New paper layout

Although the new paper layout looks lovely, it is causing me problems parsing, and I'll explain why...

My original idea was to parse the paper directly, from the pdf, or a file exported by Adobe Reader. However, due to the non-linear nature of the layout, this proved impossible. So, I took the path of copying the text manually out of the paper, creating a format of text file which then can be automatically parsed into the wiki format we all now know and love.

This was great, and worked. But, the new format is missing the markers I used to start new stories. Also, the name of the place where the story happens doesn't seem to be text, as I can select all the text on any given page, save the placenames. I even exported the new format to text, and the placenames just don't come out.

So, for older format papers, LGDRParser v1.0 works fine. But, I will need to produce a new version of LGDRParser for the new format, after I've worked out a way of extracting all the information I need. New format papers will therefore not appear as I had promised, and I'll keep the community posted when I manage to work a new version out.

I think the new paper lay-out is visually better and I remain a big fan of it - the main problem I am having (and not having the greatest eyesite doesnt help!) is with the paper displayed at 100% (it downloads at about 75%) it seems to be slightly too large for my screen so as I scroll down the columns it jumps to the next page....when I want to read the top of the next column on the previous page.

A bit annoying. Im sure there must be a way around it, I just havent figured it out yet.

So far I'm liking it but can understand the problems Count is having could be a pain.

Must be honest, it never occurred to me that the format would give tek those sort of problems! Afraid I am no teckkie so can't help but if you needed someone to do some sort of grunt work (even if it meant retyping parts of a newspaper into a word or text document so you can do clever things with it) do shout!

Sadly, I've just confirmed that the placenames are indeed images. This would explain why I can't select them. In conjunction with the story markers being missing, this is going to be extremely difficult to code for.

Therefore, it seems I'll have to do things a little more manually than I already was. I'll have to type in the placenames myself, and put the new story markers in as needed. This way, LGDRParser will work, at the additional cost of a few minutes per paper. I guess this is not so bad.

Just wish Richard could produce something that I can throw at an application and parse into wiki markup.