[PD] [PD-announce] deken v0.3.0: new fileformat for library packages

Roman Haefeli reduzent at gmail.com
Thu Mar 8 21:44:36 CET 2018


At the risk of not being helpful to the issue by adding another
opinion, I disagree. Or rather: I do not really follow you. I don't see
any problem with deken installing into a hidden folder. It's a pretty
common thing on all OSs. Just a have a look at the contents of
~/Library, %AppData%\, ~/.local/ .

You say: "Newbies have troubles finding the folder"

I say: "Newbies shouldn't have to know about it all"

Ideally (in my notion of how things would ideally be), there is nothing
to be done manually in that folder. Ideally, deken is a management
interface  for the user to install, enable, disable, remove libraries.
It doesn't matter where it does all at. You would install a library and
this is enables you to simply do [declare -std[path|lib] yourlib] to
use it in your patch. You shouldn't have to know more. It looks like
we're getting closer to that with all the efforts done recently.

Roman 

On Don, 2018-03-08 at 14:27 -0300, Alexandre Torres Porres wrote:
> > - set installation path (suggesting Pd's default search paths, and
> > offering to create those that are missing)
> I wouldn't know if, officially, there are such things as "Pd's
> default search paths". But then it's clear you mean "standard paths",
> instead of "default search paths", right? Then again, this raises a
> few issues. Cause, well, under a recent discussion, it was considered
> that only the "extra" folder was supposed to be the standard folder,
> so there are not supposed to be other "standard paths". In other
> words, it ain't plural, just a singular standard path option, and one
> that comes always with Pd and doesn't need to be created. 
> 
> So yeah, what had the need to be created was one folder like that
> weird and hidden mac folder (~/Library/Pd)... but access to it would
> still be difficult (as it is invisible). Hence, we have a new folder
> now (~/Documents/Pd) but that is not a "standard path", just a
> "regular" path. And I see now "deken" is offering me to create such
> folders like ~/Library/Pd, after a long discussion that led to the
> idea we shouldn't use them no more. See this: https://github.com/pure
> -data/pure-data/pull/139
> 
> I know I keep bringing this up here and there, but I think it's a
> very important detail that can cause many confusions and there is
> still a lot of inconclusiveness out there, and we're not doing much
> about it and letting the confusion grow. Now, I was also one of the
> people who was requesting such a feature several times in Deken, it's
> all I ever wanted, but the game has really changed because of such
> recent discussions. 
> 
> So now we can finally easily create this folder, but it is still
> invisible and inaccessible to newbies, so not much advantage there.
> Cause if you need to know how to navigate to that folder, you also
> know how to create it. So this partially fixes the issue. What was
> needed there was create it and easily navigate to it, but instead, it
> was just easier to go to a new folder instead, like ~/Documents/Pd -
> as we did!
> 
> So we have this new standard now that was designed exactly to fix
> this issue. So this new feature is pointless now, as a new solution
> (using ~/Documents/Pd) was made, so it is trying to fix an issue (but
> not in fully fixing it) that has already been fixed and has been in
> use since 0.48-0! But more importantly, this extra folder has been
> explicitly declared as a non intentional extra "standard path". See
> how much confusion this promotes?
> 
> So, given all that, I'm even suggesting a Pull Request to clear this
> up for good, it's here: https://github.com/pure-data/pure-data/pull/1
> 83
> 
> See also: https://github.com/pure-data/pure-data/issues/184 which
> discusses the behaviour of [declare] and how we could just use "-
> path" and forget about "-stdpath" 
> 
> And leaving "~/Library/Pd" aside, the other extra standard path
> option here on the mac: "/Library/Pd" doesn't even get created with
> the new deken plugin, I figure because it may require admin powers,
> so it doesn't really help either.
> 
> Now, I don't know how the plugin works, but I may assume it gets
> these paths from Pd, so maybe if my Pull Request gets accepted, no
> change needs to be made in the Deken plugin and it'll just stop
> showing them...
> 
> Anyway, what I see from this update with this issue is that we have
> added more unnecessary complexity. Things were cleaner and simpler
> before. But well, I don't consider this a "deken" issue, but a deeper
> Pd issue that is popping up in deken and other different places. 
> 
> That is to say that I'm really happy with the new improvements in
> deken, congratulations, it is quite exciting!
> 
> cheers
> 
> 
> 
> 2018-03-08 13:09 GMT-03:00 IOhannes m zmoelnig <zmoelnig at iem.at>:
> > we are proud to announce a major overhaul of the deken fileformat.
> > 
> > TL;DR to download new deken packages, you will need at least
> > version
> > v0.3.0 of the deken-plugin (Pd-0.48-1 includes deken-v0.2.4).
> > When searching externals with an old (incompatible) version of the
> > plugin, you will always find a suggestion to install the new plugin
> > with
> > something like:
> > 
> > > deken-plugin/0.3.0
> > >  (Deken externals downloader - REQUIRED FOR NEW PACKAGES)
> > 
> > once you have installed the new version (and/or Pd includes a
> > compatible
> > version), the nagging will stop :-)
> > 
> > 
> > # fileformat
> > 
> > the new fileformat fixes a number of issues we had with the
> > existing one.
> > things that are better:
> > - cool filename extension '.dek'
> > - consistent archival options for all platforms (ZIP!)
> > - proper detection of the version string for all libraries (even if
> > the
> > library has a weird name like "a-v12"
> > - support for double-precision externals (once Pd supports that!)
> > - extensible
> > 
> > 
> > # Pd-integration
> > the updated "deken-plugin" has a number of nifty features, you
> > don't
> > want to miss:
> > - support for the new fileformat
> > - support for the old (legacy) fileformat
> > - (optionally) uninstall libraries before re-installing them
> > - (optionally) show a README that is included in the package on
> > successful installation
> > - set installation path (suggesting Pd's default search paths, and
> > offering to create those that are missing)
> > - (optionally) hides the search-results for incompatible platforms
> > - (optionally) overrides the "compatible" Pd-architecture
> > - nice preferences
> > - bugfixes
> > - much more
> > 
> > 
> > # Developer tools
> > if you are a developer of libraries, we have also updated our
> > 'deken'
> > cmdline tool.
> > things you always wanted to have and which are now free:
> > - support for the new fileformat (default)
> > - support for the old (legacy) fileformat
> > - better detection of binary architectures
> > - better detection of included sources
> > - (experimental) support for double-precision externals
> > - pre-compiled & self-contained Windows binaries
> > - bugfixes
> > - much more
> > 
> > 
> > if you want to know more, see
> > - http://puredata.info/downloads/deken/releases/0.3.0/
> > - https://lists.puredata.info/pipermail/pd-dev/2018-02/021513.html
> > - https://github.com/pure-data/deken/issues/161
> > 
> > 
> > i have submitted also submitted a Pull-Request to include the new
> > deken-plugin into Pd proper:
> > - https://github.com/pure-data/pure-data/pull/320
> > 
> > and another Pull-Request for Pd that is required to make the
> > double-precision detection work:
> > - https://github.com/pure-data/pure-data/pull/300
> > 
> > 
> > happy patching.
> > fgm,asdr
> > IOhannes
> > 
> > 
> > _______________________________________________
> > Pd-announce mailing list
> > Pd-announce at lists.iem.at
> > https://lists.puredata.info/listinfo/pd-announce
> > 
> > _______________________________________________
> > Pd-list at lists.iem.at mailing list
> > UNSUBSCRIBE and account-management -> https://lists.puredata.info/l
> > istinfo/pd-list
> > 
> _______________________________________________
> Pd-list at lists.iem.at mailing list
> UNSUBSCRIBE and account-management -> https://lists.puredata.info/lis
> tinfo/pd-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 473 bytes
Desc: This is a digitally signed message part
URL: <http://lists.puredata.info/pipermail/pd-list/attachments/20180308/a550623a/attachment.sig>


More information about the Pd-list mailing list