HTML contains structure and presentation
A useful HTML document might contain a heading, two paragraphs, a list, a labelled link and a small table. The same file can also contain fonts, colours, grid layout, navigation, forms, scripts and tracking images. A rich-text editor needs the first group. Copying the second group into a working document usually creates mismatched type, strange spacing or content that does not belong in the document at all.
The safe conversion is not a screenshot of the webpage and it is not a blind copy of the source DOM. It reads semantic elements, turns them into a portable document model, removes active or decorative parts, then writes a representation the destination can understand. Pasteful uses Markdown as that intermediate model because headings, lists, links, quotes, code and simple tables remain inspectable before anything reaches another app.
- Keep document meaning such as headings, paragraphs and list order
- Remove page layout, fonts, colours, classes and IDs
- Do not execute scripts or load remote resources
- Show flattened table cells, removed links and missing images as losses
- Write both rich and plain clipboard representations where the browser allows it
Raw HTML and rendered webpage copy are different inputs
If you can see tags such as <h1> and <ul>, you have raw HTML source. Choose HTML code or file in Pasteful, paste the tags or open a local .html file. The editor keeps the source visible while the preview shows the document recovered from it. This is useful for exports from another system, saved webpages, CMS fragments and generated email bodies.
If you select text on a webpage and copy it, the clipboard may already contain text/plain and text/html representations. Use Auto-detect or Rich text for that job. Pasteful compares the visible text with the rich representation and prefers the safe structure only when the two agree. Keeping these routes separate avoids treating literal tags as prose or pretending a rendered selection is a complete HTML file.
MDN: ClipboardItem · MDN: read clipboard representations · Convert HTML to Markdown
What Pasteful keeps from HTML
The current converter keeps six heading levels, paragraphs, strong and emphasised text, strikethrough, line breaks, horizontal rules, ordered and unordered lists, block quotes, inline code, code blocks, safe external links and simple tables. A small set of inline styles is read only when it clearly represents bold, italic or strikethrough text. It is not copied as styling.
Simple tables remain tables. A cell that spans rows or columns is flattened into a rectangular grid because Markdown and receiving editors do not share one merged-cell contract. Pasteful reports that change. Remote images are not fetched, embedded pages and forms are removed, and unsafe or credential-bearing links become visible text. The preview therefore accounts for what will travel instead of hiding a lossy conversion behind a polished render.
Formatting compatibility and limitations · Pasteful's reproducible formatting corpus
Choose the output for the destination
Microsoft Word, Google Docs and Notion are all rich editors, but they make their own paste decisions. Pasteful prepares semantic HTML plus a readable plain-text fallback when the browser supports rich clipboard writes. The receiving app chooses what it accepts and applies its own document styles. That is why the useful promise is clean structure, not a pixel-perfect copy of the webpage.
For Word, copy rich text into a document that is already open or download an editable DOCX when the content should become a separate file. For Google Docs and Notion, copy the rich result and inspect heading levels, list depth, links and table cells in the target page. For a repository, static-site generator or plain-text notes app, choose Markdown and download the .md file instead.
| If you need to | Use this route |
|---|---|
| An open Microsoft Word document | Copy semantic rich text so the content adopts the document's styles. |
| A new editable Word file | Download DOCX and inspect headings, lists and tables in Word. |
| Google Docs or Notion | Copy rich text, paste into the target page and check the resulting blocks. |
| A local text file or code repository | Convert to Markdown and download the UTF-8 .md file. |
Copy rich text into Google Docs · Copy rich text into Notion · Microsoft's HTML clipboard format
Security is part of formatting
HTML is an active web format. A converter should not insert it into the current page and hope for the best. Pasteful parses the source as a document, limits its size, depth and node count, and permits only the semantic structures it knows how to represent. Scripts, stylesheets, forms, frames, embedded objects and controls are not executed or carried into the result.
Links require an explicit HTTP, HTTPS or mail address. Relative URLs are not guessed because their destination depends on the original website. URLs containing credentials are removed. Images are never fetched during conversion, which prevents a source file from quietly contacting a remote server. When a useful representation cannot be produced safely, the converter falls back to visible text or stops with a complexity warning.
Check the paste in under a minute
Use a short fixture before trusting an unfamiliar destination. Include one heading, bold and italic text, a labelled link, a two-level list, a quote, a code sample and a two-column table. Compare the Pasteful preview with the result inside the real editor. This checks the destination rather than assuming every browser and app version makes the same clipboard choice.
Confirm that headings are recognised as headings, lists continue correctly when you press Enter, links open the intended address and table values remain in the right cells. Read every item in the loss report. The conversion is complete only when the destination contains the right meaning, not merely when it looks tidy in the preview.
- Check the first and deepest list item
- Open the first and last link
- Confirm code is separate from prose
- Read tables across rows and down columns
- Keep the source until the receiving document is verified