Start to fetch page content

src/controllers/posts_controller.rs

@@ -6,7 +6,6 @@ use crate::api_data::post_api::PostAPI;
 use crate::api_data::posts_targets_api::PostsTargets;
 use crate::api_data::res_create_post::ResCreatePost;
 use crate::constants::{PATH_POST_IMAGES, PATH_POST_PDF};
-use crate::routes::RequestResult;
 use crate::data::base_request_handler::BaseRequestHandler;
 use crate::data::error::{ExecError, ResultBoxError};
 use crate::data::group::GroupAccessLevel;
@@ -15,9 +14,11 @@ use crate::data::new_survey::NewSurvey;
 use crate::data::notification::NotifEventType;
 use crate::data::post::{Post, PostAccessLevel, PostFile, PostKind, PostPageKind, PostVisibilityLevel, PostWebLink};
 use crate::helpers::{friends_helper, groups_helper, notifications_helper, posts_helper, survey_helper, user_helper};
+use crate::routes::RequestResult;
 use crate::utils::date_utils::time;
 use crate::utils::string_utils::{check_string_before_insert, check_youtube_id};
 use crate::utils::user_data_utils::user_data_path;
+use crate::utils::webpage_utils::get_post_web_link;
 
 impl PostFile {
     /// Initialize a `PostFile` instance based on a file that have just been created
@@ -163,14 +164,18 @@ pub fn create_post(r: &mut HttpRequestHandler) -> RequestResult {
             let url = r.post_url_opt("url", true)?
                 .ok_or(ExecError::new("Missing url!"))?;
 
-            // For now, for safety, we do not fetch page content
-            // But this might change in the future
-            PostKind::POST_KIND_WEBLINK(PostWebLink {
-                url,
-                title: None,
-                description: None,
-                image: None,
-            })
+            match get_post_web_link(&url) {
+                Ok(info) => PostKind::POST_KIND_WEBLINK(info),
+                Err(e) => {
+                    eprintln!("Error while fetching page metadata: {}", e);
+                    PostKind::POST_KIND_WEBLINK(PostWebLink {
+                        url,
+                        title: None,
+                        description: None,
+                        image: None,
+                    })
+                }
+            }
         }
 
         "pdf" => {

src/utils/mod.rs

@@ -10,4 +10,5 @@ pub mod string_utils;
 pub mod pdf_utils;
 pub mod mp3_utils;
 pub mod mp4_utils;
-pub mod zip_utils;
+pub mod zip_utils;
+pub mod webpage_utils;

src/utils/webpage_utils.rs

@@ -0,0 +1,32 @@
+//! # Webpage utilities
+//!
+//! @author Pierre Hubert
+
+use std::time::Duration;
+
+use webpage::{Webpage, WebpageOptions};
+
+use crate::data::error::{ExecError, Res};
+use crate::data::post::PostWebLink;
+
+/// Attempt to get information about a URL given by a user
+pub fn get_post_web_link(url: &str) -> Res<PostWebLink> {
+    let mut options = WebpageOptions::default();
+    options.max_redirections = 2;
+    options.follow_location = true;
+    options.useragent = "OpenGraph Fetcher v1.0".to_string();
+    options.timeout = Duration::from_secs(5);
+
+    let page = Webpage::from_url(url, options)?;
+
+    /*if !page.http.content_type.to_lowercase().starts_with("text/html") {
+        return Err(ExecError::boxed_string(format!("The page at {} as an invalid content type: {}!", url, page.http.content_type)));
+    }*/
+
+    Ok(PostWebLink {
+        url: url.to_string(),
+        title: page.html.opengraph.properties.get("title").map(String::to_string),
+        description: page.html.opengraph.properties.get("description").map(String::to_string),
+        image: page.html.opengraph.images.first().map(|m| m.url.to_string()),
+    })
+}