You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@perl.apache.org by Gerald Richter <ri...@ecos.de> on 2001/09/25 10:28:42 UTC
make test fails with Apache::Const not found
Hi,
after a long time where I didn't had the time to look closer at mod_perl 2.0
I now fetched the newest CVS version and compiled it sucessfully. When I try
to run make test, I get the following error:
/usr/bin/perl5.6.1 -Iblib/arch/Apache2 -Iblib/lib/Apache2 \
t/TEST -clean
setting ulimit to allow core files
ulimit -c unlimited
exec t/TEST -clean
Can't locate Apache/Const.pm in @INC (@INC contains:
/usr/msrc/cvs/modperl-2.0/lib /usr/msrc/cvs/modperl-2.0/blib/lib
/usr/msrc/cvs/modperl-2.0/blib/arch Apache-Test/lib ./Apache-Test/lib
../Apache-Test/lib /opt/perlt5.6.1/lib/5.6.1/i686-linux-thread-multi
/opt/perlt5.6.1/lib/5.6.1
/opt/perlt5.6.1/lib/site_perl/5.6.1/i686-linux-thread-multi
/opt/perlt5.6.1/lib/site_perl/5.6.1 /opt/perlt5.6.1/lib/site_perl .) at
/usr/msrc/cvs/modperl-2.0/t/response/TestDirective/perlrequire.pm line 7.
As far as I can see the xs directory should also be in @INC to solve this
problem, but I didn't found the right place to set it. I guess the error
will also disapear once mod_perl is installed, but in this case the
installed version is used instead of the version from the build directory,
which is not the way it should work.
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Gerald Richter <ri...@ecos.de>.
>
> > These two things would need a extention of the xxx_types.map files. I
would
> > suggest simply append two addtional (optional) parameters for each line.
The
> > first holds the name used in the typemap file (%typemap) and the second
the
> > alias for the sv_convert.h file (%convert)
>
> all sounds good to me, do you want to work on the patch?
>
Yes, I do that. It will come together with the separation of the mod_perl
specific part and the generic part. (Will give a real huge patch, but the XS
generation stuff is really great and worth not only using it for mod_perl
:-)
Gerald
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Doug MacEachern <do...@covalent.net>.
On Sun, 7 Oct 2001, Gerald Richter wrote:
> I think we could easily remove this line and put the package name for these
> three functions in the map file. That would make much more sense to me. If
> you agree to that, I will do it and you don't have to take care about it.
already taken care of :)
> These two things would need a extention of the xxx_types.map files. I would
> suggest simply append two addtional (optional) parameters for each line. The
> first holds the name used in the typemap file (%typemap) and the second the
> alias for the sv_convert.h file (%convert)
all sounds good to me, do you want to work on the patch?
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Gerald Richter <ri...@ecos.de>.
>
> putting some trace statements in there, there's only 3 places where
> return_type is actually used:
> using Apache::Filter for ap_add_input_filter
> using Apache::RequestRec for ap_read_request
> using APR::IpSubnet for apr_ipsubnet_create
>
> i suppose we could probably get by without it, i'll work on getting rid of
> that line.
>
I think we could easily remove this line and put the package name for these
three functions in the map file. That would make much more sense to me. If
you agree to that, I will do it and you don't have to take care about it.
There is another thing that should going into the map files:
* TypeMap.pm:
#anything needed that mod_perl.h does not already include
#XXX: .maps should INCLUDE= these
my @includes = qw{
apr_uuid.h
apr_sha1.h
apr_md5.h
apr_base64.h
apr_getopt.h
apr_hash.h
apr_lib.h
apr_general.h
apr_signal.h
util_script.h
};
Your comment suggest it already...
* TypeMap.pm:
my %convert_alias = (
Apache__RequestRec => 'r',
Apache__Server => 'server',
Apache__Connection => 'connection',
APR__Table => 'table',
APR__UUID => 'uuid',
apr_status_t => 'status',
);
* WrapXS.pm
my %typemap = (
'Apache::RequestRec' => 'T_APACHEOBJ',
'apr_time_t' => 'T_APR_TIME',
'APR::Table' => 'T_HASHOBJ',
);
These two things would need a extention of the xxx_types.map files. I would
suggest simply append two addtional (optional) parameters for each line. The
first holds the name used in the typemap file (%typemap) and the second the
alias for the sv_convert.h file (%convert)
What do you think ?
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Doug MacEachern <do...@covalent.net>.
On Sun, 7 Oct 2001, Gerald Richter wrote:
> I do that anyway, because the first_class method contains other Apache
> specific things (like checking for pool). I have done this in many places to
> spearate generic code from mod_perl specific. The difference here is, that I
> don't understand under which circumstances it makes sense to take the return
> type instead of the first argument for guessing the target package of the
> function. At the moment I can't imagine a situation where this makes sense,
> so I failed to understand the reason why you did it. Because of that I can't
> decide how to handle this for the generic situation (mod_dav was only an
> example), because the generic code should of course behave in a way to
> catches most cases and we override for the specfic ones (of course). If you
> have a example function where this takeing of return type makes sense, it
> would help me very much.
putting some trace statements in there, there's only 3 places where
return_type is actually used:
using Apache::Filter for ap_add_input_filter
using Apache::RequestRec for ap_read_request
using APR::IpSubnet for apr_ipsubnet_create
i suppose we could probably get by without it, i'll work on getting rid of
that line.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Gerald Richter <ri...@ecos.de>.
>
> > sub first_class {
>
> couldn't you just override this method to make it do what works best for
> dav?
>
I do that anyway, because the first_class method contains other Apache
specific things (like checking for pool). I have done this in many places to
spearate generic code from mod_perl specific. The difference here is, that I
don't understand under which circumstances it makes sense to take the return
type instead of the first argument for guessing the target package of the
function. At the moment I can't imagine a situation where this makes sense,
so I failed to understand the reason why you did it. Because of that I can't
decide how to handle this for the generic situation (mod_dav was only an
example), because the generic code should of course behave in a way to
catches most cases and we override for the specfic ones (of course). If you
have a example function where this takeing of return type makes sense, it
would help me very much.
Thanks
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Doug MacEachern <do...@covalent.net>.
On Sat, 6 Oct 2001, Gerald Richter wrote:
> sub first_class {
couldn't you just override this method to make it do what works best for
dav?
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Gerald Richter <ri...@ecos.de>.
>
> if it makes things easier to run through cpp first, by all means do.
I will rethink what's the best solution....
Another question. I have gotten very far in separting the mod_perl specific
code and the generic code. So I tried to run what I have so far over mod_dav
1.x and the result is really nice (also still a lot of work until I can use
it), but I found that when I don't specify the package where to put
functions in, WrapXS makes really strange guesses. It mostly takes the
return type, which doesn't make much sense to me. (For example function in
mod_dav often return a dav_error struct, so all these functions will go into
the Apache::DAV::Error package). The source is the function
*::TypeMap::first_class:
sub first_class {
my($self, $func) = @_;
return $func->{return_type} if $func->{return_type} =~ /::/;
for my $e (@{ $func->{args} }) {
next unless $e->{type} =~ /::/;
#there are alot of util functions that take an APR::Pool
#that do not belong in the APR::Pool class
next if $e->{type} eq 'APR::Pool' and $func->{name} !~ /^apr_pool/;
return $e->{type};
}
return $func->{name} =~ /^apr_/ ? 'APR' : 'Apache';
}
which returns the return_type if it is an object. From my point of view the
return ... if in line 4 should be removed, but I guess it is/was good for
something ?
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Doug MacEachern <do...@covalent.net>.
On Fri, 5 Oct 2001, Gerald Richter wrote:
> I've done so, but it's hard work to get all the details right, so it's still
> under construction, also the results already comes very close to the tables
> in the CVS.
great!!
> There is one issue I like to get some feedback before continuing. C::Scan
> has used the C preprocessor, thatway it didn't had to care about C-macro
> definitions. For example there is a macro called AP_DECLARE_HOOK, which
> declares a ap_hook_foo and a ap_run_foo. My current aproach is to only use
> Parse::RecDescent and not the C preprocessor, that makes it independend of
> installed compilers etc., but I have to find out all these macros and
> reimplement them in my RecDescent grammer, to get the correct result. Also
> there are some parts inside of #if and the current aproach ignores #if at
> all. As far as I see the second point isn't so important, but the first
> maybe, because any change in these macros inside the Apache sources must
> also be reflected in the RecDescent grammer.
if it makes things easier to run through cpp first, by all means do.
we will still ship a copy of the generated tables with the distribution,
running source_scan will be optional.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Gerald Richter <ri...@ecos.de>.
>
> I have played along this line a few hours, taken the grammer from
Inline::C
> and Inline::Struct and now have a nearly working replacement for C::Scan.
I
> think this is the right way to go, since the RecDecent parser grammer is
> much easier to extent and modify (and it works on all platforms, also it
> will be slower than C::Scan).
>
> I continue to work on this during the next days and try to integrate it
into
> Apache::ParseSource, by making a abstract base class and
Apache::ParseSource
> will contain only Apache spcific things.
>
I've done so, but it's hard work to get all the details right, so it's still
under construction, also the results already comes very close to the tables
in the CVS.
There is one issue I like to get some feedback before continuing. C::Scan
has used the C preprocessor, thatway it didn't had to care about C-macro
definitions. For example there is a macro called AP_DECLARE_HOOK, which
declares a ap_hook_foo and a ap_run_foo. My current aproach is to only use
Parse::RecDescent and not the C preprocessor, that makes it independend of
installed compilers etc., but I have to find out all these macros and
reimplement them in my RecDescent grammer, to get the correct result. Also
there are some parts inside of #if and the current aproach ignores #if at
all. As far as I see the second point isn't so important, but the first
maybe, because any change in these macros inside the Apache sources must
also be reflected in the RecDescent grammer.
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Doug MacEachern <do...@covalent.net>.
On Thu, 27 Sep 2001, Gerald Richter wrote:
> I have played along this line a few hours, taken the grammer from Inline::C
> and Inline::Struct and now have a nearly working replacement for C::Scan. I
> think this is the right way to go, since the RecDecent parser grammer is
> much easier to extent and modify (and it works on all platforms, also it
> will be slower than C::Scan).
fantastic news gerald!! not worried if it is slower than C::Scan.
> I continue to work on this during the next days and try to integrate it into
> Apache::ParseSource, by making a abstract base class and Apache::ParseSource
> will contain only Apache spcific things.
sounds good.
> I come back to the list when it is working and we can see how the best way
> is to get this into the current modperl developement.
looking forward to it!
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Gerald Richter <ri...@ecos.de>.
Doug,
> gerald, i would be happy to see the source parser usable outside of
> modperl. first i would like to solve the C::Scan problem. it just isn't
> usable on every system.
Agreed
> i think one option would be to ditch C::Scan and
> see if Parse::RecDecent can be used in place of it. i looked at
> Inline.pm, they have an Inline::C::grammar that should be a good
> start. looks like it should be usable for generating the
> Function/Structure tables in the same format they are currently in.
> we would not depend on Inline.pm, but rather borrow C::grammar.pm and turn
> it into what we need.
> anybody interested in taking a shot at this?
>
I have played along this line a few hours, taken the grammer from Inline::C
and Inline::Struct and now have a nearly working replacement for C::Scan. I
think this is the right way to go, since the RecDecent parser grammer is
much easier to extent and modify (and it works on all platforms, also it
will be slower than C::Scan).
I continue to work on this during the next days and try to integrate it into
Apache::ParseSource, by making a abstract base class and Apache::ParseSource
will contain only Apache spcific things.
I come back to the list when it is working and we can see how the best way
is to get this into the current modperl developement.
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by brian moseley <bc...@maz.org>.
you gotta figure that *::gramar.pm are generally useful
(think of lxr) and deserve their own namespace and
packaging.
On Wed, 26 Sep 2001, Doug MacEachern wrote:
> gerald, i would be happy to see the source parser usable outside of
> modperl. first i would like to solve the C::Scan problem. it just isn't
> usable on every system. i think one option would be to ditch C::Scan and
> see if Parse::RecDecent can be used in place of it. i looked at
> Inline.pm, they have an Inline::C::grammar that should be a good
> start. looks like it should be usable for generating the
> Function/Structure tables in the same format they are currently in.
> we would not depend on Inline.pm, but rather borrow C::grammar.pm and turn
> it into what we need.
> anybody interested in taking a shot at this?
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
> For additional commands, e-mail: dev-help@perl.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: ParseSource etc.
Posted by Doug MacEachern <do...@covalent.net>.
gerald, i would be happy to see the source parser usable outside of
modperl. first i would like to solve the C::Scan problem. it just isn't
usable on every system. i think one option would be to ditch C::Scan and
see if Parse::RecDecent can be used in place of it. i looked at
Inline.pm, they have an Inline::C::grammar that should be a good
start. looks like it should be usable for generating the
Function/Structure tables in the same format they are currently in.
we would not depend on Inline.pm, but rather borrow C::grammar.pm and turn
it into what we need.
anybody interested in taking a shot at this?
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
ParseSource etc.
Posted by Gerald Richter <ri...@ecos.de>.
Hi,
some time ago we discussed to make ParseSource and the other XS generation
modules more generic. My goal is to also use them outside of mod_perl. At
the moment I like to create a perl wrapper for mod_dav 1.x and don't like to
code anything by hand. I also expect using this method will make porting to
mod_dav/_perl 2.0 much more easier.
I have started by dividing Apache/ParseSource.pm into two modules. A
ExtUtils::XSBuilder::ParseSource and the Apache::ParseSource which inherits
from the first. The Apache::ParseSource only need to override some methods
to supply any needed informations, like include dirs, prefixes etc.
I append what I written so far. perldoc ParseSource.pm gives a list of
overrideable methods. I just written this down, without any testing, so
don't expect it to work out of the box. Before I continue to work on that, I
want to have a feedback, if you think this is the right way to go.
If yes, Doug, how should we handle the changes. Should I send my changes to
the list and you review and commit them or should I directly commit them (of
course I only commit after testing and made sure it doesn't break anything
as far as I can see)
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
Re: make test fails with Apache::Const not found
Posted by Gerald Richter <ri...@ecos.de>.
> On Tue, 25 Sep 2001, Gerald Richter wrote:
>
> > MP_INST_APACHE2=1 ...
>
> that could be the problem, haven't tested in a while.
Yes, without that setting the INST_LIB is set correctly, however make test
fails:
/usr/bin/perl5.6.1 -Iblib/arch -Iblib/lib \
t/TEST -clean
setting ulimit to allow core files
ulimit -c unlimited
exec t/TEST -clean
APACHE_USER= APXS= APACHE_PORT= APACHE_GROUP= APACHE= \
/usr/bin/perl5.6.1 -Iblib/arch -Iblib/lib \
t/TEST
setting ulimit to allow core files
ulimit -c unlimited
exec t/TEST
/opt/apache2/bin/httpd -d /usr/msrc/cvs/modperl-2.0/t -f
/usr/msrc/cvs/modperl-2.0/t/conf/httpd.conf -DAPACHE2 -DPERL_USEITHREADS
using Apache/2.0.26-dev (threaded MPM)
waiting for server to warm up...ok
server localhost:8529 started
apache/cgihandler...FAILED tests 1-2
Failed 2/2 tests, 0.00% okay
apache/compat.......FAILED tests 1-2
Failed 2/2 tests, 0.00% okay
apache/constants....ok
apache/post.........FAILED test 2
Failed 1/2 tests, 50.00% okay
apache/read.........FAILED test 1
Failed 1/1 tests, 0.00% okay
apache/scanhdrs.....NOK 2Use of uninitialized value in string eq at
apache/scanhdrs.t line 18.
apache/scanhdrs.....dubious
Test returned status 9 (wstat 2304, 0x900)
DIED. FAILED tests 1-4
Failed 4/4 tests, 0.00% okay
api/access..........FAILED before any test output arrived
api/aplog...........FAILED before any test output arrived
api/conn_rec........FAILED before any test output arrived
api/lookup_uri......FAILED before any test output arrived
api/module..........FAILED before any test output arrived
api/r_subclass......FAILED before any test output arrived
api/request_rec.....FAILED before any test output arrived
api/response........FAILED before any test output arrived
api/rutil...........FAILED before any test output arrived
api/send_fd.........FAILED test 2
Failed 1/3 tests, 66.67% okay
api/sendfile........FAILED test 2
Failed 1/3 tests, 66.67% okay
api/server_rec......FAILED before any test output arrived
api/uri.............FAILED before any test output arrived
apr/constants.......ok
directive/perlrequire.Use of uninitialized value in concatenation (.) or
string at directive/perlrequire.t line 25.
directive/perlrequire.dubious
Test returned status 255 (wstat 65280, 0xff00)
DIED. FAILED tests 1-2
Failed 2/2 tests, 0.00% okay
filter/input_body...FAILED tests 1-2
Failed 2/2 tests, 0.00% okay
filter/input_msg....Use of uninitialized value in concatenation (.) or
string at filter/input_msg.t line 11.
Use of uninitialized value in concatenation (.) or string at
/usr/msrc/cvs/modperl-2.0/blib/lib/Apache/TestRequest.pm line 101.
filter/input_msg....dubious
Test returned status 29 (wstat 7424, 0x1d00)
filter/lc...........FAILED test 1
Failed 1/1 tests, 0.00% okay
hooks/access........FAILED tests 2-3
Failed 2/4 tests, 50.00% okay
hooks/authen........FAILED tests 2-3
Failed 2/4 tests, 50.00% okay
hooks/authz.........FAILED tests 2-3
Failed 2/4 tests, 50.00% okay
hooks/trans.........FAILED tests 2-3
Failed 2/3 tests, 33.33% okay
modules/cgi.........NOK 5Use of uninitialized value in string eq at
modules/cgi.t line 26.
modules/cgi.........dubious
Test returned status 9 (wstat 2304, 0x900)
DIED. FAILED tests 2-6
Failed 5/6 tests, 16.67% okay
modules/cgiupload...FAILED tests 1-2
Failed 2/2 tests, 0.00% okay
protocol/echo.......NOK 1Can't use an undefined value as a symbol reference
at protocol/echo.t line 17.
protocol/echo.......dubious
Test returned status 22 (wstat 5632, 0x1600)
DIED. FAILED tests 1-3
Failed 3/3 tests, 0.00% okay
protocol/echo_filter.NOK 1Can't use an undefined value as a symbol reference
at protocol/echo_filter.t line 17.
protocol/echo_filter.dubious
Test returned status 22 (wstat 5632, 0x1600)
DIED. FAILED tests 1-3
Failed 3/3 tests, 0.00% okay
protocol/eliza......skipped test on this platform
Failed Test Status Wstat Total Fail Failed List of Failed
----------------------------------------------------------------------------
----
apache/cgihandler.t 2 2 100.00% 1-2
apache/compat.t 2 2 100.00% 1-2
apache/post.t 2 1 50.00% 2
apache/read.t 1 1 100.00% 1
apache/scanhdrs.t 9 2304 4 4 100.00% 1-4
api/access.t ?? ?? % ??
api/aplog.t ?? ?? % ??
api/conn_rec.t ?? ?? % ??
api/lookup_uri.t ?? ?? % ??
api/module.t ?? ?? % ??
api/r_subclass.t ?? ?? % ??
api/request_rec.t ?? ?? % ??
api/response.t ?? ?? % ??
api/rutil.t ?? ?? % ??
api/send_fd.t 3 1 33.33% 2
api/sendfile.t 3 1 33.33% 2
api/server_rec.t ?? ?? % ??
api/uri.t ?? ?? % ??
directive/perlrequire.t 255 65280 2 2 100.00% 1-2
filter/input_body.t 2 2 100.00% 1-2
filter/input_msg.t 29 7424 ?? ?? % ??
filter/lc.t 1 1 100.00% 1
hooks/access.t 4 2 50.00% 2-3
hooks/authen.t 4 2 50.00% 2-3
hooks/authz.t 4 2 50.00% 2-3
hooks/trans.t 3 2 66.67% 2-3
modules/cgi.t 9 2304 6 5 83.33% 2-6
modules/cgiupload.t 2 2 100.00% 1-2
protocol/echo.t 22 5632 3 3 100.00% 1-3
protocol/echo_filter.t 22 5632 3 3 100.00% 1-3
1 test skipped.
server localhost:8529 shutdown
error running tests (please examine t/logs/error_log)
Failed 30/33 test scripts, 9.09% okay. 38/67 subtests failed, 43.28% okay.
and the error log contains
[Tue Sep 25 19:57:38 2001] [notice] Apache/2.0.26-dev (Unix)
mod_perl/1.99_01-dev Perl/v5.6.1 configured -- resuming normal operations
[Tue Sep 25 19:57:38 2001] [info] Server built: Sep 25 2001 09:10:37
[Tue Sep 25 19:57:41 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestApache::cgihandler
[Tue Sep 25 19:57:41 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestApache::cgihandler
[Tue Sep 25 19:57:44 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestApache::compat
[Tue Sep 25 19:57:44 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestApache::compat
[Tue Sep 25 19:57:47 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestApache::post
[Tue Sep 25 19:57:47 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestApache::post
[Tue Sep 25 19:57:50 2001] [error] server reached MaxClients setting,
consider raising the MaxClients setting
[Tue Sep 25 19:57:52 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestApache::read
[Tue Sep 25 19:57:54 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestApache::scanhdrs
[Tue Sep 25 19:57:57 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestAPI::access
[Tue Sep 25 19:58:00 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestAPI::aplog
[Tue Sep 25 19:58:03 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestAPI::conn_rec
[Tue Sep 25 19:58:07 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestAPI::lookup_uri
[Tue Sep 25 19:58:11 2001] [error] [client 127.0.0.1] File does not exist:
/usr/msrc/cvs/modperl-2.0/t/htdocs/TestAPI::module
This looks to me like still something with @INC is wrong....
If anybody has a quick idea how to fix this or where to look for the problem
let me know. Otherwise I try to investigate myself.
>
> no, t/conf/mime.types should either be inherited from the httpd.conf found
> by t/TEST (using apxs) or it should be generated by t/TEST.
>
I have retried, no t/conf/mime.types is generated and Apache doesn't start
without copying that file into t/conf.
Gerald
P.S. Here is perl -V if this helps:
Summary of my perl5 (revision 5.0 version 6 subversion 1) configuration:
Platform:
osname=linux, osvers=2.4.2, archname=i686-linux-thread-multi
uname='linux venus 2.4.2 #2 thu mar 22 14:15:00 cet 2001 i686 unknown '
config_args='-des -Uinstallusrbinperl -Dusethreads -Dprefix=/opt/perlt5.6.1'
hint=recommended, useposix=true, d_sigaction=define
usethreads=define use5005threads=undef useithreads=define
usemultiplicity=define
useperlio=undef d_sfio=undef uselargefiles=define usesocks=undef
use64bitint=undef use64bitall=undef uselongdouble=undef
Compiler:
cc='cc', ccflags
='-D_REENTRANT -fno-strict-aliasing -I/usr/local/include -D_LARGEFILE_SOURCE
-D_FILE_OFFSET_BITS=64',
optimize='-O2',
cppflags='-D_REENTRANT -fno-strict-aliasing -I/usr/local/include'
ccversion='', gccversion='2.95.2 19991024 (release)', gccosandvers=''
intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=12
ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t',
lseeksize=8
alignbytes=4, usemymalloc=n, prototype=define
Linker and Libraries:
ld='cc', ldflags =' -L/usr/local/lib'
libpth=/usr/local/lib /lib /usr/lib
libs=-lnsl -lndbm -lgdbm -ldl -lm -lpthread -lc -lcrypt -lutil
perllibs=-lnsl -ldl -lm -lpthread -lc -lcrypt -lutil
libc=, so=so, useshrplib=false, libperl=libperl.a
Dynamic Linking:
dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-rdynamic'
cccdlflags='-fpic', lddlflags='-shared -L/usr/local/lib'
Characteristics of this binary (from libperl):
Compile-time options: MULTIPLICITY USE_ITHREADS USE_LARGE_FILES
PERL_IMPLICIT_CONTEXT
Built under linux
Compiled at Apr 23 2001 06:50:06
%ENV:
PERL5DB="BEGIN { require Devel::ptkdb }"
@INC:
/opt/perlt5.6.1/lib/5.6.1/i686-linux-thread-multi
/opt/perlt5.6.1/lib/5.6.1
/opt/perlt5.6.1/lib/site_perl/5.6.1/i686-linux-thread-multi
/opt/perlt5.6.1/lib/site_perl/5.6.1
/opt/perlt5.6.1/lib/site_perl
.
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Gerald Richter <ri...@ecos.de>.
>
> > threaded
>
> does threaded work without modperl on your system?
I can start the same Apache binary with the httpd.conf that comes with
Apache and serve static pages.
> in any case, i would
> try perfork first.
>
I just done what's in pod/modperl_dev.pod :-)
ok, I try prefork first
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Doug MacEachern <do...@covalent.net>.
On Wed, 26 Sep 2001, Gerald Richter wrote:
> threaded
does threaded work without modperl on your system? in any case, i would
try perfork first.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Gerald Richter <ri...@ecos.de>.
>
> is this with prefork or threaded mpm?
>
threaded
>
>
> PerlTrace in the config file.
>
ok, that works now. I have tested t/TEST -start and what I see is, the
server is starting, but not listeing to any port. (I don't see it in
netstat -a, but I see the running server in ps ). Also the last lines of
error_log (with PerlTrace all) are
modperl_interp_destroy: interp == 0xxxxxx
Is it correct that the interpreters are destroyed or show this that apache
is shutting down, but hangs somewhere during shutdown ?
Gerald
P.S. localhost resolves correctly on that machine
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Stas Bekman <st...@stason.org>.
> But test still doesn't do much....
>
> /usr/bin/perl5.6.1 -Iblib/arch -Iblib/lib \
> t/TEST -clean
> setting ulimit to allow core files
> ulimit -c unlimited
> exec t/TEST -clean
> APACHE_USER= APXS= APACHE_PORT= APACHE_GROUP= APACHE= \
> /usr/bin/perl5.6.1 -Iblib/arch -Iblib/lib \
> t/TEST
> setting ulimit to allow core files
> ulimit -c unlimited
> exec t/TEST
> /opt/apache2/bin/httpd -d /usr/msrc/cvs/modperl-2.0/t -f
> /usr/msrc/cvs/modperl-2.0/t/conf/httpd.conf -DAPACHE2 -DPERL_USEITHREADS
> using Apache/2.0.26-dev (threaded MPM)
> waiting for server to warm up............giving up
> server failed to start! (please examine t/logs/error_log)
>
> venus:/usr/msrc/cvs/modperl-2.0 #cat t/logs/error_log
> [Wed Sep 26 06:12:29 2001] [info] 15 Apache:: modules loaded
> [Wed Sep 26 06:12:29 2001] [info] 5 APR:: modules loaded
> [Wed Sep 26 06:12:30 2001] [info] 15 Apache:: modules loaded
> [Wed Sep 26 06:12:30 2001] [info] 5 APR:: modules loaded
is this with prefork or threaded mpm?
> I have tried with t/TEST -verbose and by setting export MOD_PERL_TRACE=all
> (MP_TRACE=1 is set for Makefile.PL) , but no more information.
MOD_PERL_TRACE is not supported yet.
http://marc.theaimsgroup.com/?l=apache-modperl-dev&m=99506486528266&w=2
> For the test case, is it enough to set MOD_PERL_TRACE on the command line,
> or do I need it to place in some of the config files ?
PerlTrace in the config file.
_____________________________________________________________________
Stas Bekman JAm_pH -- Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide http://perl.apache.org/guide
mailto:stas@stason.org http://apachetoday.com http://eXtropia.com/
http://singlesheaven.com http://perl.apache.org http://perlmonth.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Gerald Richter <ri...@ecos.de>.
> gerald, can you cvs up and try again? i think the latest patch will fix
> the problem for you.
>
Yes, it fixes the mime.types problem.
But test still doesn't do much....
/usr/bin/perl5.6.1 -Iblib/arch -Iblib/lib \
t/TEST -clean
setting ulimit to allow core files
ulimit -c unlimited
exec t/TEST -clean
APACHE_USER= APXS= APACHE_PORT= APACHE_GROUP= APACHE= \
/usr/bin/perl5.6.1 -Iblib/arch -Iblib/lib \
t/TEST
setting ulimit to allow core files
ulimit -c unlimited
exec t/TEST
/opt/apache2/bin/httpd -d /usr/msrc/cvs/modperl-2.0/t -f
/usr/msrc/cvs/modperl-2.0/t/conf/httpd.conf -DAPACHE2 -DPERL_USEITHREADS
using Apache/2.0.26-dev (threaded MPM)
waiting for server to warm up............giving up
server failed to start! (please examine t/logs/error_log)
venus:/usr/msrc/cvs/modperl-2.0 #cat t/logs/error_log
[Wed Sep 26 06:12:29 2001] [info] 15 Apache:: modules loaded
[Wed Sep 26 06:12:29 2001] [info] 5 APR:: modules loaded
[Wed Sep 26 06:12:30 2001] [info] 15 Apache:: modules loaded
[Wed Sep 26 06:12:30 2001] [info] 5 APR:: modules loaded
I have tried with t/TEST -verbose and by setting export MOD_PERL_TRACE=all
(MP_TRACE=1 is set for Makefile.PL) , but no more information.
For the test case, is it enough to set MOD_PERL_TRACE on the command line,
or do I need it to place in some of the config files ?
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Doug MacEachern <do...@covalent.net>.
gerald, can you cvs up and try again? i think the latest patch will fix
the problem for you.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Stas Bekman <st...@stason.org>.
Doug MacEachern wrote:
> i also saw the same mime.types problem, triggered by the failed require of
> perlrequire.pm, you'll need to rm t/conf/*.conf
basically the problem occurs for failures during configure,
and the system doesn't have a way to check the the whole configuration is in place.
I was thinking about creating some checkpoint to make sure that config files
are always complete.
is there still a problem with perlrequire? i.e. APACHE_TEST_CONFIGURE?
_____________________________________________________________________
Stas Bekman JAm_pH -- Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide http://perl.apache.org/guide
mailto:stas@stason.org http://apachetoday.com http://eXtropia.com/
http://singlesheaven.com http://perl.apache.org http://perlmonth.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Doug MacEachern <do...@covalent.net>.
i also saw the same mime.types problem, triggered by the failed require of
perlrequire.pm, you'll need to rm t/conf/*.conf
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Doug MacEachern <do...@covalent.net>.
On Tue, 25 Sep 2001, Gerald Richter wrote:
> MP_INST_APACHE2=1 ...
that could be the problem, haven't tested in a while. i've also never
'make install'ed modperl-2.0, i just
'use blib qw(/home/dougm/ap/modperl-2.0);'
> after changeing the to INST_LIB = blib/lib the first test works, but httpd
> isn't starting because t/conf/mime.types are missing. I have copied this
> over from the Apache distribution and now httpd starts (we should add this
> file to the cvs, should I commit it ?) but more or less every test fails.
no, t/conf/mime.types should either be inherited from the httpd.conf found
by t/TEST (using apxs) or it should be generated by t/TEST.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Gerald Richter <ri...@ecos.de>.
>
>
> Apache2? Your @INC misses blib/lib, looks like a broken MakeMaker,
It's the MakeMaker that comes with perl5.6.1
> since
> blib/lib is always there:
>
> from my Makefile:
>
> INST_LIB = blib/lib
> TEST_LIBS = -I$(INST_ARCHLIB) -I$(INST_LIB)
>
> which means that your Makefile has
>
> INST_LIB = blib/lib/Apache2
>
Yes, that exactly what's in my Makefile and after you pointed me to that I
guess reason is that I have used:
Makefile.PL MP_INST_APACHE2=1 ...
after changeing the to INST_LIB = blib/lib the first test works, but httpd
isn't starting because t/conf/mime.types are missing. I have copied this
over from the Apache distribution and now httpd starts (we should add this
file to the cvs, should I commit it ?) but more or less every test fails.
I try to rebuild without MP_INST_APACHE2=1 and see if this changes
anything...
Gerald
-------------------------------------------------------------
Gerald Richter ecos electronic communication services gmbh
Internetconnect * Webserver/-design/-datenbanken * Consulting
Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
E-Mail: richter@ecos.de Voice: +49 6133 925131
WWW: http://www.ecos.de Fax: +49 6133 925152
-------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org
Re: make test fails with Apache::Const not found
Posted by Stas Bekman <st...@stason.org>.
Gerald Richter wrote:
> Hi,
>
> after a long time where I didn't had the time to look closer at mod_perl 2.0
> I now fetched the newest CVS version and compiled it sucessfully. When I try
> to run make test, I get the following error:
>
> /usr/bin/perl5.6.1 -Iblib/arch/Apache2 -Iblib/lib/Apache2 \
Apache2? Your @INC misses blib/lib, looks like a broken MakeMaker, since
blib/lib is always there:
from my Makefile:
INST_LIB = blib/lib
TEST_LIBS = -I$(INST_ARCHLIB) -I$(INST_LIB)
which means that your Makefile has
INST_LIB = blib/lib/Apache2
I've just tested a build with 5.6.1 on linux, and it'd correct. What do
you see?
> t/TEST -clean
> setting ulimit to allow core files
> ulimit -c unlimited
> exec t/TEST -clean
> Can't locate Apache/Const.pm in @INC (@INC contains:
> /usr/msrc/cvs/modperl-2.0/lib /usr/msrc/cvs/modperl-2.0/blib/lib
> /usr/msrc/cvs/modperl-2.0/blib/arch Apache-Test/lib ./Apache-Test/lib
> ../Apache-Test/lib /opt/perlt5.6.1/lib/5.6.1/i686-linux-thread-multi
> /opt/perlt5.6.1/lib/5.6.1
> /opt/perlt5.6.1/lib/site_perl/5.6.1/i686-linux-thread-multi
> /opt/perlt5.6.1/lib/site_perl/5.6.1 /opt/perlt5.6.1/lib/site_perl .) at
> /usr/msrc/cvs/modperl-2.0/t/response/TestDirective/perlrequire.pm line 7.
>
> As far as I can see the xs directory should also be in @INC to solve this
> problem, but I didn't found the right place to set it. I guess the error
> will also disapear once mod_perl is installed, but in this case the
> installed version is used instead of the version from the build directory,
> which is not the way it should work.
see above, it's not xs :)
> Gerald
>
> -------------------------------------------------------------
> Gerald Richter ecos electronic communication services gmbh
> Internetconnect * Webserver/-design/-datenbanken * Consulting
>
> Post: Tulpenstrasse 5 D-55276 Dienheim b. Mainz
> E-Mail: richter@ecos.de Voice: +49 6133 925131
> WWW: http://www.ecos.de Fax: +49 6133 925152
> -------------------------------------------------------------
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
> For additional commands, e-mail: dev-help@perl.apache.org
>
--
_____________________________________________________________________
Stas Bekman JAm_pH -- Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide http://perl.apache.org/guide
mailto:stas@stason.org http://apachetoday.com http://eXtropia.com/
http://singlesheaven.com http://perl.apache.org http://perlmonth.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org