From 0c21f99aeed52afb7d8785492710e2ba3ef0c9c7 Mon Sep 17 00:00:00 2001
From: Clark Hensley <clarkhensley@protonmail.com>
Date: Tue, 25 Apr 2023 18:34:11 -0500
Subject: [PATCH 01/13] Sorta fixing tables. Definitely fixed emojis

---
 Markdown_HTML.js               | 11 ++---------
 Markdown_MD.js                 |  2 +-
 client/src/tinymce-webpack.mts |  4 +++-
 server/src/webserver.rs        |  3 +--
 4 files changed, 7 insertions(+), 13 deletions(-)

diff --git a/Markdown_HTML.js b/Markdown_HTML.js
index a98d463..8175233 100644
--- a/Markdown_HTML.js
+++ b/Markdown_HTML.js
@@ -16,7 +16,7 @@
 // <h5>h5 Text</h5>
 // <h6>h6 Text</h6>
 // <p><span style="text-decoration: underline;">Underlined Text</span></p>
-// <p><s><span style="text-decoration: underline;">Strikethrough Text</span></s>
+// <p><s>Strikethrough Text</s>
 // </p>
 // <ul>
 //     <li>Unordered List 1</li>
@@ -119,14 +119,7 @@
 //         height="314" allowfullscreen="allowfullscreen"></iframe></p>
 // <p>&nbsp;</p>
 // <p>Some Emoji (more unicode)</p>
-// <p><img style="width: 1em; height: 1em; margin: 0 .05em 0 .1em; vertical-align: -.1em;"
-//         draggable="false"
-//         src="https://twemoji.maxcdn.com/v/13.0.1/72x72/1f4af.png" alt="💯"
-//         data-emoticon="true"> <img
-//         style="width: 1em; height: 1em; margin: 0 .05em 0 .1em; vertical-align: -.1em;"
-//         draggable="false"
-//         src="https://twemoji.maxcdn.com/v/13.0.1/72x72/1f171.png" alt="🅱️"
-//         data-emoticon="true"></p>
+// <p>💯🅱️</p>
 // <p>&nbsp;</p>
 // <p>And some images:</p>
 // <p><img src="CodeChat1.png" alt=""></p>
diff --git a/Markdown_MD.js b/Markdown_MD.js
index ff322ad..2b0b890 100644
--- a/Markdown_MD.js
+++ b/Markdown_MD.js
@@ -96,7 +96,7 @@
 //
 // Some Emoji (more unicode)
 //
-// ![💯](https://twemoji.maxcdn.com/v/13.0.1/72x72/1f4af.png) ![🅱️](https://twemoji.maxcdn.com/v/13.0.1/72x72/1f171.png)
+// 💯🅱️
 //
 // And some images:
 //
diff --git a/client/src/tinymce-webpack.mts b/client/src/tinymce-webpack.mts
index 4d50a43..254113d 100644
--- a/client/src/tinymce-webpack.mts
+++ b/client/src/tinymce-webpack.mts
@@ -42,7 +42,6 @@ import "tinymce/plugins/code/index.js";
 import "tinymce/plugins/directionality/index.js";
 import "tinymce/plugins/emoticons/index.js";
 import "tinymce/plugins/emoticons/js/emojis.js";
-import "tinymce/plugins/emoticons/js/emojiimages.js";
 import "tinymce/plugins/help/index.js";
 import "tinymce/plugins/image/index.js";
 import "tinymce/plugins/link/index.js";
@@ -76,6 +75,9 @@ export const tinymce_init = async (
             // </p>
             plugins:
                 "advlist anchor charmap directionality emoticons help image link lists media nonbreaking pagebreak quickbars searchreplace table visualblocks visualchars",
+            // <p>Using this to disable the emoji images and force the unicode instead
+            // </p>
+            emoticons_database: "emojis",
             // <p>The imports above apply the skins; don't try to dynamically
             //     load the skin's CSS.</p>
             skin: false,
diff --git a/server/src/webserver.rs b/server/src/webserver.rs
index 3b0cf9d..590d88b 100644
--- a/server/src/webserver.rs
+++ b/server/src/webserver.rs
@@ -646,8 +646,7 @@ async fn serve_file(
     };
 
     // <p>Convert doc blocks from Markdown to HTML</p>
-    let mut options = Options::empty();
-    options.insert(Options::ENABLE_STRIKETHROUGH);
+    let options = Options::all();
     for code_doc_block in &mut code_doc_block_arr {
         if let CodeDocBlock::DocBlock(ref mut doc_block) = code_doc_block {
             let parser = Parser::new_ext(&doc_block.contents, options);

From 9ea18d58701c30db63e81e8d19c446d20e4fe7a1 Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Wed, 26 Apr 2023 23:38:03 -0500
Subject: [PATCH 02/13] Add: Custom Class to help Tables

---
 client/src/tinymce-webpack.mts       | 64 ++++++++++++----------------
 client/static/css/CodeChatEditor.css | 10 +++++
 2 files changed, 37 insertions(+), 37 deletions(-)

diff --git a/client/src/tinymce-webpack.mts b/client/src/tinymce-webpack.mts
index 254113d..7e7ece7 100644
--- a/client/src/tinymce-webpack.mts
+++ b/client/src/tinymce-webpack.mts
@@ -1,40 +1,34 @@
-// <details>
-//     <summary>Copyright (C) 2022 Bryan A. Jones.</summary>
-//     <p>This file is part of the CodeChat Editor.</p>
-//     <p>The CodeChat Editor is free software: you can redistribute it and/or
-//         modify it under the terms of the GNU General Public License as
-//         published by the Free Software Foundation, either version 3 of the
-//         License, or (at your option) any later version.</p>
-//     <p>The CodeChat Editor is distributed in the hope that it will be useful,
-//         but WITHOUT ANY WARRANTY; without even the implied warranty of
-//         MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
-//         General Public License for more details.</p>
-//     <p>You should have received a copy of the GNU General Public License
-//         along with the CodeChat Editor. If not, see <a
-//             href="http://www.gnu.org/licenses/">http://www.gnu.org/licenses/</a>.
-//     </p>
-// </details>
-// <h1><code>tinymce-webpack.ts</code> &mdash; imports the TinyMCE editor from
-//     NPM packages using webpack</h1>
-// <p>Import TinyMCE.</p>
+// Copyright (C) 2022 Bryan A. Jones.
+//
+// This file is part of the CodeChat Editor.
+//
+// The CodeChat Editor is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
+//
+// The CodeChat Editor is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+//
+// You should have received a copy of the GNU General Public License along with the CodeChat Editor. If not, see [http://www.gnu.org/licenses/](http://www.gnu.org/licenses/).
+//
+// `tinymce-webpack.ts` — imports the TinyMCE editor from NPM packages using webpack
+// =================================================================================
+//
+// Import TinyMCE.
 import tinymce, { RawEditorOptions } from "tinymce";
 export { tinymce };
 
-// <p>Default icons are required for TinyMCE 5.3 or above.</p>
+// Default icons are required for TinyMCE 5.3 or above.
 import "tinymce/icons/default/index.js";
 
-// <p>A theme is also required.</p>
+// A theme is also required.
 import "tinymce/themes/silver/index.js";
 
-// <p>Import the skin to use; use're using an inline editor, so load the inline
-//     CSS.</p>
+// Import the skin to use; use're using an inline editor, so load the inline CSS.
 import "tinymce/skins/ui/oxide/skin.css";
 import "tinymce/skins/ui/oxide/content.inline.css";
 
-// <p>Without this, TinyMCE produces errors.</p>
+// Without this, TinyMCE produces errors.
 import "tinymce/models/dom/index.js";
 
-// <p>Import plugins.</p>
+// Import plugins.
 import "tinymce/plugins/advlist/index.js";
 import "tinymce/plugins/anchor/index.js";
 import "tinymce/plugins/charmap/index.js";
@@ -55,31 +49,27 @@ import "tinymce/plugins/table/index.js";
 import "tinymce/plugins/visualblocks/index.js";
 import "tinymce/plugins/visualchars/index.js";
 
-// <p>Import premium plugins. NOTE: Download separately and add these to
-//     /src/plugins.</p>
+// Import premium plugins. NOTE: Download separately and add these to /src/plugins.
 /// import './plugins/checklist/plugin';
 /// import './plugins/powerpaste/plugin';
 /// import './plugins/powerpaste/js/wordimport';
 
-// <p>Initialize TinyMCE.</p>
+// Initialize TinyMCE.
 export const tinymce_init = async (
-    // <p>Provide editor options; don't set ``plugins`` or ``skin``, since these
-    //     must be accompanied by the correct imports.</p>
+    // Provide editor options; don't set \`\`plugins\`\` or \`\`skin\`\`, since these must be accompanied by the correct imports.
     options: RawEditorOptions
 ) =>
     /// @ts-ignore
     tinymce.init(
         Object.assign({}, options, {
-            // <p>See the list of <a
-            //         href="https://www.tiny.cloud/docs/tinymce/6/plugins/">plugins</a>.
-            // </p>
+            // See the list of [plugins](https://www.tiny.cloud/docs/tinymce/6/plugins/).
             plugins:
                 "advlist anchor charmap directionality emoticons help image link lists media nonbreaking pagebreak quickbars searchreplace table visualblocks visualchars",
-            // <p>Using this to disable the emoji images and force the unicode instead
-            // </p>
+            // Using this to disable the emoji images and force the unicode instead
             emoticons_database: "emojis",
-            // <p>The imports above apply the skins; don't try to dynamically
-            //     load the skin's CSS.</p>
+            // Add this option to add a custom class to tables. This class adds a black border to the table. 
+            visual_table_class : "CodeChat-TinyMCE-TableDefault",
+            // The imports above apply the skins; don't try to dynamically load the skin's CSS.
             skin: false,
         })
     );
diff --git a/client/static/css/CodeChatEditor.css b/client/static/css/CodeChatEditor.css
index a2cbed3..b2c2d24 100644
--- a/client/static/css/CodeChatEditor.css
+++ b/client/static/css/CodeChatEditor.css
@@ -207,3 +207,13 @@ body {
     margin-bottom: 0px;
     padding-bottom: 0px;
 }
+
+.CodeChat-TinyMCE-TableDefault { 
+    border: 1px solid black;
+    border-collapse: collapse;
+}
+
+.CodeChat-TinyMCE-TableDefault th, td{ 
+    border: 1px solid black;
+    border-collapse: collapse;
+}
\ No newline at end of file

From 537a4419663b8166a971c66987090381c75b7cb5 Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Wed, 26 Apr 2023 23:55:59 -0500
Subject: [PATCH 03/13] Final Writeup

---
 HTMLtoMarkdown_Prewrite.cchtml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/HTMLtoMarkdown_Prewrite.cchtml b/HTMLtoMarkdown_Prewrite.cchtml
index 0d3d378..cf07bfe 100644
--- a/HTMLtoMarkdown_Prewrite.cchtml
+++ b/HTMLtoMarkdown_Prewrite.cchtml
@@ -125,3 +125,8 @@
     preserve style tags. A final PR would, hopefully, pass all tests and be as
     efficient as possible in how we preserve formatting, perhaps by
     implementing a turndown filter, or by exploring other avenues.</p>
+
+<h2> Final PR Results </h2>
+<p> All of the standard HTML to Markdown conversions are working along with non-standard Markdown conversions tables, task lists, and strikethrough. The style attributes are not currently being saved so many styling features do not save. The tables have been added and fixed so that they can properly convert both ways, and a custom style class has been added so that a black border is added. On converting back to HTML, the width for the tables is lost and reset to auto. The emojis have been fixed so they can properly convert as well. </p>
+
+<p>We were unable to figure out how to add custom rules into pulldown-cmark's library. We attempted to create custom events, but could not get the custom events to properly work. The custom rules could easily be added to the HTML to Markdown side, but not to the Markdown to HTML side.</p>
\ No newline at end of file

From de7c0369f843d4ed9514b73868c343a7465408dd Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Sat, 29 Apr 2023 01:33:35 -0500
Subject: [PATCH 04/13] Add way to insert new Markdown conversions

---
 Markdown_MD.js                       |  2 +-
 client/src/CodeChatEditor.mts        | 10 ++++++
 client/src/markdown_conversions.json | 18 +++++++++++
 client/tsconfig.json                 |  3 +-
 server/src/webserver.rs              | 48 ++++++++++++++++++++++++++--
 5 files changed, 77 insertions(+), 4 deletions(-)
 create mode 100644 client/src/markdown_conversions.json

diff --git a/Markdown_MD.js b/Markdown_MD.js
index 2b0b890..58fa353 100644
--- a/Markdown_MD.js
+++ b/Markdown_MD.js
@@ -42,7 +42,7 @@
 // *   [x] Task list 2
 // *   [ ] Task list 3
 //
-// Text in superscript and subscript
+// Text in ^superscript^ and ~subscript~
 //
 // `for good measure, here's a code-block`
 //
diff --git a/client/src/CodeChatEditor.mts b/client/src/CodeChatEditor.mts
index 4514ecf..14193f4 100644
--- a/client/src/CodeChatEditor.mts
+++ b/client/src/CodeChatEditor.mts
@@ -35,6 +35,7 @@ import { html_beautify } from "js-beautify";
 import { tinymce, tinymce_init } from "./tinymce-webpack.mjs";
 import { gfm } from '@joplin/turndown-plugin-gfm';
 import TurndownService from "turndown";
+import MarkdownConversions from "../src/markdown_conversions.json";
 
 
 // ### CSS
@@ -48,6 +49,15 @@ const turndownService = new TurndownService();
 
 // Add the plugins from [turndown-plugin-gfm](https://github.com/laurent22/joplin/tree/dev/packages/turndown-plugin-gfm) to enable conversions for tables, task lists, and strikethroughs.
 turndownService.use(gfm);
+for (let i = 0; i < MarkdownConversions.conversions.length; i++) {
+    turndownService.addRule(MarkdownConversions.conversions[i].name, {
+        filter: [MarkdownConversions.conversions[i].filter_tag],
+        replacement: function (content: string) {
+          return MarkdownConversions.conversions[i].markdown + content + MarkdownConversions.conversions[i].markdown
+        }
+      })
+}
+
 
   
 // Load code when the DOM is ready.
diff --git a/client/src/markdown_conversions.json b/client/src/markdown_conversions.json
new file mode 100644
index 0000000..b38b281
--- /dev/null
+++ b/client/src/markdown_conversions.json
@@ -0,0 +1,18 @@
+{
+   "conversions": [
+        {
+            "name": "subscript",
+            "filter_tag": "sub",
+            "open_tag": "<sub>",
+            "close_tag": "</sub>",
+            "markdown": "~"
+        },
+        {
+            "name": "superscript",
+            "filter_tag": "sup",
+            "open_tag": "<sup>",
+            "close_tag": "</sup>",
+            "markdown": "^"
+        }
+    ]
+}
\ No newline at end of file
diff --git a/client/tsconfig.json b/client/tsconfig.json
index 34843ea..b41bd08 100644
--- a/client/tsconfig.json
+++ b/client/tsconfig.json
@@ -36,7 +36,8 @@
         "outDir": "out",
         "rootDir": "src",
         "strict": true,
-        "target": "ES2022"
+        "target": "ES2022",
+        "resolveJsonModule": true
     },
     "exclude": ["node_modules", "static"]
 }
diff --git a/server/src/webserver.rs b/server/src/webserver.rs
index 590d88b..e628bf8 100644
--- a/server/src/webserver.rs
+++ b/server/src/webserver.rs
@@ -646,10 +646,54 @@ async fn serve_file(
     };
 
     // <p>Convert doc blocks from Markdown to HTML</p>
-    let options = Options::all();
+    #[derive(Debug, Deserialize)]
+    struct Conversion {
+        conversions: Vec<Entry>
+    }
+    #[derive(Debug, Deserialize)]
+    struct Entry {
+        open_tag: String,
+        close_tag: String,
+        markdown: char,
+    }
+    let mut file = File::open("../client/src/markdown_conversions.json").await.unwrap();
+    let mut buff = String::new();
+    file.read_to_string(&mut buff).await.unwrap();
+ 
+    let conversion: Conversion = serde_json::from_str(&buff).unwrap();
+
     for code_doc_block in &mut code_doc_block_arr {
         if let CodeDocBlock::DocBlock(ref mut doc_block) = code_doc_block {
-            let parser = Parser::new_ext(&doc_block.contents, options);
+            let mut output_contents = Vec::new();
+            let mut tag_open = false;
+            for c in doc_block.contents.chars() { 
+                let mut replaced = false;
+                for entry in &conversion.conversions {
+                    if c == entry.markdown {
+                        if tag_open {
+                            for tag_char in entry.close_tag.chars() {
+                                output_contents.push(tag_char);
+                            }
+                        } 
+                        else {
+                            for tag_char in entry.open_tag.chars() {
+                                output_contents.push(tag_char);
+                            }
+                        }
+                        tag_open = !tag_open;
+                        replaced = true;
+                    } 
+                }
+                if !replaced {
+                    {
+                        output_contents.push(c);
+                    }
+                }
+            }
+            let output: String = output_contents.into_iter().map(|i| i.to_string()).collect::<String>();
+            
+            let options = Options::all();
+            let parser = Parser::new_ext(&output, options);
             let mut html_output = String::new();
             html::push_html(&mut html_output, parser);
             doc_block.contents = html_output;

From 47138e6514933dd5f6dfca954c702f339ca5e8c5 Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Sat, 29 Apr 2023 02:18:44 -0500
Subject: [PATCH 05/13] Add documentation & fix strikethrough

---
 HTMLtoMarkdown_Prewrite.cchtml |   3 +-
 client/src/CodeChatEditor.mts  |  46 ++---
 server/src/webserver.rs        | 299 ++++++++++++++-------------------
 3 files changed, 153 insertions(+), 195 deletions(-)

diff --git a/HTMLtoMarkdown_Prewrite.cchtml b/HTMLtoMarkdown_Prewrite.cchtml
index cf07bfe..90a8943 100644
--- a/HTMLtoMarkdown_Prewrite.cchtml
+++ b/HTMLtoMarkdown_Prewrite.cchtml
@@ -128,5 +128,4 @@
 
 <h2> Final PR Results </h2>
 <p> All of the standard HTML to Markdown conversions are working along with non-standard Markdown conversions tables, task lists, and strikethrough. The style attributes are not currently being saved so many styling features do not save. The tables have been added and fixed so that they can properly convert both ways, and a custom style class has been added so that a black border is added. On converting back to HTML, the width for the tables is lost and reset to auto. The emojis have been fixed so they can properly convert as well. </p>
-
-<p>We were unable to figure out how to add custom rules into pulldown-cmark's library. We attempted to create custom events, but could not get the custom events to properly work. The custom rules could easily be added to the HTML to Markdown side, but not to the Markdown to HTML side.</p>
\ No newline at end of file
+<p> Some non standard conversions have been implemented. These conversions are subscript & superscript. More conversions can easily be added with a specific structure. Currently, only one character symbols can be added for new Markdown conversions. This is due to how the code iterates through the doc blocks when converting the Markdown back to HTML. 
diff --git a/client/src/CodeChatEditor.mts b/client/src/CodeChatEditor.mts
index 14193f4..24cbc1b 100644
--- a/client/src/CodeChatEditor.mts
+++ b/client/src/CodeChatEditor.mts
@@ -49,6 +49,8 @@ const turndownService = new TurndownService();
 
 // Add the plugins from [turndown-plugin-gfm](https://github.com/laurent22/joplin/tree/dev/packages/turndown-plugin-gfm) to enable conversions for tables, task lists, and strikethroughs.
 turndownService.use(gfm);
+
+// Add each new rule for HTML to Markdown conversions into turndown. These rules are stored and added in from the [markdown\_conversions.json](markdown_conversions.json) file.
 for (let i = 0; i < MarkdownConversions.conversions.length; i++) {
     turndownService.addRule(MarkdownConversions.conversions[i].name, {
         filter: [MarkdownConversions.conversions[i].filter_tag],
@@ -62,9 +64,9 @@ for (let i = 0; i < MarkdownConversions.conversions.length; i++) {
   
 // Load code when the DOM is ready.
 export const page_init = (all_source: any) => {
-    // Use [URLSearchParams](https://developer.mozilla.org/en-US/docs/Web/API/URLSearchParams) to parse out the search parameters of this window's URL.
+    // Use [URLSearchParams](https://developer.mozilla.org/en-US/docs/Web/API/URLSearchParams) to parse out the search parameters of this window’s URL.
     const urlParams = new URLSearchParams(window.location.search);
-    // Get the mode from the page's query parameters. Default to edit using the [nullish coalescing operator](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Nullish_coalescing_operator). This works, but TypeScript marks it as an error. Ignore this error by including the [@ts-ignore directive](https://www.typescriptlang.org/docs/handbook/intro-to-js-ts.html#ts-check).
+    // Get the mode from the page’s query parameters. Default to edit using the [nullish coalescing operator](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Nullish_coalescing_operator). This works, but TypeScript marks it as an error. Ignore this error by including the [@ts-ignore directive](https://www.typescriptlang.org/docs/handbook/intro-to-js-ts.html#ts-check).
     /// @ts-ignore
     const editorMode = EditorMode[urlParams.get("mode") ?? "edit"];
     on_dom_content_loaded(() => open_lp(all_source, editorMode));
@@ -73,7 +75,7 @@ export const page_init = (all_source: any) => {
 // This is copied from [MDN](https://developer.mozilla.org/en-US/docs/Web/API/Document/DOMContentLoaded_event#checking_whether_loading_is_already_complete).
 const on_dom_content_loaded = (on_load_func: () => void) => {
     if (document.readyState === "loading") {
-        // Loading hasn't finished yet.
+        // Loading hasn’t finished yet.
         document.addEventListener("DOMContentLoaded", on_load_func);
     } else {
         // `DOMContentLoaded` has already fired.
@@ -81,7 +83,7 @@ const on_dom_content_loaded = (on_load_func: () => void) => {
     }
 };
 
-// Define all possible editor modes; these are passed as a [query string](https://en.wikipedia.org/wiki/Query_string) (`http://path/to/foo.py?mode=toc`, for example) to the page's URL.
+// Define all possible editor modes; these are passed as a [query string](https://en.wikipedia.org/wiki/Query_string) (`http://path/to/foo.py?mode=toc`, for example) to the page’s URL.
 enum EditorMode {
     // Display the source code using CodeChat, but disallow editing.
     view,
@@ -93,7 +95,7 @@ enum EditorMode {
     raw,
 }
 
-// This function is called on page load to "load" a file. Before this point, the server has already lexed the source file into code and doc blocks; this function transforms the code and doc blocks into HTML and updates the current web page with the results.
+// This function is called on page load to “load” a file. Before this point, the server has already lexed the source file into code and doc blocks; this function transforms the code and doc blocks into HTML and updates the current web page with the results.
 const open_lp = (
     // A data structure provided by the server, containing the source and associated metadata. See [`AllSource`](#AllSource).
     all_source: AllSource,
@@ -105,7 +107,7 @@ const open_lp = (
     const code_doc_block_arr = all_source["code_doc_block_arr"];
     let html;
     if (is_doc_only()) {
-        // Special case: a CodeChat Editor document's HTML doesn't need lexing; it only contains HTML. Instead, its structure is always: `[["", "", HTML]]`. Therefore, the HTML is at item \[0\]\[2\].
+        // Special case: a CodeChat Editor document’s HTML doesn’t need lexing; it only contains HTML. Instead, its structure is always: `[["", "", HTML]]`. Therefore, the HTML is at item \[0\]\[2\].
         html = `<div class="CodeChat-TinyMCE">${code_doc_block_arr[0][2]}</div>`;
     } else {
         html = classified_source_to_html(code_doc_block_arr);
@@ -133,7 +135,7 @@ type AllSource = {
 
 // Store the lexer info for the currently-loaded language.
 //
-// This mirrors the data provided by the server -- see [SourceFileMetadata](../../server/src/webserver.rs#SourceFileMetadata).
+// This mirrors the data provided by the server – see [SourceFileMetadata](../../server/src/webserver.rs#SourceFileMetadata).
 let current_metadata: {
     mode: string;
 };
@@ -155,7 +157,7 @@ const make_editors = async (
 ) => {
     return new Promise((accept) => {
         setTimeout(async () => {
-            // In view mode, don't use TinyMCE, since we already have HTML. Raw mode doesn't use TinyMCE at all, or even render doc blocks as HTML.
+            // In view mode, don’t use TinyMCE, since we already have HTML. Raw mode doesn’t use TinyMCE at all, or even render doc blocks as HTML.
             if (editorMode === EditorMode.edit) {
                 // Instantiate the TinyMCE editor for doc blocks. Wait until this finishes before calling anything else, to help keep the UI responsive. TODO: break this up to apply to each doc block, instead of doing them all at once.
                 await make_doc_block_editor(".CodeChat-TinyMCE");
@@ -200,15 +202,15 @@ const make_doc_block_editor = (
     selector: string
 ) => {
     return tinymce_init({
-        // Enable the [browser-supplied spellchecker](https://www.tiny.cloud/docs/tinymce/6/spelling/#browser_spellcheck), since TinyMCE's spellchecker is a premium feature.
+        // Enable the [browser-supplied spellchecker](https://www.tiny.cloud/docs/tinymce/6/spelling/#browser_spellcheck), since TinyMCE’s spellchecker is a premium feature.
         browser_spellcheck: true,
-        // Put more buttons on the [quick toolbar](https://www.tiny.cloud/docs/tinymce/6/quickbars/) that appears when text is selected. TODO: add a button for code format (can't find this one -- it's only on the [list of menu items](https://www.tiny.cloud/docs/tinymce/6/available-menu-items/#the-core-menu-items) as `codeformat`).
+        // Put more buttons on the [quick toolbar](https://www.tiny.cloud/docs/tinymce/6/quickbars/) that appears when text is selected. TODO: add a button for code format (can’t find this one – it’s only on the [list of menu items](https://www.tiny.cloud/docs/tinymce/6/available-menu-items/#the-core-menu-items) as `codeformat`).
         quickbars_selection_toolbar:
             "align | bold italic underline | quicklink h2 h3 blockquote",
         // Place the Tiny MCE menu bar at the top of the screen; otherwise, it floats in front of text, sometimes obscuring what the user wants to edit. See the [docs](https://www.tiny.cloud/docs/configure/editor-appearance/#fixed_toolbar_container).
         fixed_toolbar_container: "#CodeChat-menu",
         inline: true,
-        // When true, this still prevents hyperlinks to anchors on the current page from working correctly. There's an onClick handler that prevents links in the current page from working -- need to look into this. See also [a related GitHub issue](https://github.com/tinymce/tinymce/issues/3836).
+        // When true, this still prevents hyperlinks to anchors on the current page from working correctly. There’s an onClick handler that prevents links in the current page from working – need to look into this. See also [a related GitHub issue](https://github.com/tinymce/tinymce/issues/3836).
         //readonly: true  // Per the comment above, this is commented out.
         // TODO: Notes on this setting.
         relative_urls: true,
@@ -247,7 +249,7 @@ const make_code_block_editor = (
         highlightGutterLine: false,
         maxLines: 1e10,
         mode: `ace/mode/${current_metadata["mode"]}`,
-        // TODO: this still allows cursor movement. Need something that doesn't show an edit cursor / can't be selected; arrow keys should scroll the display, not move the cursor around in the editor.
+        // TODO: this still allows cursor movement. Need something that doesn’t show an edit cursor / can’t be selected; arrow keys should scroll the display, not move the cursor around in the editor.
         readOnly:
             editorMode === EditorMode.view || editorMode == EditorMode.toc,
         showPrintMargin: false,
@@ -263,7 +265,7 @@ const make_code_block_editor = (
 const doc_block_indent_on_before_input = (event: InputEvent) => {
     // Only modify the behavior of inserts.
     if (event.data) {
-        // Block any insert that's not an insert of spaces. TODO: need to support tabs.
+        // Block any insert that’s not an insert of spaces. TODO: need to support tabs.
         if (event.data !== " ".repeat(event.data.length)) {
             event.preventDefault();
         }
@@ -358,11 +360,11 @@ const classified_source_to_html = (
             // prettier-ignore
             html.push(
                 '<div class="CodeChat-doc">',
-                    // TODO: Add spaces matching the number of digits in the ACE gutter's line number. Currently, this is three spaces, assuming a file length of 100-999 lines.
+                    // TODO: Add spaces matching the number of digits in the ACE gutter’s line number. Currently, this is three spaces, assuming a file length of 100-999 lines.
                     '<div class="CodeChat-ACE-gutter-padding ace_editor">   </div>',
                     // This is a thin margin which matches what ACE does.
                     '<div class="CodeChat-ACE-padding"></div>',
-                    // This doc block's indent. TODO: allow paste, but must only allow pasting whitespace.
+                    // This doc block’s indent. TODO: allow paste, but must only allow pasting whitespace.
                     `<div class="ace_editor CodeChat-doc-indent" contenteditable onpaste="return false">${indent}</div>`,
                     // The contents of this doc block.
                     `<div class="CodeChat-TinyMCE" data-CodeChat-comment="${delimiter}" id="mce-${line}">`,
@@ -400,13 +402,13 @@ const editor_to_code_doc_blocks = () => {
         if (code_or_doc_tag.classList.contains("CodeChat-ACE")) {
             // See if the Ace editor was applied to this element.
             full_string =
-                // TypeScript knows that an element doesn't have a `env` attribute; ignore this, since Ace elements do.
+                // TypeScript knows that an element doesn’t have a `env` attribute; ignore this, since Ace elements do.
                 /// @ts-ignore
                 code_or_doc_tag.env === undefined
                     ? unescapeHTML(code_or_doc_tag.innerHTML)
                     : ace.edit(code_or_doc_tag).getValue();
         } else if (code_or_doc_tag.classList.contains("CodeChat-TinyMCE")) {
-            // Get the indent from the previous table cell. For a CodeChat Editor document, there's no indent (it's just a doc block). Likewise, get the delimiter; leaving it blank for a CodeChat Editor document causes the next block of code to leave off the comment delimiter, which is what we want.
+            // Get the indent from the previous table cell. For a CodeChat Editor document, there’s no indent (it’s just a doc block). Likewise, get the delimiter; leaving it blank for a CodeChat Editor document causes the next block of code to leave off the comment delimiter, which is what we want.
             if (!is_doc_only()) {
                 indent =
                     code_or_doc_tag.previousElementSibling!.textContent ?? "";
@@ -415,7 +417,7 @@ const editor_to_code_doc_blocks = () => {
                     code_or_doc_tag.getAttribute("data-CodeChat-comment") ??
                     null;
             }
-            // See [`get`](https://www.tiny.cloud/docs/tinymce/6/apis/tinymce.root/#get) and [`getContent()`](https://www.tiny.cloud/docs/tinymce/6/apis/tinymce.editor/#getContent). If this element wasn't managed by TinyMCE, it returns `null`, in which case we can directly get the `innerHTML`.
+            // See [`get`](https://www.tiny.cloud/docs/tinymce/6/apis/tinymce.root/#get) and [`getContent()`](https://www.tiny.cloud/docs/tinymce/6/apis/tinymce.editor/#getContent). If this element wasn’t managed by TinyMCE, it returns `null`, in which case we can directly get the `innerHTML`.
             //
             // Ignore the missing `get` type definition.
             /// @ts-ignore
@@ -435,7 +437,7 @@ const editor_to_code_doc_blocks = () => {
             throw `Unexpected class for code or doc block ${code_or_doc_tag}.`;
         }
 
-        // There's an implicit newline at the end of each block; restore it.
+        // There’s an implicit newline at the end of each block; restore it.
         full_string += "\n";
 
         // Merge this with previous classified line if indent and delimiter are the same; otherwise, add a new entry.
@@ -456,7 +458,7 @@ const editor_to_code_doc_blocks = () => {
 // Helper functions
 // ----------------
 //
-// Given text, escape it so it formats correctly as HTML. Because the solution at [SO](https://stackoverflow.com/a/48054293) transforms newlines in odd ways (see [innerText](https://developer.mozilla.org/en-US/docs/Web/API/HTMLElement/innerText)), it's not usable with code. Instead, this is a translation of Python's `html.escape` function.
+// Given text, escape it so it formats correctly as HTML. Because the solution at [SO](https://stackoverflow.com/a/48054293) transforms newlines in odd ways (see [innerText](https://developer.mozilla.org/en-US/docs/Web/API/HTMLElement/innerText)), it’s not usable with code. Instead, this is a translation of Python’s `html.escape` function.
 const escapeHTML = (unsafeText: string): string => {
     // Must be done first!
     unsafeText = unsafeText.replaceAll("&", "&amp;");
@@ -478,13 +480,13 @@ const is_doc_only = () => {
     return current_metadata["mode"] === "codechat-html";
 };
 
-// Per [MDN](https://developer.mozilla.org/en-US/docs/Web/API/Navigator/platform#examples), here's the least bad way to choose between the control key and the command key.
+// Per [MDN](https://developer.mozilla.org/en-US/docs/Web/API/Navigator/platform#examples), here’s the least bad way to choose between the control key and the command key.
 const os_is_osx =
     navigator.platform.indexOf("Mac") === 0 || navigator.platform === "iPhone"
         ? true
         : false;
 
-// A great and simple idea taken from [SO](https://stackoverflow.com/a/54116079): wrap all testing exports in a single variable. This avoids namespace pollution, since only one name is exported, and it's clearly marked for testing only. Test code still gets access to everything it needs.
+// A great and simple idea taken from [SO](https://stackoverflow.com/a/54116079): wrap all testing exports in a single variable. This avoids namespace pollution, since only one name is exported, and it’s clearly marked for testing only. Test code still gets access to everything it needs.
 export const exportedForTesting = {
     editor_to_code_doc_blocks,
     EditorMode,
diff --git a/server/src/webserver.rs b/server/src/webserver.rs
index e628bf8..d457191 100644
--- a/server/src/webserver.rs
+++ b/server/src/webserver.rs
@@ -1,25 +1,22 @@
-/// <details>
-///     <summary>Copyright (C) 2022 Bryan A. Jones.</summary>
-///     <p>This file is part of the CodeChat Editor.</p>
-///     <p>The CodeChat Editor is free software: you can redistribute it and/or
-///         modify it under the terms of the GNU General Public License as
-///         published by the Free Software Foundation, either version 3 of the
-///         License, or (at your option) any later version.</p>
-///     <p>The CodeChat Editor is distributed in the hope that it will be
-///         useful, but WITHOUT ANY WARRANTY; without even the implied warranty
-///         of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
-///         General Public License for more details.</p>
-///     <p>You should have received a copy of the GNU General Public License
-///         along with the CodeChat Editor. If not, see <a
-///             href="http://www.gnu.org/licenses/">http://www.gnu.org/licenses/</a>.
-///     </p>
-/// </details>
-/// <h1><code>webserver.rs</code> &mdash; Serve CodeChat Editor Client webpages
-/// </h1>
-/// <p>TODO: auto-reload when the current file changes on disk. Use <a
-///         href="https://docs.rs/notify/latest/notify/">notify</a>.</p>
-/// <h2>Imports</h2>
-/// <h3>Standard library</h3>
+/// Copyright (C) 2022 Bryan A. Jones.
+///
+/// This file is part of the CodeChat Editor.
+///
+/// The CodeChat Editor is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
+///
+/// The CodeChat Editor is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+///
+/// You should have received a copy of the GNU General Public License along with the CodeChat Editor. If not, see [http://www.gnu.org/licenses/](http://www.gnu.org/licenses/).
+///
+/// `webserver.rs` — Serve CodeChat Editor Client webpages
+/// ======================================================
+///
+/// TODO: auto-reload when the current file changes on disk. Use [notify](https://docs.rs/notify/latest/notify/).
+///
+/// Imports
+/// -------
+///
+/// ### Standard library
 use std::{
     collections::HashMap,
     env,
@@ -27,7 +24,7 @@ use std::{
     path::{Path, PathBuf},
 };
 
-// <h3>Third-party</h3>
+// ### Third-party
 use actix_files;
 use actix_web::{
     get,
@@ -48,69 +45,63 @@ use urlencoding::{self, encode};
 #[cfg(target_os = "windows")]
 use win_partitions::win_api::get_logical_drive;
 
-// <h3>Local</h3>
+// ### Local
 use super::lexer::compile_lexers;
 use super::lexer::supported_languages::LANGUAGE_LEXER_ARR;
 use crate::lexer::{source_lexer, CodeDocBlock, DocBlock, LanguageLexersCompiled};
 
-/// <h2>Data structures</h2>
+/// Data structures
+/// ---------------
 #[derive(Serialize, Deserialize)]
-/// <p><a id="SourceFileMetadata"></a>Metadata about a source file sent along
-///     with it both to and from the client.</p>
+/// Metadata about a source file sent along with it both to and from the client.
 struct SourceFileMetadata {
     mode: String,
 }
 
 #[derive(Serialize, Deserialize)]
-/// <p><a id="ClientSourceFile"></a>A simple structure for accepting JSON input
-///     to the <code>save_source</code> endpoint. Use a tuple since serdes can
-///     auto-generate a deserializer for it.</p>
+/// A simple structure for accepting JSON input to the `save_source` endpoint. Use a tuple since serdes can auto-generate a deserializer for it.
 struct ClientSourceFile {
     metadata: SourceFileMetadata,
-    // <p>TODO: implement a serdes deserializer that would convert this directly
-    //     to a CodeDocBlock?</p>
+    // TODO: implement a serdes deserializer that would convert this directly to a CodeDocBlock?
     code_doc_block_arr: Vec<(String, Option<String>, String)>,
 }
 
 #[derive(Serialize)]
-/// <p><a id="LexedSourceFile"></a>Define the structure of JSON responses when
-///     sending a source file from the <code>/fs</code> endpoint.</p>
+/// Define the structure of JSON responses when sending a source file from the `/fs` endpoint.
 struct LexedSourceFile {
     metadata: SourceFileMetadata,
     code_doc_block_arr: Vec<CodeDocBlock>,
 }
 
-/// <p>This defines the structure of JSON responses from the
-///     <code>save_source</code> endpoint.</p>
+/// This defines the structure of JSON responses from the `save_source` endpoint.
 #[derive(Serialize)]
 struct ErrorResponse {
     success: bool,
     message: String,
 }
 
-// <h2>Globals</h2>
+// Globals
+// -------
 lazy_static! {
-    /// <p>Matches a bare drive letter.</p>
+    /// Matches a bare drive letter.
     static ref DRIVE_LETTER_REGEX: Regex = Regex::new("^[a-zA-Z]:$").unwrap();
-    /// <p>Match the lexer directive in a source file.</p>
+    /// Match the lexer directive in a source file.
     static ref LEXER_DIRECTIVE: Regex = Regex::new(r#"CodeChat Editor lexer: (\w+)"#).unwrap();
 }
 
-/// <h2>Save endpoint</h2>
+/// Save endpoint
+/// -------------
 #[put("/fs/{path:.*}")]
-/// <p>The Save button in the CodeChat Editor Client posts to this endpoint with
-///     the path of the file to save.</p>
+/// The Save button in the CodeChat Editor Client posts to this endpoint with the path of the file to save.
 async fn save_source(
-    // <p>The path to save this file to.</p>
+    // The path to save this file to.
     encoded_path: web::Path<String>,
-    // <p>The file to save plus metadata, stored in the
-    //     <code>ClientSourceFile</code></p>
+    // The file to save plus metadata, stored in the `ClientSourceFile`
     client_source_file: web::Json<ClientSourceFile>,
-    // <p>Lexer info, needed to transform the <code>ClientSourceFile</code> into
-    //     source code.</p>
+    // Lexer info, needed to transform the `ClientSourceFile` into source code.
     language_lexers_compiled: web::Data<LanguageLexersCompiled<'_>>,
 ) -> impl Responder {
-    // <p>Given the mode, find the lexer.</p>
+    // Given the mode, find the lexer.
     let lexer = match language_lexers_compiled
         .map_mode_to_lexer
         .get(client_source_file.metadata.mode.as_str())
@@ -119,8 +110,7 @@ async fn save_source(
         None => return save_source_response(false, "Invalid mode"),
     };
 
-    // <p>Turn this back into code and doc blocks by filling in any missing
-    //     comment delimiters.</p>
+    // Turn this back into code and doc blocks by filling in any missing comment delimiters.
     let inline_comment = lexer.language_lexer.inline_comment_delim_arr.first();
     let block_comment = lexer.language_lexer.block_comment_delim_arr.first();
     let mut code_doc_block_vec: Vec<CodeDocBlock> = Vec::new();
@@ -132,12 +122,11 @@ async fn save_source(
         } else {
             CodeDocBlock::DocBlock(DocBlock {
                 indent: cdb.0.to_string(),
-                // <p>If no delimiter is provided, use an inline comment (if
-                //     available), then a block comment.</p>
+                // If no delimiter is provided, use an inline comment (if available), then a block comment.
                 delimiter: match &cdb.1 {
-                    // <p>The delimiter was provided. Simply use that.</p>
+                    // The delimiter was provided. Simply use that.
                     Some(v) => v.to_string(),
-                    // <p>No delimiter was provided -- fill one in.</p>
+                    // No delimiter was provided – fill one in.
                     None => {
                         if let Some(ic) = inline_comment {
                             ic.to_string()
@@ -156,24 +145,20 @@ async fn save_source(
         });
     }
 
-    // <p>Turn this vec of code/doc blocks into a string of source code.</p>
+    // Turn this vec of code/doc blocks into a string of source code.
     let mut file_contents = String::new();
     for code_doc_block in code_doc_block_vec {
         match code_doc_block {
             CodeDocBlock::DocBlock(doc_block) => {
-                // <p>Append a doc block, adding a space between the opening
-                //     delimiter and the contents when necessary.</p>
+                // Append a doc block, adding a space between the opening delimiter and the contents when necessary.
                 let mut append_doc_block = |indent: &str, delimiter: &str, contents: &str| {
                     file_contents += indent;
                     file_contents += delimiter;
-                    // <p>Add a space between the delimiter and comment body,
-                    //     unless the comment was a newline or we're at the end
-                    //     of the file.</p>
+                    // Add a space between the delimiter and comment body, unless the comment was a newline or we’re at the end of the file.
                     if contents.is_empty() || contents == "\n" {
-                        // <p>Nothing to append in this case.</p>
+                        // Nothing to append in this case.
                     } else {
-                        // <p>Put a space between the delimiter and the
-                        //     contents.</p>
+                        // Put a space between the delimiter and the contents.
                         file_contents += " ";
                     }
                     file_contents += contents;
@@ -184,16 +169,14 @@ async fn save_source(
                     .inline_comment_delim_arr
                     .contains(&doc_block.delimiter.as_str());
 
-                // <p>Build a comment based on the type of the delimiter.</p>
+                // Build a comment based on the type of the delimiter.
                 if is_inline_delim {
-                    // <p>Split the contents into a series of lines, adding the
-                    //     inline comment delimiter to each line.</p>
+                    // Split the contents into a series of lines, adding the inline comment delimiter to each line.
                     for content_line in doc_block.contents.split_inclusive('\n') {
                         append_doc_block(&doc_block.indent, &doc_block.delimiter, content_line);
                     }
                 } else {
-                    // <p>Determine the closing comment delimiter matching the
-                    //     provided opening delimiter.</p>
+                    // Determine the closing comment delimiter matching the provided opening delimiter.
                     let block_comment_closing_delimiter = match lexer
                         .language_lexer
                         .block_comment_delim_arr
@@ -211,32 +194,26 @@ async fn save_source(
                             )
                         }
                     };
-                    // <p>Produce the resulting block comment. They should
-                    //     always end with a newline.</p>
+                    // Produce the resulting block comment. They should always end with a newline.
                     assert!(&doc_block.contents.ends_with('\n'));
                     append_doc_block(
                         &doc_block.indent,
                         &doc_block.delimiter,
-                        // <p>Omit the newline, so we can instead put on the
-                        //     closing delimiter, then the newline.</p>
+                        // Omit the newline, so we can instead put on the closing delimiter, then the newline.
                         &doc_block.contents[..&doc_block.contents.len() - 1],
                     );
                     file_contents = file_contents + " " + block_comment_closing_delimiter + "\n";
                 }
             }
             CodeDocBlock::CodeBlock(contents) =>
-            // <p>This is code. Simply append it (by definition, indent and
-            //     delimiter are empty).</p>
+            // This is code. Simply append it (by definition, indent and delimiter are empty).
             {
                 file_contents += &contents
             }
         }
     }
 
-    // <p>Save this string to a file. Add a leading slash for Linux: this
-    //     changes from <code>foo/bar.c</code> to <code>/foo/bar.c</code>.
-    //     Windows already starts with a drive letter, such as
-    //     <code>C:\foo\bar.c</code>, so no changes are needed.</p>
+    // Save this string to a file. Add a leading slash for Linux: this changes from `foo/bar.c` to `/foo/bar.c`. Windows already starts with a drive letter, such as `C:\foo\bar.c`, so no changes are needed.
     let save_file_path = if cfg!(windows) {
         "".to_string()
     } else {
@@ -255,9 +232,7 @@ async fn save_source(
     save_source_response(true, "")
 }
 
-/// <p>A convenience method to fill out then return the
-///     <code>ErrorResponse</code> struct from the <code>save_source</code>
-///     endpoint.</p>
+/// A convenience method to fill out then return the `ErrorResponse` struct from the `save_source` endpoint.
 fn save_source_response(success: bool, message: &str) -> HttpResponse {
     let response = ErrorResponse {
         success,
@@ -275,17 +250,17 @@ fn save_source_response(success: bool, message: &str) -> HttpResponse {
     }
 }
 
-/// <h2>Load endpoints</h2>
-/// <p>Redirect from the root of the filesystem to the actual root path on this
-///     OS.</p>
+/// Load endpoints
+/// --------------
+///
+/// Redirect from the root of the filesystem to the actual root path on this OS.
 async fn _root_fs_redirect() -> impl Responder {
     HttpResponse::TemporaryRedirect()
         .insert_header((header::LOCATION, "/fs/"))
         .finish()
 }
 
-/// <p>The load endpoint: dispatch to support functions which serve either a
-///     directory listing, a CodeChat Editor file, or a normal file.</p>
+/// The load endpoint: dispatch to support functions which serve either a directory listing, a CodeChat Editor file, or a normal file.
 #[get("/fs/{path:.*}")]
 async fn serve_fs(
     req: HttpRequest,
@@ -294,29 +269,20 @@ async fn serve_fs(
 ) -> impl Responder {
     let mut fixed_path = orig_path.to_string();
     #[cfg(target_os = "windows")]
-    // <p>On Windows, a path of <code>drive_letter:</code> needs a
-    //     <code>/</code> appended.</p>
+    // On Windows, a path of `drive_letter:` needs a `/` appended.
     if DRIVE_LETTER_REGEX.is_match(&fixed_path) {
         fixed_path += "/";
     } else if fixed_path.is_empty() {
-        // <p>If there's no drive letter yet, we will always use
-        //     <code>dir_listing</code> to select a drive.</p>
+        // If there’s no drive letter yet, we will always use `dir_listing` to select a drive.
         return dir_listing("", Path::new("")).await;
     }
-    // <p>All other cases (for example, <code>C:\a\path\to\file.txt</code>) are
-    //     OK.</p>
+    // All other cases (for example, `C:\a\path\to\file.txt`) are OK.
 
-    // <p>For Linux/OS X, prepend a slash, so that
-    //     <code>a/path/to/file.txt</code> becomes
-    //     <code>/a/path/to/file.txt</code>.</p>
+    // For Linux/OS X, prepend a slash, so that `a/path/to/file.txt` becomes `/a/path/to/file.txt`.
     #[cfg(not(target_os = "windows"))]
     let fixed_path = "/".to_string() + &fixed_path;
 
-    // <p>On Windows, the returned path starts with <code>\\?\</code> per the <a
-    //         href="https://learn.microsoft.com/en-us/windows/win32/fileio/naming-a-file#win32-file-namespaces">docs</a>.
-    //     Handle any <a
-    //         href="https://doc.rust-lang.org/std/fs/fn.canonicalize.html#errors">errors</a>.
-    // </p>
+    // On Windows, the returned path starts with `\\?\` per the [docs](https://learn.microsoft.com/en-us/windows/win32/fileio/naming-a-file#win32-file-namespaces). Handle any [errors](https://doc.rust-lang.org/std/fs/fn.canonicalize.html#errors).
     let canon_path = match Path::new(&fixed_path).canonicalize() {
         Ok(p) => p,
         Err(err) => {
@@ -332,22 +298,21 @@ async fn serve_fs(
         return serve_file(&canon_path, &req, language_lexers_compiled).await;
     }
 
-    // <p>It's not a directory or a file...we give up. For simplicity, don't
-    //     handle symbolic links.</p>
+    // It’s not a directory or a file…we give up. For simplicity, don’t handle symbolic links.
     html_not_found(&format!(
         "<p>The requested path <code>{}</code> is not a directory or a file.</p>",
         path_display(&canon_path)
     ))
 }
 
-/// <h3>Directory browser</h3>
-/// <p>Create a web page listing all files and subdirectories of the provided
-///     directory.</p>
+/// ### Directory browser
+///
+/// Create a web page listing all files and subdirectories of the provided directory.
 async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
-    // <p>Special case on Windows: list drive letters.</p>
+    // Special case on Windows: list drive letters.
     #[cfg(target_os = "windows")]
     if dir_path == Path::new("") {
-        // <p>List drive letters in Windows</p>
+        // List drive letters in Windows
         let mut drive_html = String::new();
         let logical_drives = match get_logical_drive() {
             Ok(v) => v,
@@ -370,7 +335,7 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
         )));
     }
 
-    // <p>List each file/directory with appropriate links.</p>
+    // List each file/directory with appropriate links.
     let mut unwrapped_read_dir = match fs::read_dir(dir_path).await {
         Ok(p) => p,
         Err(err) => {
@@ -382,7 +347,7 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
         }
     };
 
-    // <p>Get a listing of all files and directories</p>
+    // Get a listing of all files and directories
     let mut files: Vec<DirEntry> = Vec::new();
     let mut dirs: Vec<DirEntry> = Vec::new();
     loop {
@@ -402,7 +367,7 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
                     if file_type.is_file() {
                         files.push(dir_entry);
                     } else {
-                        // <p>Group symlinks with dirs.</p>
+                        // Group symlinks with dirs.
                         dirs.push(dir_entry);
                     }
                 } else {
@@ -414,7 +379,7 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
             }
         };
     }
-    // <p>Sort them -- case-insensitive on Windows, normally on Linux/OS X.</p>
+    // Sort them – case-insensitive on Windows, normally on Linux/OS X.
     #[cfg(target_os = "windows")]
     let file_name_key = |a: &DirEntry| {
         Ok::<String, std::ffi::OsString>(a.file_name().into_string()?.to_lowercase())
@@ -425,7 +390,7 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
     files.sort_unstable_by_key(file_name_key);
     dirs.sort_unstable_by_key(file_name_key);
 
-    // <p>Put this on the resulting webpage. List directories first.</p>
+    // Put this on the resulting webpage. List directories first.
     let mut dir_html = String::new();
     for dir in dirs {
         let dir_name = match dir.file_name().into_string() {
@@ -441,14 +406,9 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
         dir_html += &format!(
             "<li><a href='/fs/{}{}{}'>{}</a></li>\n",
             web_path,
-            // <p>If this is a raw drive letter, then the path already ends with
-            //     a slash, such as <code>C:/</code>. Don't add a second slash
-            //     in this case. Otherwise, add a slash to make
-            //     <code>C:/foo</code> into <code>C:/foo/</code>.</p>
-            // <p>Likewise, the Linux root path of <code>/</code> already ends
-            //     with a slash, while all other paths such a <code>/foo</code>
-            //     don't. To detect this, look for an empty
-            //     <code>web_path</code>.</p>
+            // If this is a raw drive letter, then the path already ends with a slash, such as `C:/`. Don’t add a second slash in this case. Otherwise, add a slash to make `C:/foo` into `C:/foo/`.
+            //
+            // Likewise, the Linux root path of `/` already ends with a slash, while all other paths such a `/foo` don’t. To detect this, look for an empty `web_path`.
             if web_path.ends_with('/') || web_path.is_empty() {
                 ""
             } else {
@@ -459,7 +419,7 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
         );
     }
 
-    // <p>List files second.</p>
+    // List files second.
     let mut file_html = String::new();
     for file in files {
         let file_name = match file.file_name().into_string() {
@@ -496,15 +456,14 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
     HttpResponse::Ok().body(body)
 }
 
-// <h3>Serve a CodeChat Editor Client webpage</h3>
+// ### Serve a CodeChat Editor Client webpage
 async fn serve_file(
     file_path: &Path,
     req: &HttpRequest,
     language_lexers_compiled: web::Data<LanguageLexersCompiled<'_>>,
 ) -> HttpResponse {
     let raw_dir = file_path.parent().unwrap();
-    // <p>Use a lossy conversion, since this is UI display, not filesystem
-    //     access.</p>
+    // Use a lossy conversion, since this is UI display, not filesystem access.
     let dir = escape_html(path_display(raw_dir).as_str());
     let name = escape_html(&file_path.file_name().unwrap().to_string_lossy());
     let ext = &file_path
@@ -512,7 +471,7 @@ async fn serve_file(
         .unwrap_or_else(|| OsStr::new(""))
         .to_string_lossy();
 
-    // <p>Get the <code>mode</code> and <code>test</code> query parameters.</p>
+    // Get the `mode` and `test` query parameters.
     let empty_string = "".to_string();
     let query_params = web::Query::<HashMap<String, String>>::from_query(req.query_string());
     let (mode, is_test_mode) = match query_params {
@@ -523,7 +482,7 @@ async fn serve_file(
         Err(_err) => (empty_string, false),
     };
 
-    // <p>Read the file.</p>
+    // Read the file.
     let mut file_contents = String::new();
     let read_ret = match File::open(file_path).await {
         Ok(fc) => fc,
@@ -540,15 +499,14 @@ async fn serve_file(
     
     // Categorize the file:
     //
-    // - A binary file (meaning we can't read the contents as UTF-8): just serve it raw. Assume this is an image/video/etc.
-    // - A text file - first determine the type. Based on the type:
-    //   - If it's an unknown type (such as a source file we don't know or a plain text file): just serve it raw.
-    //   - If the client requested a table of contents, then serve it wrapped in a CodeChat TOC.
-    //   - If it's Markdown or CCHTML, serve it wrapped in a CodeChat Document Editor.
-    //   - Otherwise, it must be a recognized file type. Serve it wrapped in a CodeChat Editor.
+    // *   A binary file (meaning we can’t read the contents as UTF-8): just serve it raw. Assume this is an image/video/etc.
+    // *   A text file - first determine the type. Based on the type:
+    //     *   If it’s an unknown type (such as a source file we don’t know or a plain text file): just serve it raw.
+    //     *   If the client requested a table of contents, then serve it wrapped in a CodeChat TOC.
+    //     *   If it’s Markdown or CCHTML, serve it wrapped in a CodeChat Document Editor.
+    //     *   Otherwise, it must be a recognized file type. Serve it wrapped in a CodeChat Editor.
     if let Err(_err) = read_ret {
-        // <p>TODO: make a better decision, don't duplicate code. The file type
-        //     is unknown. Serve it raw, assuming it's an image/video/etc.</p>
+        // TODO: make a better decision, don’t duplicate code. The file type is unknown. Serve it raw, assuming it’s an image/video/etc.
         match actix_files::NamedFile::open_async(file_path).await {
             Ok(v) => {
                 let res = v
@@ -569,9 +527,7 @@ async fn serve_file(
         }
     }
 
-    // <p>The TOC is a simplified web page requiring no additional processing.
-    //     The script ensures that all hyperlinks target the enclosing page, not
-    //     just the iframe containing this page.</p>
+    // The TOC is a simplified web page requiring no additional processing. The script ensures that all hyperlinks target the enclosing page, not just the iframe containing this page.
     if mode == "toc" {
         return HttpResponse::Ok().body(format!(
             r#"<!DOCTYPE html>
@@ -600,9 +556,9 @@ async fn serve_file(
         ));
     }
 
-    // <p>Determine the lexer to use for this file.</p>
+    // Determine the lexer to use for this file.
     let ace_mode;
-    // <p>First, search for a lexer directive in the file contents.</p>
+    // First, search for a lexer directive in the file contents.
     let lexer = if let Some(captures) = LEXER_DIRECTIVE.captures(&file_contents) {
         ace_mode = captures[1].to_string();
         match language_lexers_compiled
@@ -613,15 +569,14 @@ async fn serve_file(
             None => return html_not_found(&format!("<p>Unknown lexer type {}.</p>", &ace_mode)),
         }
     } else {
-        // <p>Otherwise, look up the lexer by the file's extension.</p>
+        // Otherwise, look up the lexer by the file’s extension.
         if let Some(llc) = language_lexers_compiled
             .map_ext_to_lexer_vec
             .get(ext.as_ref())
         {
             llc.first().unwrap()
         } else {
-            // <p>The file type is unknown. Serve it raw, assuming it's an
-            //     image/video/etc.</p>
+            // The file type is unknown. Serve it raw, assuming it’s an image/video/etc.
             match actix_files::NamedFile::open_async(file_path).await {
                 Ok(v) => {
                     let res = v.into_response(req);
@@ -638,14 +593,14 @@ async fn serve_file(
         }
     };
 
-    // <p>Lex the code and put it in a JSON structure.</p>
+    // Lex the code and put it in a JSON structure.
     let mut code_doc_block_arr = if lexer.language_lexer.ace_mode == "codechat-html" {
         vec![CodeDocBlock::CodeBlock(file_contents)]
     } else {
         source_lexer(&file_contents, lexer)
     };
 
-    // <p>Convert doc blocks from Markdown to HTML</p>
+    // Read non-standard markdown conversions from [markdown\_conversions.json](../../client/src/markdown_conversions.json) file
     #[derive(Debug, Deserialize)]
     struct Conversion {
         conversions: Vec<Entry>
@@ -659,17 +614,20 @@ async fn serve_file(
     let mut file = File::open("../client/src/markdown_conversions.json").await.unwrap();
     let mut buff = String::new();
     file.read_to_string(&mut buff).await.unwrap();
- 
     let conversion: Conversion = serde_json::from_str(&buff).unwrap();
 
     for code_doc_block in &mut code_doc_block_arr {
         if let CodeDocBlock::DocBlock(ref mut doc_block) = code_doc_block {
+            
+            // For each all document block contents, iterate through each character to look for new Markdown conversions. If symbol is found, insert corresponding HTML tag.
             let mut output_contents = Vec::new();
             let mut tag_open = false;
-            for c in doc_block.contents.chars() { 
+            let doc_block_contents = doc_block.contents.chars().collect::<Vec<char>>();
+            for i in 0..doc_block_contents.len() { 
                 let mut replaced = false;
                 for entry in &conversion.conversions {
-                    if c == entry.markdown {
+                    // Check if character matches a markdown conversion. Also check to ensure that subscript isn't converting '~~' so that strikethrough still works.
+                    if doc_block_contents[i] == entry.markdown && doc_block_contents[i+1] != entry.markdown && doc_block_contents[i-1] != entry.markdown {
                         if tag_open {
                             for tag_char in entry.close_tag.chars() {
                                 output_contents.push(tag_char);
@@ -686,12 +644,13 @@ async fn serve_file(
                 }
                 if !replaced {
                     {
-                        output_contents.push(c);
+                        output_contents.push(doc_block_contents[i]);
                     }
                 }
             }
             let output: String = output_contents.into_iter().map(|i| i.to_string()).collect::<String>();
-            
+           
+            // Convert doc blocks from Markdown to HTML
             let options = Options::all();
             let parser = Parser::new_ext(&output, options);
             let mut html_output = String::new();
@@ -716,14 +675,12 @@ async fn serve_file(
             ))
         }
     };
-    // <p>Look for any script tags and prevent these from causing problems.</p>
+    // Look for any script tags and prevent these from causing problems.
     let lexed_source_file_string = lexed_source_file_string.replace("</script>", "<\\/script>");
 
-    // <p>Look for a project file by searching the current directory, then all
-    //     its parents, for a file named <code>toc.cchtml</code>.</p>
+    // Look for a project file by searching the current directory, then all its parents, for a file named `toc.cchtml`.
     let mut is_project = false;
-    // <p>The number of directories between this file to serve (in
-    //     <code>path</code>) and the toc file.</p>
+    // The number of directories between this file to serve (in `path`) and the toc file.
     let mut path_to_toc = PathBuf::new();
     let mut current_dir = file_path.to_path_buf();
     loop {
@@ -741,7 +698,7 @@ async fn serve_file(
         path_to_toc.push("../");
     }
 
-    // <p>For project files, add in the sidebar.</p>
+    // For project files, add in the sidebar.
     let (sidebar_iframe, sidebar_css) = if is_project {
         (
             format!(
@@ -754,7 +711,7 @@ async fn serve_file(
         ("".to_string(), "")
     };
 
-    // <p>Add in content when testing.</p>
+    // Add in content when testing.
     let testing_src = if is_test_mode {
         r#"
         <link rel="stylesheet" href="https://unpkg.com/mocha/mocha.css" />
@@ -764,7 +721,7 @@ async fn serve_file(
         ""
     };
 
-    // <p>Build and return the webpage.</p>
+    // Build and return the webpage.
     HttpResponse::Ok().body(format!(
         r##"<!DOCTYPE html>
 <html lang="en">
@@ -812,8 +769,10 @@ async fn serve_file(
     ))
 }
 
-// <h2>Utilities</h2>
-// <p>Given a <code>Path</code>, transform it into a displayable string.</p>
+// Utilities
+// ---------
+//
+// Given a `Path`, transform it into a displayable string.
 fn path_display(p: &Path) -> String {
     let path_orig = p.to_string_lossy();
     if cfg!(windows) {
@@ -823,12 +782,12 @@ fn path_display(p: &Path) -> String {
     }
 }
 
-// <p>Return a Not Found (404) errors with the provided HTML body.</p>
+// Return a Not Found (404) errors with the provided HTML body.
 fn html_not_found(msg: &str) -> HttpResponse {
     HttpResponse::NotFound().body(html_wrapper(msg))
 }
 
-// <p>Wrap the provided HTML body in DOCTYPE/html/head tags.</p>
+// Wrap the provided HTML body in DOCTYPE/html/head tags.
 fn html_wrapper(body: &str) -> String {
     format!(
         "<!DOCTYPE html>
@@ -846,8 +805,7 @@ fn html_wrapper(body: &str) -> String {
     )
 }
 
-// <p>Given text, escape it so it formats correctly as HTML. This is a
-//     translation of Python's <code>html.escape</code> function.</p>
+// Given text, escape it so it formats correctly as HTML. This is a translation of Python’s `html.escape` function.
 fn escape_html(unsafe_text: &str) -> String {
     unsafe_text
         .replace('&', "&amp;")
@@ -855,31 +813,30 @@ fn escape_html(unsafe_text: &str) -> String {
         .replace('>', "&gt;")
 }
 
-// <h2>Webserver startup</h2>
+// Webserver startup
+// -----------------
 #[actix_web::main]
 pub async fn main() -> std::io::Result<()> {
     HttpServer::new(|| {
-        // <p>Get the path to this executable. Assume that static files for the
-        //     webserver are located relative to it.</p>
+        // Get the path to this executable. Assume that static files for the webserver are located relative to it.
         let exe_path = env::current_exe().unwrap();
         let exe_dir = exe_path.parent().unwrap();
         let mut client_static_path = PathBuf::from(exe_dir);
         client_static_path.push("../../../client/static");
         client_static_path = client_static_path.canonicalize().unwrap();
 
-        // <p>Start the server.</p>
+        // Start the server.
         App::new()
             .app_data(web::Data::new(compile_lexers(LANGUAGE_LEXER_ARR)))
-            // <p>Serve static files per the <a
-            //         href="https://actix.rs/docs/static-files">docs</a>.</p>
+            // Serve static files per the [docs](https://actix.rs/docs/static-files).
             .service(actix_files::Files::new(
                 "/static",
                 client_static_path.as_os_str(),
             ))
-            // <p>This endpoint serves the filesystem.</p>
+            // This endpoint serves the filesystem.
             .service(serve_fs)
             .service(save_source)
-            // <p>Reroute to the filesystem for typical user-requested URLs.</p>
+            // Reroute to the filesystem for typical user-requested URLs.
             .route("/", web::get().to(_root_fs_redirect))
             .route("/fs", web::get().to(_root_fs_redirect))
     })

From 06e78c728b3b0cbeab4203e7ee907525d4d26e78 Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Sat, 29 Apr 2023 03:07:14 -0500
Subject: [PATCH 06/13] Add link to explain use for new table class

---
 client/static/css/CodeChatEditor.css | 157 +++++++++++----------------
 1 file changed, 65 insertions(+), 92 deletions(-)

diff --git a/client/static/css/CodeChatEditor.css b/client/static/css/CodeChatEditor.css
index b2c2d24..5d24a8f 100644
--- a/client/static/css/CodeChatEditor.css
+++ b/client/static/css/CodeChatEditor.css
@@ -1,36 +1,32 @@
-/* <details>
-    <summary>License</summary>
-    <p>Copyright (C) 2022 Bryan A. Jones.</p>
-    <p>This file is part of the CodeChat Editor.</p>
-    <p>The CodeChat Editor is free software: you can redistribute it and/or
-        modify it under the terms of the GNU General Public License as
-        published by the Free Software Foundation, either version 3 of the
-        License, or (at your option) any later version.</p>
-    <p>The CodeChat Editor is distributed in the hope that it will be useful,
-        but WITHOUT ANY WARRANTY; without even the implied warranty of
-        MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
-        General Public License for more details.</p>
-    <p>You should have received a copy of the GNU General Public License
-        along with the CodeChat Editor. If not, see <a
-            href="http://www.gnu.org/licenses/">http://www.gnu.org/licenses/</a>.
-    </p>
-</details>
-<h1><code>CodeChatEditor.css</code> &mdash; Styles for the CodeChat Editor
-</h1>
-<p>This stylesheet is used by the HTML generated by <a
-        href="..\..\src\CodeChatEditor.mts">CodeChatEditor.mts</a>.</p>
-<p>TODO: do a much better job of grouping common styles. Rename styles based
-    on whether they style a code or doc block.</p>
-<h2>Styles for the entire page layout</h2>
-<p>This is used only to store a reused variable value. See the <a
-        href="https://drafts.csswg.org/css-variables/">CSS docs</a>.</p> */
+/* License
+
+Copyright (C) 2022 Bryan A. Jones.
+
+This file is part of the CodeChat Editor.
+
+The CodeChat Editor is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
+
+The CodeChat Editor is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with the CodeChat Editor. If not, see [http://www.gnu.org/licenses/](http://www.gnu.org/licenses/).
+
+`CodeChatEditor.css` — Styles for the CodeChat Editor
+=====================================================
+
+This stylesheet is used by the HTML generated by [CodeChatEditor.mts](....%5Csrc%5CCodeChatEditor.mts).
+
+TODO: do a much better job of grouping common styles. Rename styles based on whether they style a code or doc block.
+
+Styles for the entire page layout
+---------------------------------
+
+This is used only to store a reused variable value. See the [CSS docs](https://drafts.csswg.org/css-variables/). */
 :root {
     --top-height: 6.7rem;
     --body-padding: 0.2rem;
 }
 
-/* <p>See <a href="https://css-tricks.com/box-sizing/">box sizing</a> for the
-    following technique to use <code>border-box</code> sizing.</p> */
+/* See [box sizing](https://css-tricks.com/box-sizing/) for the following technique to use `border-box` sizing. */
 html {
     box-sizing: border-box;
 }
@@ -42,40 +38,42 @@ html {
 }
 
 body {
-    /* <p>For box model simplicity, switch the padding and margin.</p> */
+    /* For box model simplicity, switch the padding and margin. */
     padding: var(--body-padding);
     margin: 0px;
-    /* <p>The default of transparent makes VS Code's black show up and confuse
-    everything. For now, make it white.</p> */
+    /* The default of transparent makes VS Code’s black show up and confuse everything. For now, make it white. */
     background: white;
 }
 
-/* <p>Provide space at the top of the screen for the filename and TinyMCE menu
-    bar.</p> */
+/* Provide space at the top of the screen for the filename and TinyMCE menu bar. */
 #CodeChat-top {
     height: var(--top-height);
 }
 
-/* <p>The rest of the screen is the editor area.</p> */
+/* The rest of the screen is the editor area. */
 #CodeChat-body {
     height: calc(100vh - var(--top-height) - 2 * var(--body-padding));
     overflow: auto;
 }
 
-/* <h2>Misc styling</h2>
-<p>Mark a hotkey.</p> */
+/* Misc styling
+------------
+
+Mark a hotkey. */
 .CodeChat-hotkey {
     text-decoration: underline;
 }
 
-/* <p>Make the filename compact.</p> */
+/* Make the filename compact. */
 #CodeChat-filename p {
     margin: 0px;
     white-space: nowrap;
 }
 
-/* <h2>Code block styling</h2>
-<p>Before the ACE Editor loads, use a monotype font.</p> */
+/* Code block styling
+------------------
+
+Before the ACE Editor loads, use a monotype font. */
 .CodeChat-ACE {
     font-family: monospace;
     white-space: pre;
@@ -83,97 +81,72 @@ body {
 }
 
 .CodeChat-code {
-    /* <p>Use <a
-        href="https://developer.mozilla.org/en-US/docs/Web/CSS/CSS_Flexible_Box_Layout/Basic_Concepts_of_Flexbox">flexbox
-        layout</a> to style code blocks. The goal of this layout is:</p>
-<p>&lt;max line numbers spaces&gt; &lt;code&gt;</p>
-<p>where the max line numbers spaces cause this code block's line numbers
-    to line up with all other code block line numbering. For example, if
-    this code block is lines 9-23 (which the Ace editor will treat as a
-    two-digit line number), but the largest line is 453 (three digits),
-    then we need to insert one extra space to correct for the extra digit
-    when the line number is &gt; 99. A <code>&lt;div
-        class="CodeChat-ACE-gutter"&gt;</code> implements this &lt;max
-    line numbers spaces&gt; indent.</p> */
+    /* Use [flexbox layout](https://developer.mozilla.org/en-US/docs/Web/CSS/CSS_Flexible_Box_Layout/Basic_Concepts_of_Flexbox) to style code blocks. The goal of this layout is:
+
+where the max line numbers spaces cause this code block’s line numbers to line up with all other code block line numbering. For example, if this code block is lines 9-23 (which the Ace editor will treat as a two-digit line number), but the largest line is 453 (three digits), then we need to insert one extra space to correct for the extra digit when the line number is > 99. A `<div class="CodeChat-ACE-gutter">` implements this indent. */
     display: flex;
 }
 
-/* <p>Style the max line numbers spaces element discussed above.</p> */
+/* `Style the max line numbers spaces element discussed above.` */
 .CodeChat-ACE-gutter {
     font-family: monospace;
     white-space: pre;
     background-color: #f0f0f0;
-    /* <p>Make the &lt;max line number spaces&gt; not expand or shrink, but take
-    exactly the width required by the text (spaces) it contains. Defaults
-    for the &lt;code&gt; element cause it to take the rest of the space
-    on the line.</p> */
+    /* `` Make the not expand or shrink, but take exactly the width required by the text (spaces) it contains. Defaults for the `element cause it to take the rest of the space on the line.` `` */
     flex: 0 0 auto;
 }
 
-/* <p>Hide the ACE Editor's cursor when it's inactive.</p> */
+/* `` `Hide the ACE Editor’s cursor when it’s inactive.` `` */
 .ace_hidden-cursors {
     opacity: 0;
 }
 
-/* <h2>Doc block styling</h2> */
+/* `` `Doc block styling` ``
+------------------------- */
 .CodeChat-doc {
-    /* <p>Use flexbox layout to style doc blocks. The approach is similar to
-    code blocks; the overall goal is:</p>
-<p>&lt;max line number spaces&gt; &lt;padding to match the ACE editor&gt;
-    &lt;doc block indent&gt; &lt;doc block contents&gt;</p>
-<p>where:</p>
-<ul>
-    <li><code>&lt;div class="CodeChat-ACE-gutter-padding"&gt;</code>
-        contains &lt;max line number spaces&gt;</li>
-    <li><code>&lt;div class="CodeChat-ACE-padding"&gt;</code> contains
-        &lt;padding to match the ACE editor&gt;</li>
-    <li><code>&lt;div class="CodeChat-doc-indent"&gt;</code> contains
-        &lt;doc block indent&gt;</li>
-    <li><code>&lt;div class="CodeChat-TinyMCE"&gt;</code> contains the
-        &lt;doc block contents&gt;</li>
-</ul> */
+    /* `` `Use flexbox layout to style doc blocks. The approach is similar to code blocks; the overall goal is:` ``
+
+where:
+
+*   `<div class="CodeChat-ACE-gutter-padding">` contains
+*   `<div class="CodeChat-ACE-padding">` contains
+*   `<div class="CodeChat-doc-indent">` contains
+*   `<div class="CodeChat-TinyMCE">` contains the */
     display: flex;
 }
 
-/* <p>Match the ACE Editor gutter for doc blocks.</p> */
+/* `` `Match the ACE Editor gutter for doc blocks.` `` */
 .CodeChat-ACE-gutter-padding {
     padding-left: 33px;
     background-color: #f0f0f0;
-    /* <p>Make this &lt;max line number spaces&gt; not expand or shrink, but
-    take exactly the width required by the text (spaces) it contains.</p> */
+    /* `` `Make this not expand or shrink, but take exactly the width required by the text (spaces) it contains.` `` */
     flex: 0 0 auto;
     font-family: monospace;
     white-space: pre;
 }
 
-/* <p>The ACE editor puts a bit of left padding in; match this so that code and
-    doc blocks line up.</p> */
+/* `` `The ACE editor puts a bit of left padding in; match this so that code and doc blocks line up.` `` */
 .CodeChat-ACE-padding {
     padding-left: 3px;
 }
 
-/* <p>Preserve whitespace in the indent of a doc block.</p> */
+/* `` `Preserve whitespace in the indent of a doc block.` `` */
 .CodeChat-doc-indent {
-    /* <p>Make this &lt;doc block indent&gt; not expand or shrink, but take
-    exactly the width required by the text (spaces) it contains.</p> */
+    /* `` `Make this not expand or shrink, but take exactly the width required by the text (spaces) it contains.` `` */
     flex: 0 0 auto;
     white-space: pre;
     tab-size: 4;
 }
 
-/* <p>Remove the editor's border when it's selected, since this hides the cursor
-    when the cursot is at the beginning of a line and isn't necessary (the
-    entire screen is an editor, not just that region.)</p> */
+/* `` `Remove the editor’s border when it’s selected, since this hides the cursor when the cursot is at the beginning of a line and isn’t necessary (the entire screen is an editor, not just that region.)` `` */
 .CodeChat-TinyMCE.mce-edit-focus {
     outline-width: 0px;
 }
 
-/* <h2>Combined code/doc block syling</h2>
-<p><span id="remove-space">Remove space between a code block followed by a
-        doc block. Doc block elements typically have top margin and/or
-        padding that produce this undesired space; remove it on the first
-        element in the doc block, the first element of the first element in
-        the doc block, etc.</span></p> */
+/* `` `Combined code/doc block syling` ``
+--------------------------------------
+
+Remove space between a code block followed by a doc block. Doc block elements typically have top margin and/or padding that produce this undesired space; remove it on the first element in the doc block, the first element of the first element in the doc block, etc. */
 .CodeChat-TinyMCE > *:first-child,
 .CodeChat-TinyMCE > *:first-child > *:first-child,
 .CodeChat-TinyMCE > *:first-child > *:first-child > *:first-child,
@@ -192,8 +165,7 @@ body {
     padding-top: 0px;
 }
 
-/* <p><a href="remove-space">Remove space</a> between a doc block followed by a
-    code block.</p> */
+/* [Remove space](remove-space) between a doc block followed by a code block. */
 .CodeChat-TinyMCE > *:last-child,
 .CodeChat-TinyMCE > *:last-child > *:last-child,
 .CodeChat-TinyMCE > *:last-child > *:last-child > *:last-child,
@@ -208,6 +180,7 @@ body {
     padding-bottom: 0px;
 }
 
+/* Add a border around tables for converting back to HTML from Markdown. This class is used in [tinymce-webpack.mts](../../src/tinymce-webpack.mts) to add to all tables on initialization */
 .CodeChat-TinyMCE-TableDefault { 
     border: 1px solid black;
     border-collapse: collapse;

From db604d450d652edceb823c210dc2065fc131f1f5 Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Sat, 29 Apr 2023 09:55:37 -0500
Subject: [PATCH 07/13] Add check for closing tag for correct conversions

---
 HTMLtoMarkdown_Prewrite.cchtml |  2 +-
 Markdown_HTML.js               |  4 +-
 Markdown_MD.js                 |  4 +-
 server/src/webserver.rs        | 73 ++++++++++++++++++++++++----------
 4 files changed, 58 insertions(+), 25 deletions(-)

diff --git a/HTMLtoMarkdown_Prewrite.cchtml b/HTMLtoMarkdown_Prewrite.cchtml
index 90a8943..5888c47 100644
--- a/HTMLtoMarkdown_Prewrite.cchtml
+++ b/HTMLtoMarkdown_Prewrite.cchtml
@@ -128,4 +128,4 @@
 
 <h2> Final PR Results </h2>
 <p> All of the standard HTML to Markdown conversions are working along with non-standard Markdown conversions tables, task lists, and strikethrough. The style attributes are not currently being saved so many styling features do not save. The tables have been added and fixed so that they can properly convert both ways, and a custom style class has been added so that a black border is added. On converting back to HTML, the width for the tables is lost and reset to auto. The emojis have been fixed so they can properly convert as well. </p>
-<p> Some non standard conversions have been implemented. These conversions are subscript & superscript. More conversions can easily be added with a specific structure. Currently, only one character symbols can be added for new Markdown conversions. This is due to how the code iterates through the doc blocks when converting the Markdown back to HTML. 
+<p> Some non standard conversions have been implemented. These conversions are subscript & superscript. More conversions can easily be added with a specific structure. Currently, only one character symbols can be added for new Markdown conversions. This is due to how the code iterates through the doc blocks when converting the Markdown back to HTML. </p>
diff --git a/Markdown_HTML.js b/Markdown_HTML.js
index 8175233..16fb2b3 100644
--- a/Markdown_HTML.js
+++ b/Markdown_HTML.js
@@ -57,10 +57,10 @@
 //         nice colors</span></p>
 // <p style="line-height: 1;">&nbsp;</p>
 // <p style="line-height: 1;"><span
-//         style="background-color: rgb(132, 63, 161); color: rgb(241, 196, 15);">Text
+//         style="background-color: rgb(132, 63, 161); color: rgb(241, 196, 15);">~~Text
 //         in less-nice colors</span></p>
 // <p style="line-height: 1;">&nbsp;</p>
-// <p>A nice little table</p>
+// <p>A nice little table~~</p>
 // <table style="border-collapse: collapse; width: 100.024%;" border="1">
 //     <colgroup>
 //         <col style="width: 20.0122%;">
diff --git a/Markdown_MD.js b/Markdown_MD.js
index 58fa353..9b26a5c 100644
--- a/Markdown_MD.js
+++ b/Markdown_MD.js
@@ -72,9 +72,9 @@
 //
 // Text in nice colors
 //
-// Text in less-nice colors
+// ~~Text in less-nice colors
 //
-// A nice little table
+// A nice little table~~
 //
 // |     |     |     |     |     |
 // | --- | --- | --- | --- | --- |    
diff --git a/server/src/webserver.rs b/server/src/webserver.rs
index d457191..6ba1e78 100644
--- a/server/src/webserver.rs
+++ b/server/src/webserver.rs
@@ -622,31 +622,64 @@ async fn serve_file(
             // For each all document block contents, iterate through each character to look for new Markdown conversions. If symbol is found, insert corresponding HTML tag.
             let mut output_contents = Vec::new();
             let mut tag_open = false;
-            let doc_block_contents = doc_block.contents.chars().collect::<Vec<char>>();
-            for i in 0..doc_block_contents.len() { 
-                let mut replaced = false;
-                for entry in &conversion.conversions {
-                    // Check if character matches a markdown conversion. Also check to ensure that subscript isn't converting '~~' so that strikethrough still works.
-                    if doc_block_contents[i] == entry.markdown && doc_block_contents[i+1] != entry.markdown && doc_block_contents[i-1] != entry.markdown {
-                        if tag_open {
-                            for tag_char in entry.close_tag.chars() {
-                                output_contents.push(tag_char);
+            let doc_block_lines = doc_block.contents.lines();
+            
+            for line in doc_block_lines {
+                let doc_block_contents = line.chars().collect::<Vec<char>>();
+                for i in 0..doc_block_contents.len() {
+                    let mut replaced = false;
+                    for entry in &conversion.conversions {
+                        // Check if character matches a markdown conversion.
+                        if doc_block_contents[i] == entry.markdown {
+                            let mut convert_to_html = false;
+
+                            if tag_open {
+                                for tag_char in entry.close_tag.chars() {
+                                    output_contents.push(tag_char);
+                                }
+                                tag_open = false;
+                                replaced = true;
+                            } 
+                            else {
+                                // Check to see if characters need converted to HTML. Only convert when the text has the correct format & not just anytime the symbol appears. 
+                                for j in i..doc_block_contents.len() {
+                                    if doc_block_contents[j] == entry.markdown {
+                                        convert_to_html = true;
+                                    }
+                                }
+                                // This is in place to ensure that subscript isn't converting '~~' so that strikethrough still works.
+                                if i == 0 {
+                                    if doc_block_contents[i+1] == entry.markdown {
+                                            convert_to_html = false;
+                                    }  
+                                }
+                                else if i == doc_block_contents.len() - 1 {
+                                    if doc_block_contents[i-1] == entry.markdown {
+                                        convert_to_html = false
+                                    }
+                                }
+                                else {
+                                    if doc_block_contents[i+1] == entry.markdown || doc_block_contents[i-1] == entry.markdown {
+                                        convert_to_html = false;
+                                    }
+                                }
+                                if convert_to_html {
+                                    for tag_char in entry.open_tag.chars() {
+                                        output_contents.push(tag_char);
+                                    }
+                                    tag_open = true;
+                                    replaced = true;
+                                }
                             }
                         } 
-                        else {
-                            for tag_char in entry.open_tag.chars() {
-                                output_contents.push(tag_char);
-                            }
+                    }
+                    if !replaced {
+                        {
+                            output_contents.push(doc_block_contents[i]);
                         }
-                        tag_open = !tag_open;
-                        replaced = true;
-                    } 
-                }
-                if !replaced {
-                    {
-                        output_contents.push(doc_block_contents[i]);
                     }
                 }
+                output_contents.push('\n');
             }
             let output: String = output_contents.into_iter().map(|i| i.to_string()).collect::<String>();
            

From d1ab3e1871bfd2f0e06a3def04c61554021d12d3 Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Sat, 29 Apr 2023 10:24:57 -0500
Subject: [PATCH 08/13] Clean up console logs & unneeded file

---
 KEEP_Markdown.js              | 137 ----------------------------------
 client/src/CodeChatEditor.mts |   2 -
 2 files changed, 139 deletions(-)
 delete mode 100644 KEEP_Markdown.js

diff --git a/KEEP_Markdown.js b/KEEP_Markdown.js
deleted file mode 100644
index 6a6513b..0000000
--- a/KEEP_Markdown.js
+++ /dev/null
@@ -1,137 +0,0 @@
-// In this file, I will test as many markdown things as I can, to ensure that the conversion works correctly:  
-//   
-// **Bold Text**
-//
-// _Italic Text_
-//
-// Left-Justified Text
-//
-// Center-Justified Text
-//
-// Right-Justified Text
-//
-// Fully-Justified Text. Obviously this one needs to be a bit longer. Lorem Ipsum dolor sit amet. The Quick Brown Fox Jumps Over The Lazy Dog. tHE qUICK bROWN fOX jUMPS oVER tHE lAZY dOG. foo bar baz qux quem etc.
-//
-// Big h1 Text
-// ===========
-//
-// h2 Text 
-// --------
-//
-// ### h3 Text
-//
-// #### h4 Text
-//
-// ##### h5 Text
-//
-// ###### h6 Text
-//
-// Underlined Text
-//
-// Strikethrough Text
-//
-// Text in superscript and subscript
-//
-// `for good measure, here's a code-block`
-//
-// > Blockquote Text
-//
-// Div Text
-//
-// Pre Text  
-//   
-// Formatting Time:
-//
-// Different Fonts
-//
-// Fonts
-//
-// Fonts
-//
-// tiny font
-//
-// LARGE FONT
-//
-// large line height
-//
-// like this all should be
-//
-// double spaced
-//
-// Text in nice colors
-//
-// Text in less-nice colors
-//
-// A nice little table
-//
-//     
-//
-// a
-//
-// b
-//
-// c
-//
-// d
-//
-// e
-//
-// b
-//
-// c
-//
-// d
-//
-// e
-//
-// f
-//
-// c
-//
-// d
-//
-// e
-//
-// f
-//
-// g
-//
-// d
-//
-// e
-//
-// f
-//
-// g
-//
-// h
-//
-// e
-//
-// f
-//
-// g
-//
-// h
-//
-// i
-//
-// [Here's a nice link](https://github.com)
-//
-// ∫ some unicode symbols δ ¾ 
-//
-// A line
-//
-// * * *
-//
-// Embedded Video
-//
-// Some Emoji (more unicode)
-//
-// ![💯](https://twemoji.maxcdn.com/v/13.0.1/72x72/1f4af.png) ![🅱️](https://twemoji.maxcdn.com/v/13.0.1/72x72/1f171.png)
-//
-// And some images:
-//
-// ![](CodeChat1.png)
-//
-// ![](CodeChat2.png)![](CodeChat3.png)
diff --git a/client/src/CodeChatEditor.mts b/client/src/CodeChatEditor.mts
index 24cbc1b..c0e75de 100644
--- a/client/src/CodeChatEditor.mts
+++ b/client/src/CodeChatEditor.mts
@@ -287,7 +287,6 @@ export const on_keydown = (event: KeyboardEvent) => {
 // Save CodeChat Editor contents.
 export const on_save = async () => {
     // This is the data to write — the source code. First, transform the HTML back into code and doc blocks.
-    console.log(turndownService);
     const source_code = editor_to_code_doc_blocks();
     // Then, wrap these in a [struct the server expects](../server/src/webserver.rs#ClientSourceFile) and send it.
     await save({
@@ -432,7 +431,6 @@ const editor_to_code_doc_blocks = () => {
                     80 - indent.length - (delimiter?.length ?? 1) - 1,
             });
             full_string = turndownService.turndown(full_string);
-            console.log(full_string);
         } else {
             throw `Unexpected class for code or doc block ${code_or_doc_tag}.`;
         }

From 502968302f68577ef011feaae10f3aa8bed4381c Mon Sep 17 00:00:00 2001
From: Clark Hensley <clarkhensley@protonmail.com>
Date: Wed, 3 May 2023 00:38:10 -0500
Subject: [PATCH 09/13] A mostly working fix on code tags in Markdown
 Conversion

---
 client/src/CodeChatEditor.mts | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/client/src/CodeChatEditor.mts b/client/src/CodeChatEditor.mts
index c0e75de..676c5b8 100644
--- a/client/src/CodeChatEditor.mts
+++ b/client/src/CodeChatEditor.mts
@@ -46,9 +46,11 @@ import "./../static/css/CodeChatEditor.css";
 //
 // Instantiate [turndown](https://github.com/mixmark-io/turndown) for HTML to Markdown conversion
 const turndownService = new TurndownService();
+const baseTurndownService = new TurndownService();
 
 // Add the plugins from [turndown-plugin-gfm](https://github.com/laurent22/joplin/tree/dev/packages/turndown-plugin-gfm) to enable conversions for tables, task lists, and strikethroughs.
 turndownService.use(gfm);
+baseTurndownService.use(gfm);
 
 // Add each new rule for HTML to Markdown conversions into turndown. These rules are stored and added in from the [markdown\_conversions.json](markdown_conversions.json) file.
 for (let i = 0; i < MarkdownConversions.conversions.length; i++) {
@@ -60,7 +62,28 @@ for (let i = 0; i < MarkdownConversions.conversions.length; i++) {
       })
 }
 
-
+// Adapted from [this issue](https://github.com/mixmark-io/turndown/issues/152)
+interface HTMLEntityMap {
+    "<": string,
+    ">": string
+};
+turndownService.addRule("angle brackets",
+    {
+        filter: function (node: any, options: any){
+            const re = /(.*<.*>.*)+/g;
+            return(re.test(node.innerText));
+        },
+        replacement: function (content: string, node: any) {
+            let htmlEntityMap: HTMLEntityMap = {
+                "<": " &lt; ",
+                ">": " &gt; "
+            };
+            let formattedHTML: string = baseTurndownService.turndown(node.outerHTML);
+            let finalHTML: string = formattedHTML.replace(/[<>]/g, function(m) {return htmlEntityMap[m as keyof HTMLEntityMap]});
+            console.log(finalHTML);
+            return finalHTML;
+        }
+    });
   
 // Load code when the DOM is ready.
 export const page_init = (all_source: any) => {

From f091a041083544d50811a7d73be28857bd0dc3a0 Mon Sep 17 00:00:00 2001
From: Clark Hensley <clarkhensley@protonmail.com>
Date: Wed, 3 May 2023 00:47:54 -0500
Subject: [PATCH 10/13] Documentation for previous PR

---
 client/src/CodeChatEditor.mts | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/client/src/CodeChatEditor.mts b/client/src/CodeChatEditor.mts
index 676c5b8..17e70b5 100644
--- a/client/src/CodeChatEditor.mts
+++ b/client/src/CodeChatEditor.mts
@@ -46,10 +46,14 @@ import "./../static/css/CodeChatEditor.css";
 //
 // Instantiate [turndown](https://github.com/mixmark-io/turndown) for HTML to Markdown conversion
 const turndownService = new TurndownService();
+
+// Generate a "base" version of the turndown plugin that will be used within the main one. This way, we can use the functionality during the special rules we add.
 const baseTurndownService = new TurndownService();
 
 // Add the plugins from [turndown-plugin-gfm](https://github.com/laurent22/joplin/tree/dev/packages/turndown-plugin-gfm) to enable conversions for tables, task lists, and strikethroughs.
 turndownService.use(gfm);
+
+// Add the gfm capabilities to the base version too
 baseTurndownService.use(gfm);
 
 // Add each new rule for HTML to Markdown conversions into turndown. These rules are stored and added in from the [markdown\_conversions.json](markdown_conversions.json) file.
@@ -63,6 +67,7 @@ for (let i = 0; i < MarkdownConversions.conversions.length; i++) {
 }
 
 // Adapted from [this issue](https://github.com/mixmark-io/turndown/issues/152)
+// This rule is more complex than the above ones, so it is not in the .json file.
 interface HTMLEntityMap {
     "<": string,
     ">": string
@@ -78,9 +83,13 @@ turndownService.addRule("angle brackets",
                 "<": " &lt; ",
                 ">": " &gt; "
             };
+            // Convert the outer HTML of this node. This initial conversion handles things like
+            // converting "\<code\>" blocks to "\`backtick\`" blocks
             let formattedHTML: string = baseTurndownService.turndown(node.outerHTML);
+            // then, we take that turndown'd version, and replace the "\<" and "\>" characters
+            // with their escaped html entity counterparts. This keeps them from potentially
+            // acting as html tags when they are not.
             let finalHTML: string = formattedHTML.replace(/[<>]/g, function(m) {return htmlEntityMap[m as keyof HTMLEntityMap]});
-            console.log(finalHTML);
             return finalHTML;
         }
     });

From 2670b338147ade92b793a55abbd23108d4c6ac73 Mon Sep 17 00:00:00 2001
From: Clark Hensley <clarkhensley@protonmail.com>
Date: Wed, 3 May 2023 21:40:09 -0500
Subject: [PATCH 11/13] Fixing broken anchor tags, working to improve greater
 and less-than symbols

---
 client/src/CodeChatEditor.mts        |  13 +--
 client/static/css/CodeChatEditor.css |  55 ++++++-------
 potential_codeblock_fix.js           |  27 +++++++
 server/src/webserver.rs              | 116 +++++----------------------
 4 files changed, 76 insertions(+), 135 deletions(-)
 create mode 100644 potential_codeblock_fix.js

diff --git a/client/src/CodeChatEditor.mts b/client/src/CodeChatEditor.mts
index 17e70b5..600a6c0 100644
--- a/client/src/CodeChatEditor.mts
+++ b/client/src/CodeChatEditor.mts
@@ -80,20 +80,23 @@ turndownService.addRule("angle brackets",
         },
         replacement: function (content: string, node: any) {
             let htmlEntityMap: HTMLEntityMap = {
-                "<": " &lt; ",
-                ">": " &gt; "
+                "<": "&lt;",
+                ">": "&gt;"
             };
             // Convert the outer HTML of this node. This initial conversion handles things like
-            // converting "\<code\>" blocks to "\`backtick\`" blocks
+            // converting "&lt;code&gt;" blocks to "\`backtick\`" blocks
             let formattedHTML: string = baseTurndownService.turndown(node.outerHTML);
-            // then, we take that turndown'd version, and replace the "\<" and "\>" characters
+            // then, we take that turndown'd version, and replace the "&lt;" and "&gt;" characters
             // with their escaped html entity counterparts. This keeps them from potentially
             // acting as html tags when they are not.
             let finalHTML: string = formattedHTML.replace(/[<>]/g, function(m) {return htmlEntityMap[m as keyof HTMLEntityMap]});
             return finalHTML;
         }
     });
-  
+
+// Keep <a> anchor tags, because they get removed from Markdown otherwise
+turndownService.keep(["a"]);
+
 // Load code when the DOM is ready.
 export const page_init = (all_source: any) => {
     // Use [URLSearchParams](https://developer.mozilla.org/en-US/docs/Web/API/URLSearchParams) to parse out the search parameters of this window’s URL.
diff --git a/client/static/css/CodeChatEditor.css b/client/static/css/CodeChatEditor.css
index 5d24a8f..95f8184 100644
--- a/client/static/css/CodeChatEditor.css
+++ b/client/static/css/CodeChatEditor.css
@@ -13,7 +13,7 @@ You should have received a copy of the GNU General Public License along with the
 `CodeChatEditor.css` — Styles for the CodeChat Editor
 =====================================================
 
-This stylesheet is used by the HTML generated by [CodeChatEditor.mts](....%5Csrc%5CCodeChatEditor.mts).
+This stylesheet is used by the HTML generated by [CodeChatEditor.mts](..\..\src\CodeChatEditor.mts).
 
 TODO: do a much better job of grouping common styles. Rename styles based on whether they style a code or doc block.
 
@@ -41,7 +41,7 @@ body {
     /* For box model simplicity, switch the padding and margin. */
     padding: var(--body-padding);
     margin: 0px;
-    /* The default of transparent makes VS Code’s black show up and confuse everything. For now, make it white. */
+    /* The default of transparent makes VS Code's black show up and confuse everything. For now, make it white. */
     background: white;
 }
 
@@ -83,68 +83,70 @@ Before the ACE Editor loads, use a monotype font. */
 .CodeChat-code {
     /* Use [flexbox layout](https://developer.mozilla.org/en-US/docs/Web/CSS/CSS_Flexible_Box_Layout/Basic_Concepts_of_Flexbox) to style code blocks. The goal of this layout is:
 
-where the max line numbers spaces cause this code block’s line numbers to line up with all other code block line numbering. For example, if this code block is lines 9-23 (which the Ace editor will treat as a two-digit line number), but the largest line is 453 (three digits), then we need to insert one extra space to correct for the extra digit when the line number is > 99. A `<div class="CodeChat-ACE-gutter">` implements this indent. */
+&lt;max line numbers spaces&gt; &lt;code&gt;where the max line numbers spaces cause this code block's line numbers to line up with all other code block line numbering. For example, if this code block is lines 9-23 (which the Ace editor will treat as a two-digit line number), but the largest line is 453 (three digits), then we need to insert one extra space to correct for the extra digit when the line number is &gt; 99. A `&lt;div class="CodeChat-ACE-gutter"&gt;` implements this &lt;max line numbers spaces&gt; indent. */
     display: flex;
 }
 
-/* `Style the max line numbers spaces element discussed above.` */
+/* Style the max line numbers spaces element discussed above. */
 .CodeChat-ACE-gutter {
     font-family: monospace;
     white-space: pre;
     background-color: #f0f0f0;
-    /* `` Make the not expand or shrink, but take exactly the width required by the text (spaces) it contains. Defaults for the `element cause it to take the rest of the space on the line.` `` */
+    /* Make the &lt;max line number spaces&gt; not expand or shrink, but take exactly the width required by the text (spaces) it contains. Defaults for the &lt;code&gt; element cause it to take the rest of the space on the line. */
     flex: 0 0 auto;
 }
 
-/* `` `Hide the ACE Editor’s cursor when it’s inactive.` `` */
+/* Hide the ACE Editor's cursor when it's inactive. */
 .ace_hidden-cursors {
     opacity: 0;
 }
 
-/* `` `Doc block styling` ``
-------------------------- */
+/* Doc block styling
+----------------- */
 .CodeChat-doc {
-    /* `` `Use flexbox layout to style doc blocks. The approach is similar to code blocks; the overall goal is:` ``
+    /* Use flexbox layout to style doc blocks. The approach is similar to code blocks; the overall goal is:
+
+&lt;max line number spaces&gt; &lt;padding to match the ACE editor&gt; &lt;doc block indent&gt; &lt;doc block contents&gt;
 
 where:
 
-*   `<div class="CodeChat-ACE-gutter-padding">` contains
-*   `<div class="CodeChat-ACE-padding">` contains
-*   `<div class="CodeChat-doc-indent">` contains
-*   `<div class="CodeChat-TinyMCE">` contains the */
+*   `&lt;div class="CodeChat-ACE-gutter-padding"&gt;` contains &lt;max line number spaces&gt;
+*   `&lt;div class="CodeChat-ACE-padding"&gt;` contains &lt;padding to match the ACE editor&gt;
+*   `&lt;div class="CodeChat-doc-indent"&gt;` contains &lt;doc block indent&gt;
+*   `&lt;div class="CodeChat-TinyMCE"&gt;` contains the &lt;doc block contents&gt; */
     display: flex;
 }
 
-/* `` `Match the ACE Editor gutter for doc blocks.` `` */
+/* Match the ACE Editor gutter for doc blocks. */
 .CodeChat-ACE-gutter-padding {
     padding-left: 33px;
     background-color: #f0f0f0;
-    /* `` `Make this not expand or shrink, but take exactly the width required by the text (spaces) it contains.` `` */
+    /* Make this &lt;max line number spaces&gt; not expand or shrink, but take exactly the width required by the text (spaces) it contains. */
     flex: 0 0 auto;
     font-family: monospace;
     white-space: pre;
 }
 
-/* `` `The ACE editor puts a bit of left padding in; match this so that code and doc blocks line up.` `` */
+/* The ACE editor puts a bit of left padding in; match this so that code and doc blocks line up. */
 .CodeChat-ACE-padding {
     padding-left: 3px;
 }
 
-/* `` `Preserve whitespace in the indent of a doc block.` `` */
+/* Preserve whitespace in the indent of a doc block. */
 .CodeChat-doc-indent {
-    /* `` `Make this not expand or shrink, but take exactly the width required by the text (spaces) it contains.` `` */
+    /* Make this &lt;doc block indent&gt; not expand or shrink, but take exactly the width required by the text (spaces) it contains. */
     flex: 0 0 auto;
     white-space: pre;
     tab-size: 4;
 }
 
-/* `` `Remove the editor’s border when it’s selected, since this hides the cursor when the cursot is at the beginning of a line and isn’t necessary (the entire screen is an editor, not just that region.)` `` */
+/* Remove the editor's border when it's selected, since this hides the cursor when the cursot is at the beginning of a line and isn't necessary (the entire screen is an editor, not just that region.) */
 .CodeChat-TinyMCE.mce-edit-focus {
     outline-width: 0px;
 }
 
-/* `` `Combined code/doc block syling` ``
---------------------------------------
+/* Combined code/doc block syling
+------------------------------
 
 Remove space between a code block followed by a doc block. Doc block elements typically have top margin and/or padding that produce this undesired space; remove it on the first element in the doc block, the first element of the first element in the doc block, etc. */
 .CodeChat-TinyMCE > *:first-child,
@@ -179,14 +181,3 @@ Remove space between a code block followed by a doc block. Doc block elements ty
     margin-bottom: 0px;
     padding-bottom: 0px;
 }
-
-/* Add a border around tables for converting back to HTML from Markdown. This class is used in [tinymce-webpack.mts](../../src/tinymce-webpack.mts) to add to all tables on initialization */
-.CodeChat-TinyMCE-TableDefault { 
-    border: 1px solid black;
-    border-collapse: collapse;
-}
-
-.CodeChat-TinyMCE-TableDefault th, td{ 
-    border: 1px solid black;
-    border-collapse: collapse;
-}
\ No newline at end of file
diff --git a/potential_codeblock_fix.js b/potential_codeblock_fix.js
new file mode 100644
index 0000000..4a5ef7c
--- /dev/null
+++ b/potential_codeblock_fix.js
@@ -0,0 +1,27 @@
+// This is a code snippet of a turndown rule that
+// might help fix issues with "&lt;" and "&gt;" being rendered as
+// those escape sequences rather than the corresponding angle
+// bracket characters.
+// This does perform the correct operations to correctly convert code blocks.
+// Unfortunately, the parser starts at the most-nested elements, so
+// the parent elements of the <code> elements then undo this conversion.
+// One possible (BAD!) solution would be to create another instance of
+// the turndownService in such a way that this rule is its only rule, and
+// then apply that to the entire document at the end. Definitely a last resort.
+interface AngleBracketMap {
+    "&lt;": string,
+    "&gt;": string
+};
+turndownService.addRule("code blocks angle brackets",
+    {
+        filter: ["code"],
+        replacement: function (content: string, node: any) {
+            let angleBracketMap: AngleBracketMap = {
+                "&lt;": "<",
+                "&gt;": ">"
+            };
+            let formattedString: string =  "`" + content.replace(/(&lt;)|(&gt;)/g, function(m) {return angleBracketMap[m as keyof AngleBracketMap]}) + "`";
+            console.log(formattedString);
+            return formattedString;
+        }
+    });
diff --git a/server/src/webserver.rs b/server/src/webserver.rs
index 6ba1e78..e98ac2a 100644
--- a/server/src/webserver.rs
+++ b/server/src/webserver.rs
@@ -53,13 +53,13 @@ use crate::lexer::{source_lexer, CodeDocBlock, DocBlock, LanguageLexersCompiled}
 /// Data structures
 /// ---------------
 #[derive(Serialize, Deserialize)]
-/// Metadata about a source file sent along with it both to and from the client.
+/// <a id="SourceFileMetadata"></a>Metadata about a source file sent along with it both to and from the client.
 struct SourceFileMetadata {
     mode: String,
 }
 
 #[derive(Serialize, Deserialize)]
-/// A simple structure for accepting JSON input to the `save_source` endpoint. Use a tuple since serdes can auto-generate a deserializer for it.
+/// <a id="ClientSourceFile"></a>A simple structure for accepting JSON input to the `save_source` endpoint. Use a tuple since serdes can auto-generate a deserializer for it.
 struct ClientSourceFile {
     metadata: SourceFileMetadata,
     // TODO: implement a serdes deserializer that would convert this directly to a CodeDocBlock?
@@ -67,7 +67,7 @@ struct ClientSourceFile {
 }
 
 #[derive(Serialize)]
-/// Define the structure of JSON responses when sending a source file from the `/fs` endpoint.
+/// <a id="LexedSourceFile"></a>Define the structure of JSON responses when sending a source file from the `/fs` endpoint.
 struct LexedSourceFile {
     metadata: SourceFileMetadata,
     code_doc_block_arr: Vec<CodeDocBlock>,
@@ -126,7 +126,7 @@ async fn save_source(
                 delimiter: match &cdb.1 {
                     // The delimiter was provided. Simply use that.
                     Some(v) => v.to_string(),
-                    // No delimiter was provided – fill one in.
+                    // No delimiter was provided -- fill one in.
                     None => {
                         if let Some(ic) = inline_comment {
                             ic.to_string()
@@ -154,7 +154,7 @@ async fn save_source(
                 let mut append_doc_block = |indent: &str, delimiter: &str, contents: &str| {
                     file_contents += indent;
                     file_contents += delimiter;
-                    // Add a space between the delimiter and comment body, unless the comment was a newline or we’re at the end of the file.
+                    // Add a space between the delimiter and comment body, unless the comment was a newline or we're at the end of the file.
                     if contents.is_empty() || contents == "\n" {
                         // Nothing to append in this case.
                     } else {
@@ -273,7 +273,7 @@ async fn serve_fs(
     if DRIVE_LETTER_REGEX.is_match(&fixed_path) {
         fixed_path += "/";
     } else if fixed_path.is_empty() {
-        // If there’s no drive letter yet, we will always use `dir_listing` to select a drive.
+        // If there's no drive letter yet, we will always use `dir_listing` to select a drive.
         return dir_listing("", Path::new("")).await;
     }
     // All other cases (for example, `C:\a\path\to\file.txt`) are OK.
@@ -298,7 +298,7 @@ async fn serve_fs(
         return serve_file(&canon_path, &req, language_lexers_compiled).await;
     }
 
-    // It’s not a directory or a file…we give up. For simplicity, don’t handle symbolic links.
+    // It's not a directory or a file...we give up. For simplicity, don't handle symbolic links.
     html_not_found(&format!(
         "<p>The requested path <code>{}</code> is not a directory or a file.</p>",
         path_display(&canon_path)
@@ -379,7 +379,7 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
             }
         };
     }
-    // Sort them – case-insensitive on Windows, normally on Linux/OS X.
+    // Sort them -- case-insensitive on Windows, normally on Linux/OS X.
     #[cfg(target_os = "windows")]
     let file_name_key = |a: &DirEntry| {
         Ok::<String, std::ffi::OsString>(a.file_name().into_string()?.to_lowercase())
@@ -406,9 +406,9 @@ async fn dir_listing(web_path: &str, dir_path: &Path) -> HttpResponse {
         dir_html += &format!(
             "<li><a href='/fs/{}{}{}'>{}</a></li>\n",
             web_path,
-            // If this is a raw drive letter, then the path already ends with a slash, such as `C:/`. Don’t add a second slash in this case. Otherwise, add a slash to make `C:/foo` into `C:/foo/`.
+            // If this is a raw drive letter, then the path already ends with a slash, such as `C:/`. Don't add a second slash in this case. Otherwise, add a slash to make `C:/foo` into `C:/foo/`.
             //
-            // Likewise, the Linux root path of `/` already ends with a slash, while all other paths such a `/foo` don’t. To detect this, look for an empty `web_path`.
+            // Likewise, the Linux root path of `/` already ends with a slash, while all other paths such a `/foo` don't. To detect this, look for an empty `web_path`.
             if web_path.ends_with('/') || web_path.is_empty() {
                 ""
             } else {
@@ -506,7 +506,7 @@ async fn serve_file(
     //     *   If it’s Markdown or CCHTML, serve it wrapped in a CodeChat Document Editor.
     //     *   Otherwise, it must be a recognized file type. Serve it wrapped in a CodeChat Editor.
     if let Err(_err) = read_ret {
-        // TODO: make a better decision, don’t duplicate code. The file type is unknown. Serve it raw, assuming it’s an image/video/etc.
+        // TODO: make a better decision, don't duplicate code. The file type is unknown. Serve it raw, assuming it's an image/video/etc.
         match actix_files::NamedFile::open_async(file_path).await {
             Ok(v) => {
                 let res = v
@@ -569,14 +569,14 @@ async fn serve_file(
             None => return html_not_found(&format!("<p>Unknown lexer type {}.</p>", &ace_mode)),
         }
     } else {
-        // Otherwise, look up the lexer by the file’s extension.
+        // Otherwise, look up the lexer by the file's extension.
         if let Some(llc) = language_lexers_compiled
             .map_ext_to_lexer_vec
             .get(ext.as_ref())
         {
             llc.first().unwrap()
         } else {
-            // The file type is unknown. Serve it raw, assuming it’s an image/video/etc.
+            // The file type is unknown. Serve it raw, assuming it's an image/video/etc.
             match actix_files::NamedFile::open_async(file_path).await {
                 Ok(v) => {
                     let res = v.into_response(req);
@@ -600,92 +600,12 @@ async fn serve_file(
         source_lexer(&file_contents, lexer)
     };
 
-    // Read non-standard markdown conversions from [markdown\_conversions.json](../../client/src/markdown_conversions.json) file
-    #[derive(Debug, Deserialize)]
-    struct Conversion {
-        conversions: Vec<Entry>
-    }
-    #[derive(Debug, Deserialize)]
-    struct Entry {
-        open_tag: String,
-        close_tag: String,
-        markdown: char,
-    }
-    let mut file = File::open("../client/src/markdown_conversions.json").await.unwrap();
-    let mut buff = String::new();
-    file.read_to_string(&mut buff).await.unwrap();
-    let conversion: Conversion = serde_json::from_str(&buff).unwrap();
-
+    // Convert doc blocks from Markdown to HTML
+    let mut options = Options::empty();
+    options.insert(Options::ENABLE_STRIKETHROUGH);
     for code_doc_block in &mut code_doc_block_arr {
         if let CodeDocBlock::DocBlock(ref mut doc_block) = code_doc_block {
-            
-            // For each all document block contents, iterate through each character to look for new Markdown conversions. If symbol is found, insert corresponding HTML tag.
-            let mut output_contents = Vec::new();
-            let mut tag_open = false;
-            let doc_block_lines = doc_block.contents.lines();
-            
-            for line in doc_block_lines {
-                let doc_block_contents = line.chars().collect::<Vec<char>>();
-                for i in 0..doc_block_contents.len() {
-                    let mut replaced = false;
-                    for entry in &conversion.conversions {
-                        // Check if character matches a markdown conversion.
-                        if doc_block_contents[i] == entry.markdown {
-                            let mut convert_to_html = false;
-
-                            if tag_open {
-                                for tag_char in entry.close_tag.chars() {
-                                    output_contents.push(tag_char);
-                                }
-                                tag_open = false;
-                                replaced = true;
-                            } 
-                            else {
-                                // Check to see if characters need converted to HTML. Only convert when the text has the correct format & not just anytime the symbol appears. 
-                                for j in i..doc_block_contents.len() {
-                                    if doc_block_contents[j] == entry.markdown {
-                                        convert_to_html = true;
-                                    }
-                                }
-                                // This is in place to ensure that subscript isn't converting '~~' so that strikethrough still works.
-                                if i == 0 {
-                                    if doc_block_contents[i+1] == entry.markdown {
-                                            convert_to_html = false;
-                                    }  
-                                }
-                                else if i == doc_block_contents.len() - 1 {
-                                    if doc_block_contents[i-1] == entry.markdown {
-                                        convert_to_html = false
-                                    }
-                                }
-                                else {
-                                    if doc_block_contents[i+1] == entry.markdown || doc_block_contents[i-1] == entry.markdown {
-                                        convert_to_html = false;
-                                    }
-                                }
-                                if convert_to_html {
-                                    for tag_char in entry.open_tag.chars() {
-                                        output_contents.push(tag_char);
-                                    }
-                                    tag_open = true;
-                                    replaced = true;
-                                }
-                            }
-                        } 
-                    }
-                    if !replaced {
-                        {
-                            output_contents.push(doc_block_contents[i]);
-                        }
-                    }
-                }
-                output_contents.push('\n');
-            }
-            let output: String = output_contents.into_iter().map(|i| i.to_string()).collect::<String>();
-           
-            // Convert doc blocks from Markdown to HTML
-            let options = Options::all();
-            let parser = Parser::new_ext(&output, options);
+            let parser = Parser::new_ext(&doc_block.contents, options);
             let mut html_output = String::new();
             html::push_html(&mut html_output, parser);
             doc_block.contents = html_output;
@@ -838,7 +758,7 @@ fn html_wrapper(body: &str) -> String {
     )
 }
 
-// Given text, escape it so it formats correctly as HTML. This is a translation of Python’s `html.escape` function.
+// Given text, escape it so it formats correctly as HTML. This is a translation of Python's `html.escape` function.
 fn escape_html(unsafe_text: &str) -> String {
     unsafe_text
         .replace('&', "&amp;")

From e37f30732b8988c349e95541cc6d0f919c46e4e9 Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Wed, 3 May 2023 22:22:11 -0500
Subject: [PATCH 12/13] Fix: Remove Smart Punctuation & add regex for escaping
 symbols

---
 client/static/css/CodeChatEditor.css | 300 ++++++++++++++-------------
 server/Cargo.toml                    |   1 +
 server/src/webserver.rs              |  21 +-
 3 files changed, 169 insertions(+), 153 deletions(-)

diff --git a/client/static/css/CodeChatEditor.css b/client/static/css/CodeChatEditor.css
index 5d24a8f..439b63f 100644
--- a/client/static/css/CodeChatEditor.css
+++ b/client/static/css/CodeChatEditor.css
@@ -21,164 +21,168 @@ Styles for the entire page layout
 ---------------------------------
 
 This is used only to store a reused variable value. See the [CSS docs](https://drafts.csswg.org/css-variables/). */
-:root {
-    --top-height: 6.7rem;
-    --body-padding: 0.2rem;
-}
-
-/* See [box sizing](https://css-tricks.com/box-sizing/) for the following technique to use `border-box` sizing. */
-html {
-    box-sizing: border-box;
-}
-
-*,
-*:before,
-*:after {
-    box-sizing: inherit;
-}
-
-body {
-    /* For box model simplicity, switch the padding and margin. */
-    padding: var(--body-padding);
-    margin: 0px;
-    /* The default of transparent makes VS Code’s black show up and confuse everything. For now, make it white. */
-    background: white;
-}
-
-/* Provide space at the top of the screen for the filename and TinyMCE menu bar. */
-#CodeChat-top {
-    height: var(--top-height);
-}
-
-/* The rest of the screen is the editor area. */
-#CodeChat-body {
-    height: calc(100vh - var(--top-height) - 2 * var(--body-padding));
-    overflow: auto;
-}
-
-/* Misc styling
+        :root {
+            --top-height: 6.7rem;
+            --body-padding: 0.2rem;
+        }
+        
+        /* See [box sizing](https://css-tricks.com/box-sizing/) for the following technique to use `border-box` sizing. */
+        html {
+            box-sizing: border-box;
+        }
+        
+        *,
+        *:before,
+        *:after {
+            box-sizing: inherit;
+        }
+        
+        body {
+            /* For box model simplicity, switch the padding and margin. */
+            padding: var(--body-padding);
+            margin: 0px;
+            /* The default of transparent makes VS Code's black show up and confuse everything. For now, make it white. */
+            background: white;
+        }
+        
+        /* Provide space at the top of the screen for the filename and TinyMCE menu bar. */
+        #CodeChat-top {
+            height: var(--top-height);
+        }
+        
+        /* The rest of the screen is the editor area. */
+        #CodeChat-body {
+            height: calc(100vh - var(--top-height) - 2 * var(--body-padding));
+            overflow: auto;
+        }
+        
+        /* Misc styling
 ------------
 
 Mark a hotkey. */
-.CodeChat-hotkey {
-    text-decoration: underline;
-}
-
-/* Make the filename compact. */
-#CodeChat-filename p {
-    margin: 0px;
-    white-space: nowrap;
-}
-
-/* Code block styling
+        .CodeChat-hotkey {
+            text-decoration: underline;
+        }
+        
+        /* Make the filename compact. */
+        #CodeChat-filename p {
+            margin: 0px;
+            white-space: nowrap;
+        }
+        
+        /* Code block styling
 ------------------
 
 Before the ACE Editor loads, use a monotype font. */
-.CodeChat-ACE {
-    font-family: monospace;
-    white-space: pre;
-    width: 100%;
-}
-
-.CodeChat-code {
-    /* Use [flexbox layout](https://developer.mozilla.org/en-US/docs/Web/CSS/CSS_Flexible_Box_Layout/Basic_Concepts_of_Flexbox) to style code blocks. The goal of this layout is:
-
-where the max line numbers spaces cause this code block’s line numbers to line up with all other code block line numbering. For example, if this code block is lines 9-23 (which the Ace editor will treat as a two-digit line number), but the largest line is 453 (three digits), then we need to insert one extra space to correct for the extra digit when the line number is > 99. A `<div class="CodeChat-ACE-gutter">` implements this indent. */
-    display: flex;
-}
-
-/* `Style the max line numbers spaces element discussed above.` */
-.CodeChat-ACE-gutter {
-    font-family: monospace;
-    white-space: pre;
-    background-color: #f0f0f0;
-    /* `` Make the not expand or shrink, but take exactly the width required by the text (spaces) it contains. Defaults for the `element cause it to take the rest of the space on the line.` `` */
-    flex: 0 0 auto;
-}
-
-/* `` `Hide the ACE Editor’s cursor when it’s inactive.` `` */
-.ace_hidden-cursors {
-    opacity: 0;
-}
-
-/* `` `Doc block styling` ``
-------------------------- */
-.CodeChat-doc {
-    /* `` `Use flexbox layout to style doc blocks. The approach is similar to code blocks; the overall goal is:` ``
+        .CodeChat-ACE {
+            font-family: monospace;
+            white-space: pre;
+            width: 100%;
+        }
+        
+        .CodeChat-code {
+            /* Use [flexbox layout](https://developer.mozilla.org/en-US/docs/Web/CSS/CSS_Flexible_Box_Layout/Basic_Concepts_of_Flexbox) to style code blocks. The goal of this layout is:
+
+<max line numbers spaces> <code>
+
+where the max line numbers spaces cause this code block's line numbers to line up with all other code block line numbering. For example, if this code block is lines 9-23 (which the Ace editor will treat as a two-digit line number), but the largest line is 453 (three digits), then we need to insert one extra space to correct for the extra digit when the line number is > 99. A `<div class="CodeChat-ACE-gutter">` implements this <max line numbers spaces> indent. */
+            display: flex;
+        }
+        
+        /* Style the max line numbers spaces element discussed above. */
+        .CodeChat-ACE-gutter {
+            font-family: monospace;
+            white-space: pre;
+            background-color: #f0f0f0;
+            /* Make the <max line number spaces> not expand or shrink, but take exactly the width required by the text (spaces) it contains. Defaults for the <code> element cause it to take the rest of the space on the line. */
+            flex: 0 0 auto;
+        }
+        
+        /* Hide the ACE Editor's cursor when it's inactive. */
+        .ace_hidden-cursors {
+            opacity: 0;
+        }
+        
+        /* Doc block styling
+----------------- */
+        .CodeChat-doc {
+            /* Use flexbox layout to style doc blocks. The approach is similar to code blocks; the overall goal is:
+
+<max line number spaces> <padding to match the ACE editor> <doc block indent> <doc block contents>
 
 where:
 
-*   `<div class="CodeChat-ACE-gutter-padding">` contains
-*   `<div class="CodeChat-ACE-padding">` contains
-*   `<div class="CodeChat-doc-indent">` contains
-*   `<div class="CodeChat-TinyMCE">` contains the */
-    display: flex;
-}
-
-/* `` `Match the ACE Editor gutter for doc blocks.` `` */
-.CodeChat-ACE-gutter-padding {
-    padding-left: 33px;
-    background-color: #f0f0f0;
-    /* `` `Make this not expand or shrink, but take exactly the width required by the text (spaces) it contains.` `` */
-    flex: 0 0 auto;
-    font-family: monospace;
-    white-space: pre;
-}
-
-/* `` `The ACE editor puts a bit of left padding in; match this so that code and doc blocks line up.` `` */
-.CodeChat-ACE-padding {
-    padding-left: 3px;
-}
-
-/* `` `Preserve whitespace in the indent of a doc block.` `` */
-.CodeChat-doc-indent {
-    /* `` `Make this not expand or shrink, but take exactly the width required by the text (spaces) it contains.` `` */
-    flex: 0 0 auto;
-    white-space: pre;
-    tab-size: 4;
-}
-
-/* `` `Remove the editor’s border when it’s selected, since this hides the cursor when the cursot is at the beginning of a line and isn’t necessary (the entire screen is an editor, not just that region.)` `` */
-.CodeChat-TinyMCE.mce-edit-focus {
-    outline-width: 0px;
-}
-
-/* `` `Combined code/doc block syling` ``
---------------------------------------
+*   `<div class="CodeChat-ACE-gutter-padding">` contains <max line number spaces>
+*   `<div class="CodeChat-ACE-padding">` contains <padding to match the ACE editor>
+*   `<div class="CodeChat-doc-indent">` contains <doc block indent>
+*   `<div class="CodeChat-TinyMCE">` contains the <doc block contents> */
+            display: flex;
+        }
+        
+        /* Match the ACE Editor gutter for doc blocks. */
+        .CodeChat-ACE-gutter-padding {
+            padding-left: 33px;
+            background-color: #f0f0f0;
+            /* Make this <max line number spaces> not expand or shrink, but take exactly the width required by the text (spaces) it contains. */
+            flex: 0 0 auto;
+            font-family: monospace;
+            white-space: pre;
+        }
+        
+        /* The ACE editor puts a bit of left padding in; match this so that code and doc blocks line up. */
+        .CodeChat-ACE-padding {
+            padding-left: 3px;
+        }
+        
+        /* Preserve whitespace in the indent of a doc block. */
+        .CodeChat-doc-indent {
+            /* Make this <doc block indent> not expand or shrink, but take exactly the width required by the text (spaces) it contains. */
+            flex: 0 0 auto;
+            white-space: pre;
+            tab-size: 4;
+        }
+        
+        /* Remove the editor's border when it's selected, since this hides the cursor when the cursot is at the beginning of a line and isn't necessary (the entire screen is an editor, not just that region.) */
+        .CodeChat-TinyMCE.mce-edit-focus {
+            outline-width: 0px;
+        }
+        
+        /* Combined code/doc block syling
+------------------------------
 
 Remove space between a code block followed by a doc block. Doc block elements typically have top margin and/or padding that produce this undesired space; remove it on the first element in the doc block, the first element of the first element in the doc block, etc. */
-.CodeChat-TinyMCE > *:first-child,
-.CodeChat-TinyMCE > *:first-child > *:first-child,
-.CodeChat-TinyMCE > *:first-child > *:first-child > *:first-child,
-.CodeChat-TinyMCE
-    > *:first-child
-    > *:first-child
-    > *:first-child
-    > *:first-child,
-.CodeChat-TinyMCE
-    > *:first-child
-    > *:first-child
-    > *:first-child
-    > *:first-child
-    > *:first-child {
-    margin-top: 0px;
-    padding-top: 0px;
-}
-
-/* [Remove space](remove-space) between a doc block followed by a code block. */
-.CodeChat-TinyMCE > *:last-child,
-.CodeChat-TinyMCE > *:last-child > *:last-child,
-.CodeChat-TinyMCE > *:last-child > *:last-child > *:last-child,
-.CodeChat-TinyMCE > *:last-child > *:last-child > *:last-child > *:last-child,
-.CodeChat-TinyMCE
-    > *:last-child
-    > *:last-child
-    > *:last-child
-    > *:last-child
-    > *:last-child {
-    margin-bottom: 0px;
-    padding-bottom: 0px;
-}
+        .CodeChat-TinyMCE > *:first-child,
+        .CodeChat-TinyMCE > *:first-child > *:first-child,
+        .CodeChat-TinyMCE > *:first-child > *:first-child > *:first-child,
+        .CodeChat-TinyMCE
+            > *:first-child
+            > *:first-child
+            > *:first-child
+            > *:first-child,
+        .CodeChat-TinyMCE
+            > *:first-child
+            > *:first-child
+            > *:first-child
+            > *:first-child
+            > *:first-child {
+            margin-top: 0px;
+            padding-top: 0px;
+        }
+        
+        /* [Remove space](remove-space) between a doc block followed by a code block. */
+        .CodeChat-TinyMCE > *:last-child,
+        .CodeChat-TinyMCE > *:last-child > *:last-child,
+        .CodeChat-TinyMCE > *:last-child > *:last-child > *:last-child,
+        .CodeChat-TinyMCE > *:last-child > *:last-child > *:last-child > *:last-child,
+        .CodeChat-TinyMCE
+            > *:last-child
+            > *:last-child
+            > *:last-child
+            > *:last-child
+            > *:last-child {
+            margin-bottom: 0px;
+            padding-bottom: 0px;
+        }
 
 /* Add a border around tables for converting back to HTML from Markdown. This class is used in [tinymce-webpack.mts](../../src/tinymce-webpack.mts) to add to all tables on initialization */
 .CodeChat-TinyMCE-TableDefault { 
@@ -189,4 +193,4 @@ Remove space between a code block followed by a doc block. Doc block elements ty
 .CodeChat-TinyMCE-TableDefault th, td{ 
     border: 1px solid black;
     border-collapse: collapse;
-}
\ No newline at end of file
+}
diff --git a/server/Cargo.toml b/server/Cargo.toml
index 145a330..820c2a1 100644
--- a/server/Cargo.toml
+++ b/server/Cargo.toml
@@ -34,6 +34,7 @@ actix-files = "0.6"
 actix-web = "4"
 lazy_static = "1"
 regex = "1"
+fancy-regex = "0.11.0"
 serde = { version = "1", features = ["derive"] }
 serde_json = "1"
 tokio = { version = "1", features = ["full"] }
diff --git a/server/src/webserver.rs b/server/src/webserver.rs
index 6ba1e78..45c0ce1 100644
--- a/server/src/webserver.rs
+++ b/server/src/webserver.rs
@@ -34,7 +34,8 @@ use actix_web::{
 };
 use lazy_static::lazy_static;
 use pulldown_cmark::{html, Options, Parser};
-use regex::Regex;
+use regex::{Regex};
+use fancy_regex::{Captures};
 use serde::{Deserialize, Serialize};
 use serde_json;
 use tokio::{
@@ -87,6 +88,8 @@ lazy_static! {
     static ref DRIVE_LETTER_REGEX: Regex = Regex::new("^[a-zA-Z]:$").unwrap();
     /// Match the lexer directive in a source file.
     static ref LEXER_DIRECTIVE: Regex = Regex::new(r#"CodeChat Editor lexer: (\w+)"#).unwrap();
+    // Matches all < and > symbols not within backticks. This allows the HTML to escape only the necessary symbols
+    static ref ESCAPE_HTML_REGEX: fancy_regex::Regex = fancy_regex::Regex::new("[<>](?![^`]*`)").unwrap();
 }
 
 /// Save endpoint
@@ -126,7 +129,7 @@ async fn save_source(
                 delimiter: match &cdb.1 {
                     // The delimiter was provided. Simply use that.
                     Some(v) => v.to_string(),
-                    // No delimiter was provided – fill one in.
+                    // No delimiter was provided -- fill one in.
                     None => {
                         if let Some(ic) = inline_comment {
                             ic.to_string()
@@ -625,7 +628,14 @@ async fn serve_file(
             let doc_block_lines = doc_block.contents.lines();
             
             for line in doc_block_lines {
-                let doc_block_contents = line.chars().collect::<Vec<char>>();
+                let line_escaped = &ESCAPE_HTML_REGEX.replace_all(line, |cap: &Captures| {
+                    match &cap[0] {
+                        ">" => "&gt;",
+                        "<" => "&lt;",
+                        _ => panic!("Escape HTML Regex replace failed"),
+                    }.to_string()
+                }).to_string();
+                let doc_block_contents = line_escaped.chars().collect::<Vec<char>>();
                 for i in 0..doc_block_contents.len() {
                     let mut replaced = false;
                     for entry in &conversion.conversions {
@@ -641,7 +651,7 @@ async fn serve_file(
                                 replaced = true;
                             } 
                             else {
-                                // Check to see if characters need converted to HTML. Only convert when the text has the correct format & not just anytime the symbol appears. 
+                                // Check to see if characters need converted to HTML. Only convert when the text has the correct format & not just anytime the symbol appears.
                                 for j in i..doc_block_contents.len() {
                                     if doc_block_contents[j] == entry.markdown {
                                         convert_to_html = true;
@@ -684,7 +694,8 @@ async fn serve_file(
             let output: String = output_contents.into_iter().map(|i| i.to_string()).collect::<String>();
            
             // Convert doc blocks from Markdown to HTML
-            let options = Options::all();
+            let mut options = Options::all();
+            options.remove(Options::ENABLE_SMART_PUNCTUATION);
             let parser = Parser::new_ext(&output, options);
             let mut html_output = String::new();
             html::push_html(&mut html_output, parser);

From 21f5a85653358ea36d19732c90ab55e4463175c5 Mon Sep 17 00:00:00 2001
From: mellenberger <kenzie.ellenberger22@gmail.com>
Date: Thu, 4 May 2023 00:11:40 -0500
Subject: [PATCH 13/13] Fix regex to ensure it convert all symbols not within
 code blocks

---
 client/src/CodeChatEditor.mts | 54 +++++++++++++++++------------------
 server/src/webserver.rs       |  4 +--
 2 files changed, 28 insertions(+), 30 deletions(-)

diff --git a/client/src/CodeChatEditor.mts b/client/src/CodeChatEditor.mts
index a0511f0..007bd97 100644
--- a/client/src/CodeChatEditor.mts
+++ b/client/src/CodeChatEditor.mts
@@ -36,7 +36,6 @@ import { tinymce, tinymce_init } from "./tinymce-webpack.mjs";
 import { gfm } from '@joplin/turndown-plugin-gfm';
 import TurndownService from "turndown";
 import MarkdownConversions from "../src/markdown_conversions.json";
-import MarkdownConversions from "../src/markdown_conversions.json";
 
 
 // ### CSS
@@ -67,33 +66,32 @@ for (let i = 0; i < MarkdownConversions.conversions.length; i++) {
       })
 }
 
-// Adapted from [this issue](https://github.com/mixmark-io/turndown/issues/152)
-// This rule is more complex than the above ones, so it is not in the .json file.
-interface HTMLEntityMap {
-    "<": string,
-    ">": string
-};
-turndownService.addRule("angle brackets",
-    {
-        filter: function (node: any, options: any){
-            const re = /(.*<.*>.*)+/g;
-            return(re.test(node.innerText));
-        },
-        replacement: function (content: string, node: any) {
-            let htmlEntityMap: HTMLEntityMap = {
-                "<": "&lt;",
-                ">": "&gt;"
-            };
-            // Convert the outer HTML of this node. This initial conversion handles things like
-            // converting "&lt;code&gt;" blocks to "\`backtick\`" blocks
-            let formattedHTML: string = baseTurndownService.turndown(node.outerHTML);
-            // then, we take that turndown'd version, and replace the "&lt;" and "&gt;" characters
-            // with their escaped html entity counterparts. This keeps them from potentially
-            // acting as html tags when they are not.
-            let finalHTML: string = formattedHTML.replace(/[<>]/g, function(m) {return htmlEntityMap[m as keyof HTMLEntityMap]});
-            return finalHTML;
-        }
-    });
+// Adapted from [this issue](https://github.com/mixmark-io/turndown/issues/152) This rule is more complex than the above ones, so it is not in the .json file. Add flag to test the client side solution for HTML escaping. Setting to false to test server side regex solution.
+const test_HTML_Escape = false;
+if (test_HTML_Escape) {
+    interface HTMLEntityMap {
+        "<": string,
+        ">": string
+    };
+    turndownService.addRule("angle brackets",
+        {
+            filter: function (node: any, options: any){
+                const re = /(.*<.*>.*)+/g;
+                return(re.test(node.innerText));
+            },
+            replacement: function (content: string, node: any) {
+                let htmlEntityMap: HTMLEntityMap = {
+                    "<": "&lt;",
+                    ">": "&gt;"
+                };
+                // Convert the outer HTML of this node. This initial conversion handles things like converting "<code>" blocks to "\`backtick\`" blocks
+                let formattedHTML: string = baseTurndownService.turndown(node.outerHTML);
+                // then, we take that turndown'd version, and replace the "<" and ">" characters with their escaped html entity counterparts. This keeps them from potentially acting as html tags when they are not.
+                let finalHTML: string = formattedHTML.replace(/[<>]/g, function(m) {return htmlEntityMap[m as keyof HTMLEntityMap]});
+                return finalHTML;
+            }
+        });
+}
 
 // Keep <a> anchor tags, because they get removed from Markdown otherwise
 turndownService.keep(["a"]);
diff --git a/server/src/webserver.rs b/server/src/webserver.rs
index 1781dc6..9c90f35 100644
--- a/server/src/webserver.rs
+++ b/server/src/webserver.rs
@@ -88,8 +88,8 @@ lazy_static! {
     static ref DRIVE_LETTER_REGEX: Regex = Regex::new("^[a-zA-Z]:$").unwrap();
     /// Match the lexer directive in a source file.
     static ref LEXER_DIRECTIVE: Regex = Regex::new(r#"CodeChat Editor lexer: (\w+)"#).unwrap();
-    // Matches all < and > symbols not within backticks. This allows the HTML to escape only the necessary symbols
-    static ref ESCAPE_HTML_REGEX: fancy_regex::Regex = fancy_regex::Regex::new("[<>](?![^`]*`)").unwrap();
+    // Matches all < and > symbols not within backticks. This allows the HTML to escape only the necessary symbols. This regex was found on [StackOverflow](https://stackoverflow.com/questions/64561041/regular-expression-to-match-a-word-but-not-inside-backticks).
+    static ref ESCAPE_HTML_REGEX: fancy_regex::Regex = fancy_regex::Regex::new("[<>](?=(?:[^`]*`[^`]*`)*[^`]*$)").unwrap();
 }
 
 /// Save endpoint