feat: support line-break (#111)

rasendubi · May 19, 2024 · b45baf9 · b45baf9
1 parent dbf6452
commit b45baf9
Show file tree

Hide file tree

Showing 12 changed files with 161 additions and 3 deletions.
diff --git a/.changeset/quiet-berries-sniff.md b/.changeset/quiet-berries-sniff.md
@@ -0,0 +1,10 @@
+---
+'uniorg-parse': major
+'uniorg-stringify': minor
+'uniorg-rehype': minor
+'uniorg': minor
+---
+
+Support `line-break` in uniorg, uniorg-parse, uniorg-rehype, and uniorg-stringify.
+
+This is a breaking change for uniorg-parse as it may output nodes unknown to downstream users (uniorg-rehype and uniorg-stringify). If you upgrade uniorg-parse, you should also upgrade uniorg-rehype and uniorg-stringify to the corresponding versions.
diff --git a/README.md b/README.md
@@ -46,8 +46,6 @@ However, there are a couple of places I haven't finished yet:
 - babel-call, inline-babel-call, inline-src-block
 - dynamic-block
 - target, radio-target
-- line-break
-- export-snippet
 - macro
 - switches and parameters in src-block and example-block
 - repeater/warning props in timestamp

diff --git a/packages/uniorg-parse/src/__snapshots__/parser.spec.ts.snap b/packages/uniorg-parse/src/__snapshots__/parser.spec.ts.snap
@@ -2438,6 +2438,79 @@ children:
         value: "."
 `;
 
+exports[`org/parser line-break fake line break (must be at end of line) 1`] = `
+type: "org-data"
+contentsBegin: 0
+contentsEnd: 27
+children:
+  - type: "paragraph"
+    affiliated: {}
+    contentsBegin: 0
+    contentsEnd: 27
+    children:
+      - type: "text"
+        value: "some text\\\\"
+      - type: "entity"
+        useBrackets: false
+        name: "not"
+        latex: "\\\\textlnot{}"
+        requireLatexMath: false
+        html: "&not;"
+        ascii: "[angled dash]"
+        latin1: "¬"
+        utf8: "¬"
+      - type: "text"
+        value: " a line break"
+`;
+
+exports[`org/parser line-break fake line break (triple backslash) 1`] = `
+type: "org-data"
+contentsBegin: 0
+contentsEnd: 26
+children:
+  - type: "paragraph"
+    affiliated: {}
+    contentsBegin: 0
+    contentsEnd: 26
+    children:
+      - type: "text"
+        value: "some text\\\\\\\\\\\\\\nnot next line"
+`;
+
+exports[`org/parser line-break valid line break 1`] = `
+type: "org-data"
+contentsBegin: 0
+contentsEnd: 22
+children:
+  - type: "paragraph"
+    affiliated: {}
+    contentsBegin: 0
+    contentsEnd: 22
+    children:
+      - type: "text"
+        value: "some text"
+      - type: "line-break"
+      - type: "text"
+        value: "line break"
+`;
+
+exports[`org/parser line-break whitespace after \\\\ 1`] = `
+type: "org-data"
+contentsBegin: 0
+contentsEnd: 24
+children:
+  - type: "paragraph"
+    affiliated: {}
+    contentsBegin: 0
+    contentsEnd: 24
+    children:
+      - type: "text"
+        value: "some text"
+      - type: "line-break"
+      - type: "text"
+        value: "line break"
+`;
+
 exports[`org/parser links ./ as start of file link 1`] = `
 type: "org-data"
 contentsBegin: 0

diff --git a/packages/uniorg-parse/src/parser.spec.ts b/packages/uniorg-parse/src/parser.spec.ts
@@ -1106,4 +1106,25 @@ more text
 
     itParses('fake export snippet, missing backend', '@@<b>@@');
   });
+
+  describe('line-break', () => {
+    itParses(
+      'valid line break',
+      `some text\\\\
+line break`
+    );
+
+    itParses('whitespace after \\\\', `some text\\\\  \nline break`);
+
+    itParses(
+      'fake line break (must be at end of line)',
+      `some text\\\\not a line break`
+    );
+
+    itParses(
+      'fake line break (triple backslash)',
+      `some text\\\\\\
+not next line`
+    );
+  });
 });
diff --git a/packages/uniorg-parse/src/parser.ts b/packages/uniorg-parse/src/parser.ts
@@ -53,6 +53,7 @@ import {
   CitationSuffix,
   CitationCommonSuffix,
   ExportSnippet,
+  LineBreak,
 } from 'uniorg';
 
 import { getOrgEntity } from './entities.js';
@@ -638,7 +639,9 @@ class Parser {
         break;
       case '\\':
         if (c[1] === '\\') {
-          // TODO: line break parser
+          if (restriction.has('line-break')) {
+            return this.parseLineBreak();
+          }
         } else {
           const offset = this.r.offset();
           const entity = restriction.has('entity') && this.parseEntity();
@@ -1707,6 +1710,19 @@ class Parser {
     return u('latex-fragment', { value, contents: contents ?? value });
   }
 
+  private parseLineBreak(): LineBreak | null {
+    const m = this.r.lookingAt(/\\\\[ \t]*$/m);
+    if (!m) return null;
+
+    // check character before linebreak
+    this.r.backoff(1);
+    if (this.r.peek(1) === '\\') return null;
+
+    this.r.advance(this.r.line());
+
+    return u('line-break');
+  }
+
   private parseFootnoteReference(): FootnoteReference | null {
     const begin = this.r.offset();
     const m = this.r.match(footnoteRe);

diff --git a/packages/uniorg-rehype/src/__snapshots__/org-to-hast.spec.ts.snap b/packages/uniorg-rehype/src/__snapshots__/org-to-hast.spec.ts.snap
@@ -354,6 +354,12 @@ exports[`org/org-to-hast latex-fragment 1`] = `
 
 `;
 
+exports[`org/org-to-hast line-break 1`] = `
+
+<p>hello<br>world</p>
+
+`;
+
 exports[`org/org-to-hast link 1`] = `
 
 <p><a href="https://example.com">https://example.com</a></p>

diff --git a/packages/uniorg-rehype/src/org-to-hast.spec.ts b/packages/uniorg-rehype/src/org-to-hast.spec.ts
@@ -474,6 +474,12 @@ either $$ a=+\\sqrt{2} $$ or \\[ a=-\\sqrt{2} \\].`
     `[cite/style:common prefix; prefix @key suffix; @key2; common suffix]`
   );
 
+  hastTest(
+    'line-break',
+    `hello\\\\
+world`
+  );
+
   test('respects hProperties', () => {
     const s = unified()
       .use(orgParse)

diff --git a/packages/uniorg-rehype/src/org-to-hast.ts b/packages/uniorg-rehype/src/org-to-hast.ts
@@ -84,6 +84,9 @@ const defaultHandlers: Handlers = {
     if (org.backEnd !== 'html') return null;
     return u('raw', org.value) as any;
   },
+  'line-break': function (org) {
+    return this.h(org, 'br');
+  },
 };
 
 function renderAsChildren(

diff --git a/packages/uniorg-stringify/src/__snapshots__/stringify.spec.ts.snap b/packages/uniorg-stringify/src/__snapshots__/stringify.spec.ts.snap
@@ -295,6 +295,18 @@ either $$ a=+\\\\sqrt{2} $$ or \\\\[ a=-\\\\sqrt{2} \\\\].
 "
 `;
 
+exports[`stringify line-break 1`] = `
+"hello\\\\\\\\
+world!
+"
+`;
+
+exports[`stringify line-break with trailing whitespace 1`] = `
+"hello\\\\\\\\
+world!
+"
+`;
+
 exports[`stringify links angle link 1`] = `
 "<http://example.com>
 "

diff --git a/packages/uniorg-stringify/src/stringify.spec.ts b/packages/uniorg-stringify/src/stringify.spec.ts
@@ -514,4 +514,11 @@ some text
   });
 
   test('handle export-snippet', `@@backend:custom value@@`);
+
+  test(
+    'line-break',
+    `hello\\\\
+world!`
+  );
+  test('line-break with trailing whitespace', `hello\\\\  \nworld!`);
 });
diff --git a/packages/uniorg-stringify/src/stringify.ts b/packages/uniorg-stringify/src/stringify.ts
@@ -18,6 +18,7 @@ export type StringifyOptions = {
 const defaultOptions: StringifyOptions = {
   handlers: {
     'export-snippet': (org) => `@@${org.backEnd}:${org.value}@@`,
+    'line-break': () => `\\\\\n`,
   },
 };
 

diff --git a/packages/uniorg/src/index.ts b/packages/uniorg/src/index.ts
@@ -77,6 +77,7 @@ export type ObjectType =
   | LatexFragment
   | Entity
   | ExportSnippet
+  | LineBreak
   | TableCell;
 
 export type OrgNode = GreaterElementType | ElementType | ObjectType;
@@ -197,6 +198,10 @@ export interface ExportSnippet extends Node {
   value: string;
 }
 
+export interface LineBreak extends Node {
+  type: 'line-break';
+}
+
 export interface List extends GreaterElement {
   type: 'plain-list';
   listType: 'ordered' | 'unordered' | 'descriptive';