Remove extra spaces from text

Spaces you cannot see

Text gathers spaces from everywhere it has been:

• Double spaces after a full stop, a habit from typewriters. Chicago and APA style both ask for one.
• Tabs from documents and spreadsheets, which look like a few spaces but are one character.
• Non-breaking spaces from web pages, Word and PDFs. They look exactly like a space but stop a line from breaking there, and a search for a normal space does not find them.
• Trailing spaces at the ends of lines, invisible until something compares two lines.
• Zero-width characters, which take no room at all. Clean does not catch these; find them with Pattern on and [\u200B\u200C\u200D\uFEFF], and replace them with nothing.

Collapse extra spaces handles the first four in one go: every run of spaces, tabs and non-breaking spaces becomes a single ordinary space, and spaces left at the end of a line are removed.

Trim or collapse?

The two main choices in Clean do different jobs:

• Trim spaces at the ends of lines removes spaces and tabs, non-breaking ones included, before and after the text on every line, and leaves the spaces between words alone. That includes indents at the start, so it would flatten code, poetry or a nested list.
• Collapse extra spaces works anywhere in a line, but only shrinks: an indent of eight spaces becomes one, not none. It turns every tab into a space as well, which merges the columns of anything copied from a spreadsheet.

Run both, Collapse first, for clean single-spaced lines with no indents. When part of the text must keep its layout, such as a table or a code sample, select only the rest before you choose. Empty lines are a separate matter: Remove empty lines is in the Lines menu.

More in the Clean menu

• Straighten quotes: “curly” and ‘single’ quotes become straight " and ' marks, which code, CSV files and many forms expect. The low opening quotes used in German, „ and ‚, are straightened too.
• Remove accents: café becomes cafe and naïve becomes naive. Letters that are not an accent on a base letter, such as ø, ł and ß, stay as they are; the slug generator spells those out.
• Remove HTML tags: deletes everything from each < to the next > and keeps the text between tags. Entities such as &amp; stay as written, and so does the code inside script and style blocks. A lone < in ordinary text can make it remove more than a tag. To turn entities back into characters, use the HTML entities tool in the Dev Toolbox.
• Tabs to spaces: each tab becomes four spaces.

Why the spaces matter

Extra spaces are more than untidy. A name with a trailing space does not match the same name without one, so a spreadsheet lookup fails and a duplicate slips through. A non-breaking space between two words hides them from a search. Double spaces inflate a character count that has to stay under a limit, and in code, trailing whitespace shows up as noise in every change.

Clean first and the rest of the work goes better: remove duplicate lines, sort or compare the result, and lines that differed only by a space now match.

To see the effect, watch the character count in the bottom bar: it drops by exactly the number of spaces removed. Typing two spaces into Find, with Pattern off, shows how many double spaces are left anywhere in the text.

One kind of stray space survives all of this: a space before a comma or a full stop, as in “word ,”. Find a space followed by a comma, replace it with a comma alone, and do the same for full stops.