viewvc access broken?

viewvc access broken?

[Per Bothner]

I have various links to:
   http://sourceware.org/viewvc/kawa/
most prominently from:
   http://www.gnu.org/software/kawa/Getting-Kawa.html
I also have some links from proposals on srfi.schemers.org.

Currently, the viewvc URL is broken.  Any chance this
would be revived?
-- 
	--Per Bothner
per@bothner.com   http://per.bothner.com/

Re: viewvc access broken?

[Frank Ch. Eigler]

Hi -

> I have various links to:
>   http://sourceware.org/viewvc/kawa/
> most prominently from:
>   http://www.gnu.org/software/kawa/Getting-Kawa.html
> I also have some links from proposals on srfi.schemers.org.
> Currently, the viewvc URL is broken.  Any chance this
> would be revived?

It's back via changes to /etc/viewvc.conf, and invoked via WSGI rather
than CGI.  One drawback is that the sourceware & gcc repos appear
together in a single list, for I haven't figured out how to give them
separate config files yet.

- FChE

Re: viewvc access broken?

[Per Bothner]

On 03/23/2013 05:57 AM, Frank Ch. Eigler wrote:
> Hi -
>
>> I have various links to:
>>    http://sourceware.org/viewvc/kawa/

> It's back via changes to /etc/viewvc.conf, and invoked via WSGI rather
> than CGI.  One drawback is that the sourceware & gcc repos appear
> together in a single list, for I haven't figured out how to give them
> separate config files yet.

http://sourceware.org/viewvc/kawa/ is currently showing a
six-year-old snapshot.  That's pretty useless.

Also, the links are different.  Before, I could use a link:
http://sourceware.org/viewvc/kawa/trunk/gnu/expr/ApplyExp.java?view=co
http://sourceware.org/viewvc/kawa/kawa/gnu/expr/ApplyExp.java

Also, there is no way to view the raw (down-loadable) file.

Now I realize if I'm the only one who cares, it is unreasonable
to expect much effort.  It might be time to migrate to Mercurial
and a different host (most likely Savannah) at the same time.
However, if it's an easy fix, I'll put off the migration a
bit longer ...
-- 
	--Per Bothner
per@bothner.com   http://per.bothner.com/

Re: viewvc access broken?

[Frank Ch. Eigler]

Hi -

> http://sourceware.org/viewvc/kawa/ is currently showing a
> six-year-old snapshot.  That's pretty useless.

Sorry about that.  viewvc was sicced onto the old cvs repo 
instead of the svn one.  This is corrected.


> [...] It might be time to migrate to Mercurial and a different host
> (most likely Savannah) at the same time.  However, if it's an easy
> fix, I'll put off the migration a bit longer ...

You are of course welcome to migrate if/when you like.

- FChE

Re: viewvc access broken?

[Per Bothner]

On 03/23/2013 10:06 AM, Frank Ch. Eigler wrote:
> Hi -
>
>> http://sourceware.org/viewvc/kawa/ is currently showing a
>> six-year-old snapshot.  That's pretty useless.
>
> Sorry about that.  viewvc was sicced onto the old cvs repo
> instead of the svn one.  This is corrected.

Thanks!  One more request/suggestion:  Could "checkout view"
be re-enabled?  I used to be able to go to:
http://sourceware.org/viewvc/kawa/trunk/testsuite/bad-srfi-109.scm?view=co
which now yields:

     An Exception Has Occurred
     Checkout view is disabled
     HTTP Response Status
     403 Forbidden

Leaving off the "?view=co" does work:
http://sourceware.org/viewvc/kawa/trunk/testsuite/bad-srfi-109.scm
-- 
	--Per Bothner
per@bothner.com   http://per.bothner.com/

Re: viewvc access broken?

[Per Bothner]

On 03/23/2013 02:19 PM, Frank Ch. Eigler wrote:
> Hi -
>
>> Thanks!  One more request/suggestion:  Could "checkout view"
>> be re-enabled?  [...]
>
> Please try now.

Thanks!

One "feature" I noticed that's a little half-baked.  I'll phrase
it in the context of gcc, to make it non-Kawa-specific:

If you go to:
http://gcc.gnu.org/viewcvs/gcc/trunk/libcpp/errors.c
you see the head and various revisions have "as text" links,
addition to the "download" link.  They have the same effect,
except (it appears) "as text" forces a "text/plain" mime-type.

OTOH, if you go to
http://gcc.gnu.org/viewcvs/gcc/trunk/libcpp/Makefile.in
there are no "as text" links, just the "download" link.

I assume this has something to do with file extensions
and inferred mime types, but it seems like a weird "feature"
to make this distinction in this semi-random manner.  No
biggie - the "download" link is functional enough.
-- 
	--Per Bothner
per@bothner.com   http://per.bothner.com/