PDF search checkboxes
Hello! I did not find a similar thread, so I'm writing here.
I love Zotero. However, for me, search in big PDFs in Zotero's default reader have been slow (and slower than other PDF readers on either Windows or iOS). I tried doing multiple things, including deleting the full-text index, but didn't help. I think there were other threads describing this but I can't find them now.
But today I found something very interesting. The search bar in Zotero desktop has three checkboxes underneath. The first one, "Highlight all" seems to be the culprit. It seems that the speed is greatly affected when Zotero tries to highlight all the responses in a PDF.
The problem is, this option is checked by default and I have to turn it off every time I search something. So, is there any setting to address this, have it turned off by default? Or can this be a feature suggestion?
On iOS there are no checkboxes for this. But it's the same: any other pdf viewer is much faster. Maybe it has the same cause as the desktop version?
https://s3.amazonaws.com/zotero.org/images/forums/u10271061/4uvy4b6sqdug2x31tzkf.png
I love Zotero. However, for me, search in big PDFs in Zotero's default reader have been slow (and slower than other PDF readers on either Windows or iOS). I tried doing multiple things, including deleting the full-text index, but didn't help. I think there were other threads describing this but I can't find them now.
But today I found something very interesting. The search bar in Zotero desktop has three checkboxes underneath. The first one, "Highlight all" seems to be the culprit. It seems that the speed is greatly affected when Zotero tries to highlight all the responses in a PDF.
The problem is, this option is checked by default and I have to turn it off every time I search something. So, is there any setting to address this, have it turned off by default? Or can this be a feature suggestion?
On iOS there are no checkboxes for this. But it's the same: any other pdf viewer is much faster. Maybe it has the same cause as the desktop version?
https://s3.amazonaws.com/zotero.org/images/forums/u10271061/4uvy4b6sqdug2x31tzkf.png
After turning off the "Highlight all" option as described by @educatedspice99, the warning window didn't pop up.
I found two main reasons:
1. The search is automatically enabled even if the desired texts are not fully typed, a shorted search pattern always means more matched results and longer execution time.
2. Rendering the highlighting feature may be a performance bottleneck.
It would be great that it the automatic search is disable if the PDF file is too large, or only the current pages shown on the screen are rendered.
Tang, K., Adcock, T. A. A., & Mostert, W. (2025). Droplet bag formation in turbulent airflows. Physical Review Fluids, 10(3), 033604. https://doi.org/10.1103/PhysRevFluids.10.033604
takes 20 seconds to return 291 results the first time after loading the PDF file.
It shows the spinning wheel most of that time, without showing any search result:
https://s3.amazonaws.com/zotero.org/images/forums/u265723/ym3o7g2m7yz8t33lgcms.png
It takes 15 seconds without the "Highlight all" option.
It takes less than a second to show the same search results in Firefox.
Zotero 7.1-beta.43+59730c0fc (64-bit)
Windows 10
Edit: If I wait 30 seconds scrolling through the PDF file just after loading it, and then only do the search, the search results appear in less than 1 or 2 seconds.
So the problem seems to be that the initial processes done after loading the PDF file are blocking the search results.
@educatedspice99: Do you see the slow searches only after loading the PDF file, or also after reading it for a few minutes?
Do you have an example of an Open Access PDF file with slow search results?
Thanks to everyone for the contributions. I had a hard time reproducing the issue because, I have to admit, this got better in the last few months—in the sense that it used to happen all the time, now it only happens after loading the PDF file. Actually, it happens every time I load the PDF file, even if I close and open again the same file.
@mjthoraval I don't have a open access pdf because it is easier to reproduce with big books. I'll leave here a drive link with some screen captures where I tried to reproduce the issue on both windows and iOS, with the used PDFs. I didn't actually fully explored the iOS behavior, but I think I showcased some weirdness. The tested PDF has no issue in other apps.
https://drive.google.com/drive/folders/1BM9vU8QFaK6tCPRNe02cMwFeLSKSgX1z?usp=drive_link