On Tue, Jan 18, 2022 at 2:39 PM Luca Burgazzoli <lburgazz...@gmail.com> wrote:
>
> On Tue, Jan 18, 2022 at 2:31 PM Claus Ibsen <claus.ib...@gmail.com> wrote:
>
> > On Tue, Jan 18, 2022 at 12:39 AM Marat Gubaidullin
> > <marat.gubaidul...@gmail.com> wrote:
> > >
> > > 1. I could omit kebab cased properties in generator while we have both.
> > > 2. When read YAML Karavan "camelize" properties and DSL. Internal Karavan
> > > Camel Definitions are in camel case. So output YAML is also camel case.
> > >
> >
> > Yeah we can do that and for Camel 3.15 have both camelCase and
> > kebab-case in the yaml-dsl.
> > Where kebab-case is deprecated.
> >
>
> I'm not very sure we can support both kebab anc camel case in
> camel-yaml-dsl.json, (as
> example,. there may be an issue with required properties) however we may
> think to just
> deprecate the kebab case in the json schema but we can keep support in the
> yaml loader.
>

Ah so could we just generate two schema files, one with kebab-case and
another with camelCase.
Then Marat can use the camelCase for the Karavan tool.

The current yaml-dsl is all kebab-case based, in the generated model
serializers.

The parser is conforming to kebab-case, eg such as (so we support both
cases in the parser/loader)

final String propertyName =
StringHelper.camelCaseToDash(key.getValue()).toLowerCase(Locale.US);




>
> > This gives us that "wriggle room" to let kamelets, camel-k and others
> > migrate to camelCase.
> > Then in Camel 3.16 og 3.17 we can drop kebab and when we are fully
> > migrated.
> >
> >
> >
> > > On Mon., Jan. 17, 2022, 09:49 Claus Ibsen <claus.ib...@gmail.com> wrote:
> > >
> > > > On Mon, Jan 17, 2022 at 3:17 PM Marat Gubaidullin
> > > > <marat.gubaidul...@gmail.com> wrote:
> > > > >
> > > > > Hello Claus,
> > > > >
> > > > > 1. There is a typo (gzipDeflator instead of gzipDeflater) in the data
> > > > > format name
> > > > >
> > > >
> > https://github.com/apache/camel/blob/4829e4b7c9fec2ea785411a1ccad65339e080520/core/camel-core-model/src/main/java/org/apache/camel/model/dataformat/DataFormatsDefinition.java#L57
> > > > >
> > > > > 2. I have fetched, built and still have data format names (and
> > almost all
> > > > > other properties) in kabab case in camel-yaml-dsl.json
> > > > >
> > > >
> > > > One problem with having both camelCase and kebab-case in the
> > > > camel-yaml-dsl.json is that you would have double properties for
> > > > everything.
> > > > Then in the tooling you would need to filter out the duplicated kebab
> > > > based properties.
> > > >
> > > > Or we switch over to camelCase only. For the existing kamelets then we
> > > > need to update those, but they are almost all in camelCase anyway.
> > > > There is a few EIPs like set-header and set-property they are using,
> > > > which should be setHeader and setProperty.
> > > > So the migration is easy.
> > > >
> > > >
> > > > >
> > > > > On Mon, Jan 17, 2022 at 2:51 AM Claus Ibsen <claus.ib...@gmail.com>
> > > > wrote:
> > > > >
> > > > > > Hi
> > > > > >
> > > > > > Just a heads up that I have squashed and merged this to the main
> > > > branch.
> > > > > >
> > > > > > On Sun, Jan 16, 2022 at 3:12 PM Claus Ibsen <claus.ib...@gmail.com
> > >
> > > > wrote:
> > > > > > >
> > > > > > > Hi
> > > > > > >
> > > > > > > Ticket: https://issues.apache.org/jira/browse/CAMEL-17499
> > > > > > >
> > > > > > > The data formats have over the time become inconsistent in the
> > model
> > > > > > > and their names and also name vs camel-xxx JAR name etc.
> > > > > > >
> > > > > > > So after the 3.14 LTS release and where we drop JDK8 then we are
> > > > doing
> > > > > > > some house cleaning.
> > > > > > > We also have some components to be marked as deprecated and to be
> > > > > > removed later.
> > > > > > >
> > > > > > > The data format renaming is a more complex "thing" as we have a
> > bunch
> > > > > > > of source code generated files based on those names.
> > > > > > >
> > > > > > > So a rename of eg zipfile to zipFile (to use camel case) that
> > causes
> > > > > > > the generated files to keep the existing file name because the
> > lower
> > > > > > > case name existed. So to "fix" this a commit is needed to delete
> > the
> > > > > > > first, and then another to regenerate the files.
> > > > > > >
> > > > > > > So for all of this work I did this today on a quiet day but
> > there is
> > > > > > > about 50 commits in total as there are many regens in this line
> > of
> > > > > > > spirit to make it all work.
> > > > > > >
> > > > > > > I am not sure if a git quash would work?
> > > > > > >
> > > > > > > I pushed the code to a branch
> > > > > > > https://github.com/apache/camel/tree/dataformat-rename
> > > > > > >
> > > > > > > And a bit PR (we can try to squash this one)
> > > > > > > https://github.com/apache/camel/pull/6760
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > Claus Ibsen
> > > > > > > -----------------
> > > > > > > http://davsclaus.com @davsclaus
> > > > > > > Camel in Action 2: https://www.manning.com/ibsen2
> > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Claus Ibsen
> > > > > > -----------------
> > > > > > http://davsclaus.com @davsclaus
> > > > > > Camel in Action 2: https://www.manning.com/ibsen2
> > > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Claus Ibsen
> > > > -----------------
> > > > http://davsclaus.com @davsclaus
> > > > Camel in Action 2: https://www.manning.com/ibsen2
> > > >
> >
> >
> >
> > --
> > Claus Ibsen
> > -----------------
> > http://davsclaus.com @davsclaus
> > Camel in Action 2: https://www.manning.com/ibsen2
> >



-- 
Claus Ibsen
-----------------
http://davsclaus.com @davsclaus
Camel in Action 2: https://www.manning.com/ibsen2

Reply via email to