pdf search text question

shogun1234 · 09-16-2007, 11:56 AM

i use ps2pdf to create a pdf file (the ps file is downloaded from the internet - http://user.it.uu.se/~joachim/intro.ps). the command i use is `ps2pdf in.ps out.pdf'

however, whilst opening the pdf and try to search the text, i found out the matched text (e.g., An) will be highlighted and the area of the highlight is full of the page.

what cause such problem?

thank you very much

jschiwal · 09-16-2007, 08:24 PM

I downloaded the Postscript file and converted it myself. When I looked at the contents in vim, is looks there aren't more than two consecutive letters side by side. This document went from tex -> dvi -> ps -> pdf. I tried searching for more than one word in a "gawk.pdf" document that I produced from source, and was able to.

shogun1234 · 09-17-2007, 06:33 AM

Quote:

Originally Posted by jschiwal

I downloaded the Postscript file and converted it myself. When I looked at the contents in vim, is looks there aren't more than two consecutive letters side by side. This document went from tex -> dvi -> ps -> pdf. I tried searching for more than one word in a "gawk.pdf" document that I produced from source, and was able to.

Sorry I do not catch the point. How can I search a word (e.g., process) in that generated pdf without copy words in other lines?
Or what steps you've done enables to select/ search word wanted? I can't do that. If I search the word (e.g., process), the pdf only rely `... the find item was not found.'; however, the word just appear in front of me or it might hightlight words found with whole page if I chose to find other words such as `and'.

Thanks for reply.

jschiwal · 09-17-2007, 08:07 AM

The search works for other pdfs but not this one. You can't even cut & paste text. It comes up all garbled. Looking at the document properties, it uses 31 different fonts with custom encoding. Without knowing the encoding, I don't think that a pdf viewer program can equate a word you are entering with the characters in the document.

You could run ps2ascii to obtain some of the text. You won't get the diagrams and the equations won't be accurate.

shogun1234 · 09-19-2007, 01:11 PM

Quote:

Originally Posted by jschiwal

The search works for other pdfs but not this one. You can't even cut & paste text. It comes up all garbled. Looking at the document properties, it uses 31 different fonts with custom encoding. Without knowing the encoding, I don't think that a pdf viewer program can equate a word you are entering with the characters in the document.

You could run ps2ascii to obtain some of the text. You won't get the diagrams and the equations won't be accurate.

I see. It seems like to use ps2ascii is the best way to obtain the text.

Thanks your help. I appreciate it.