-
Notifications
You must be signed in to change notification settings - Fork 9
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Last commit broke connect #14
Comments
@michaelfliegner Thanks for this! Sorry, my bad, I copied from the MySQL adaptor and missed these differences! |
Mmm, actually the solution is not 100% correct - the objective was to allow defining connection parameters using ENV variables. Now these are not available for dbname and user. |
I see. can fix that to support You, if you want. |
@michaelfliegner I pushed the changes on master - can you please check that it works fine? If good, I'll tag a new version. No idea re the CI job... we need to check. |
There are no test jobs configured. Looks like they haven't been setup for Postgres. |
I can provide a CI-Workflow soon, a test stub and just a test on connect :-) Others can easily contribute tests then. OK? |
There is an PR open but now useless because of the ongoing work made in the past months. The main parts are ready and need only to be copied into a new PR based on the actual version. This is really straight forward. The first test are also there. |
@FrankUrbach can you please detail how you see this module? |
@essenciary This modul should contain the all general functions which every adapter should implement in the same fashion. E.g. connect, disconnect, query etc. So it would be possible to get the same results from each and every adapter. If there are differences we can pin it at the documentation. So we have the ability to test the conformity of the api you've created in the past. |
couldn't it be submodule of SearchLight? |
Yes of course. That was the idea behind my answer. |
That's what I try to understand. Why would a testing module become part of the library? Why not a file in |
From the point of view of the adapter Searchlight is a different package an is not reachable if you don't know where the source is placed. In order to get what we want, we must know in each and every environment where the source of the tests of SearchLight are located. Is this possible? I don't think so. Therefore the idea to create a test module inside Searchlight. I can imagine that you are not happy about a module which isn't necessary in production. But to be honest, this isnt that much code we will need for testing. |
The way I see it, there are multiple things to consider - and we need clarity re what we want to achieve. 1/ if we talk about pure 2/ For 1/ unit testing can be put directly into SearchLight tests. |
Now we are at the same page. If you set up the package SearchlightTest.jl I will try to fill it with live. The rest is work. |
@FrankUrbach @michaelfliegner I have created a new blank repo at https://github.com/GenieFramework/SearchLightTests.jl and invited you as maintainers. |
@FrankUrbach <https://github.com/FrankUrbach> I take it, that You take the
lead, so I wait for Your suggestions.
Michael
…On Tue, Aug 16, 2022 at 9:34 AM Adrian Salceanu ***@***.***> wrote:
@FrankUrbach <https://github.com/FrankUrbach> @michaelfliegner
<https://github.com/michaelfliegner> I have created a new blank repo at
https://github.com/GenieFramework/SearchLightTests.jl and invited you as
maintainers.
—
Reply to this email directly, view it on GitHub
<#14 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AC6Y2USJBRVDUWP5V6TDSLLVZNAADANCNFSM56E47A6A>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
haskey(conn_data, "database") && push!(dns, string("dbname=", conn_data["database"]))
was deleted, but is needed, as LibPQ expects an option
dbname
instead ofdatabase
same with
username
/user
The text was updated successfully, but these errors were encountered: