Skip to content

Job application tracker and AI-powered job search assistant. Helps job seekers manage their search journey with AI resume review, job matching, task logging, and application analytics, all while keeping your data private.

feat: enhance HTML entity decoding in text processing and resume conversion

Khuram Niaz9a44052bd42a
diff --git a/__tests__/ats-runner.spec.ts b/__tests__/ats-runner.spec.ts
index 742061d..99381fe 100644
--- a/__tests__/ats-runner.spec.ts
+++ b/__tests__/ats-runner.spec.ts
@@ -169,4 +169,38 @@ describe("runAutomation (lever)", () => {
expect(result.errorMessage).toBe("no_companies");
expect((searchLeverJobs as any).mock.calls.length).toBe(0);
});
+
+ it("strips HTML and decodes entities from the resume before matching", async () => {
+ (prisma.resume.findUnique as any).mockResolvedValue({
+ id: "resume1",
+ title: "My Resume",
+ ContactInfo: null,
+ ResumeSections: [
+ {
+ sectionType: "experience",
+ workExperiences: [
+ {
+ description: "<p>Sales &amp; Marketing</p>",
+ startDate: new Date("2020-01-01"),
+ endDate: null,
+ Company: { label: "Acme" },
+ jobTitle: { label: "Engineer" },
+ location: { label: "Remote" },
+ },
+ ],
+ },
+ ],
+ });
+ (searchLeverJobs as any).mockResolvedValue({
+ jobs: [makeJob("Frontend Engineer", "React")],
+ errors: [],
+ });
+
+ await runAutomation(leverAutomation);
+
+ const prompt = (generateText as any).mock.calls[0][0].prompt as string;
+ expect(prompt).toContain("Sales & Marketing");
+ expect(prompt).not.toContain("&amp;");
+ expect(prompt).not.toContain("<p>");
+ });
});
diff --git a/__tests__/taskDescription.spec.ts b/__tests__/taskDescription.spec.ts
index 6d59072..e3aa590 100644
--- a/__tests__/taskDescription.spec.ts
+++ b/__tests__/taskDescription.spec.ts
@@ -25,6 +25,12 @@ describe("toPlainText", () => {
it("decodes non-breaking spaces to whitespace", () => {
expect(toPlainText("<p>a&nbsp;b</p>")).toBe("a b");
});
+
+ it("decodes HTML entities like &amp; back to their characters", () => {
+ expect(toPlainText("<p>Sales &amp; Marketing</p>")).toBe(
+ "Sales & Marketing",
+ );
+ });
});
describe("hasDescription", () => {
diff --git a/__tests__/text-processing.spec.ts b/__tests__/text-processing.spec.ts
index 835fce0..b4eb760 100644
--- a/__tests__/text-processing.spec.ts
+++ b/__tests__/text-processing.spec.ts
@@ -65,6 +65,18 @@ describe("removeHtmlTags", () => {
const result = removeHtmlTags(html);
expect(result).toContain("• Content");
});
+
+ it("decodes HTML entities left over from tag stripping", () => {
+ const html = "<p>Baker Hughes, &amp; EPIROC</p>";
+ const result = removeHtmlTags(html);
+ expect(result).toBe("Baker Hughes, & EPIROC");
+ });
+
+ it("decodes common named and numeric entities", () => {
+ const html = "<p>&lt;tag&gt; &quot;quoted&quot; &#39;it&#39;s&#39;</p>";
+ const result = removeHtmlTags(html);
+ expect(result).toBe('<tag> "quoted" \'it\'s\'');
+ });
});
describe("normalizeWhitespace", () => {
diff --git a/src/lib/ai/tools/text-processing.ts b/src/lib/ai/tools/text-processing.ts
index 65ef229..392887a 100644
--- a/src/lib/ai/tools/text-processing.ts
+++ b/src/lib/ai/tools/text-processing.ts
@@ -6,17 +6,39 @@
// HTML AND WHITESPACE NORMALIZATION
+const NAMED_ENTITIES: Record<string, string> = {
+ "&lt;": "<",
+ "&gt;": ">",
+ "&quot;": '"',
+ "&apos;": "'",
+ "&#39;": "'",
+ "&nbsp;": " ",
+ "&amp;": "&", // keep ampersand last so it does not undo other entities
+};
+
+const decodeHtmlEntities = (text: string): string => {
+ let out = text.replace(/&#(\d+);/g, (_, code) =>
+ String.fromCodePoint(Number(code)),
+ );
+ for (const [entity, char] of Object.entries(NAMED_ENTITIES)) {
+ out = out.split(entity).join(char);
+ }
+ return out;
+};
+
export const removeHtmlTags = (description: string | undefined): string => {
if (!description) return "";
- return description
- .replace(/<li[^>]*>/gi, "• ")
- .replace(/<\/(li|p|div|br)[^>]*>/gi, "\n")
- .replace(/<br\s*\/?>/gi, "\n")
- .replace(/<[^>]+>/g, " ")
- .replace(/\s+/g, " ")
- .replace(/\n\s*\n/g, "\n")
- .trim();
+ return decodeHtmlEntities(
+ description
+ .replace(/<li[^>]*>/gi, "• ")
+ .replace(/<\/(li|p|div|br)[^>]*>/gi, "\n")
+ .replace(/<br\s*\/?>/gi, "\n")
+ .replace(/<[^>]+>/g, " ")
+ .replace(/\s+/g, " ")
+ .replace(/\n\s*\n/g, "\n")
+ .trim(),
+ );
};
export const normalizeWhitespace = (text: string): string => {
diff --git a/src/lib/scraper/runner.ts b/src/lib/scraper/runner.ts
index 86950d7..02672c5 100644
--- a/src/lib/scraper/runner.ts
+++ b/src/lib/scraper/runner.ts
@@ -1211,7 +1211,7 @@ async function convertResumeForMatch(
for (const section of resume.ResumeSections) {
if (section.sectionType === "summary" && section.summary?.content) {
- parts.push("## SUMMARY", section.summary.content);
+ parts.push("## SUMMARY", removeHtmlTags(section.summary.content));
}
if (
@@ -1224,7 +1224,7 @@ async function convertResumeForMatch(
`Company: ${exp.Company.label}`,
`Job Title: ${exp.jobTitle.label}`,
`Location: ${exp.location.label}`,
- `Description: ${exp.description}`,
+ `Description: ${removeHtmlTags(exp.description)}`,
"",
);
}
@@ -1237,7 +1237,7 @@ async function convertResumeForMatch(
`Institution: ${edu.institution}`,
`Degree: ${edu.degree}`,
`Field: ${edu.fieldOfStudy}`,
- edu.description ? `Description: ${edu.description}` : "",
+ edu.description ? `Description: ${removeHtmlTags(edu.description)}` : "",
"",
);
}
diff --git a/src/lib/tasks/description.ts b/src/lib/tasks/description.ts
index e808cc3..728df27 100644
--- a/src/lib/tasks/description.ts
+++ b/src/lib/tasks/description.ts
@@ -7,6 +7,11 @@ export function toPlainText(html?: string | null): string {
return html
.replace(/<[^>]+>/g, " ")
.replace(/&nbsp;/g, " ")
+ .replace(/&lt;/g, "<")
+ .replace(/&gt;/g, ">")
+ .replace(/&quot;/g, '"')
+ .replace(/&#39;/g, "'")
+ .replace(/&amp;/g, "&")
.replace(/\s+/g, " ")
.trim();
}