obsidian-plugin/keathmilligan_obsidian-paste-reformatter
1
0
Fork 0
mirror of https://github.com/keathmilligan/obsidian-paste-reformatter.git synced 2026-07-22 05:43:44 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Refactor, add styles

Browse source
This commit is contained in:
Keath Milligan 2025-03-24 20:20:11 -05:00
parent 45226a32ed
commit bbe04ac12b
No known key found for this signature in database
GPG key ID: 70CFBEC9F9DCCAC0
9 changed files with 542 additions and 286 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
.gitignore vendored
View file

@ -18,6 +18,7 @@ node_modules/
# Distribution files
dist/
dist
build/
# Dependency lock files (keep package.json)

158
cascade-rules.md Normal file
View file

@ -0,0 +1,158 @@
# Heading Cascade Rules
## Scanario: Cascade Heading Levels is Disabled
When Cascade Heading Levels is not enabled, then the Max Heading Level setting simply caps the maximum heading value that is allowed. For instance, if it is set to H3, then H1 and H2 would simply be changed to H3. Otherwise, heading values are not affected.
### Example
If Max heading Level is set to H3, then the following text when pasted:
```
# Heading 1
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
# Heading 1
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
```
Will be transformed into:
```
### Heading 1
### Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
​
### Heading 1
### Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
```
## Scenario: Cascade Heading Levels is Enabled
When Cascade Heading Levels is enabled, the heading values that are greater (in this case "greater" means H1 is greater than H2, H2 is greater than H3, etc.) than the max heading level setting are changed to the max heading level.
In addition subsequent headings values are demoted ("cascaded") to the heading level below.
### Example
When max heading level is set to H3 and the following text was pasted:
```
# Heading 1
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
# Heading 1
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
```
The result would be:
```
### Heading 1
#### Heading 2
##### Heading 3
###### Heading 4
###### Heading 5
###### Heading 6
### Heading 1
#### Heading 2
##### Heading 3
###### Heading 4
###### Heading 5
###### Heading 6
```
(heading levels will never be demoted below H6, e.g. "######")
### Another Example
When Max Heading Level is set to H2. The following should remain unchanged:
```
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
```
# Contextual Cascade Rules
Contextual Cascade is completely independent of Cascade Heading Levels.
## Scenario: Contextual Cascade is Disabled
When Contextual Cascade is disabled, then the normal Max Heading Level and Cascade Heading Levels rules apply as usual.
## Scenario: Contextual Cascade is Enabled
When Contextual Cascade is enabled, then the Max Heading Level and Cascade Heading Levels rules are superceded by the Contextual Cascade rules:
### Example
When text is pasted into an H2 section, for example, then headings are cascaded down from H3. For example, if the following text was pasted into an H2 section:
```
# Heading 1
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
# Heading 1
## Heading 2
### Heading 3
#### Heading 4
##### Heading 5
###### Heading 6
```
The result would be:
```
### Heading 1
#### Heading 2
##### Heading 3
###### Heading 4
###### Heading 5
###### Heading 6
### Heading 1
#### Heading 2
##### Heading 3
###### Heading 4
###### Heading 5
###### Heading 6
```
As with regular cascading, the heading levels will never be demoted below H6, e.g. "######".

34
copy-styles.mjs Normal file
View file

@ -0,0 +1,34 @@
// copy-styles.mjs
import { promises as fs } from 'fs';
import path from 'path';
// Ensure the dist directory exists
async function ensureDir(dir) {
try {
await fs.mkdir(dir, { recursive: true });
} catch (error) {
if (error.code !== 'EEXIST') {
throw error;
}
}
}
async function copyFiles() {
const distDir = path.resolve('./dist');
await ensureDir(distDir);
try {
// Copy styles.css to dist directory
await fs.copyFile('./styles.css', path.join(distDir, 'styles.css'));
console.log('Successfully copied styles.css to dist directory');
// Copy manifest.json to dist directory
await fs.copyFile('./manifest.json', path.join(distDir, 'manifest.json'));
console.log('Successfully copied manifest.json to dist directory');
} catch (error) {
console.error('Error copying files:', error);
process.exit(1);
}
}
copyFiles();

4
esbuild.config.mjs
View file

@ -15,7 +15,7 @@ const context = await esbuild.context({
banner: {
js: banner,
},
entryPoints: ["main.ts"],
entryPoints: ["src/main.ts"],
bundle: true,
external: [
"obsidian",
@ -37,7 +37,7 @@ const context = await esbuild.context({
logLevel: "info",
sourcemap: prod ? false : "inline",
treeShaking: true,
outfile: "main.js",
outfile: "dist/main.js",
minify: prod,
});

6
package.json
View file

@ -2,10 +2,10 @@
"name": "obsidian-sample-plugin",
"version": "1.0.0",
"description": "This is a sample plugin for Obsidian (https://obsidian.md)",
"main": "main.js",
"main": "dist/main.js",
"scripts": {
"dev": "node esbuild.config.mjs",
"build": "tsc -noEmit -skipLibCheck && node esbuild.config.mjs production",
"dev": "node esbuild.config.mjs && node copy-styles.mjs",
"build": "tsc -noEmit -skipLibCheck && node esbuild.config.mjs production && node copy-styles.mjs",
"version": "node version-bump.mjs && git add manifest.json versions.json"
},
"keywords": [],

115
src/htmlTransformer.ts Normal file
View file

@ -0,0 +1,115 @@
// Paste Reformatter - A plugin that re-formats pasted HTML text in Obsidian.
// Copyright (C) 2025 by Keath Milligan.
/**
* Transforms the HTML content before converting it to Markdown
* @param html The HTML content to transform
* @param settings The settings to use for transformation
* @returns The transformed HTML content
*/
export function transformHTML(
html: string,
settings: {
htmlRegexReplacements: Array<{pattern: string, replacement: string}>,
stripLineBreaks: boolean,
removeEmptyElements: boolean
}
): string {
// Apply regex replacements first
if (settings.htmlRegexReplacements && settings.htmlRegexReplacements.length > 0) {
for (const replacement of settings.htmlRegexReplacements) {
try {
const regex = new RegExp(replacement.pattern, 'g');
html = html.replace(regex, replacement.replacement);
} catch (error) {
console.error(`Error applying regex replacement: ${error}`);
}
}
}
// Create a temporary DOM element to parse the HTML
const parser = new DOMParser();
const doc = parser.parseFromString(html, 'text/html');
// Process line breaks if strip line breaks is enabled
if (settings.stripLineBreaks) {
// Find all <br> elements and remove them
const brElements = doc.querySelectorAll('br');
brElements.forEach(br => {
br.remove();
});
} else {
// If we're not stripping line breaks, convert them to special paragraph tags
// that will be preserved even if empty elements are removed
const brElements = doc.querySelectorAll('br');
brElements.forEach(br => {
// Create a special paragraph with a class that marks it as a line break
const lineBreakP = doc.createElement('p');
lineBreakP.className = 'preserve-line-break';
lineBreakP.setAttribute('data-preserve', 'true');
// Add a non-breaking space as placeholder content
// This ensures the paragraph isn't considered empty when Remove Empty Lines is enabled
// The unicode character will be invisible but ensures the line isn't empty
lineBreakP.textContent = '\u200B'; // Zero-width space
// Replace the <br> with our special paragraph
br.parentNode?.replaceChild(lineBreakP, br);
});
}
// Remove empty elements if enabled
if (settings.removeEmptyElements) {
// Function to check if an element is empty (no text content and no meaningful children)
const isElementEmpty = (element: Element): boolean => {
// Skip certain elements that are meaningful even when empty
if (['img', 'hr', 'br', 'input', 'iframe'].includes(element.tagName.toLowerCase())) {
return false;
}
// Skip our special line break paragraphs
if (element.hasAttribute('data-preserve')) {
return false;
}
// Check if it has any text content (excluding whitespace)
if (element.textContent && element.textContent.trim().length > 0) {
return false;
}
// Check if it has any non-empty children
for (let i = 0; i < element.children.length; i++) {
if (!isElementEmpty(element.children[i])) {
return false;
}
}
return true;
};
// Find and remove empty elements
// We need to use a while loop because the DOM changes as we remove elements
let emptyElementsFound = true;
while (emptyElementsFound) {
emptyElementsFound = false;
// Target common empty elements
const potentialEmptyElements = doc.querySelectorAll('p:not([data-preserve]), div, span, li, ul, ol, table, tr, td, th');
potentialEmptyElements.forEach(element => {
if (isElementEmpty(element)) {
element.remove();
emptyElementsFound = true;
}
});
// If no more empty elements are found, exit the loop
if (!emptyElementsFound) {
break;
}
}
}
// Return the modified HTML
const serializer = new XMLSerializer();
return serializer.serializeToString(doc.body);
}

317
main.ts → src/main.ts
View file

@ -2,6 +2,8 @@
// Copyright (C) 2025 by Keath Milligan.
import { App, MarkdownView, Plugin, PluginSettingTab, Setting, htmlToMarkdown, Notice } from 'obsidian';
import { transformHTML } from './htmlTransformer';
import { transformMarkdown } from './markdownTransformer';
interface RegexReplacement {
pattern: string;
@ -53,236 +55,6 @@ export default class PasteReformatter extends Plugin {
async saveSettings() {
await this.saveData(this.settings);
}
/**
* Transforms the HTML content before converting it to Markdown
* @param html The HTML content to transform
* @returns The transformed HTML content
*/
transformHTML(html: string): string {
// Apply regex replacements first
if (this.settings.htmlRegexReplacements && this.settings.htmlRegexReplacements.length > 0) {
for (const replacement of this.settings.htmlRegexReplacements) {
try {
const regex = new RegExp(replacement.pattern, 'g');
html = html.replace(regex, replacement.replacement);
} catch (error) {
console.error(`Error applying regex replacement: ${error}`);
}
}
}
// Create a temporary DOM element to parse the HTML
const parser = new DOMParser();
const doc = parser.parseFromString(html, 'text/html');
// Process line breaks if strip line breaks is enabled
if (this.settings.stripLineBreaks) {
// Find all <br> elements and remove them
const brElements = doc.querySelectorAll('br');
brElements.forEach(br => {
br.remove();
});
} else {
// If we're not stripping line breaks, convert them to special paragraph tags
// that will be preserved even if empty elements are removed
const brElements = doc.querySelectorAll('br');
brElements.forEach(br => {
// Create a special paragraph with a class that marks it as a line break
const lineBreakP = doc.createElement('p');
lineBreakP.className = 'preserve-line-break';
lineBreakP.setAttribute('data-preserve', 'true');
// Add a non-breaking space as placeholder content
// This ensures the paragraph isn't considered empty when Remove Empty Lines is enabled
// The unicode character will be invisible but ensures the line isn't empty
lineBreakP.innerHTML = '&#8203;'; // Zero-width space
// Replace the <br> with our special paragraph
br.parentNode?.replaceChild(lineBreakP, br);
});
}
// Remove empty elements if enabled
if (this.settings.removeEmptyElements) {
// Function to check if an element is empty (no text content and no meaningful children)
const isElementEmpty = (element: Element): boolean => {
// Skip certain elements that are meaningful even when empty
if (['img', 'hr', 'br', 'input', 'iframe'].includes(element.tagName.toLowerCase())) {
return false;
}
// Skip our special line break paragraphs
if (element.hasAttribute('data-preserve')) {
return false;
}
// Check if it has any text content (excluding whitespace)
if (element.textContent && element.textContent.trim().length > 0) {
return false;
}
// Check if it has any non-empty children
for (let i = 0; i < element.children.length; i++) {
if (!isElementEmpty(element.children[i])) {
return false;
}
}
return true;
};
// Find and remove empty elements
// We need to use a while loop because the DOM changes as we remove elements
let emptyElementsFound = true;
while (emptyElementsFound) {
emptyElementsFound = false;
// Target common empty elements
const potentialEmptyElements = doc.querySelectorAll('p:not([data-preserve]), div, span, li, ul, ol, table, tr, td, th');
potentialEmptyElements.forEach(element => {
if (isElementEmpty(element)) {
element.remove();
emptyElementsFound = true;
}
});
// If no more empty elements are found, exit the loop
if (!emptyElementsFound) {
break;
}
}
}
// Return the modified HTML
return doc.body.innerHTML;
}
/**
* Transforms the markdown content based on the plugin settings
* @param markdown The markdown content to transform
* @param contextLevel The current heading level for contextual cascade (0 if not in a heading section)
* @returns The transformed markdown content
*/
transformMarkdown(markdown: string, contextLevel: number = 0): string {
// Apply regex replacements if defined
if (this.settings.markdownRegexReplacements && this.settings.markdownRegexReplacements.length > 0) {
for (const replacement of this.settings.markdownRegexReplacements) {
try {
const regex = new RegExp(replacement.pattern, 'g');
markdown = markdown.replace(regex, replacement.replacement);
} catch (error) {
console.error(`Error applying markdown regex replacement: ${error}`);
}
}
}
// Find all heading lines
const headingRegex = /^(#{1,6})\s/gm;
// Process headings based on settings
if (this.settings.contextualCascade && contextLevel > 0) {
// Contextual cascade is enabled and we have a context level
markdown = markdown.replace(headingRegex, (match, hashes) => {
const currentLevel = hashes.length;
let newLevel;
// Start headings one level deeper than the context
const baseLevel = contextLevel + 1;
// Calculate the offset from H1
const offset = currentLevel - 1;
// Apply the offset to the base level, ensuring we don't exceed H6
newLevel = Math.min(baseLevel + offset, 6);
// Return the new heading with the adjusted level
return '#'.repeat(newLevel) + ' ';
});
} else if (this.settings.maxHeadingLevel > 1) {
// Only process max heading level if contextual cascade is not active
markdown = markdown.replace(headingRegex, (match, hashes) => {
const currentLevel = hashes.length;
let newLevel = currentLevel;
if (this.settings.cascadeHeadingLevels) {
// When cascade is enabled, heading values greater than max level are changed to max level
// and subsequent heading values are demoted to the level below
// For example, if max level is 3:
// H1 -> H3 (max level)
// H2 -> H4 (max level + 1)
// H3 -> H5 (max level + 2)
// H4+ -> H6 (capped at H6)
// Calculate the offset from H1
const offset = currentLevel - 1;
// Apply the offset to the max level, ensuring we don't exceed H6
newLevel = Math.min(this.settings.maxHeadingLevel + offset, 6);
} else {
// Only max heading level is enabled, cascade is disabled
if (currentLevel < this.settings.maxHeadingLevel) {
// If heading level is less than max, increase it to max level
newLevel = this.settings.maxHeadingLevel;
} else {
// Keep headings at their original level if they're already at or deeper than max
newLevel = currentLevel;
}
}
// Return the new heading with the adjusted level
return '#'.repeat(newLevel) + ' ';
});
}
// First handle the special line break markers
let preserveLineBreaks = !this.settings.stripLineBreaks;
// If we're not stripping line breaks, we need to handle the special markers
if (preserveLineBreaks) {
// Replace special line break markers with a unique placeholder that won't be affected by empty line removal
const lineBreakPlaceholder = '___LINE_BREAK_PLACEHOLDER___';
markdown = markdown.replace(/<p class="preserve-line-break"[^>]*>.*?<\/p>/g, lineBreakPlaceholder);
markdown = markdown.replace(/<p data-preserve="true"[^>]*>.*?<\/p>/g, lineBreakPlaceholder);
// Remove empty lines if enabled
if (this.settings.removeEmptyLines) {
// First, normalize line endings to ensure consistent processing
markdown = markdown.replace(/\r\n/g, '\n');
// Split the content into lines
const lines = markdown.split('\n');
// Filter out empty lines, but keep our placeholders
const filteredLines = lines.filter(line => {
return line.trim() !== '' || line.includes(lineBreakPlaceholder);
});
// Join the filtered lines back together
markdown = filteredLines.join('\n');
}
// Now replace our placeholders with actual line breaks
markdown = markdown.replace(new RegExp(lineBreakPlaceholder, 'g'), '\n');
} else {
// If we're stripping line breaks, just remove empty lines normally
if (this.settings.removeEmptyLines) {
// First, normalize line endings to ensure consistent processing
markdown = markdown.replace(/\r\n/g, '\n');
// Split the content into lines
const lines = markdown.split('\n');
// Filter out all empty lines
const filteredLines = lines.filter(line => line.trim() !== '');
// Join the filtered lines back together
markdown = filteredLines.join('\n');
}
}
return markdown;
}
onPaste(event: ClipboardEvent) {
// Check if there's content in the clipboard
@ -314,7 +86,7 @@ export default class PasteReformatter extends Plugin {
// console.log("HTML content:", html);
// Transform HTML before converting to Markdown
const transformedHTML = this.transformHTML(html);
const transformedHTML = transformHTML(html, this.settings);
// Use Obsidian's built-in htmlToMarkdown function
markdown = htmlToMarkdown(transformedHTML);
@ -336,7 +108,7 @@ export default class PasteReformatter extends Plugin {
}
// Apply settings to transform the markdown
markdown = this.transformMarkdown(markdown, contextLevel);
markdown = transformMarkdown(markdown, this.settings, contextLevel);
// Replace the current selection with the converted markdown
editor.replaceSelection(markdown);
@ -376,7 +148,6 @@ export default class PasteReformatter extends Plugin {
// No heading found above the cursor
return 0;
}
}
class PasteReformmatterSettingsTab extends PluginSettingTab {
@ -446,14 +217,12 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
});
// Create a container for the regex replacement rows
const regexContainer = containerEl.createDiv('regex-replacements-container');
regexContainer.style.marginLeft = '40px';
regexContainer.style.marginBottom = '20px';
const regexContainer = containerEl.createDiv();
regexContainer.addClass('regex-replacements-container');
// Create a table for the regex replacements
const table = regexContainer.createEl('table');
table.style.width = '100%';
table.style.borderCollapse = 'collapse';
table.addClass('regex-table');
// Create the header row
const thead = table.createEl('thead');
@ -462,21 +231,19 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
// Pattern header
const patternHeader = headerRow.createEl('th');
patternHeader.setText('Pattern');
patternHeader.style.textAlign = 'left';
patternHeader.style.padding = '5px';
patternHeader.style.width = '40%';
patternHeader.addClass('regex-th');
patternHeader.addClass('regex-th-pattern');
// Replacement header
const replacementHeader = headerRow.createEl('th');
replacementHeader.setText('Replacement');
replacementHeader.style.textAlign = 'left';
replacementHeader.style.padding = '5px';
replacementHeader.style.width = '40%';
replacementHeader.addClass('regex-th');
replacementHeader.addClass('regex-th-replacement');
// Actions header
const actionsHeader = headerRow.createEl('th');
actionsHeader.style.width = '20%';
actionsHeader.style.padding = '5px';
actionsHeader.addClass('regex-th');
actionsHeader.addClass('regex-th-actions');
// Create the table body
const tbody = table.createEl('tbody');
@ -487,14 +254,14 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
// Pattern cell
const patternCell = row.createEl('td');
patternCell.style.padding = '5px';
patternCell.addClass('regex-td');
// Pattern input
const patternInput = document.createElement('input');
patternInput.type = 'text';
patternInput.value = replacement.pattern;
patternInput.placeholder = 'Regular expression pattern';
patternInput.style.width = '100%';
patternInput.addClass('regex-input');
patternInput.addEventListener('change', async () => {
this.plugin.settings.htmlRegexReplacements[index].pattern = patternInput.value;
await this.plugin.saveSettings();
@ -503,14 +270,14 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
// Replacement cell
const replacementCell = row.createEl('td');
replacementCell.style.padding = '5px';
replacementCell.addClass('regex-td');
// Replacement input
const replacementInput = document.createElement('input');
replacementInput.type = 'text';
replacementInput.value = replacement.replacement;
replacementInput.placeholder = 'Replacement value (can use $1, $2, etc.)';
replacementInput.style.width = '100%';
replacementInput.addClass('regex-input');
replacementInput.addEventListener('change', async () => {
this.plugin.settings.htmlRegexReplacements[index].replacement = replacementInput.value;
await this.plugin.saveSettings();
@ -519,8 +286,8 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
// Actions cell
const actionsCell = row.createEl('td');
actionsCell.style.padding = '5px';
actionsCell.style.textAlign = 'center';
actionsCell.addClass('regex-td');
actionsCell.addClass('regex-td-actions');
// Remove button
const removeButton = document.createElement('button');
@ -538,11 +305,7 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
const emptyRow = tbody.createEl('tr');
const emptyCell = emptyRow.createEl('td');
emptyCell.colSpan = 3;
emptyCell.style.textAlign = 'center';
emptyCell.style.padding = '10px';
emptyCell.style.fontStyle = 'italic';
emptyCell.style.color = 'var(--text-faint)';
emptyCell.style.opacity = '0.7';
emptyCell.addClass('regex-empty-message');
emptyCell.setText('No replacements defined. Click "Add Replacement" to add one.');
}
@ -634,14 +397,12 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
});
// Create a container for the regex replacement rows
const markdownRegexContainer = containerEl.createDiv('markdown-regex-replacements-container');
markdownRegexContainer.style.marginLeft = '40px';
markdownRegexContainer.style.marginBottom = '20px';
const markdownRegexContainer = containerEl.createDiv();
markdownRegexContainer.addClass('regex-replacements-container');
// Create a table for the regex replacements
const markdownRegexTable = markdownRegexContainer.createEl('table');
markdownRegexTable.style.width = '100%';
markdownRegexTable.style.borderCollapse = 'collapse';
markdownRegexTable.addClass('regex-table');
// Create the header row
const markdownRegexThead = markdownRegexTable.createEl('thead');
@ -650,21 +411,19 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
// Pattern header
const markdownRegexPatternHeader = markdownRegexHeaderRow.createEl('th');
markdownRegexPatternHeader.setText('Pattern');
markdownRegexPatternHeader.style.textAlign = 'left';
markdownRegexPatternHeader.style.padding = '5px';
markdownRegexPatternHeader.style.width = '40%';
markdownRegexPatternHeader.addClass('regex-th');
markdownRegexPatternHeader.addClass('regex-th-pattern');
// Replacement header
const markdownRegexReplacementHeader = markdownRegexHeaderRow.createEl('th');
markdownRegexReplacementHeader.setText('Replacement');
markdownRegexReplacementHeader.style.textAlign = 'left';
markdownRegexReplacementHeader.style.padding = '5px';
markdownRegexReplacementHeader.style.width = '40%';
markdownRegexReplacementHeader.addClass('regex-th');
markdownRegexReplacementHeader.addClass('regex-th-replacement');
// Actions header
const markdownRegexActionsHeader = markdownRegexHeaderRow.createEl('th');
markdownRegexActionsHeader.style.width = '20%';
markdownRegexActionsHeader.style.padding = '5px';
markdownRegexActionsHeader.addClass('regex-th');
markdownRegexActionsHeader.addClass('regex-th-actions');
// Create the table body
const markdownRegexTbody = markdownRegexTable.createEl('tbody');
@ -675,14 +434,14 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
// Pattern cell
const patternCell = row.createEl('td');
patternCell.style.padding = '5px';
patternCell.addClass('regex-td');
// Pattern input
const patternInput = document.createElement('input');
patternInput.type = 'text';
patternInput.value = replacement.pattern;
patternInput.placeholder = 'Regular expression pattern';
patternInput.style.width = '100%';
patternInput.addClass('regex-input');
patternInput.addEventListener('change', async () => {
this.plugin.settings.markdownRegexReplacements[index].pattern = patternInput.value;
await this.plugin.saveSettings();
@ -691,14 +450,14 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
// Replacement cell
const replacementCell = row.createEl('td');
replacementCell.style.padding = '5px';
replacementCell.addClass('regex-td');
// Replacement input
const replacementInput = document.createElement('input');
replacementInput.type = 'text';
replacementInput.value = replacement.replacement;
replacementInput.placeholder = 'Replacement value (can use $1, $2, etc.)';
replacementInput.style.width = '100%';
replacementInput.addClass('regex-input');
replacementInput.addEventListener('change', async () => {
this.plugin.settings.markdownRegexReplacements[index].replacement = replacementInput.value;
await this.plugin.saveSettings();
@ -707,8 +466,8 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
// Actions cell
const actionsCell = row.createEl('td');
actionsCell.style.padding = '5px';
actionsCell.style.textAlign = 'center';
actionsCell.addClass('regex-td');
actionsCell.addClass('regex-td-actions');
// Remove button
const removeButton = document.createElement('button');
@ -726,11 +485,7 @@ class PasteReformmatterSettingsTab extends PluginSettingTab {
const emptyRow = markdownRegexTbody.createEl('tr');
const emptyCell = emptyRow.createEl('td');
emptyCell.colSpan = 3;
emptyCell.style.textAlign = 'center';
emptyCell.style.padding = '10px';
emptyCell.style.fontStyle = 'italic';
emptyCell.style.color = 'var(--text-faint)';
emptyCell.style.opacity = '0.7';
emptyCell.addClass('regex-empty-message');
emptyCell.setText('No replacements defined. Click "Add Replacement" to add one.');
}
}

144
src/markdownTransformer.ts Normal file
View file

@ -0,0 +1,144 @@
// Paste Reformatter - A plugin that re-formats pasted HTML text in Obsidian.
// Copyright (C) 2025 by Keath Milligan.
/**
* Transforms the markdown content based on the plugin settings
* @param markdown The markdown content to transform
* @param settings The settings to use for transformation
* @param contextLevel The current heading level for contextual cascade (0 if not in a heading section)
* @returns The transformed markdown content
*/
export function transformMarkdown(
markdown: string,
settings: {
markdownRegexReplacements: Array<{pattern: string, replacement: string}>,
contextualCascade: boolean,
maxHeadingLevel: number,
cascadeHeadingLevels: boolean,
stripLineBreaks: boolean,
removeEmptyLines: boolean
},
contextLevel: number = 0
): string {
// Apply regex replacements if defined
if (settings.markdownRegexReplacements && settings.markdownRegexReplacements.length > 0) {
for (const replacement of settings.markdownRegexReplacements) {
try {
const regex = new RegExp(replacement.pattern, 'g');
markdown = markdown.replace(regex, replacement.replacement);
} catch (error) {
console.error(`Error applying markdown regex replacement: ${error}`);
}
}
}
// Find all heading lines
const headingRegex = /^(#{1,6})\s/gm;
// Process headings based on settings
if (settings.contextualCascade && contextLevel > 0) {
let delta = -1;
let cascading = false;
// Contextual cascade is enabled and we have a context level
markdown = markdown.replace(headingRegex, (match, hashes) => {
const currentLevel = hashes.length;
let newLevel = currentLevel;
if (cascading) {
// Cascade subsequent levels below the context level
newLevel = Math.min(currentLevel + delta, 6);
console.log(`contextual cascade: delta ${delta}`);
} else if (currentLevel <= contextLevel) {
// Intiate contextual cascading
newLevel = Math.min(contextLevel + 1, 6);
delta = newLevel - currentLevel;
cascading = true;
console.log(`contextual cascade initiated: delta: ${delta}`);
} // else nothing to do
console.log(`result: current level: ${currentLevel}, new level: ${newLevel}`);
// Return the new heading with the adjusted level
return '#'.repeat(newLevel) + ' ';
});
} else if (settings.maxHeadingLevel > 1) {
let delta = -1;
let cascading = false;
markdown = markdown.replace(headingRegex, (match, hashes) => {
const currentLevel = hashes.length;
let newLevel = currentLevel;
if (settings.cascadeHeadingLevels) {
// If cascading is enabled, start cascading subsequent headings down if needed
if (cascading) {
// Cascade subsequent headers down, don't go deeper than H6
newLevel = Math.min(currentLevel + delta, 6);
console.log(`cascading: delta: ${delta}`);
} else if (currentLevel < settings.maxHeadingLevel) {
newLevel = settings.maxHeadingLevel;
delta = newLevel - currentLevel;
cascading = true; // we need to cascade
console.log(`cascade initiated: delta: ${delta}`);
} // else nothing to do, heading is good as is
} else {
// Cascading not enabled, just cap heading levels at max
newLevel = Math.max(currentLevel, settings.maxHeadingLevel)
}
console.log(`result: current level: ${currentLevel}, new level: ${newLevel}`);
// Return the new heading with the adjusted level
return '#'.repeat(newLevel) + ' ';
});
}
// First handle the special line break markers
let preserveLineBreaks = !settings.stripLineBreaks;
// If we're not stripping line breaks, we need to handle the special markers
if (preserveLineBreaks) {
// Replace special line break markers with a unique placeholder that won't be affected by empty line removal
const lineBreakPlaceholder = '___LINE_BREAK_PLACEHOLDER___';
markdown = markdown.replace(/<p class="preserve-line-break"[^>]*>.*?<\/p>/g, lineBreakPlaceholder);
markdown = markdown.replace(/<p data-preserve="true"[^>]*>.*?<\/p>/g, lineBreakPlaceholder);
// Remove empty lines if enabled
if (settings.removeEmptyLines) {
// First, normalize line endings to ensure consistent processing
markdown = markdown.replace(/\r\n/g, '\n');
// Split the content into lines
const lines = markdown.split('\n');
// Filter out empty lines, but keep our placeholders
const filteredLines = lines.filter(line => {
return line.trim() !== '' || line.includes(lineBreakPlaceholder);
});
// Join the filtered lines back together
markdown = filteredLines.join('\n');
}
// Now replace our placeholders with actual line breaks
markdown = markdown.replace(new RegExp(lineBreakPlaceholder, 'g'), '\n');
} else {
// If we're stripping line breaks, just remove empty lines normally
if (settings.removeEmptyLines) {
// First, normalize line endings to ensure consistent processing
markdown = markdown.replace(/\r\n/g, '\n');
// Split the content into lines
const lines = markdown.split('\n');
// Filter out all empty lines
const filteredLines = lines.filter(line => line.trim() !== '');
// Join the filtered lines back together
markdown = filteredLines.join('\n');
}
}
return markdown;
}

49
styles.css
View file

@ -6,3 +6,52 @@ available in the app when your plugin is enabled.
If your plugin does not need CSS, delete this file.
*/
/* Regex Replacement Container Styles */
.regex-replacements-container {
margin-left: 40px;
margin-bottom: 20px;
}
/* Regex Replacement Table Styles */
.regex-table {
width: 100%;
border-collapse: collapse;
}
/* Table Header Styles */
.regex-th {
text-align: left;
padding: 5px;
}
.regex-th-pattern, .regex-th-replacement {
width: 40%;
}
.regex-th-actions {
width: 20%;
}
/* Table Cell Styles */
.regex-td {
padding: 5px;
}
.regex-td-actions {
text-align: center;
}
/* Input Field Styles */
.regex-input {
width: 100%;
}
/* Empty Message Styles */
.regex-empty-message {
text-align: center;
padding: 10px;
font-style: italic;
color: var(--text-muted);
opacity: 0.7;
}