For HTML transcripts, just use the next segment's startTime for the e…

…ndTime instead of adding the startTime to it. Resolves #14
stevencrader · Apr 2, 2023 · af759a2 · af759a2
1 parent 344bbfe
commit af759a2
Show file tree

Hide file tree

Showing 8 changed files with 87 additions and 89 deletions.
diff --git a/src/formats/html.ts b/src/formats/html.ts
@@ -148,8 +148,7 @@ const getSegmentsFromHTMLElements = (elements: Array<HTMLElement>): Array<Segmen
                 // update endTime of previous Segment
                 const totalSegments = outSegments.length
                 if (totalSegments > 0) {
-                    outSegments[totalSegments - 1].endTime =
-                        outSegments[totalSegments - 1].startTime + s.segment.startTime
+                    outSegments[totalSegments - 1].endTime = s.segment.startTime
                     outSegments[totalSegments - 1].endTimeFormatted = timestampFormatter.format(
                         outSegments[totalSegments - 1].endTime
                     )

diff --git a/test/html.test.ts b/test/html.test.ts
@@ -1,7 +1,7 @@
 import { describe, expect, test } from "@jest/globals"
 
+import { Segment } from "../src"
 import { parseHTML } from "../src/formats/html"
-import { Segment } from "../src/types"
 
 import { readFile, TestFiles } from "./test_utils"
 
@@ -225,8 +225,8 @@ describe("Bad HTML data", () => {
                 {
                     startTime: 1,
                     startTimeFormatted: "00:00:01.000",
-                    endTime: 31,
-                    endTimeFormatted: "00:00:31.000",
+                    endTime: 30,
+                    endTimeFormatted: "00:00:30.000",
                     speaker: "Kevin",
                     body: "It is so stinking nice to like, show up and record this show. And Travis has already put together an outline. Kevin's got suggestions, I throw my thoughts into the mix. And then Travis goes and does all the work from there, too. It's out into the wild. And I don't see anything. That's an absolute joy for at least two thirds of the team. Yeah, I mean, exactly.",
                 },

diff --git a/test/json.test.ts b/test/json.test.ts
@@ -1,7 +1,7 @@
 import { describe, expect, test } from "@jest/globals"
 
+import { Segment } from "../src"
 import { parseJSON } from "../src/formats/json"
-import { Segment } from "../src/types"
 
 import { readFile, TestFiles } from "./test_utils"
 

diff --git a/test/srt.test.ts b/test/srt.test.ts
@@ -1,7 +1,7 @@
 import { describe, expect, jest, test } from "@jest/globals"
 
+import { Segment } from "../src"
 import { parseSRT, parseSRTSegment, SRTSegment } from "../src/formats/srt"
-import { Segment } from "../src/types"
 
 import { readFile, TestFiles } from "./test_utils"
 

diff --git a/test/test.ts b/test/test.ts
@@ -1,7 +1,6 @@
 import { describe, expect, test } from "@jest/globals"
 
-import { combineSingleWordSegments, convertFile, determineFormat } from "../src"
-import { Segment, TranscriptFormat } from "../src/types"
+import { combineSingleWordSegments, convertFile, determineFormat, Segment, TranscriptFormat } from "../src"
 
 import { readFile, TestFiles } from "./test_utils"
 

diff --git a/test/test_files/buzzcast_html_parsed.json b/test/test_files/buzzcast_html_parsed.json
diff --git a/test/test_files/podnews_weekly_review_html_parsed.json b/test/test_files/podnews_weekly_review_html_parsed.json
diff --git a/test/test_utils.ts b/test/test_utils.ts
@@ -1,7 +1,7 @@
 import { existsSync, mkdirSync, readFileSync, writeFileSync } from "node:fs"
 import * as path from "node:path"
 
-import { Segment } from "../src/types"
+import { Segment } from "../src"
 
 export enum TestFiles {
     TRANSCRIPT_HTML_BUZZCAST = "buzzcast.html",