Experimental in the sense that there is no use found for them at this moment. We will not publish any modules until there is a) consensus and b) a valid preferential use case over the existing system.
Sent from my phone. Apologies for brevity and typos.On 3 Nov 2016 1:52 pm, DM Smith <dmsm...@crosswire.org> wrote: > > lzss is an early compression supported by SWORD and JSword. There is no need, > other than policy, to refrain from using it. > > xz and bzip2 are fully supported by JSword. But not all frontends in the wild > have it. > > I worked with Chris Little on both xz and bzip2 support. Him adding it to > SWORD lib, me to JSword and osis2mod. I’m not sure in what way they are > experimental. AFAICT, they are complete and work. > > DM > > > On Nov 3, 2016, at 3:57 AM, Jaak Ristioja <j...@ristioja.ee> wrote: > > > > Thanks you, Peter! Ok, so XzCompress is experimental and the most common > > ZipCompress is considered stable. But what about LZSSCompress and > > Bzip2Compress? > > > > Will the experimental compression methods be removed from stable Sword > > releases in the future? > > > > J > > > > On 03.11.2016 09:45, Peter von Kaehne wrote: > >> I can categorically say that there are no released CW modules using any > >> compression other than zip. There might be some in our experimental repo. > >> But they would not be released so. > >> > >> I would think the same can be confirmed rapidly with all other official > >> repos, Xiphos, IBT and eBible. I am convinced they will say the same. > >> > >> It is unlikely that anyone else has done so, but who knows? I would not > >> worry about it. > >> > >> Peter > >> > >> Sent from my phone. Apologies for brevity and typos.On 3 Nov 2016 07:31, > >> Jaak Ristioja <j...@ristioja.ee> wrote: > >>> > >>> I just tried `./configure && make` Sword 1.7.4 and it did compile in > >>> xzcomprs.cpp. So I guess the stable releases DO include the experimental > >>> code. > >>> > >>> Usually it helps when experimental features have their own feature > >>> branches or similar. This not being a common practice in the Sword > >>> project, is probably one cause for other experimental code reaching > >>> trunk as well (see the "Infinite loop in SWModule multilemma window > >>> search" thread on this mailing list). > >>> > >>> Anyway, are you saying that removing/disabling this code will not break > >>> anything for end-users? Because there are no (known?) modules which have > >>> XZ compression? I think there's a risk it has already been used for > >>> production. Can you please elaborate? > >>> > >>> Best regards, > >>> J > >>> > >>> PS: Unrelated to the issue at hand, but this article might be of > >>> interest to the Sword project: "Xz format inadequate for long-term > >>> archiving" http://www.nongnu.org/lzip/xz_inadequate.html > >>> > >>> On 03.11.2016 08:51, Peter von Kaehne wrote: > >>>> Leaving aside the question of a bug, xzcompress.cpp is experimental > >>>> code. It is not included in any releases so far AFAIK and should not be > >>>> compiled into releases or used by frontends. > >>>> > >>>> At least under autotools you need to ask to include it. > >>>> > >>>> We have no modules created with it and would currently not allow any > >>>> either. The only currently used and acceptable compression is zip. > >>>> > >>>> HTH > >>>> > >>>> Peter > >>>> Sent from my phone. Apologies for brevity and typos.On 3 Nov 2016 02:50, > >>>> TS <outofthec...@icloud.com> wrote: > >>>>> > >>>>> In xzcomprs.cpp, at line 30, there’s the code: > >>>>> > >>>>> #include <lzma.h> > >>>>> > >>>>> Xcode was not happy about this since there is no header named this. It > >>>>> seems that in the previous version of PocketSword, the fix was just to > >>>>> set Xcode as to not include this file when compiling so that’s what I’m > >>>>> doing too. Since the SWORD framework does not include lzma.h as a file, > >>>>> this seems like a bug to me, but perhaps there's a reason for it? > >>>>> > >>>>> I couldn’t find any reference regarding this matter in specific in the > >>>>> wiki or mailing list. > >>>>> > >>>>> > >>>>> -TS > >>>> _______________________________________________ > >>>> sword-devel mailing list: sword-devel@crosswire.org > >>>> http://www.crosswire.org/mailman/listinfo/sword-devel > >>>> Instructions to unsubscribe/change your settings at above page > >>>> > >>> > >>> _______________________________________________ > >>> sword-devel mailing list: sword-devel@crosswire.org > >>> http://www.crosswire.org/mailman/listinfo/sword-devel > >>> Instructions to unsubscribe/change your settings at above page > >> _______________________________________________ > >> sword-devel mailing list: sword-devel@crosswire.org > >> http://www.crosswire.org/mailman/listinfo/sword-devel > >> Instructions to unsubscribe/change your settings at above page > >> > > > > > > _______________________________________________ > > sword-devel mailing list: sword-devel@crosswire.org > > http://www.crosswire.org/mailman/listinfo/sword-devel > > Instructions to unsubscribe/change your settings at above page > > _______________________________________________ > sword-devel mailing list: sword-devel@crosswire.org > http://www.crosswire.org/mailman/listinfo/sword-devel > Instructions to unsubscribe/change your settings at above page _______________________________________________ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
