Forum Discussion
line chart visual bug
| DATE_FROM | DATE_TO | DATE | COMPRESSED_RANGE |
|---|---|---|---|
| 01.01.2020 | 01.01.2020 | 2020-01-01:null | |
| 01.01.2020 | 02.01.2020 | 2020-01-01:null | |
| 01.01.2020 | 03.01.2020 | 2020-01-01:null | |
| 01.01.2020 | 04.01.2020 | 2020-01-01:null | |
| 01.01.2020 | 05.01.2020 | 2020-01-01:null | |
| 02.01.2020 | 02.01.2020 | 2020-01-02:null | |
| 02.01.2020 | 03.01.2020 | 2020-01-02:null | |
| 02.01.2020 | 04.01.2020 | 2020-01-02:null | |
| 02.01.2020 | 05.01.2020 | 2020-01-02:null | |
| 02.01.2020 | 02.01.2020 | 02.01.2020 | 2020-01-02:2020-01-02 |
| 03.01.2020 | 03.01.2020 | 03.01.2020 | 2020-01-03:2020-01-03 |
| 04.01.2020 | 04.01.2020 | 2020-01-04:null | |
| 04.01.2020 | 05.01.2020 | 2020-01-04:null |
| date |
|---|
| 01.01.2020 |
| 02.01.2020 |
| 03.01.2020 |
| 04.01.2020 |
| BO_ID | ATTR_0 | DATE_FROM | DATE_TO | COMPRESSED_RANGE |
|---|---|---|---|---|
| 0 | C | 02.01.2020 | 02.01.2020 | 2020-01-02:2020-01-02 |
| 0 | I | 03.01.2020 | 03.01.2020 | 2020-01-03:2020-01-03 |
| 0 | M | 04.01.2020 | 2020-01-04:null | |
| 1 | Z | 02.01.2020 | 2020-01-02:null | |
| 4 | N | 06.01.2020 | 2020-01-06:null | |
| 5 | C | 01.01.2020 | 2020-01-01:null | |
| 7 | U | 01.01.2020 | 2020-01-01:null |
Your pasting didn't work because the forum doesn't know what to do with thead and tbody tags. Removing these seems to work.
Now - what's the expected outcome based on this data?
thanks,
this is the expected outcome
it depitcts the count of "covering ranges" from `fact_table_range` for each date from `calendar_daily` ie something like count of rows returned from this select:
select * from calendar_daily inner join fact_table_range on calendar_daily.date >= DATE_FROM and (DATE_TO is null or calendar_daily.date <= DATE_TO)
for given calendar_daily.date. This "logic" is done via the `dimension_table_range`, which translates any date from `calendar_daily` to all possible "covering ranges" from `fact_table_range`, as seen here.
It works fine with such small dataset, the problem starts when i use greater dataset with three years worth of days in `calendar_daily` and millions of rows in `fact_table_range` (the size of `dimension_table_range` is determined by unique tuples of (`date_from`, `date_to`) in the fact table, the cap for dataset tracking three years is somewhere around 210K rows). The outcome for a dataset of that size is
as shown my original post
your 'dimension_table_range' table does not look happy. It is clearly not a dimension table. I think you want to review and refactor your data model before you work on the visual issues.
