Michael Heiming's bits of Mon, 10 Sep 2001 translated to:
>you're right, I'm using pdftohtml, changed the patch
>xpdf-0.90-fefe-diff (tried both available) so it applied to the
>pdftohtml source, which includes much of the xpdf source. 'make all'
>ran fine until (see below), seems to be a problem with the md5 header
>file, it came with the patch, but sadly it doesn't work.
[snip]
>-I./../goo -I. -o pdftohtml Array.o Catalog.o Dict.o Error.o
>FontEncoding.o FontFile.o Gfx.o GfxFont.o GfxState.o Lexer.o Link.o
>Object.o OutputDev.o Page.o Params.o Parser.o PDFDoc.o HtmlOutputDev.o
>Stream.o XRef.o HtmlFonts.o HtmlLinks.o pdftohtml.o \
> -L../goo -lGoo -lm
>Parser.o: In function `Parser::getEncryptedObj(Object *, int, int)':
>/usr/local/src/pdftohtml_0_31/xpdf/Parser.cc:169: undefined reference
>to `MD5::MD5(void)'
>/usr/local/src/pdftohtml_0_31/xpdf/Parser.cc:222: undefined reference
>to `MD5::update(unsigned char *, unsigned int)'
[...]
Hi - The problem is that new files generated by the patch (e.g. md5.h,
md5.cc, rc4.h, rc4.cc) are not accounted for in the Makefile included in
the xpdf directory. Thus the undefined references. I just built the two
object files by hand (e.g. g++ -Wall -c md5.cc rc4.cc) and added them to
the link line in the Makefile. I did use the resulting pdftohtml to
convert an 800 page encrypted document using the complex html option.
Jim
_______________________________________________
htdig-general mailing list <[EMAIL PROTECTED]>
To unsubscribe, send a message to <[EMAIL PROTECTED]> with a
subject of unsubscribe
FAQ: http://htdig.sourceforge.net/FAQ.html
