We are pleased to introduce you the new version of Relational pipes. It
brings several important features:
*XML input and output*: offers lossless bidirectional conversion between
relational format and text. Can be used for manual modifications or
reviews. XML form of relational data can be generated or consumed by
XQuery or XSLT processors or other XML tools.
*CSV input and output*: CSV is a standard format defined in RFC 4180. It
is very simple, lacks data types and can contain only a single relation.
Despite this, it is very useful when integrating Relational pipes with
other tools. At the end of our pipeline, we can put relpipe-out-csv,
feed the relation to e.g. GNU R and compute some statistics or generate
some nice graphs.
*Filesystem input*: this input filter reads file paths (usually
generated by the find command) and fetches metadata of given files.
Supports various name/path formats, size, owner, symlinks… and extended
attributes (xattr). The output can be customized using CLI arguments.
This tool can be used to catalogue disk contents, search files or to
collect data for some statistics like: How much space does files in
particular formats occupy? or From which domains I download files most
often? Other use case is: save the output of relpipe-in-filesystem in a
file and use it as an index of our files (even on offline media).
See the examples page for details.
Installation instructions are here: