Hi there!
Well, then my comment was not completely wrong.
Sqlite was a quick&dirty way to test if my ideas would work. For my future
productive system I really want to use a mix of couchdb and maybe postgres.
i.e. a JSON document storage for the component information and sql for
inventory managem
>Sqlite was a quick&dirty way to test if my ideas would work.
There are ODBC wrappers for SQLite
> I mean libreoffice would do for the management.
Yes, and you know what you would use? Not CSV files.
LibreOffice Base / Microsoft Access.
This is the office suite database that's basically SQ
Putting aside the fact that I think this feature isn't really aimed at
hobbyists, I would not be opposed to people wanting to extend it beyond
ODBC but that comes with extra complexity that must be handled.
With ODBC, KiCad just needs to know about the interface to retrieve the
data.
With a CSV f
I would most likely reject any solution that was tied to a particular
database. All this would do is open up a Pandora's box of complaints as
to why we didn't use database A over database B. ODBC is the most
flexible solution that I am aware of and allows users to choose their
preferred database.
Surely there must be an open source impl of an ODBC interface on a CSV file?
Although I’m not sure of the desire to avoid MySQL. It’s remarkably easy to
set up an instance (or auto-deploy one with an app).
Apologies if we’ve already talked about that; I’ll confess to not having
followed this
On 8/29/20 9:21 AM, Jeff Young wrote:
> Surely there must be an open source impl of an ODBC interface on a CSV file?
I'm pretty sure this already exists.
>
> Although I’m not sure of the desire to avoid MySQL. It’s remarkably
> easy to set up an instance (or auto-deploy one with an app).
My
I think this thread has kind of blown up to use cases beyond the original
subject.
While I don't think it's wrong to consider *other* proposals for how people
want to get library data into KiCad, I want to limit this particular thread
(and the linked spec doc to one in particular):
An external OD
> Surely there must be an open source impl of an ODBC interface on a CSV
file?
Yep they do. On Windows, Microsoft actually installs their "Microsoft Text
Driver" with MS Office which allows ODBC to CSV files.
>Although I’m not sure of the desire to avoid MySQL. It’s remarkably easy
to set up an
Nobody is suggesting that we bundle any database engine.
If your company doesn't want to use MySQL, and they want to use this
feature, I would suggest Postgres or MariaDB :)
On Sat, Aug 29, 2020 at 9:52 AM Mark Roszko wrote:
> > Surely there must be an open source impl of an ODBC interface on a
https://calcite.apache.org/docs/adapter.html
Apache calcite is an SQL query engine for non-sql data sources.
On Sat, Aug 29, 2020, 21:21 Jeff Young wrote:
> Surely there must be an open source impl of an ODBC interface on a CSV
> file?
>
> Although I’m not sure of the desire to avoid MySQL.
> Other than that, bundling a database engine is incredibly messy and a world
> of packaging hell from conflicts, IT policy, MS Group policy, etc. Not worth
> bundling.
>
Even SQLite? I’ve incorporated it into other projects with a minimum of fuss.
Not that I’m advocating for KiCAD to neces
BINGO! That didn't take long. I knew the non-sql database question
would rear it's ugly head. It just happened quicker than I thought it
would ;)
On 8/29/20 9:58 AM, Tim Hawkins wrote:
> https://calcite.apache.org/docs/adapter.html
>
> Apache calcite is an SQL query engine for non-sql data sour
My point about MySQL wasn’t that *we* should bundle it, but rather why some of
the contributors to this thread wanted to avoid it (or other things that
support ODBC) and instead use CSV.
___
Mailing list: https://launchpad.net/~kicad-developers
Post
I think the people suggesting CSV are missing the point of the database
feature (as I see it): collaboration.
Having 5 different people working from a single CSV file is not fun.
-Jon
On Sat, Aug 29, 2020 at 10:26 AM Jeff Young wrote:
> My point about MySQL wasn’t that *we* should bundle it, b
Calcite is a classic open source application. No ODBC driver but they
support queries in pig latin.
On Sat, Aug 29, 2020 at 7:03 AM Wayne Stambaugh
wrote:
> BINGO! That didn't take long. I knew the non-sql database question
> would rear it's ugly head. It just happened quicker than I thought
>>I think the people suggesting CSV are missing the point of the database
feature (as I see it): collaboration.
But if there is an application which acts between CSV and OBDC it is fine
and everybody is happy :)
I updated the diagram to make it more clear. It can be seen that the part
selector wid
Awesome. You guys might lose me so I can go work on that….
> On 29 Aug 2020, at 16:07, Seth Hillbrand wrote:
>
> Calcite is a classic open source application. No ODBC driver but they
> support queries in pig latin.
>
___
Mailing list: https://laun
Hello all,
probably a repeated question, but does any one is able to build KiCad in a
distributed build such as distcc or icecc ?
Mario
___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubs
Hi Mário-
You can use the cmake flag -DUSE_DISTCC. I frequently use this.
Best-
Seth
On Sat, Aug 29, 2020 at 3:15 PM Mário Luzeiro wrote:
> Hello all,
>
> probably a repeated question, but does any one is able to build KiCad in a
> distributed build such as distcc or icecc ?
>
> Mario
>
> ___
## Seth Hillbrand (s...@kipro-pcb.com):
> You can use the cmake flag -DUSE_DISTCC. I frequently use this.
Does distcc speed up the build that much? (Here the full package
build - chroot cloning, dependencies installation, kicad build and
packaging - takes between 8 and 9 minutes).
Regards,
Chri
ah thanks! seems to be working .. (I was trying to manually change the CC flag
to distcc)
How do I increase the jobs on the machines? It is using me 4 on each machine.
looks the -j make flag does not work here..
on the server side I set it to some high number..
Mario
___
ouch! it takes to me 20 minutes or probably more (8 CPU AMD, 8 GB RAM)
..without distcc.. so I'm looking for some help to reduce this time :)
From: Kicad-developers
on behalf of
Christoph Moench-Tegeder
Sent: 29 August 2020 23:42
To: kicad-developers@l
## Mário Luzeiro (mrluze...@ua.pt):
> ouch! it takes to me 20 minutes or probably more (8 CPU AMD, 8 GB RAM)
> ..without distcc.. so I'm looking for some help to reduce this time :)
Is that with SSD or spinning rust? RAM could help, too - as a database
person, I'd recommend 4 to 8 GB per CPU core
On Sat, Aug 29, 2020 at 3:43 PM Christoph Moench-Tegeder
wrote:
> ## Seth Hillbrand (s...@kipro-pcb.com):
>
> > You can use the cmake flag -DUSE_DISTCC. I frequently use this.
>
> Does distcc speed up the build that much? (Here the full package
> build - chroot cloning, dependencies installation
On Sat, Aug 29, 2020 at 3:43 PM Mário Luzeiro wrote:
> ah thanks! seems to be working .. (I was trying to manually change the CC
> flag to distcc)
>
> How do I increase the jobs on the machines? It is using me 4 on each
> machine.
> looks the -j make flag does not work here..
> on the server side
Hi Seth,
I managed to make it work, on my setup I had to change the .distcc/hosts file
where I set the maximum allowed jobs by adding IP/NJOBS
It looks the localhost is limited to 2 jobs and remotes to 4 by default
so I had to tuned it a bit..
Mario
From
26 matches
Mail list logo
