- Description: Adobe's PDF Document Viewer
- Summary: Working OK, but needs some fixes from Adobe for more compelling access
Using Adobe's Acrobat Reader
We are currently working on a script for Adobe's Acrobat Reader, and the initial version has been contributed to Orca v2.17.5. Note that there is still work to do, and we are very encouraged by Adobe's commitment to accessibility.
If you get Adobe Acrobat Reader directly from the Adobe site, you may find that the software will not launch. The symptoms: If launched from the Applications menu, it appears that nothing has happened. If launched from a terminal window, you receive a long series of syntax error messages. Should this occur, you will need to edit your /usr/bin/acroread file as described in this Ubuntu forum thread.
If you instead choose to install Acrobat Reader from your distribution's packages, be sure to install both acroread and acroread-plugins. Without the plugins package, you will not have the access to the "Reading" category in the "Edit->Preferences" dialog.
Note: If you use Ubuntu 7.04 and cannot locate the acroread package, you will likely need to edit your /etc/apt/sources.list to include the 'multiverse' repository. Then run apt-get update. Having done so, you should then be able to install both acroread and acroread-plugins.
If using Ubuntu 7.10, you will need to enable the medibuntu repository. Please enter the commands exactly as they are shown below in a console:
- First, add the repository. When prompted, type your root password.
sudo wget http://www.medibuntu.org/sources.list.d/gutsy.list -O /etc/apt/sources.list.d/medibuntu.list
- Next, import the GPG key and refresh your package list.
wget -q http://packages.medibuntu.org/medibuntu-key.gpg -O- | sudo apt-key add - && sudo apt-get update
- finally, install acroread and acroread-plugins using apt-get.
sudo apt-get install acroread acroread-plugins
Enabling Accessibility and Caret Navigation
Once you've installed Reader, you need to enable accessibility:
In the "Accessibility" category of the "Edit->Preferences" dialog, make sure "Always display the keyboard selection cursor" (Alt+d) is checked.
In the "Reading" category of the "Edit->Preferences" dialog, make sure "Enable document accessibility" (Alt+e) is checked.
Working with Untagged Documents
When you open an untagged document, you will be presented with a dialog box asking you to specify the reading order and the reading mode options. The default reading order is "Infer reading order from document." This setting works well for most documents, but not all of them.
The default reading mode will vary depending upon the length of the document: For short documents, the default is "Read the entire document"; for long documents it is "Read the currently visible pages only." Orca seems to provide much better access when "Read the entire document" is selected.
If you find that a document seems to be missing parts of the text, try changing the reading options. To do so, get into the Document menu and select "Change Accessibility Reading Options..." or use the shortcut Control+Shift+5.
The best page layout for reading a document with Orca is "single page". When "continuous" is chosen, moving focus to text as the page is scrolling can cause that text not to be read. You can change the page layout to "single page" by getting into the View menu and selecting the Page Layout submenu.
Regarding tables: As far as we know there's not currently a whole lot of keyboard support for tables in Acrobat Reader. This, by the way, is not an Orca thing; it is an Acrobat Reader thing.
Basically, there are the arrow keys. Their behavior is to move you first within the cell with focus, then to the next cell with data in it. Note that currently, Up and Down Arrow do *not* seem to move you vertically among cells -- at least that's been our experience with the tables we've tried. Instead, Up and Down seem to move you horizontally until you run out of cells on the current row.
Cut from Evince and Paste into GEdit
Francisco Javier Dorado Martínez has posted this trick for accessing PDF files via a combination of evine and gedit
I am using the following combination for reading pdf's. 1. Open evince and a pdf file 2. Press control + a for selecting all the text. 3. Press control + c for copying all selected. 4. Open gedit 5. Press control + v for pasting 6. Press control + home for go begining of document and read it. All these steps I have put in a python script and used the dogtail api (install python-dogtail package first) #!/usr/bin/python import os import sys from dogtail.procedural import * os.environ['LANG']='C' argumentos=' ' + sys.argv run('evince', argumentos) click('Edit', roleName='menu') click('Select All', roleName='menu item') click('Edit', roleName='menu') click('Copy', roleName='menu item') click('File', roleName='menu') click('Close', roleName='menu item') run("gedit") click('Edit', roleName='menu') click('Paste', roleName='menu item') #end of file. So when you run this script with a pdf filename as parameter, you get a gedit window with your text converted.
From Orca list member Leena Chourey (For Accessibility Team, CDAC Mumbai): usage: pdftohtml -s <file.pdf>
This enhancement in poppler library is done to make pdftohtml option more accessible and usable. New option '-s' for pdftohtml is defined to generate complex html with a single <file>-html.html including all pages. Steps to add poppler patch for option pdftohtml -s <file.pdf> 1.Get the latest source code of poppler using "git clone git://git.freedesktop.org/git/poppler/poppler" OR from http://poppler.freedesktop.org/ 2.Install it using following steps: o cd ~/poppler o ./autogen.sh o make o sudo make install 3.Download patch file from http://126.96.36.199/~access/0001-accesspdf_1Sept.patch and Apply patch by using command "git am <file.patch>". o Link for details: [[http://live.gnome.org/Git/Developers#Applying_Patches]] 4.Again build poppler 5.Run pdftohtml -s <file.pdf> . This generates two .html files (<file-html.html> & <files.html>) and one .png file corresponding to each page of pdf file. 6.Execute file-html.html file to access content of pdf file in html format. Troubleshooting: o If already have libpoppler4/5/6 then un-install it (check this from synaptic package manager), otherwise poppler installation may give error with following error message: "error while loading shared libraries: libpoppler.so.6: cannot open shared object file: No such file or directory". NOTE: note down the list of dependent software those you have to reinstall again from synaptic package manager. Give your feedback on firstname.lastname@example.org
From Orca list member Leena Chourey (For Accessibility Team, CDAC Mumbai):
Our Solution to access pdf is: We convert the PDF file into Complex HTML format using pdftohtml and open the same automatically in Firefox.
== HOW TO USE THIS SOFTWARE == You will first need to insatll the necessary software before you can use the utility. Follow the below steps (under Installation of the necessary software) === Installation of the necessary software === (Please note that while performing these steps if you get some error please refer to the troubleshooting section at the end of this document to get assistance.) 1.Download accesspdf.tar.gz from [[ http://188.8.131.52/~access/ ]] 2.Open a terminal window 3.Run command tar-zxvf accesspdf.tar.gz to extract required software. 4.Run command cd accesspdf 5.Run command tar-zxvf poppler-0.12.4.tar.gz to extract poppler-012.4 library. 6.Run command cd poppler-0.12.4 Then run the following commands: automake ./configure make sudo make install – This command will ask you the root password, please provide the same. Now poppler is installed and ready for use. 7.Run cd .. 8.Run command tar-zxvf orca-2.31.3pre.tar.gz to extract orca-2.31.3pre – This version of Orca contains our enhancements – we will soon provide this as a patch for use. 9.Run command cd orca-2.31.3pre and then run the following commands: automake ./configure make sudo make install - This will ask you the root password, please provide the same. Orca is now installed to use. 10.Run command cd .. 11.Change permissions for accesspdf.sh using command sudo chmod 777 accesspdf.sh . You have now installed all the necessary software and you can now run the same. Follow the steps in the How to Run section below to run the enhancement and access PDF document. === How to Run === 1.Run orca 2.Run accesspdf.sh - You can do this in the folowing 2 ways a) By selecting the accesspdf.sh and press enter. It will prompt asking what you want to do with the file - Run on Terminal, Display, Cancel or Run. Choose Run option. b) Else open a terminal and run the command ./accesspdf.sh 3.On running this command you will get a file browser to browse and open the PDF file you want to read. Select your corresponding PDF file and select OK. 4.The file will now open in Firefox and Orca will read it. Regular Firefox commands like sayall, whereamI, arrowkeys, etc. will work to move through the document. === Troubleshooting Instructions for Orca-2.31.3pre: === While ./configure you may get following dependencies. You can solve it by installing respective packages 1.configure: error: Your intltool is too old. You need intltool 0.40.0 or later. Solution: install intltool from synaptic package manager 2.NOTE: Mouse review requires wnck, which was not found. Solution : install pyathon-wnck from synaptic package manager Give your feedback on email@example.com
From Orca list member Jason White:
My solution to PDF files is: pdftotext -layout file.pdf or if it's in columns: pdftotext -raw file.pdf then read it with less or in an editor. pdfinfo file.pdf will tell me whether it's tagged; they almost always aren't. The most promising project heading in the right direction appears to be http://www.gnupdf.org/ which aims to implement the entirety of the ISO PDF standard, based in turn on PDF 1.7.
See also the Acroread Bugs page for a list of Acroread bugs that prevent Orca from doing as good a job as it could do.
The information on this page and the other Orca-related pages on this site are distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.