On Sun, Jan 28, 2018 at 9:01 PM, Martin Landa <landa.martin@gmail.com> wrote:
Hi Markus,
2018-01-28 20:45 GMT+01:00 Markus Neteler <neteler@osgeo.org>:
Do you plan to add reprojection on the fly through r.import?
r.sentinel.import -o
I see. Perhaps I got confused because without -o it tells me:
...
In case of no significant differences in the projection definitions,
use the -o flag to ignore them and use current location definition.
Consider generating a new location from the input dataset using the
'location' parameter.
So it would be good to suppress this message which likely comes from a
r.in.gdal call and replace it with one related to
-o Reproject raster data using r.import if needed
the reprojections is done when using -o (as advertised in the flag description).
A related idea, if you don't mind:
GRASS 7.4.1svn (nc_spm_08_grass7):~ > r.sentinel.import
input=/home/mneteler/tmp/ pattern='B0[4|8]' -o
...
Estimated target resolution for input band <T17SQV_20180124T160541_B04>:
9.79587588067
It would probably be best to fetch the nominal resolution from the S2
band description and apply that rather than estimating it (at least,
most users will expect to get 10m in this example). Or, optionally
make it a flag like e.g. "-k keep nominal band resolution when in
metric CRS" or similar which could then also imply bilinear resampling
rather than the standard nearest neighbor resampling.
Very nice work! This will simplify some of my workflows significantly!
One parameter I was missing was however "area_relation". Please find attached a diff, in case you agree that it would be a useful addition (the python script is successfully tested, though I did not check the html changes).
Two things I was wondering are:
1. it might make sense to also provide the "url" as an option, cause there are some national mirrors of ESAs scihub around (at least this is the case for Norway) that expose the same API but are faster / cause have less traffic. What do you think?
2. the tools currently only support Sentinel-2, right. Do you plan to extend them to Sentinel-1/3... in future?
Cheers
Stefan
-----Original Message-----
From: grass-user [mailto:grass-user-bounces@lists.osgeo.org] On Behalf Of Martin Landa
Sent: fredag 26. januar 2018 23.36
To: GRASS users list <grass-user@lists.osgeo.org>
Subject: [GRASS-user] New AddOns for downloading and importing Sentinel products
Hi,
today I have uploaded a new toolset for downloading and importing Sentinel products, namely two modules:
r.sentinel.download [1] and r.sentinel.import [2].
The first module requires Sentinelsat and Pandas library, the second GDAL with OpenJPEG.
2018-01-28 21:15 GMT+01:00 Markus Neteler <neteler@osgeo.org>:
So it would be good to suppress this message which likely comes from a
r.in.gdal call and replace it with one related to
-o Reproject raster data using r.import if needed
done, projection check is done by `r.in.gdal -j`. I have also rename
-o flag to -r. Flag -o is used for overriding projection check which
is this case. Ma
2018-01-28 21:36 GMT+01:00 Markus Neteler <neteler@osgeo.org>:
It would probably be best to fetch the nominal resolution from the S2
band description and apply that rather than estimating it (at least,
most users will expect to get 10m in this example). Or, optionally
implemented in r72198 by reading input file using GDAL.
make it a flag like e.g. "-k keep nominal band resolution when in
metric CRS" or similar which could then also imply bilinear resampling
rather than the standard nearest neighbor resampling.
I would prefer avoid adding new flags if not necessary Ma
2018-01-29 9:11 GMT+01:00 Stefan Blumentrath <Stefan.Blumentrath@nina.no>:
Very nice work! This will simplify some of my workflows significantly!
nice to hear.
One parameter I was missing was however "area_relation". Please find attached a diff, in case you agree that it would be a useful addition (the python script is successfully tested, though I did not check the html changes).
Patch applied (r72199)
Two things I was wondering are:
1. it might make sense to also provide the "url" as an option, cause there are some national mirrors of ESAs scihub around (at least this is the case for Norway) that expose the same API but are faster / cause have less traffic. What do you think?
Would make sense.
2. the tools currently only support Sentinel-2, right. Do you plan to extend them to Sentinel-1/3... in future?
- installing GRASS using osgeo4w gives you sentinelsat, but pandas
doesn't seem to be installed in the correct version as with
import pandas
you get an error message about a missing natype module (something like
that, am not at the windows computer, now).
quick test here on my OSGeo4W 64bit:
run o-help for a list of available commands
C:\>python
Python 2.7.5 (default, May 15 2013, 22:44:16) [MSC v.1500 64 bit (AMD64)] on
win32
Type "help", "copyright", "credits" or "license" for more information.
2018-02-16 16:17 GMT+01:00 Moritz Lennert <mlennert@club.worldonline.be>:
Shouldn't these be better called i.sentinel.* ? They are clearly imagery
related.
you are probably right. Than we should also rename r.modis and
probably also other modules. What do you think? Ma
+1
Concerning r.sentinel.download: A colleague has been trying to use it
under windows, but:
- installing GRASS using osgeo4w gives you sentinelsat, but pandas
doesn't seem to be installed in the correct version as with
import pandas
you get an error message about a missing natype module (something like
that, am not at the windows computer, now).
- installing GRASS using the standalone installer, my colleague has
difficulty installing sentinelsat as 'pip install sentinelsat' fails
Has anyone else tried this on Windows ?
P.S.
I consider it unideal practice to write passwords in clear on the command line. I don't know if it would be very complicated to work with a sentinel scihub password file...
2018-02-19 14:22 GMT+01:00 Moritz Lennert <mlennert@club.worldonline.be>:
I consider it unideal practice to write passwords in clear on the command
line. I don't know if it would be very complicated to work with a sentinel
scihub password file...
what is a structure of such file? Thanks for pointers, Ma
Le Mon, 19 Feb 2018 20:13:51 +0100,
Martin Landa <landa.martin@gmail.com> a écrit :
Hi,
2018-02-19 14:22 GMT+01:00 Moritz Lennert
<mlennert@club.worldonline.be>:
> I consider it unideal practice to write passwords in clear on the
> command line. I don't know if it would be very complicated to work
> with a sentinel scihub password file...
what is a structure of such file? Thanks for pointers, Ma
Following the example of .pgpass for PostgreSQL, it could be a simple
text file with:
login:password
and the module would only use it if it is chmod 600, i.e. only readable
by the owner.
Ma