Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    damarsuyitno
    @damarsuyitno
    Okay thanks, nope its old product @JVankat
    Olesya
    @lesyaelf_gitlab
    @damarsuyitno I've replied by email. In short, this page isn't linked from anywhere else on your site. If you re-index it manually, the sku search will work. In my email, I've also described a couple of options to make sure all your products can get indexed.
    Golf Empire
    @opengolfcomps_twitter
    Hi @JVankat - I'm not sure if this is on your road map for the future but it would be great if I could use SS360 to be able to do a radius search. For example the user will input their location and results would only be shown with x kilometers (they could select their maximum distance). Each record would obviously have the longitute & latitude .
    Jaroslav Vaňkát
    @JVankat
    Hey @opengolfcomps_twitter - we've recently discussed location based search, and decided not to put it on our roadmap just yet. But I'll make sure to let you know if we'd decide to start working on this anytime soon.
    Golf Empire
    @opengolfcomps_twitter

    Hey @opengolfcomps_twitter - we've recently discussed location based search, and decided not to put it on our roadmap just yet. But I'll make sure to let you know if we'd decide to start working on this anytime soon.

    OK, & thanks for the quick reply!

    Dimitri KOPRIWA
    @kopax
    hello
    Jaroslav Vaňkát
    @JVankat
    Hey @kopax
    Jannette Pazer
    @sociallyclimb_twitter
    Hi @JVankat! Question - how does search choose which image to use? We have 2 images on a page, first one is a photo, second one is a Facebook logo. Don't know why search results displays the FB logo instead of the 1st image. (There is no official featured image on this CMS). Any way we can change image for this particular page whenever it shows up in site search results?
    Jaroslav Vaňkát
    @JVankat
    Hi @sociallyclimb_twitter - we try to detect the image if no XPath is set up, but that doesn't always work perfectly. You can change the Image XPath on the Crawler Settings page and re-index. Just let me know if you need any help setting this up (I'd need your siteId and a sample page from your site).
    Jannette Pazer
    @sociallyclimb_twitter
    Hi, not sure if that will work - it's only for 1 page. Here's the page, and I was able to see it in search results by searching for "accepted students". https://www.ciachef.edu/accepted-students/. Siteid: www.ciachef.edu_1
    (note: currently reindexing - so may need to wait a bit)
    Jaroslav Vaňkát
    @JVankat
    @sociallyclimb_twitter okay, since it's only this one page, let me check it tomorrow (it's midnight in our timezone) so I can make sure the XPath doesn't break any other pages. I'll get back to you once I've figured it out.
    Jaroslav Vaňkát
    @JVankat
    @sociallyclimb_twitter okay, in the end I've added an exclude content xpath for you, so the facebook logo is extracted. I've reindexed this page and the image is now a different one.
    Jannette Pazer
    @sociallyclimb_twitter
    Wow, excellent! Thank you! So what you did was create an xpath to exclude any facebook logos?
    So in general, there's no definitive pattern to which image gets used in results? Curious if there's a pattern we need to be aware of.
    Jaroslav Vaňkát
    @JVankat
    @sociallyclimb_twitter exactly, I've added a content exclude XPath so the logo would be ignored with our image detection. I think you could specify the following Image XPath: //div[contains(@class, 'm-callout')]//img/@src - that one seems to be the one we do usually detect, but I've checked just a few pages from your index, so can't tell you whether this would work globally.
    Jannette Pazer
    @sociallyclimb_twitter
    Don't think so, most of our images aren't in m-callout, just happens to be the case on some pages. Just don't know why it wouldn't have taken the 1st image instead of the 2nd image on the page.
    Jaroslav Vaňkát
    @JVankat
    @sociallyclimb_twitter tbh I'm not sure, I'll try to debug this later today.
    Jaroslav Vaňkát
    @JVankat
    @sociallyclimb_twitter okay, I did debug this and the order on the page isn't persisted when trying to detect the image. That's unfortunately a behavior we cannot update immediately, because that might change the image extraction for other customers, but I'm creating a ticket and we'll try to look into it within the next few weeks.
    Jannette Pazer
    @sociallyclimb_twitter
    Thanks so much for all of your help!
    adwhite75
    @adwhite75
    Anyone know why the magnifying glass next to the search bar still shows on our site despite me trying to hide it using the Config Designer?
    Jaroslav Vaňkát
    @JVankat
    Hi @adwhite75 - can you send me a link to your site? I will take a look.
    adwhite75
    @adwhite75
    You'll notice it also looks funky on Mobile too
    Jaroslav Vaňkát
    @JVankat
    @adwhite75 I see you've also sent an email to our support, the support team is looking into it at the moment and should get back to you soon.
    adwhite75
    @adwhite75
    Thanks! One question - to boost newer pages in the search results it says use the timestamp "Another use case: to boost the pages that are newer, use a timestamp or the year as a data point." - what format should that be in? I dont see an example
    Jaroslav Vaňkát
    @JVankat
    @adwhite75 (sorry for the delay, forgot to hit enter...) - you should be able to use the full date - e.g. the //time[contains(@class, 'date-container')] XPath should do it in your case (based on https://unofficialsf.com/validation-checker-flow-action/) - we should automatically extract the timestamp. You'll need to re-index after changing this.
    Dalton
    @Daltonloll_twitter
    Recently I was told to reindex my website by SS360 which I did, this caused ~1k pages to disappear from the search results. Now when I've tried to recrawl those pages they won't index, and when I try to manually submit them 1 at a time I get told it's configured to be noindex, or is a cloaked link.
    Neither of these are true, and nothing is being configured to be noindex within SS60.... what's going on? Why is it telling me my pages are configured in a way they are not?
    Dalton
    @Daltonloll_twitter
    And why did this reindex disable cross-site searching?
    Jaroslav Vaňkát
    @JVankat
    @Daltonloll_twitter can you please send me your siteId? I will take a look.
    5 replies
    Jannette Pazer
    @sociallyclimb_twitter
    Hi @JVankat! Question. We're seeing lots of duplicate results because of URL parameters. But some of those parameters identify the page (the website is a CRM front-end) and others do not. Is there a way to ignore only certain parameters, and/or de-dup the results?
    To see it, go to https://www.ciaalumninetwork.com/ and search on "bourdain" for an example.
    Jaroslav Vaňkát
    @JVankat

    Hi @sociallyclimb_twitter we'd usually recommend having a sitemap, or using canonical meta tags - but I suppose this is not super easy to implement on your side.

    Url replacements might be an option, e.g. replacing &crid=(\d)+ with an empty string.

    Another option might be blacklisting by a certain url params, but in that case some pages might get removed from the index (e.g. if you blacklist by an url param, but the url without the param isn't linked anywhere on your site).

    Jannette Pazer
    @sociallyclimb_twitter
    No sitemap, no canonical tags. I tried a URL replacement in crawler settings. Didn't work, but maybe it's my regex. I don't think blacklisting will work, as the extraneous URL parameters are quite varied, and I have no idea where they're linked from.
    Adjusted the regex in crawler settings, reindexing, see if that works.
    Jaroslav Vaňkát
    @JVankat
    Okay, understood, please let me know if that worked for you.
    Jannette Pazer
    @sociallyclimb_twitter
    @JVankat Unfortunately didn't work.
    Jaroslav Vaňkát
    @JVankat
    @sociallyclimb_twitter actually it seems to work, I just forgot to mention that this modification requires either a fresh re-index (emptying the index first), or removing the old urls from index via Remove all matching URLs after the re-index. --> now if I manually re-index the page https://www.ciaalumninetwork.com/s/898/17/interior.aspx?sid=898&gid=1&pgid=252&cid=3985&ecid=3985&crid=0&calpgid=61&calcid=3438 the crid query param is removed from the url of the indexed page.
    Jannette Pazer
    @sociallyclimb_twitter
    Hmmm, thanks, I'll look into it.
    Jannette Pazer
    @sociallyclimb_twitter
    @JVankat Well it's not perfect, but I excluded a bunch of dup-forming URL parameters in the "Remove all matching URLs". It's enough to remove most of the dups, and possibly will have to be revisited and updated frequently, but it's definitely better than before. Thanks for your quick response, as always.
    Jaroslav Vaňkát
    @JVankat
    @sociallyclimb_twitter glad this worked for you in the end!
    bbroekhuizen
    @bbroekhuizen

    Hello @JVankat , i have an issue where i don't return the activeFilterOptions from the API. Im sending these parameters with Axios:

    filterOptions: true
    filters: "[{"fid#1":["Jeugd"]}]"
    includeContentGroups: "["Programma's","Thema's","Podcasts","Over EO","EO NL"]"
    limit: 50
    limitPerGroup: true
    query: "testen"

    Am i missing something?

    In my response i get:
    activeFilterOptions: Array(1)
    0: {
    key: null
    max: null
    min: null
    }

    bbroekhuizen
    @bbroekhuizen
    Is there anyone else who can answer my question?
    Jaroslav Vaňkát
    @JVankat
    Hey @bbroekhuizen can you please also send me your siteId (or maybe the full request url)? I will take a look.
    bbroekhuizen
    @bbroekhuizen
    @JVankat Sure, one moment
    As you can see the program "Topdoks" with the applied filter, should also have the content property
    Jaroslav Vaňkát
    @JVankat
    @bbroekhuizen perfect, thanks for the requests, I'll get back to you in about 20 mins.
    bbroekhuizen
    @bbroekhuizen
    @JVankat Thanks!