Pseudolocalisation test values
The standard internationalisation-testing technique, reduced to values you can paste into a field. 10 values, each with what it breaks and what a correct application does instead.
Type them one at a time with AltShiftN
in the palette, or take them all into a script with nkb emit pseudolocalization.
Pseudolocalised text in brackets
pseudolocalization/pseudo-bracketed
[!!!Ŕéšöûŕčéš!!!]
[!!! Ŕéšöûŕčéš !!!]
- 19 graphemes
- 19 code points
- 28 bytes
- 19 UTF-16 units
- What it breaks
- The brackets reveal truncation instantly: if either is missing on screen, the text was cut.
- What a correct application does
- Both brackets visible wherever the value is displayed.
Accented pseudolocalised word
pseudolocalization/pseudo-accented
Ŕéšöûŕčéš
Ŕéšöûŕčéš
- 9 graphemes
- 9 code points
- 18 bytes
- 9 UTF-16 units
- What it breaks
- Still readable as English, so a tester spots it in a screen where a string was never translated.
- What a correct application does
- Displayed with every accent intact, in every layer.
Pseudolocalised text, 30 per cent longer
pseudolocalization/pseudo-expanded-30
[!!!Ŕéšöûŕčéšééééé!!!]
[!!! Ŕéšöûŕčéš ééééé !!!]
- 25 graphemes
- 25 code points
- 39 bytes
- 25 UTF-16 units
- What it breaks
- Thirty per cent longer, the typical growth when English is translated into German.
- What a correct application does
- The layout holds without clipping or overlapping.
Pseudolocalised text, twice as long
pseudolocalization/pseudo-expanded-100
[!!!Ŕéšöûŕčéšéééééééééééééééééé!!!]
[!!! Ŕéšöûŕčéš éééééééééééééééééé !!!]
- 38 graphemes
- 38 code points
- 65 bytes
- 38 UTF-16 units
- What it breaks
- Twice the length, for labels and buttons sized to the English word.
- What a correct application does
- Wrapped or resized rather than clipped.
Very short pseudolocalised text
pseudolocalization/pseudo-short
[!Ŕ!]
[!Ŕ!]
- 5 graphemes
- 5 code points
- 6 bytes
- 5 UTF-16 units
- What it breaks
- The opposite case: a very short translation leaves a layout built around a long one looking broken.
- What a correct application does
- The layout holds when the value is short.
Pseudolocalised text with direction marks
pseudolocalization/pseudo-rtl
[!!!Ŕéšöûŕčéš!!!]
[!!! \u200FŔéšöûŕčéš\u200F !!!]
- 2 characters that change the direction of text
- 21 graphemes
- 21 code points
- 34 bytes
- 21 UTF-16 units
- What it breaks
- Pseudolocalisation with direction marks, for checking mirrored layouts without knowing a right-to-left language.
- What a correct application does
- The interface mirrors correctly and the brackets stay in the right places.
Number inside pseudolocalised text
pseudolocalization/pseudo-numbers
[!!!1234,56!!!]
[!!! 1234,56 !!!]
- 17 graphemes
- 17 code points
- 17 bytes
- 17 UTF-16 units
- What it breaks
- Numbers inside a pseudolocalised string reveal formatting that was hard-coded next to the text.
- What a correct application does
- Numbers formatted by locale, not concatenated.
Placeholder inside pseudolocalised text
pseudolocalization/pseudo-with-placeholder
[!!!Ŕéšöûŕčéš{0}!!!]
[!!! Ŕéšöûŕčéš {0} !!!]
- 23 graphemes
- 23 code points
- 32 bytes
- 23 UTF-16 units
- What it breaks
- A placeholder that must survive translation; losing it produces a message with a hole in it.
- What a correct application does
- The placeholder preserved and filled.
Markup inside pseudolocalised text
pseudolocalization/pseudo-html
[!!!<b>Ŕéšöûŕčéš</b>!!!]
[!!! <b>Ŕéšöûŕčéš</b> !!!]
- 26 graphemes
- 26 code points
- 35 bytes
- 26 UTF-16 units
- What it breaks
- Markup inside a translated string is either escaped or rendered, and both are wrong in different places.
- What a correct application does
- Handled the same way as any other user-provided markup in that field.
Three scripts in one string
pseudolocalization/pseudo-mixed-scripts
[!!!Ŕéšöûŕčéš中文العربية!!!]
[!!! Ŕéšöûŕčéš 中文 العربية !!!]
- 30 graphemes
- 30 code points
- 50 bytes
- 30 UTF-16 units
- What it breaks
- Three scripts in one string: font fallback, line breaking and direction all at once.
- What a correct application does
- Every script rendered, with no missing-glyph boxes.