I like it, too. More below...

On 03.04.2012 19:55:40 Andreas Lehmkuehler wrote:
> Hi,
> 
> I like your proposal, comments see below ...
> 
> Am 03.04.2012 02:34, schrieb Tharaka Nayanajith Wijebandara:
> > Hi,
> >
> > I have completed some parts of the the GSoC proposal and I need some
> > feedback from the community about the that. Here a I used swing approach
> > since I have some doubts about RCP. Anyway I also understand RCP will
> > easy approach rather than this and if someone going to help me on that I'm
> > still ready to go in that way. Still I have to prepare 'Project Timeline
> > and Deliverables' part and I will send those part soon.
> >
> > *Project Title*
> >
> > *=============*
> >
> >
> >
> > Implement PDF object inspector and enhance features of PDFReader GUI
> >
> >
> >
> >
> >
> > *Abstract*
> >
> > *========*
> >
> >
> >
> > PDFReader is small GUI application which comes with Apache PDFBox pure java
> > PDF library. Currently it is only capable of rendering PDFs. This project
> > is to redesign PDFReader GUI with plugin framework and implement some
> > developer oriented features for it. Primary objectives of project are as
> > follows.
> >
> >
> >
> > (1). Design a plugin framework for the PDFReader.
> +1

+1

> > (2). Implement PDF basic viewing features. (Zooming, rotating, page
> > thumbnail viewer)
> +1

+1

> > (3). Implement a PDF object inspector to display PDF object properties.
> Could be complicated ...

Possible, so I suggest to stick to the easy targets first. Prioritize.
Getting information about an XObject should be fairly easy and allows to
create the basic infrastructure (finding the bounding box, painting a
selection rectangle etc.) to support additional objects such as text
passages. However, it may be a good idea to switch order/priority between
point 3 and 4.

> > (4). Implement a type-writer feature to add/remove text from a content
> > stream.
> IMO, you should consider to drop this idea. The search/replace feature was 
> discussed many times on the mailing list. As PDF is not a text oriented 
> format, 
> it is even difficult to find text but much more complicated to replace it.

How about just implementing the "add text" feature? That should be
relatively easy. Add a new optional content group to the page and add
all text as part of that. That makes this a cool tool to fill out forms
that were scanned, or that are not AcroForms. I fully agree that the
replace feature is going to be tough, relatively speaking.

I wouldn't put the bar too high for this project. Get some low-hanging
fruit that adds some cool and useful functionality to the GUI. More can
always be added later.

More ideas for low-hanging fruit (as additional tasks if you end up with
spare time after the main objectives above):
- Allow the "Save As..." menu item not only save a modified PDF, but
also bitmaps (PNG, JPEG), plain text and HTML (i.e. from text extraction).
That would make a plug-in for each output format, so we could add others
later.
- Add a "Print..." menu item.
- Integrate preflight as a plug-in into the application.
- Split, concatenate, page reorder, page rotation (not just viewing but
modifying the PDF), overlay/underlay.
- Menu item: "Create PDF from image(s)" and "Add page from image".
- A dialog/view listing the fonts in the PDF.
- A dialog/view for the basic PDF properties like in Acrobat's
Properties Dialog.
- A dialog/view listing all images with their properties and effective
resolution in the PDF.
- I could go on and on... :-) I'm seeing a serious itch going to catch
me once you've set up a good base for that GUI. I'm so looking forward
to this!

Anyway, points 1 and 2 will be the most important tasks because they
provide the basis for easily adding additional functionality.

> <SNIP>
> 
> 
> BR
> Andreas Lehmkühler




Jeremias Maerki

Reply via email to