Problem with parameter default when redeploying report

I've written a rss script file to automate publishing of reports to
our server. I've encountered a problem when republishing reports that
have a default value set for a parameter.
If I change the default value of the parameter in the report rdl and
then try to republish, that parameter's default value is not getting
changed on the server. If I add or remove parameters then the server
gets updated correctly. It's only if I change the default value that
the update is not happening.
I've tried with both the CreateReport and SetReportDefinition
functions. The only way I've gotten this to work is to delete the
report and then republish but this is not an acceptable solution
because it also deletes report history and subscriptions.
I'm using RS2005.
Any help is appreciated.Default values for parameters are a little like datasources, in that you
have to explicitly indicate that you want to override a previous definition
of a datasource when you re-publish a report to a server. Basically the idea
is that your testbed, from which you publish, may not be the same as the
server environment, and you want to keep those things separate, and I'm
saying that parameters' defaults are treated like datasources in this
respect.
OK so far?
If you were handling this interactively using the Report Manager interface,
and assuming you have appropriate rights, you know that you can see the Data
Sources and configure them from the Properties tab of a report. Again, the
assumption is not made that the data source information for this report is
re-deployable and automatically written from your test bed.
Similarly, if a report has parameters, when you have selected the Properties
tab, you should see a Parameters item in the left-hand menu along with Data
Sources. Here you can set the default values differently from how they are
currently set -- I don't really understand whether "Override default" works
all the time or not, you will see it in the dialog, though. Never mind.
Here is where you can fix whatever you don't like about how the report got
re-published.
However, you say "I've tried with both the CreateReport and
SetReportDefinition functions", indicating that you are using web services
rather than interactively publishing. I understand this -- just consider
the above explanation a way to conceptualize *why* the parameters work the
way they do and require a separate step, rather than the way you expected.
I think that you may need to use the .SetReportParameters method here,
explicitly providing the new information to indicate that, yes, you want to
change the default values. If not, it may be the .SetProperties method.
I hope this works for you -- if not, you may be able to use the explanation
above to figure out the correct web service approach <s>.
>L<
<bruce42@.gmail.com> wrote in message
news:1175872439.784150.127590@.e65g2000hsc.googlegroups.com...
> I've written a rss script file to automate publishing of reports to
> our server. I've encountered a problem when republishing reports that
> have a default value set for a parameter.
> If I change the default value of the parameter in the report rdl and
> then try to republish, that parameter's default value is not getting
> changed on the server. If I add or remove parameters then the server
> gets updated correctly. It's only if I change the default value that
> the update is not happening.
> I've tried with both the CreateReport and SetReportDefinition
> functions. The only way I've gotten this to work is to delete the
> report and then republish but this is not an acceptable solution
> because it also deletes report history and subscriptions.
> I'm using RS2005.
> Any help is appreciated.
>|||So, from what you're saying, it's intentional that the parameter
defaults are not being overriden. You mention an "override defaults",
can you be more specific? I can see an "OverwriteDataSources" if I
publish straight from VS, but I've found nothing regarding overwriting
parameter defaults.
I'm familiar with the SetReportParameters method, what I'm trying to
do is avoid writing specific scripts for each report that I need to
publish. The rss script I use to publish now is a generic script that
I can use against any of my reports. Setting the default values
manually via the report manager interface is not really an option for
me. For one reason, it introduces the possibility of me setting
values differently than what was actually used in our test
environment, and two, we are using query based defaults and the report
manager interface doesn't give you enough detail to even be able to
make these changes (i.e. it doesn't show dataset or value field).
Currently the best idea I have for how to resolve this issue is to
publish my report to a temporary location (where it does not already
exist), loop through all of the parameters and capture the defaults, I
can then publish the report to it's normal location and do a
SetReportParameters using the default values I collected. I'm not
especially happy with this approach, it feels a little kludgy to me,
but at least it keeps me from having to write report specific rss
scripts.
Any other ideas?
thanks for your response...
-bruce
On Apr 6, 12:43 pm, "Lisa Slater Nicholls" <l...@.spacefold.com> wrote:
> Defaultvalues for parameters are a little like datasources, in that you
> have to explicitly indicate that you want to override a previous definition
> of a datasource when you re-publish a report to a server. Basically the idea
> is that your testbed, from which you publish, may not be the same as the
> server environment, and you want to keep those things separate, and I'm
> saying that parameters' defaults are treated like datasources in this
> respect.
> OK so far?
> If you were handling this interactively using the Report Manager interface,
> and assuming you have appropriate rights, you know that you can see the Data
> Sources and configure them from the Properties tab of a report. Again, the
> assumption is not made that the data source information for this report is
> re-deployable and automatically written from your test bed.
> Similarly, if a report has parameters, when you have selected the Properties
> tab, you should see a Parameters item in the left-hand menu along with Data
> Sources. Here you can set thedefaultvalues differently from how they are
> currently set -- I don't really understand whether "Overridedefault" works
> all the time or not, you will see it in the dialog, though. Never mind.
> Here is where you can fix whatever you don't like about how the report got
> re-published.
> However, you say "I've tried with both the CreateReport and
> SetReportDefinition functions", indicating that you are using web services
> rather than interactively publishing. I understand this -- just consider
> the above explanation a way to conceptualize *why* the parameters work the
> way they do and require a separate step, rather than the way you expected.
> I think that you may need to use the .SetReportParameters method here,
> explicitly providing the new information to indicate that, yes, you want to
> change thedefaultvalues. If not, it may be the .SetProperties method.
> I hope this works for you -- if not, you may be able to use the explanation
> above to figure out the correct web service approach <s>.
> >L<
> <bruc...@.gmail.com> wrote in message
> news:1175872439.784150.127590@.e65g2000hsc.googlegroups.com...
>
> > I've written a rss script file to automate publishing of reports to
> > our server. I've encountered a problem when republishing reports that
> > have adefaultvalue set for aparameter.
> > If I change thedefaultvalue of theparameterin the report rdl and
> > then try to republish, thatparameter'sdefaultvalue is not getting
> > changed on the server. If I add or remove parameters then the server
> > gets updated correctly. It's only if I change thedefaultvalue that
> > the update is not happening.
> > I've tried with both the CreateReport and SetReportDefinition
> > functions. The only way I've gotten this to work is to delete the
> > report and then republish but this is not an acceptable solution
> > because it also deletes report history and subscriptions.
> > I'm using RS2005.
> > Any help is appreciated.- Hide quoted text -
> - Show quoted text -|||Hi Bruce,
>> Any other ideas?
Forgive my lateness of reply (I will CC your e-mail to make sure you see
this) -- I don't get to the forum all that often.
AFAIK you are correct about not having the Overwrite parameter option to
match the Overwrite datasources when you publish straight from VS. When I
mentioned the "override defaults" I was talking only about the interactive
manager interface.
I do agree with you not only that it is confusing but also that, in most
cases, it is not advisable to set values differently in the test environment
than you would set in production. However -- bear with me, I am trying to
envision what was supposed to be the purpose of this "feature" -- we can
imagine that the designers of this system thought it *was* a good idea to
have an "override defaults" so that you could manage the report differently
on different servers, whether for test versus production or deployment of a
generic report to different customers.
For example there might be a sample size used for the test box that would be
different from production, or some sort of customer-specific value that you
wanted to use to brand a generic report.
Now to address your question...
For your generic script purposes, you might need to do something similar to
reflection to "publish" each report.
So far, I'm just restating something that you may have already tried with
your "publish to a temporary location". You could obviously pull the
parameters out of the appropriate Catalog field on the server, or use
GetReportParameters, if you've done that.
However, you don't really need to do that. Remember that the parameters
exist in the RDL, without publication, as a set of XML nodes. So, without
temporarily publishing anywhere, you should be able to read them out of the
RDL and issue the appropriate calls to set them properly on the target.
I hope this makes sense. If you like, you can e-mail me to discuss further
if it doesn't <g>. Again, I don't get here all that much...
>L<
<bruce42@.gmail.com> wrote in message
news:1176117653.014697.204110@.w1g2000hsg.googlegroups.com...
> So, from what you're saying, it's intentional that the parameter
> defaults are not being overriden. You mention an "override defaults",
> can you be more specific? I can see an "OverwriteDataSources" if I
> publish straight from VS, but I've found nothing regarding overwriting
> parameter defaults.
> I'm familiar with the SetReportParameters method, what I'm trying to
> do is avoid writing specific scripts for each report that I need to
> publish. The rss script I use to publish now is a generic script that
> I can use against any of my reports. Setting the default values
> manually via the report manager interface is not really an option for
> me. For one reason, it introduces the possibility of me setting
> values differently than what was actually used in our test
> environment, and two, we are using query based defaults and the report
> manager interface doesn't give you enough detail to even be able to
> make these changes (i.e. it doesn't show dataset or value field).
> Currently the best idea I have for how to resolve this issue is to
> publish my report to a temporary location (where it does not already
> exist), loop through all of the parameters and capture the defaults, I
> can then publish the report to it's normal location and do a
> SetReportParameters using the default values I collected. I'm not
> especially happy with this approach, it feels a little kludgy to me,
> but at least it keeps me from having to write report specific rss
> scripts.
> Any other ideas?
> thanks for your response...
> -bruce
> On Apr 6, 12:43 pm, "Lisa Slater Nicholls" <l...@.spacefold.com> wrote:
>> Defaultvalues for parameters are a little like datasources, in that you
>> have to explicitly indicate that you want to override a previous
>> definition
>> of a datasource when you re-publish a report to a server. Basically the
>> idea
>> is that your testbed, from which you publish, may not be the same as the
>> server environment, and you want to keep those things separate, and I'm
>> saying that parameters' defaults are treated like datasources in this
>> respect.
>> OK so far?
>> If you were handling this interactively using the Report Manager
>> interface,
>> and assuming you have appropriate rights, you know that you can see the
>> Data
>> Sources and configure them from the Properties tab of a report. Again,
>> the
>> assumption is not made that the data source information for this report
>> is
>> re-deployable and automatically written from your test bed.
>> Similarly, if a report has parameters, when you have selected the
>> Properties
>> tab, you should see a Parameters item in the left-hand menu along with
>> Data
>> Sources. Here you can set thedefaultvalues differently from how they are
>> currently set -- I don't really understand whether "Overridedefault"
>> works
>> all the time or not, you will see it in the dialog, though. Never mind.
>> Here is where you can fix whatever you don't like about how the report
>> got
>> re-published.
>> However, you say "I've tried with both the CreateReport and
>> SetReportDefinition functions", indicating that you are using web
>> services
>> rather than interactively publishing. I understand this -- just consider
>> the above explanation a way to conceptualize *why* the parameters work
>> the
>> way they do and require a separate step, rather than the way you
>> expected.
>> I think that you may need to use the .SetReportParameters method here,
>> explicitly providing the new information to indicate that, yes, you want
>> to
>> change thedefaultvalues. If not, it may be the .SetProperties method.
>> I hope this works for you -- if not, you may be able to use the
>> explanation
>> above to figure out the correct web service approach <s>.
>> >L<
>> <bruc...@.gmail.com> wrote in message
>> news:1175872439.784150.127590@.e65g2000hsc.googlegroups.com...
>>
>> > I've written a rss script file to automate publishing of reports to
>> > our server. I've encountered a problem when republishing reports that
>> > have adefaultvalue set for aparameter.
>> > If I change thedefaultvalue of theparameterin the report rdl and
>> > then try to republish, thatparameter'sdefaultvalue is not getting
>> > changed on the server. If I add or remove parameters then the server
>> > gets updated correctly. It's only if I change thedefaultvalue that
>> > the update is not happening.
>> > I've tried with both the CreateReport and SetReportDefinition
>> > functions. The only way I've gotten this to work is to delete the
>> > report and then republish but this is not an acceptable solution
>> > because it also deletes report history and subscriptions.
>> > I'm using RS2005.
>> > Any help is appreciated.- Hide quoted text -
>> - Show quoted text -
>