Improved hints, by implementing the hover service

Author: Joan Giner

src/extension.ts

@@ -6,6 +6,8 @@ import {
 } from 'vscode-languageclient/node';
 import { DocumentationGenerator } from './generator-service/dataset-descriptor-documentation';
 import { DatasetUploader } from './uploader-service/dataset-descriptor-uploader';
+import { HintsService } from './hints-service/hints-service';
+
 
 
 let client: LanguageClient;
@@ -29,6 +31,18 @@ export function activate(context: vscode.ExtensionContext): void {
             });
     }));
 
+    context.subscriptions.push(
+        vscode.languages.registerHoverProvider(
+            'dataset-descriptor', {
+                provideHover(document, position, token) {
+                    let hints = new HintsService();
+                    let content = hints.populateHints(document, position);
+                
+                    if (content != "empty") return new vscode.Hover(new vscode.MarkdownString(content));
+                    return null;
+            }
+      }));
+
     // Here we register the HTML generation service
     context.subscriptions.push(vscode.commands.registerCommand('datadesc.generateDocumentation', async () => {
         await generatorHTMLService(context);

src/hints-service/hints-service.ts

@@ -0,0 +1,155 @@
+/******************************************************************************
+ * Copyright 2022 SOM Research
+ * This program and the accompanying materials are made available under the
+ * terms of the MIT License, which is available in the project root.
+ ******************************************************************************/
+import { TextDocument } from 'vscode';
+
+
+ 
+ 
+ export interface Hints {
+     // Load the Abstract Syntax Tree of the .descML active file
+     populateHints(Declaration : string | TextDocument, position: any) : string | undefined;
+ }
+ 
+ /**
+  */
+ export class HintsService implements Hints {
+ 
+ 
+     constructor() {
+       
+    }
+ 
+     populateHints(document : TextDocument, position: any) : string {
+        const wordRange = document.getWordRangeAtPosition(position);
+        const word = document.getText(wordRange);
+        switch (word) {
+            case "Metadata":
+                return `## Metadata
+In this section authors are expected to provide the metadata of the dataset
+                `
+            case "Dates":
+                return "Set the release, published and last updated date in DD-MM-YYYY format";
+            case "Citation":
+                return "Set the citations of the dataset";
+                // Description
+            case "Description":
+                return `### Description
+In this section authors are expected to provide a description of the dataset
+
+#### Purposes:
+For what propose was the dataser created?
+
+#### Tasks:
+For what tasks this dataset is inteded for
+
+#### Gaps:
+Was there specific gap that needed to be filled? Please provide a description
+                            `;
+                            case "Purposes":
+                                return "For what propose was the dataser created?";
+                            case "Tasks":
+                                return "For what tasks this dataset is inteded for";
+                            case "Gaps":
+                                return "Was there specific gap that needed to be filled?\nPlease provide a description";
+            case "Tags":
+                return "Set the tags separated by a whitespace";
+            case "Areas":
+                return "Set the areas separated by a whitespace";
+        // Distribution
+            case "Distribution":
+                return `## Distribution
+In this section authors are expected to indicate the distribution of the dataset
+
+### Licenses: 
+Set the licence of the dataset.
+
+## Rights stand-alone:
+Choose the level of distribution of the stand-alone data.
+
+## Rights of the models: 
+Choose the level of distribution of the models trained with the data.
+                        `;
+            case "Licenses":
+                return "Indicate in 'others:' if any other policy is applied to the data";
+        // Applications
+            case "Applications":
+                return `## Applications 
+In this section authors are expected to indicate the recommneded and non-recommneded uses of the dataset
+
+### Benchmarking
+If the dataset have been used in the past, authors are expected to indicate the benchmarking results
+Models names, and results should be provided (accuracy, precision, recall, F1-score)
+                `;
+        // Authoring
+        case "Authoring":
+            return `## Authoring 
+In this section authors are expected to indicate who created the dataset and who funded the dataset
+Please provide information about the organization grating the work
+
+### Maintenance
+Who maintains the dataset, but also the contribution policies, if theere is any erratum, and the data life cycle should be informed in this chapter
+            `;
+            case "Funders":
+                return "Who founded the creation of the dataset?\n2 - If is there any associated grant, please provide the number and the name of the grantor and the gran name and number \n Set a `_` or a `-` as a white spaces in the name e.g: 'John_Smith'? ";
+            case "Authors":
+                return "Who is the author of the dataset?";
+            case "Maintainers":
+                return "Who maintan the dataset? How can be contacted?";
+        // Composition
+            case "Composition":
+            return `## Composition 
+Please provide information about the composition of the dataset. The type of files (data instances), it's number, and information regarding attributes 
+
+### Statistics 
+A set of statistics can be provided for each attribute and at a data instance level. Please provide only the statistics that are relevant for the specific dataset use case.
+
+### Consistency rules
+The Consistency rules can be expressed following OCL. OCL is a language for expressing constraints on models. It is based on the Object Constraint Language (OCL) defined by OMG. OCL is a language for expressing constraints on models. It is based on the Object Constraint Language (OCL) defined by OMG. \n
+            
+            `;
+         // Provenance
+         case "Provenance":
+            return `## Provenance 
+In this section authors are expected to fill information about the process applied to create the dataset 
+
+### Curation Rationale 
+This explanation intend to be a shor and comprhensive enumartion of the processes applied over the data, and to provide specific use-case details for this dataset
+
+### Gathering 
+How the dataset has been gathered? Who gathered the dataset? Which are the sources of the data?
+
+### Annotation 
+How the dataset has been annotated? Who annotated the dataset? Which are the infrastructure used to annotate the data?
+
+### Data preparation 
+Indicate the process done to prepare the data, and it's type
+            
+            `;
+        // Social Concers
+        case "Concerns":
+        return `
+## Social Concerns
+In this section authors are expected to fill information about the social concerns of the data. Is expected to inform 4 types of social concerns \n
+
+### Bias concers 
+Whether the dataset may be biased against a specific social group 
+
+### Representativeness concerns 
+Whether the dataset could misrepresent any specific social group
+
+### Sensitivity concerns 
+Does the dataset contains data that can offend a social group?
+
+### Privacy Concerns 
+Is there any privacy concerns on the data?
+        
+        `;
+           default:
+            return "empty"
+        }
+     }
+ }
+