A good web frontend to Cologne and Mobile apps

[drdhaval2785]

Dear friends, the following announcement may be of immense importance to us. I have checked iOS app and it works just fine. Good news us that it uses data from csl-orig repository and can have updated data via a pipeline periodically. Give it a try and let us know your feedback here.

QUOTE

नमः सदसे !

I'd like to introduce this group to a recently launched website and app: https://sanskritkosha.com. It provides a user-friendly interface for convenient access to over 50 koshas at your fingertips.

Besides using it as a standalone website, you can also install the app on both Android and iOS by following the instructions at https://sanskritkosha.com/mobile_apps.

The kosha data is sourced from various GitHub repositories maintained by scholars worldwide. I express my sincere gratitude to these invaluable contributors who played a crucial role in curating these dictionaries.

This app shares the same framework as ashtadhyadyi.com, ensuring that the navigation should already feel familiar and comfortable.

युगपत्सर्वकोषाणां यथावद्योजनं कृतम् । समेषामुपकाराय भवेदत्र न संशयः ॥

Feel free to give it a go!

इति सादरं नीलेशः

QUOTE

[funderburkjim]

First thoughts:

• Amazingly responsive
• Multi-dictionary
  • What is the relation between Cologne data and the data displayed at Sanskrit-kosha? Similarly question for the other sources
• Various headword spellings are accepted
  • gaNgA, gaMgA (fewer dicts) गङ्गा , Gaṅga, gaṅga
  • अश्व, अश्वः (fewer dicts), aSva (slp1), azva (hk), áśva (None - accent?) , aśva (many)
• relation between headword spellling and dictionaries are not obvious. Maybe should be examined further.
• Should there be links on the Cologne homepage to various of these nice applications that @drdhaval2785 has mentioned recently? And maybe some others (e.g. from Huet).

how did they do that?

• hosted at namecheap.com ? How funded?
• Who is the development team?
• view-page-source obscure.
• No idea how developed. Is it based on react - is it node-js on the server?
• https://github.com/sanskritkosha-com/ seems to be the home of the data
  • except for the data repository, this seems to be a PRIVATE repository
• Why is the code private? This privacy bothers me. Maybe I should just enjoy the fruits and not worry about the hidden parts.

[drdhaval2785]

@neeleshb may be able to answer these technical questions. As far as I understand, there is no private data directory. The website and app runs some pipeline which takes data from the current sanskrit-lexicon/csl-orig/, sanskrit-kosha/kosha and indic-dict/stardict-sanskrit repositories and presents them in a user friendly way. I am nog sure what is the frequency of fetching updated data from github repos.

[neeleshb]

Answers to the questions

• What is the relation between Cologne data and the data displayed at Sanskrit-kosha?

The app pulls the data from 3 different github repos (see https://sanskritkosha.com/about for details). Specifically talking about the Cologne data, most of the files in the csl-orig/ are being processed.

• relation between headword spellling and dictionaries are not obvious. Maybe should be examined further.

The headwords present in the source files are taken. The app does not create many "variations" at the moment. This is easy to implement, however.

• hosted at namecheap.com ? How funded?

Domain and hosting are on namecheap. Funded personally by me.

• Who is the development team?

Myself (neelesh.bodas@gmail.com).

• view-page-source obscure.

Oh that's because the CSS and JS are obfuscated to reduce the size.

• No idea how developed. Is it based on react - is it node-js on the server?

It's a "Progressive Web App" (PWA) . Plain vanilla HTML-CSS-JS with a few libraries like JQuery and Bootstrap.

• https://github.com/sanskritkosha-com/ seems to be the home of the data

Correct. All the data is hosted here.

• except for the data repository, this seems to be a PRIVATE repository

No, everything is public. All data is hosted on github.

• Why is the code private? This privacy bothers me. Maybe I should just enjoy the fruits and not worry about the hidden parts.

LOL :) What is not published is just the HTML/CSS/JS code that renders the UI. I would assume that's not a big issue? As soon as code is made public I will get numerous pull requests for features which I don't want to clutter the app with ! And another reason to not open up the source is that it shares a lot of code with Ashtadhyayi.com (Another of my website) which is too popular to be touched !!

[neeleshb]

The website and app runs some pipeline which takes data from the current sanskrit-lexicon/csl-orig/, sanskrit-kosha/kosha and indic-dict/stardict-sanskrit repositories and presents them in a user friendly way.

Yes, that's exactly what the website does.

I am not sure what is the frequency of fetching updated data from github repos.

As of now I fetched it just once a few days ago. We can fetch it as frequently as we want. I am thinking once a month or so.

[mediabuff]

thanks for the details. Could you please describe the technical architecture a little bit? do you use search engine ? how is it indexed ? are the headwords normalized across dictionaries ? stem form or nominative singular form. I see bunch of json files actually hosted in https://github.com/ashtadhyayi-com/data/tree/master/kosha