incubator-connectors-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Karl Wright <>
Subject Re: Release?
Date Fri, 03 Dec 2010 03:02:44 GMT
OK, so I will do the appropriate things to make (1), (4), and maybe
(5) happen.  Does anyone want to help with (2), (3), and (8)?

On Thu, Dec 2, 2010 at 9:59 PM, Grant Ingersoll <> wrote:
> On Dec 2, 2010, at 9:54 PM, Karl Wright wrote:
>> Hi Grant,
>> In offline conversation you clarified that for (1) you are looking for
>> the top level dir in the zip/tar to be named "apache-manifoldcf-0.1".
>> You also seem to be asking for a number of other fixes that are
>> specific to a release, that I presume would NOT be in sources on trunk
>> (e.g. CHANGES.txt).  Are you envisioning that we make these specific
>> changes in the release branch only?
> It's perfectly fine for CHANGES.txt to be on trunk.  You make the change marking it
as 0.1.  Once the release is out, you add a new section at the top for trunk again.
> Later, as we mature, we will likely have branches, etc. for this stuff, but for now let's
just assume trunk is under code freeze and the only changes that can be made are those related
to release.
>> Karl
>> On Thu, Dec 2, 2010 at 9:45 PM, Grant Ingersoll <> wrote:
>>> We're close, but I think we've got a few more things to do.  I did get it to
>>> Notes:
>>> 1. We should package the stuff all under apache-manifold-0.1 so that when we
unzip it's all in one folder.
>>> 2. Many of the libs require an entry in the NOTICE.txt file
>>> 3.  All licenses for those libs need to be appended on to the end of the LICENSE.txt
file (See Solr's for instance)
>>> 4. The CHANGES.txt file should reflect that it is a release and not trunk (not
critical to fix)
>>> 5. Is there anyway to make the package smaller?  Maybe we don't need to ship
both PDF and HTML for the docs.  Anything else we can trim?
>>> 6. What's json/org/json all about?
>>> 7. I still see Memex stuff in connectors dir.  I didn't check other places.
>>> 8. We should hook in RAT (see Solr's build file) to verify that all source files
have appropriate license headers
>>> Other than that, some other eyes on it would be good.
>>> -Grant
>>> On Dec 2, 2010, at 8:51 PM, Karl Wright wrote:
>>>> Done
>>>> Karl
>>>> On Thu, Dec 2, 2010 at 8:49 PM, Karl Wright <> wrote:
>>>>> ok - I might move it there
>>>>> Karl
>>>>> On Thu, Dec 2, 2010 at 8:31 PM, Grant Ingersoll <>
>>>>>> Weird, ~kwright doesn't resolve for me on people.a.o, but I can get
to /x1/home/kwright
>>>>>> FWIW, if you have a public_html directory in your directory and then
place the files there, everyone can download them and check them out at
>>>>>> -Grant
>>>>>> On Nov 23, 2010, at 1:00 PM, Karl Wright wrote:
>>>>>>> While I was looking for a solution, an upload attempt succeeded!
>>>>>>> So there is now an RC0 out on
>>>>>>> [kwright@minotaur:~]$ ls -lt manifoldcf-0.1.*
>>>>>>> -rw-r--r--  1 kwright  kwright         63 Nov 23 17:57
>>>>>>> -rw-r--r--  1 kwright  kwright         60 Nov 23 17:57
>>>>>>> -rw-r--r--  1 kwright  kwright  158734230 Nov 23 17:55
>>>>>>> -rw-r--r--  1 kwright  kwright  156742315 Nov 23 17:06 manifoldcf-0.1.tar.gz
>>>>>>> [kwright@minotaur:~]$
>>>>>>> Please let me know what you think.
>>>>>>> Karl
>>>>>>> On Tue, Nov 23, 2010 at 11:25 AM, Karl Wright <>
>>>>>>>> The upload has failed repeatedly for me, so I'll clearly
have to find
>>>>>>>> another way.
>>>>>>>> Karl
>>>>>>>> On Tue, Nov 23, 2010 at 10:47 AM, Karl Wright <>
>>>>>>>>> I'm uploading a release candidate now.  But someone
needs to feed the
>>>>>>>>> hamsters turning the wheels or something, because the
upload speed to
>>>>>>>>> that machine is 51KB/sec, so it's going to take 3 hours
to get the
>>>>>>>>> candidate up there, if my network connection doesn't
bounce in the
>>>>>>>>> interim.  Is there any other place available?
>>>>>>>>> Karl
>>>>>>>>> On Fri, Nov 19, 2010 at 8:34 AM, Grant Ingersoll <>
>>>>>>>>>> On Nov 19, 2010, at 6:18 AM, Karl Wright wrote:
>>>>>>>>>>> I've created a signing key, and checked in a
KEYS file.  Apache
>>>>>>>>>>> instructions for this are actually decent, so
I didn't have to make
>>>>>>>>>>> much stuff up.  Glad about that.
>>>>>>>>>> Yep, sorry, have been in meetings.
>>>>>>>>>>> Last remaining release issue is getting the release
files to a
>>>>>>>>>>> download mirror.  Maybe I can find some doc
for that too.
>>>>>>>>>> Next steps would be to generate a candidate release
which the rest of us can download.  Put it up on and
then send a note to the list saying where to locate it.  Rather than call a vote right away,
just ask us to check it out and try it as there will likely be issues for the first release.
 Once we all feel we have a decent candidate, we can call a vote, which should be a formality.
>>>>>>>>>> See for more info.
>>>>>>>>>>> Karl
>>>>>>>>>>> On Fri, Nov 19, 2010 at 4:13 AM, Karl Wright
<> wrote:
>>>>>>>>>>>> The build changes are complete.  I removed
the modules level from the
>>>>>>>>>>>> hierarchy because it served no useful purpose
and complicated matters.
>>>>>>>>>>>>  The outer level build.xml now allows you
build code, docs, and run
>>>>>>>>>>>> tests separately from one another, and gives
you help as a default.
>>>>>>>>>>>> "ant image" builds you the deliverable .zip
and tar.gz files.  Online
>>>>>>>>>>>> site has been polished so that it now contains
complete javadoc, as
>>>>>>>>>>>> does the built and delivered .zip and tar.gz's.
 In short,  we *could*
>>>>>>>>>>>> actually do a release now, if only we had
(and incorporated) the KEYS
>>>>>>>>>>>> file I alluded to earlier, which I do not
know how to build or obtain.
>>>>>>>>>>>>  I believe this needs to be both generated
and registered.  The site
>>>>>>>>>>>> also needs to refer to a download location/list
of mirrors before it
>>>>>>>>>>>> could go out the door.
>>>>>>>>>>>> Help? Grant?
>>>>>>>>>>>> Karl
>>>>>>>>>>>> On Wed, Nov 17, 2010 at 9:50 PM, Karl Wright
<> wrote:
>>>>>>>>>>>>> Hearing nothing, went ahead and made
the port of documentation to the
>>>>>>>>>>>>> site official.  I also now include the
generated site in the release
>>>>>>>>>>>>> tar.gz and .zip.
>>>>>>>>>>>>> Issues still to address before release:
>>>>>>>>>>>>> (1) source tar.gz and zip in outer-level
build.xml, which I will try
>>>>>>>>>>>>> to address shortly.
>>>>>>>>>>>>> (2) vehicle for release downloads, and
naming thereof.  In short,
>>>>>>>>>>>>> where do I put these things so people
can download them??
>>>>>>>>>>>>> (3) Voting procedures for release.  I've
seen this done as a vote in
>>>>>>>>>>>>> - is that actually
>>>>>>>>>>>>> (4) Release branch and tag.  Do we want
both?  What is the correct
>>>>>>>>>>>>> naming for each in apache?
>>>>>>>>>>>>> (5) Legal requirements.  CHANGES.txt,
LICENSE.txt, etc.  Do these need
>>>>>>>>>>>>> to be included in the release tar.gz,
or just the source tar.gz?  I
>>>>>>>>>>>>> suspect both, but please confirm.  Also,
if there is a typical
>>>>>>>>>>>>> organization of the release tar.gz in
relation to the source tar.gz
>>>>>>>>>>>>> this would be a good time to make that
>>>>>>>>>>>>> Thanks,
>>>>>>>>>>>>> Karl
>>>>>>>>>>>>> On Tue, Nov 16, 2010 at 5:44 PM, Karl
Wright <> wrote:
>>>>>>>>>>>>>> What I've done here is taken all
the pages that I originally put in
>>>>>>>>>>>>>> the Wiki, describing how to set up
and run ManifoldCF, and converted
>>>>>>>>>>>>>> them to xdocs that are part of the
ManifoldCF site.  These documents
>>>>>>>>>>>>>> have no user content other than stuff
Grant or I added, according to
>>>>>>>>>>>>>> their logs, so I feel that is safe
to do.  I've left the wiki pages
>>>>>>>>>>>>>> around but am thinking we'll want
them to go away at some point.  Not
>>>>>>>>>>>>>> sure exactly what to do with all
the user comments to them, however.
>>>>>>>>>>>>>> Is this a reasonable way to proceed?
 We should avoid using the wiki
>>>>>>>>>>>>>> in the future for documentation,
seems to me, but otherwise I can see
>>>>>>>>>>>>>> no issues here.
>>>>>>>>>>>>>> Karl
>>>>>>>>>>>>>> On Tue, Nov 16, 2010 at 5:36 PM,
Grant Ingersoll <> wrote:
>>>>>>>>>>>>>>> On Nov 15, 2010, at 1:23 PM,
Jack Krupansky wrote:
>>>>>>>>>>>>>>>> I didn't mean to imply that
the wiki needs to be physically included in the release zip/tar, just that snapshotting and
versioning of the wiki should be done, if feasible, so that a user who is on an older release
can still see the doc for that release. I am just thinking ahead for future releases. So,
0.1 does not need this right now.
>>>>>>>>>>>>>>> Right, and I'm saying that we
can't include user generated content in a release unless we have explicitly asked for permission
on it in the form of patches and then committed by a committer.  Since we don't lock down
our wiki, we can't do it.
>>>>>>>>>>>>>>>> -- Jack Krupansky
>>>>>>>>>>>>>>>> -----Original Message-----
From: Grant Ingersoll
>>>>>>>>>>>>>>>> Sent: Monday, November 15,
2010 10:23 AM
>>>>>>>>>>>>>>>> To:
>>>>>>>>>>>>>>>> Subject: Re: Release?
>>>>>>>>>>>>>>>> On Nov 10, 2010, at 1:22
AM, Jack Krupansky wrote:
>>>>>>>>>>>>>>>>> And the wiki doc is also
part of the release. Does this stuff get a version/release as well? Presumably we want doc
for currently supported releases, and the doc can vary between releases. Can we easily snapshot
the wiki?
>>>>>>>>>>>>>>>> You can't put Wiki in a release,
as their is no way to track whether the person has permission to donate it..
>>>>>>>>>>>>>>>>> Will we have nightly
builds in place? I think a 0.1 can get released without a nightly build, but it would be nice
to say that we also have a "rolling trunk release" which is just the latest build off trunk
and the latest wiki/doc as well. So, some people may want the official 0.1, but others may
want to run straight from trunk/nightly build.
>>>>>>>>>>>>>>>>> -- Jack Krupansky
>>>>>>>>>>>>>>>>> -----Original Message-----
From: Karl Wright
>>>>>>>>>>>>>>>>> Sent: Tuesday, November
09, 2010 1:56 PM
>>>>>>>>>>>>>>>>> To:
>>>>>>>>>>>>>>>>> Subject: Re: Release?
>>>>>>>>>>>>>>>>> Proposal:  Release to
consist of two things: tar and zip of a complete
>>>>>>>>>>>>>>>>> source tree, and tar
and zip of the modules/dist area after the build.
>>>>>>>>>>>>>>>>> The implied way people
are to work with this is:
>>>>>>>>>>>>>>>>> - to use just the distribution,
untar or unzip the distribution
>>>>>>>>>>>>>>>>> zip/tar into a work area,
and either use the multiprocess version, or
>>>>>>>>>>>>>>>>> the quickstart example.
>>>>>>>>>>>>>>>>> - to add a connector,
untar or unzip the source zip/tar into a work
>>>>>>>>>>>>>>>>> area, and integrate your
connector into the build.
>>>>>>>>>>>>>>>>> Is this acceptable for
a 0.1 release?
>>>>>>>>>>>>>>>>> Karl
>>>>>>>>>>>>>>>>> On Tue, Nov 9, 2010 at
10:22 AM, Jack Krupansky
>>>>>>>>>>>>>>>>> <>
>>>>>>>>>>>>>>>>>> Oh, I wasn't intending
to disparage the RSS or other connectors, just giving
>>>>>>>>>>>>>>>>>> my own priority list
of "must haves." By all means, the "well-supported"
>>>>>>>>>>>>>>>>>> connector list should
be whatever list you want to feel is appropriate and
>>>>>>>>>>>>>>>>>> exclude only those
where "we" feel that "we" would not be able to provide
>>>>>>>>>>>>>>>>>> sufficient support
and assistance online.
>>>>>>>>>>>>>>>>>> That's great that
qBase is offering access.
>>>>>>>>>>>>>>>>>> BTW, I was just thinking
that maybe we should try to keep logs of each
>>>>>>>>>>>>>>>>>> connector type in
action so that people have a reference to consult when
>>>>>>>>>>>>>>>>>> debugging their own
connector-related problems. In other words, what a
>>>>>>>>>>>>>>>>>> successful connection
session is supposed to look like. So, have a test and
>>>>>>>>>>>>>>>>>> its "reference" log.
>>>>>>>>>>>>>>>>>> -- Jack Krupansky
>>>>>>>>>>>>>>>>>> -----Original Message-----
From: Karl Wright
>>>>>>>>>>>>>>>>>> Sent: Tuesday, November
09, 2010 9:46 AM
>>>>>>>>>>>>>>>>>> To:
>>>>>>>>>>>>>>>>>> Subject: Re: Release?
>>>>>>>>>>>>>>>>>> If you can claim
"well supported" for the web connector, you certainly
>>>>>>>>>>>>>>>>>> should be able to
claim it for the RSS connector.  You could also
>>>>>>>>>>>>>>>>>> reasonably include
the JDBC connector because it does not require a
>>>>>>>>>>>>>>>>>> proprietary system
to test.
>>>>>>>>>>>>>>>>>> But if your definition
is that tests exist for all the "well
>>>>>>>>>>>>>>>>>> supported" ones,
somebody has some work to do.  I'd like to see a plan
>>>>>>>>>>>>>>>>>> on how we get from
where we are now to a more comprehensive set of
>>>>>>>>>>>>>>>>>> tests.  I've gotten
qBase to agree to let me have access to their Q/A
>>>>>>>>>>>>>>>>>> infrastructure (which
used to be MetaCarta's), but that's only going
>>>>>>>>>>>>>>>>>> to be helpful for
diagnosing problems and doing development, not for
>>>>>>>>>>>>>>>>>> automated tests that
anyone can run.
>>>>>>>>>>>>>>>>>> Karl
>>>>>>>>>>>>>>>>>> On Tue, Nov 9, 2010
at 9:38 AM, Jack Krupansky
>>>>>>>>>>>>>>>>>> <>
>>>>>>>>>>>>>>>>>>> And one of the
issues on the list should be to define the "well-supported"
>>>>>>>>>>>>>>>>>>> connectors for
0.5 (or whatever) as opposed to the "code is there and
>>>>>>>>>>>>>>>>>>> thought to work,
you are on your own for testing/support" connectors.
>>>>>>>>>>>>>>>>>>> Longer
>>>>>>>>>>>>>>>>>>> term, "we" should
get most/all connectors into the well-supported
>>>>>>>>>>>>>>>>>>> category,
>>>>>>>>>>>>>>>>>>> but I wouldn't
use that as the bar for even 1.0.
>>>>>>>>>>>>>>>>>>> My personal minimum
"well-supported" connector list for a 0.5 would be
>>>>>>>>>>>>>>>>>>> file
>>>>>>>>>>>>>>>>>>> system, web,
and SharePoint*.
>>>>>>>>>>>>>>>>>>> * Oh... there
is the issue of SharePoint 2010 or whatever the latest is,
>>>>>>>>>>>>>>>>>>> but
>>>>>>>>>>>>>>>>>>> current MCF support
should be good enough for a 0.5 release, I think.
>>>>>>>>>>>>>>>>>>> (Got to keep
up with Google Connectors!)
>>>>>>>>>>>>>>>>>>> -- Jack Krupansky
>>>>>>>>>>>>>>>>>>> -----Original
Message----- From: Karl Wright
>>>>>>>>>>>>>>>>>>> Sent: Tuesday,
November 09, 2010 9:28 AM
>>>>>>>>>>>>>>>>>>> To:
>>>>>>>>>>>>>>>>>>> Subject: Re:
>>>>>>>>>>>>>>>>>>> I'm in favor
of a release.  I'm not sure, though, what the release
>>>>>>>>>>>>>>>>>>> parameters ought
to be.  I think the minimum is that we need to build
>>>>>>>>>>>>>>>>>>> a release infrastructure
and plan, set up a release process, and
>>>>>>>>>>>>>>>>>>> decide what the
release packaging should look like (zip's, tar's,
>>>>>>>>>>>>>>>>>>> sources, deliverables)
and where the javadoc will be published online.
>>>>>>>>>>>>>>>>>>> (It's possible
that we may, for instance, decide to change the way
>>>>>>>>>>>>>>>>>>> the ant build
scripts work to make it easier for people to build the
>>>>>>>>>>>>>>>>>>> proprietary connectors
after the fact, for instance.  Or we could
>>>>>>>>>>>>>>>>>>> claim that the
release is just the sources, either way.)
>>>>>>>>>>>>>>>>>>> After that, we
need to figure out what tickets we still want done
>>>>>>>>>>>>>>>>>>> before the release
occurs.  I'd argue for more testing, and I'm also
>>>>>>>>>>>>>>>>>>> trying to figure
out issues pertaining to Documentum and FileNet,
>>>>>>>>>>>>>>>>>>> because these
connectors require sidecar processes that are not well
>>>>>>>>>>>>>>>>>>> supported in
the example.  We could go substantially beyond that, but
>>>>>>>>>>>>>>>>>>> I agree with
Jack that 0.1 would be useful if we only get that far.
>>>>>>>>>>>>>>>>>>> Thoughts?
>>>>>>>>>>>>>>>>>>> Karl
>>>>>>>>>>>>>>>>>>> On Tue, Nov 9,
2010 at 8:58 AM, Jack Krupansky
>>>>>>>>>>>>>>>>>>> <>
>>>>>>>>>>>>>>>>>>>> At least
get a release 0.1 dry-run with code as-is out ASAP to flush out
>>>>>>>>>>>>>>>>>>>> release process
issues. This would help to send out a message to the rest
>>>>>>>>>>>>>>>>>>>> of
>>>>>>>>>>>>>>>>>>>> the world
that MCF is an available product rather than purely
>>>>>>>>>>>>>>>>>>>> development/incubation.
>>>>>>>>>>>>>>>>>>>> Then come
up with a list of issues that people strongly feel need to be
>>>>>>>>>>>>>>>>>>>> resolved
before a true, squeaky-clean 1.0 release. Maybe that is the
>>>>>>>>>>>>>>>>>>>> original
list of tasks, including better testing, but some
>>>>>>>>>>>>>>>>>>>> review/decisions
>>>>>>>>>>>>>>>>>>>> are probably
needed. That will be the ultimate target.
>>>>>>>>>>>>>>>>>>>> Then decide
on a "close enough" subset of issues that would constitute
>>>>>>>>>>>>>>>>>>>> what
>>>>>>>>>>>>>>>>>>>> people consider
a "solid beta" and target that as a release 0.5 and focus
>>>>>>>>>>>>>>>>>>>> on
>>>>>>>>>>>>>>>>>>>> that as the
near-term target (after getting 0.1 out ASAP.) I personally
>>>>>>>>>>>>>>>>>>>> do
>>>>>>>>>>>>>>>>>>>> not have
any major issues on the top of my head that I would hold out as
>>>>>>>>>>>>>>>>>>>> "blockers"
for a 0.5.
>>>>>>>>>>>>>>>>>>>> Or, get 0.1
out and then move on to a 0.2, etc. on a monthly/bi-monthly
>>>>>>>>>>>>>>>>>>>> basis as
progress is made.
>>>>>>>>>>>>>>>>>>>> In short,
get MCF as-is 0.1 out ASAP, have a very short list for MCF 0.5
>>>>>>>>>>>>>>>>>>>> to
>>>>>>>>>>>>>>>>>>>> get it out
reasonably soon, and then revisit what 1.0 really means versus
>>>>>>>>>>>>>>>>>>>> 0.6, etc.
>>>>>>>>>>>>>>>>>>>> -- Jack Krupansky
>>>>>>>>>>>>>>>>>>>> -----Original
Message----- From: Grant Ingersoll
>>>>>>>>>>>>>>>>>>>> Sent: Tuesday,
November 09, 2010 8:38 AM
>>>>>>>>>>>>>>>>>>>> To:
>>>>>>>>>>>>>>>>>>>> Subject:
>>>>>>>>>>>>>>>>>>>> Now that
we have NTLM figured out and the Memex stuff behind us, how do
>>>>>>>>>>>>>>>>>>>> people feel
about working towards a release?
>>>>>>>>>>>>>>>>>>>> -Grant
>>>>>>>>>>>>>>>> --------------------------
>>>>>>>>>>>>>>>> Grant Ingersoll
>>>>>>>>>>>>>>> --------------------------
>>>>>>>>>>>>>>> Grant Ingersoll
>>>>>>>>>> --------------------------
>>>>>>>>>> Grant Ingersoll
>>>>>> --------------------------
>>>>>> Grant Ingersoll
>>> --------------------------
>>> Grant Ingersoll
> --------------------------
> Grant Ingersoll

View raw message