Expose target URL content type

Added a `Scoop.targetUrlContentType` property, which keeps track of the content-type detected in `Scoop.#detectAndCaptureNonWebContent()`.

This property is exposed by `Scoop.summary()`, and therefore present in the JSON file generated by the `--json-summary-output` CLI option.

Scoop.js

@@ -77,6 +77,13 @@ export class Scoop {
    */
   targetUrlIsWebPage = true
 
+  /**
+   * Content-type of the target url.
+   * Assumed `text/html` unless detected otherwise.
+   * @type {string}
+   */
+  targetUrlContentType = 'text/html; charset=utf-8'
+
   /**
    * Current settings.
    * @type {ScoopOptions}
@@ -624,6 +631,8 @@ export class Scoop {
    * - Captures it via a curl behind our proxy
    * - Sets capture state to `PARTIAL`
    *
+   * Populates `this.targetUrlIsWebPage` and `this.targetUrlContentType`.
+   *
    * @param {Page} page - A Playwright [Page]{@link https://playwright.dev/docs/api/class-page} object
    * @returns {Promise<void>}
    * @private
@@ -676,7 +685,13 @@ export class Scoop {
       return
     }
 
-    // If text/html, continue capture as normal
+    // Capture content-type
+    if (contentType) {
+      this.targetUrlContentType = contentType
+    }
+
+    // If text/html, bail from non-web content capture process.
+    // Scoop.capture will go based on the value of `this.targeredUrlIsWebPage`.
     if (contentType?.startsWith('text/html')) {
       this.log.info('Requested URL is a web page')
       return
@@ -1370,6 +1385,7 @@ export class Scoop {
    * @property {string[]} states - Zero-indexed Scoop.states values.
    * @property {string} targetUrl
    * @property {boolean} targetUrlIsWebPage
+   * @property {string} targetUrlContentType
    * @property {ScoopOptions} options
    * @property {string} startedAt - ISO-formated date
    * @property {string[]} blockedRequests
@@ -1399,6 +1415,7 @@ export class Scoop {
       states: Object.keys(Scoop.states), // So summary.states[summary.state] = 'NAME-OF-STATE'
       targetUrl: this.url,
       targetUrlIsWebPage: this.targetUrlIsWebPage,
+      targetUrlContentType: this.targetUrlContentType,
       options: this.options,
       startedAt: this.startedAt,
       blockedRequests: [],

Scoop.test.js

@@ -93,6 +93,7 @@ await test('Scoop - capture of a web page.', async (t) => {
     const summary = await capture.summary()
     assert(summary)
     assert.equal(summary.targetUrl, capture.url)
+    assert.equal(summary.targetUrlContentType, 'text/html; charset=UTF-8')
     assert.equal(summary.state, Scoop.states.COMPLETE)
     assert.equal(summary.exchangeUrls.length, capture.exchanges.length)
     assert.equal(summary.attachments.provenanceSummary, 'provenance-summary.html')
@@ -141,6 +142,16 @@ await test('Scoop - capture of a non-web resource.', async (t) => {
     // assert.notEqual(html.response.body, testPdfFixture)
   })
 
+  await t.test('Scoop.summary() returns a valid object', async (_t) => {
+    const capture = await Scoop.capture(`${URL}/test.pdf`, options)
+    const summary = await capture.summary()
+    assert(summary)
+    assert.equal(summary.targetUrl, capture.url)
+    assert.equal(summary.targetUrlContentType, 'application/pdf')
+    assert.equal(summary.state, Scoop.states.PARTIAL)
+    assert.equal(summary.exchangeUrls.length, capture.exchanges.length)
+  })
+
   /*
    * TEARDOWN
    */