You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@camel.apache.org by Claus Ibsen <cl...@gmail.com> on 2020/04/16 15:05:47 UTC
Maintaining information for camel-catalog in an easier way for the
huge number of components we have
Hi
The information we put together for camel-catalog are assembled from
source code, pom.xml, and potentially other places.
I have thought of having a csv file (or whatever format) in the root
of components folder, where we can override the information that are
being used. As this allows to easily maintain that information instead
of having to go to each .java source code for the @UriEndpoiint and
whatnot.
So the csv file could be something like
artifactId=camel-ftp,name=ftps,description=Bla bla,supportLevel=Preview
We can of course use other formats. Or have different files for
components, languages, data formats, others, etc and then have a fixed
order, so you can easily edit this from even a spreadsheet
Any thoughts?
--
Claus Ibsen
-----------------
http://davsclaus.com @davsclaus
Camel in Action 2: https://www.manning.com/ibsen2
Re: Maintaining information for camel-catalog in an easier way for
the huge number of components we have
Posted by Peter Palaga <pp...@redhat.com>.
I am open to discuss whether some particular bits should be hosted
elsewhere from their current location, but moving all to a new file
sounds like un-rooting many of those pieces from their natural biotope.
E.g. for artifactId: everybody knows the master value is in pom.xml,
copying it to a new file will introduce a new problem of keeping the two
in sync. -- P
On 16/04/2020 17:05, Claus Ibsen wrote:
> Hi
>
> The information we put together for camel-catalog are assembled from
> source code, pom.xml, and potentially other places.
>
> I have thought of having a csv file (or whatever format) in the root
> of components folder, where we can override the information that are
> being used. As this allows to easily maintain that information instead
> of having to go to each .java source code for the @UriEndpoiint and
> whatnot.
>
> So the csv file could be something like
>
> artifactId=camel-ftp,name=ftps,description=Bla bla,supportLevel=Preview
>
> We can of course use other formats. Or have different files for
> components, languages, data formats, others, etc and then have a fixed
> order, so you can easily edit this from even a spreadsheet
>
> Any thoughts?
>
>
Re: Maintaining information for camel-catalog in an easier way for
the huge number of components we have
Posted by David Jencks <da...@gmail.com>.
I fear that this will make this information impossible to maintain rather than just difficult and confusing, as, IIUC, there will now be two possible sources of data for every item in the catalog, and there’s a 50% to 100% chance one of them is wrong. Can you provide a complete end-to-end use case where this is a good idea?
i have no idea if it’s remotely practical, but I wonder if a tool to go from an item in the catalog to it’s source would be possible.
Making it easier to find and edit this information is certainly a good idea!
David Jencks
> On Apr 16, 2020, at 8:05 AM, Claus Ibsen <cl...@gmail.com> wrote:
>
> Hi
>
> The information we put together for camel-catalog are assembled from
> source code, pom.xml, and potentially other places.
>
> I have thought of having a csv file (or whatever format) in the root
> of components folder, where we can override the information that are
> being used. As this allows to easily maintain that information instead
> of having to go to each .java source code for the @UriEndpoiint and
> whatnot.
>
> So the csv file could be something like
>
> artifactId=camel-ftp,name=ftps,description=Bla bla,supportLevel=Preview
>
> We can of course use other formats. Or have different files for
> components, languages, data formats, others, etc and then have a fixed
> order, so you can easily edit this from even a spreadsheet
>
> Any thoughts?
>
>
> --
> Claus Ibsen
> -----------------
> http://davsclaus.com @davsclaus
> Camel in Action 2: https://www.manning.com/ibsen2
Re: Maintaining information for camel-catalog in an easier way for
the huge number of components we have
Posted by Andrea Cosentino <an...@gmail.com>.
I think it's a good idea to have information in a central location, but I
do believe the csv approach is a bit error prone.
Maybe we could focus on improving the information we already have in terms
of @UriEndpoint and java source code
Il giorno gio 16 apr 2020 alle ore 17:06 Claus Ibsen <cl...@gmail.com>
ha scritto:
> Hi
>
> The information we put together for camel-catalog are assembled from
> source code, pom.xml, and potentially other places.
>
> I have thought of having a csv file (or whatever format) in the root
> of components folder, where we can override the information that are
> being used. As this allows to easily maintain that information instead
> of having to go to each .java source code for the @UriEndpoiint and
> whatnot.
>
> So the csv file could be something like
>
> artifactId=camel-ftp,name=ftps,description=Bla bla,supportLevel=Preview
>
> We can of course use other formats. Or have different files for
> components, languages, data formats, others, etc and then have a fixed
> order, so you can easily edit this from even a spreadsheet
>
> Any thoughts?
>
>
> --
> Claus Ibsen
> -----------------
> http://davsclaus.com @davsclaus
> Camel in Action 2: https://www.manning.com/ibsen2
>