wrapper: fix not downloading photo with relative urls

2 files changed, 21 insertions, 7 deletions

diff --git a/src/weepinbell/processor/wrapper.clj b/src/weepinbell/processor/wrapper.clj
index 202ac02..8e2a8e2 100644
--- a/src/weepinbell/processor/wrapper.clj
+++ b/src/weepinbell/processor/wrapper.clj
@@ -3,7 +3,7 @@
   It adds microformats and extra data and return complete webmention map ready for storage."

   (:require [clojure.java.io :as io]

             [weepinbell.http.downloader :refer [download-file]]

-            [weepinbell.utils.url :refer [valid-http-url?]]

+            [weepinbell.utils.url :refer [valid-http-url? get-base relative? normalize]]

             [weepinbell.microformats.searcher :refer [use-items search-photo find-h-entry find-h-card]]

             [mf2clj.core :as mf-parser]

             [taoensso.timbre :as timbre]

@@ -38,7 +38,7 @@
   (or (->> mf use-items find-h-entry search-photo)

       (->> mf use-items find-h-card search-photo)))

 

-(defn- proxy-mf-photo [raw-photo base-url]

+(defn- proxy-mf-photo [photo-base raw-photo]

   (if raw-photo

     (let [photo-url (str raw-photo)

           file-ext (or (re-find #"\.(?:jpg|jpeg|png|gif|webp)" photo-url) ".jpg")

@@ -49,22 +49,28 @@
       (when-not (.exists (io/file local-path))

         (timbre/info "Downloading microformats photo to proxy external image.")

         (download-file photo-url local-path))

-      (str base-url public-path))

+      (str photo-base public-path))

     nil))

 

+(defn complete-url [source-base url]

+  (if (relative? url)

+    (normalize (str source-base url))

+    url))

+

 (defn- wrap-webmention

   "Parse microformats and collect essential info. 

   Wrap source, target and collected data into a map and return it."

-  [id source-html source target base-url]

+  [id source-html source target photo-base]

   (timbre/info "parsing microformats and wrapping up webmention" id)

   (let [mf (-> source-html mf-parser/parse :items)

         wm {:source source

             :target target

             :id (digest (str source target))

             :modified (System/currentTimeMillis)}

-        proxied-photo (-> mf

-                          find-mf-photo

-                          (proxy-mf-photo base-url))

+        proxied-photo (->> mf

+                           find-mf-photo

+                           (complete-url (get-base source))

+                           (proxy-mf-photo photo-base))

         data {:webmention wm

               :microformats mf

               :extra {:proxied-photo proxied-photo}}]

diff --git a/src/weepinbell/utils/url.clj b/src/weepinbell/utils/url.clj
index a0de304..265509c 100644
--- a/src/weepinbell/utils/url.clj
+++ b/src/weepinbell/utils/url.clj
@@ -16,6 +16,10 @@
     (catch MalformedURLException _ false)

     (catch URISyntaxException _ false)))

 

+(defn relative? [url]

+  (or (string/starts-with? url "/")

+      (string/starts-with? url ".")))

+

 (defn get-domain [url]

   (when (some? url)

     (-> url URI/create .getHost)))

@@ -24,6 +28,10 @@
   (when (some? url)

     (-> url URI/create .getPath)))

 

+(defn get-base [url]

+  (let [uri (URI. url)]

+    (str (.getScheme uri) "://" (.getAuthority uri))))

+

 (defn get-target [wm]

   (get-in wm [:webmention :target]))