Grid file is written as formatted data by writeECLdata() about opm-core HOT 13 CLOSED

Well, function writeECLData() is designed to handle data of restart type (i.e., one output per (report) time step) and doesn't write a grid file at all. Rather it is EclipseGridParser::saveEGRID() that produces the (formatted) .FEGRID file.

I fully agree that there should be some overall consistency between the two and I have actually started a slight consolidation project (branch 'devel' at http://github.com/bska/opm-core) but it is nowhere near complete. The .INIT file is low priority, however, so: Patches Welcome[tm]

from opm-core.

Hmm - I thought there wad INIT file support, will look at it. Joakim
Den 8. mai 2013 12:31 skrev "Bård Skaflestad" [email protected]
følgende:

Well, function writeECLData() is designed to handle data of restart type
(i.e., one output per (report) time step) and doesn't write a grid file at
all. Rather it is EclipseGridParser::saveEGRID() that produces the
(formatted) .FEGRID file.

I fully agree that there should be some overall consistency between the
two and I have actually started a slight consolidation project (branch
'devel' at http://github.com/bska/opm-corehttps://github.com/bska/opm-core)
but it is nowhere near complete. The .INIT file is low priority, however,
so: Patches Welcome[tm]

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/236#issuecomment-17597425
.

from opm-core.

Hmm - I thought there wad INIT file support

You're right. There is (function EclipseGridParser::saveEGRID_INIT()). That function's apparently not used, though. At least not in any of the modules I have easy access to on disk right now.

from opm-core.

I fully agree that there should be some overall consistency between the two

and I have actually started a slight consolidation project (branch 'devel'
at http://github.com/bska/opm-core https://github.com/bska/opm-core)
but it is nowhere near complete. The .INIT file is low priority, however,
so: Patches Welcome[tm]

The existing function EclipseGridParser::saveEGRID_INIT() will write both
the EGRID file and an INIT file. The created INIT file will contain the
properties PERMX, PERMZ, PERMY and FIPNUM,SATNUM and EQLNUM. This is a
quite arbitrary hardcoded list of data, and improvements should probably go
in the direction Bård mentions.

As for Bårds devel work (have not looked at it yet), which summary vectors
to store is certainly a piece of information which should be assembled
somewhere.

Joakim

On Wed, May 8, 2013 at 12:31 PM, Bård Skaflestad
[email protected]:

Well, function writeECLData() is designed to handle data of restart type
(i.e., one output per (report) time step) and doesn't write a grid file at
all. Rather it is EclipseGridParser::saveEGRID() that produces the
(formatted) .FEGRID file.

I fully agree that there should be some overall consistency between the
two and I have actually started a slight consolidation project (branch
'devel' at http://github.com/bska/opm-corehttps://github.com/bska/opm-core)
but it is nowhere near complete. The .INIT file is low priority, however,
so: Patches Welcome[tm]

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/236#issuecomment-17597425
.

from opm-core.

Following PR #286, what is the state of this issue?

from opm-core.

Following PR #286 #286, what is
the state of this issue?

Well - there are two issues left:

from opm-core.

Well - there are two issues left:

And those are:

1. ERT must be fixed to handle simulation dates before the dawn of unix
   time, i.e. 1.th of Januara 1970. That is high on my list; then
   Resinsight must be built against the new ert version.
2. opm-upscaling must be updated to write the results in a ECLIPSE
   compatible manner, essentally all files having the same base name.

Joakim

from opm-core.

Cool. Thanks a lot for the update.

from opm-core.

We'll tentatively defer this issue to the next (2014.03) release of opm-core.

from opm-core.

The ERT 1970 date issue was a pure Windows problem, but anyway that as well should be fixed in this PR: Ensembles/ert#203

from opm-core.

The ERT 1970 date issue was a pure Windows problem, but anyway that as well should be fixed in [a PR]

Cool, I look forward to seeing how this pans out.

from opm-core.

Is this issue still relevant? Since our current eclipse-format output capabilities are far better than they were at the time this issue was created, I suspect that it can be closed.

from opm-core.

I consider this issue resolved/completely out of date.

Closing.

from opm-core.