> -----Original Message-----
> From: Anne-Gert Bultena (ELN)
> [mailto:[EMAIL PROTECTED]]
> Sent: Monday, July 08, 2002 9:39 AM
> To: 'Plucker Development List'
> Subject: new options? - usage problems
>
>
> I'm just using plucker for a couple of weeks now (and 1.2 for
> just a week or so). I had some questions though. Hopefully,
> they are as easy to answer as the one about blue anchors.
> (PS: I did RTFM, but couldn't find M :-) ).
>
> 1) Has the documentation been updated for version 1.2? (or
> anything after 1.1.13, which is the last I got).
>
> 2) I'm using windows and I've tried the different options for
> image_parser in the plucker.ini. Most of them don't work,
> except pil2 and windowspil.
> But
>
> 2a) windowspil doesn't make a link from small pictures (the
> maxwidth x maxheight ones) to the larger versions (the
> alt-maxwidth x alt-maxheight ones). I've tried to figure out
> the source-code (I'm new to python, so that was fun) and I
> found that some image parser are derived from ImageParser,
> which seems to be the baseclass, but others aren't
> class ImageParser:
> class ImageMagickImageParser:
> class NetPBMImageParser:
> class NewNetPBMImageParser(ImageParser):
> class PythonImagingLibraryParser:
> class NewPythonImagingLibraryParser(ImageParser):
> class WindowsImageParser:
> class WindowsPILImageParser:
> Note that pil2 (NewPythonImagingLibraryParser) is using ImageParser.
> I also found that at the bottom of
> ImageParser::get_plucker_doc(), there is a piece of code that
> seems to gather the bigger picture by calling
> _related_images(). I also saw that these images are added by
> a call to PluckerImageDocument::add_related_image().
> Then I looked at where add_related_image() is used and found
> two places: it's definition and it's use at the bottom of
> ImageParser::get_plucker_doc(). Now you probably see the
> relevance how I started: windowspil (WindowsPILImageParser)
> isn't derived from ImageParser AND it doesn't generate those
> related images.
>
> 2b) pil2 does a call to add_related_image(), but has another
> problem: it crashes when I use bpp=16.
>
> C:\>"c:\program files\plucker\python\python" "c:\program
> files\plucker\pyplucker
> \spider.py" "--pluckerhome=c:\program files\plucker"
> --home-url=file:c:/x.html -
> -doc-file=c:\x --bpp=16
> Pluckerdir is 'c:\program files\plucker'...
> ---- 0 collected, 1 to do ----
> Processing file:C:/x.html...
> Retrieved ok.
> Parsed ok; 1 image.
> ---- 1 collected, 1 to do ----
> Processing file:C:\x.jpg...
> Retrieved ok.
> Error: Unknown error parsing document file:C:\x.jpg:
> Traceback (innermost last):
> File "C:\Program Files\plucker\PyPlucker\Parser.py", line
> 45, in generic_parse
> r
> return parsed.get_plucker_doc ()
> File "C:\Program Files\plucker\PyPlucker\ImageParser.py",
> line 218, in get_plu
> cker_doc
> raise ImageSize ("Image data too large (%d bytes) for a
> Plucker image record
> "
> ImageSize: Image data too large (209024 bytes) for a Plucker
> image record (max 6
> 1440 bytes) when plucked at 500x209x16! Scale it down.
> Parsed ok.
> ---- all 3 pages retrieved and parsed ----
>
> Writing out collected data...
> Writing document 'x' to file c:\x.pdb
> Traceback (innermost last):
> File "c:\program files\plucker\pyplucker\spider.py", line 1512, in ?
> sys.exit(realmain())
> File "c:\program files\plucker\pyplucker\spider.py", line
> 1505, in realmain
> retval = main (config, exclusion_lists)
> File "c:\program files\plucker\pyplucker\spider.py", line
> 1041, in main
> mapping = writer.write (verbose=verbosity, alias_list=alias_list)
> File "C:\Program Files\plucker\PyPlucker\Writer.py", line
> 518, in write
> result = Writer.write (self, verbose, alias_list=alias_list)
> File "C:\Program Files\plucker\PyPlucker\Writer.py", line
> 310, in write
> self._mapper = Mapper(self._collection, alias_list.as_dict())
> File "C:\Program Files\plucker\PyPlucker\Writer.py", line
> 102, in __init__
> self._get_id_for_doc(doc)
> File "C:\Program Files\plucker\PyPlucker\Writer.py", line
> 112, in _get_id_for_
> doc
> id = self._url_to_id_mapping.get(doc.get_url())
> AttributeError: 'None' object has no attribute 'get_url'
>
> I think _get_id_for_doc should return if doc==None, which
> would be a bit more graceful.
>
> I hope anyone can help me here.
>
> agb
