Keep Tamil and other non-Latin titles intact in your ebook library
A title in Tamil stays in Tamil. Readwyn stores the title and authors exactly as the book file gives them, in any script, and keeps a title you type in the edit details sheet exactly as you typed it. Converting a MOBI, an AZW3 or an EPUB 2 to EPUB 3 doesn't change a single character of its text, and the library shows a Tamil title in full wherever it fits.
Editing the details yourself
“Edit details” on the book screen opens the edit details sheet. Nothing is stored until you save. Until then, each changed field shows “Was” with the old value struck through, so you can check the old title against the new one.
The Language field sets the book's main language. Only English books are hyphenated in the reader, so in a book set to Tamil the lines break between words.
Long titles
In the list view, a Tamil title that fits is shown whole, and a longer one ends in “…”. Readwyn measures the line as it's actually drawn, and it cuts only between whole letters, so a vowel sign never comes away from its consonant.
When a book has no cover image, Readwyn draws a plain cover with the title on it, set at 17 px. If one word is too wide for the cover, the whole title gets smaller, down to 11 px, until that word fits. A Tamil title then wraps at a space rather than inside a word, unless one word is too long even at 11 px.
Conversion keeps every character
The conversion to EPUB 3 is tested on a Tamil book, one of 10 in the round-trip tests. It comes from archive.org, because Project Gutenberg has no Tamil books. Other tools write each book out as MOBI 6, as AZW3 and as a combo file, and Readwyn converts those back to EPUB 3. The text has to match the source, apart from a few known changes that the writing tools make, such as an added contents page. One of them touches Tamil: a tool stores some vowel signs in composed form, so ெ followed by ௗ comes back as ௌ.
Highlights in Tamil books handle both forms too. When Readwyn has to find a highlight again by its text, it looks for the text as the book has it and in composed form.
Search and word counts in other scripts
Search ignores case in any script that has it. It compares lowercase forms, so Greek “λόγος” finds “ΛΌΓΟΣ”, and the final sigma “ς” counts as “σ”, in the library search and in Find in the book (⌘F in the reader) alike. In Find in the book, accents have to match: “ville” won't find “villé”.
A Tamil book is counted at its spaces, like an English one, and in Chinese or Japanese text each ideograph counts as one word.