Dies,
Message ids can not have a space as that will cause an exception when we
try to use the message id as a key.
If you would like to help that would be great. I know that Sekhar was
compiling a list of messages and ids that should be added but I don't
have that list. the thing that needs to be done when adding the message
ids is to figure out which logger is used and make sure that the message
id is unique in the package where it is used. Sekhar and I thought it
would be easiest to just assign a range of numbers for each module
(actually logger from LogDomains) and that would minimize duplicates. I
can publish that list tomorrow. Note that there are cases where will
may not be able to figure out the logger being used and will have to use
come up with a solution.
Let me know if you have questions.
Carla
Gail Risdal wrote:
> Hi Dies,
>
> One other thing ...
>
> I just heard back from the doc tools developer, who says #3 and #4 in
> your list would break the tool we're using to harvest the error messages
> for the documentation.
>
> Thanks,
> Gail
>
> Dies Koper wrote:
>> Hi,
>>
>> We had several threads about the messages: that they need message IDs,
>> diag info and (not much discussed) be externalized for localization.
>> Sekhar has even prepared a tool to help.
>>
>> Since then I think everybody has been too busy to actually do the work.
>>
>> I'd like to help out where I can, and already sent out a patch to
>> Shalini directly for jdbc, but felt we still have different opinions
>> about what to do. I'd like to move the discussion here to explain my
>> intentions so that all the module owners for whom I might prepare
>> similar patches are in agreement with it.
>>
>> For now, I'd like to fix the following issues:
>>
>> 1. Messages with no ID
>>
>> 2. Typos (just the obvious ones, I'm not a professional proofreader)
>>
>> 3. Message ID (layout) issues (such as space in "RAR 7014")
>>
>> 4. Single apostrophes in messages with arguments (see issue #9896)
>>
>> For most messages I can't supply the diag info, it would take too much
>> time to investigate each and find meaningful check/cause explanations.
>> Is that okay?
>>
>> Although from the discussions I understood that message IDs on INFO
>> messages are "not required", I did not take it to mean INFO messages
>> CANNOT have IDs. In many cases it will be easier for me to just add IDs
>> to all messages, as cross-referencing with the logging code to check the
>> level would take a lot of time. Is that okay?
>> (Of course I would be careful not to add IDs to messages that should not
>> have IDs, and you'll have a chance to double-check when you review the
>> patches.)
>>
>> Also, it is my understanding that the doc team will run a tool at the
>> end to pick up all messages from the property files, so I do not need to
>> worry about updating the docs. Is that correct?
>>
>> Thanks,
>> Dies
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe_at_glassfish.dev.java.net
>> For additional commands, e-mail: dev-help_at_glassfish.dev.java.net
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe_at_glassfish.dev.java.net
> For additional commands, e-mail: dev-help_at_glassfish.dev.java.net
>