Merge pull request #761 from CloudCannon/fix/stable-output

Stabilize filename hashes for fragments
CloudCannon · Dec 17, 2024 · a138dd1 · a138dd1
2 parents 36358de + da3c0f9
commit a138dd1
Show file tree

Hide file tree

Showing 7 changed files with 90 additions and 28 deletions.
diff --git a/pagefind/integration_tests/base/stable-output.toolproof.yml b/pagefind/integration_tests/base/stable-output.toolproof.yml
@@ -0,0 +1,61 @@
+name: Base Tests > Stable Output
+steps:
+  - ref: ./background.toolproof.yml
+  - step: I have a "public/cat/index.html" file with the content {html}
+    html: >-
+      <!DOCTYPE html><html lang="en"><head></head><body>
+        <h1>Hello world</h1>
+        <p data-pagefind-meta="animal">Cat</p>
+        <p data-pagefind-filter="species">Feline</p>
+        <p data-pagefind-sort="order">1</p>
+      </body></html>
+  - step: I have a "public/dog/index.html" file with the content {html}
+    html: >-
+      <!DOCTYPE html><html lang="en"><head></head><body>
+        <h1>Dog world</h1>
+        <p data-pagefind-meta="animal">Dog</p>
+        <p data-pagefind-filter="species">Canine</p>
+        <p data-pagefind-sort="order">2</p>
+      </body></html>
+  - macro: I run Pagefind
+  - step: stdout should contain "Running Pagefind"
+  - step: The file "public/pagefind/pagefind.js" should not be empty
+  # -----------------------------------------
+  # TODO: Stabilise the `pf_meta` file hashes
+  # -----------------------------------------
+  # - I run "ls public/pagefind"
+  # - snapshot: stdout
+  #   snapshot_content: |-
+  #     ╎filter
+  #     ╎fragment
+  #     ╎index
+  #     ╎pagefind-entry.json
+  #     ╎pagefind-highlight.js
+  #     ╎pagefind-modular-ui.css
+  #     ╎pagefind-modular-ui.js
+  #     ╎pagefind-ui.css
+  #     ╎pagefind-ui.js
+  #     ╎pagefind.en_3918d9ab34.pf_meta
+  #     ╎pagefind.js
+  #     ╎wasm.en.pagefind
+  #     ╎wasm.unknown.pagefind
+  # -----------------------------------------
+  # TODO: Stabilise the `pf_meta` file hashes
+  # -----------------------------------------
+  # - I run "ls public/pagefind/filter"
+  # - snapshot: stdout
+  #   snapshot_content: |-
+  #     ╎en_8d32c48.pf_filter
+  # -----------------------------------------
+  # TODO: Stabilise the `pf_meta` file hashes
+  # -----------------------------------------
+  # - I run "ls public/pagefind/index"
+  # - snapshot: stdout
+  #   snapshot_content: |-
+  #     ╎en_b2167ad.pf_index
+  - I run "ls public/pagefind/fragment"
+  - snapshot: stdout
+    snapshot_content: |-
+      ╎en_282213b.pf_fragment
+      ╎en_4375818.pf_fragment
+      ╎en_571daca.pf_fragment
diff --git a/pagefind/src/fossick/mod.rs b/pagefind/src/fossick/mod.rs
@@ -8,6 +8,7 @@ use lazy_static::lazy_static;
 use pagefind_stem::{Algorithm, Stemmer};
 use path_slash::PathExt as _;
 use regex::Regex;
+use std::collections::BTreeMap;
 use std::io::Error;
 use std::ops::Mul;
 use std::path::{Path, PathBuf};
@@ -46,7 +47,7 @@ pub struct FossickedData {
     pub url: String,
     pub fragment: PageFragment,
     pub word_data: HashMap<String, Vec<FossickedWord>>,
-    pub sort: HashMap<String, String>,
+    pub sort: BTreeMap<String, String>,
     pub has_custom_body: bool,
     pub force_inclusion: bool,
     pub has_html_element: bool,

diff --git a/pagefind/src/fossick/parser.rs b/pagefind/src/fossick/parser.rs
@@ -1,9 +1,9 @@
-use hashbrown::HashMap;
 use lazy_static::lazy_static;
 use lol_html::html_content::Element;
 use lol_html::{element, text, HtmlRewriter, Settings};
 use regex::Regex;
 use std::cell::RefCell;
+use std::collections::BTreeMap;
 use std::default::Default;
 use std::rc::Rc;
 
@@ -53,11 +53,11 @@ pub struct DomParser<'a> {
 #[derive(Default, Debug)]
 struct DomParserData {
     current_node: Rc<RefCell<DomParsingNode>>,
-    filters: HashMap<String, Vec<String>>,
-    sort: HashMap<String, String>,
-    meta: HashMap<String, String>,
-    default_meta: HashMap<String, String>,
-    anchor_content: HashMap<String, String>,
+    filters: BTreeMap<String, Vec<String>>,
+    sort: BTreeMap<String, String>,
+    meta: BTreeMap<String, String>,
+    default_meta: BTreeMap<String, String>,
+    anchor_content: BTreeMap<String, String>,
     language: Option<String>,
     has_html_element: bool,
     has_old_bundle_reference: bool,
@@ -104,10 +104,10 @@ struct DomParsingNode {
 #[derive(Debug)]
 pub struct DomParserResult {
     pub digest: String,
-    pub filters: HashMap<String, Vec<String>>,
-    pub sort: HashMap<String, String>,
-    pub meta: HashMap<String, String>,
-    pub anchor_content: HashMap<String, String>,
+    pub filters: BTreeMap<String, Vec<String>>,
+    pub sort: BTreeMap<String, String>,
+    pub meta: BTreeMap<String, String>,
+    pub anchor_content: BTreeMap<String, String>,
     pub has_custom_body: bool,
     pub force_inclusion: bool, // Include this page even if there is no body
     pub has_html_element: bool,

diff --git a/pagefind/src/fragments/mod.rs b/pagefind/src/fragments/mod.rs
@@ -1,4 +1,4 @@
-use hashbrown::HashMap;
+use std::collections::BTreeMap;
 
 use serde::Serialize;
 
@@ -15,8 +15,8 @@ pub struct PageFragmentData {
     pub url: String,
     pub content: String,
     pub word_count: usize,
-    pub filters: HashMap<String, Vec<String>>,
-    pub meta: HashMap<String, String>,
+    pub filters: BTreeMap<String, Vec<String>>,
+    pub meta: BTreeMap<String, String>,
     pub anchors: Vec<PageAnchorData>,
 }
 

diff --git a/pagefind/src/service/api.rs b/pagefind/src/service/api.rs
@@ -35,9 +35,8 @@
 
 pub use crate::output::SyntheticFile;
 use anyhow::{bail, Result};
-use hashbrown::HashMap;
 use rust_patch::Patch;
-use std::path::PathBuf;
+use std::{collections::BTreeMap, path::PathBuf};
 
 use crate::{
     fossick::{parser::DomParserResult, Fossicker},
@@ -49,7 +48,7 @@ use crate::{
 pub struct IndexedFileResponse {
     pub page_word_count: u32,
     pub page_url: String,
-    pub page_meta: HashMap<String, String>,
+    pub page_meta: BTreeMap<String, String>,
 }
 
 pub struct PagefindIndex {
@@ -126,16 +125,16 @@ impl PagefindIndex {
         url: String,
         content: String,
         language: String,
-        meta: Option<HashMap<String, String>>,
-        filters: Option<HashMap<String, Vec<String>>>,
-        sort: Option<HashMap<String, String>>,
+        meta: Option<BTreeMap<String, String>>,
+        filters: Option<BTreeMap<String, Vec<String>>>,
+        sort: Option<BTreeMap<String, String>>,
     ) -> Result<IndexedFileResponse> {
         let data = DomParserResult {
             digest: content,
             filters: filters.unwrap_or_default(),
             sort: sort.unwrap_or_default(),
             meta: meta.unwrap_or_default(),
-            anchor_content: HashMap::new(),
+            anchor_content: BTreeMap::new(),
             has_custom_body: false,
             force_inclusion: true,
             has_html_element: true,
@@ -214,7 +213,6 @@ impl PagefindIndex {
 #[cfg(test)]
 mod tests {
     use super::*;
-    use tokio;
 
     #[tokio::test]
     async fn test_add_file() {

diff --git a/pagefind/src/service/requests.rs b/pagefind/src/service/requests.rs
@@ -1,4 +1,5 @@
-use hashbrown::HashMap;
+use std::collections::BTreeMap;
+
 use serde::{Deserialize, Serialize};
 
 use crate::options::PagefindServiceConfig;
@@ -26,9 +27,9 @@ pub(super) enum RequestAction {
         url: String,
         content: String,
         language: String,
-        meta: Option<HashMap<String, String>>,
-        filters: Option<HashMap<String, Vec<String>>>,
-        sort: Option<HashMap<String, String>>,
+        meta: Option<BTreeMap<String, String>>,
+        filters: Option<BTreeMap<String, Vec<String>>>,
+        sort: Option<BTreeMap<String, String>>,
     },
     AddDir {
         index_id: u32,

diff --git a/pagefind/src/service/responses.rs b/pagefind/src/service/responses.rs
@@ -1,4 +1,5 @@
-use hashbrown::HashMap;
+use std::collections::BTreeMap;
+
 use serde::{Deserialize, Serialize};
 
 #[derive(Debug, Deserialize, Serialize)]
@@ -20,7 +21,7 @@ pub(super) enum ResponseAction {
     IndexedFile {
         page_word_count: u32,
         page_url: String,
-        page_meta: HashMap<String, String>,
+        page_meta: BTreeMap<String, String>,
     },
     IndexedDir {
         page_count: u32,