Dashboards & looks
Building reports & visualizations in Looker
- 1,772 Topics
- 4,200 Replies
I’ve set a PDT to rebuild on a datagroup_trigger, which itself is a check built in RedShift to switch to a ‘1’ value when the data is ready from our ETL process. I’d like to schedule the emailing of a report using this PDT as a data source, not at a specific time, but rather as soon as the PDT has successfully rebuilt so the email is going out with fresh data as soon as it’s available. Question:If I map both the PDT rebuild and the email schedule to the same datagroup_trigger, will the email wait until the PDT has rebuilt, or will both just trigger at the same time? Is there a clever way of doing this? Many thanksScott
This works:Dashboard, two panels, both with a single queryEdit Dashboard, Edit first panel, edit an axis to hide the label, Save.Edit the second panel, hide an axis, Save.Save Dashboard and done. All good. This doesn’t work:Dashboard, two panels, one with a single query, one with a merged query.Edit Dashboard, Edit panel with single query, edit an axis to hide the label, Save.Edit the second panel, with the merged query (now it opens a new tab/window, where I think the problem comes in), hide an axis, Save.I am now in a new tab, with the original Dashboard, but the first, single query panel I edited is now reverted. Basically if I want to edit different panels, some single, some merged query, I have to first edit the single ones, save the dashboard, then edit the dashboard again and do the merge query panels one by one. Not the end of the world, but still annoying when doing a last cleaning up of a dashboard.Could Looker just let me edit a merged query panel in a “popup” on the same pa
I am using the radial gauge marketplace visualization and I would love to get it to work how I am desiring it to - which is to have the first measure value represent the needle value (as it currently does) and have the second measure represent the max of the range for the radial gauge (which it does not currently do). Best I can do is add a target source and have it be the 2nd measure, but I’m curious why when the visualization asks for 2 measures by default it doesn’t just use the 2nd measure for the max of the range? Moreover there is a max range override setting, but it does not allow a measure value, just a static hard coded number it seems. Has anyone had any luck w/ having this max range be represented by the 2nd measure in this viz?
Hello all,As an actuary I quite often look at the same metrics, displayed in the same format, but split under different dimensions. Therefore one of my prefered dashboard is showing 10 times the same tile but just changing the dimensions. Currently I have 2 big issues to work correctly :For no reason sometimes the displayed format of my tiles changes, and I need to correct it 10 times to have a dashboard easy to read again. If I want to change something in the dashboard, adding/removing a metric, changing a title, I need to do it in each single tile, which is quite time consumingI would love to work in the following way:I create a "template" where I would setup all the metrics I want to see + their format (size, name, colour, etc.) but no dimension I create a dashboard with several tiles linked to the template. When I choose to use the template, I can only select the dimensions I want to use + their format - everything related to metrics is locked Everytime I change the template it au
After a fair amount of work, I managed to use pivots to format my data in such a way that it would be far more efficient for use in my custom visualisation. Custom visualisations receive a JSON-formatted chunk of data from the server, typically organised in rows.Whilst doing this legwork, I tried to see what would come from my efforts by using the ‘Download’ option on the Look, then viewing the resulting JSON. Everything looked fine, with a wonderfully nested structure matching the pivots on the table in the Look UI.However, once I checked to see what my custom visualisation was receiving, I was shocked to discover that that data looked nothing like the downloaded sample; most of the nested structure was absent, replaced with flat row sets with utterly unexpected value naming.Is there any way to get JSON sent to my visualisation that matches the structure shown when i choose to download a sample??
We are in the process of creating some Looker System Activity Dashboard metrics, unfortunately Looker doesnt expose the Model to create custom dimensions / groups. We are in the process of identifying the Min (Created Date) & Max(Last Query Date) for each Dashboard as shown here, any suggestion is really appreciated. Min(Created Date) , Max(Updated Date)
I have a plot with a reference line with a range, it seems that data points that fall completely under that range are no longer hoverable with the mouse. Apparently because the reference line is rendered on top of the data points.Is it possible to configure the reference line to show up on the background of the chart?
Using Dashboard-next, while downloading the dashboard to PDF the filter names of the dashboard are inproper
Using the 'dashboards-next' type, while exporting the dashboard into PDF, the dashboard filters not showing the 'title' Name of the filter but display as - 'name:' Ex: Original displayed name is 'Contract ID' in the dashboard after downloading to PDF it display the name as 'contract_id_flt'Is this a bug in the looker ?My current looker version is 21.8.24Did anyone has faced this issue and is there a way to resolve this ?
Many of us are using Persistent Data Tables (PDTs) and the new Incremental PDTs. Our Company would like to be able to display how “fresh” the data is from a PDT without resorting to using the see_system_activity value, which requires a Standard license to view.So we would like to see a widget added to Looker that shows the date/time stamp of when a PDT was last updated/refreshed.
Hello Community, I am highlighting Countries with color and tooltip details, But Where us Map Visualization is not at all reflected with color pattern and all ? Would you please suggest me what necessary steps to fallow? My Condition is : USA, MEXICO, CANADA AND Russian Federation Countries with stock outs as a measure. Thanks in Advace,Mamatha
Avoid installing chromium 92 or later on customer-hosted Looker instances to prevent rendering issues
Chromium is required for scheduling, sending or downloading Looks and dashboards in data formats like PDF or PNG. On Looker-hosted customer instances, the chromium installation is managed by the Looker team. However, customers need to manage the chromium installation themselves on customer-hosted Looker instances. Recently, we have discovered a bug with the latest chromium 92 beta version when rendering Looks or Dashboards to PDF format when the Looks or Dashboards contain any map visualisations. The rendered PDF does not contain the map visualisation present in the Dashboard or the Look and instead contains a blank space in place of the map. In the past, Looker has generally recommended installing the latest version of chromium. However, because of the issue described above, we are now recommending that customers do not install Chromium 92 or later until this issue is resolved. The stable release of chromium 92 is scheduled for July 20th, 2021 as per https://chromiumdash.appspot.com
Right now I have a filter for messages in a certain category.The problem is our source data provides categories and boards by their uid.So, right now my filter looks like this:This translates to /root/top/Chat, where Chat(314) is the category.Is it possible to change the above to this?Category | Is Equal To | <dropdown of options>And if the user chooses Chat in the dropdown, the filter actually does ancestor path contains /1/3/314
Hi,Can anyone tell me whether the field widths of the Details drill through are alterable? If not, can anyone tell me what determines the auto-sizing behaviour - are there certain characters that cause line breaks within the cell?(I’m experiencing different behaviours in text-rich columns; some opt for tall skinny others short wide)Thanks in advance,Casey.
Hello,I wanted to share a small issue I notice quite often while building dashboard next.The timeframe filter (which looks great) always take into consideration the current day or period selected. This is really annoying because when you don’t have data available for the current day, it shows an empty data point. For that reason, most of the time, I have to revert back to a basic advanced filter when I can select “past 7 complete days” as a default value, which is a shame because the Timeframe filter is really nice and much more user friendly.I think a simple checkbox “include current period” would do the work here :)Antho
Hi,I am trying to create a table that shows how the percentage split between refundable and non-refundable bookings has changed over the last quarter. What I am struggling with is creating a dimension that shows YTD data per week over the last 12 weeks. I have created YTD data before but creating a dimension that does this for each of the previous 12 weeks is proving a challenge, any help here would be much appreciated. Thanks, Dan
Hi there, I am getting error whenever I try to edit the settings on a Donut Multiples tiles. It crashes and any work not saved on the dashboard is lost. After checking the sql, I find one of the alias z__rank is missing a underscore. Is this a known bug? any workaround? Thanks! SELECT * FROM (SELECT *, DENSE_RANK() OVER (ORDER BY z___min_rank) as z___pivot_row_rank, RANK() OVER (PARTITION BY z__pivot_col_rank ORDER BY z___min_rank) as z__pivot_col_ordering, CASE WHEN z___min_rank = z__rank THEN 1 ELSE 0 END AS z_is_highest_ranked_cell FROM (
Let’s say i set my dashboard filter to a certain date , i want all my looks to bring in the past 6 months data from that dateIs there a way to do this in looker. I tried setting up 2 date filters on dashboard and tried to select filters to update by another date but i dont think it works.
Hi all, I’m trying to create a column similar to the one on the right, QoQ. I have 10 metrics that I would like to see QoQ performance. The only way I can see doing it is creating a calculated field using offset for each metric, but that’s going to take a while. Wondering if there’s a way to do it all at once.
Already have an account? Login
Login to the community
No account yet? Create an account
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.