Reporting Services, Charts with time-scale axis

Hi all,
Firstly sorry if this is the wrong place to be posting
this question ... it's the most appropriate newsgroup I
could find.
I'm creating a report using reporting services and I'm
including a graph thats shows number of times something
is accessed per day over time. Data coming from an SQL
Server 2000 database. The dataset will only have data
records in if the item is accessed ... i.e. if it is not
accessed in a particular day then no record is present.
I've therefore made the chart time-scale in the x-axis to
ensure the scaling is correct. I've even included a union
query to a blank set of data for todays date such that the
graph will always extend up to the current date. The
report has a parameter such that a filter can be applied
to the data to show particular sections only (i.e.
accesses by a particular individual) This all works fine
for when there is a relatively large set of records post
filter and the charts look good.
The problem comes when there is only a small amount of
data, say three points, over a large period of time. The
scaling on the x-axis is correct and runs from the first
data items date up until the current date. The problem is
that the width of each column is massive and spans maybe 7
days. It looks to me as though the width of the columns is
calculated according to the number of the points (in this
example 3) and not according to the number of "time-sacle
points" maybe 100. The result is three hugely wide bars
each spanning a weeks worth of dates with the left hand
edge on the date of the actual event.
I could try and introduce zero data into the data set for
each of the intervening dates but didn't want to go to
through the overhead of calculating all this unnecessary
data ... that is after all the point of a time-scale axis.
Would appreciate it is anyone has any experience in this
or ideas I could try.
Thanks in advance (and for reading this far!)
Matt.Matt,
Just to let you know that there is an RS group: .reportingsvcs. I think you have much better change
to get good help there.
Yesterday I read a post referring to editing the rdl file and pitting in the width there. Check the
archives, search for <PointWidth>.
--
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
"Matt Williams" <anonymous@.discussions.microsoft.com> wrote in message
news:064501c49af9$4753e070$a401280a@.phx.gbl...
> Hi all,
> Firstly sorry if this is the wrong place to be posting
> this question ... it's the most appropriate newsgroup I
> could find.
> I'm creating a report using reporting services and I'm
> including a graph thats shows number of times something
> is accessed per day over time. Data coming from an SQL
> Server 2000 database. The dataset will only have data
> records in if the item is accessed ... i.e. if it is not
> accessed in a particular day then no record is present.
> I've therefore made the chart time-scale in the x-axis to
> ensure the scaling is correct. I've even included a union
> query to a blank set of data for todays date such that the
> graph will always extend up to the current date. The
> report has a parameter such that a filter can be applied
> to the data to show particular sections only (i.e.
> accesses by a particular individual) This all works fine
> for when there is a relatively large set of records post
> filter and the charts look good.
> The problem comes when there is only a small amount of
> data, say three points, over a large period of time. The
> scaling on the x-axis is correct and runs from the first
> data items date up until the current date. The problem is
> that the width of each column is massive and spans maybe 7
> days. It looks to me as though the width of the columns is
> calculated according to the number of the points (in this
> example 3) and not according to the number of "time-sacle
> points" maybe 100. The result is three hugely wide bars
> each spanning a weeks worth of dates with the left hand
> edge on the date of the actual event.
> I could try and introduce zero data into the data set for
> each of the intervening dates but didn't want to go to
> through the overhead of calculating all this unnecessary
> data ... that is after all the point of a time-scale axis.
> Would appreciate it is anyone has any experience in this
> or ideas I could try.
> Thanks in advance (and for reading this far!)
> Matt.|||Thank you very much for your help ... it gives me
somewhere to start ...
Matt.
>--Original Message--
>Matt,
>Just to let you know that there is an RS
group: .reportingsvcs. I think you have much better
change
>to get good help there.
>Yesterday I read a post referring to editing the rdl
file and pitting in the width there. Check the
>archives, search for <PointWidth>.
>--
>Tibor Karaszi, SQL Server MVP
>http://www.karaszi.com/sqlserver/default.asp
>http://www.solidqualitylearning.com/
>
>"Matt Williams" <anonymous@.discussions.microsoft.com>
wrote in message
>news:064501c49af9$4753e070$a401280a@.phx.gbl...
>> Hi all,
>> Firstly sorry if this is the wrong place to be posting
>> this question ... it's the most appropriate newsgroup I
>> could find.
>> I'm creating a report using reporting services and I'm
>> including a graph thats shows number of times something
>> is accessed per day over time. Data coming from an SQL
>> Server 2000 database. The dataset will only have data
>> records in if the item is accessed ... i.e. if it is
not
>> accessed in a particular day then no record is present.
>> I've therefore made the chart time-scale in the x-axis
to
>> ensure the scaling is correct. I've even included a
union
>> query to a blank set of data for todays date such that
the
>> graph will always extend up to the current date. The
>> report has a parameter such that a filter can be
applied
>> to the data to show particular sections only (i.e.
>> accesses by a particular individual) This all works
fine
>> for when there is a relatively large set of records
post
>> filter and the charts look good.
>> The problem comes when there is only a small amount of
>> data, say three points, over a large period of time.
The
>> scaling on the x-axis is correct and runs from the
first
>> data items date up until the current date. The problem
is
>> that the width of each column is massive and spans
maybe 7
>> days. It looks to me as though the width of the
columns is
>> calculated according to the number of the points (in
this
>> example 3) and not according to the number of "time-
sacle
>> points" maybe 100. The result is three hugely wide bars
>> each spanning a weeks worth of dates with the left hand
>> edge on the date of the actual event.
>> I could try and introduce zero data into the data set
for
>> each of the intervening dates but didn't want to go to
>> through the overhead of calculating all this
unnecessary
>> data ... that is after all the point of a time-scale
axis.
>> Would appreciate it is anyone has any experience in
this
>> or ideas I could try.
>> Thanks in advance (and for reading this far!)
>> Matt.
>
>.
>