Advanced Scripting for Sisense’s Simply Ask NLQ: Adding Currency Symbols to Pivot Tables
Sisense’s Simply Ask feature empowers users to create visualizations instantly by asking natural language questions, simplifying data exploration without needing technical expertise or manual widget configuration. This Natural Language Query (NLQ) capability accelerates insights generation without the need to manually create new Sisense widgets through the standard UI for each question or data relationship of interest. In a previously published article , it was discussed how to customize Simply Ask NLQ-generated widgets using scripting, providing multiple examples of basic modifications and more detail on the basic mechanics of scripting Simply Ask widgets. In this article a more complex example wil be shared to show how widgets can be customized in more complex conditional ways: programmatically adding currency symbols to specific data columns in pivot table widgets generated by Simply Ask.Modifying Simply Ask Natural Language Query Generated Widgets with Scripting
Sisense's Simply Ask feature empowers users to create visualizations instantly by asking natural language questions, simplifying data exploration without needing technical expertise or manual widget configuration. This Natural Language Query (NLQ) capability accelerates data exploration and insights generation without needing to manually create new Sisense widgets through the standard UI for each question or data relationship of interest. However, these automatically generated widgets differ from standard widgets in that they do not trigger typical widget scripting events, which poses a challenge for applying custom modifications. This limitation can hinder the application of standard dashboard and widget scripting customizations directly to these widgets. Fortunately, it is possible to modify NLQ-generated widgets programmatically by using JavaScript to manipulate the HTML of the fully rendered widgets without relying on Sisense-specific widget scripting events. This approach involves observing changes in the Simply Ask modal DOM and applying custom code when new widgets are rendered. This article explores how to implement such customizations, providing code examples and notes to help customize NLQ-generated Simply Ask widgets.Plugin - customLogoLink
Plugin - Set Custom Link Overriding default In Top Left Logo Icon Description This plugin allows changing the link of the top-left Sisense or white-labeled logo on the server's home page to a custom URL. Disabling the plugin will restore the logo link to its default behavior. Installation To install this plugin, download and unzip the attachment. Then drop the disableLogoLink folder into your plugins folder (/opt/sisense/storage/plugins). Enable the plugin on the Add-ons tab in the Admin section, wait for the plugin to build, and the plugin will be enabled. The plugin API and file management UI can also be used. Config Options - URL - URL navigated to on logo click - iframeLogoLinkChangeParentAddress - Boolean option allows logo click to change the parent address of a Sisense Iframe. It should be noted that JTD uses an iframe Notes This plugin is similar to the DisableLogoLink plugin which is documented here this page includes notes on creating plugins similar to this one. Learn more: https://docs.sisense.com/main/SisenseLinux/pre-post-build-plugins.htm https://academy.sisense.com/master-class-advanced-dashboards-with-plug-ins-and-scripts
Sisense Compose SDK
Sisense Compose SDK The Future is Composable In the fast-paced world of data-driven applications, developers are constantly seeking innovative ways to create customized data products that seamlessly integrate into their projects. Sisense has stepped up to the challenge with its groundbreaking Compose SDK, which has fundamentally changed the game. This powerful toolkit empowers developers to harness the full potential of Sisense components, enabling them to craft data products that are not only highly customized but also perfectly tailored to their applications. Traditionally, the embedded analytics market constrained developers within the boundaries of their chosen tools. However, Compose SDK breaks free from this limitation. Instead of merely embedding a third-party tool, developers now have the freedom to build bespoke data products by selecting from a rich library of Sisense components and seamlessly integrating them with a wide range of other tools within their Integrated Development Environment (IDE). A Practical Example In the world of UI design, consistency is key. Imagine your company has chosen the Material UI framework to style your application. It provides a sleek and uniform look, but now you need to populate Material UI tables with dynamic data from Sisense. Enter Compose SDK. With it, you can seamlessly merge Material UI's aesthetics with Sisense's analytical power. Start with Material UI tables that align perfectly with your design guidelines. Then, using the Compose SDK's ExecuteQuery component, you effortlessly infuse these tables with real-time data from Sisense.   The result? Stunning, data-rich components that not only adhere to your design principles but also deliver invaluable insights. Your users get the best of both worlds - visually appealing design and powerful data. Get Started with Compose SDK Excited to explore the potential of the Compose SDK? Good news - it's currently in Beta and available to all Sisense partners, ready for you to harness its capabilities today. To embark on your journey, you can begin by cloning the example repository from GitHub listed below. To install the necessary packages for your sample application, follow the user-friendly Compose SDK Quickstart Guide. Keep in mind that, for the time being, these packages are accessible on GitHub, requiring a personal access token for access. In conclusion, Sisense's Compose SDK represents a revolutionary leap forward in the world of data product development. It empowers developers to break free from the constraints of traditional embedded analytics and build highly customized, seamlessly integrated data products. With the Compose SDK, the possibilities are limitless, and the power to transform your applications lies at your fingertips. So, why wait? Dive into the world of Compose SDK, unleash your creativity, and revolutionize your data products today. Your applications will thank you for it! Credit: Sample Application initially developed by Sisense Senior Director of Field Engineering: Tom Linton Compose SDK -- Material UI Example Sisense Quick Start Guide Sisense GithubConditional Coloring with Scripting in Sisense Polar Bar Charts
Enhance Sisense polar charts with custom conditional coloring using a script. Apply dynamic colors to bars based on value thresholds (e.g., red for high, yellow for medium, green for low) to improve data visualization and interpretation.Demonstrating ComposeSDK Styling Of Sisense Dashboard Widgets
Demonstrating ComposeSDK Styling Of Existing Sisense Dashboard Widgets When developing ComposeSDK embedded applications, there are three principal techniques for embedding or converting an existing Sisense widget. Discussing each method, and some of the benefits associated with them: Direct Rendering existing Sisense Dashboard Widgets with the DashboardWidget Function: The expedited approach involves rendering an existing Sisense widget directly using the DashboardWidget function. While this method ensures swift integration, it does not allow all data options of the widget to be edited within ComposeSDK or defined as variables, one of the features of native ComposeSDK widgets. Nevertheless, customization of other parameters, including styling, remains feasible based on the props used. Mandatory parameters include the dashboard and widget ID, which allow the rendering of an existing Sisense widget. Widget Recreation using ComposeSDK to Create Native ComposeSDK widgets An alternative method, detailed in this article, entails recreating the widget as a native ComposeSDK widget by employing an existing Sisense widget as a template for a new not directly linked ComposeSDK widget. Though potentially more time-intensive, this approach yields a fully customizable native ComposeSDK widget. It stands independent of any specific widget or dashboard on the Sisense server, enabling complete independence to changes made to the original dashboard or widget, or complete deletion of the widget used as the model. Loading Widget Metadata with the useGetWidgetModel function Leveraging the useGetWidgetModel ComposeSDK function provides a middle option, detailed in this article. It allows automating the return of widget metadata from an existing Sisense widget, facilitating dynamic modifications within ComposeSDK. This method balances somewhat the autonomy of entirely recreating a widget as a native ComposeSDK widget and rendering a widget as a Dashboard Widget. In this article, we will demonstrate and discuss the DashboardWidget rendering feature, a powerful capability within ComposeSDK that allows the embedding and rendering of existing widgets. The focus will be on exploring the large number of styling options provided by this feature. Among the properties of a DashboardWidget component, the styleOptions parameter determines the styling. This parameter accepts a DashboardWidgetStyleOptions object, which includes a large number of parameters documented in detail below and in this documentation page. To demonstrate a practical implementation of using these styling options, the following code example showcases a Compose DashboardWidget utilizing all documented parameters within the DashboardWidgetStyleOptions object. import { DashboardWidget } from "@sisense/sdk-ui"; export function ComposeSDKChart(props) { let styleOptions = { "backgroundColor": "lightblue", "border": true, "borderColor": "green", "cornerRadius": "Large", "shadow": "Dark", "spaceAround": "Large", "header": { "hidden": false, "titleAlignment": "Center", "backgroundColor": "lightblue", "titleTextColor": "blue" }, "height": 200, "width": 200 }; return (<DashboardWidget widgetOid={'65ab8958857ff900335db870'} dashboardOid={'65ab8948857ff900335db86e'} styleOptions={styleOptions} title={"My Chart"} />) } The final result in this example is the widget below, showing the effect of this styling in this specific case: Discussing each style option individually, including the type of parameter accepted by each style setting. backgroundColor: Type: string - Specifically a string containing an HTML color Description: This option sets the background color of the widget. In the example, the background color is set to "lightblue," a preset standard HTML color code that is equal to #ADD8E6 as a hex color code. border: Type: boolean Description: This option determines whether the widget container has a border or not. In the example, the border is enabled with the value set to true, adding a defined boundary to the widget that can be styled. borderColor: Type: string - Specifically a string containing an HTML color, Description: Specifies the color of the widget container's border. cornerRadius: Type: "Large" | "Medium" | "Small" - Three specific options set as specific strings Description: Defines the corner radius of the widget container, similar to the CSS property of the same name, allowing for customization of the widget container's shape. header: Type: object- Contains various parameters that control the styling of the header and title. Description: A style object to customize the widget header. It includes options like background color, divider line toggle, divider line color, header visibility toggle, title alignment, and title text color. header.backgroundColor: Type: string - Specifically a string containing an HTML color Description: Sets the background color of the widget header. header.dividerLine: Type: boolean Description: Controls the visibility of the divider line between the widget header and the chart. header.dividerLineColor Type: string - Specifically a string containing an HTML color Description: Specifies the color of the divider line, if visible. header.hidden Type: boolean Description: Toggles the visibility of the header and title. header.titleAlignment Type: "Left" | "Center" - Two specific options set as specific strings Description: Alignment of the title within the header. The example centers the title using the value "Center. header.titleTextColor Type: string- Specifically a string containing an HTML color Description: Specifies the text color of the header title. height: Type: number - Number in pixels Description: Sets the total height of the widget in pixels. If not explicitly set will use the height of the container. width: Type: number - Number in pixels Description: Sets the total width of the widget in pixels. If not explicitly set will use the width of the container. shadow: Type: "Medium" | "Light" | "Dark" - Three specific options set as specific strings Description: Defines the shadow level of the widget container, similar to the CSS style of the same name. Effects styling only when spaceAround is defined. spaceAround: Type: "Large" | "Medium" | "Small" - Three specific options set as specific strings Description: Specifies the space between the widget container edge and the chart. With these styling options, one can modify the visual styling of a ComposeSDK DashboardWidget, enabling a large amount of style customization for an otherwise unchanged existing dashboard widget selected from an existing dashboard on the Sisense server. While ComposeSDK DashboardWidgets may not offer the extensive modification capabilities available to native ComposeSDK widgets, such as the ability to completely modify all of widget data options, they still provide a large degree of extensive visual customization through the use of styling options. Share your experience in the comments!Plugin - Custom Coloring of Bar and Column Chart Type Widgets (barChartCustomColor)
Plugin - Custom Coloring of Bar and Column chart-type widgets This plugin modifies the color scheme of bar and column chart-type widgets to match the current color palette of the dashboard. Installation To install this plugin, download and unzip the attachment. Then drop the barChartCustomColor folder into your plugins folder (/opt/sisense/storage/plugins). Enable the plugin on the Add-ons tab in the Admin section, wait for the plugin to build, and the plugin will be enabled. The plugin API can also be used, as well as the file management UI. Notes When a bar or column chart type Sisense widget is very straightforward and focused and does not have multiple values, breaks by's, categories, or conditional coloring, it can sometimes appear relatively mono-color but still be the most effective way to quickly visually communicate important data. While Sisense includes numerous powerful and varied methods and options to change widget styling and includes functionality to set bar color to vary based on the bar value, it sometimes may be desired to use the dashboard palette to vary the bar colors independently of bar value and guarantee a colorful widget regardless of the data and current filter state of the current dashboard and widget. This plugin allows the current dashboard palette to be used in this manner in a simple and quick matter and allows this to be applied quickly to an entire dashboard at once. Linking the widget coloring to the dashboard palette allows this color scheme to be quickly changed in the native UI without modifying individual widgets. This plugin also allows using a color dictionary config option to modify the coloring of a value in all widgets this plugin is enabled in. This plugin uses the processresult widget event to modify the color parameter of each bar or column in a series. The dashboard palette is retrieved using the getPalette() function within the dashboard style parameter, which returns the current dashboard palette. This can be used in other plugins and scripts. dashboard.style.getPalette() This plugin is enabled via dashboard or widget scripts that set the widget parameter changeColor to true for a widget. For a widget script, this is as simple as adding this one line to the script: widget.changeColor = true; Adding this three-line dashboard script will set this parameter to true for all widgets in the dashboard: dashboard.on('widgetinitialized', function (_, dashObj) { dashObj.widget.changeColor = true; }); This can be modified with custom logic as needed, for example, based on widget title or ordering in the dashboard. If this option is enabled for a type of widget this plugin does not apply to, it will simply be ignored by the plugin. This plugin ignores widgets with Break By's or Date Categories, or widgets that are not bar or column chart-type widgets. The color scheme is based on the current dashboard palette. Changing the dashboard palette using the standard Sisense palette UI will result in a matching change of all widgets in the dashboard where this plugin is enabled. The config file of this plugin includes an option called colorDictionary, this can be used to override the standard palette color for all instances this value appears as the bar category, for all widgets this plugin is enabled for, regardless of the dashboard palette. Any standard HTML color naming format may be used in the dictionary config value. For example, if the colorDictionary is set to: colorDictionary : { 'Bikes': '#AA6C39', 'ABC' : 'yellow' } Then the ABC bar or column will be yellow in the widget, regardless of the current widget palette or positioning. This will apply to all widgets where this plugin is active. This plugin can be used as a basic template for your own Sisense plugins that run on a specific widget or dashboard event, that can be enabled or disabled via widget script, and that includes a configuration file to modify settings without modifying code files. How did the plugin work for you? What other type of plugin are you looking to learn more about? Let me know in the comments!
Sankey Diagrams With Plot.Ly In Periscope
Sankey diagrams are a very easy-to-digest way to look at the flow of objects between different points. The example we will walk through here is a basic Sankey overview, leveraging data from a fictional gaming company. The emphasis of this post is focused on how to get your pandas dataframe data into something that can be used by plot.ly in a Sankey diagram. The Sankey set up itself is modeled after the first example in the documentation here. If you want to look at different Sankey plots from a visualization standpoint, plot.ly has a great library or resources. I've linked some of those resources at the bottom of this post for further reference Without any further ado, here is the annotated code! Note that you would need the Periscope Python and R integration to use our Plot.ly integration. Our SQL output is a dataframe that contains 3 columns: Source - The source that the user came from Platform - The user's platform of choice Count - The number of users for that source-platform combination The SQL to generate this is noted below for reference select source , platform , count(*) from users group by 1 , 2 And here is the Python 3.6 code, annotated up top where we set up our pandas dataframe df for use in plot.ly. # SQL output is imported as a dataframe variable called 'df' import pandas as pd import plotly.plotly as py # First, we get a list of all of sources, remove duplicates, and make this a list sources = df["source"].drop_duplicates().tolist() # Then, we get a list of all of platforms (our targets), remove duplicates, and make this a list platforms = df["platform"].drop_duplicates().tolist() # Finally, create a list of all our nodes. We will use this for giving an id to each node for plot.ly all_nodes = sources + platforms # Keeping the size of our dataframes, this would be useful for applying the same color for each "node" and "link" of our sankey diagram, if we so choose to do so n = len(all_nodes) n2 = len(df["source"]) # Create a dataframe that has all of the node ids. We will join this to the original data frame to reference later df1 = pd.DataFrame(all_nodes, columns = ['node']) df1 = df1.reset_index() df2 = pd.merge(pd.merge(df, df1, how = 'inner', left_on = "source", right_on ="node"), df1, how = 'inner', left_on = "platform", right_on ="node", suffixes = ('_source','_target')) # Setting up the data in the plotly "data" argument. # The nodes are described in the "node" dictionary (these are the vertical rectangles in the diagram) # The links are described in the "link" dictionary. These have 3 attributes, the "source" (the index of the node they start at), the "target" (the index of the node they end at), and the "value" the thickness of the band. Additional attributes, such as color can also be specified. data = dict( type='sankey', node = dict( pad = 15, thickness = 20, line = dict( color = "#435951", width = 0.5 ), label = all_nodes, color = ["#84baa6"] * n ), link = dict( source = df2["index_source"], target = df2["index_target"], value = df2["count"], color = ['#bdf9e5'] * n2 )) # Setting up the layout settings in the "layout" argument layout = dict( title = "An Example Sankey Diagram", font = dict( size = 12 ) ) fig = dict(data=[data], layout=layout) # Use Periscope to visualize a dataframe, text, or an image by passing data to periscope.table(), periscope.text(), or periscope.image() respectively. periscope.plotly(fig) Plot.ly Sankey Diagrams - 1 Plot.ly Sankey Diagrams - 2 Plot.ly Sankey Diagrams - 3Confidence Interval Printout - Python
Let's say we want a printout of our confidence interval for an entire sample (Note, if you're looking for a visual of a confidence interval over time, check out the post here!) The solution here requires Periscope Data's Python/R Integration as we'll be using Python's Scipy package and matplotlib text charts. Our SQL output is a dataframe (df) with one column of a list of values. Notice that the function created in the Python snippet also takes in the following optional parameters for added customization: interval: the width of the Confidence Interval. Default is set to 0.95 (95%) method: whether we use the t distribution or a z/normal distribution to calculate the confidence interval. Default is set to a 't' distribution. We recommend verifying that your data is normally distributed before using the z distribution statistic. The snippet is written in Python 3.6 # SQL output is imported as a pandas dataframe variable called "df" import pandas as pd from scipy import stats import math import matplotlib.pyplot as plt import numpy as np #Function: CI_prinout, a function that outputs a number overlay expressing a sample's Confidence Interval #Inputs: a dataframe with one column of values. Optional paramater interval for the size of the confidence interval (default is 0.95). Option parameter method that specifies whether the confidence interval will be calculating using the t distribution or a z/normal distribution. #Outputs: a matplotlib text chart with the % confidence interval and the lower and upper bounds def CI_printout(series, interval = 0.95, method = 't'): mean_val = series.mean() n = series.count() stdev = series.std() if method == 't': test_stat = stats.t.ppf((interval + 1)/2, n) elif method == 'z': test_stat = stats.norm.ppf((interval + 1)/2) lower_bound = mean_val - test_stat * stdev / math.sqrt(n) upper_bound = mean_val + test_stat * stdev / math.sqrt(n) fig = plt.figure() plt.axis('off') plt.gcf().set_size_inches(8, 2) plt.xticks([]) plt.yticks([]) plt.text(.5, .75, str(round(interval * 100))+ '% Confidence Interval', fontsize=25, color='black', ha='center') plt.text(.5, .35, str(round(lower_bound[0],2)) + ' to ' + str(round(upper_bound[0],2)), fontsize=25, color='black', ha='center') return plt # Use Periscope to visualize a dataframe or an image by passing data to periscope.output() periscope.output(CI_printout(df)) For the R equivalent of this post, check out the community page here! Found this useful? Let us know in the comments!
