Is search in PDF documents possible?

[tello]

Hi everybody;


Nowadays ,after installing our OJS system successfully, we are trying to create our back issues. My question is that ; Our system has no capability to make search in PDF documents. Is there any way to make it possible.In forum I found somethings but I did not understand anything. It is very important for us because we will not produce HTML forms of our back issues.

Thanks in advance, tello

[asmecher]

Hi tello,

OJS uses external tools to extract text from PDF files for searching. In OJS's config.inc.php file, you need to specify where OJS can find these tools. For example, the stock configuration file includes the lines:

Code: Select all

; PDF
; index[application/pdf] = "/usr/bin/pstotext %s"
; index[application/pdf] = "/usr/bin/pdftotext %s -"

These lines are commented out by default -- if your server has the tool /usr/bin/pdftotext installed, un-comment that line:

Code: Select all

; PDF
; index[application/pdf] = "/usr/bin/pstotext %s"
index[application/pdf] = "/usr/bin/pdftotext %s -"

Once you've changed the configuration file, you can rebuild the text index by running "php tools/rebuildSearchIndex.php" from the command line.

Regards,
Alec Smecher
Open Journal Systems Team

[tello]

Thanks Alec ;


I have tried your what you told but it did not work. I think as you sad

These lines are commented out by default -- if your server has the tool /usr/bin/pdftotext installed, un-comment that line:

maybe our server doesn't have this option installed how can I be sure whether it is installed or not in my server.

Also I told you that I tried but I think only uncommenting the lines is nt enough because in my server the path is also different. What can we do further?.

Thanks in advance,tello

[asmecher]

Hi tello,

You're probably best off asking a server administrator if and where these tools are installed. OJS will support any command-line tool that converts PDFs into text, such as pdftotext (part of the xpdf toolset) and ps2ascii (part of the ghostscript toolset). One or both of these are often installed on *NIX servers, but the location and details depend on the operating system. If you're using a Windows server, you'll probably have to install one of these manually. I'd suggest getting the xpdf toolset (see http://www.foolabs.com/xpdf/download.html).

Regards,
Alec Smecher
Open Journal Systems Team