Hi
On Tue, Jun 7, 2016 at 8:02 AM, Sandeep Thakkar <
sandeep.thak...@enterprisedb.com> wrote:
> Hi Dave
>
> I used few APIs from Importing Modules and Modules Objects to import the
> pgAdmin4 module and then get it's location but the application crashed on
> running. I didn't debug it and
Hi Dave
I used few APIs from Importing Modules and Modules Objects to import the
pgAdmin4 module and then get it's location but the application crashed on
running. I didn't debug it and instead created a new function in Server.cpp
to get the python path and the webpath which uses python command.
On Mon, Jun 6, 2016 at 2:44 PM, Dave Page wrote:
> Hmm, virtualenv's a good point.
>
> I wonder if for the RPMs (and DEBs) we're just trying too hard. Is there
> any good reason to support SxS there? Stability I suppose, but then we
> don't support back-branches long term
--Adding back Devrim :-)
On Mon, Jun 6, 2016 at 2:44 PM, Dave Page wrote:
> Hmm, virtualenv's a good point.
>
> I wonder if for the RPMs (and DEBs) we're just trying too hard. Is there
> any good reason to support SxS there? Stability I suppose, but then we
> don't support
Hmm, virtualenv's a good point.
I wonder if for the RPMs (and DEBs) we're just trying too hard. Is there
any good reason to support SxS there? Stability I suppose, but then we
don't support back-branches long term anyway.
Does anyone think we need to support side-by-side RPM/DEB installation of
I already asked them about the APIs, though I didn't ask them about what is
the best way to handle SxS installation. Will check with them.
PIP package for pgadmin4 doesn't support SxS as it creates the directory
with the name 'pgadmin4' only. Googling about the SxS with PIP says that
people use
I have no idea. I would ask one of the Python guru's sitting next to you
(as well as whether the way we'd handle side-by-side packages is
appropriate). Also, look at what the PIP package does (does that even work
properly in a SxS scenario? I don't know if we thought to check that).
BTW; on the
Yeah, I got the point. To distinguish between v1 and v2, we can have blank
__init__.py in the v1 and v2 directories. I tried it and could successfully
import the pgAdmin4 using "import pgadmin4_web_v1.pgAdmin4" and "import
pgadmin4_web_v2.pgAdmin4". Please note that I had to rename hyphen to
My point is that the runtime uses the platform supplied Python interpreter,
which presumably knows where to search for packages. Mind you, I suppose
the issue there is that it wouldn't be able to distinguish between v1 and
v2 then...
I don't have a major issue with your suggested code - I just
Do you mean to say when a python app is launched, it imports some modules
automatically and in that sense it knows about where it's site-packages
are? May be, but how the pgAdmin4 runtime will know where the Web App is
installed?
The changes that I have done to the runtime is to let it know the
Hi,
Well, I have to wonder why we need the changes to the runtime? We're
linking the runtime with the same build of Python that's already on the
system - doesn't it know where it's site-packages are already? I could see
us needing this is we were using a distro-independent build of Python and
Hi Dave,
how about changes in the pgadmin4 source code for conf.py and Server.cpp?
Looks okay?
On Fri, Jun 3, 2016 at 2:41 PM, Sandeep Thakkar <
sandeep.thak...@enterprisedb.com> wrote:
> Thanks Dave.
>
> On Fri, Jun 3, 2016 at 2:08 PM, Dave Page wrote:
>
>> Hi
>>
>> On Thu,
Thanks Dave.
On Fri, Jun 3, 2016 at 2:08 PM, Dave Page wrote:
> Hi
>
> On Thu, Jun 2, 2016 at 4:23 PM, Sandeep Thakkar
> wrote:
> > Hi Devrim, Hi Dave,
> >
> > I have updated the patch. The earlier patch may fail because of app
> bundle
> >
Hi
On Thu, Jun 2, 2016 at 4:23 PM, Sandeep Thakkar
wrote:
> Hi Devrim, Hi Dave,
>
> I have updated the patch. The earlier patch may fail because of app bundle
> commit in git.
>
> For testing, you may define the source tarball location as :
> Source0:
>
Hi Devrim, Hi Dave,
I have updated the patch. The earlier patch may fail because of app bundle
commit in git.
For testing, you may define the source tarball location as :
Source0:
http://bugatti.pn.in.enterprisedb.com/temp/pgadmin4/%{name}-v%{version}.tar.gz
Known issue that I'm still working
Hi
Few changes in the updated patch:
- added the missing modules in the specfile. The unavailable modules are
still commented.
- added changelog in specfile
- added dependency of pgadmin4-doc on pgadmin4-web
On Wed, Jun 1, 2016 at 2:57 PM, Sandeep Thakkar <
sandeep.thak...@enterprisedb.com>
Hi Devrim,
I have attached the patch for review. Please note that right now I have
commented the python dependencies (in Requires) which you are building.
Please review and let me know if specfile or anything else needs some
changes. Once the rpms are built, please let me know how to install them
[Adding Devrim]
On Fri, May 27, 2016 at 1:55 PM, Sandeep Thakkar
wrote:
>
>
> On Mon, May 9, 2016 at 6:35 PM, Dave Page wrote:
>>
>> Hi
>>
>> Initial eyeball review comments below...
>>
>> On Fri, Apr 22, 2016 at 11:57 AM, Sandeep Thakkar
>>
On Mon, May 9, 2016 at 6:35 PM, Dave Page wrote:
> Hi
>
> Initial eyeball review comments below...
>
> On Fri, Apr 22, 2016 at 11:57 AM, Sandeep Thakkar <
> sandeep.thak...@enterprisedb.com> wrote:
>
>> Hi Team, Dave,
>>
>> Attached herewith are two patches.
>>
>>
Hi
Initial eyeball review comments below...
On Fri, Apr 22, 2016 at 11:57 AM, Sandeep Thakkar <
sandeep.thak...@enterprisedb.com> wrote:
> Hi Team, Dave,
>
> Attached herewith are two patches.
>
> *pgadmin4-rpm.patch* - This is the main patch that includes scripts,
> makefiles and spec to
Hi Team, Dave,
Attached herewith are two patches.
*pgadmin4-rpm.patch* - This is the main patch that includes scripts,
makefiles and spec to create RPMs for RHEL6/RHEL7/F-22/F-23/F-24.
It will create two RPMs i.e pgadmin4 and pgadmin4-web. The pgadmin4 tpm
depends on web and the web rpm
21 matches
Mail list logo