From c451a12e44c40fcc61d0c5594d5bd5bad9e5143d Mon Sep 17 00:00:00 2001
From: Olivier 'reivilibre <olivier@librepush.net>
Date: Sun, 12 Jun 2022 15:26:44 +0100
Subject: [PATCH] Pass the bytes through when extracting HTML

---
 quickpeep_raker/src/raking.rs                 | 8 +++-----
 quickpeep_raker/src/raking/page_extraction.rs | 7 ++++---
 2 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/quickpeep_raker/src/raking.rs b/quickpeep_raker/src/raking.rs
index 6aed87e..b98fc06 100644
--- a/quickpeep_raker/src/raking.rs
+++ b/quickpeep_raker/src/raking.rs
@@ -396,7 +396,7 @@ impl Raker {
         {
             // We don't try any fallbacks for an HTML page
             return Ok(self
-                .rake_html_page(&content, url, is_cf, &headers)
+                .rake_html_page(content, url, is_cf, &headers)
                 .await
                 .context("Raking HTML page")?);
         }
@@ -445,16 +445,14 @@ impl Raker {
 
     pub async fn rake_html_page(
         &self,
-        content: &[u8],
+        content: Vec<u8>,
         url: &Url,
         is_cf: bool,
         headers: &HeaderMap,
     ) -> anyhow::Result<RakeOutcome> {
-        let content_str = std::str::from_utf8(content)?.to_owned();
-
         match self
             .page_extraction
-            .extract(content_str, url.clone(), headers.clone(), is_cf)
+            .extract(content, url.clone(), headers.clone(), is_cf)
             .await?
         {
             ExtractedPage::Success {
diff --git a/quickpeep_raker/src/raking/page_extraction.rs b/quickpeep_raker/src/raking/page_extraction.rs
index 9422726..1f0c213 100644
--- a/quickpeep_raker/src/raking/page_extraction.rs
+++ b/quickpeep_raker/src/raking/page_extraction.rs
@@ -29,7 +29,7 @@ pub struct PageExtractionService {
 }
 
 pub struct ExtractionTask {
-    content: String,
+    content: Vec<u8>,
     url: Url,
     headers: HeaderMap,
     is_cf: bool,
@@ -39,7 +39,7 @@ pub struct ExtractionTask {
 impl PageExtractionService {
     pub async fn extract(
         &self,
-        content: String,
+        content: Vec<u8>,
         url: Url,
         headers: HeaderMap,
         is_cf: bool,
@@ -111,11 +111,12 @@ struct PageExtractionServiceInternal {
 impl PageExtractionServiceInternal {
     fn extract_page(
         &self,
-        content_str: String,
+        content_bytes: Vec<u8>,
         url: Url,
         headers: HeaderMap,
         is_cf: bool,
     ) -> anyhow::Result<ExtractedPage> {
+        let content_str: &str = todo!();
         let root_node: NodeRef = kuchiki::parse_html().one(content_str.as_ref());
 
         // See whether this page is at the canonical URL for the page.