[gatsby-source-wordpress] Handle all media types

packages/gatsby-source-wordpress/README.md

@@ -42,16 +42,16 @@ We welcome PRs adding support for data from other plugins.
 // In your gatsby-config.js
 plugins: [
   /*
-     * Gatsby's data processing layer begins with “source”
-     * plugins. Here the site sources its data from Wordpress.
-     */
+   * Gatsby's data processing layer begins with “source”
+   * plugins. Here the site sources its data from Wordpress.
+   */
   {
     resolve: "gatsby-source-wordpress",
     options: {
       /*
-        * The base URL of the Wordpress site without the trailingslash and the protocol. This is required.
-        * Example : 'gatsbyjsexamplewordpress.wordpress.com' or 'www.example-site.com'
-        */
+       * The base URL of the Wordpress site without the trailingslash and the protocol. This is required.
+       * Example : 'gatsbyjsexamplewordpress.wordpress.com' or 'www.example-site.com'
+       */
       baseUrl: "gatsbyjsexamplewordpress.wordpress.com",
       // The protocol. This can be http or https.
       protocol: "http",

packages/gatsby-source-wordpress/src/__tests__/request-in-queue.js

@@ -4,7 +4,9 @@ const requestInQueue = require(`../request-in-queue`)
 const axios = require(`axios`)
 
 axios.mockImplementation(opts => {
-  if (opts.throw) { throw new Error(opts.throw) }
+  if (opts.throw) {
+    throw new Error(opts.throw)
+  }
 
   return opts.url.slice(opts.url.lastIndexOf(`/`) + 1)
 })
@@ -28,7 +30,7 @@ describe(`requestInQueue`, () => {
   it(`runs all requests in queue`, async () => {
     await requestInQueue(requests)
 
-    requests.forEach((req) => {
+    requests.forEach(req => {
       expect(axios).toHaveBeenCalledWith(req)
     })
   })

packages/gatsby-source-wordpress/src/fetch.js

@@ -207,7 +207,15 @@ async function fetchData({
   if (_verbose) console.time(`Fetching the ${type} took`)
 
   let routeResponse = await getPages(
-    { url, _perPage, _hostingWPCOM, _auth, _accessToken, _verbose, _concurrentRequests },
+    {
+      url,
+      _perPage,
+      _hostingWPCOM,
+      _auth,
+      _accessToken,
+      _verbose,
+      _concurrentRequests,
+    },
     1
   )
 
@@ -270,7 +278,15 @@ async function fetchData({
  * @returns
  */
 async function getPages(
-  { url, _perPage, _hostingWPCOM, _auth, _accessToken,  _concurrentRequests, _verbose },
+  {
+    url,
+    _perPage,
+    _hostingWPCOM,
+    _auth,
+    _accessToken,
+    _concurrentRequests,
+    _verbose,
+  },
   page = 1
 ) {
   try {
@@ -320,9 +336,13 @@ async function getPages(
     }
 
     // We got page 1, now we want pages 2 through totalPages
-    const pageOptions = _.range(2, totalPages + 1).map(getPage => getOptions(getPage))
+    const pageOptions = _.range(2, totalPages + 1).map(getPage =>
+      getOptions(getPage)
+    )
 
-    const pages = await requestInQueue(pageOptions, { concurrent: _concurrentRequests })
+    const pages = await requestInQueue(pageOptions, {
+      concurrent: _concurrentRequests,
+    })
 
     const pageData = pages.map(page => page.data)
     pageData.forEach(list => {
@@ -383,10 +403,17 @@ function getValidRoutes({
           console.log(
             colorized.out(`Invalid route.`, colorized.color.Font.FgRed)
           )
-      } else if (_excludedRoutes.some(excludedRoute => minimatch(routePath, excludedRoute))) {
+      } else if (
+        _excludedRoutes.some(excludedRoute =>
+          minimatch(routePath, excludedRoute)
+        )
+      ) {
         if (_verbose)
           console.log(
-            colorized.out(`Excluded route from excludedRoutes pattern.`, colorized.color.Font.FgYellow)
+            colorized.out(
+              `Excluded route from excludedRoutes pattern.`,
+              colorized.color.Font.FgYellow
+            )
           )
       } else {
         if (_verbose)
@@ -475,8 +502,7 @@ const getRawEntityType = route =>
  * @param {any} baseUrl The base site URL that should be removed
  * @param {any} fullUrl The full URL to retrieve the route path from
  */
-const getRoutePath = (baseUrl, fullUrl) =>
-  fullUrl.replace(baseUrl, ``)
+const getRoutePath = (baseUrl, fullUrl) => fullUrl.replace(baseUrl, ``)
 
 /**
  * Extract the route manufacturer

packages/gatsby-source-wordpress/src/normalize.js

@@ -223,11 +223,13 @@ exports.mapTagsCategoriesToTaxonomies = entities =>
     return e
   })
 
-exports.mapElementsToParent = entities => entities.map(e => {
+exports.mapElementsToParent = entities =>
+  entities.map(e => {
     if (e.wordpress_parent) {
       // Create parent_element with a link to the parent node of type.
-      e.parent_element___NODE = entities
-        .find(t => t.wordpress_id === e.wordpress_parent && t.__type === e.__type).id
+      e.parent_element___NODE = entities.find(
+        t => t.wordpress_id === e.wordpress_parent && t.__type === e.__type
+      ).id
     }
     return e
   })
@@ -295,10 +297,8 @@ exports.mapEntitiesToMedia = entities => {
         ? true
         : false
 
-    const photoRegex = /\.(gif|jpg|jpeg|tiff|png)$/i
-    const isPhotoUrl = filename =>
-      _.isString(filename) && photoRegex.test(filename)
-    const isPhotoUrlAlreadyProcessed = key => key == `source_url`
+    const isURL = value => _.isString(value) && value.startsWith(`http`)
+    const isMediaUrlAlreadyProcessed = key => key == `source_url`
     const isFeaturedMedia = (value, key) =>
       (_.isNumber(value) || _.isBoolean(value)) && key === `featured_media`
     // ACF Gallery and similarly shaped arrays
@@ -308,7 +308,7 @@ exports.mapEntitiesToMedia = entities => {
 
     // Try to get media node from value:
     //  - special case - check if key is featured_media and value is photo ID
-    //  - check if value is photo url
+    //  - check if value is media url
     //  - check if value is ACF Image Object
     //  - check if value is ACF Gallery
     const getMediaFromValue = (value, key) => {
@@ -319,7 +319,7 @@ exports.mapEntitiesToMedia = entities => {
             : null,
           deleteField: true,
         }
-      } else if (isPhotoUrl(value) && !isPhotoUrlAlreadyProcessed(key)) {
+      } else if (isURL(value) && !isMediaUrlAlreadyProcessed(key)) {
         const mediaNodeID = getMediaItemID(
           media.find(m => m.source_url === value)
         )

packages/gatsby-source-wordpress/src/request-in-queue.js

@@ -35,14 +35,16 @@ async function handleQueue(task, cb) {
  * @param  {Options}  opts   Options that will be given to better-queue
  * @return {Promise}         Resolves with the accumulated values from the tasks
  */
-module.exports = function requestInQueue (tasks, opts = {}) {
+module.exports = function requestInQueue(tasks, opts = {}) {
   return new Promise((res, rej) => {
     const q = new Queue(handleQueue, { ..._defaults, ...opts })
 
-    const taskMap = new Map(tasks.map((t) => {
-      q.push(t)
-      return [t.url, null]
-    }))
+    const taskMap = new Map(
+      tasks.map(t => {
+        q.push(t)
+        return [t.url, null]
+      })
+    )
 
     q.on(`task_failed`, (id, err) => {
       rej(`${id} failed with err: ${err}`)