On Mon, May 8, 2017 at 7:51 AM, Justin Stay <[email protected]> wrote:
> Daniel,
>
> I don't quite follow your code in your last post. Could you explain
> conceptually what you are doing? Where and how are dt and mergedDatatable
> defined?
>
> I think what you are saying is that for the worst case where you had N
> lines, each with say M number of x (domain) values, you'd have N columns,
> M*X rows, and most of the values would be NULL? That seems very
> inefficient.
>
Yes, it sounds like you do understand what my code was intended to make
more explicit. You would need to wrap that loop in another loop over the N
datatables. Before that, just create a new empty datatable with your N + 1
columns (don't forget the domain column).
But instead of filling with nulls, you could leave values undefined, as
long as you make each row the same length by adding a null value in yet
another column.
Yes this is inefficient for space, but you may want to merge rows that have
the same domain value.
I'm thinking of adding a 'series' role column that would allow you to do
what you want more directly. This would be the equivalent of putting each
data value in a separate column.
If the data size inefficiency is a problem, there is another way you could
do this, though you wouldn't get a legend of N series. You could simply
append all your rows in one table, but for each row, add a 'style' role
column to specify the color of the line. Also, between each set of rows,
add one extra row with a null value which will break the line at that point
if you don't interpolateNulls.
> On Sunday, May 7, 2017 at 1:02:51 PM UTC-4, Daniel LaLiberte wrote:
>>
>> Ah, I see. Since you want n lines, one for each of your datatables,
>> then, yes, you need to put them in different columns. It would not be a
>> join in that case. Just add more rows to the one datatable. For each row
>> of data, just use nulls for all the columns where you don't have a value.
>> It might be good enough to leave null values undefined, but I suspect the
>> rows have to be the same length. You could ensure that by just adding a
>> last column.
>>
>> for (i = 0; i < numRows; i++) {
>> var row = [];
>> row[0] = dt.getValue(i, 0);
>> row[i] = dt.getValue(i, 1);
>> row[numCols] = null;
>> }
>> mergedDatatable.addRow(row);
>>
>>
>> On Sun, May 7, 2017 at 12:54 PM, Daniel LaLiberte <[email protected]>
>> wrote:
>>
>>> Hi Justin,
>>>
>>> I'm not sure I understand what you mean by "This approach requires the
>>> dt1columns argument is different for each joining of an additional data
>>> set.", but that sounds wrong. You could join the same columns in different
>>> datatables each time. The result of each join is a new datatable, so you
>>> could do a loop like this:
>>>
>>> dtJoin = dtables[0];
>>> for (i = 1; i < dtables.length; i++) {
>>> dtJoin = google.visualization.data.join(dtJoin, dtables[i],
>>> joinMethod, joinKeys, 1, 1);
>>> }
>>>
>>> On Sun, May 7, 2017 at 11:28 AM, Justin Stay <[email protected]> wrote:
>>>
>>>> I have many (2+, number unknown at time of generation) of 1D data sets
>>>> that I would like to plot together. This data is being queried from a
>>>> MySQL database. The generation of these datasets are captured through N
>>>> queries (where there are N lines to plot on the line chart). For example:
>>>>
>>>> Dataset #1: MySQL query results a data set with M1 (x,y) data points
>>>> Dataset #2: M2 (x,y) data points
>>>> etc . . .
>>>>
>>>> Such that you end with some number of datasets that are not the same
>>>> length nor share the same x values.
>>>>
>>>> I have seen a few questions/examples that seem to address this question
>>>> using the join method (https://developers.google.com
>>>> /chart/interactive/docs/reference#join). This approach requires the
>>>> dt1columns argument is different for each joining of an additional data
>>>> set.
>>>>
>>>> Is this approach the only method for solving this problem? Is there a
>>>> better approach? No other plotting tool I've used requires that each data
>>>> set be the same size.
>>>>
>>>> Thanks!
>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "Google Visualization API" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to [email protected].
>>>> To post to this group, send email to [email protected].
>>>> Visit this group at https://groups.google.com/grou
>>>> p/google-visualization-api.
>>>> To view this discussion on the web visit https://groups.google.com/d/ms
>>>> gid/google-visualization-api/a6ee493c-11db-4303-9268-ac56e1
>>>> cdefb5%40googlegroups.com
>>>> <https://groups.google.com/d/msgid/google-visualization-api/a6ee493c-11db-4303-9268-ac56e1cdefb5%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>> .
>>>> For more options, visit https://groups.google.com/d/optout.
>>>>
>>>
>>>
>>>
>>> --
>>> Daniel LaLiberte <https://plus.google.com/100631381223468223275?prsrc=2>
>>> [email protected] 5CC, Cambridge MA
>>>
>>
>>
>>
>> --
>> Daniel LaLiberte <https://plus.google.com/100631381223468223275?prsrc=2>
>> [email protected] 5CC, Cambridge MA
>>
> --
> You received this message because you are subscribed to the Google Groups
> "Google Visualization API" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To post to this group, send email to google-visualization-api@
> googlegroups.com.
> Visit this group at https://groups.google.com/
> group/google-visualization-api.
> To view this discussion on the web visit https://groups.google.com/d/
> msgid/google-visualization-api/945f7b11-6ee1-4006-a88d-
> d8a409438472%40googlegroups.com
> <https://groups.google.com/d/msgid/google-visualization-api/945f7b11-6ee1-4006-a88d-d8a409438472%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>
> For more options, visit https://groups.google.com/d/optout.
>
--
Daniel LaLiberte <https://plus.google.com/100631381223468223275?prsrc=2>
[email protected] <[email protected]> 5CC, Cambridge MA
--
You received this message because you are subscribed to the Google Groups
"Google Visualization API" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/google-visualization-api.
To view this discussion on the web visit
https://groups.google.com/d/msgid/google-visualization-api/CAOtcSJM5spa41FU4DyYmK_CaYHxU04gRTk2qRR3H2KGR72pb-Q%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
