only show files that still exist in the knowledge gaps report

Author: anton-107

.eslintrc.cjs

@@ -12,7 +12,7 @@ module.exports = {
     "@typescript-eslint/no-unused-vars": "error",
     "max-depth": ["error", 2],
     "max-nested-callbacks": ["error", 2],
-    "max-lines-per-function": ["error", 60],
+    "max-lines-per-function": ["error", 70],
     "max-statements": ["error", 26],
     "max-params": ["error", 3],
   },

src/git-reader/git-repository.ts

@@ -1,10 +1,17 @@
+import { stat } from "node:fs/promises";
+
 import fs from "fs";
 import git from "isomorphic-git";
 import { Readable } from "stream";
 
 import { ResumableProcessor } from "../cache/resumable-processor.js";
 import { time, timeLog } from "../index.js";
-import { ChangedFile, Commit, ExpandedCommit } from "../interfaces.js";
+import {
+  ChangedFile,
+  ChangedFileType,
+  Commit,
+  ExpandedCommit,
+} from "../interfaces.js";
 
 interface GitReadOptions {
   stream?: Readable;
@@ -77,11 +84,11 @@ export class GitRepository {
     prevOID: string,
     nextOID: string
   ): Promise<ChangedFile[]> {
-    return await git.walk({
+    const files = await git.walk({
       fs,
       dir: this.repoPath,
       trees: [git.TREE({ ref: prevOID }), git.TREE({ ref: nextOID })],
-      map: async function (filepath, [commitA, commitB]) {
+      map: async (filepath, [commitA, commitB]) => {
         let aOID = "";
         let bOID = "";
         let aType = "";
@@ -115,7 +122,7 @@ export class GitRepository {
         }
 
         // determine modification type:
-        let type = "equal";
+        let type: ChangedFileType = "equal";
         if (aOID !== bOID) {
           type = "modify";
         }
@@ -132,5 +139,20 @@ export class GitRepository {
         };
       },
     });
+    // check if file exists:
+    for (const file of files) {
+      file.isExistingFile = await this.checkFileExists(
+        `${this.repoPath}${file.path}`
+      );
+    }
+    return files;
+  }
+  private async checkFileExists(path: string): Promise<boolean> {
+    try {
+      const stats = await stat(path);
+      return stats.isFile();
+    } catch {
+      return false;
+    }
   }
 }

src/index.ts

@@ -5,7 +5,6 @@ import { SummaryDashboard } from "./dashboard/summary-dashboard.js";
 import { GitRepository } from "./git-reader/git-repository.js";
 import { ExpandedCommit } from "./interfaces.js";
 import { ListOfContributorsPerFileAggregate } from "./stats/aggregate/list-of-contributors-per-file-aggregate.js";
-import { getListOfContributorsPerFile } from "./stats/list-of-contributors-per-file.js";
 import { getNumberOfChangesPerFile } from "./stats/number-of-changes-per-file.js";
 import { getNumberOfCommitsByAuthor } from "./stats/number-of-commits-by-author.js";
 import { getNumberOfContributorsPerFile } from "./stats/number-of-contributors-per-file.js";
@@ -53,25 +52,27 @@ async function collectHotFiles(commitsWithChangedFiles: ExpandedCommit[]) {
 async function collectKnowledgeGaps(commitsWithChangedFiles: ExpandedCommit[]) {
   const contributorsPerFile = getNumberOfContributorsPerFile(
     commitsWithChangedFiles
-  );
+  ).filter((x) => x.isExistingFile);
   log(
     "knowledge gaps (files with least number of contributors)",
-    Object.keys(contributorsPerFile)
-      .map((x) => {
-        return [x, contributorsPerFile[x]];
+    contributorsPerFile
+      .sort((a, b) => {
+        if (a.contributorsNames.length === b.contributorsNames.length) {
+          return a.lastChange.getTime() - b.lastChange.getTime();
+        }
+        return a.contributorsNames.length - b.contributorsNames.length;
       })
-      .sort((a, b) => Number(a[1]) - Number(b[1]))
       .slice(0, 50)
   );
 }
-function collectDetailedContributorsPerFile(
-  commitsWithChangedFiles: ExpandedCommit[]
-) {
-  const listOfContributorsPerFile = getListOfContributorsPerFile(
-    commitsWithChangedFiles
-  );
-  log("detailed contributors for each file", listOfContributorsPerFile);
-}
+// function collectDetailedContributorsPerFile(
+//   commitsWithChangedFiles: ExpandedCommit[]
+// ) {
+//   const listOfContributorsPerFile = getListOfContributorsPerFile(
+//     commitsWithChangedFiles
+//   );
+//   log("detailed contributors for each file", listOfContributorsPerFile);
+// }
 
 /* eslint-disable-next-line max-lines-per-function, max-statements */
 async function main() {
@@ -152,6 +153,6 @@ async function main() {
   });
   await collectHotFiles(commitsWithChangedFiles);
   await collectKnowledgeGaps(commitsWithChangedFiles);
-  collectDetailedContributorsPerFile(commitsWithChangedFiles);
+  // collectDetailedContributorsPerFile(commitsWithChangedFiles);
 }
 main();

src/interfaces.ts

@@ -9,6 +9,7 @@ export type ChangedFileType = "modify" | "add" | "remove" | "equal";
 export interface ChangedFile {
   type: ChangedFileType;
   path: string;
+  isExistingFile: boolean;
 }
 
 export interface Commit {

src/stats/number-of-contributors-per-file.ts

@@ -1,24 +1,51 @@
 import { ExpandedCommit } from "../interfaces.js";
+interface ContributorsSummary {
+  filePath: string;
+  contributorsNames: string[];
+  firstChange: Date;
+  lastChange: Date;
+  isExistingFile: boolean;
+}
 
 export function getNumberOfContributorsPerFile(
   commits: ExpandedCommit[]
-): Record<string, number> {
-  const changedFiles: Record<string, string[]> = {};
+): ContributorsSummary[] {
+  const changedFiles: Record<string, Set<string>> = {};
+  const existingFiles: Record<string, boolean> = {};
+  const firstChange: Record<string, number> = {};
+  const lastChange: Record<string, number> = {};
 
   commits.forEach((commit) => {
     commit.changedFiles.forEach((file) => {
-      if (!changedFiles[file.path]) {
-        changedFiles[file.path] = [];
+      if (file.type === "equal") {
+        return;
       }
-      if (!changedFiles[file.path].includes(commit.commit.commit.author.name)) {
-        changedFiles[file.path].push(commit.commit.commit.author.name);
+      if (!changedFiles[file.path]) {
+        changedFiles[file.path] = new Set();
       }
+      changedFiles[file.path].add(commit.commit.commit.author.name);
+      existingFiles[file.path] = file.isExistingFile;
+
+      firstChange[file.path] = Math.min(
+        firstChange[file.path] || 1_000_000_000_000,
+        commit.commit.commit.author.timestamp
+      );
+      lastChange[file.path] = Math.max(
+        lastChange[file.path] || 0,
+        commit.commit.commit.author.timestamp
+      );
     });
   });
 
-  const results: Record<string, number> = {};
+  const results: ContributorsSummary[] = [];
   Object.keys(changedFiles).forEach((filePath) => {
-    results[filePath] = changedFiles[filePath].length;
+    results.push({
+      filePath,
+      contributorsNames: Array.from(changedFiles[filePath].values()),
+      firstChange: new Date(firstChange[filePath] * 1000),
+      lastChange: new Date(lastChange[filePath] * 1000),
+      isExistingFile: existingFiles[filePath],
+    });
   });
 
   return results;