Collapse Spaces: Cleaning Gaps from Pasted Text

When you actually need it

Web page text is the usual source: multiple spaces were inserted in the markup for alignment and survive the copy. PDF extraction leaves irregular gaps between words, so a paragraph looks loose and the line length jumps around.

Columns copied from a spreadsheet, chat logs, and OCR output carry the same clutter. Collapsing it before pasting into an email, a form, or a document makes the layout predictable and stops lookups from failing over a difference in space count.

• Runs of spaces after copying from a page or PDF
• Irregular word gaps in OCR output
• Tidying text before pasting into email or a form

What it changes and what it ignores

Runs of spaces and tabs become a single space, and nothing else is touched. Leading indentation is compressed to one space rather than removed, so a small gap remains on the left; run a trim afterwards if you want a hard left edge.

Several kinds of whitespace are left alone. Line breaks survive, so the line structure is unchanged. The ideographic space U+3000 is not a half-width space and is not collapsed, which is why wide gaps in Chinese text look untouched, and the non-breaking space U+00A0 stays as well.

• Space and tab runs become one space
• Leading indentation becomes a single space
• U+3000 and U+00A0 are not collapsed

Alternatives and when to avoid it

Do not run it on code. Indentation compressed to one space breaks Python immediately and ruins readability elsewhere, and anything aligned with spaces loses its layout. Two trailing spaces in Markdown mean a hard line break, which also disappears.

To remove only trailing spaces, a regex replace in the editor is more precise. To clear full-width spaces too, convert them to half-width first and then collapse. When paragraph structure matters, delete empty lines and collapse spaces as separate steps.