On Mon, 1 Dec 2003, Stas Bekman wrote:
> [...]
>
> I'm still unhappy about whatever_url() working with fs
> paths. I guess I'm taking my words on using Unix-> back.
> At least we know that we work with paths and not urls. How
> about this:
>
> # concat a dir/file using unix path separators
> # no
>
> BTW, why Finfo doesn't use uid/gid but user/group? because of the C data
> struct names? It's probably more intuitive to rename them to be uid/gid,
> since they return the ids and not names.
>
I don't think we have typically been renaming struct slots but rather
carrying them directly over
Stas Bekman wrote:
[...]
Cool, I've missed that. It just looked weird. I guess they use 2*32-1
for nobody, so they can accomodate 2*32-2 users ;)
s/2\*32/2**32/ of course
__
Stas BekmanJAm_pH --> Just Another mod_perl
Geoffrey Young wrote:
I just tested and it seems to work fine for me under 'root'. Notice that
all files under t are chowned to the user/group the server is running
under (not-root) before the tests are run.
yes, for me too, but on linux. did you test any OSX variant, though?
richard, are you o
> I just tested and it seems to work fine for me under 'root'. Notice that
> all files under t are chowned to the user/group the server is running
> under (not-root) before the tests are run.
yes, for me too, but on linux. did you test any OSX variant, though?
richard, are you on darwin or pant
Stas Bekman wrote:
Geoffrey Young wrote:
geoff,
fair enuf!
(yeah, yeah -- RTFM ;-), altho for the life of me i can't find where it
says run "make test" as non-root ...)
it might not in the docs. doing only install as root is considered by
many
to be a best practice. I'll try to remember to a
Geoffrey Young wrote:
geoff,
fair enuf!
(yeah, yeah -- RTFM ;-), altho for the life of me i can't find where it
says run "make test" as non-root ...)
it might not in the docs. doing only install as root is considered by many
to be a best practice. I'll try to remember to add something in the
a
> geoff,
>
> fair enuf!
> (yeah, yeah -- RTFM ;-), altho for the life of me i can't find where it
> says run "make test" as non-root ...)
it might not in the docs. doing only install as root is considered by many
to be a best practice. I'll try to remember to add something in the
appropriate p
% t/TEST t/apr/finfo.t -v
*** setting ulimit to allow core files
ulimit -c unlimited; t/TEST 't/apr/finfo.t' -v
*** root mode: changing the files ownership to 'nobody'
(4294967294:4294967294)
are you doing all this as root? that could be part of the problem.
please 't/TEST t/apr/finfo
> % t/TEST t/apr/finfo.t -v
> *** setting ulimit to allow core files
> ulimit -c unlimited; t/TEST 't/apr/finfo.t' -v
> *** root mode: changing the files ownership to 'nobody'
> (4294967294:4294967294)
are you doing all this as root? that could be part of the problem.
please 't/TEST
geoff,
after staring at the results i'd just sent you, i wanted to take a look into the httpd.conf's that each test used ...
specifically, in the case of:
(a) httpd-2.0.48
/usr/local/httpd2048/sbin/httpd -d /Volumes/develop/ports/modperl2048/t -f
/Volumes/develop/ports/modperl2048/
;-)
i'll report back with results ...
geoff,
ok, here we go ... (i've included all the following info for completeness/posterity's
sake ...)
bottom line: the results of " t/TEST t/apr/finfo.t -v" for
(a) httpd-2.0.48+modperl
(b) httpd-2.1.0 (dev) +modperl
are, in fact, different
Geoffrey Young wrote:
It'd be better to commit this separately, as it has nothing to do with
the rest of the commit.
well, the goal for me was to get things working with 2.1. before commit was
not working, after commit was.
it's ok if it was committed by mistake, but if you simply don't care t
># testing : $r->finfo->user()
># expected: 4294967294
># received: -2
>not ok 7
># testing : $r->finfo->group()
># expected: 4294967294
># received: -2
>not ok 8
ack!
this is apache 2.1, right?
if you're interested in helping OSX
>> +
>> +
>> +# 2.1
>> +
>> +http://@servername@:@port@/*>
>> +Order Deny,Allow
>> +Deny from all
>> +Allow from @servername@
>> +
>> +ProxyRequests Off
>> +
OpenMacNews wrote:
hi geoff,
i can verify that mod_perl compiles (thanks!) and runs w/ Apache2.10
(dev) on MacOSX 10.2.8 & 10.3.1:
Apache/2.1.0-dev (Unix) mod_perl/1.99_12-dev Perl/v5.8.2 DAV/2
PHP/4.3.4
great! thanks again!
as fyi -- "make test" gives:
Failed Test Stat Wstat Total Fail
[EMAIL PROTECTED] wrote:
geoff 2003/12/02 07:49:22
Modified:t/filter/TestFilter both_str_req_proxy.pm
t/response/TestAPI rutil.pm
xs/maps apr_functions.map
xs/tables/current/Apache ConstantsTable.pm FunctionTable.pm
Log:
remove deprecat
OpenMacNews wrote:
> hi geoff,
>
> i can verify that mod_perl compiles (thanks!) and runs w/ Apache2.10
> (dev) on MacOSX 10.2.8 & 10.3.1:
>
>Apache/2.1.0-dev (Unix) mod_perl/1.99_12-dev Perl/v5.8.2 DAV/2
> PHP/4.3.4
>
> great! thanks again!
>
> as fyi -- "make test" gives:
>
> Fai
hi geoff,
i can verify that mod_perl compiles (thanks!) and runs w/ Apache2.10 (dev) on MacOSX 10.2.8 & 10.3.1:
Apache/2.1.0-dev (Unix) mod_perl/1.99_12-dev Perl/v5.8.2 DAV/2 PHP/4.3.4
great! thanks again!
as fyi -- "make test" gives:
Failed Test Stat Wstat Total Fail Failed List of
>>> 2.1 has dropped quite a few deprecated features. If we can do that
>>> without any special #ifdef (i.e. 2.0.46 and higher support this API),
>>> then I see no harm in adjusting it. Patches are very welcome.
>>
>>
>> I'll look into this and fix it if possible. hopefully there's no reason
>> t
Barbara, you have a strange mail client, it keeps on adding RE: to the
subject, and breaking the thread.
Hmm, I guess that explains it:
X-Mailer: Microsoft Outlook, Build 10.0.2627
Here is what I get. Maybe the problem comes from my user in httpd.conf
not being "nobody" ? (but I doubt it could be
Hi Stas,
Here is what I get. Maybe the problem comes from my user in httpd.conf
not being "nobody" ? (but I doubt it could be so).
su-2.05b# killall httpd
su-2.05b# t/TEST -start
*** root mode: changing the files ownership to 'nobody' (65534:65534)
*** /usr/bin/perl -e ' require POSIX; PO
22 matches
Mail list logo
