;&Ograve;?&#7;&egrave;&#39;&#135;&Uuml;&sup2;&frac14;&macr;&Uuml;?&#18;m&atilde;^&sect;&#3;F&copy;&#4;S/&#129;&#155;&amp;&#8;&ucirc;&frac34;&iexcl;h*T&#135;&#11;<br/>S&nbsp;(<br/>&cedil;&Aring;&iexcl;&ccedil;g&aring;&middot;&aelig;&ntilde;J&#159;B&ccedil;&#148;&plusmn;&reg;v&#158;J%&Aacute;&ecirc;&#26;P&ugrave;t&#20;&ouml;&Uacute;&#16;&#140;bQ&#2;&#139;x&#20;U&#3;&sup3;&Ugrave;&plusmn;&#144;I&atilde;&Icirc;&middot;&Ccedil;&#16;&ograve;&sup1;&aacute;&ETH;l&#152;&#156;&#153;A&yacute;{&Yacute;&auml;O!}&acute;&#15;&Otilde;&Aacute;&#149;2&#16;&aring;&ocirc;&#19;&#143;&eth;&Otilde;{{&#16;y&eacute;&#21;L &ouml;&Atilde;&oslash;;&#131;&Egrave;&para;&iacute;&acute;&sup2;&oslash;&#158;R:&#5;&Aring;n&Ccedil;&eth;&#27;/&Atilde;&#157;L&Agrave;&iacute;tJ&#127;&acirc;&macr;8&cent;&pound;&#147;r&gt;&Oslash;\&#131;`]=&#1;R&#0;&Oacute;C&yacute;&lt;&oslash;&Igrave;F\xs;&#138;&plusmn;&#31;&Aacute;&#8;&Egrave;&#138;&lt;&#140;&otilde;&szlig;&#157;&#132;&Yacute;W&#129;&euml;&Ccedil;&#14;&aacute;&ouml;&ccedil;&#159;&Agrave;[&#21;&#128;&Ucirc;&iacute;&#129;Y,&#128;&#17;H&lt;&#30;&Ccedil;&#11;&#19;&#25;&#24;&#133;&#2;&#134;&#30;&#11;&quot;X_O&#15;&quot;@B&aring;&#133;D&#28;&Atilde;&#157;&Iacute;&uml;nl&#148;a&#23;r9&#20;k&Atilde;&Oslash;p&acirc;&#140;&#140;`x&times;&#14;&cedil;&AElig;~E&#6;*&ordf;&ordm;&para;&Agrave;&#18;)!&agrave;&para;/&#14;&Atilde;&cent;&frac14;&#143;&igrave;y&#7;&ordf;&times;/9R&cedil;&iexcl;&oacute;&Acirc;T&#2;&#23;6u&nbsp;2&#20;&#146;,&Ograve;-&Egrave;&auml;&ograve;h&yuml;&uacute;8&uuml;Mk&eth;{&iuml;g&Egrave;~u&#24;&lt;&sup2;<br/>&euml;O&#156;&frac12;&#159;&plusmn;2&#143;f&#25;&#6;Tb&Eacute;&nbsp;D&#155;&#6;1L&divide;&#136;&#155;&#152;e&quot;=rQ&#2;&ordf;.7,&#11;0s&sup3;&oslash;&oacute;&#155;&pound;0&ccedil;&aelig;&Agrave;KE&not;~&divide;}p:?&thorn;&aring;A&#137;!H&Ocirc;&#8;&#5;&#26;%&Uacute;&Otilde;&Uacute;&#129;&acirc;&micro;+4&#39;i&ETH;&#130;nrbW&#147;w&ccedil;&Ccedil;&yuml;&#0;h&reg;&#17;&Egrave;&auml;&Aacute;&yacute;2W&ugrave;TZ&#2;Z&ordm;&#142;&#27;&#7;&ordm;&aacute;}($&divide;2! &brvbar;&ordf;hx&otilde;5$&#147;I&egrave;&oacute;y&Igrave;S&#14;&#11;tQx&oacute;&#139;&#152;&#155;&#24;C&gt;v&#14;&Uuml;f&#3;&sect;&ETH;&#153;&Ccedil;&#3;&aelig;&otilde;B&otilde;&#137;&#136;@&Ntilde;&raquo;&egrave;&Ucirc;&#7;&#133;&uuml;&Acirc;&Ocirc;&times;&#27;#=&#129;&Ouml;u&eth;5=&#142;&Ecirc;&aelig;&micro;H&szlig;&frac12;&#11;&micro;n&#21;&Uacute;?&iacute;&#133;&#153;&Iuml;&aacute;&ccedil;&#29;[&eacute;&ugrave;&#22;,&brvbar;&Ecirc;j&#146;&amp;DM: t&lt;&#137;&Igrave;&yen;&#139;&#152;&#30;8&#5;fw&Agrave;&#18;,&#159;l&copy;&atilde;U&#27;&cent;h &iacute;9kj&aacute;Z&plusmn;&#18;&ouml;&ecirc;&#26;,&Igrave;L&aacute;&igrave;&para;MP&eacute;&nbsp;B&ugrave;&frac12;gP&curren;Bsz6C&#12;2&Oslash;&uuml;&#21;&sup2;R&auml;&#147;5&#18;m&ecirc;L?&Icirc;&iquest;&frac12;&#19;?l&icirc;Dz&ocirc;&#18;@&#137;&Icirc;&#141;&#143;!{&ucirc;&amp;L&Uacute;hp&euml;^&pound;&#136;E&curren;&Igrave;_ &aelig;&oacute;&Euml;&sup1;A&gt;a&#140;r&#15;F&micro;l#&#17;&laquo;&#14;&#7;RW&#127;&#147;&#128;+Z&Uacute;&deg;&aelig;&pound;&yacute;(.&#148;`P&acirc;M&#1;&ograve;?&amp;&Agrave;&Auml;:+&auml;&sup2;H&#39;&#18;0&Uuml;^Twn&#132;&gt;&#155;&#3;4b&times;2&ETH;&uuml;&Ouml;n&not;&yacute;&cedil;&#7;&oacute;T &brvbar;n&Egrave;&Uuml;=&Egrave;&not;&yen;5&aring;&#151;}{y&Euml;&reg;&Yacute;&ETH;I&ugrave;&Ugrave;k&ocirc;&Igrave;&#27;&atilde;(&#17;&Euml;&aacute;&ccedil;&raquo;&#16;x&acute;&#137;&Acirc;W&#145;&frac34;r&#25;&sect;vR-&ucirc;&#136;I&Ecirc;&Ugrave;&#131;&#134;,<br/>&yacute;&Icirc;-&#30;;&deg;&#15;&Oacute;&ccedil;&#6;&#23;&pound;&pound;&deg;9&#153;&#131;&#136;Y&divide;A7&Acirc;&Iuml;F&Aacute;&#28;N&#140;&ouml;&otilde;&acirc;&ecirc;&#145;&gt;&Oslash;I&auml;&Euml;&#141;&atilde;&iquest;&#128;&Otilde;&sect;&reg;&#143;&ouml;$&#134;&#6;&uml;M1&Ugrave;U&#20;&cent;_&iexcl;\&#154;T&macr;&#19;&Ccedil;&iquest;&#135;B:&shy;%Yy&Atilde;&#17;&Uuml;&#26;&egrave;&#151;:&aring;&cent;&#129;&#16;&#17;&aring;&amp;X&#23;&Iuml;f7&#135;&Iuml;CuRiQ&#15;&#17;&ugrave;&#17;Na&#156;&#158;j&macr;&#8;&agrave;&ograve;&#145;CH&Auml;~&#130;&#135;&uacute;&curren;F&ograve;0 &Ograve;d&Ugrave;&frac34;&Aring;&Uuml; &#134;&#23;5&Ecirc;&Uacute;&divide;|&#8;&#31;&otilde;2&Oacute;&ETH;e&#127;&#23;7-&#155;p0&#138;&#156;S&#141;&#151;&Ograve;TE&Ocirc;&iexcl;E&#158;&Ecirc;&divide;&#136;&ugrave;&#146;S&uacute;&yuml;&#1;&iquest;&egrave;U&Atilde;mn,%&#0;&#0;&#0;&#0;IEND&reg;B`&#130; https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8129.html Fri, 10 Nov 2017 19:53:13 +0000 Re: minor changes in a parser by jerry This kind of stuff is trivial in Perl. You&#39;ve chosen a good language.<br/><br/>my $url =~ s|/bar$||;<br/><br/>...Which means: &quot;Find any occurence of &quot;/bar&quot; at the very end of the URL <br/>and replace it<br/>with a nothing. This is called a &quot;regex&quot; ( short for &quot;regular <br/>expression&quot; ). We usually do regexes with forward slashes, but you can <br/>use other characters ( like &quot;|&quot; when the target string contains forward <br/>slashes.<br/><br/> Do a web search for &quot;Perl regex&quot;.<br/><br/> - Jerry Kaidor<br/><br/><br/><br/><br/><br/>On 11/10/2017 06:00, Martin Kaspar wrote:<br/>&gt; hello dear perl-experts,<br/>&gt; <br/>&gt; I&#39;m pretty new to Programming and OO programming especially.<br/>&gt; Nonetheless, I&#39;m trying to get done a very simple Spider for web<br/>&gt; crawling.<br/>&gt; <br/>&gt; the script below - is what i got to work<br/>&gt; <br/>&gt; it runs nicely : now i want to modify the script a bit - tailoring and<br/>&gt; tinkering is the way to learn. I want to fetch urls with a certain<br/>&gt; content in the URL-string<br/>&gt; <br/>&gt; &quot;http://www.foo.com/bar&quot;<br/>&gt; <br/>&gt; in other words: what is aimed, i need to fetch all the urls that<br/>&gt; contains the term &quot; /bar&quot;<br/>&gt; - then i want to extract the &quot;bar&quot; so that it remains the url:<br/>&gt; http://www.foo.com<br/>&gt; -<br/>&gt; <br/>&gt; is this doable?<br/>&gt; <br/>&gt; love to hear from you<br/>&gt; Martin<br/>&gt; <br/>&gt; #!C:\Perl\bin\perl<br/>&gt; <br/>&gt; use strict; # You always want to include both strict and warnings<br/>&gt; use warnings;<br/>&gt; <br/>&gt; use LWP::Simple;<br/>&gt; use LWP::UserAgent;<br/>&gt; use HTTP::Request;<br/>&gt; use HTTP::Response;<br/>&gt; use HTML::LinkExtor;<br/>&gt; <br/>&gt; # There was no reason for this to be in a BEGIN block (and there<br/>&gt; # are a few good reasons for it not to be)<br/>&gt; open my $file1,&quot;+&gt;&gt;&quot;, (&quot;links.txt&quot;);<br/>&gt; select($file1);<br/>&gt; <br/>&gt; #The Url I want it to start at;<br/>&gt; # Note that I&#39;ve made this an array, @urls, rather than a scalar, $URL<br/>&gt; <br/>&gt; my @urls = (&#39;https://the url goes in here&#39;);<br/>&gt; my %visited; # The % sigil indicates it&#39;s a hash<br/>&gt; my $browser = LWP::UserAgent-&gt;new();<br/>&gt; $browser-&gt;timeout(5);<br/>&gt; <br/>&gt; while (@urls) {<br/>&gt; my $url = shift @urls;<br/>&gt; <br/>&gt; # Skip this URL and go on to the next one if we&#39;ve<br/>&gt; # seen it before<br/>&gt; next if $visited{$url};<br/>&gt; <br/>&gt; my $request = HTTP::Request-&gt;new(GET =&gt; $url);<br/>&gt; my $response = $browser-&gt;request($request);<br/>&gt; <br/>&gt; # No real need to invoke printf if we&#39;re not doing<br/>&gt; # any formatting<br/>&gt; if ($response-&gt;is_error()) {print $response-&gt;status_line, &quot;\n&quot;;}<br/>&gt; my $contents = $response-&gt;content();<br/>&gt; <br/>&gt; # Now that we&#39;ve got the url&#39;s content, mark it as<br/>&gt; # visited<br/>&gt; $visited{$url} = 1;<br/>&gt; <br/>&gt; my ($page_parser) = HTML::LinkExtor-&gt;new(undef, $url);<br/>&gt; $page_parser-&gt;parse($contents)-&gt;eof;<br/>&gt; my @links = $page_parser-&gt;links;<br/>&gt; <br/>&gt; foreach my $link (@links) {<br/>&gt; print &quot;$$link[2]\n&quot;;<br/>&gt; push @urls, $$link[2];<br/>&gt; }<br/>&gt; sleep 60;<br/>&gt; }<br/>&gt; <br/>&gt; On Wed, Oct 4, 2017 at 10:49 PM, Dan Book &lt;grinnz@gmail.com&gt; wrote:<br/>&gt; <br/>&gt;&gt; How can we proceed from here?<br/>&gt;&gt; -Dan<br/>&gt;&gt; <br/>&gt;&gt; On Mon, Sep 18, 2017 at 1:17 PM, Patrick M. Galbraith<br/>&gt;&gt; &lt;patg@patg.net&gt; wrote:<br/>&gt;&gt; <br/>&gt;&gt; Pali,<br/>&gt;&gt; <br/>&gt;&gt; Great! Now we can start moving forward.<br/>&gt;&gt; <br/>&gt;&gt; Sorry if my responses have been intermittent - first week at new<br/>&gt;&gt; job.<br/>&gt;&gt; <br/>&gt;&gt; Regards,<br/>&gt;&gt; <br/>&gt;&gt; Patrick<br/>&gt;&gt; <br/>&gt;&gt; On 9/16/17 4:35 AM, pali@cpan.org wrote:<br/>&gt;&gt; <br/>&gt;&gt; I prepared branch master-new, which is based on current DBD-mysql<br/>&gt;&gt; master<br/>&gt;&gt; branch and revert state to pre-4.043 version, including all changes<br/>&gt;&gt; done<br/>&gt;&gt; after 4.043 release to master branch. I have this master-new branch<br/>&gt;&gt; in<br/>&gt;&gt; my fork. If you want you can use it...<br/>&gt;&gt; <br/>&gt;&gt; https://github.com/pali/DBD-mysql/tree/master-new [1]<br/>&gt; <br/>&gt; --<br/>&gt; <br/>&gt; [2] [3]<br/>&gt; <br/>&gt; Links:<br/>&gt; ------<br/>&gt; [1] https://github.com/pali/DBD-mysql/tree/master-new<br/>&gt; [2] http://www.facebook.com/martin.kaspar.547<br/>&gt; [3] https://plus.google.com/u/0/104428351748591530426<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8128.html Fri, 10 Nov 2017 19:53:07 +0000 Re: DBD::mysql next steps by Patrick M. Galbraith Darren,<br/><br/>Yes, the other plan definitely had its merits. Credit due to Michiel on <br/>that plan who really helped lay it out and explain the pros and cons of <br/>each and helped me to get that idea out.<br/><br/>Kind Regards,<br/><br/>Patrick<br/><br/>On 11/10/17 12:24 PM, Darren Duncan wrote:<br/>&gt; I agree in principle with Patrick&#39;s plan.<br/>&gt;<br/>&gt; My strong recommendation for continuing development under a different <br/>&gt; module name was based on the assumption (but not the knowledge) that <br/>&gt; the Unicode/Blob problems were rooted in the DBD::mysql codebase in <br/>&gt; such a way that they blocked the ability to use the newer <br/>&gt; MySQL/MariaDB/etc client libraries properly and that maintaining <br/>&gt; support for both behaviors user-configurable would be too difficult to <br/>&gt; do without making bugs worse.<br/>&gt;<br/>&gt; However, I also believe that Patrick&#39;s proposal (a single DBD::mysql <br/>&gt; under that name where the incompatible new behavior is toggled on a <br/>&gt; per-connection switch) is actually the best and most elegant solution <br/>&gt; for satisfying all parties under the assumption that there are savvy <br/>&gt; developers who fully understand the problem and are able and willing <br/>&gt; to support such a more-complicated codebase.<br/>&gt;<br/>&gt; -- Darren Duncan<br/>&gt;<br/>&gt; On 2017-11-10 7:13 AM, Patrick M. Galbraith wrote:<br/>&gt;&gt; Greetings all!<br/>&gt;&gt;<br/>&gt;&gt; Michiel and I have been talking, weighing options of what course to <br/>&gt;&gt; take in<br/>&gt;&gt; dealing with moving forward-- with the goal of both offering both <br/>&gt;&gt; stability<br/>&gt;&gt; and the choice to have the latest functionality and bug fixes as well as<br/>&gt;&gt; give contributors the opportunity to be part of overall improvements <br/>&gt;&gt; to the<br/>&gt;&gt; driver.<br/>&gt;&gt;<br/>&gt;&gt; What we are going to do is:<br/>&gt;&gt;<br/>&gt;&gt; Add to the connection the ability to turn on proper UTF handling with<br/>&gt;&gt; &#39;mysql_enable_proper_unicode&#39;. This gives the user volition the <br/>&gt;&gt; option to knowingly<br/>&gt;&gt; toggle whether they want the new functionality and understand that<br/>&gt;&gt; data structures returned or accepted by the driver might be different<br/>&gt;&gt; than without this setting.<br/>&gt;&gt;<br/>&gt;&gt; The other options had their merits, but we think this will solve the <br/>&gt;&gt; issue<br/>&gt;&gt; while keeping the driver unified and prevent divergence.<br/>&gt;&gt;<br/>&gt;&gt; Thank you for your input over the last couple months-- we look <br/>&gt;&gt; forward to<br/>&gt;&gt; moving ahead!<br/>&gt;&gt;<br/>&gt;&gt; Patrick and Michiel<br/>&gt;&gt;<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8127.html Fri, 10 Nov 2017 18:56:39 +0000 Re: DBD::mysql next steps by Darren Duncan I agree in principle with Patrick&#39;s plan.<br/><br/>My strong recommendation for continuing development under a different module <br/>name was based on the assumption (but not the knowledge) that the Unicode/Blob <br/>problems were rooted in the DBD::mysql codebase in such a way that they blocked <br/>the ability to use the newer MySQL/MariaDB/etc client libraries properly and <br/>that maintaining support for both behaviors user-configurable would be too <br/>difficult to do without making bugs worse.<br/><br/>However, I also believe that Patrick&#39;s proposal (a single DBD::mysql under that <br/>name where the incompatible new behavior is toggled on a per-connection switch) <br/>is actually the best and most elegant solution for satisfying all parties under <br/>the assumption that there are savvy developers who fully understand the problem <br/>and are able and willing to support such a more-complicated codebase.<br/><br/>-- Darren Duncan<br/><br/>On 2017-11-10 7:13 AM, Patrick M. Galbraith wrote:<br/>&gt; Greetings all!<br/>&gt;<br/>&gt; Michiel and I have been talking, weighing options of what course to take in<br/>&gt; dealing with moving forward-- with the goal of both offering both stability<br/>&gt; and the choice to have the latest functionality and bug fixes as well as<br/>&gt; give contributors the opportunity to be part of overall improvements to the<br/>&gt; driver.<br/>&gt;<br/>&gt; What we are going to do is:<br/>&gt;<br/>&gt; Add to the connection the ability to turn on proper UTF handling with<br/>&gt; &#39;mysql_enable_proper_unicode&#39;. This gives the user volition the option to knowingly<br/>&gt; toggle whether they want the new functionality and understand that<br/>&gt; data structures returned or accepted by the driver might be different<br/>&gt; than without this setting.<br/>&gt;<br/>&gt; The other options had their merits, but we think this will solve the issue<br/>&gt; while keeping the driver unified and prevent divergence.<br/>&gt;<br/>&gt; Thank you for your input over the last couple months-- we look forward to<br/>&gt; moving ahead!<br/>&gt;<br/>&gt; Patrick and Michiel<br/>&gt;<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8126.html Fri, 10 Nov 2017 17:24:44 +0000 DBD::mysql next steps by Patrick M. Galbraith Greetings all!<br/><br/>Michiel and I have been talking, weighing options of what course to take in<br/>dealing with moving forward-- with the goal of both offering both stability<br/>and the choice to have the latest functionality and bug fixes as well as<br/>give contributors the opportunity to be part of overall improvements to the<br/>driver.<br/><br/>What we are going to do is:<br/><br/>Add to the connection the ability to turn on proper UTF handling with<br/>&#39;mysql_enable_proper_unicode&#39;. This gives the user volition the option to knowingly<br/>toggle whether they want the new functionality and understand that<br/>data structures returned or accepted by the driver might be different<br/>than without this setting.<br/><br/>The other options had their merits, but we think this will solve the issue<br/>while keeping the driver unified and prevent divergence.<br/><br/>Thank you for your input over the last couple months-- we look forward to<br/>moving ahead!<br/><br/>Patrick and Michiel<br/><br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8125.html Fri, 10 Nov 2017 15:14:15 +0000 Re: DBD::mysql path forward by pali Technical: Reason why it does not help nor solve the main problem: Due<br/>to internals, DBD::mysql pseudo-randomly decide if input bind variable<br/>is encoded to UTF-8 or not. And do it independently of how is option<br/>enable_utf8 configured. As we know there are applications which misuse<br/>this internal. Calling perlapi functions may change this internal state<br/>under some circumstances and IIRC perl (itself) can change. Therefore it<br/>is not safe to touching code around. Fixing Unicode support needs to<br/>totally replace this code.<br/><br/>Night Light wrote that enable_utf8=0 does no encoding of input and this<br/>is not truth. It is also root of all problems and applications which<br/>misuse it.<br/><br/>Also we have no idea what else applications misuse. And because we were<br/>not already able to detect it in testing period, I really doubt that we<br/>would be able to do it in future...<br/><br/>On Friday 10 November 2017 09:34:19 pali@cpan.org wrote:<br/>&gt; I do not understand what (and how) your proposal with =2 solve. Probably<br/>&gt; nothing and people would again start after final release again<br/>&gt; complaining...<br/>&gt; <br/>&gt; On Friday 10 November 2017 09:24:35 Night Light wrote:<br/>&gt; &gt; Forking would take away my concerns (thank you all for suggesting that) but<br/>&gt; &gt; one thing pops into my mind:<br/>&gt; &gt; A decision to fork a mature module is not a light decision and should only<br/>&gt; &gt; be considered as a last resort (when ran out of all options).<br/>&gt; &gt; So far it is all based on the opinion of only one developer who finds it<br/>&gt; &gt; hard or probably impossible to implement.<br/>&gt; &gt; <br/>&gt; &gt; Can before forking at least a second person with C knowledge open dbdimp.c<br/>&gt; &gt; and measure if it is indeed unfeasible to make the module backward<br/>&gt; &gt; compatible by adding an option as suggested?<br/>&gt; &gt; <br/>&gt; &gt; Michiel, do you know C? Can you do that 5-minute check?<br/>&gt; &gt; <br/>&gt; &gt; The breaking is due to the fact that the prepared SQL statement and the<br/>&gt; &gt; bound parameters will be apart from what is returned from MySQL is encoded<br/>&gt; &gt; as well.<br/>&gt; &gt; The mysql_enable_utf8/mysql_enable_utf8mb4 flags already arrange logic to<br/>&gt; &gt; set if utf8 encoding/decoding should be done or not.<br/>&gt; &gt; I&#39;m not a C expert but this is what I find in dbdimp.c of 4.042 about it:<br/>&gt; &gt; <br/>&gt; &gt; dbd_st_prepare_sv function:<br/>&gt; &gt; bool enable_utf8 = (imp_dbh-&gt;enable_utf8 || imp_dbh-&gt;enable_utf8mb4);<br/>&gt; &gt; get_statement(aTHX_ statement_sv, enable_utf8, &amp;statement, &amp;statement_len);<br/>&gt; &gt; <br/>&gt; &gt; dbd_bind_ph function:<br/>&gt; &gt; bool enable_utf8 = (imp_dbh-&gt;enable_utf8 || imp_dbh-&gt;enable_utf8mb4);<br/>&gt; &gt; bind_param(&amp;imp_sth-&gt;params[idx], value, sql_type, idx+1, enable_utf8);<br/>&gt; &gt; <br/>&gt; &gt; <br/>&gt; &gt; Would changing that into the following make the module backwards compatible?<br/>&gt; &gt; bool enable_utf8 = (imp_dbh-&gt;enable_utf8 == 2 || imp_dbh-&gt;enable_utf8mb4 ==<br/>&gt; &gt; 2);<br/>&gt; &gt; <br/>&gt; &gt; enable_utf8 0 = no encoding of input and only decoding of output<br/>&gt; &gt; enable_utf8 1 = decode only output (4.043 mode)<br/>&gt; &gt; enable_utf8 2 = encoded input and decode output (4.042 mode)<br/>&gt; &gt; <br/>&gt; &gt; On Fri, Nov 10, 2017 at 8:37 AM, Darren Duncan &lt;darren@darrenduncan.net&gt;<br/>&gt; &gt; wrote:<br/>&gt; &gt; <br/>&gt; &gt; &gt; Michael, why can&#39;t you accept moving forward under a new module name? Why<br/>&gt; &gt; &gt; does it have to be under the old name? When people purposefully want to<br/>&gt; &gt; &gt; upgrade they purposefully choose the new module name in order to do so.<br/>&gt; &gt; &gt; What is the actual problem in that? -- Darren Duncan<br/>&gt; &gt; &gt;<br/>&gt; &gt; &gt; On 2017-11-09 10:59 PM, Michiel Beijen wrote:<br/>&gt; &gt; &gt;<br/>&gt; &gt; &gt;&gt; On Fri, Nov 10, 2017 at 7:16 AM, Darren Duncan &lt;darren@darrenduncan.net&gt;<br/>&gt; &gt; &gt;&gt; wrote:<br/>&gt; &gt; &gt;&gt;<br/>&gt; &gt; &gt;&gt;&gt; I agree with everything Dan said here. Its what I proposed, in fewer<br/>&gt; &gt; &gt;&gt;&gt; words.<br/>&gt; &gt; &gt;&gt;&gt; Do all new development under a new name, including all of Pali&#39;s work,<br/>&gt; &gt; &gt;&gt;&gt; and<br/>&gt; &gt; &gt;&gt;&gt; leave the current name for a product with no further effort applied to<br/>&gt; &gt; &gt;&gt;&gt; develop it. -- Darren Duncan<br/>&gt; &gt; &gt;&gt;&gt;<br/>&gt; &gt; &gt;&gt;<br/>&gt; &gt; &gt;&gt; This is NOT an option to me - it simply can&#39;t because the world moves<br/>&gt; &gt; &gt;&gt; forward and because of bitrot. The &#39;old&#39; version - the version that<br/>&gt; &gt; &gt;&gt; works for most people, the current version of DBD::mysql, the one<br/>&gt; &gt; &gt;&gt; which would then receive no more maintenance as it is no longer<br/>&gt; &gt; &gt;&gt; compiles with the latest version of libmysqlclient and it does not<br/>&gt; &gt; &gt;&gt; compile with libmariadb. This will only get worse in the future.<br/>&gt; &gt; &gt;&gt;<br/>&gt; &gt; &gt;&gt; I&#39;ll stick with my earlier proposal - I&#39;ll propose to go back to the<br/>&gt; &gt; &gt;&gt; *current* latest DBD::mysql release which does not break backcompat<br/>&gt; &gt; &gt;&gt; for our users; add the patches that we discarded when we rolled back<br/>&gt; &gt; &gt;&gt; one by one, such as testing on many different lib/db options, memory<br/>&gt; &gt; &gt;&gt; leaks and so on, and make a new release so we can be on the move<br/>&gt; &gt; &gt;&gt; again.<br/>&gt; &gt; &gt;&gt;<br/>&gt; &gt; &gt;&gt; If possible I&#39;d want to add back the *breaking* unicode changes that<br/>&gt; &gt; &gt;&gt; were introduced but they should be either in a separate namespace OR<br/>&gt; &gt; &gt;&gt; under a specific configuration option.<br/>&gt; &gt; &gt;&gt; Currently this whole thing has cost us loosing MONTHS of progress and<br/>&gt; &gt; &gt;&gt; then MONTHS of nothing and that is simply not good.<br/>&gt; &gt; &gt;&gt;<br/>&gt; &gt; &gt;&gt; Patrick: let me know if you&#39;re OK with this and then let&#39;s get start<br/>&gt; &gt; &gt;&gt; again!<br/>&gt; &gt; &gt;&gt;<br/>&gt; &gt; &gt;<br/> https://www.nntp.perl.org/group/perl.dbi.dev/2017/11/msg8124.html Fri, 10 Nov 2017 15:03:42 +0000