@Masumrezarock100 proposed implementation:

• Create a helper function, to retrieve the list of all SUL wikis with an account attached
• Add an option to the settings page, in addition to "add a site", to "add all SUL sites"
• When selected, retrieve all SUL wikis, and set the list of sites to be equivalent to the list retrieved

I'm not sure what the "refresh the previous search results" refers to - this would be run each time the option was selected

Getting all SUL wikis with merged accounts:

new mw.Api().get( {
    action: 'query',
    meta: 'globaluserinfo',
    guiprop: 'merged',
} ).done( function( response ) { console.log( response ); } );

In T235453#5573787, @DannyS712 wrote:

I'm not sure what the "refresh the previous search results" refers to - this would be run each time the option was selected

I meant "clear the stored search results list, and request a new list using API". For users like me, and you who have unified their accounts all Wikimedia projects (have created local accounts on all Wikimedia sites), the list would be pretty big and it will take some time to retrieve the data from the server. So that is why I am saying it should store the search results in a personal JS file and retrieve the list from there instead of the server. And refresh it when the user tells the script to.

I wasn't going to store the result unless it was saved - it only took a few seconds to run via the console

@Masumrezarock100 for now, from the command line, you can use GlobalWatchlist.config.siteList = (await GlobalWatchlist.API('get', { action: 'query', meta: 'globaluserinfo', guiprop: 'merged' }, 'getAttachedAccounts')).query.globaluserinfo.merged.map((s) => s.url.replace(/https:\/\/|\.org/g, '')) and then hit the reset changes button

@Masumrezarock100 what about adding a parameter, allsites=, that, if set, results in (that time) going over all account sites, rather than the list (i.e. https://meta.wikimedia.org/wiki/Special:BlankPage/GlobalWatchlist?allsites=true allows overriding temporarily)

In T235453#5681027, @DannyS712 wrote:

@Masumrezarock100 what about adding a parameter, allsites=, that, if set, results in (that time) going over all account sites, rather than the list (i.e. https://meta.wikimedia.org/wiki/Special:BlankPage/GlobalWatchlist?allsites=true allows overriding temporarily)

Good idea!

In T235453#5681034, @Masumrezarock100 wrote:

In T235453#5681027, @DannyS712 wrote:

@Masumrezarock100 what about adding a parameter, allsites=, that, if set, results in (that time) going over all account sites, rather than the list (i.e. https://meta.wikimedia.org/wiki/Special:BlankPage/GlobalWatchlist?allsites=true allows overriding temporarily)

Good idea!

Yeah, except I just tried it, and it stalled around ln.wikipedia - I don't think this is going to be possible

In T235453#5681051, @DannyS712 wrote:

In T235453#5681034, @Masumrezarock100 wrote:

In T235453#5681027, @DannyS712 wrote:

@Masumrezarock100 what about adding a parameter, allsites=, that, if set, results in (that time) going over all account sites, rather than the list (i.e. https://meta.wikimedia.org/wiki/Special:BlankPage/GlobalWatchlist?allsites=true allows overriding temporarily)

Good idea!

Yeah, except I just tried it, and it stalled around ln.wikipedia - I don't think this is going to be possible

Does that mean if I try to add all sites manually, the script would not be able to handle it?

@Masumrezarock100 it'll work at first, but will likely stall out

In T235453#5682269, @DannyS712 wrote:

@Masumrezarock100 it'll work at first, but will likely stall out

Hmm. So API query times out. Would it be possible for your script to get Database access so that it can query watchlist entries directly?

In T235453#5682273, @Masumrezarock100 wrote:

In T235453#5682269, @DannyS712 wrote:

@Masumrezarock100 it'll work at first, but will likely stall out

Hmm. So API query times out. Would it be possible for your script to get Database access so that it can query watchlist entries directly?

Its not the API - I think its browser memory at that point. Scripts cannot query the databases

@DannyS712 Hey but you can do the alternative proposal, right? I mean adding an auto-complete feature?

Maybe - no idea how to make text autocomplete. As for toolforge - access to personal watchlists via grants might work, but its a very different direction for the script