Comments on "File and operating system steps for XProc" of 2 August 2013

1. pf:copy

It worries me that by deliberately starting simple, the WG will end up
having to make piecemeal additions as users discover more aspects of
'cp' (or even the MS-DOS 'copy' command) that they'd just like added
to pf:copy to make that user's work that little bit easier.

1.1 What happens when @href refers to a directory, e.g., when a
    user wants to copy a directory of images?

1.2 Is it always the case that pf:copy overwrites an existing @target
    file (or file in @target directory)?

1.3 Should there be an option to 'force' the copy to succeed (e.g., in
    the manner of 'cp -f'), especially since there is currently no way
    to set file or directory permissions?

1.4 What happens on filesystems that have symbolic links when @target
    refers to an existing symbolic link?  Does it overwrite the target
    of the link or the link itself?

1.5 What is the timestamp of the copied file?


2. pf:touch

The definition does not make it clear that an empty file will be
created if the file named in @href does not exist.


3. Editorial

3.1 The definitions of pf:move, pf:tempfile, and pf:touch use an
    undefined 'pxf' namespace prefix.

3.2 In section 3.2, pf:delete, in "the recursive option has the value
    false", 'false' could be in monospace text.  (XProc 1.0 is
    inconsistent in whether 'false' is in plain text, quoted, or in
    monospace text.)

Regards,


Tony Graham                                   tgraham@mentea.net
Consultant                                 http://www.mentea.net
Mentea       13 Kelly's Bay Beach, Skerries, Co. Dublin, Ireland
 --  --  --  --  --  --  --  --  --  --  --  --  --  --  --  --
    XML, XSL-FO and XSLT consulting, training and programming

Received on Tuesday, 27 August 2013 14:09:28 UTC