Squish surrounding whitespace in node text

Author: Wouter Coppieters

Cargo.lock

@@ -1,6 +1,6 @@
 [package]
 name = "htmltoadf"
-version = "0.1.4"
+version = "0.1.5"
 edition = "2021"
 license = "MIT"
 description = "An HTML to Atlassian Document Format (ADF) converter"

Cargo.toml

@@ -31,20 +31,20 @@ https://wouterken.github.io/htmltoadf/
 ### Install Binary from Crates.io with `cargo install`
 ```
 $ cargo install htmltoadf
-    installing htmltoadf v0.1.4 (/usr/src/html2adf)
+    installing htmltoadf v0.1.5 (/usr/src/html2adf)
     Updating crates.io index
  Downloading crates ...
   Downloaded lock_api v0.4.6
 --snip--
-      Compiling htmltoadf v0.1.4
+      Compiling htmltoadf v0.1.5
     Finished release [optimized] target(s) in 1m 42s
   Installing ~/.cargo/bin/htmltoadf
-   Installed package `htmltoadf v0.1.4` (executable `html2adf`)
+   Installed package `htmltoadf v0.1.5` (executable `html2adf`)
 ```
 
 ### Download Binary file from Github
 Pre-built binaries can be downloaded from here:
-https://github.com/wouterken/htmltoadf/releases/tag/0.1.2
+https://github.com/wouterken/htmltoadf/releases/tag/0.1.5
 
 ### Docker Image
 **Docker Repo:**
@@ -54,10 +54,10 @@ https://hub.docker.com/r/wouterken/html2adf
 **Usage**
 
 ```bash
-$ echo "<h1>Hello world<p>Test</p></h1>" | docker run --rm -i wouterken/html2adf:0.1.4
+$ echo "<h1>Hello world<p>Test</p></h1>" | docker run --rm -i wouterken/html2adf:0.1.5
 {"version":1,"type":"doc","content":[{"type":"heading","attrs":{"level":1},"content":[{"type":"text","text":"Hello world"},{"type":"text","text":"Test"}]}]}
 
-$ echo "<h1>Hello world<p>Test</p></h1>" | docker run --rm -i wouterken/html2adf:0.1.4 | jq
+$ echo "<h1>Hello world<p>Test</p></h1>" | docker run --rm -i wouterken/html2adf:0.1.5 | jq
 {
   "version": 1,
   "type": "doc",
@@ -90,7 +90,7 @@ $ echo "<h1>Hello world<p>Test</p></h1>" | docker run --rm -i wouterken/html2adf
 
 ```toml
 [dependencies]
-htmltoadf = "0.1.4"
+htmltoadf = "0.1.5"
 ```
 
 **Code**
@@ -201,6 +201,7 @@ The following features are implemented:
 * Compile binaries and create release
 * Build and push Docker image
 * Build and push WASM NPM package
+* Push crate
 * Update dependency in demo page
 * Push to VCS
 

README.md

@@ -78,7 +78,7 @@
 
   </style>
   <script defer type="module">
-   import init, {convert} from "https://unpkg.com/htmltoadf@0.1.4/htmltoadf.js";
+   import init, {convert} from "https://unpkg.com/htmltoadf@0.1.5/htmltoadf.js";
 
    let editor;
 
@@ -106,7 +106,7 @@ <h1>Header 1</h1>
        var str = '<span class=json-string>';
        var r = pIndent || '';
        if (pKey)
-        r = r + key + pKey.replace(/[": ]/g, '') + '</span>: ';
+        r = r + '"'+key + pKey.replace(/[": ]/g, '') + '"</span>: ';
       if (pVal)
         r = r + (pVal[0] == '"' ? str : val) + pVal + '</span>';
       return r + (pEnd || '');

docs/index.html

@@ -22,6 +22,14 @@ pub fn esc_hr(hrstr: String) -> String {
         .to_string();
 }
 
+/**
+ * Squish surrounding whitespace to a single space if it exists.
+ */
+pub fn squish_surrounding_whitespace(input: &str) -> String {
+    let re = Regex::new(r"^\s+|\s+$").unwrap();
+    re.replace_all(input, " ").to_string()
+}
+
 /**
  * We parse a raw scraper::HTML and return a
  * list of leaf doc nodes  (each with a linked list pointer to the root)
@@ -38,22 +46,22 @@ pub fn extract_leaves(fragment: &Html) -> Vec<DocNode> {
                     if element.value().name() == "iframe" || element.value().name() == "img" {
                         leaf_nodes.push(DocNode {
                             name: element.value().name().trim(),
-                            text: "".trim(),
+                            text: "".trim().to_owned(),
                             node,
                         })
                     } else if element.value().name() == HRBR_PLACEHOLDER {
                         leaf_nodes.push(DocNode {
                             name: "hr",
-                            text: "".trim(),
+                            text: "".trim().to_owned(),
                             node,
                         })
                     }
                 } else if let Node::Text(text_node) = node.value() {
                     if !text_node.text.trim().is_empty() {
                         leaf_nodes.push(DocNode {
                             name: "text",
-                            text: text_node.text.trim(),
-                            node: node,
+                            text: squish_surrounding_whitespace(&text_node.text),
+                            node,
                         })
                     }
                 }

src/extractor.rs

@@ -22,7 +22,7 @@ fn top_level() {
                         </li>
                         <li>
                             <div style='color: #0F0'>
-                                <span> With some blue text inside </span>
+                                <span>With some blue text inside</span>
                             </div>
                         </li>
                         <li>

src/tests/combination.rs

@@ -167,7 +167,7 @@ fn nested_1() {
                       "content": [
                         {
                           "type": "text",
-                          "text": "Nested List"
+                          "text": " Nested List "
                         }
                       ]
                     },

src/tests/lists.rs

@@ -5,7 +5,7 @@ use std::fmt;
 
 pub struct DocNode<'a> {
     pub name: &'a str,
-    pub text: &'a str,
+    pub text: String,
     pub node: NodeRef<'a, Node>,
 }