[Docutils-users] More include woes

[Brett g P. <bgp...@ac...>]

I've stumbled on one thing that looks like a real problem, and one th=
at's
just making me scratch my head.

1) Directory problems:

I am assembling a master document that pulls in some source from a
subdirectory, like:

/main_dir
    masterdoc.txt
    ch1.txt
    <<etc.>>
    /reference_dir
        refch1.txt
        refch2.txt
        <<etc.>>

The reference chapter begins with some introductory text, then brings=
 in
each of the functions that are being documented. My first pass at thi=
s had
that intro file in the main_dir directory, and then included the ref =
files
like:

<<ref_intro.txt>>

text here, etc.

=2E. include:: reference_dir/refch1.txt

=2E. include:: reference_dir/refch2.txt

Then I got the bright idea that the ref_intro doc should live down in=
 the
reference_dir directory, so I moved it, and tried patching the paths =
so it
was including refch*.txt as if they were in the current directory.

That didn't work (error was that there was no file `refch1.txt') so I
restored the paths as if the current working directory at the context=
 of the
include was still the top-level /main_dir -- then the error was even =
odder,
that there was no file `reference_dir/reference_dir/refch1.txt'

On the one hand, directories are doubly-expanded, but on the other ha=
nd they
dont' seem to be exapnded at all.

This isn't a killer for me (because it works fine with the intro text=
 in the
/main_dir directory), but I wanted to report it nonetheless.

2) Problems including a file.

I have a file that works fine when rendered on its own. If I copy its=
 body
into my master document at the point where I would like to .. include=
:: it,
the resulting master doc renders fine. However, any attempt to includ=
e this
file in another (even an otherwise empty file that just includes it) =
gives
me this error:
Traceback (most recent call last):
  File "c:\bin\docutils\tools\html.py", line 25, in ?
    publish_cmdline(writer_name=3D'html', description=3Ddescription)
  File "C:\bin\Python22\Lib\site-packages\docutils\core.py", line 230=
, in
publish_cmdline
    pub.publish(argv, usage, description, settings_spec, settings_ove=
rrides)
  File "C:\bin\Python22\Lib\site-packages\docutils\core.py", line 173=
, in
publish
    output =3D self.writer.write(document, self.destination)
  File "C:\bin\Python22\Lib\site-packages\docutils\writers\__init__.p=
y",
line 52
, in write
    output =3D self.destination.write(self.output)
  File "C:\bin\Python22\Lib\site-packages\docutils\io.py", line 208, =
in
write
    output =3D self.encode(data)
  File "C:\bin\Python22\Lib\site-packages\docutils\io.py", line 126, =
in
encode
    return data.encode(self.settings.output_encoding or '')
UnicodeError: ASCII decoding error: ordinal not in range(128)

I've tried narrowing down the problem by whittling the problematic fi=
le
down, but can't reproduce the problem or locate the source. I'm hopin=
g that
the traceback will give you a clue as to what the problem is.

thx



--
//  Today's Oblique Strategy (=A9 Brian Eno/Peter Schmidt):
//  Do nothing for as long as possible
//  Brett g Porter * BgP...@ac...
//  http://mywebpages.comcast.net/bgporter/