feat: add performance investigation

.cargo/config.toml

@@ -0,0 +1,115 @@
+# add the below section to `.cargo/config.toml`
+
+[target.'cfg(all())']
+rustflags = [
+    # BEGIN - Embark standard lints v6 for Rust 1.55+
+    # do not change or add/remove here, but one can add exceptions after this section
+    # for more info see: <https://github.com/EmbarkStudios/rust-ecosystem/issues/59>
+    "-Dunsafe_code",
+    "-Wclippy::all",
+    "-Wclippy::await_holding_lock",
+    "-Wclippy::char_lit_as_u8",
+    "-Wclippy::checked_conversions",
+    "-Wclippy::dbg_macro",
+    "-Wclippy::debug_assert_with_mut_call",
+    "-Wclippy::doc_markdown",
+    "-Wclippy::empty_enum",
+    "-Wclippy::enum_glob_use",
+    "-Wclippy::exit",
+    "-Wclippy::expl_impl_clone_on_copy",
+    "-Wclippy::explicit_deref_methods",
+    "-Wclippy::explicit_into_iter_loop",
+    "-Wclippy::fallible_impl_from",
+    "-Wclippy::filter_map_next",
+    "-Wclippy::flat_map_option",
+    "-Wclippy::float_cmp_const",
+    "-Wclippy::fn_params_excessive_bools",
+    "-Wclippy::from_iter_instead_of_collect",
+    "-Wclippy::if_let_mutex",
+    "-Wclippy::implicit_clone",
+    "-Wclippy::imprecise_flops",
+    "-Wclippy::inefficient_to_string",
+    "-Wclippy::invalid_upcast_comparisons",
+    "-Wclippy::large_digit_groups",
+    "-Wclippy::large_stack_arrays",
+    "-Wclippy::large_types_passed_by_value",
+    "-Wclippy::let_unit_value",
+    "-Wclippy::linkedlist",
+    "-Wclippy::lossy_float_literal",
+    "-Wclippy::macro_use_imports",
+    "-Wclippy::manual_ok_or",
+    "-Wclippy::map_err_ignore",
+    "-Wclippy::map_flatten",
+    "-Wclippy::map_unwrap_or",
+    "-Wclippy::match_on_vec_items",
+    "-Wclippy::match_same_arms",
+    "-Wclippy::match_wild_err_arm",
+    "-Wclippy::match_wildcard_for_single_variants",
+    "-Wclippy::mem_forget",
+    "-Wclippy::mismatched_target_os",
+    "-Wclippy::missing_enforced_import_renames",
+    "-Wclippy::mut_mut",
+    "-Wclippy::mutex_integer",
+    "-Wclippy::needless_borrow",
+    "-Wclippy::needless_continue",
+    "-Wclippy::needless_for_each",
+    "-Wclippy::option_option",
+    "-Wclippy::path_buf_push_overwrite",
+    "-Wclippy::ptr_as_ptr",
+    "-Wclippy::rc_mutex",
+    "-Wclippy::ref_option_ref",
+    "-Wclippy::rest_pat_in_fully_bound_structs",
+    "-Wclippy::same_functions_in_if_condition",
+    "-Wclippy::semicolon_if_nothing_returned",
+    "-Wclippy::single_match_else",
+    "-Wclippy::string_add_assign",
+    "-Wclippy::string_add",
+    "-Wclippy::string_lit_as_bytes",
+    "-Wclippy::string_to_string",
+    "-Wclippy::todo",
+    "-Wclippy::trait_duplication_in_bounds",
+    "-Wclippy::unimplemented",
+    "-Wclippy::unnested_or_patterns",
+    "-Wclippy::unused_self",
+    "-Wclippy::useless_transmute",
+    "-Wclippy::verbose_file_reads",
+    "-Wclippy::zero_sized_map_values",
+    "-Wfuture_incompatible",
+    "-Wnonstandard_style",
+    "-Wrust_2018_idioms",
+    # END - Embark standard lints v6 for Rust 1.55+
+    "-Wclippy::pedantic",
+    "-Aclippy::must_use_candidate",
+    "-Aclippy::missing_errors_doc",
+    "-Wclippy::unwrap_used",
+    "-Wclippy::nursery",
+    "-Wabsolute_paths_not_starting_with_crate",
+    "-Welided_lifetimes_in_paths",
+    "-Wexplicit_outlives_requirements",
+    "-Wkeyword_idents",
+    "-Wmacro_use_extern_crate",
+    "-Wmeta_variable_misuse",
+    "-Wmissing_abi",
+    "-Wmissing_copy_implementations",
+    "-Wmissing_debug_implementations",
+    "-Wnon_ascii_idents",
+    "-Wnoop_method_call",
+    "-Wpointer_structural_match",
+    "-Wrust_2021_incompatible_closure_captures",
+    "-Wrust_2021_incompatible_or_patterns",
+    "-Wrust_2021_prefixes_incompatible_syntax",
+    "-Wrust_2021_prelude_collisions",
+    "-Wsingle_use_lifetimes",
+    "-Wtrivial_casts",
+    "-Wtrivial_numeric_casts",
+    "-Wunsafe_code",
+    "-Wunsafe_op_in_unsafe_fn",
+    "-Wunstable_features",
+    "-Wunused_extern_crates",
+    "-Wunused_import_braces",
+    "-Wunused_lifetimes",
+    "-Wunused_macro_rules",
+    "-Wunused_qualifications",
+    "-Wunused_results",
+    "-Wvariant_size_differences",
+]

Cargo.toml

@@ -25,3 +25,10 @@ tracing-subscriber = { version = "0.3", default-features = false, features = ["f
 
 [features]
 aws-lambda = [ "dep:lambda_runtime", "dep:tokio", "dep:tracing", "dep:tracing-subscriber" ]
+
+[dev-dependencies]
+criterion = { version = "0.4.0", features = [ "html_reports" ]}
+
+[[bench]]
+name = "bench_main"
+harness = false

benches/bench_main.rs

@@ -0,0 +1,9 @@
+use criterion::criterion_main;
+
+mod benchmarks;
+
+criterion_main! {
+  benchmarks::map_to_html::map_to_html_bench,
+  benchmarks::download_blogs::download_blogs_bench,
+  benchmarks::get_page::get_page_bench,
+}

benches/benchmarks/download_blogs.rs

@@ -0,0 +1,14 @@
+use criterion::{Criterion, criterion_group};
+use rss2email::download_blogs;
+
+pub fn criterion_benchmark(c: &mut Criterion) {
+  let mut group = c.benchmark_group("download blogs");
+  group.bench_function("download blogs", |f| f.iter(|| download_blogs(1)));
+  group.finish();
+}
+
+criterion_group! {
+    name = download_blogs_bench;
+    config = Criterion::default().sample_size(10);
+    targets = criterion_benchmark
+}

benches/benchmarks/get_page.rs

@@ -0,0 +1,38 @@
+use std::time::Duration;
+
+use criterion::{BenchmarkId, Criterion, SamplingMode, criterion_group};
+use regex::Regex;
+use rss2email::{get_page, read_feeds};
+
+pub fn criterion_benchmark(c: &mut Criterion) {
+  let mut group = c.benchmark_group("get page");
+  group.sampling_mode(SamplingMode::Flat);
+
+  let feeds = read_feeds();
+
+  // Ok, for this it's probably best to check here first https://regex101.com/r/KyY0vd/1
+  // but basically, I wanted to get some parts of the URLs that hopefully won't cause
+  // collisions in the benchmark IDs.
+  let seperator = Regex::new(r"://([a-zA-Z0-9.?-]+)/?([a-zA-Z0-9?-]+)?(?:[^=\n]+)?(?:=(\w+))?")
+    .expect("Invalid regex");
+
+  for feed in feeds {
+    let captures = seperator.captures(&feed).unwrap();
+    let p1 = captures.get(1).map_or("Regex failed", |m| m.as_str());
+    let p2 = captures.get(2).map_or("", |m| m.as_str());
+    let p3 = captures.get(3).map_or("", |m| m.as_str());
+    let name = p1.to_owned() + "/" + p2 + "/" + p3;
+
+    group.bench_with_input(BenchmarkId::from_parameter(&name), &feed, |b, feed| {
+      b.iter(|| get_page(feed));
+    });
+  }
+
+  group.finish();
+}
+
+criterion_group! {
+    name = get_page_bench;
+    config = Criterion::default().sample_size(10).measurement_time(Duration::from_secs(4));
+    targets = criterion_benchmark
+}

benches/benchmarks/map_to_html.rs

@@ -0,0 +1,29 @@
+use chrono::{FixedOffset, TimeZone};
+use criterion::{criterion_group, Criterion};
+use rss2email::{map_to_html, Blog, Post};
+
+fn mock_blog() -> Vec<Blog> {
+  let date_time = FixedOffset::east(1000).ymd(1970, 3, 22).and_hms(1, 1, 1);
+  let dummy_str = "a";
+
+  let p = Post {
+    title: dummy_str.to_string(),
+    link: dummy_str.to_string(),
+    description: Some(dummy_str.to_string()),
+    last_build_date: date_time,
+  };
+
+  vec![Blog {
+    title: dummy_str.to_string(),
+    last_build_date: date_time,
+    posts: vec![p],
+  }]
+}
+
+pub fn criterion_benchmark(c: &mut Criterion) {
+  let blogs = mock_blog();
+
+  c.bench_function("map to html", |b| b.iter(|| map_to_html(&blogs)));
+}
+
+criterion_group!(map_to_html_bench, criterion_benchmark);

benches/benchmarks/mod.rs

@@ -0,0 +1,5 @@
+#![allow(unused_results, clippy::unwrap_used, clippy::string_add)]
+
+pub mod download_blogs;
+pub mod get_page;
+pub mod map_to_html;

src/util.rs → src/lib.rs

@@ -7,10 +7,11 @@ use log::{info, warn};
 use regex::Regex;
 use std::fmt::Write as _;
 
-use crate::{
-  blog::{Blog, Post},
-  xml::parse_web_feed,
-};
+pub use blog::{Blog, Post};
+mod blog;
+mod xml;
+
+use crate::xml::parse_web_feed;
 
 /// Downloads all the RSS feeds specified in `feeds.txt` and converts them to `Blog`s.
 pub fn download_blogs(days: i64) -> Vec<Blog> {
@@ -55,7 +56,7 @@ pub fn download_blogs(days: i64) -> Vec<Blog> {
 ///
 /// Assumed one link per line. Any text between a `#` and a line end
 /// is considered a comment.
-fn read_feeds() -> Vec<String> {
+pub fn read_feeds() -> Vec<String> {
   let links = fs::read_to_string("feeds.txt").expect("Error in reading the feeds.txt file");
 
   // Not really necessary but yes
@@ -101,7 +102,7 @@ fn within_n_days(n: i64, date: &DateTime<FixedOffset>) -> bool {
 }
 
 /// Helper function for downloading the contents of a web page.
-fn get_page(url: &str) -> Result<String, ureq::Error> {
+pub fn get_page(url: &str) -> Result<String, ureq::Error> {
   let body: String = ureq::get(url)
     .set("Example-Header", "header value")
     .call()?

src/main.rs

@@ -1,53 +1,11 @@
-#![warn(
-  rust_2018_idioms,
-  clippy::pedantic,
-  clippy::unwrap_used,
-  clippy::nursery,
-  absolute_paths_not_starting_with_crate,
-  elided_lifetimes_in_paths,
-  explicit_outlives_requirements,
-  keyword_idents,
-  macro_use_extern_crate,
-  meta_variable_misuse,
-  missing_abi,
-  missing_copy_implementations,
-  missing_debug_implementations,
-  non_ascii_idents,
-  noop_method_call,
-  pointer_structural_match,
-  rust_2021_incompatible_closure_captures,
-  rust_2021_incompatible_or_patterns,
-  rust_2021_prefixes_incompatible_syntax,
-  rust_2021_prelude_collisions,
-  single_use_lifetimes,
-  trivial_casts,
-  trivial_numeric_casts,
-  unsafe_code,
-  unsafe_op_in_unsafe_fn,
-  unstable_features,
-  unused_crate_dependencies,
-  unused_extern_crates,
-  unused_import_braces,
-  unused_lifetimes,
-  unused_macro_rules,
-  unused_qualifications,
-  unused_results,
-  variant_size_differences
-)]
-
 use crate::email::email_provider::get_email_provider;
 use crate::email::email_provider::EmailProvider;
-use crate::util::{download_blogs, map_to_html};
 use dotenv::dotenv;
 use env_logger::Env;
 use log::{error, info};
+use rss2email::{download_blogs, map_to_html, time_func};
 
-use crate::util::time_func;
-
-mod blog;
 mod email;
-mod util;
-mod xml;
 
 fn core_main() -> Result<(), String> {
   env_logger::Builder::from_env(Env::default().default_filter_or("info")).init();
@@ -89,7 +47,7 @@ fn core_main() -> Result<(), String> {
     // Only load email related variables if ran on release
     let api_key = std::env::var("API_KEY").expect("API_KEY must be set.");
     let address = std::env::var("EMAIL_ADDRESS").expect("EMAIL_ADDRESS must be set.");
-    
+
     get_email_provider().send_email(&address, &api_key, &html);
   }