[Scribus] PDF Export -- Searchable Text?

[Gregory Pittman]

Plinnell wrote:
> On Thursday 24 August 2006 04:46, Jennifer Baughman wrote:
>   
>> Would someone be able to tell me if there is a plan to implement
>> searchable text in exported PDFs? There are list mails from 2004
>> indicating that's not possible with the 1.2.x builds, but I'm
>> wondering if it has been (or is going to be) implemented in the
>> 1.3.3.x or 1.3.4+ versions. If not, is there a workaround?
>>
>> Thank you,
>>
>> Jennifer Baughman
>> _______________________________________________
>> Scribus mailing list
>> Scribus at nashi.altmuehlnet.de
>> http://nashi.altmuehlnet.de/mailman/listinfo/scribus
>>     
>
> 1.3.x is no different and probably will remain that way until we redo 
> certain parts of the PDF export.
>
> Why is this so ?
>
> The way text in PDF is laid by Scribus is not efficient  - we know 
> this.  However, this gives us extremely precise glyph location, so 
> the layout matches on screen.
>
> Workarounds ? None that I know of at the moment.
>   
You can see the practical difficulty by trying File > Save as Text in 
Adobe Reader. You will likely get all the text in terms of not missing 
(many) letters, but there will be odd spaces or missing spaces that 
create a large editing task to fix so that you could then suitably use a 
search program.
A PDF has lost all representation of characters as characters, but 
instead has the information that a printer can use to draw the letters 
and everything else on the page. In order to convert a PDF to text, it's 
something like OCR -- you have to turn the visual representation of 
letters back into letters.

Greg