Pasting a short paragraph into a converter is easy. Converting a 40-page Word file with tables, headings, bold text, page numbers and a few English words is a different job. If you copy everything out, convert it and paste it back, the formatting is usually the first casualty. This guide explains why, and how to do it properly.
Why plain copy-and-paste ruins the document
A Word file is not one long string of text. It is a set of runs: small chunks, each with its own font, size, bold or italic settings. A typical Kruti Dev document might contain:
- Hindi body text in Kruti Dev 010,
- English words, numbers and email addresses in Times New Roman or Calibri,
- headings in a larger Kruti Dev size,
- table cells, headers, footers and footnotes, each with their own runs.
If you select everything and convert it as one block, two things go wrong. The English parts get converted too (so “India” turns into Devanagari nonsense, because the converter assumes every letter is legacy text), and all the run-level formatting is flattened to whatever your paste setting keeps.
The approach that works: convert run by run
Only text that is set in a legacy font should be converted. Everything else should be left exactly alone. That is what the Word file converter is for: it looks at the font of each run, converts the runs set in a recognised legacy font, and leaves the rest of the file as it was, so tables, bold, page layout and English text stay put. The conversion runs in your browser, so the file is not uploaded.
Step 1: Find out what is in the file
Before converting, check what you’re dealing with.
- Open the document in Word.
- Press Ctrl+A to select everything and look at the font box on the Home tab. A name means one font is used throughout. A blank box means mixed fonts.
- Click inside a Hindi paragraph and read the font name. It should be Kruti Dev 010 (or a similar name). If it says something else, such as DevLys or Chanakya, use that converter instead. See the font detector if you are not sure.
- Check headers, footers, footnotes and text boxes. These are separate stories in the file and are easy to miss.
If the file is an old .doc, open it in Word and use File → Save As → Word Document (.docx) first.
Step 2: Convert the file
- Open the Word file converter and choose your .docx.
- Leave the source set to automatic detection unless you know the exact font. If you do, pick Kruti Dev.
- Download the converted .docx.
Step 3: Check the result
Open the converted file next to the original and check:
- Short-i matra and reph. Look at words such as हिन्दी, कार्य, निर्णय and धर्म. These are where bad converters fail, because the signs are stored in a different order in a legacy file.
- Headings and numbering. Make sure heading levels, numbered lists and bullets are intact.
- Tables. Column widths and borders should be as before, with Hindi text in the cells.
- Mixed English. Names, numbers, dates and URLs should read as before.
- Headers, footers and footnotes.
Step 4: Change the font in the styles
After conversion, the text uses a Unicode font, but the document’s styles may still point to Kruti Dev for new text. Open Home → Styles, right-click Normal and choose Modify, then set the font to Mangal, Nirmala UI or another Unicode Devanagari font. Do the same for any heading styles. This stops the old font from creeping back when someone adds a new paragraph.
What can still go wrong
- Text inside images and scanned PDFs. If a page is a picture of text, there is nothing to convert. Use OCR first.
- Text in text boxes and WordArt. These are sometimes missed by tools that only read the main document. Check them by hand.
- Mixed legacy fonts. A file typed partly in Kruti Dev and partly in DevLys needs each part to be converted with the right profile.
- Smart quotes. In Kruti Dev, the apostrophe and double-quote keys type श् and ष् but Word’s AutoCorrect sometimes turns them into curly quotes. The converter handles both forms.
- Characters Unicode has but the font did not, or the opposite. If you convert back later, read the loss report.
If you only need a paragraph or two
For short passages, the quick route is fine: select the Hindi text in Word, copy, paste into Kruti Dev to Unicode, and paste the result back with Paste Special → Keep Text Only on a line that is already set in a Unicode font such as Mangal. You will lose character formatting inside that passage (bold, italics), so reapply it by hand.
Going the other way
If a printer wants the document in Kruti Dev, don’t edit the converted .docx by hand. Convert the Unicode text with Unicode to Kruti Dev, set the font to Kruti Dev 010, and export a PDF with fonts embedded so the press does not need the font installed.
A checklist you can keep
- Save a copy of the original. Never convert your only copy.
- Identify the font(s) in use.
- Convert only the legacy-font runs.
- Spot-check reordered words, tables, headers and footers.
- Change the document styles to a Unicode font.
- Keep the original file until someone else has opened and approved the converted one.