Hi there,
I’ve just started trying out DEVONthink, and I’m so far finding it useful for my purposes. However, there is one thing that is creating some frustration, and I would love to hear if I’m doing something wrong or if there is a fix. Please bear with me if the question is ‘noobie’. I’ve searched the forum but haven’t been able to find a relevant thread. In case it helps, I’m currently using the trial version of DT4.
So when I search for a keyword within a specific group containing OCRed PDFs, I get a series of hits. However, I notice that not all of the hits actually contain the keyword that I’m searching for. Sometimes the ‘relevance’ bar is pretty full, but still the keyword doesn’t exist, while some PDFs with lover relevance according to DT does contain the keyword.
Is there someway to make DT only show the PDFs that contain that specific keyword? Also, what is the relevance based on if the keyword I’m searching for isn’t contained in the PDF?
Many thanks.
By default DEVONthink searches also metadata (e.g. name), you could use the search prefix text: to avoid this.
Many thanks for the swift reply. Unfortunately the problem persists even when I add the text: prefix. Also - the keyword doesn’t appear anywhere in the metadata either. I suppose the only thing that I can think of is that they are filed in the same group as other PDFs that do contain that keyword.
Which toolbar search options are enabled?
Not sure if this is what you asked for, but I’ve attached a screenshot. So this search will yield results that do not include ‘Poland’ in the text or content. In fact, it I will get the same items whether I use the prefix text: or not.
A screenshot of the popup of the toolbar search field (see loupe icon) would be great, thanks.
Does disabling Partial matches while typing make a difference? If not, then please convert one of the found PDF documents to plain text. Does the text contain the word?
Hi again. Partial matches while typing didn’t make a change, however I did convert the several of the PDFs to plain text, and the text does in fact contain the word! What could this be due to, and do you know if there’s a fix? Many thanks for the help
Either due to OCR errors or to the way these documents were created.
Thank you for the prompt reply. I’ll try and redo the OCR and see if that solves the problem.
