Black screen

[D'Lanor]

Ok, I figured out why the fni files that PyPRP writes are wrong.

Current uruprp_export.py code:

Code: Select all

    #generate the funny file
    print "Writing %s" %(agename + ".fni")
    initxt=alcFindBlenderText("init")
    fnitxt=""
    for line in initxt.asLines():
        fnitxt=fnitxt + line + "\n"
    if fnitxt!="":
        age.setInit(fnitxt)


It is the old Carriage return/Linefeed on Windows thing again. So we need to add a \r (Carriage return). The function which writes the age file already adds this btw.

My fix:

Code: Select all

    #generate the funny file
    print "Writing %s" %(agename + ".fni")
    initxt=alcFindBlenderText("init")
    fnitxt=""
    for line in initxt.asLines():
        fnitxt=fnitxt + line + "\r\n"
    fnitxt=fnitxt[:-2]
    if fnitxt!="":
        age.setInit(fnitxt)


I also stripped the stuff that is added after the last line. This will make sure that no fni file is written when there is no fni info, instead of a fni file with an empty line as it does now. And which may be causing the problem reported here: viewtopic.php?f=10&t=1402

Edit: added to the SVN contrib section now.

[D'Lanor]

Bump. Could one of the developers evaluate my contribution and merge it to the trunk?

[Robert The Rebuilder]

You might have better luck re-posting your request in the PYPRP forum, D'Lanor.

[D'Lanor]

Maybe, but I shouldn't have to post on the forum at all if it is on the SVN. You cannot miss a revision there if you use a proper SVN client.

[Paradox]

Ack. Sorry D'Lanor.
Usually I'm the one that looks over small fixes and merges them, or pokes other people to check them and merge them. I assure you that your fix is correct and that it will be merged as soon as I recover my PyPRP repo from my corrupt drive (hopefully sometime this evening).

[D'Lanor]

Thanks.

[D'Lanor]

Paradox, you have not implemented my second fix for the "empty" fni file: fnitxt=fnitxt[:-2]

This strips the carriage return and linefeed characters which are added after the last line. If you do not strip those characters then the next condition if fnitxt!="": will never be true and a fni file will be written even if there is nothing in the init text. This fni file only contains a new line and it will cause a warning message to pop up when loading the age.

I have re-submitted this fix to the SVN contrib.

Edit: I have no idea why you removed this, but there is no need for a fni file to end with a new line. If you look at the Cyan fni files you will see that not all of them have a new line at the end.

[Nadnerb]

People really shouldn't be using empty .fni files.. it will just cause the settings from the previous age to carry over, and that is a Bad Thingâ„¢ Besides, anyone who has created a new book with a recent version of the plugin will not have an empty .fni file, because the new book script automatically creates it for them.

[D'Lanor]

Actually not every age has a fni file. There are Cyan ages which do not have one. Ages inherit fni settings from the previously loaded fni file. So if an age can only be reached from another age which already has the correct settings then you do not need a fni file.

Now if I wanted to do some nitpicking I could mention that it does not make sense to set an if-statement which is never true. And that it is a bad habit not to clean up string appendages after for-statements. But then again, the bug is not that serious. And I can still implement the fix in my local copy.