From e2af729061f4065ef78e3202dbaf3293a432aedb Mon Sep 17 00:00:00 2001
From: j-mendez <jeff@a11ywatch.com>
Date: Wed, 27 Dec 2023 07:16:20 -0500
Subject: [PATCH] chore(builder): add depth and caching

---
 Cargo.toml          |  4 ++--
 book/src/website.md | 28 ++++++++++++++++++++++++++++
 src/page.rs         | 15 ++++++++++++++-
 src/website.rs      |  8 +++++++-
 4 files changed, 51 insertions(+), 4 deletions(-)

diff --git a/Cargo.toml b/Cargo.toml
index dd9dd2b..e1d23a9 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,7 +1,7 @@
 [package]
 edition = "2021"
 name = "spider_rs"
-version = "0.0.11"
+version = "0.0.12"
 description = "The fastest web crawler written in Rust ported to nodejs."
 repository = "https://github.com/spider-rs/spider-nodejs"
 
@@ -11,7 +11,7 @@ crate-type = ["cdylib"]
 [dependencies]
 indexmap = "2.1.0"
 num_cpus = "1.16.0"
-spider = { version = "1.80.24", features = ["budget", "cron", "regex", "cookies", "socks", "chrome", "control", "smart", "chrome_intercept" ] }
+spider = { version = "1.80.26", features = ["budget", "cron", "regex", "cookies", "socks", "chrome", "control", "smart", "chrome_intercept", "cache" ] }
 pyo3 = { version = "0.20.0", features = ["extension-module"] }
 pyo3-asyncio = { version = "0.20", features = ["attributes", "tokio-runtime"] }
 
diff --git a/book/src/website.md b/book/src/website.md
index 6b12a28..0147899 100644
--- a/book/src/website.md
+++ b/book/src/website.md
@@ -136,6 +136,34 @@ async def main():
 asyncio.run(main())
 ```
 
+### Depth Limit
+
+Set the depth limit for the amount of forward pages.
+
+```ts
+import asyncio
+from spider_rs import Website
+
+async def main():
+    website = Website("https://choosealicense.com").with_depth(3)
+
+asyncio.run(main())
+```
+
+### Cache
+
+Enable HTTP caching, this useful when using the spider on a server.
+
+```py
+import asyncio
+from spider_rs import Website
+
+async def main():
+    website = Website("https://choosealicense.com").with_caching(True)
+
+asyncio.run(main())
+```
+
 ### Delays
 
 Add delays between pages. Defaults to none.
diff --git a/src/page.rs b/src/page.rs
index 3f3b4cb..9202ba4 100644
--- a/src/page.rs
+++ b/src/page.rs
@@ -34,9 +34,22 @@ impl Page {
 
   /// get the page content
   pub fn fetch(mut slf: PyRefMut<'_, Self>) -> PyRefMut<'_, Self> {
+    use spider::{
+      lazy_static::lazy_static, reqwest::Client, reqwest_middleware::ClientWithMiddleware,
+      ClientBuilder,
+    };
+    lazy_static! {
+      /// top level single page client to re-use.
+      pub static ref PAGE_CLIENT: ClientWithMiddleware = {
+        let reqwest_client = Client::builder().build().unwrap_or_default();
+        let client = ClientBuilder::new(reqwest_client).build();
+
+        client
+      };
+    }
     let s = pyo3_asyncio::tokio::get_runtime()
       .block_on(async move {
-        let page = spider::page::Page::new_page(&slf.url, &Default::default()).await;
+        let page = spider::page::Page::new_page(&slf.url, &PAGE_CLIENT).await;
         slf.status_code = page.status_code.into();
         slf.inner = Some(page);
         slf.selectors = spider::page::get_page_selectors(
diff --git a/src/website.rs b/src/website.rs
index 94827bf..d3f563d 100644
--- a/src/website.rs
+++ b/src/website.rs
@@ -746,12 +746,18 @@ impl Website {
     slf
   }
 
-  /// Set a crawl depth limit. If the value is 0 there is no limit. This does nothing without the feat flag [budget] enabled.
+  /// Set a crawl depth limit. If the value is 0 there is no limit.
   pub fn with_depth(mut slf: PyRefMut<'_, Self>, depth: usize) -> PyRefMut<'_, Self> {
     slf.inner.with_depth(depth);
     slf
   }
 
+  /// Cache the page following HTTP rules.
+  pub fn with_caching(mut slf: PyRefMut<'_, Self>, cache: bool) -> PyRefMut<'_, Self> {
+    slf.inner.with_caching(cache);
+    slf
+  }
+
   /// add external domains
   pub fn with_external_domains(
     mut slf: PyRefMut<'_, Self>,