Re: "scan for new items" always scans the entire collection

Diego <[email protected]> Thu, 9 Jul 2026 10:33:45 -0500
Newsgroups gmane.comp.kde.digikam.user
Message-ID <[email protected]>
This is a multi-part message in MIME format.
--------------dpxvyWSSYYBNSZc4XA5bE0nE
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Thanks, Gilles. I'm not sure is related. As I read it, it was resolved 
"not a bug" since it's behaving as expected for face scanning.

In our case (no faces involved), "scan for new items" should only scan 
new items; without this option checked, it should scan the entire 
collection. The problem reported here (for DK 9.1.0) is that regardless 
of whether you ask to "scan for new items" or not, it will scan the 
entire collection. This is not the expected behavior.

Best,

On 7/9/26 6:21 AM, Gilles Caulier wrote:
> Hi,
>
> Look at the comments in is report :
>
> https://bugs.kde.org/show_bug.cgi?id=522695
>
> I'm not sure if it's close to the problems listed here.
>
> Best regards
>
> Gilles Caulier
>
>
>
> Le jeu. 9 juil. 2026 à 12:41, Thomas <[email protected]> a écrit :
>
>
>     On 2026-07-09 07.33, VALÍK Robert wrote:
>     > Hey there!
>     >
>     > After upgrading Digikam to version 9.1, I have also noticed this
>     weird
>     > behavior.
>     >
>
>
>     Oh, I thought it was only me. I experience this too. Hopefully,
>     this can
>     be fixed.
>
>     A complete scan on my collection takes several days. So this has
>     effectively been preventing me from using DK for some time.
>
>     In fact I think a lot could be done to improve the whole experience
>     regarding scanning collection. For one, the progress reporting
>     could be
>     MUCH better. Currently, it has an almost useless progress bar. It
>     gives
>     a percentage but nothing in terms of how many pictures/folders it has
>     scanned and how many remaining. (I am aware that exact number of
>     remaining objects might be impossible to give, but then estimate
>     it like
>     other tools does, e.g. rsync will just update the number of remaining
>     items as it learns about them...). There is also no estimate of
>     how long
>     it will take or the processing rate (how many items per minute, etc.)
>     All these infos would be useful. Especially if trying to make it
>     faster
>     and see what works.
>
>     I also think the scanning could be sped up quite a bit by using
>     multiple
>     threads. I have my collection on a NAS. I believe a lot of the time
>     spent on scanning/indexing is really spent on network latency:
>     waiting
>     for the samba server to list files, open file handle, etc.
>
>     When I look at resources on the DK client machine, it is mostly
>     idle on
>     both CPU, network and disk I/O. Same with the NAS server, it is
>     mostly
>     idle on those resources as well. So I think a lot of time is pent
>     waiting for network packages. If I could allow 2, 4, 8, .., parallel
>     threads, I think we could see this wait time reduced significantly
>     and
>     CPU/net/disk utilization increased.
>
>
>     BR
>
>     Thomas
>
--------------dpxvyWSSYYBNSZc4XA5bE0nE
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<!DOCTYPE html>
<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <p>Thanks, Gilles. I'm not sure is related. As I read it, it was
      resolved "not a bug" since it's behaving as expected for face
      scanning.</p>
    <p>In our case (no faces involved), "scan for new items" should only
      scan new items; without this option checked, it should scan the
      entire collection. The problem reported here (for DK 9.1.0) is
      that regardless of whether you ask to "scan for new items" or not,
      it will scan the entire collection. This is not the expected
      behavior. </p>
    <p>Best,</p>
    <div class="moz-cite-prefix">On 7/9/26 6:21 AM, Gilles Caulier
      wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:CAHFG6sG7=3KwwkPEOPuEqbiFdkykuDGjxav1eSmXGkeQcAvYcQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div dir="ltr">
        <div>Hi,</div>
        <div><br>
        </div>
        <div>Look at the comments in is report :</div>
        <div><br>
        </div>
        <div><a href="https://bugs.kde.org/show_bug.cgi?id=522695"
            moz-do-not-send="true" class="moz-txt-link-freetext">https://bugs.kde.org/show_bug.cgi?id=522695</a></div>
        <div><br>
        </div>
        <div>I'm not sure if it's close to the problems listed here.</div>
        <div><br>
        </div>
        <div>Best regards</div>
        <div><br>
        </div>
        <div>Gilles Caulier</div>
        <div><br>
        </div>
        <div><br>
        </div>
      </div>
      <br>
      <div class="gmail_quote gmail_quote_container">
        <div dir="ltr" class="gmail_attr">Le jeu. 9 juil. 2026 à 12:41,
          Thomas &lt;<a href="mailto:[email protected]"
            moz-do-not-send="true" class="moz-txt-link-freetext">[email protected]</a>&gt;
          a écrit :<br>
        </div>
        <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><br>
          On 2026-07-09 07.33, VALÍK Robert wrote:<br>
          &gt; Hey there!<br>
          &gt;<br>
          &gt; After upgrading Digikam to version 9.1, I have also
          noticed this weird <br>
          &gt; behavior.<br>
          &gt;<br>
          <br>
          <br>
          Oh, I thought it was only me. I experience this too.
          Hopefully, this can <br>
          be fixed.<br>
          <br>
          A complete scan on my collection takes several days. So this
          has <br>
          effectively been preventing me from using DK for some time.<br>
          <br>
          In fact I think a lot could be done to improve the whole
          experience <br>
          regarding scanning collection. For one, the progress reporting
          could be <br>
          MUCH better. Currently, it has an almost useless progress bar.
          It gives <br>
          a percentage but nothing in terms of how many pictures/folders
          it has <br>
          scanned and how many remaining. (I am aware that exact number
          of <br>
          remaining objects might be impossible to give, but then
          estimate it like <br>
          other tools does, e.g. rsync will just update the number of
          remaining <br>
          items as it learns about them...). There is also no estimate
          of how long <br>
          it will take or the processing rate (how many items per
          minute, etc.) <br>
          All these infos would be useful. Especially if trying to make
          it faster <br>
          and see what works.<br>
          <br>
          I also think the scanning could be sped up quite a bit by
          using multiple <br>
          threads. I have my collection on a NAS. I believe a lot of the
          time <br>
          spent on scanning/indexing is really spent on network latency:
          waiting <br>
          for the samba server to list files, open file handle, etc.<br>
          <br>
          When I look at resources on the DK client machine, it is
          mostly idle on <br>
          both CPU, network and disk I/O. Same with the NAS server, it
          is mostly <br>
          idle on those resources as well. So I think a lot of time is
          pent <br>
          waiting for network packages. If I could allow 2, 4, 8, ..,
          parallel <br>
          threads, I think we could see this wait time reduced
          significantly and <br>
          CPU/net/disk utilization increased.<br>
          <br>
          <br>
          BR<br>
          <br>
          Thomas<br>
          <br>
        </blockquote>
      </div>
    </blockquote>
  </body>
</html>

--------------dpxvyWSSYYBNSZc4XA5bE0nE--