add test comment 😅

* [REFACTORING]Rename whitespace_tokenizer to tag_tokenizer for
registration

Name representing its purpose is preferred.

* Add lindera-tantivy to plume-model's dependencies

* Install lindera-tantivy

* Add SearchTokenizerConfig struct

* Add search tokenizers to config option

* Use CONFIG for tokenizers

* Use enum to hold tokenizer config instead of initializing on config phase

* Use guard instead of duplicate default values

* Use as_deref() instead of guard

* Move SearchTokenizer from plume-models to plume-models::search::tokenizer

* Rename SearchTokenizer to TokenizerKind

* Define SearchTokenierConfig::determine_tokenizer()

* Use determine_tokenizer in SearchTokenizerConfig::init()

* Pass tokenizer config to Searcher methods

* Add LowerCase filter to Lindera tokenizer

* Add test for Lindera tokenizer

* Define SEARCH_LANG env to specify tokenizers set

* Run cargo fmt

* Make Lindera tokenizer optional

* Fix typos

Cargo.toml

@@ -24,7 +24,6 @@ rocket = "0.4.2"
 rocket_contrib = { version = "0.4.2", features = ["json"] }
 rocket_i18n = { git = "https://github.com/Plume-org/rocket_i18n", rev = "e922afa7c366038b3433278c03b1456b346074f2" }
 rpassword = "4.0"
-runtime-fmt = "0.4.0"
 scheduled-thread-pool = "0.2.2"
 serde = "1.0"
 serde_json = "1.0"
@@ -79,6 +78,7 @@ postgres = ["plume-models/postgres", "diesel/postgres"]
 sqlite = ["plume-models/sqlite", "diesel/sqlite"]
 debug-mailer = []
 test = []
+search-lindera = ["plume-models/search-lindera"]
 
 [workspace]
 members = ["plume-api", "plume-cli", "plume-models", "plume-common", "plume-front", "plume-macro"]

plume-cli/Cargo.toml

@@ -23,3 +23,4 @@ path = "../plume-models"
 [features]
 postgres = ["plume-models/postgres", "diesel/postgres"]
 sqlite = ["plume-models/sqlite", "diesel/sqlite"]
+search-lindera = ["plume-models/search-lindera"]

plume-cli/src/search.rs

@@ -82,7 +82,7 @@ fn init<'a>(args: &ArgMatches<'a>, conn: &Connection) {
         }
     };
     if can_do || force {
-        let searcher = Searcher::create(&path).unwrap();
+        let searcher = Searcher::create(&path, &CONFIG.search_tokenizers).unwrap();
         refill(args, conn, Some(searcher));
     } else {
         eprintln!(
@@ -98,7 +98,8 @@ fn refill<'a>(args: &ArgMatches<'a>, conn: &Connection, searcher: Option<Searche
         Some(path) => Path::new(path).join("search_index"),
         None => Path::new(&CONFIG.search_index).to_path_buf(),
     };
-    let searcher = searcher.unwrap_or_else(|| Searcher::open(&path).unwrap());
+    let searcher =
+        searcher.unwrap_or_else(|| Searcher::open(&path, &CONFIG.search_tokenizers).unwrap());
 
     searcher.fill(conn).expect("Couldn't import post");
     println!("Commiting result");

plume-common/Cargo.toml

@@ -22,6 +22,7 @@ serde_json = "1.0"
 shrinkwraprs = "0.2.1"
 syntect = "3.3"
 tokio = "0.1.22"
+regex-syntax = { version = "0.6.17", default-features = false, features = ["unicode-perl"] }
 
 [dependencies.chrono]
 features = ["serde"]

plume-common/src/utils.rs

@@ -1,6 +1,7 @@
 use heck::CamelCase;
 use openssl::rand::rand_bytes;
 use pulldown_cmark::{html, Event, Options, Parser, Tag};
+use regex_syntax::is_word_character;
 use rocket::{
     http::uri::Uri,
     response::{Flash, Redirect},
@@ -200,6 +201,12 @@ fn process_image<'a, 'b>(
     }
 }
 
+#[derive(Default, Debug)]
+struct DocumentContext {
+    in_code: bool,
+    in_link: bool,
+}
+
 /// Returns (HTML, mentions, hashtags)
 pub fn md_to_html<'a>(
     md: &str,
@@ -223,13 +230,21 @@ pub fn md_to_html<'a>(
         .map(|evt| process_image(evt, inline, &media_processor))
         // Ignore headings, images, and tables if inline = true
         .scan((vec![], inline), inline_tags)
-        .scan(false, |in_code, evt| match evt {
+        .scan(&mut DocumentContext::default(), |ctx, evt| match evt {
             Event::Start(Tag::CodeBlock(_)) | Event::Start(Tag::Code) => {
-                *in_code = true;
+                ctx.in_code = true;
                 Some((vec![evt], vec![], vec![]))
             }
             Event::End(Tag::CodeBlock(_)) | Event::End(Tag::Code) => {
-                *in_code = false;
+                ctx.in_code = false;
+                Some((vec![evt], vec![], vec![]))
+            }
+            Event::Start(Tag::Link(_, _)) => {
+                ctx.in_link = true;
+                Some((vec![evt], vec![], vec![]))
+            }
+            Event::End(Tag::Link(_, _)) => {
+                ctx.in_link = false;
                 Some((vec![evt], vec![], vec![]))
             }
             Event::Text(txt) => {
@@ -269,7 +284,7 @@ pub fn md_to_html<'a>(
                                 }
                             }
                             State::Hashtag => {
-                                let char_matches = c.is_alphanumeric() || "-_".contains(c);
+                                let char_matches = c == '-' || is_word_character(c);
                                 if char_matches && (n < (txt.chars().count() - 1)) {
                                     text_acc.push(c);
                                     (events, State::Hashtag, text_acc, n + 1, mentions, hashtags)
@@ -300,7 +315,7 @@ pub fn md_to_html<'a>(
                                 }
                             }
                             State::Ready => {
-                                if !*in_code && c == '@' {
+                                if !ctx.in_code && !ctx.in_link && c == '@' {
                                     events.push(Event::Text(text_acc.into()));
                                     (
                                         events,
@@ -310,7 +325,7 @@ pub fn md_to_html<'a>(
                                         mentions,
                                         hashtags,
                                     )
-                                } else if !*in_code && c == '#' {
+                                } else if !ctx.in_code && !ctx.in_link && c == '#' {
                                     events.push(Event::Text(text_acc.into()));
                                     (
                                         events,
@@ -399,6 +414,7 @@ mod tests {
             ("not_a@mention", vec![]),
             ("`@helo`", vec![]),
             ("```\n@hello\n```", vec![]),
+            ("[@atmark in link](https://example.org/)", vec![]),
         ];
 
         for (md, mentions) in tests {
@@ -424,6 +440,9 @@ mod tests {
             ("with some punctuation #test!", vec!["test"]),
             (" #spaces     ", vec!["spaces"]),
             ("not_a#hashtag", vec![]),
+            ("#نرم‌افزار_آزاد", vec!["نرم‌افزار_آزاد"]),
+            ("[#hash in link](https://example.org/)", vec![]),
+            ("#zwsp\u{200b}inhash", vec!["zwsp"]),
         ];
 
         for (md, mentions) in tests {

plume-front/src/main.rs

@@ -19,6 +19,7 @@ init_i18n!(
     en,
     eo,
     es,
+    fa,
     fr,
     gl,
     hi,
@@ -45,6 +46,9 @@ lazy_static! {
         let lang = js! { return navigator.language }.into_string().unwrap();
         let lang = lang.splitn(2, '-').next().unwrap_or("en");
 
+        // Force a language (Add .env setting)
+        //let lang = "fa"; ???
+
         let english_position = catalogs
             .iter()
             .position(|(language_code, _)| *language_code == "en")

plume-models/Cargo.toml

@@ -22,7 +22,7 @@ scheduled-thread-pool = "0.2.2"
 serde = "1.0"
 serde_derive = "1.0"
 serde_json = "1.0"
-tantivy = "0.10.1"
+tantivy = "0.12.0"
 url = "2.1"
 walkdir = "2.2"
 webfinger = "0.4.1"
@@ -30,6 +30,7 @@ whatlang = "0.7.1"
 shrinkwraprs = "0.2.1"
 diesel-derive-newtype = "0.1.2"
 glob = "0.3.0"
+lindera-tantivy = { version = "0.1.2", optional = true }
 
 [dependencies.chrono]
 features = ["serde"]
@@ -54,3 +55,4 @@ diesel_migrations = "1.3.0"
 [features]
 postgres = ["diesel/postgres", "plume-macro/postgres" ]
 sqlite = ["diesel/sqlite", "plume-macro/sqlite" ]
+search-lindera = ["lindera-tantivy"]

plume-models/src/blogs.rs

@@ -498,6 +498,7 @@ pub(crate) mod tests {
     use super::*;
     use crate::{
         blog_authors::*,
+        config::CONFIG,
         instance::tests as instance_tests,
         medias::NewMedia,
         search::tests::get_searcher,
@@ -767,7 +768,9 @@ pub(crate) mod tests {
         conn.test_transaction::<_, (), _>(|| {
             let (_, blogs) = fill_database(conn);
 
-            blogs[0].delete(conn, &get_searcher()).unwrap();
+            blogs[0]
+                .delete(conn, &get_searcher(&CONFIG.search_tokenizers))
+                .unwrap();
             assert!(Blog::get(conn, blogs[0].id).is_err());
             Ok(())
         })
@@ -777,7 +780,7 @@ pub(crate) mod tests {
     fn delete_via_user() {
         let conn = &db();
         conn.test_transaction::<_, (), _>(|| {
-            let searcher = get_searcher();
+            let searcher = get_searcher(&CONFIG.search_tokenizers);
             let (user, _) = fill_database(conn);
 
             let b1 = Blog::insert(

plume-models/src/config.rs

@@ -1,3 +1,4 @@
+use crate::search::TokenizerKind as SearchTokenizer;
 use rocket::config::Limits;
 use rocket::Config as RocketConfig;
 use std::env::{self, var};
@@ -11,7 +12,10 @@ pub struct Config {
     pub base_url: String,
     pub database_url: String,
     pub db_name: &'static str,
+    pub db_max_size: Option<u32>,
+    pub db_min_idle: Option<u32>,
     pub search_index: String,
+    pub search_tokenizers: SearchTokenizerConfig,
     pub rocket: Result<RocketConfig, RocketError>,
     pub logo: LogoConfig,
     pub default_theme: String,
@@ -186,6 +190,56 @@ impl Default for LogoConfig {
     }
 }
 
+pub struct SearchTokenizerConfig {
+    pub tag_tokenizer: SearchTokenizer,
+    pub content_tokenizer: SearchTokenizer,
+    pub property_tokenizer: SearchTokenizer,
+}
+
+impl SearchTokenizerConfig {
+    pub fn init() -> Self {
+        use SearchTokenizer::*;
+
+        match var("SEARCH_LANG").ok().as_deref() {
+            Some("ja") => {
+                #[cfg(not(feature = "search-lindera"))]
+                panic!("You need build Plume with search-lindera feature, or execute it with SEARCH_TAG_TOKENIZER=ngram and SEARCH_CONTENT_TOKENIZER=ngram to enable Japanese search feature");
+                #[cfg(feature = "search-lindera")]
+                Self {
+                    tag_tokenizer: Self::determine_tokenizer("SEARCH_TAG_TOKENIZER", Lindera),
+                    content_tokenizer: Self::determine_tokenizer(
+                        "SEARCH_CONTENT_TOKENIZER",
+                        Lindera,
+                    ),
+                    property_tokenizer: Ngram,
+                }
+            }
+            _ => Self {
+                tag_tokenizer: Self::determine_tokenizer("SEARCH_TAG_TOKENIZER", Whitespace),
+                content_tokenizer: Self::determine_tokenizer("SEARCH_CONTENT_TOKENIZER", Simple),
+                property_tokenizer: Ngram,
+            },
+        }
+    }
+
+    fn determine_tokenizer(var_name: &str, default: SearchTokenizer) -> SearchTokenizer {
+        use SearchTokenizer::*;
+
+        match var(var_name).ok().as_deref() {
+            Some("simple") => Simple,
+            Some("ngram") => Ngram,
+            Some("whitespace") => Whitespace,
+            Some("lindera") => {
+                #[cfg(not(feature = "search-lindera"))]
+                panic!("You need build Plume with search-lindera feature to use Lindera tokenizer");
+                #[cfg(feature = "search-lindera")]
+                Lindera
+            }
+            _ => default,
+        }
+    }
+}
+
 lazy_static! {
     pub static ref CONFIG: Config = Config {
         base_url: var("BASE_URL").unwrap_or_else(|_| format!(
@@ -193,12 +247,21 @@ lazy_static! {
             var("ROCKET_PORT").unwrap_or_else(|_| "7878".to_owned())
         )),
         db_name: DB_NAME,
+        db_max_size: var("DB_MAX_SIZE").map_or(None, |s| Some(
+            s.parse::<u32>()
+                .expect("Couldn't parse DB_MAX_SIZE into u32")
+        )),
+        db_min_idle: var("DB_MIN_IDLE").map_or(None, |s| Some(
+            s.parse::<u32>()
+                .expect("Couldn't parse DB_MIN_IDLE into u32")
+        )),
         #[cfg(feature = "postgres")]
         database_url: var("DATABASE_URL")
             .unwrap_or_else(|_| format!("postgres://plume:plume@localhost/{}", DB_NAME)),
         #[cfg(feature = "sqlite")]
         database_url: var("DATABASE_URL").unwrap_or_else(|_| format!("{}.sqlite", DB_NAME)),
         search_index: var("SEARCH_INDEX").unwrap_or_else(|_| "search_index".to_owned()),
+        search_tokenizers: SearchTokenizerConfig::init(),
         rocket: get_rocket_config(),
         logo: LogoConfig::default(),
         default_theme: var("DEFAULT_THEME").unwrap_or_else(|_| "default-light".to_owned()),

plume-models/src/lib.rs

@@ -320,7 +320,7 @@ mod tests {
     pub fn rockets() -> super::PlumeRocket {
         super::PlumeRocket {
             conn: db_conn::DbConn((*DB_POOL).get().unwrap()),
-            searcher: Arc::new(search::tests::get_searcher()),
+            searcher: Arc::new(search::tests::get_searcher(&CONFIG.search_tokenizers)),
             worker: Arc::new(ScheduledThreadPool::new(2)),
             user: None,
         }

plume-models/src/search/mod.rs

@@ -3,10 +3,11 @@ mod searcher;
 mod tokenizer;
 pub use self::query::PlumeQuery as Query;
 pub use self::searcher::*;
+pub use self::tokenizer::TokenizerKind;
 
 #[cfg(test)]
 pub(crate) mod tests {
-    use super::{Query, Searcher};
+    use super::{Query, Searcher, TokenizerKind};
     use diesel::Connection;
     use plume_common::utils::random_hex;
     use std::env::temp_dir;
@@ -14,18 +15,20 @@ pub(crate) mod tests {
 
     use crate::{
         blogs::tests::fill_database,
+        config::SearchTokenizerConfig,
         post_authors::*,
         posts::{NewPost, Post},
         safe_string::SafeString,
         tests::db,
+        CONFIG,
     };
 
-    pub(crate) fn get_searcher() -> Searcher {
+    pub(crate) fn get_searcher(tokenizers: &SearchTokenizerConfig) -> Searcher {
         let dir = temp_dir().join(&format!("plume-test-{}", random_hex()));
         if dir.exists() {
-            Searcher::open(&dir)
+            Searcher::open(&dir, tokenizers)
         } else {
-            Searcher::create(&dir)
+            Searcher::create(&dir, tokenizers)
         }
         .unwrap()
     }
@@ -100,27 +103,27 @@ pub(crate) mod tests {
     fn open() {
         let dir = temp_dir().join(format!("plume-test-{}", random_hex()));
         {
-            Searcher::create(&dir).unwrap();
+            Searcher::create(&dir, &CONFIG.search_tokenizers).unwrap();
         }
-        Searcher::open(&dir).unwrap();
+        Searcher::open(&dir, &CONFIG.search_tokenizers).unwrap();
     }
 
     #[test]
     fn create() {
         let dir = temp_dir().join(format!("plume-test-{}", random_hex()));
 
-        assert!(Searcher::open(&dir).is_err());
+        assert!(Searcher::open(&dir, &CONFIG.search_tokenizers).is_err());
         {
-            Searcher::create(&dir).unwrap();
+            Searcher::create(&dir, &CONFIG.search_tokenizers).unwrap();
         }
-        Searcher::open(&dir).unwrap(); //verify it's well created
+        Searcher::open(&dir, &CONFIG.search_tokenizers).unwrap(); //verify it's well created
     }
 
     #[test]
     fn search() {
         let conn = &db();
         conn.test_transaction::<_, (), _>(|| {
-            let searcher = get_searcher();
+            let searcher = get_searcher(&CONFIG.search_tokenizers);
             let blog = &fill_database(conn).1[0];
             let author = &blog.list_authors(conn).unwrap()[0];
 
@@ -180,10 +183,69 @@ pub(crate) mod tests {
         });
     }
 
+    #[cfg(feature = "search-lindera")]
+    #[test]
+    fn search_japanese() {
+        let conn = &db();
+        conn.test_transaction::<_, (), _>(|| {
+            let tokenizers = SearchTokenizerConfig {
+                tag_tokenizer: TokenizerKind::Lindera,
+                content_tokenizer: TokenizerKind::Lindera,
+                property_tokenizer: TokenizerKind::Ngram,
+            };
+            let searcher = get_searcher(&tokenizers);
+            let blog = &fill_database(conn).1[0];
+
+            let title = random_hex()[..8].to_owned();
+
+            let post = Post::insert(
+                conn,
+                NewPost {
+                    blog_id: blog.id,
+                    slug: title.clone(),
+                    title: title.clone(),
+                    content: SafeString::new("ブログエンジンPlumeです。"),
+                    published: true,
+                    license: "CC-BY-SA".to_owned(),
+                    ap_url: "".to_owned(),
+                    creation_date: None,
+                    subtitle: "".to_owned(),
+                    source: "".to_owned(),
+                    cover_id: None,
+                },
+                &searcher,
+            )
+            .unwrap();
+
+            searcher.commit();
+
+            assert_eq!(
+                searcher.search_document(conn, Query::from_str("ブログエンジン").unwrap(), (0, 1))
+                    [0]
+                .id,
+                post.id
+            );
+            assert_eq!(
+                searcher.search_document(conn, Query::from_str("Plume").unwrap(), (0, 1))[0].id,
+                post.id
+            );
+            assert_eq!(
+                searcher.search_document(conn, Query::from_str("です").unwrap(), (0, 1))[0].id,
+                post.id
+            );
+            assert_eq!(
+                searcher.search_document(conn, Query::from_str("。").unwrap(), (0, 1))[0].id,
+                post.id
+            );
+
+            Ok(())
+        });
+    }
+
     #[test]
     fn drop_writer() {
-        let searcher = get_searcher();
+        let searcher = get_searcher(&CONFIG.search_tokenizers);
         searcher.drop_writer();
-        get_searcher();
+        get_searcher(&CONFIG.search_tokenizers);
     }
 }

plume-models/src/search/searcher.rs

@@ -1,18 +1,14 @@
 use crate::{
-    instance::Instance,
-    posts::Post,
-    schema::posts,
-    search::{query::PlumeQuery, tokenizer},
-    tags::Tag,
-    Connection, Result,
+    config::SearchTokenizerConfig, instance::Instance, posts::Post, schema::posts,
+    search::query::PlumeQuery, tags::Tag, Connection, Result,
 };
 use chrono::Datelike;
 use diesel::{ExpressionMethods, QueryDsl, RunQueryDsl};
 use itertools::Itertools;
 use std::{cmp, fs::create_dir_all, path::Path, sync::Mutex};
 use tantivy::{
-    collector::TopDocs, directory::MmapDirectory, schema::*, tokenizer::*, Index, IndexReader,
-    IndexWriter, ReloadPolicy, Term,
+    collector::TopDocs, directory::MmapDirectory, schema::*, Index, IndexReader, IndexWriter,
+    ReloadPolicy, Term,
 };
 use whatlang::{detect as detect_lang, Lang};
 
@@ -34,7 +30,7 @@ impl Searcher {
     pub fn schema() -> Schema {
         let tag_indexing = TextOptions::default().set_indexing_options(
             TextFieldIndexing::default()
-                .set_tokenizer("whitespace_tokenizer")
+                .set_tokenizer("tag_tokenizer")
                 .set_index_option(IndexRecordOption::Basic),
         );
 
@@ -70,15 +66,7 @@ impl Searcher {
         schema_builder.build()
     }
 
-    pub fn create(path: &dyn AsRef<Path>) -> Result<Self> {
-        let whitespace_tokenizer = tokenizer::WhitespaceTokenizer.filter(LowerCaser);
-
-        let content_tokenizer = SimpleTokenizer
-            .filter(RemoveLongFilter::limit(40))
-            .filter(LowerCaser);
-
-        let property_tokenizer = NgramTokenizer::new(2, 8, false).filter(LowerCaser);
-
+    pub fn create(path: &dyn AsRef<Path>, tokenizers: &SearchTokenizerConfig) -> Result<Self> {
         let schema = Self::schema();
 
         create_dir_all(path).map_err(|_| SearcherError::IndexCreationError)?;
@@ -90,9 +78,9 @@ impl Searcher {
 
         {
             let tokenizer_manager = index.tokenizers();
-            tokenizer_manager.register("whitespace_tokenizer", whitespace_tokenizer);
-            tokenizer_manager.register("content_tokenizer", content_tokenizer);
-            tokenizer_manager.register("property_tokenizer", property_tokenizer);
+            tokenizer_manager.register("tag_tokenizer", tokenizers.tag_tokenizer);
+            tokenizer_manager.register("content_tokenizer", tokenizers.content_tokenizer);
+            tokenizer_manager.register("property_tokenizer", tokenizers.property_tokenizer);
         } //to please the borrow checker
         Ok(Self {
             writer: Mutex::new(Some(
@@ -109,31 +97,38 @@ impl Searcher {
         })
     }
 
-    pub fn open(path: &dyn AsRef<Path>) -> Result<Self> {
-        let whitespace_tokenizer = tokenizer::WhitespaceTokenizer.filter(LowerCaser);
-
-        let content_tokenizer = SimpleTokenizer
-            .filter(RemoveLongFilter::limit(40))
-            .filter(LowerCaser);
-
-        let property_tokenizer = NgramTokenizer::new(2, 8, false).filter(LowerCaser);
-
-        let index =
+    pub fn open(path: &dyn AsRef<Path>, tokenizers: &SearchTokenizerConfig) -> Result<Self> {
+        let mut index =
             Index::open(MmapDirectory::open(path).map_err(|_| SearcherError::IndexOpeningError)?)
                 .map_err(|_| SearcherError::IndexOpeningError)?;
 
         {
             let tokenizer_manager = index.tokenizers();
-            tokenizer_manager.register("whitespace_tokenizer", whitespace_tokenizer);
-            tokenizer_manager.register("content_tokenizer", content_tokenizer);
-            tokenizer_manager.register("property_tokenizer", property_tokenizer);
+            tokenizer_manager.register("tag_tokenizer", tokenizers.tag_tokenizer);
+            tokenizer_manager.register("content_tokenizer", tokenizers.content_tokenizer);
+            tokenizer_manager.register("property_tokenizer", tokenizers.property_tokenizer);
         } //to please the borrow checker
-        let mut writer = index
+        let writer = index
             .writer(50_000_000)
             .map_err(|_| SearcherError::WriteLockAcquisitionError)?;
-        writer
-            .garbage_collect_files()
+
+        // Since Tantivy v0.12.0, IndexWriter::garbage_collect_files() returns Future.
+        // To avoid conflict with Plume async project, we don't introduce async now.
+        // After async is introduced to Plume, we can use garbage_collect_files() again.
+        // Algorithm stolen from Tantivy's SegmentUpdater::list_files()
+        use std::collections::HashSet;
+        use std::path::PathBuf;
+        let mut files: HashSet<PathBuf> = index
+            .list_all_segment_metas()
+            .into_iter()
+            .flat_map(|segment_meta| segment_meta.list_files())
+            .collect();
+        files.insert(Path::new("meta.json").to_path_buf());
+        index
+            .directory_mut()
+            .garbage_collect(|| files)
             .map_err(|_| SearcherError::IndexEditionError)?;
+
         Ok(Self {
             writer: Mutex::new(Some(writer)),
             reader: index

plume-models/src/search/tokenizer.rs

@@ -1,5 +1,34 @@
+#[cfg(feature = "search-lindera")]
+use lindera_tantivy::tokenizer::LinderaTokenizer;
 use std::str::CharIndices;
-use tantivy::tokenizer::{Token, TokenStream, Tokenizer};
+use tantivy::tokenizer::*;
+
+#[derive(Clone, Copy)]
+pub enum TokenizerKind {
+    Simple,
+    Ngram,
+    Whitespace,
+    #[cfg(feature = "search-lindera")]
+    Lindera,
+}
+
+impl From<TokenizerKind> for TextAnalyzer {
+    fn from(tokenizer: TokenizerKind) -> TextAnalyzer {
+        use TokenizerKind::*;
+
+        match tokenizer {
+            Simple => TextAnalyzer::from(SimpleTokenizer)
+                .filter(RemoveLongFilter::limit(40))
+                .filter(LowerCaser),
+            Ngram => TextAnalyzer::from(NgramTokenizer::new(2, 8, false)).filter(LowerCaser),
+            Whitespace => TextAnalyzer::from(WhitespaceTokenizer).filter(LowerCaser),
+            #[cfg(feature = "search-lindera")]
+            Lindera => {
+                TextAnalyzer::from(LinderaTokenizer::new("decompose", "")).filter(LowerCaser)
+            }
+        }
+    }
+}
 
 /// Tokenize the text by splitting on whitespaces. Pretty much a copy of Tantivy's SimpleTokenizer,
 /// but not splitting on punctuation
@@ -12,15 +41,13 @@ pub struct WhitespaceTokenStream<'a> {
     token: Token,
 }
 
-impl<'a> Tokenizer<'a> for WhitespaceTokenizer {
-    type TokenStreamImpl = WhitespaceTokenStream<'a>;
-
-    fn token_stream(&self, text: &'a str) -> Self::TokenStreamImpl {
-        WhitespaceTokenStream {
+impl Tokenizer for WhitespaceTokenizer {
+    fn token_stream<'a>(&self, text: &'a str) -> BoxTokenStream<'a> {
+        BoxTokenStream::from(WhitespaceTokenStream {
             text,
             chars: text.char_indices(),
             token: Token::default(),
-        }
+        })
     }
 }
 impl<'a> WhitespaceTokenStream<'a> {

plume-models/src/timeline/query.rs

@@ -406,8 +406,8 @@ impl Bool {
                 }
             }
             Bool::HasCover => Ok(post.cover_id.is_some()),
-            Bool::Local => Ok(post.get_blog(&rocket.conn)?.is_local()),
-            Bool::All => Ok(true),
+            Bool::Local => Ok(post.get_blog(&rocket.conn)?.is_local() && kind == Kind::Original),
+            Bool::All => Ok(kind == Kind::Original),
         }
     }
 }

plume-models/src/users.rs

@@ -760,7 +760,7 @@ impl User {
                     mime_type: None,
                     href: None,
                     template: Some(format!(
-                        "https://{}/remote_interact?{{uri}}",
+                        "https://{}/remote_interact?target={{uri}}",
                         self.get_instance(conn)?.public_domain
                     )),
                 },
@@ -1026,6 +1026,7 @@ impl NewUser {
 pub(crate) mod tests {
     use super::*;
     use crate::{
+        config::CONFIG,
         instance::{tests as instance_tests, Instance},
         search::tests::get_searcher,
         tests::{db, rockets},
@@ -1122,7 +1123,9 @@ pub(crate) mod tests {
             let inserted = fill_database(conn);
 
             assert!(User::get(conn, inserted[0].id).is_ok());
-            inserted[0].delete(conn, &get_searcher()).unwrap();
+            inserted[0]
+                .delete(conn, &get_searcher(&CONFIG.search_tokenizers))
+                .unwrap();
             assert!(User::get(conn, inserted[0].id).is_err());
             Ok(())
         });

po/plume-front/fr.po

@@ -10,7 +10,7 @@ msgstr ""
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=UTF-8\n"
 "Content-Transfer-Encoding: 8bit\n"
-"Plural-Forms: nplurals=2; plural=(n > 1);\n"
+"Plural-Forms: nplurals=2; plural=(n != 1);\n"
 "X-Crowdin-Project: plume\n"
 "X-Crowdin-Language: fr\n"
 "X-Crowdin-File: /master/po/plume-front/plume-front.pot\n"

po/plume/fr.po

@@ -10,7 +10,7 @@ msgstr ""
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=UTF-8\n"
 "Content-Transfer-Encoding: 8bit\n"
-"Plural-Forms: nplurals=2; plural=(n > 1);\n"
+"Plural-Forms: nplurals=2; plural=(n != 1);\n"
 "X-Crowdin-Project: plume\n"
 "X-Crowdin-Language: fr\n"
 "X-Crowdin-File: /master/po/plume/plume.pot\n"

src/main.rs

@@ -6,8 +6,6 @@ extern crate gettext_macros;
 #[macro_use]
 extern crate rocket;
 #[macro_use]
-extern crate runtime_fmt;
-#[macro_use]
 extern crate serde_json;
 #[macro_use]
 extern crate validator_derive;
@@ -28,7 +26,8 @@ use std::sync::{Arc, Mutex};
 use std::time::Duration;
 
 init_i18n!(
-    "plume", ar, bg, ca, cs, de, en, eo, es, fr, gl, hi, hr, it, ja, nb, pl, pt, ro, ru, sr, sk, sv
+    "plume", ar, bg, ca, cs, de, en, eo, es, fa, fr, gl, hi, hr, it, ja, nb, pl, pt, ro, ru, sr,
+    sk, sv
 );
 
 mod api;
@@ -55,10 +54,13 @@ fn init_pool() -> Option<DbPool> {
     }
 
     let manager = ConnectionManager::<Connection>::new(CONFIG.database_url.as_str());
-    let pool = DbPool::builder()
+    let mut builder = DbPool::builder()
         .connection_customizer(Box::new(PragmaForeignKey))
-        .build(manager)
-        .ok()?;
+        .min_idle(CONFIG.db_min_idle);
+    if let Some(max_size) = CONFIG.db_max_size {
+        builder = builder.max_size(max_size);
+    };
+    let pool = builder.build(manager).ok()?;
     Instance::cache_local(&pool.get().unwrap());
     Some(pool)
 }
@@ -97,7 +99,7 @@ Then try to restart Plume.
     let workpool = ScheduledThreadPool::with_name("worker {}", num_cpus::get());
     // we want a fast exit here, so
     #[allow(clippy::match_wild_err_arm)]
-    let searcher = match UnmanagedSearcher::open(&CONFIG.search_index) {
+    let searcher = match UnmanagedSearcher::open(&CONFIG.search_index, &CONFIG.search_tokenizers) {
         Err(Error::Search(e)) => match e {
             SearcherError::WriteLockAcquisitionError => panic!(
                 r#"

src/routes/blogs.rs

@@ -1,5 +1,4 @@
 use activitypub::collection::{OrderedCollection, OrderedCollectionPage};
-use atom_syndication::{Entry, FeedBuilder};
 use diesel::SaveChangesDsl;
 use rocket::{
     http::ContentType,
@@ -361,20 +360,13 @@ pub fn outbox_page(
 pub fn atom_feed(name: String, rockets: PlumeRocket) -> Option<Content<String>> {
     let blog = Blog::find_by_fqn(&rockets, &name).ok()?;
     let conn = &*rockets.conn;
-    let feed = FeedBuilder::default()
-        .title(blog.title.clone())
-        .id(Instance::get_local()
-            .ok()?
-            .compute_box("~", &name, "atom.xml"))
-        .entries(
-            Post::get_recents_for_blog(&*conn, &blog, 15)
-                .ok()?
-                .into_iter()
-                .map(|p| super::post_to_atom(p, &*conn))
-                .collect::<Vec<Entry>>(),
-        )
-        .build()
-        .ok()?;
+    let entries = Post::get_recents_for_blog(&*conn, &blog, 15).ok()?;
+    let uri = Instance::get_local()
+        .ok()?
+        .compute_box("~", &name, "atom.xml");
+    let title = &blog.title;
+    let default_updated = &blog.creation_date;
+    let feed = super::build_atom_feed(entries, &uri, title, default_updated, conn);
     Some(Content(
         ContentType::new("application", "atom+xml"),
         feed.to_string(),

src/routes/mod.rs

@@ -1,6 +1,9 @@
 #![warn(clippy::too_many_arguments)]
 use crate::template_utils::Ructe;
-use atom_syndication::{ContentBuilder, Entry, EntryBuilder, LinkBuilder, Person, PersonBuilder};
+use atom_syndication::{
+    ContentBuilder, Entry, EntryBuilder, Feed, FeedBuilder, LinkBuilder, Person, PersonBuilder,
+};
+use chrono::naive::NaiveDateTime;
 use plume_models::{posts::Post, Connection, CONFIG, ITEMS_PER_PAGE};
 use rocket::{
     http::{
@@ -115,13 +118,46 @@ pub struct RemoteForm {
     pub remote: String,
 }
 
-pub fn post_to_atom(post: Post, conn: &Connection) -> Entry {
+pub fn build_atom_feed(
+    entries: Vec<Post>,
+    uri: &str,
+    title: &str,
+    default_updated: &NaiveDateTime,
+    conn: &Connection,
+) -> Feed {
+    let updated = if entries.is_empty() {
+        default_updated
+    } else {
+        &entries[0].creation_date
+    };
+
+    FeedBuilder::default()
+        .title(title)
+        .id(uri)
+        .updated(updated.format("%Y-%m-%dT%H:%M:%SZ").to_string())
+        .entries(
+            entries
+                .into_iter()
+                .map(|p| post_to_atom(p, conn))
+                .collect::<Vec<Entry>>(),
+        )
+        .links(vec![LinkBuilder::default()
+            .href(uri)
+            .rel("self")
+            .mime_type("application/atom+xml".to_string())
+            .build()
+            .expect("Atom feed: link error")])
+        .build()
+        .expect("user::atom_feed: Error building Atom feed")
+}
+
+fn post_to_atom(post: Post, conn: &Connection) -> Entry {
+    let formatted_creation_date = post.creation_date.format("%Y-%m-%dT%H:%M:%SZ").to_string();
     EntryBuilder::default()
         .title(format!("<![CDATA[{}]]>", post.title))
         .content(
             ContentBuilder::default()
                 .value(format!("<![CDATA[{}]]>", *post.content.get()))
-                .src(post.ap_url.clone())
                 .content_type("html".to_string())
                 .build()
                 .expect("Atom feed: content error"),
@@ -141,8 +177,9 @@ pub fn post_to_atom(post: Post, conn: &Connection) -> Entry {
         )
         // Using RFC 4287 format, see https://tools.ietf.org/html/rfc4287#section-3.3 for dates
         // eg: 2003-12-13T18:30:02Z (Z is here because there is no timezone support with the NaiveDateTime crate)
-        .published(post.creation_date.format("%Y-%m-%dT%H:%M:%SZ").to_string())
-        .id(post.id.to_string())
+        .published(formatted_creation_date.clone())
+        .updated(formatted_creation_date)
+        .id(post.ap_url.clone())
         .links(vec![LinkBuilder::default()
             .href(post.ap_url)
             .build()

src/routes/posts.rs

@@ -643,7 +643,7 @@ pub fn remote_interact_post(
         .and_then(|blog| Post::find_by_slug(&rockets.conn, &slug, blog.id))?;
     if let Some(uri) = User::fetch_remote_interact_uri(&remote.remote)
         .ok()
-        .and_then(|uri| rt_format!(uri, uri = target.ap_url).ok())
+        .map(|uri| uri.replace("{uri}", &target.ap_url))
     {
         Ok(Redirect::to(uri).into())
     } else {

src/routes/user.rs

@@ -2,7 +2,6 @@ use activitypub::{
     activity::Create,
     collection::{OrderedCollection, OrderedCollectionPage},
 };
-use atom_syndication::{Entry, FeedBuilder};
 use diesel::SaveChangesDsl;
 use rocket::{
     http::{ContentType, Cookies},
@@ -202,15 +201,14 @@ pub fn follow_not_connected(
         if let Some(uri) = User::fetch_remote_interact_uri(&remote_form)
             .ok()
             .and_then(|uri| {
-                rt_format!(
-                    uri,
-                    uri = format!(
+                Some(uri.replace(
+                    "{uri}",
+                    &format!(
                         "{}@{}",
                         target.fqn,
                         target.get_instance(&rockets.conn).ok()?.public_domain
-                    )
-                )
-                .ok()
+                    ),
+                ))
             })
         {
             Ok(Redirect::to(uri).into())
@@ -619,20 +617,13 @@ pub fn ap_followers(
 pub fn atom_feed(name: String, rockets: PlumeRocket) -> Option<Content<String>> {
     let conn = &*rockets.conn;
     let author = User::find_by_fqn(&rockets, &name).ok()?;
-    let feed = FeedBuilder::default()
-        .title(author.display_name.clone())
-        .id(Instance::get_local()
-            .unwrap()
-            .compute_box("@", &name, "atom.xml"))
-        .entries(
-            Post::get_recents_for_author(conn, &author, 15)
-                .ok()?
-                .into_iter()
-                .map(|p| super::post_to_atom(p, conn))
-                .collect::<Vec<Entry>>(),
-        )
-        .build()
-        .expect("user::atom_feed: Error building Atom feed");
+    let entries = Post::get_recents_for_author(conn, &author, 15).ok()?;
+    let uri = Instance::get_local()
+        .ok()?
+        .compute_box("@", &name, "atom.xml");
+    let title = &author.display_name;
+    let default_updated = &author.creation_date;
+    let feed = super::build_atom_feed(entries, &uri, title, default_updated, conn);
     Some(Content(
         ContentType::new("application", "atom+xml"),
         feed.to_string(),

templates/posts/details.rs.html

@@ -40,7 +40,7 @@
             <div class="article-info" dir="auto">
                 <span class="author">
                     @Html(i18n!(ctx.1, "Written by {0}"; format!("<a href=\"{}\">{}</a>",
-                                        uri!(user::details: name = &author.fqn),
+                                        escape(&uri!(user::details: name = &author.fqn).to_string()),
                                         escape(&author.name()))))
                 </span>
                 &mdash;
@@ -103,8 +103,8 @@
             </section>
         } else {
             <p class="center">@Html(i18n!(ctx.1, "{0}Log in{1}, or {2}use your Fediverse account{3} to interact with this article";
-                format!("<a href='{}'>", uri!(session::new: m = _)), "</a>",
-                format!("<a href='{}'>", uri!(posts::remote_interact: blog_name = &blog.fqn, slug = &article.slug)), "</a>"
+                format!("<a href='{}'>", escape(&uri!(session::new: m = _).to_string())), "</a>",
+                format!("<a href='{}'>", escape(&uri!(posts::remote_interact: blog_name = &blog.fqn, slug = &article.slug).to_string())), "</a>"
                 ))
             </p>
             <section class="actions">