No announcement yet.

ACDSee Ultimate 2021 - Edit Metadata Action's steps

  • Filter
  • Time
  • Show
Clear All
new posts

  • ACDSee Ultimate 2021 - Edit Metadata Action's steps

    How can I edit (or at least view) the contents of an old Metadata Action on a very detailed level?
    I would need to trace and change or remove metadata elements that I cannot find in the Metadata View listings.

  • #2
    Originally posted by Erik Törnroos View Post
    How can I edit (or at least view) the contents of an old Metadata Action on a very detailed level?
    I would need to trace and change or remove metadata elements that I cannot find in the Metadata View listings.
    Just to be clear, are you talking about recorded Actions from Edit mode, or custom Metadata Presets.
    If Metadata Presets, then they can be edited from the Manage Presets option in the Metadata Presets pull down menu.

    If it is a recorded action, then AFAIK you can only view the major steps via the action browser, but not edit them, and it doesn't display the individual settings.
    I know of no way of editing recorded actions. I have made it a practice to document all the steps and individual settings for custom actions when I first develop the steps, and then work from that documentation when I record them. That way at a later date, I can revise the steps and re-record the action if needed.


    • #3
      Thank you for your very comprehensive answer. I had found the major steps but, as you predict, not the individual, detailed settings.

      Unfortunately my question was on a recorded action so either I need to find my old notes or I have to re-develop it from scratch.
      In labeling my actions I have indicated which elements are included - but I cannot find all the elements in the listing of Metadata elements.
      And I initially expected that it would be possible to copy and edit previous actions.

      Most elements I can find in the Properties pane's Metadata and, in my case, primarily in the IPTC Metadata.
      But some information can obviously come from other elements: in my example at hand one is the File Name and antoher I have noted as "writ".

      Where are they stored and what other elements are there that can be used in recording actions?


      • #4
        Erik Törnroos
        Erik Törnroos

        I'm not sure I fully understand your question, so sorry if this reply does not help.

        "Recorded Actions" are recorded from Edit mode, and record the individual tools and their settings used. They can then be played back from Manage mode. The Edit tool that most comes to mind as being able to directly use metadata is the Text tool. That tool can add text taken from metadata. The list of available metadata types, and the individual fields available is presented in a dialog when you select the text tool and then click on the "Ins" button.

        You can view some of the ACDSee metadata fields, some of the Exif metadata fields, and some of the IPTC metadata fields from the Properties Pane, Metadata tab, when "Default Metadata View" is selected. To see more fields, you can change the selection to "All Exif" or "All IPTC".

        The "canned" recorded actions for U2021 (that are installed with the program" are stored at C:\User\Username\Appdata\Local\ACD Systems\Actions\140Ult\V10.

        The custom user recorded actions are stored at C:\User\Username\Appdata\Local\ACD Systems\Actions\V10.

        Neither are in a form that you could edit.

        I hope that helps, but if not, perhaps you could describe what the action is required to do, and what metadata field you want it to use.


        • #5
          Many thanks for your clarifications.
          Our major project is to add explaining text to most of our pictures in order to give them a chance to survive the transition to the next generation. For this I make a few Metadata Actions that add one or several white banners beneath the picture and write in the necessary Metadata elements. One to three lines depending on how much data is needed.

          I think that my biggest problem was solved through your comment on the use of the "Ins" button. That way I found the "invisible" elements that I was missing. And using the Text tool makes it possible to edit -at least on a single picture- the text in the banner, even at a later date.
          Still, how much easier would it have been if only one could edit user recorded actions.
          Thanks again for the help.


          • #6
            Erik Törnroos
            Erik Törnroos

            You can possibly use Batch Edit, to do what you want and it can both be used on a group of selected images from manage mode , and be edited

            For example In the attached screen shot, it is adding a white band at the bottom, and placing the ACDSee Caption and the Exif Date/Time Original text in black on that band.

            Click image for larger version  Name:	Batch edit.jpg Views:	0 Size:	241.5 KB ID:	58988
            By putting a Shift +Enter between the two metadata fields instead of a space, and changing the width of the bar, you can split it into two lines.

            Click image for larger version

Name:	Batch Edit 2.jpg
Views:	71
Size:	239.4 KB
ID:	58989
            Pro Member
            Last edited by Greyfox; 01-10-2022, 07:21 PM.


            • #7
              Your enlightening example of using ACDSee’s Batch Edit function helped a lot. I had abt a year ago tried it out, but abandoned it as I did not see how I could make it do what I wanted. And ACDSee suggested that one rather should use Batch Editing. Neither is it perfect now, but I found out a way to make it work well enough. One advantage of using Batch Edit is that the process is simple and it can be edited – which is not possible with the Recorded Action.
              Your example shows how to add a white (or transparent) bar onto which I can add text. But the drawback is that the bar and text in fact overlays / hides a part of the picture – which I do not want.

              Through some experimenting I found out that I can create a white bar outside the proper picture by enlarging the picture. Slightly stupidly (IMO) it seems that I can only create two synchronised bars – one above and one below the picture. I could not find a way to create (or allocate) the bar to only one side of the picture.

              Still, this I can actually live with as this version of the picture mainly is for explaining the contents and persons in the picture and the picture as such stays unchanged. It could, for that reason, actually work with the text partly on top of the picture – but not as neat (and there is the risk that this could be the only version left of the picture).
              Another way to go would be to first use a Recorded Action adding a white bar underneath the picture and then use the Batch Edit to add the Text Overlay into the bar area. Though this would require two rounds of batch processing.
              One cannot always get the perfect solution – but this can be considered to be good enough.
              My sincere thanks for your help and the time you put into showing me how to make use of some rather tricky functions in ACDSee.

              During this process I did run into another of ACDSee’s properties I have to circumvent.
              In registering keywords for a picture – naming the persons – one would like to have them printed out in a certain order – e.g. persons from left to right. But keywords entered into the Metadata field is presented in alphabethical order in the picture, which isn’t very helpful.
              This can be evaded by numbering the persons in the keyword field (1 alfa, 2 beta ...) or by just writing out all the names as a string without hitting space between the names (this string works well alone but it will be alphabetised as one “name” if anything comma separated is added as further keywords ).
              Both work, but neither is really good. Again we'll accept "good enough".


              • #8
                Erik Törnroos
                Erik Törnroos

                I started using black bars for captioning some years ago for images that would be shown on a TV. The images are placed on a USB flash drive, plugged into a PVR, and played using the slide show function of the PVR.

                To cut a long story short, I used recorded actions to create a black bar at the bottom of the images, taking into account there were both 6"x 4" and 6" x 4.5" images of different pixel sizes from different cameras, and whilst those actions have been refined over time, I still use the two process method. Add the bar as a recorded action, and place the captions as a Batch edit.

                I did experiment with using the border tool in Ultimate's Edit mode (which can be configured to produce a single bar on either top, bottom or left or right sides), but found it could not do exactly what I needed. The border tool calculates border width dependent on the the aspect ratio of the individual image, so does not give consistent results across different aspect ratio images.

                When this subject came up again late last year, I raised the border tools issue with ACDSee Support, and in December 2021 they advised that they would raise an improvement request to give an option in the border tool to use exact pixel size as well as the current method.

                Not being able to edit recorded actions is not for me a big issue. Most of my recorded actions are by nature relative simple, and because I document all the steps, I find it is just as quick to re-record them. I usually do a manual step though anyhow to check that the proposed changes are going to work as expected.

                For the "left to right" identification of people you mentioned, I use the ACDSee caption (or the IPTC title). My images can contain many keywords, not just names, and whilst the Batch Edit's text overlay allows for the inclusion of multiple metadata fields, it doesn't allow for selection of only parts of a particular metadata field. I find it is just as simple to write in a "left to right" caption as it would be to write a "left to right" multiple name keyword, and it doesn't have any side effects.