Delete (or change) many elements with the same object number or text

[MedBooster]

image.png

How can I delete about 1000 items that all have this object number? "7,393"

edit:
the number changed now, but it is still the same for all the items I want to delete
FYI, they are text boxes, but they are not "comments" they're actually part of the original document.

image(1).png

[MedBooster]

well this did not work ... maybe someone has a javascript script for this ?

THe object number is 992

Code: Select all

let objects = [/* your array of objects */];

for (let i = objects.length - 1; i >= 0; i--) {
  if (objects[i].objectNumber === 992) {
    objects.splice(i, 1);
  }
}

console.log(objects); // this will output the modified array of objects

[MedBooster]

alright.. I managed to delete what I needed to delete with the redacting tool because it contained text,

I just had to change the redaction color to "none"

image.png

edit:
no that did not work because "none" actually completely removes everything behind what I am trying to remove as well, and I don't want it to do that.

And
I am still curious to hear if you can do something similar for shapes, or highlights of a particular color, maybe with javascript, or redacting

Though... I guess if you can't filter for comments, you won't be able to mass delete them either
However, I am also referring to shapes that are part of the original comment, not just shapes you have added yourself as "comments"
viewtopic.php?t=40233


Maybe anyone has been able to use AutoHotKey for something like this?
https://superuser.com/questions/811143/removing-identical-element-from-all-pages-of-a-pdf

[MedBooster]

image.png

consider this a suggestion to add an option to "delete" all results, not redact, in the advanced search tool.

image(1).png

[MedBooster]

Does PDF-xce have a find (search) and replace feature? Maybe you could replace it with nothing to delete it?

[Tracker Supp-Stefan]

Hello MedBooster,

There is no find and replace feature for now. The best is "find and Redact" - which does remove more than what you want it would seem!
I am afraid that I can not think of a quick way to delete multiple base content objects in an automated way.
The JS you shared above is incomplete - as you are not specifying the "objects" array, and as far as I am aware JS inside PDF can't really control base content, and is mostly working with annotations.

Kind regards,
Stefan

[MedBooster]

Hello MedBooster,

There is no find and replace feature for now. The best is "find and Redact" - which does remove more than what you want it would seem!
I am afraid that I can not think of a quick way to delete multiple base content objects in an automated way.
The JS you shared above is incomplete - as you are not specifying the "objects" array, and as far as I am aware JS inside PDF can't really control base content, and is mostly working with annotations.

Kind regards,
Stefan

why is it even that redact removes all that is below it? Doesn't deleting the base content delete the information completely...

edit: I guess I just have to convert the PDF to Word, find and replace to nothing, and then convert it back to PDF. Would have been nice not having to go through all that hassle though.

[Tracker Supp-Stefan]

Hello MedBooster,

The "Redaction tool" is intended to "redact out" or "fully erase" the content you mark - so that if there is any sensitive or confidential information in the file - you can be sure that it will be removed with the correct redaction applied. So indeed this tool is intended for permanent removal of information from a PDF file and should be used with caution.

Kind regards,
Stefan

[David.P]

If it's about removing text, you could search for the text and then apply a white highlighter to all search results (if the background color is white, otherwise use whatever the background color is for the highlights), then select all the highlights in the Comments Pane, and finally set their blend mode from "Multiply" to "Normal":

image(1).png

Maybe flatten all the highlights afterward.

It then looks like this (I have removed, i.e. covered up, all occurrences of "that" and "and" in the text):

image.png

[MedBooster]

If it's about removing text, you could search for the text and then apply a white highlighter to all search results (if the background color is white, otherwise use whatever the background color is for the highlights), then select all the highlights in the Comments Pane, and finally set their blend mode from "Multiply" to "Normal":


image(1).png


Maybe flatten all the highlights afterward.

It then looks like this (I have removed, i.e. covered up, all occurrences of "that" and "and" in the text):


image.png

well , the redacted tool could also do that if you set the color to "none", but the thing is that as I already stated – it also removes everything below. and in my case I want to keep what is below.

It is strange, because when I am manually deleting these objects what is below *stays*

[TrackerSupp-Daniel]

Hello, MedBooster

So, the purpose of the redaction tool is to ensure sensitive information is properly removed, while yes you can simply select and delete the content manually, Sometimes (spelled "surprisingly often") PDF will have duplicate sets of that information, especiall if OCR or printing from a web browser was ever performed on this document. In many cases, it is critical that all this information is fully removed, and so the redaction tool, till delete any any all objects in a selected area, as well as edit any images that happen to overlap, "cropping out" a section of the image, and even delete comments in the area. This is very much a tool intended to "be sure I actually got rid of it" instead of relying on you to check every single possible content type, and cycle through the sheet with various tools to ensure you have removed it all, then select your images and open them up in an external editor to cut out that part of the graphic.

In short, it is something to use when you just want to be sure its really "gone" (and is very useful in particular for legal fields).

If you do not need to truly remove these items, or as you say, you do still want the text below to be searchable than yes, using a comment such as a page colored rectangle and "flattening" it would allow you to obscure the content without deleting any of it.

Kind regards,

[David.P]

it also removes everything below

Yep, that's what the Redact Tool does. If you cover stuff up (e.g. with intransparent highlights), the background does not get deleted. Depending on how the background looks, it can be looking covered up as well.

[MedBooster]

can you move all objects that are in the advanced search results to a new layer maybe?

Then I would just be able to create a new layer, delete it, and be done with it ...

edit: somehow that also did not work. SO: I am trying to convert the PDF to HTML, delete what I want removed in the HTML, and then convert back to PDF.

[TrackerSupp-Daniel]

Hello, MedBooster

To be sure, what exactly is the purpose here, are you trying to ensure that the text cannot be searched/seen or anything like that ever again, or are you just trying to mask it but ensure it is still searchable?

If you want to fully remove it all, Find and Redact is what you are looking for, it does what it says on the tin, searches for text strings, and places a redaction annotation on them, so you can apply it when you are ready.

If simply masking it, then using Davids suggested "Highlight" settings and setting them as the default, then using advanced Search and the "apply highlight to search results" option (in the advanced search options dropdown) would be the simplest solution (now that I think of it, the advanced search also offers a "redact results" option, so that is another way to do the former).

Kind regards,

[MedBooster]

Hello, MedBooster

To be sure, what exactly is the purpose here, are you trying to ensure that the text cannot be searched/seen or anything like that ever again, or are you just trying to mask it but ensure it is still searchable?

If you want to fully remove it all, Find and Redact is what you are looking for, it does what it says on the tin, searches for text strings, and places a redaction annotation on them, so you can apply it when you are ready.

If simply masking it, then using Davids suggested "Highlight" settings and setting them as the default, then using advanced Search and the "apply highlight to search results" option (in the advanced search options dropdown) would be the simplest solution (now that I think of it, the advanced search also offers a "redact results" option, so that is another way to do the former).

Kind regards,

Even PyMuPDF (Python) did not seem to let me delete the base content I want to delete. So I just have to do it manually I guess.

Well,
the "redacted" tool is great when the goal is to redact and get rid of sensitive information, BUT, my goal here is more related to the document appearance, so information trails don't really matter, because this is not about redacting sensitive information.
This is about when there are some shapes or text covering text in the background (base content) you don't want to get deleted (which happens when you redact what is in the front)

Furthermore,
as already mentioned a mass-deletion option of certain search results would be nice, also for comments which may contain certain keywords, or just be of a particular object number / color / type.

[TrackerSupp-Daniel]

Hello, MedBooster

So, the solution David proposed will work for your needs in the time being then? I should also mention that a find and replace function is planned for the future, though I cannot say how far along that may be. Finding objects by the individual object number (which to my knowledge is never the same for multiple objects in the document, even if they are the same type and share similar properties) and deleting them, will not likely be added, but I can pass the idea along for consideration.

Kind regards,

[MedBooster]

Another IRL example of why I would want to do this.
A 100 pages of lecture slides... all with the same terrible background...
Which makes it impossible to read the text...
It would have been nice to just check the object number in the properties and basically delete the background from all the slides

However... as you can see in the screenshot below.. the weird background seems to be recognized as a hyperlink... and there is no object number.

image.png

[Tracker Supp-Stefan]

Hello MedBooster,

That's the thing - that those object numbers might not match - or some objects might not have them at all!
So for such specific use cases like yours where you want to remove a specific object, however not redact out the area of that page fully - then only manually removing the objects one by one seems to be a viable option.

Kind regards,
Stefan

[David.P]

Hello all,

To my knowledge, a link cannot contain an image, or even formatting (other than the link border format). Here, that link is probably placed over the background, which is why the background may be difficult to select.

In other words, the link itself is probably not the unwanted background object here. If this were so however, the link could be removed relatively easily from the entire document at once using the Link Pane.

Best regards
David

[Tracker Supp-Stefan]

Hello David.P,

Thanks for your input!
Indeed you are probably right! If MedBooster can share a sample with us - I will be happy to take a look at one of those files.
There could be some masks or something else (besides the link) that is e.g. altering the background image as well.

Kind regards,
Stefan

[MedBooster]

remove annoying background.pdf

Go ahead give it a shot,
I would be curious to see if there would be a way to batch remove the background from 100+ pages.

In this case I just duplicated some of the pages

[David.P]

With some luck and depending on what else you have in your document that you'd want to keep, there is a chance that you could succeed with this command (that also can be found in the "Contents" pane behind the gear icon button):

image.png

[MedBooster]

With some luck and depending on what else you have in your document that you'd want to keep, there is a chance that you could succeed with this command (that also can be found in the "Contents" pane behind the gear icon button):


image.png

Hello David,
I'm not sure I understand what you mean.
Sure I am able to select and delete it

I am asking for a way to batch-delete identical elements like these.

Maybe they would not have the same base content ID... but I assume a PDF editor would be able to see that they are the same size, color etc...

[David.P]

If you use that exact command, it will select all shapes in the entire document with just one click. When you're lucky (i.e. when there are no shapes in the document that you'll want to keep), then this should be what you are after.

[MedBooster]

If you use that exact command, it will select all shapes in the entire document with just one click. When you're lucky (i.e. when there are no shapes in the document that you'll want to keep), then this should be what you are after.

Maybe there are shapes I would like to keep though

That's the reason for the thread.

With efficient and productive software such as PDF-xce there should be no "if you are lucky" , but simply you are lucky to be using PDF-xce

(nevertheless thank you for the input, it might of course be a good workaround)

[Tracker Supp-Stefan]

Hello MedBooster,

David was trying to help with a suggestion that might work with a specific subset of files. So yes it probably can be called "luck", or I'd rather use 'coincidence to have the right type of file'
Anyway that is semantics.

Thanks for the kind words about our software once again. However on this specific topic - we do not really have an automated solution for you, and you might be looking at doing that particular task manually.

Kind regards,
Stefan

[rakunavi]

Hello all,

I would like to comment on how to remove the background, as it may not apply to all cases, but perhaps it may be helpful.

The content pane of the PDF-XChange Editor is basically displayed in order from the bottom of the Z-order, so if the PDF file has only one background object at the bottom of the Z-order, it is relatively easy to delete that object with RPA tools like autohotkey.

The keystrokes required are shown in the autohotkey script below for reference. The number of times the left and right arrow keys must be pressed must be equal to or greater than the number of times it takes to reach the first leaf (object) of the tree in the content pane. Pressing more than the required number of times is not particularly problematic, it only slows down the process.

Sample A

Code: Select all

SendMode Event
Loop, 5
	Send {Right 4}{Delete}{Left 4}{Down}
return

• Animation1.gif

• SampleFiles.zip

As shown in the video above, both the attached sample file and the file attached by MedBooster can be deleted with this script, but it may not work depending on the structure of the PDF file. In that case, you should be able to delete them with the following keystrokes, albeit somewhat less efficiently.

Sample B

Code: Select all

PagesCount := 5
Loop, % PagesCount
{
	SendMode Event
	SetKeyDelay 40
	Send {Right 4}{Delete}{Home}
	SendMode Input
	Loop, % A_Index
		Send {Left}{Down}
}
return

• Animation2.gif

If there are multiple objects to be deleted, starting from the very bottom of the page, you can run these scripts multiple times. If you review the keystrokes, you may be able to delete them more efficiently. Since this is just an RPA tool, you need to adjust the wait time appropriately according to the performance of your hardware, but it may save you some time compared to deleting everything by hand.

Best regards,
rakunavi

[MedBooster]

Hello all,

I would like to comment on how to remove the background, as it may not apply to all cases, but perhaps it may be helpful.

The content pane of the PDF-XChange Editor is basically displayed in order from the bottom of the Z-order, so if the PDF file has only one background object at the bottom of the Z-order, it is relatively easy to delete that object with RPA tools like autohotkey.

The keystrokes required are shown in the autohotkey script below for reference. The number of times the left and right arrow keys must be pressed must be equal to or greater than the number of times it takes to reach the first leaf (object) of the tree in the content pane. Pressing more than the required number of times is not particularly problematic, it only slows down the process.

Sample A

Code: Select all

SendMode Event
Loop, 5
	Send {Right 4}{Delete}{Left 4}{Down}
return

• Animation1.gif

• SampleFiles.zip

As shown in the video above, both the attached sample file and the file attached by MedBooster can be deleted with this script, but it may not work depending on the structure of the PDF file. In that case, you should be able to delete them with the following keystrokes, albeit somewhat less efficiently.

Sample B

Code: Select all

PagesCount := 5
Loop, % PagesCount
{
	SendMode Event
	SetKeyDelay 40
	Send {Right 4}{Delete}{Home}
	SendMode Input
	Loop, % A_Index
		Send {Left}{Down}
}
return

• Animation2.gif

If there are multiple objects to be deleted, starting from the very bottom of the page, you can run these scripts multiple times. If you review the keystrokes, you may be able to delete them more efficiently. Since this is just an RPA tool, you need to adjust the wait time appropriately according to the performance of your hardware, but it may save you some time compared to deleting everything by hand.

Best regards,
rakunavi

Thanks a lot, I've been trying to understand how to use Javascript for things like these

[Tracker Supp-Stefan]

Hello MedBooster,

JS can not really work with base content objects (other than to e.g. perform a search) - and the code Rakunavi shared is for AutoHotKey I believe.

Kind regards,
Stefan

[MedBooster]

On a similar note, it would be cool if you were able to batch change the color for example of all elements. Let's say turning all elements with a specific color

I am referring to base content here, as it is of course easily done with comments.

[Dimitar - Tracker Supp]

Thanks for the input.

This will really be a useful feature to have.

Regards.

[MedBooster]

Thanks for the input.

This will really be a useful feature to have.

Regards.

Maybe this is something Javascript could do... changing a particular HeX color code to another one for all base content elements.

[TrackerSupp-Daniel]

Hello, MedBooster

Sadly, JS cannot directly interact with existing Base content objects. It is able to create base content in some specific cases, by leveraging the "flatten comments" functions, and using or creating other comment objects beforehand, but it cannot grab a base content item and modify it directly.

Kind regards,

[rakunavi]

Hello MedBooster,

On a similar note, it would be cool if you were able to batch change the color for example of all elements. Let's say turning all elements with a specific color

I am referring to base content here, as it is of course easily done with comments.

While not directly changing the color of the base content itself, the following steps can be taken to achieve the same on-screen results. This would satisfy most of the demands.

1. Convert to grayscale using the Convert Colors function. Specifically, specify 'Black & White' or 'Gray Gamma 2.2' as the Conversion Profile.
2. Specify white as the background color of the document.
3. In the foreground of the document, draw a rectangular comment large enough to cover all of the base content with the color you want to change as the fill color, and change the blend mode to 'Lighten'.
4. Flatten the rectangular comment.

If the document is originally in black and white or grayscale, Step 1 is not necessary. Also, if you do not want to flatten the added rectangular comment, you do not need to perform Step 2 and Step 4. In the end, the point of this method is to change the final color that reaches the eye, just like putting on a pair of colored glasses, by structuring the document as follows.

• figure.png

Therefore, as long as this point is taken into account, there are other methods, not limited to Step 3 and Step 4. For example, you can create a blank page with the background color set to the desired color and the blend mode of the corresponding shape set to Lighten, and then add the page as an overlay or watermark to achieve the same result.

If you want to change the color of content on multiple pages at once, you can do so by duplicating the comments created in step 3 on all pages using the copy function. You can more easily apply the changes to all pages by using the overlay and watermark functions.

See the video for more details.

• Animation.gif

Incidentally, the reason why Step 2 is necessary can be found in the following topic.

• When a comment is set to a specific blend mode, preview and actual display are different when background is specified
  viewtopic.php?t=40920

Best regards,
rakunavi

[MedBooster]

Hello Rakunavi
hmmi interesting, as always, thanks for the video and explanation, it's not exactly what I am asking about, as I am more referring to batch-changing properties... however this might also be useful in some cases, and I wouldn't be able to come up with it myself.
Kind regards

[rakunavi]

Hello MedBooster, thank you for your reply.

I am well aware that this is not the very function you are looking for.

Lighten in Blend Mode is effective for partial coloring of monochrome scan images. There is no need to set the background to white when working with images. Since monochrome documents will exist for a long time to come, knowing how to use this blend mode will make such documents easier to read. In my Comment Styles Palette, I have set styles with Blend Mode set to Lighten for most comment tools.

I hope that one day this post will be useful to someone who reads it.

Best regards,
rakunavi

[MedBooster]

Hello MedBooster, thank you for your reply.

I am well aware that this is not the very function you are looking for.

Lighten in Blend Mode is effective for partial coloring of monochrome scan images. There is no need to set the background to white when working with images. Since monochrome documents will exist for a long time to come, knowing how to use this blend mode will make such documents easier to read. In my Comment Styles Palette, I have set styles with Blend Mode set to Lighten for most comment tools.

I hope that one day this post will be useful to someone who reads it.

Best regards,
rakunavi

The reason I am asking for this is simply because some PDF documents or slideshows might be in this color combination that kind of strains your eyes when you're reading. So I appreciate any way to basically adjust this.

Using "override document colors" is another option, but that feature does not affect images (yet?)

[TrackerSupp-Daniel]

Hello, MedBooster

At this time, I do not believe there are plans for color overrides to affect images, only base content and some comment types. That may change in the future, but I cannot make any promises on that front.

Kind regards,