I just opened a Jira issue for this task, including Jeff's suggestion:
http://objectstyle.org/jira/secure/ViewIssue.jspa?key=CAY-402
Cris, we don't have an Ant task for reverse engineering yet. I can
see how such task can be useful - it would allow to sync your objects
directly against the DB, bypassing the Modeler. Is this what you had
in mind?
Looks like this thing is growing big. I marked the issue as "AFTER 1.2".
Andrus
On Nov 20, 2005, at 10:08 PM, Cris Daniluk wrote:
> Yeah... it doesn't seem like it would be a terribly big deal to
> squeeze into 1.2. While having it in preferences makes the most sense
> from the UI perspective, I do think its important that it not be
> stored in the preferences DB - this way it can be used in the ant
> tasks too.
>
> The Wiki could serve as a holding place for pre-canned rules that
> people provide, as well...
>
> On 11/20/05, Andrus Adamchik <andru..bjectstyle.org> wrote:
>> Cris,
>>
>> I think you may be on to something. We can create a new panel in
>> Preferences where users would enter regular expression rules. Each
>> rule in a chain can be applied before or after basic conversion of
>> "ABC_XYZ" to "abcXyz" done by Cayenne.
>>
>> E.g. Ahmed's example may look like this:
>>
>> Search: ([^s])s$
>> Replace: $1
>> After Conversion: Yes
>>
>> Stripping a "DB_" prefix from table name would look like this:
>>
>> Search: ^DB_
>> Replace:
>> After Conversion: No
>>
>> This is the most flexible solution I can think of... and a pretty
>> straightforward too.
>>
>> Andrus
>>
>>
>>
>> On Nov 20, 2005, at 8:35 PM, Cris Daniluk wrote:
>>> You know... I don't know that this is necessarily as difficult as we
>>> are making it out to be. It seems like it shouldn't be a blind
>>> "remove
>>> the s" type system, but rather a mapping file. A simple XML mapping
>>> file could send Products to Product, etc, and grow with time.
>>>
>>> Users could then select whether or not to use it when generating via
>>> the modeler or the ant task and even specify their own alternate
>>> file.
>>> I've long wanted a similar feature - to be able to systematically
>>> remove field prefixes/suffixes.
>>>
>>> Cris
>>
>>
>
This archive was generated by hypermail 2.0.0 : Wed Nov 30 2005 - 02:46:13 EST