geteilt von: https://lemmy.world/post/1004415

cross-posted from: https://lemmy.world/post/963301

I shared bits and pieces of this before, but it’s officially up and running now: https://www.search-lemmy.com/

This is an enhanced search engine for Lemmy. With a few primary goals:

  • You can choose a preferred instance. After choosing what your primary instance is, and performing a search ALL links will open in that instance.
  • This aims to be a replacement for using site:reddit.com in Google, but just for the fediverse.
  • You can filter the search results by:
    • Instance – This will filter the results to only show communities that belong to a particular instance. Just type something like instance:lemmy.wrold or instance:https://lemmy.world/. This is separate from your preferred instance, such that you can search for posts on lemmy.world while still opening them on lemmy.ml.
    • Community – You can refine the search by a specific community. You use the same syntax that you’d use here community:!fediverse@lemmy.world.
    • Author – Similar to the above you can also filter by a specific author such as: author:@marsara9@lemmy.world.
  • The entire thing is open-source. You can view the code and even host your own instance… See more details here: https://github.com/marsara9/lemmy-search.

NOTE: This only supports Lemmy instances for now. Other fediverse type instances may be in the future depending on how this works out.

I’ve been working on this over just the last few weeks, so it hasn’t had a chance to crawl much of the fediverse yet. For now it only supports lemmy.world and lemmy.ml but other preferred-instances will come online as time goes by.

If anyone finds any bugs, and I’m sure you will, or if anyone has any suggestions PLEASE raise an issue on GitHub for me to track. Lastly, if anyone wants to help contribute please feel free to reach out.

NOTE TO SERVER ADMINS: You can prevent your site from being crawled by adding lemmy-search to your robots.txt for the user-agent.

  • tarjeezy
    link
    fedilink
    English
    arrow-up
    5
    ·
    1 year ago

    The home instance selector works as expected if you use the autocomplete, but it gets weird if you manually type the instance address and don’t include the https://, and the / at the end.

    It looks like the code is expecting those pieces to be there and truncating the first 8 and last 1 characters. So, if I only type “lemmy.world”, the home_instance URL parameter gets set to just “rl”, and no results are returned. I’m sure there are some premade functions out there that can extract the hostname from a URL to better handle this.

    • tarjeezy
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      By the way, neat project! It looks like it’s filling a pretty big need that people having been asking for. Keep up the good work!

    • marsara9@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      There’s still some bugs with this especially as I try and improve the UI. If anyone has any good JavaScript experience I could use the help.

      This use to be a combo box that you couldn’t type into but I’m exploring using a freeform box with auto complete, to make it easier to narrow the list down. …needless to say it’s not perfect yet. But if anyone has any ideas …

      • tarjeezy
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        It doesn’t seem like you actually need the https:// to parse the instance name, since the text xxxxxxxxlemmy.worldx still searches successfully. Maybe just exclude those parts off the URL and simply use lemmy.world, lemmy.ml, etc in the autocomplete list.

  • Tvkan@feddit.de
    link
    fedilink
    English
    arrow-up
    3
    ·
    1 year ago

    This aims to be a replacement for using site:reddit.com in Google, but just for the fediverse.

    Oh shit, da habe ich gar nicht drüber nachgedacht, site:reddit.com habe ich auch immer gerne genutzt. Ein direktes Lemmy-Analogon dazu wäre super, aber vermutlich(?) schwierig umzusetzen - tendenziell sind die Suchergebnisse von DuckDuckGo bzw. Google dann doch besser als irgendwelche selbstprogrammierten (z. B. die von reddit).