Re: svn commit: r1747056 - in /httpd/httpd/branches/2.4.x: ./ STATUS modules/mappers/mod_rewrite.c

[William A Rowe Jr]

On Tue, Jun 7, 2016 at 1:23 AM, Marion & Christophe JAILLET <
christophe.jail...@wanadoo.fr> wrote:

> Le 07/06/2016 à 06:59, William A Rowe Jr a écrit :
>
> On Mon, Jun 6, 2016 at 3:10 PM, Marion & Christophe JAILLET <
> christophe.jail...@wanadoo.fr> wrote:
>
>> ap_casecmpstr breaks 2.4.x build.
>> Keep strncasecmp for now? (waiting for ap_cstr_casecmp[n])
>>
>> Le 06/06/2016 à 21:11, j...@apache.org a écrit :
>>
>>> Author: jim
>>> Date: Mon Jun  6 19:11:20 2016
>>> New Revision: 1747056
>>>
>>> URL: http://svn.apache.org/viewvc?rev=1747056=rev
>>> Log:
>>> Merge r1744206 from trunk:
>>>
>>> mod_rewrite: adding h2:// and h2c:// proxy schemes to absolute uri
>>> detection, patch by Evgeny Kotkov
>>> Submitted by: icing
>>> Reviewed/backported by: jim
>>>
>>> Modified:
>>>  httpd/httpd/branches/2.4.x/   (props changed)
>>>  httpd/httpd/branches/2.4.x/STATUS
>>>  httpd/httpd/branches/2.4.x/modules/mappers/mod_rewrite.c
>>>
>>>
>>> Modified: httpd/httpd/branches/2.4.x/STATUS
>>> URL:
>>> http://svn.apache.org/viewvc/httpd/httpd/branches/2.4.x/STATUS?rev=1747056=1747055=1747056=diff
>>>
>>> ==
>>> --- httpd/httpd/branches/2.4.x/STATUS (original)
>>> +++ httpd/httpd/branches/2.4.x/STATUS Mon Jun  6 19:11:20 2016
>>> @@ -115,12 +115,6 @@ PATCHES ACCEPTED TO BACKPORT FROM TRUNK:
>>> [ start all new proposals below, under PATCHES PROPOSED. ]
>>> - *) mod_rewrite: support new h2:// and h2c:// proxy url schemes in
>>> detection
>>> - of absolute urls:
>>> - trunk patch: http://svn.apache.org/r1744206
>>> - 2.4.x patch: trunk works
>>> - +1: icing, jim, covener
>>>
>>
> Well, contrary to the commit message that failed to credit to the two
> other
> reviewers, it did have 3 +1's... we are pretty specific on backport policy.
>
> Is that a vote against, jallettc? If not, the backport remains valid.
>
> No, it's not a vote against.
> It is just a remark that we should move forward and either backport
> ap_cstr_casecmp[n] or decide to delay it to another release.
>
> We are building 2.4.21, so breaking compilation for a few days is, IMHO,
> acceptable.
>

Agreed, I don't like a broken trunk or branch state, but it isn't
a major crisis.


> I will have limited time until end of the week to update my
> ap_cstr_casecmp[n] backport proposal.
> So if anyone wants to update it before, do not hesitate.
>

I'm happy to do so, but my preference is to favor the apr_cstr_casecmp[n]
as I detailed in my (too many times edited) original reply to the proposal.

I'll work this up in the coming hour, no worries about your schedule.

Cheers,

Bill

Re: dbmmanage

[Rich Bowen]

On 06/07/2016 08:31 AM, Rich Bowen wrote:
> 
> 
>>
>>> dbmmanage has some functionality that is lacking in htdbm - in
>>> particular, the ability to import a plain text htpasswd style password
>>> file. However, the script httxt2dbm fills that need, which is presumably
>>> (usually) a one-time thing, rather than ongoing maintenance, so that's ok.
>>
>> Right, we ship that too.  I don't recall dropping dbmmanage being 
>> controversial.
> 
> I don't know that its controversial, but we do want reality reflected in
> the docs.
> 
> To clarify, are you saying that this was just a RHEL/Fedora change, or
> do you know if it's more widespread than that? Trying to find a
> Debian/Ubuntu machine to verify ...


ok, Ubuntu does have dbmmanage, so perhaps we just need to have the docs
note that it's not available on all platforms, etc.

Now, I just need to figure out how to reproduce the the `import`
functionality without dbmmanage.

Thanks for the info, Joe.

--Rich


-- 
Rich Bowen - rbo...@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon



signature.asc
Description: OpenPGP digital signature

Re: dbmmanage

[Rich Bowen]

On 06/07/2016 08:21 AM, Joe Orton wrote:
> On Tue, Jun 07, 2016 at 06:39:46AM -0400, Rich Bowen wrote:
>> In troubleshooting something with dbmmanage, I came across this:
>>
>> http://marc.info/?l=fedora-extras-commits=137148193030744=2
>>
>> I'm sure there's more context here that I haven't unearthed yet, but
>> does anyone (Joe?) happen to remember what the comment "zombie
>> dbmmanage" means here? Is dbmmanage deprecated in some way that we
>> should reflect in the docs?
> 
> With ht*dbm we know for certain the tool can use exactly the same set of 
> DB libraries as mod_auth*_dbm - the same was never true for dbmmanage, 
> which used whatever was available to Perl.  I suspect dbmmanage was the 
> last thing creating a dependency from the httpd package to Perl as well.
> 
> We dropped dbmmanage from our packages a very long time ago, I think 
> possibly even when upgrading from 1.3 to 2.0.  RHEL3's httpd 2.0 
> packages don't have dbmmanage, from a quick check.
> 
> When updating to 2.4 a bunch of scripts moved from sbindir to bindir, 
> which fooled the spec file, dbmmanage came back to life and was 
> unintenionally - and briefly - shipped in the Fedora RPMs again.  Hence 
> the zombie killing comment!
> 
>> dbmmanage has some functionality that is lacking in htdbm - in
>> particular, the ability to import a plain text htpasswd style password
>> file. However, the script httxt2dbm fills that need, which is presumably
>> (usually) a one-time thing, rather than ongoing maintenance, so that's ok.
> 
> Right, we ship that too.  I don't recall dropping dbmmanage being 
> controversial.

I don't know that its controversial, but we do want reality reflected in
the docs.

To clarify, are you saying that this was just a RHEL/Fedora change, or
do you know if it's more widespread than that? Trying to find a
Debian/Ubuntu machine to verify ...



-- 
Rich Bowen - rbo...@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon



signature.asc
Description: OpenPGP digital signature

Re: dbmmanage

[Joe Orton]

On Tue, Jun 07, 2016 at 06:39:46AM -0400, Rich Bowen wrote:
> In troubleshooting something with dbmmanage, I came across this:
> 
> http://marc.info/?l=fedora-extras-commits=137148193030744=2
> 
> I'm sure there's more context here that I haven't unearthed yet, but
> does anyone (Joe?) happen to remember what the comment "zombie
> dbmmanage" means here? Is dbmmanage deprecated in some way that we
> should reflect in the docs?

With ht*dbm we know for certain the tool can use exactly the same set of 
DB libraries as mod_auth*_dbm - the same was never true for dbmmanage, 
which used whatever was available to Perl.  I suspect dbmmanage was the 
last thing creating a dependency from the httpd package to Perl as well.

We dropped dbmmanage from our packages a very long time ago, I think 
possibly even when upgrading from 1.3 to 2.0.  RHEL3's httpd 2.0 
packages don't have dbmmanage, from a quick check.

When updating to 2.4 a bunch of scripts moved from sbindir to bindir, 
which fooled the spec file, dbmmanage came back to life and was 
unintenionally - and briefly - shipped in the Fedora RPMs again.  Hence 
the zombie killing comment!

> dbmmanage has some functionality that is lacking in htdbm - in
> particular, the ability to import a plain text htpasswd style password
> file. However, the script httxt2dbm fills that need, which is presumably
> (usually) a one-time thing, rather than ongoing maintenance, so that's ok.

Right, we ship that too.  I don't recall dropping dbmmanage being 
controversial.

Regards, Joe

Re: dbmmanage

[Rich Bowen]

On 06/07/2016 06:39 AM, Rich Bowen wrote:
> In troubleshooting something with dbmmanage, I came across this:
> 
> http://marc.info/?l=fedora-extras-commits=137148193030744=2
> 
> I'm sure there's more context here that I haven't unearthed yet, but
> does anyone (Joe?) happen to remember what the comment "zombie
> dbmmanage" means here? Is dbmmanage deprecated in some way that we
> should reflect in the docs?
> 
> dbmmanage has some functionality that is lacking in htdbm - in
> particular, the ability to import a plain text htpasswd style password
> file. However, the script httxt2dbm fills that need, which is presumably
> (usually) a one-time thing, rather than ongoing maintenance, so that's ok.

Hmm. I take that back. httxt2dbm doesn't handle this case - it's for
files formatted for rewritemap. So converting text password files to dbm
isn't something we have a (simple) solution for other than dbmmanage.


-- 
Rich Bowen - rbo...@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon

dbmmanage

[Rich Bowen]

In troubleshooting something with dbmmanage, I came across this:

http://marc.info/?l=fedora-extras-commits=137148193030744=2

I'm sure there's more context here that I haven't unearthed yet, but
does anyone (Joe?) happen to remember what the comment "zombie
dbmmanage" means here? Is dbmmanage deprecated in some way that we
should reflect in the docs?

dbmmanage has some functionality that is lacking in htdbm - in
particular, the ability to import a plain text htpasswd style password
file. However, the script httxt2dbm fills that need, which is presumably
(usually) a one-time thing, rather than ongoing maintenance, so that's ok.

--Rich

Re: mod_macro: Control on bad nesting

[Nick Gearls]

Good point.
A global option is not a good option as it would disable this very 
useful check for all macros.


What about " wrote:


What syntax would be the best one? maybe a (one character) /option after the
Macro keyword?
Ex: "" (and later "
Kinda ugly and probably breaks the core from recognizing "

Re: svn commit: r1747056 - in /httpd/httpd/branches/2.4.x: ./ STATUS modules/mappers/mod_rewrite.c

[Marion & Christophe JAILLET]

Le 07/06/2016 à 06:59, William A Rowe Jr a écrit :
On Mon, Jun 6, 2016 at 3:10 PM, Marion & Christophe JAILLET 
> 
wrote:


ap_casecmpstr breaks 2.4.x build.
Keep strncasecmp for now? (waiting for ap_cstr_casecmp[n])

Le 06/06/2016 à 21:11, j...@apache.org  a
écrit :

Author: jim
Date: Mon Jun  6 19:11:20 2016
New Revision: 1747056

URL: http://svn.apache.org/viewvc?rev=1747056=rev
Log:
Merge r1744206 from trunk:

mod_rewrite: adding h2:// and h2c:// proxy schemes to absolute
uri detection, patch by Evgeny Kotkov
Submitted by: icing
Reviewed/backported by: jim

Modified:
 httpd/httpd/branches/2.4.x/   (props changed)
 httpd/httpd/branches/2.4.x/STATUS
 httpd/httpd/branches/2.4.x/modules/mappers/mod_rewrite.c


Modified: httpd/httpd/branches/2.4.x/STATUS
URL:

http://svn.apache.org/viewvc/httpd/httpd/branches/2.4.x/STATUS?rev=1747056=1747055=1747056=diff

==
--- httpd/httpd/branches/2.4.x/STATUS (original)
+++ httpd/httpd/branches/2.4.x/STATUS Mon Jun  6 19:11:20 2016
@@ -115,12 +115,6 @@ PATCHES ACCEPTED TO BACKPORT FROM TRUNK:
[ start all new proposals below, under PATCHES PROPOSED. ]
- *) mod_rewrite: support new h2:// and h2c:// proxy url
schemes in detection
- of absolute urls:
- trunk patch: http://svn.apache.org/r1744206
- 2.4.x patch: trunk works
- +1: icing, jim, covener


Well, contrary to the commit message that failed to credit to the two 
other
reviewers, it did have 3 +1's... we are pretty specific on backport 
policy.


Is that a vote against, jallettc? If not, the backport remains valid.



No, it's not a vote against.
It is just a remark that we should move forward and either backport 
ap_cstr_casecmp[n] or decide to delay it to another release.


We are building 2.4.21, so breaking compilation for a few days is, IMHO, 
acceptable.



I will have limited time until end of the week to update my 
ap_cstr_casecmp[n] backport proposal.

So if anyone wants to update it before, do not hesitate.

CJ