[#944] Implement authorship analysis (#2140)

A line is credited to the author who last modified it.

Another author might have written the line initially and the current
author only modified it slightly. In such a case, the current author
gets credited for work that is not entirely done by him/her.

Let's analyze how similar a line is as compared to its ancestor lines
(previous versions of the line) and give full or partial credit to the
last author based on the analysis.

build.gradle

@@ -196,7 +196,7 @@ def serveTestReportInBackground = tasks.register('serveTestReportInBackground',
     workingDir = 'build/serveTestReport'
     main = mainClassName
     classpath = sourceSets.main.runtimeClasspath
-    args = ['--config', './exampleconfig', '--since', 'd1', '--view']
+    args = ['--config', './exampleconfig', '--since', 'd1', '--view', '-A']
     String versionJvmArgs = '-Dversion=' + getRepoSenseVersion()
     jvmArgs = [ versionJvmArgs ]
     waitForPort = 9000

docs/ug/cli.md

@@ -16,16 +16,36 @@ The command `java -jar RepoSense.jar` takes several flags.
 **Examples**:
 
 An example of a command using most parameters:<br>
-`java -jar RepoSense.jar --repos https://github.com/reposense/RepoSense.git --output ./report_folder --since 31/1/2017 --until 31/12/2018 --formats java adoc xml --view --ignore-standalone-config --last-modified-date --timezone UTC+08 --find-previous-authors`
+`java -jar RepoSense.jar --repos https://github.com/reposense/RepoSense.git --output ./report_folder --since 31/1/2017 --until 31/12/2018 --formats java adoc xml --view --ignore-standalone-config --last-modified-date --timezone UTC+08 --find-previous-authors --analyze-authorship --originality-threshold 0.66`
 
 Same command as above but using most parameters in alias format:<br>
-`java -jar RepoSense.jar -r https://github.com/reposense/RepoSense.git -o ./report_folder -s 31/1/2017 -u 31/12/2018 -f java adoc xml -v -i -l -t UTC+08 -F`
+`java -jar RepoSense.jar -r https://github.com/reposense/RepoSense.git -o ./report_folder -s 31/1/2017 -u 31/12/2018 -f java adoc xml -v -i -l -t UTC+08 -F -A -ot 0.66`
 </box>
 
 The section below provides explanations for each of the flags.
 
 <!-- --------------------------◘---------------------------------------------------------------------------- -->
 
+### `--analyze-authorship`, `-A`
+
+**`--analyze-authorship`**: Performs further analysis to distinguish between partial and full credit attribution for
+lines of code assigned to the author.
+
+* Default: this feature is turned ***off*** by default and the author will receive partial credits for all lines of
+  code, as the code lines are at least partial credit but may not qualify for full credit.
+* Alias: `-A` (upper case)
+* Example: `--analyze-authorship` or `-A`
+
+<box type="info" seamless>
+
+A darker background colour represents full credit, while a lighter background colour represents partial credit.
+
+If the code is attributed to a different author by the user via `@@author` tag, then the new author will be given
+partial credit.
+</box>
+
+<!-- ------------------------------------------------------------------------------------------------------ -->
+
 ### `--assets`, `-a`
 
 <div id="section-assets">
@@ -147,6 +167,26 @@ This flag overrides the `Ignore file size limit` field in the CSV config file.
 
 <!-- ------------------------------------------------------------------------------------------------------ -->
 
+### `--originality-threshold`, `-ot`
+
+**`--originality-threshold [VALUE]`**: Specifies the cut-off point for partial and full credit
+in `--analyze-authorship`. Author will be given full credit if their contribution exceeds this threshold, else partial
+credit is given.
+
+* Parameter: `VALUE` Optional. Acceptable range: [0.0, 1.0].<br>
+  Default: `0.51`
+* Alias: `-ot`
+* Example: `--originality-threshold 0.66` or `-ot 0.66`
+
+<box type="info" seamless>
+
+* Requires `--analyze-authorship` flag.
+* An author's contribution, or `originality score`, is calculated using Levenshtein Distance (Edit Distance) algorithm.
+  We compare the difference between current code line and its previous versions.
+</box>
+
+<!-- ------------------------------------------------------------------------------------------------------ -->
+
 ### `--output`, `-o`
 
 **`--output OUTPUT_DIRECTORY`**: Indicates where to save the report generated.

frontend/cypress/tests/codeView/codeView_codeHighlighting.cy.js

@@ -38,7 +38,7 @@ describe('code highlighting works properly', () => {
     cy.get('.hljs-comment').contains('* Represents a Git Author.')
       .parent() // .line-content
       .parent() // .code
-      .should('have.css', 'background-color', 'rgb(230, 255, 237)'); // #e6ffed
+      .should('have.css', 'background-color', 'rgb(191, 246, 207)'); // #BFF6CF
   });
 
   it('should highlight code when multiple authors are merged in a repo group', () => {
@@ -62,13 +62,13 @@ describe('code highlighting works properly', () => {
     cy.get('.hljs-comment').contains('* MUI Colors module') // eugenepeh
       .parent() // .line-content
       .parent() // .code
-      .should('have.css', 'background-color', 'rgba(30, 144, 255, 0.19)') // #1e90ff, transparencyValue 30
+      .should('have.css', 'background-color', 'rgba(30, 144, 255, 0.314)') // #1e90ff, transparencyValue 50
       .then((firstAuthorColor) => {
         // eslint-disable-next-line quotes
         cy.get('.line-content').contains("'red': (") // jamessspanggg
           .parent() // .code
-          // #f08080, transparencyValue 30
-          .should('have.css', 'background-color', 'rgba(240, 128, 128, 0.19)')
+          // #f08080, transparencyValue 50
+          .should('have.css', 'background-color', 'rgba(240, 128, 128, 0.314)')
           .and('not.eq', firstAuthorColor);
       });
   });

frontend/cypress/tests/codeView/codeView_creditBackgroundColour.cy.js

@@ -0,0 +1,55 @@
+describe('credit background colour', () => {
+  it('check if background colour match the credit information for Eugene', () => {
+    // open the code panel
+    cy.get('.icon-button.fa-code')
+      .should('exist')
+      .first()
+      .click();
+
+    // src/main/java/reposense/model/Author.java
+    // line 9: full credit - #BFF6CF
+    cy.get(':nth-child(1) > .file-content > .segment-collection > :nth-child(2) > .java > .code')
+      .should('have.css', 'background-color')
+      .and('eq', 'rgb(191, 246, 207)');
+
+    // src/main/java/reposense/model/Author.java
+    // line 15: partial credit - #E6FFED
+    cy.get(':nth-child(1) > .file-content > .segment-collection > :nth-child(4) > .java > .code')
+      .should('have.css', 'background-color')
+      .and('eq', 'rgb(230, 255, 237)');
+  });
+
+  it('check if background colour match the credit information when group is merged', () => {
+    // check merge group checkbox
+    cy.get('#summary label.merge-group > input')
+      .should('be.visible')
+      .check()
+      .should('be.checked');
+
+    // open the code panel
+    cy.get('.icon-button.fa-code')
+      .should('exist')
+      .first()
+      .click();
+
+    // frontend/src/styles/_colors.scss
+    // line 35: full credit - #F0808050
+    cy.get(':nth-child(7) > .scss > :nth-child(1)')
+      .should('have.css', 'background-color')
+      .and('eq', 'rgba(240, 128, 128, 0.314)');
+
+    // FileInfoExtractor.java is too far away to be loaded, use filter to go to it directly
+    cy.get('#search')
+      .click()
+      .type('FileInfoExtractor.java');
+
+    cy.get('#submit-button')
+      .click();
+
+    // src/main/java/reposense/authorship/FileInfoExtractor.java
+    // line 23: partial credit - #1E90FF20
+    cy.get(':nth-child(10) > .java > :nth-child(1)')
+      .should('have.css', 'background-color')
+      .and('eq', 'rgba(30, 144, 255, 0.125)');
+  });
+});

frontend/src/components/c-segment.vue

@@ -1,8 +1,8 @@
 <template lang="pug">
 .segment(
-  v-bind:class="{ untouched: !segment.knownAuthor, active: isOpen }",
+  v-bind:class="{ untouched: !segment.knownAuthor, active: isOpen, isNotFullCredit: !segment.isFullCredit }",
   v-bind:style="{ 'border-left': `0.25rem solid ${authorColors[segment.knownAuthor]}` }",
-  v-bind:title="`Author: ${segment.knownAuthor || \"Unknown\"}`"
+  v-bind:title="`${segment.isFullCredit ? 'Author' : 'Co-author'}: ${segment.knownAuthor || \"Unknown\"}`"
 )
   .closer(v-if="canOpen",
     v-on:click="toggleCode", ref="topButton")
@@ -17,6 +17,7 @@
       v-bind:title="'Click to hide code'"
     )
   div(v-if="isOpen", v-hljs="path")
+    //- author color is applied only when the author color exists, else it takes the default mui color value
     .code(
       v-for="(line, index) in segment.lines", v-bind:key="index",
       v-bind:style="{ 'background-color': `${authorColors[segment.knownAuthor]}${transparencyValue}` }"
@@ -57,7 +58,7 @@ export default defineComponent({
     return {
       isOpen: (this.segment.knownAuthor !== null) || this.segment.lines.length < 5 as boolean,
       canOpen: (this.segment.knownAuthor === null) && this.segment.lines.length > 4 as boolean,
-      transparencyValue: '30' as string,
+      transparencyValue: (this.segment.isFullCredit ? '50' : '20') as string,
     };
   },
   computed: {
@@ -81,7 +82,7 @@ export default defineComponent({
   border-left: .25rem solid mui-color('green');
 
   .code {
-    background-color: mui-color('github', 'authored-code-background');
+    background-color: mui-color('github', 'full-authored-code-background');
     padding-left: 1rem;
   }
 
@@ -114,6 +115,12 @@ export default defineComponent({
     word-break: break-word;
   }
 
+  &.isNotFullCredit {
+    .code {
+      background-color: mui-color('github', 'partial-authored-code-background');
+    }
+  }
+
   &.untouched {
     $grey: mui-color('grey', '400');
     border-left: .25rem solid $grey;

frontend/src/styles/_colors.scss

@@ -303,7 +303,8 @@ $mui-colors: (
   'github': (
     'title-background': #FAFBFC,
     'border': #E1E4E8,
-    'authored-code-background': #E6FFED,
+    'full-authored-code-background': #BFF6CF,
+    'partial-authored-code-background': #E6FFED,
   ),
   'grey': (
     '50': #FAFAFA,

frontend/src/styles/hightlight-js-style.css

@@ -34,7 +34,7 @@
 
 .hljs-section,
 .hljs-name {
-    color: #63a35c;
+    color: #468C5A;
 }
 
 .hljs-tag {

frontend/src/types/types.ts

@@ -61,6 +61,7 @@ export interface Repo extends RepoRaw {
 
 export interface AuthorshipFileSegment {
   knownAuthor: string | null;
+  isFullCredit: boolean;
   lineNumbers: number[];
   lines: string[];
 }
@@ -83,3 +84,9 @@ export interface Bar {
   color?: string;
   tooltipText?: string;
 }
+
+export interface SegmentState {
+  id: number;
+  author: string | null;
+  isFullCredit: boolean;
+}

frontend/src/types/window.ts

@@ -68,6 +68,7 @@ declare global {
     repoSenseVersion: string;
     isSinceDateProvided: boolean;
     isUntilDateProvided: boolean;
+    isAuthorshipAnalyzed: boolean;
     DOMAIN_URL_MAP: DomainUrlMap;
     // eslint-disable-next-line  @typescript-eslint/no-explicit-any
     app: any;

frontend/src/types/zod/authorship-type.ts

@@ -4,6 +4,7 @@ const lineSchema = z.object({
   lineNumber: z.number(),
   author: z.object({ gitId: z.string() }),
   content: z.string(),
+  isFullCredit: z.boolean().default(false), // for backwards compatability
 });
 
 const fileResult = z.object({

frontend/src/types/zod/summary-type.ts

@@ -31,7 +31,6 @@ const urlSchema = z.object({
 const supportedDomainUrlMapSchema = z.record(urlSchema);
 
 // Contains the zod validation schema for the summary.json file
-
 export const summarySchema = z.object({
   repoSenseVersion: z.string(),
   reportGeneratedTime: z.string(),
@@ -44,6 +43,7 @@ export const summarySchema = z.object({
   untilDate: z.string(),
   isSinceDateProvided: z.boolean(),
   isUntilDateProvided: z.boolean(),
+  isAuthorshipAnalyzed: z.boolean().default(false), // for backwards compatability
   supportedDomainUrlMap: supportedDomainUrlMapSchema,
 });
 

frontend/src/utils/api.ts

@@ -217,6 +217,7 @@ window.api = {
     window.repoSenseVersion = data.repoSenseVersion;
     window.isSinceDateProvided = data.isSinceDateProvided;
     window.isUntilDateProvided = data.isUntilDateProvided;
+    window.isAuthorshipAnalyzed = data.isAuthorshipAnalyzed;
     document.title = data.reportTitle || document.title;
 
     const errorMessages: { [key: string]: ErrorMessage } = {};