- From: Geert Josten <geert.josten@dayon.nl>
- Date: Tue, 29 Jan 2013 09:35:05 +0100
- To: Norman Walsh <ndw@nwalsh.com>, XProc Dev <xproc-dev@w3.org>
Zip and probably other formats allow selecting a compression level. Is there real difference between archive and compress if you select uncompressed? About guessing format: I'd suggest allow the developer to specify a mime-type, and allow implementations to guess the mime-type in any way they feel appropriate. There are still quite a number of systems that use extensions to do so, most importantly: MarkLogic.. ;-) Cheers > -----Oorspronkelijk bericht----- > Van: Norman Walsh [mailto:ndw@nwalsh.com] > Verzonden: maandag 28 januari 2013 17:53 > Aan: XProc Dev > Onderwerp: Re: Proposed steps: pxp:gunzip and pxp:gzip > > Zearin <zearin@gonk.net> writes: > > I dislike separate steps for the different archive formats. > > > > I should be able to specify that I want to ARCHIVE or UNARCHIVE a > > file, and let the processor handle the details. Both archival and > > unarchival can be determined from the file extension. > [...] > > What say you? > > So your preference would be a pxp:archive step that could perform gzip, > compress, bzip2, etc. and a pxp:unarchive step that did the reverse? > > I suppose that's plausible. I'm not sure I like extension peeking though. > For the archive step at least, I think I'd want an attribute to control > the archive format. > > Does this beg the question of whether the pxp:zip and pxp:unzip steps > should also be combined? I think that's probably not a good idea, as > multi-file archives are quite different from compression. > > Maybe the steps should be pxp:compress and pxp:uncompress as archive > does sort of suggest zip/tar/rar etc. > > Be seeing you, > norm > > -- > Norman Walsh > Lead Engineer > MarkLogic Corporation > Phone: +1 512 761 6676 > www.marklogic.com
Received on Tuesday, 29 January 2013 08:35:34 UTC