Web search not searching in PDFs?

DEVONagent Feedback
1

If I go to google.com in Chrome and search for “devonagent web deeper deepest”, I see in the second hit the information I’m looking for, in a PDF on Amazon S3.

If I search in DEVONagent Pro, using the Google, Web (Fast), Web (Deep) or Web (Deeper) search sets, it does not find this PDF.

This makes DA less useful for me, since most of my searching is for academic articles where the search results are in PDFs, not web pages. Is there anything i need to enable to make this type of searching happen?

Thanks, John

2

Ah, I found it, I can enable it in the search set configuration.

3

Thanks for the heads up on this, I had not realised that PDFs were not searched and that you could configure search sets to include them :blush:

4

Yes, in particular you need to do two things: Select “Objects” as a searched item, and then “PDFs” as one of the objects that will be searched.

5

SO, in order for DEVONagent to search inside objects, you need to check the Objects box and check all/at least one bxo under Files?

:exclamation: The documentation does not mention this, and the UI doesn’t give a hint, either.

6

No, these are independent options. Each filetype can be enabled on its own. In addition, searching scanner objects just means that the URL of found objects (e.g. movies, webcams, thumbnails) is compared to the search term and in case of a match the page is accepted.

7

I believe we might be not talking about the same thing?

What is the sense of the URL option compared to the object option?

For reference see the DEVONagent Pro Version 3.10 Documentation, page 24.

8

If the URL option is enabled, then the URL of the page is matched. Enabling the Objects option matches the URLs of the objects of the page.