Hi Marius Adrian,
It's a ping as you was expressing the intention to maintain ext/interbase
https://wiki.php.net/rfc/removal_of_dead_sapis_and_exts#extinterbase . How
it currently looks like, ext/interbase is still broken and is not completely
ported to PHP7. Fe there's currently this bug ticket
https://bugs.php.net/bug.php?id=71175 and there are quite bad compilation
warnings (only a part linked here
https://gist.github.com/weltling/b352eace02ef5e1c4797 ). As Firebird is used
less often, it's not the case for a quick fix for everyone, as the test
environment is missing. Could you please go for the ext/interbase now? For
the particular bug and in general, it was only taken off from voting because
you've expressed the intention to maintain it
http://grokbase.com/p/php/php-internals/151mxve1t5/php-dev-rfc-discussion-re
moval-of-dead-sapis-and-extensions .
Thanks
Anatol
It's a ping as you was expressing the intention to maintain ext/interbase
https://wiki.php.net/rfc/removal_of_dead_sapis_and_exts#extinterbase . How
it currently looks like, ext/interbase is still broken and is not completely
ported to PHP7. Fe there's currently this bug ticket
https://bugs.php.net/bug.php?id=71175 and there are quite bad compilation
warnings (only a part linked here
https://gist.github.com/weltling/b352eace02ef5e1c4797 ). As Firebird is used
less often, it's not the case for a quick fix for everyone, as the test
environment is missing. Could you please go for the ext/interbase now? For
the particular bug and in general, it was only taken off from voting because
you've expressed the intention to maintain it
http://grokbase.com/p/php/php-internals/151mxve1t5/php-dev-rfc-discussion-re
moval-of-dead-sapis-and-extensions .
We need help from someone who understands exactly what has been changed :(
A lot of work was done back in February, but since then no further input
has been provided, and questions asked back then still remain
unanswered. We were I think waiting for the later PHP7 changes to be
implemented but since the extension continued to compile and work ...
For me the extension is working quite happily, but the latest
corrections posted by laruence are NOT compiling on my copy here so I
will go back to the working copy and check things out from there. I
think I am seeing the same warnings as #70071 on my working compile but
my warnings are 'comparison between pointer and integer' relating to
handles to the database links, and I am then getting lost.
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk
Hey:
On Mon, Dec 21, 2015 at 10:30 PM, Anatol Belski anatol.php@belski.net
wrote:
Hi Marius Adrian,
It's a ping as you was expressing the intention to maintain ext/interbase
https://wiki.php.net/rfc/removal_of_dead_sapis_and_exts#extinterbase . How
it currently looks like, ext/interbase is still broken and is not
completely
ported to PHP7. Fe there's currently this bug ticket
https://bugs.php.net/bug.php?id=71175 and there are quite bad compilation
warnings (only a part linked here
https://gist.github.com/weltling/b352eace02ef5e1c4797 ). As Firebird is
used
less often, it's not the case for a quick fix for everyone, as the test
environment is missing. Could you please go for the ext/interbase now? For
the particular bug and in general, it was only taken off from voting
because
you've expressed the intention to maintain ithttp://grokbase.com/p/php/php-internals/151mxve1t5/php-dev-rfc-discussion-re
moval-of-dead-sapis-and-extensions .Here is a working patch of mine:
https://gist.github.com/laruence/f3af903012902818d7da
if you are going to work on this, please base on this one.
since I still don't have setup a evn to test it... so there might be some
problems
thanks
Thanks
Anatol
--
--
Xinchen Hui
@Laruence
http://www.laruence.com/
Here is a working patch of mine:
https://gist.github.com/laruence/f3af903012902818d7daif you are going to work on this, please base on this one.
since I still don't have setup a evn to test it... so there might be some
problemsthanks
Xinchen ... working test setup should be easy, but I'm more than happy
to run tests, my problem is the core code has changed so drastically
since I ACTUALLY did bug fixes at that level that I don't understand the
current methods :(
I can understand this new patch but I still need to back-pedal the old
one. The only bit I'm not quite following is the switching between
'NULL' and '0' as I'm not sure why you are switching to '0' for some
handles that are currently 'NULL' and then defaulting others which were
'0' to 'NULL' but I think it's that we are now passing a pointer rather
than a handle?
I need to re-sync my repository with git.php.net then at least I can
track and roll back changes ... I'll get back to this later
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk
Hey：
Sent from my iPhone
Here is a working patch of mine:
https://gist.github.com/laruence/f3af903012902818d7daif you are going to work on this, please base on this one.
since I still don't have setup a evn to test it... so there might be some
problemsthanks
Xinchen ... working test setup should be easy, but I'm more than happy
to run tests, my problem is the core code has changed so drastically
since I ACTUALLY did bug fixes at that level that I don't understand the
current methods :(I can understand this new patch but I still need to back-pedal the old
one. The only bit I'm not quite following is the switching between
'NULL' and '0' as I'm not sure why you are switching to '0' for some
handles that are currently 'NULL' and then defaulting others which were
'0' to 'NULL' but I think it's that we are now passing a pointer rather
than a handle?
complier wanrning about handle are int
Anyway, this patch is not finish, I hope I can finish this whole work in this week
Will update here when it is done
Thanks
I need to re-sync my repository with git.php.net then at least I can
track and roll back changes ... I'll get back to this later--
Lester Caine - G8HFLContact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk
Hey:
Hey：
Sent from my iPhone
Here is a working patch of mine:
https://gist.github.com/laruence/f3af903012902818d7daif you are going to work on this, please base on this one.
since I still don't have setup a evn to test it... so there might be
some
problemsthanks
Xinchen ... working test setup should be easy, but I'm more than happy
to run tests, my problem is the core code has changed so drastically
since I ACTUALLY did bug fixes at that level that I don't understand the
current methods :(I can understand this new patch but I still need to back-pedal the old
one. The only bit I'm not quite following is the switching between
'NULL' and '0' as I'm not sure why you are switching to '0' for some
handles that are currently 'NULL' and then defaulting others which were
'0' to 'NULL' but I think it's that we are now passing a pointer rather
than a handle?complier wanrning about handle are int
Anyway, this patch is not finish, I hope I can finish this whole work in
this weekWill update here when it is done
Consider it done.
No compiler warnings anymore(expect the char* and ISC_UCHAR* mismatch)
However, there are some test fails, which I am not quite sure about.
Anyone who is faimilar with firebird, please help
https://gist.github.com/laruence/f3af903012902818d7da
thanks
Thanks
I need to re-sync my repository with git.php.net then at least I can
track and roll back changes ... I'll get back to this later--
Lester Caine - G8HFLContact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk--
--
Xinchen Hui
@Laruence
http://www.laruence.com/
Consider it done.
No compiler warnings anymore(expect the char* and ISC_UCHAR* mismatch)
Just on ibase_events.c which is async access to firebird/interbase This
is an area that needs more work with PHP anyway.
I'm seeing a change on github where 'efree(args);' elements have been
stripped from that file? Have compiled including them, but the patch
failed to install fully apparently due to the difference.
However, there are some test fails, which I am not quite sure about.
http://php7.lsces.org.uk/ibtest.php is running clean except for the
final drop table which is failing safe.
I'll work through the 5 make test failures later, but suspect it will
just be reply formats.
Anyone who is faimilar with firebird, please help
https://gist.github.com/laruence/f3af903012902818d7da
Thanks for the expert input!
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk
However, there are some test fails, which I am not quite sure about.
I'll work through the 5 make test failures later, but suspect it will
just be reply formats.
OK nothing particularly heavy on the failures. Actually it's mainly due
to the 'non-real-world' nature of the tests using combinations of
connections one should never use in practice. It is not normal practice
to create and delete new databases from any application when using
Firebird. One would normally be working with data stored long term, so
some of the problems are due to trying to delete a database that is
still being created asynchronously. The ibtest.php script works with one
of the existing demo databases so only relies on a stable link.
That said, I think there is a leak of some sort, or rather not clearing
down properly which is the error in both 004.phpt and 005.phpt
Both give an 'invalid statement handle' which did not make sense
initially since the same ibase_query() had been run before and after
without an error. The problem is where the same variable is used a
second time, after the first use has been cleared down. For example ...
004.phpt line 31 creates $q ... Line 60 reuses it without a problem, but
line 102 creates the 'invalid statement handle' warning. Changing that
handle from $q to $r moves the warning to line 112 which I've changed to
$s and the test now runs clean. ( line numbers for .php copy of test)
OK ... cross checking that it would seem that the problem is with
ibase_free_result() which I think should actually free the statement
handle? If I try an unset() before the next use, it's the unset which
gets the warning, then the actual query runs clean.
Any pointers where I should dig next? Will dig into the actual code now
I know where to start ... I think.
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk