From 227cc28e01cd48ce3f69d84b4d686273850d4942 Mon Sep 17 00:00:00 2001
From: David Hewitt <moshie@users.noreply.github.com>
Date: Sun, 14 Jan 2018 00:08:42 +0000
Subject: [PATCH] Search and replace the hostname in URLs. (#3498)

* Search and replace the hostname in URLs.
https://github.com/gatsbyjs/gatsby/issues/3450

* Fixed code styling issues

* Missed backticks for the timers requirement

* Updated readme with search and replace options

* Removed async, wrapped json parse in a try catch and updated blacklist function

* Updated search and replace api and updated documentation

* Added logging for invalid search and replace option

* make option more explicit, updated default value & removed redundant logging

* Updated the variable name making it clearer
---
 packages/gatsby-source-wordpress/README.md    |  5 ++
 .../src/gatsby-node.js                        |  7 +++
 .../gatsby-source-wordpress/src/normalize.js  | 49 +++++++++++++++++++
 3 files changed, 61 insertions(+)

diff --git a/packages/gatsby-source-wordpress/README.md b/packages/gatsby-source-wordpress/README.md
index 5ef5e5fd82500..6b2144a8b8678 100644
--- a/packages/gatsby-source-wordpress/README.md
+++ b/packages/gatsby-source-wordpress/README.md
@@ -83,6 +83,11 @@ plugins: [
       // Set verboseOutput to true to display a verbose output on `npm run develop` or `npm run build`
       // It can help you debug specific API Endpoints problems
       verboseOutput: false,
+      // Search and Replace Urls across WordPress content
+      searchAndReplaceContentUrls: {
+        sourceUrl: "https://source-url.com", 
+        replacementUrl: "https://replacement-url.com"
+      }
     },
   },
 ];
diff --git a/packages/gatsby-source-wordpress/src/gatsby-node.js b/packages/gatsby-source-wordpress/src/gatsby-node.js
index 31bf8dfbf67be..8e5cf5493ded1 100644
--- a/packages/gatsby-source-wordpress/src/gatsby-node.js
+++ b/packages/gatsby-source-wordpress/src/gatsby-node.js
@@ -27,6 +27,7 @@ exports.sourceNodes = async (
     auth = {},
     verboseOutput,
     perPage = 100,
+    searchAndReplaceContentUrls = {},
   }
 ) => {
   const { createNode } = boundActionCreators
@@ -92,6 +93,12 @@ exports.sourceNodes = async (
     createNode,
   })
 
+  // Search and replace Content Urls
+  entities = normalize.searchReplaceContentUrls({ 
+    entities, 
+    searchAndReplaceContentUrls,
+  })
+
   // creates nodes for each entry
   normalize.createNodesFromEntities({ entities, createNode })
 
diff --git a/packages/gatsby-source-wordpress/src/normalize.js b/packages/gatsby-source-wordpress/src/normalize.js
index 7c68b0ef24c33..cfb7d8162d7f6 100644
--- a/packages/gatsby-source-wordpress/src/normalize.js
+++ b/packages/gatsby-source-wordpress/src/normalize.js
@@ -234,6 +234,55 @@ exports.mapTagsCategoriesToTaxonomies = entities =>
     return e
   })
 
+exports.searchReplaceContentUrls = function ({ entities, searchAndReplaceContentUrls }) {
+
+  if (
+    !(_.isPlainObject(searchAndReplaceContentUrls)) ||
+    !(_.has(searchAndReplaceContentUrls, `sourceUrl`)) || 
+    !(_.has(searchAndReplaceContentUrls, `replacementUrl`)) || 
+    typeof searchAndReplaceContentUrls.sourceUrl !== `string` ||
+    typeof searchAndReplaceContentUrls.replacementUrl !== `string`
+  ) {
+    return entities
+  }
+
+  const { sourceUrl, replacementUrl } = searchAndReplaceContentUrls
+
+  const _blacklist = [
+    `_links`,
+    `__type`,
+  ]
+
+  const blacklistProperties = function (obj = {}, blacklist = []) {
+    for (var i = 0; i < blacklist.length; i++) {
+        delete obj[blacklist[i]]
+    }
+
+    return obj
+  }
+
+  return entities.map(function (entity) {
+    const original = Object.assign({}, entity)
+
+    try {
+      var whiteList = blacklistProperties(entity, _blacklist)
+      var replaceable = JSON.stringify(whiteList)
+      var replaced = replaceable.replace(new RegExp(sourceUrl, `g`), replacementUrl)
+      var parsed = JSON.parse(replaced)
+    } catch (e) {
+      console.log(
+        colorized.out(
+          e.message,
+          colorized.color.Font.FgRed
+        )
+      )
+      return original
+    }
+
+    return _.defaultsDeep(parsed, original)
+  })
+}
+
 exports.mapEntitiesToMedia = entities => {
   const media = entities.filter(e => e.__type === `wordpress__wp_media`)