That sounds like a reasonable solution. Ian
On Thu, 21 May 2020 at 14:40, Andrea Aime <andrea.a...@geo-solutions.it> wrote: > Hi, > following up on this, Even replied to the thread noting that in OGR there > is no distinction between > linestring and multilinestring, and that one can force the creation of > multi geometries by a parameter in ogr2ogr. > > Now, that still means many will get it wrong, because they may not be > aware of the parameter, or aware of the mixed > single/multi geometries, as long as their tools of choice are based on OGR. > > So, while one can be formally "right" about the geopackage being broken, > I'd still suggest to add a flag forcing > the store to treat all single geometries as multi ones, and maybe make it > a default. Most won't even notice and > will be happy to have the store work with their data, the few that do, > have a flag to turn off the behavior. > > My 2 cents :-) > > Cheers > Andrea > > On Thu, May 14, 2020 at 3:15 PM Andrea Aime <andrea.a...@geo-solutions.it> > wrote: > >> Hi all, >> got confirmation from the spec lead that the GeoPackage is indeed >> invalid, quoting my question about mixing POLYGON and MULTIPOLYGON in the >> same table: >> >> Andrea, >> Your instincts are correct here. We're using the simple features rules >> for geometries and those do not provide a way to classify geometries with a >> single dimension and unknown multiplicity. In the case you are describing, >> GEOMETRY should be used. Not doing so violates Requirement 31 >> <http://www.geopackage.org/spec121/#r31>. Perhaps we can update the text >> to be more explicit about this. >> >> >> Cheers >> Andrea >> >> >> On Wed, May 13, 2020 at 7:25 PM Jody Garnett <jody.garn...@geocat.net> >> wrote: >> >>> Noted, full table scans on those would not be fun! >>> >>> So a warning (or exception?) and a parameter: "force_multi" accepting a >>> list of tables to multi, or "*" for all of them may be a good combo. >>> >>> aside: I wish we had an extra "tab" for the datastores to perform >>> maintenance activities and checks (and in this case fix the geopackage >>> definition). >>> -- >>> Jody Garnett >>> GeoCat - Government Geographic Data Publishing >>> >>> >>> On Wed, May 13, 2020 at 10:16 AM Andrea Aime < >>> andrea.a...@geo-solutions.it> wrote: >>> >>>> On Wed, May 13, 2020 at 6:57 PM Jody Garnett <jody.garn...@gmail.com> >>>> wrote: >>>> >>>>> Andrea I am personally in favour of throwing an exception if the data >>>>> is inconsistent, but I respect that some data is not going to be fixed... >>>>> >>>>> Would a connection parameter work? Or were you thinking a system >>>>> property flag ... >>>>> >>>> >>>> Connection parameter, off by default. >>>> >>>> >>>>> Something like "check_geometry_type": do a pass through the geometry >>>>> to detect this problem and override to MultiPolygon when creating the >>>>> FeatureType. >>>>> >>>> >>>> Ouch no, this implementation would take forever to run... I'm just >>>> playing with a couple of Ordnance Survey geopackages, the small one is >>>> 10GB, >>>> the large one over 200GB. There are hundreds of millions of features >>>> inside, you don't want to actually visit them all to find out that you did >>>> not really >>>> need to expand the type ;-) >>>> Just to give you a reference, here is a summary table for OS mastermap >>>> topography, full country, single GeoPackage... the largest table has over >>>> 338 million rows...: >>>> >>>> table_schema | table_name | row_estimate | total | index | >>>> toast | table >>>> --------------+--------------------+--------------+---------+---------+------------+--------- >>>> translator2 | topographicline | 338,790,112 | 183 GB | 42 GB | >>>> 227 MB | 140 GB translator2 | topographicarea | 121,382,384 | 101 >>>> GB | 14 GB | 2138 MB | 85 GB translator2 | cartographictext | >>>> 21,865,014 | 10 GB | 2695 MB | 8192 bytes | 8022 MB translator2 | >>>> topographicpoint | 4,270,670 | 2015 MB | 453 MB | 8192 bytes | 1562 >>>> MB translator2 | cartographicsymbol | 3,705,680 | 1755 MB | 403 MB | >>>> 8192 bytes | 1353 MB translator2 | boundaryline | 516,863 | >>>> 563 MB | 55 MB | 24 MB | 484 MB >>>> >>>> >>>> >>>>> I considered a flag to force "multi" but that would end up applying to >>>>> all tables even when it is not needed. >>>>> >>>> >>>> This seems a lot more sane to me... an alternative could be to have the >>>> user list which tables to "fix", getting a comma >>>> separated list. A bit more annoying to setup though, could use a >>>> dedicated UI. >>>> >>>> Cheers >>>> Andrea >>>> >>>> == GeoServer Professional Services from the experts! Visit >>>> http://goo.gl/it488V for more information. == Ing. Andrea Aime >>>> @geowolf Technical Lead GeoSolutions S.A.S. Via di Montramito 3/A 55054 >>>> Massarosa (LU) phone: +39 0584 962313 fax: +39 0584 1660272 mob: +39 339 >>>> 8844549 http://www.geo-solutions.it http://twitter.com/geosolutions_it >>>> ------------------------------------------------------- *Con >>>> riferimento alla normativa sul trattamento dei dati personali (Reg. UE >>>> 2016/679 - Regolamento generale sulla protezione dei dati “GDPR”), si >>>> precisa che ogni circostanza inerente alla presente email (il suo >>>> contenuto, gli eventuali allegati, etc.) è un dato la cui conoscenza è >>>> riservata al/i solo/i destinatario/i indicati dallo scrivente. Se il >>>> messaggio Le è giunto per errore, è tenuta/o a cancellarlo, ogni altra >>>> operazione è illecita. Le sarei comunque grato se potesse darmene notizia. >>>> This email is intended only for the person or entity to which it is >>>> addressed and may contain information that is privileged, confidential or >>>> otherwise protected from disclosure. We remind that - as provided by >>>> European Regulation 2016/679 “GDPR” - copying, dissemination or use of this >>>> e-mail or the information herein by anyone other than the intended >>>> recipient is prohibited. If you have received this email by mistake, please >>>> notify us immediately by telephone or e-mail.* >>>> >>> >> >> -- >> >> Regards, Andrea Aime >> >> == GeoServer Professional Services from the experts! Visit >> http://goo.gl/it488V for more information. == Ing. Andrea Aime @geowolf >> Technical Lead GeoSolutions S.A.S. Via di Montramito 3/A 55054 Massarosa >> (LU) phone: +39 0584 962313 fax: +39 0584 1660272 mob: +39 339 8844549 >> http://www.geo-solutions.it http://twitter.com/geosolutions_it >> ------------------------------------------------------- *Con riferimento >> alla normativa sul trattamento dei dati personali (Reg. UE 2016/679 - >> Regolamento generale sulla protezione dei dati “GDPR”), si precisa che ogni >> circostanza inerente alla presente email (il suo contenuto, gli eventuali >> allegati, etc.) è un dato la cui conoscenza è riservata al/i solo/i >> destinatario/i indicati dallo scrivente. Se il messaggio Le è giunto per >> errore, è tenuta/o a cancellarlo, ogni altra operazione è illecita. Le >> sarei comunque grato se potesse darmene notizia. This email is intended >> only for the person or entity to which it is addressed and may contain >> information that is privileged, confidential or otherwise protected from >> disclosure. We remind that - as provided by European Regulation 2016/679 >> “GDPR” - copying, dissemination or use of this e-mail or the information >> herein by anyone other than the intended recipient is prohibited. If you >> have received this email by mistake, please notify us immediately by >> telephone or e-mail.* >> > > > -- > > Regards, Andrea Aime > > == GeoServer Professional Services from the experts! Visit > http://goo.gl/it488V for more information. == Ing. Andrea Aime @geowolf > Technical Lead GeoSolutions S.A.S. Via di Montramito 3/A 55054 Massarosa > (LU) phone: +39 0584 962313 fax: +39 0584 1660272 mob: +39 339 8844549 > http://www.geo-solutions.it http://twitter.com/geosolutions_it > ------------------------------------------------------- *Con riferimento > alla normativa sul trattamento dei dati personali (Reg. UE 2016/679 - > Regolamento generale sulla protezione dei dati “GDPR”), si precisa che ogni > circostanza inerente alla presente email (il suo contenuto, gli eventuali > allegati, etc.) è un dato la cui conoscenza è riservata al/i solo/i > destinatario/i indicati dallo scrivente. Se il messaggio Le è giunto per > errore, è tenuta/o a cancellarlo, ogni altra operazione è illecita. Le > sarei comunque grato se potesse darmene notizia. This email is intended > only for the person or entity to which it is addressed and may contain > information that is privileged, confidential or otherwise protected from > disclosure. We remind that - as provided by European Regulation 2016/679 > “GDPR” - copying, dissemination or use of this e-mail or the information > herein by anyone other than the intended recipient is prohibited. If you > have received this email by mistake, please notify us immediately by > telephone or e-mail.* > _______________________________________________ > GeoTools-Devel mailing list > GeoTools-Devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/geotools-devel > -- Ian Turton
_______________________________________________ GeoTools-Devel mailing list GeoTools-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
