[Scottish] Script hates whitespace

[Keith Wyse]

Thanks Claudio, that's working great.
I'll have to read up some more on that and add it to my toolkit.

Thanks for your help!
Keith


On 19 July 2013 11:36, Claudio Calvelli <scotlug at intercal.org.uk> wrote:

> Keith Wyse writes:
> > Hi all;
> > I'm having some trouble with a little script I'm writing - it hates
> > whitespace.
> > I have a bunch of books in .pdf format, but I can read them better in
> .epub
> > because of page resizing and font resizing. I wanted to use Calibre's
> > command line ebook-convert to modify them.
> > So the main line of the script is;
> >
> > find . -name "*.pdf" -execdir sh -c 'ebook-convert {} "`basename {}
> > .pdf`.epub"' \;
>
> That results in the command being expanded several times by various shells,
> not a good combination.  You could try:
>
> find . -name '*.pdf' -print0 | \
> while read -r -d '' f; do pushd "`dirname "$f"`"; n="`basename "$f"`"; \
> ebook-convert "$n" "${n%.pdf}.epub"; popd; done
>
> which is naturally self-explanatory but in case it isn't, you have
> your shell reading each name (NUL-delimited, using "-print0" to "find"
> and "-d ''" to the shell's "read" builtin) and then use the shell's own
> string manipulation to remove the ".pdf".  I don't have ebook-convert,
> but if I replace it with "echo" I get the list I would expect.
>
> Or you could use something other than a shell one-liner :-)
>
> HTH
> C
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.lug.org.uk/pipermail/scottish/attachments/20130719/7fffeafa/attachment.html>