Implement syntax highlighting for syntax fragments

A new key 'highlighters' has been added to the toaster.conf file. The
value should be a line defining the languages to use that syntax for,
like [py/python]. The lines following are the template definitions, as
per the highlighter library.

4 files changed, 132 insertions, 2 deletions

diff --git a/Cargo.lock b/Cargo.lock
index a46e8db..ec76ce2 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -3,6 +3,49 @@
 version = 4
 
 [[package]]
+name = "aho-corasick"
+version = "1.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8e60d3430d3a69478ad0993f19238d2df97c507009a52b3c10addcd7f6bcb916"
+dependencies = [
+ "memchr",
+]
+
+[[package]]
+name = "bit-set"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "08807e080ed7f9d5433fa9b275196cfc35414f66a0c79d864dc51a0d825231a3"
+dependencies = [
+ "bit-vec",
+]
+
+[[package]]
+name = "bit-vec"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5e764a1d40d510daf35e07be9eb06e75770908c27d411ee6c92109c9840eaaf7"
+
+[[package]]
+name = "fancy-regex"
+version = "0.14.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6e24cb5a94bcae1e5408b0effca5cd7172ea3c5755049c5f3af4cd283a165298"
+dependencies = [
+ "bit-set",
+ "regex-automata",
+ "regex-syntax",
+]
+
+[[package]]
+name = "highlight"
+version = "1.0.0"
+source = "git+git://benbridle.com/highlight?tag=v1.0.0#0e9fe32e7439a597a49f9c36c56a26726100b434"
+dependencies = [
+ "fancy-regex",
+]
+
+[[package]]
 name = "inked"
 version = "1.0.0"
 source = "git+git://benbridle.com/inked?tag=v1.0.0#2954d37b638fa2c1dd3d51ff53f08f475aea6ea3"
@@ -29,6 +72,12 @@ version = "3.3.0"
 source = "git+git://benbridle.com/markdown?tag=v3.3.0#df45ffb3affb7cb1d53b567b70fef721353ccffe"
 
 [[package]]
+name = "memchr"
+version = "2.7.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "78ca9ab1a0babb1e7d5695e3530886289c18cf2f87ec19a575a0abdce112e3a3"
+
+[[package]]
 name = "paste"
 version = "1.0.15"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -40,6 +89,23 @@ version = "1.4.0"
 source = "git+git://benbridle.com/recipe?tag=v1.4.0#652aaee3130e2ee02742fdcc248ddd1bee285737"
 
 [[package]]
+name = "regex-automata"
+version = "0.4.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "809e8dc61f6de73b46c85f4c96486310fe304c434cfa43669d7b40f711150908"
+dependencies = [
+ "aho-corasick",
+ "memchr",
+ "regex-syntax",
+]
+
+[[package]]
+name = "regex-syntax"
+version = "0.8.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2b15c43186be67a4fd63bee50d0303afffcef381492ebe2c5d87f324e1b8815c"
+
+[[package]]
 name = "switchboard"
 version = "1.0.0"
 source = "git+git://benbridle.com/switchboard?tag=v1.0.0#ea70fa89659e5cf1a9d4ca6ea31fb67f7a2cc633"
@@ -61,6 +127,7 @@ dependencies = [
 name = "toaster"
 version = "1.11.0"
 dependencies = [
+ "highlight",
  "log 2.0.0",
  "markdown",
  "recipe",
diff --git a/Cargo.toml b/Cargo.toml
index 6479d3f..b0faa4f 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -9,6 +9,7 @@ markdown = { git = "git://benbridle.com/markdown", tag = "v3.3.0" }
 recipe = { git = "git://benbridle.com/recipe", tag = "v1.4.0" }
 log = { git = "git://benbridle.com/log", tag = "v2.0.0" }
 switchboard = { git = "git://benbridle.com/switchboard", tag = "v1.0.0" }
+highlight = { git = "git://benbridle.com/highlight", tag = "v1.0.0" }
 
 [profile.release]
 lto=true
diff --git a/src/collect_files.rs b/src/collect_files.rs
index 7a3c464..18d75f1 100644
--- a/src/collect_files.rs
+++ b/src/collect_files.rs
@@ -1,5 +1,6 @@
 use crate::*;
 
+use highlight::*;
 use vagabond::*;
 
 use std::collections::HashMap;
@@ -8,6 +9,7 @@ use std::collections::HashMap;
 pub struct Website {
     pub name: String,
     pub config: HashMap<String, String>,
+    pub highlighters: Highlighters,
     pub pages: Vec<Page>,
     pub redirects: Vec<Redirect>,
     pub static_files: Vec<StaticItem>,  // Redirects, !-prefixed-dir contents
@@ -60,6 +62,10 @@ pub trait LinkFrom {
     }
 }
 
+pub struct Highlighters {
+    pub languages: HashMap<String, usize>,
+    pub highlighters: Vec<Highlighter>,
+}
 
 impl Page {
     pub fn root(&self) -> String {
@@ -96,8 +102,13 @@ impl Website {
                 Err(err) => fatal!("Couldn't open {:?}: {:?}", &path, err),
             },
             config: HashMap::new(),
+            highlighters: Highlighters {
+                languages: HashMap::new(),
+                highlighters: Vec::new(),
+            },
         };
         new.collect_entry(path, path);
+        new.parse_highlighters();
         return new;
     }
 
@@ -149,7 +160,7 @@ impl Website {
             let mut key = None;
             let mut value = String::new();
             for line in config.lines() {
-                if line.starts_with(" ") {
+                if line.starts_with(" ") || line.trim().is_empty() {
                     value.push_str(line.trim());
                     value.push('\n');
                 } else {
@@ -268,6 +279,40 @@ impl Website {
         }
     }
 
+    pub fn parse_highlighters(&mut self) {
+        let mut languages = Vec::new();
+        let mut source = String::new();
+        for line in self.get_config("highlighters").lines() {
+            if let Some(line) = line.trim().strip_prefix('[') {
+                if let Some(line) = line.strip_suffix(']') {
+                    // Bank the current source.
+                    if !languages.is_empty() {
+                        let i = self.highlighters.highlighters.len();
+                        for language in languages {
+                            self.highlighters.languages.insert(language, i);
+                        }
+                        let highlighter = Highlighter::from_str(&source);
+                        self.highlighters.highlighters.push(highlighter);
+                    }
+                    languages = line.split('/').map(|s| s.trim().to_string()).collect();
+                    source.clear();
+                    continue;
+                }
+            }
+            source.push_str(line);
+            source.push('\n');
+        }
+        // Bank the current source.
+        if !languages.is_empty() {
+            let i = self.highlighters.highlighters.len();
+            for language in languages {
+                self.highlighters.languages.insert(language, i);
+            }
+            let highlighter = Highlighter::from_str(&source);
+            self.highlighters.highlighters.push(highlighter);
+        }
+    }
+
     // Ext is extension without a dot.
     // Checks if a relative link to an internal page name can be reached from
     // the current page, and returns a resolved absolute link to the page with extension.
diff --git a/src/generate_html.rs b/src/generate_html.rs
index 0de42a5..5abddf2 100644
--- a/src/generate_html.rs
+++ b/src/generate_html.rs
@@ -236,7 +236,24 @@ pub fn document_to_html(document: &MarkdownDocument, page: &Page, website: &Webs
                             warn!("Gallery-nav on page {from:?} has line without a '::' separator");
                         }
                     }),
-                    _ => wrap!("pre", format!("class='{language}'"), html!("{}", sanitize_text(content, false))),
+                    _ => wrap!("pre", format!("class='{language}'"), {
+                        if let Some(i) = website.highlighters.languages.get(language) {
+                            let mut source = String::new();
+                            let highlighter = &website.highlighters.highlighters[*i];
+                            for span in highlighter.highlight(content) {
+                                if span.tag.is_empty() {
+                                    source.push_str(&sanitize_text(&span.text, false));
+                                } else {
+                                    source.push_str(&format!("<span class='{}'>", span.tag.to_lowercase()));
+                                    source.push_str(&sanitize_text(&span.text, false));
+                                    source.push_str("</span>");
+                                }
+                            }
+                            html!("{source}");
+                        } else {
+                            html!("{}", sanitize_text(content, false))
+                        }
+                    })
                 }
             }
             Block::Break => html!("<hr>"),