mcdonnel dictionary - deal with IAST in entries better.

[vvasuki]

Entries have IAST in them -

cs. uddhāraya, cause to be extracted (arrow) by (in.); extricate, rescue, deliver, from (ab.); appropriate; hold up; des. wish to relieve or assist. abhi_ud, extricate, from (ab.); take out, draw (water); collect (gold.)

हरि hár-i, a. [√ 3. hṛ, be yellow] tawny, yellow (esp. of horses); greenish (rare, C.); m. (C.) steed (sp. of Indra); lion; monkey; N. of Indra and (more commonly) of Viṣṇu-kṛṣṇa; N. of a various men; (hári)- keśa, a. tawny-haried; -gaṇa, m. troop of horses; N.; -ghoṣa, m. N.; -candana, m. (?) Indra's sandal tree (one of the five trees in Indra's paradise); kind of sandal tree; n. yellow sandal; -candra, m. N.; -cāpa, m. n. Indra's bow, rainbow.

Now, because most of our dictionaries lack IAST headwords (IAST is unpopular among hindu users), users cannot easily long press or click on a word in the entry to go look up the word in all the dicts. This can be overcome by:

• replacing IAST with devanAgarI everywhere. (we did this in some dictionaries, probably MW is an example: हरि m. of various authors and scholars ( esp. of the poet भर्तृ-हरिas the author of the वाक्य-पदीय; also with मिश्र, भट्ट, दीक्षितetc. ) Cat.)
• Adding links directing devanAgarI lookups.

I think that the first approach above is better. I remember that it took me quite a while to get around to accepting and learning iast.

[drdhaval2785]

The difficult story with MD is that it has a lot of french words. They also have diacritics just like IAST. So converting them without affecting French is very difficult, if not impossible.

[vvasuki]

Ah I see. I myself did not find a french word while glancing around - so, even if there are many french words, they might be relatively few in number. Distinguishing french vs devanAgarI might be partially possible by checking against a list of all sanskrit words (easily constructed) while writing the program - but things will get missed anyway.

In any case, there would be no harm in linking them all for lookup.

[damooo]

for same issue with other dicts, I tried to keep devanagari names in vedic index, and created a test version here, https://github.com/sanskrit-coders/stardict-sanskrit/tree/master/sa-kAvya/vedic_indexvedic_index_customized_c221 (wrong name, as uploading from browser, just renaming cannot done through browser it seems) . here only author names like max muller etc are in IAST other than devanagari. for present i converted them too, but soon will filter out them, once plz check remaining and suggest improvements or inform faulty things?

[vvasuki]

Thanks for the update, dAmodara.

How exactly have you updated vedic_index? "vedic_index updated" wasn't clear - you reduced the styling further? @drdhaval2785 since you're a colordict user, could you see if the new version look fine?

Regarding your test dict- Renamed it as needed. A simple tip for omitting the mlecCha name conversion - they tend to appear in footnotes - so do conversion only in the body, not footnotes. Another trick would be to look if the target string contains letters like ä, which aren't in IAST.

[drdhaval2785]

@vvasuki It doesn't seem to be updated in updater app.

[vvasuki]

@vvasuki It doesn't seem to be updated in updater app.

That was a bug - fixed now..