Guide
Preeti vs Unicode
If a Nepali document opens as strings like g]kfn, nothing is corrupted. You are looking at a font problem that has a permanent fix.
On this page
What you are actually seeing
Preeti is a font, not an encoding. It was built before Unicode supported Devanagari properly, and it works by drawing Nepali letter shapes on top of Latin code points.
When a Preeti document stores न, the file does not contain न. It contains the Latin letter g. Only the Preeti font makes it look Nepali. Open the same file anywhere the font is missing and you get the Latin letters back, because that is what was there all along.
Why this matters beyond appearance
Installing Preeti makes the text readable again, and that is where most people stop. It is the wrong fix, because the underlying data is still Latin:
- Search does not work. Searching for नेपाल in a Preeti file finds nothing, because the file contains no न.
- Copy and paste breaks. Paste into email or a browser and it reverts to gibberish.
- Phones cannot display it. You cannot install Preeti on most mobile devices.
- Sorting is meaningless. Names sort by their Latin stand-ins, not Nepali alphabetical order.
- Screen readers fail entirely.
The permanent fix
Convert once, to Unicode, using the Preeti to Unicode converter. Then set the font to any Unicode Devanagari face — Kalimati, Mangal and Noto Sans Devanagari are all free and widely available.
After conversion the file contains real Nepali characters. Search, sorting, copy-paste, mobile display and screen readers all start working, and they keep working regardless of which fonts are installed.
The part most converters get wrong
Swapping characters one for one is not enough. Preeti stores text in the order the shapes are drawn on the page. Unicode stores it in the order the letters are spoken. Two marks sit in different places as a result:
- The i-kar (ि) is drawn to the left of its consonant, so Preeti stores it first. Unicode stores it after the whole cluster.
- The reph (र्) is drawn above the cluster it belongs to, so Preeti stores it last. Unicode stores it first.
A converter that ignores this returns text that renders correctly in no font at all — which is why some online converters produce output that looks almost right but is subtly broken.
Preeti is not the only legacy font
Kantipur, Sagarmatha, Himali and several others use the same trick with different mappings. Text typed in Kantipur will not convert correctly with a Preeti converter, and the result is usually a mix of correct and wrong letters.
Before converting, check the font name in the original document. In Word, click into the Nepali text and read the font box.
Frequently asked questions
Should I still use Preeti for anything?
Only when the destination requires it — some print shops and layout templates do. For anything else, Unicode is strictly better.
Will converting change how my document looks?
The letters will be identical. Line breaks and spacing may shift slightly, because Unicode fonts have different metrics than Preeti. Check page breaks after converting a long document.
Can I convert a scanned PDF?
No. A scan is an image, so there is no text to convert. That needs OCR, which is a different process.