Checking GEOS master (state 8 October)

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Checking GEOS master (state 8 October)

Roger Bivand
I've been checking the possible fallout from Overlay_NG (assuming that
building from master on 8 October uses the new code) in some R packages,
in particular sf and rgeos. I see messages such as:

CBR: result (after common-bits addition) is INVALID: Self-intersection at
or near point 368576.69999999995 864696 (368576.69999999995343 864696)

followed by the text representation of the object dumped to console. The
message and the dumped object are not passed through error or warning
handlers passed in through initGEOS_r(), and so are hard to trap and
handle. Is this a temporary feature which will be normalised for 3.9.0?

So far I don't think that there are obvious failures in R packages related
to the new code (again assuming that a straight build of master of 8
November exercises Overlay_NG), but the unexpected output was surprising.

Best wishes,

Roger

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
https://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
_______________________________________________
geos-devel mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/geos-devel
Roger Bivand
NHH Norwegian School of Economics, Bergen, Norway
Reply | Threaded
Open this post in threaded view
|

Re: Checking GEOS master (state 8 October)

Paul Ramsey


> On Oct 13, 2020, at 11:08 AM, Roger Bivand <[hidden email]> wrote:
>
> I've been checking the possible fallout from Overlay_NG (assuming that building from master on 8 October uses the new code) in some R packages, in particular sf and rgeos. I see messages such as:
>
> CBR: result (after common-bits addition) is INVALID: Self-intersection at or near point 368576.69999999995 864696 (368576.69999999995343 864696)

That looks like the old overlay handler. Have you turned NG on in your build?

P


>
> followed by the text representation of the object dumped to console. The message and the dumped object are not passed through error or warning handlers passed in through initGEOS_r(), and so are hard to trap and handle. Is this a temporary feature which will be normalised for 3.9.0?
>
> So far I don't think that there are obvious failures in R packages related to the new code (again assuming that a straight build of master of 8 November exercises Overlay_NG), but the unexpected output was surprising.
>
> Best wishes,
>
> Roger
>
> --
> Roger Bivand
> Department of Economics, Norwegian School of Economics,
> Helleveien 30, N-5045 Bergen, Norway.
> voice: +47 55 95 93 55; e-mail: [hidden email]
> https://orcid.org/0000-0003-2392-6140
> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
> _______________________________________________
> geos-devel mailing list
> [hidden email]
> https://lists.osgeo.org/mailman/listinfo/geos-devel

_______________________________________________
geos-devel mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/geos-devel
Reply | Threaded
Open this post in threaded view
|

Re: Checking GEOS master (state 8 October)

Roger Bivand
On Tue, 13 Oct 2020, Paul Ramsey wrote:

>
>
>> On Oct 13, 2020, at 11:08 AM, Roger Bivand <[hidden email]> wrote:
>>
>> I've been checking the possible fallout from Overlay_NG (assuming that
>> building from master on 8 October uses the new code) in some R
>> packages, in particular sf and rgeos. I see messages such as:
>>
>> CBR: result (after common-bits addition) is INVALID: Self-intersection
>> at or near point 368576.69999999995 864696 (368576.69999999995343
>> 864696)
>
> That looks like the old overlay handler. Have you turned NG on in your
> build?

Thanks for the quick response. First, is "master" the correct branch
(gitea)? If yes, what is the incantation - I couldn't find any reference
to NG in the code or configure.ac? I'd assumed from earlier discussion
here that NG was on by default on "master".

Roger

>
> P
>
>
>>
>> followed by the text representation of the object dumped to console.
>> The message and the dumped object are not passed through error or
>> warning handlers passed in through initGEOS_r(), and so are hard to
>> trap and handle. Is this a temporary feature which will be normalised
>> for 3.9.0?
>>
>> So far I don't think that there are obvious failures in R packages
>> related to the new code (again assuming that a straight build of master
>> of 8 November exercises Overlay_NG), but the unexpected output was
>> surprising.
>>
>> Best wishes,
>>
>> Roger
>>
>> --
>> Roger Bivand
>> Department of Economics, Norwegian School of Economics,
>> Helleveien 30, N-5045 Bergen, Norway.
>> voice: +47 55 95 93 55; e-mail: [hidden email]
>> https://orcid.org/0000-0003-2392-6140
>> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
>> _______________________________________________
>> geos-devel mailing list
>> [hidden email]
>> https://lists.osgeo.org/mailman/listinfo/geos-devel
>
>

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
https://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
_______________________________________________
geos-devel mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/geos-devel
Roger Bivand
NHH Norwegian School of Economics, Bergen, Norway
Reply | Threaded
Open this post in threaded view
|

Re: Checking GEOS master (state 8 October)

Paul Ramsey
In Cmake set DISABLE_OVERLAYNG to OFF
P

> On Oct 13, 2020, at 11:23 AM, Roger Bivand <[hidden email]> wrote:
>
> On Tue, 13 Oct 2020, Paul Ramsey wrote:
>
>>
>>
>>> On Oct 13, 2020, at 11:08 AM, Roger Bivand <[hidden email]> wrote:
>>>
>>> I've been checking the possible fallout from Overlay_NG (assuming that building from master on 8 October uses the new code) in some R packages, in particular sf and rgeos. I see messages such as:
>>>
>>> CBR: result (after common-bits addition) is INVALID: Self-intersection at or near point 368576.69999999995 864696 (368576.69999999995343 864696)
>>
>> That looks like the old overlay handler. Have you turned NG on in your build?
>
> Thanks for the quick response. First, is "master" the correct branch (gitea)? If yes, what is the incantation - I couldn't find any reference to NG in the code or configure.ac? I'd assumed from earlier discussion here that NG was on by default on "master".
>
> Roger
>
>>
>> P
>>
>>
>>>
>>> followed by the text representation of the object dumped to console. The message and the dumped object are not passed through error or warning handlers passed in through initGEOS_r(), and so are hard to trap and handle. Is this a temporary feature which will be normalised for 3.9.0?
>>>
>>> So far I don't think that there are obvious failures in R packages related to the new code (again assuming that a straight build of master of 8 November exercises Overlay_NG), but the unexpected output was surprising.
>>>
>>> Best wishes,
>>>
>>> Roger
>>>
>>> --
>>> Roger Bivand
>>> Department of Economics, Norwegian School of Economics,
>>> Helleveien 30, N-5045 Bergen, Norway.
>>> voice: +47 55 95 93 55; e-mail: [hidden email]
>>> https://orcid.org/0000-0003-2392-6140
>>> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
>>> _______________________________________________
>>> geos-devel mailing list
>>> [hidden email]
>>> https://lists.osgeo.org/mailman/listinfo/geos-devel
>>
>>
>
> --
> Roger Bivand
> Department of Economics, Norwegian School of Economics,
> Helleveien 30, N-5045 Bergen, Norway.
> voice: +47 55 95 93 55; e-mail: [hidden email]
> https://orcid.org/0000-0003-2392-6140
> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en

_______________________________________________
geos-devel mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/geos-devel
Reply | Threaded
Open this post in threaded view
|

Re: Checking GEOS master (state 8 October)

Roger Bivand
On Tue, 13 Oct 2020, Paul Ramsey wrote:

> In Cmake set DISABLE_OVERLAYNG to OFF

Thanks, never Cmake here, am trying ../configure --enable-overlayng. I'm
still puzzled about the free output to console not going through the error
or warning handler.

Roger


> P
>
>> On Oct 13, 2020, at 11:23 AM, Roger Bivand <[hidden email]> wrote:
>>
>> On Tue, 13 Oct 2020, Paul Ramsey wrote:
>>
>>>
>>>
>>>> On Oct 13, 2020, at 11:08 AM, Roger Bivand <[hidden email]> wrote:
>>>>
>>>> I've been checking the possible fallout from Overlay_NG (assuming that building from master on 8 October uses the new code) in some R packages, in particular sf and rgeos. I see messages such as:
>>>>
>>>> CBR: result (after common-bits addition) is INVALID: Self-intersection at or near point 368576.69999999995 864696 (368576.69999999995343 864696)
>>>
>>> That looks like the old overlay handler. Have you turned NG on in your build?
>>
>> Thanks for the quick response. First, is "master" the correct branch (gitea)? If yes, what is the incantation - I couldn't find any reference to NG in the code or configure.ac? I'd assumed from earlier discussion here that NG was on by default on "master".
>>
>> Roger
>>
>>>
>>> P
>>>
>>>
>>>>
>>>> followed by the text representation of the object dumped to console. The message and the dumped object are not passed through error or warning handlers passed in through initGEOS_r(), and so are hard to trap and handle. Is this a temporary feature which will be normalised for 3.9.0?
>>>>
>>>> So far I don't think that there are obvious failures in R packages related to the new code (again assuming that a straight build of master of 8 November exercises Overlay_NG), but the unexpected output was surprising.
>>>>
>>>> Best wishes,
>>>>
>>>> Roger
>>>>
>>>> --
>>>> Roger Bivand
>>>> Department of Economics, Norwegian School of Economics,
>>>> Helleveien 30, N-5045 Bergen, Norway.
>>>> voice: +47 55 95 93 55; e-mail: [hidden email]
>>>> https://orcid.org/0000-0003-2392-6140
>>>> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
>>>> _______________________________________________
>>>> geos-devel mailing list
>>>> [hidden email]
>>>> https://lists.osgeo.org/mailman/listinfo/geos-devel
>>>
>>>
>>
>> --
>> Roger Bivand
>> Department of Economics, Norwegian School of Economics,
>> Helleveien 30, N-5045 Bergen, Norway.
>> voice: +47 55 95 93 55; e-mail: [hidden email]
>> https://orcid.org/0000-0003-2392-6140
>> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
>
>

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
https://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
_______________________________________________
geos-devel mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/geos-devel
Roger Bivand
NHH Norwegian School of Economics, Bergen, Norway
Reply | Threaded
Open this post in threaded view
|

Re: Checking GEOS master (state 8 October)

Martin Davis-3
That output is coming from HeuristicOverlay.cpp:  https://github.com/libgeos/geos/blame/c4aebbbc048618e75b39f6c8d85dce8e8216cf9f/src/geom/HeuristicOverlay.cpp#L206

It shouldn't be showing up unless GEOS_DEBUG_HEURISTICOVERLAY has been enabled, however.

On Tue, Oct 13, 2020 at 11:40 AM Roger Bivand <[hidden email]> wrote:
On Tue, 13 Oct 2020, Paul Ramsey wrote:

> In Cmake set DISABLE_OVERLAYNG to OFF

Thanks, never Cmake here, am trying ../configure --enable-overlayng. I'm
still puzzled about the free output to console not going through the error
or warning handler.

Roger


> P
>
>> On Oct 13, 2020, at 11:23 AM, Roger Bivand <[hidden email]> wrote:
>>
>> On Tue, 13 Oct 2020, Paul Ramsey wrote:
>>
>>>
>>>
>>>> On Oct 13, 2020, at 11:08 AM, Roger Bivand <[hidden email]> wrote:
>>>>
>>>> I've been checking the possible fallout from Overlay_NG (assuming that building from master on 8 October uses the new code) in some R packages, in particular sf and rgeos. I see messages such as:
>>>>
>>>> CBR: result (after common-bits addition) is INVALID: Self-intersection at or near point 368576.69999999995 864696 (368576.69999999995343 864696)
>>>
>>> That looks like the old overlay handler. Have you turned NG on in your build?
>>
>> Thanks for the quick response. First, is "master" the correct branch (gitea)? If yes, what is the incantation - I couldn't find any reference to NG in the code or configure.ac? I'd assumed from earlier discussion here that NG was on by default on "master".
>>
>> Roger
>>
>>>
>>> P
>>>
>>>
>>>>
>>>> followed by the text representation of the object dumped to console. The message and the dumped object are not passed through error or warning handlers passed in through initGEOS_r(), and so are hard to trap and handle. Is this a temporary feature which will be normalised for 3.9.0?
>>>>
>>>> So far I don't think that there are obvious failures in R packages related to the new code (again assuming that a straight build of master of 8 November exercises Overlay_NG), but the unexpected output was surprising.
>>>>
>>>> Best wishes,
>>>>
>>>> Roger
>>>>
>>>> --
>>>> Roger Bivand
>>>> Department of Economics, Norwegian School of Economics,
>>>> Helleveien 30, N-5045 Bergen, Norway.
>>>> voice: +47 55 95 93 55; e-mail: [hidden email]
>>>> https://orcid.org/0000-0003-2392-6140
>>>> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
>>>> _______________________________________________
>>>> geos-devel mailing list
>>>> [hidden email]
>>>> https://lists.osgeo.org/mailman/listinfo/geos-devel
>>>
>>>
>>
>> --
>> Roger Bivand
>> Department of Economics, Norwegian School of Economics,
>> Helleveien 30, N-5045 Bergen, Norway.
>> voice: +47 55 95 93 55; e-mail: [hidden email]
>> https://orcid.org/0000-0003-2392-6140
>> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
>
>

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
https://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
_______________________________________________
geos-devel mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/geos-devel

_______________________________________________
geos-devel mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/geos-devel