sqlite build failure on BSP bsets

[Joel Sherrill]

Thank you so much Jonathon!  I have pushed this.

We are really trying to close all issues and get things ready for a
release. All help and user testing is appreciated.

--joel

On Thu, Apr 2, 2020 at 5:47 PM Jonathan Brandmeyer <jbrandmeyer at planetiq.com>
wrote:

> Try this.
>
> I changed the logic which avoids building the sqlite3 shell.  Something
> about the way it uses libtool made the old `touch`-based method fail.
> There's also a bunch of configuration notes that really ought to be more
> visible.
>
> On Thu, Apr 2, 2020 at 2:26 PM Joel Sherrill <joel at rtems.org> wrote:
>
>>
>>
>> On Thu, Apr 2, 2020 at 3:00 PM Gedare Bloom <gedare at rtems.org> wrote:
>>
>>> On Thu, Apr 2, 2020 at 1:57 PM Jonathan Brandmeyer
>>> <jbrandmeyer at planetiq.com> wrote:
>>> >
>>> >
>>> > On Thu, Apr 2, 2020 at 1:43 PM Joel Sherrill <joel at rtems.org> wrote:
>>> >>
>>> >> Thank you for the detailed feedback.
>>> >>
>>> >> I am torn between:
>>> >>
>>> >> + deleting sqlite from the BSP bset's
>>> >> + requiring tclsh to be installed
>>> >> + updating the RSB recipe for sqlite.
>>> >>
>>> >> Given the desire to get a release out the door. I am prone to the
>>> first
>>> >> option with a ticket for 5.2 to update sqlite and re-add it to the BSP
>>> >> bsets.
>>> >
>>> >
>>> > I can see where you're coming from.  That said, even having an
>>> obsolete bset in the RSB served as a helpful hint that it worked for
>>> someone once upon a time.
>>> >
>>> >>
>>> >>
>>> >> Since you obviously use sqlite, if you could tackle updating the RSB
>>> >> recipe, that would be great and avoid having to do something expedient
>>> >> to get the release out the door.
>>> >
>>> >
>>> >  I can make a pass at it.  What's the quickest way to iterate on a
>>> BSET's rules?  I'd rather not have to wait on GCC and crew to rebuild when
>>> I get it wrong the first time.
>>> >
>>> Use --dry-run option
>>>
>>
>> This is good to watch the commands quickly.  But you can just explicitly
>> build the one package.
>>
>> $ cd rtems
>> $ ../source-builder/sb-set-builder \
>>      --log=l-sqlite.txt \
>>      --prefix=XXXX \
>>       databases/sqlite
>>
>> Should be very close to working. I hacked my command for dtc so there may
>> be a typo.
>>
>> This should take seconds.
>>
>> It is interesting that you didn't know you can build a single package
>> or --dry-run. No foul to you. Just points out that we aren't letting the
>> community know the cool things the RSB can do.
>>
>> I use the source download only option to get people to download things
>> for the tools before they come to the Kick Start class. If they have
>> network
>> issues, can't fetch git, no admin, etc. then they can deal with those with
>> their IT. I can't help with someone's draconian policies. :)
>>
>>
>>>
>>> >>
>>> >> Chris will eventually wake up in Sydney and chime in. :)
>>> >>
>>> >
>>> >
>>> > --
>>> > Jonathan Brandmeyer
>>> > _______________________________________________
>>> > devel mailing list
>>> > devel at rtems.org
>>> > http://lists.rtems.org/mailman/listinfo/devel
>>>
>>
>
> --
> Jonathan Brandmeyer
> Vice President of Software Engineering
> PlanetiQ
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20200402/7c2808b0/attachment.html>