The rules for the extension are the same as Perl's (see perldoc perlrun): if the extension includes no ``*'' then it is appended to the original file name, If the extension does contain one or more ``*'' characters, then each ``*'' is replaced with the current filename.
If this option is used the XML declaration is updated (and created if there was none).
Make sure that the encoding is supported by the parser you use if you want to be able to process the pretty_printed file (XML::Parser does not support 'latin1' for example, you have to use 'iso-8859-1')
This prevents a bug (see BUGS) but uses more memory
xml_pp foo.xml > foo_pp.xml # pretty print foo.xml xml_pp < foo.xml > foo_pp.xml # pretty print from standard input
xml_pp -v -i.bak *.xml # pretty print .xml files, with backups xml_pp -v -i'orig_*' *.xml # backups are named orig_<filename>
xml_pp -i -p pre foo.xhtml # preserve spaces in pre tags
xml_pp -i.bak -p 'pre code' foo.xml # preserve spaces in pre and code tags xml_pp -i.bak -p pre -p code foo.xml # same
xml_pp -i -s record mydb_export.xml # pretty print using the record style
xml_pp -e utf8 -i foo.xml # output will be in utf8 xml_pp -e iso-8859-1 -i foo.xml # output will be in iso-8859-1
xml_pp -v -i.bak -f lof # pretty print in place files from lof
xml_pp -- -i.xml # pretty print the -i.xml file
xml_pp -l foo.xml # loads the entire file in memory # before pretty printing it
xml_pp -h # display help
will be output as
Using the "-l" option solves this bug (but uses more memory)