On Sun, Apr 4, 2021 at 1:38 PM Saq Imtiaz <saq.imt...@gmail.com> wrote:

> Hi Mohammad,


Hi Saq,

>
> So a lot of things are still up in the air. As I mentioned this is a quick
> prototype without a lot of pre-planning and I haven't had the chance to
> figure out a good workflow yet even for my specific needs. I'll try to
> answer your questions as well as I can at this point,
>


> but some things we may need to discuss and figure out together.
>

Sure!

>
>
>> One question: In Searchwikis (https://github.com/kookma/TW-Searchwikis)
>> it uses index tiddlers from target wikis to search them
>> The question is: Can I use Get Remote Tiddlers plugin to automatically
>> update the indeed tiddlers in my main wiki i.e to fetch latest changes?
>>
>
> Yes that should be possible as long as the wikis are on the same domain or
> served with CORS headers. The catch is that for every wiki, the code will
> be fetching the entire TW file to extract the tiddlers, which means more
> bandwidth used and also possibly some lag in the wiki depending on the
> number of wikis being fetched and the hardware. I want to experiment with
> using the same technique with an uglified TW with Flibbles' plugin, or a TW
> with the core in an external js file. Especially the second option should
> load considerably faster.
>

I did not know the whole wiki is fetched, then yes for many external wikis
or large wikis we will have some time lags! I use the Flibbles uglify, I am
very happy with that, but for this case still the empty.html is around
1.2MB, the external.js seems promising here!

>
> Another option that might make the most sense for Searchwikis is to
> generate and save the index as a separate file. This would need far less
> bandwidth to fetch and should overall be considerably faster. For wikis
> served by Github pages we could even try to implement an option to save the
> index file automatically each time the wiki is saved.
>

Great! Right now I have to open each target wiki separately, update the
index, export it, and import to the central wiki!

>
>
>> A more radical question is can " Get Remote Tiddlers " deprecate
>> searchwikis and lets users search other wikis from a central wiki?
>>
>
> Fetching tiddlers from a remote wiki already entails loading the entire
> wiki and extracting all its tiddlers and creating an array of tiddler
> titles. So yes, theoretically you could skip the part of Searchwikis that
> creates the index and just use Get Remote Tiddlers to create the index for
> the other wikis. However, the same caveats apply in terms of bandwidth
> usage and that with many wikis being fetched in the background it could
> slow down the wiki when that is happening. Using TW files with an external
> core.js file would be very helpful here.
>
> Having to rely on CORS headers being present is indeed a limitation but
> this overall approach (rather than the code I posted specifically) opens
> some interesting possibilities as well.
>
> If you're interested, there's a small discussion here about enabling CORS
> on TiddlyHost: https://github.com/simonbaird/tiddlyhost/issues/112
>

I will!

>
> I am also curious to see if TW on Fission opens up for similar workflows.
>
> Cheers,
> Saq
>


Thank you Saq!

> --
> You received this message because you are subscribed to the Google Groups
> "TiddlyWiki" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to tiddlywiki+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/tiddlywiki/a30adfd0-8d9f-42d8-aea6-e0824c777db6n%40googlegroups.com
> <https://groups.google.com/d/msgid/tiddlywiki/a30adfd0-8d9f-42d8-aea6-e0824c777db6n%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to tiddlywiki+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/tiddlywiki/CAAV1gMApi7STpRFofsx3Uq3inNk6B1R0u%3Dsbp%2BEqQa2xXurzJw%40mail.gmail.com.

Reply via email to