Wrong positioning of the search marks

[kenneth488]

Hi KurtCode

I have the library with different PDF files, and it does often place the search marks wrong. In my fork, there is a PDF file called amazon-dynamo-sosp2007.pdf. Searching for the string "re" will show you that the position of the marks a sometimes calculated wrong, since they are misplaced. The same goes for the string "in" and probably others.

Is this something you might be able to fix?

Have a great day

Regards,

Kenneth

[viezel]

+1 could be nice to get sorted out.

[kenneth488]

I have almost fixed the issue now, so the general solution works. Will commit and push today.

[kenneth488]

I will commit it as soon as all the bugs are fixed! Are working on it atm! :)

[kenneth488]

I will commit the fixes soon. I also implemented support for newline word searching, so we can mark search words spanning more than one line.

When I commit, you can fork it and implement your fixes.

[nattapon21]

.....

2015-06-12 16:18 GMT+07:00 TryMyBest [email protected]:

Ok. Will check your new version. Nice work guy! :)

—
Reply to this email directly or view it on GitHub
#85 (comment).

[kenneth488]

The commit is up. Now remember that there might be more bugs. If so, post them. I am still working on making it a lot better.

[kenneth488]

Yes just saw that. Though that is only in the test project. It does work when you give it the correct page size. The way Kitten sets up a page, is pretty weird. It works fine otherwise:
http://postimg.org/image/f5py9l2d7/
http://postimg.org/image/ksmb724vv/

I will see if I can fix it in Kitten.

[kenneth488]

What you need to pass to the convertPDFRectToViewRect method is the frame of the userSpace PDF page aka. the visible pdf page you are viewing. Anything else will fail.

[timeflying]

Still mark on the wrong position~

http://www.takming.edu.tw/study/upload/filedownload150318163116.pdf

[timeflying]

I use the app preview which embedded in mac os . It works great on the same pdf.