PDF reader search freezes

ForkNTalk
I've been getting a bug when searching within an open PDF from the reader search box, not from Edit/Search or Edit/Advanced Search.

As I start typing the search word, the function begins searching the text before I finish typing, which can jam things up when the word I am using begins with a commonly used letter combination; for example; so, or at. The timer icon appears and if I click too soon I get a java script error pop-uip and usually have to restart Zotero.

Is there a setting that that only allows searching after I hit Enter?

Thank you

William
  • martynas_b
    1) How many pages does the PDF file have?
    2) Does it happen with other PDF files?
    3) Which Zotero version?
  • ForkNTalk
    Thank you for your response.

    1) It usually happens on PDFs of more than 100 pages, but not all the time.
    2) It has occurred on multiple files
    3) 7.0.27 (64 bit)

    I am reading scanned and PDFd documents from the early 19th century, that were printed. Some I have download with already recognised text and some I have OCR'd myself using the Tesseract engine. If there is a poor image quality I create image files for each page and then run them through Topaz Sharpen or Topaze DeNoise, recompile them into a PDF and then OCR them. sometimes because the original image was too poor the resultant OCR, even after enhancement, is also poor- but I'm not including those in the issue.

    I wouldn't sweat it, more than anything it's annoying. It would, however, be good if there was a check box to allow for searching only after ENTER has been hit- yes I know easier said than done.

    Next time it occurs I'll check the Error Console. I have noticed from yesterday there is an entry "19:14:20.287 Missing resource in locale en-AU: reader.ftl "

    William
  • martynas_b
    There isn't any setting for this, but normally you shouldn't have any problems with just a hundred pages.

    Could you send one of those slow PDF files to support@zotero.org, including a link to this thread?
  • ForkNTalk
    Hi,
    Will do, I can't remember which PDF it was because it happens so frequently. Next time it happens I'll upload the PDF to the link. I use Zotero for several hours each day, so I'm sure it will happen again soon. Yesterday, 20 November 2025, it occurred three times.
    When I look at the Error Console I notice that I have a lot of JSON Parse errors, for example:
    11:10:02.832 SyntaxError: JSON.parse: unexpected character at line 1 column 10 of the JSON data 3

  • ForkNTalk
    edited 49 minutes ago
    I just tried sending an email with a PDF that just cased the error, but the mail server said some very impolite things before rejecting the file as too large, 25 MB 517 pages of OCR'd text under the images. Attempting to compress it doesn't reduce the file size. I sometimes achieve success by printing the PDF to PDF, but Acrobat was as equally impolite.
Sign In or Register to comment.