def setTable(imgFile, colorTable, bandNumber=1):
"""
Given either an open gdal dataset, or a filename,
sets the color table as an array.
The colorTable is given as an array of shape (4, numEntries)
where numEntries is the size of the color table. The order of indices
in the first axis is:
* Red
* Green
* Blue
* Opacity
The Red/Green/Blue values are on the range 0-255, with 255 meaning full
color, and the opacity is in the range 0-255, with 255 meaning fully
opaque.
This table is useually generated by getTable() or genTable().
"""
if isinstance(imgFile, basestring):
ds = gdal.Open(str(imgFile), gdal.GA_Update)
elif isinstance(imgFile, gdal.Dataset):
ds = imgFile
gdalBand = ds.GetRasterBand(bandNumber)
attrTbl = gdalBand.GetDefaultRAT()
if attrTbl is None:
# some formats eg ENVI return None
# here so we need to be able to cope
attrTbl = gdal.RasterAttributeTable()
isFileRAT = False
else:
isFileRAT = True
# but if it doesn't support dynamic writing
# we still ahve to call SetDefaultRAT
if not attrTbl.ChangesAreWrittenToFile():
isFileRAT = False
ncols, numEntries = colorTable.shape
attrTbl.SetRowCount(numEntries)
# set the columns based on their usage, creating
# if necessary
colorUsages = {gdal.GFU_Red : 'Red', gdal.GFU_Green : 'Green',
gdal.GFU_Blue : 'Blue', gdal.GFU_Alpha : 'Alpha'}
for idx, usage in enumerate(colorUsages):
colNum = attrTbl.GetColOfUsage(usage)
if colNum == -1:
name = colorUsages[usage]
attrTbl.CreateColumn(name, gdal.GFT_Integer, usage)
colNum = attrTbl.GetColumnCount() - 1
attrTbl.WriteArray(colorTable[idx], colNum)
if not isFileRAT:
attrTbl.SetDefaultRAT(attrTbl)
def test_hfa_unique_values_hist():
try:
gdal.RasterAttributeTable()
except:
pytest.skip()
ds = gdal.Open('data/i8u_c_i.img')
md = ds.GetRasterBand(1).GetMetadata()
expected = '12603|1|0|0|45|1|0|0|0|0|656|177|0|0|5026|1062|0|0|2|0|0|0|0|0|0|0|0|0|0|0|0|0|75|1|0|0|207|158|0|0|8|34|0|0|0|0|538|57|0|10|214|20|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|1|31|0|0|9|625|67|0|0|118|738|117|3004|1499|491|187|1272|513|1|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|16|3|0|0|283|123|5|1931|835|357|332|944|451|80|40|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|12|5|0|0|535|1029|118|0|33|246|342|0|0|10|8|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|169|439|0|0|6|990|329|0|0|120|295|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|164|42|0|0|570|966|0|0|18|152|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|45|106|0|0|16|16517|'
assert md[
'STATISTICS_HISTOBINVALUES'] == expected, 'Unexpected HISTOBINVALUES.'
assert md['STATISTICS_HISTOMIN'] == '0' and md['STATISTICS_HISTOMAX'] == '255', \
"unexpected histomin/histomax value."
# lets also check the RAT to ensure it has the BinValues column added.
rat = ds.GetRasterBand(1).GetDefaultRAT()
assert rat.GetColumnCount() == 6 and rat.GetTypeOfCol(0) == gdal.GFT_Real and rat.GetUsageOfCol(0) == gdal.GFU_MinMax, \
'BinValues column wrong.'
assert rat.GetValueAsInt(2, 0) == 4, 'BinValues value wrong.'
rat = None
ds = None
def test_mem_rat():
ds = gdal.GetDriverByName('MEM').Create('', 1, 1)
ds.GetRasterBand(1).SetDefaultRAT(gdal.RasterAttributeTable())
assert ds.GetRasterBand(1).GetDefaultRAT() is not None
ds.GetRasterBand(1).SetDefaultRAT(None)
assert ds.GetRasterBand(1).GetDefaultRAT() is None
def test_ehdr_rat():
tmpfile = '/vsimem/rat.bil'
gdal.Translate(tmpfile, 'data/int16_rat.bil', format='EHdr')
ds = gdal.Open(tmpfile)
rat = ds.GetRasterBand(1).GetDefaultRAT()
assert rat is not None
assert rat.GetColumnCount() == 4
assert rat.GetRowCount() == 25
for (idx, val) in [(0, -500), (1, 127), (2, 40), (3, 65)]:
assert rat.GetValueAsInt(0, idx) == val
for (idx, val) in [(0, 2000), (1, 145), (2, 97), (3, 47)]:
assert rat.GetValueAsInt(24, idx) == val
assert ds.GetRasterBand(1).GetColorTable() is not None
ds = None
ds = gdal.Open(tmpfile, gdal.GA_Update)
ds.GetRasterBand(1).SetDefaultRAT(None)
ds.GetRasterBand(1).SetColorTable(None)
ds = None
ds = gdal.Open(tmpfile, gdal.GA_Update)
assert not (ds.GetRasterBand(1).GetDefaultRAT() or ds.GetRasterBand(1).GetColorTable())
with gdaltest.error_handler():
ret = ds.GetRasterBand(1).SetDefaultRAT(gdal.RasterAttributeTable())
assert ret != 0
ds = None
gdal.GetDriverByName('EHDR').Delete(tmpfile)
def create_rat(in_raster, lookup, band_number=1):
"""
Create simple raster attribute table based on lookup {int: string} dict
Output RAT columns: VALUE (integer), DESCRIPTION (string)
eg: lookup = {1: "URBAN", 5: "WATER", 11: "AGRICULTURE", 16: "MINING"}
https://gis.stackexchange.com/questions/333897/read-rat-raster-attribute-table-using-gdal-or-other-python-libraries
"""
# open the raster at band
raster = gdal.Open(in_raster, gdal.GA_Update)
band = raster.GetRasterBand(band_number)
# Create and populate the RAT
rat = gdal.RasterAttributeTable()
rat.CreateColumn("VALUE", gdal.GFT_Integer, gdal.GFU_Generic)
rat.CreateColumn("DESCRIPTION", gdal.GFT_String, gdal.GFU_Generic)
i = 0
for value, description in sorted(lookup.items()):
rat.SetValueAsInt(i, 0, int(value))
rat.SetValueAsString(i, 1, str(description))
i += 1
raster.FlushCache()
band.SetDefaultRAT(rat)
raster = None
rat = None
band = None
def get_rat_from_vat(filename):
md = ogr.Open(filename)
mdl = md.GetLayer(0)
# get column definitions:
rat = gdal.RasterAttributeTable()
# use skip to adjust column index
layer_defn = mdl.GetLayerDefn()
for field_idx in range(0, layer_defn.GetFieldCount()):
field_defn = layer_defn.GetFieldDefn(field_idx)
field_type = TYPE_MAP[field_defn.GetType()]
if field_type is None:
# skip unmappable field type
continue
rat.CreateColumn(field_defn.GetName(), field_type,
USAGE_MAP[field_defn.GetName()])
for feature_idx in range(0, mdl.GetFeatureCount()):
feature = mdl.GetFeature(feature_idx)
skip = 0
for field_idx in range(0, feature.GetFieldCount()):
field_type = TYPE_MAP[feature.GetFieldType(field_idx)]
if field_type == gdal.GFT_Integer:
rat.SetValueAsInt(feature_idx, field_idx - skip,
feature.GetFieldAsInteger(field_idx))
elif field_type == gdal.GFT_Real:
rat.SetValueAsDouble(feature_idx, field_idx - skip,
feature.GetFieldAsDouble(field_idx))
elif field_type == gdal.GFT_String:
rat.SetValueAsString(feature_idx, field_idx - skip,
feature.GetFieldAsString(field_idx))
else:
# skip all unmappable field types
skip += 1
return rat
def test_rat_3():
ds = gdal.GetDriverByName('GTiff').Create('/vsimem/rat_3.tif', 1, 1)
ds.GetRasterBand(1).SetDefaultRAT(gdal.RasterAttributeTable())
ds = None
gdal.GetDriverByName('GTiff').Delete('/vsimem/rat_3.tif')
def vrtmisc_rat():
ds = gdal.Translate('/vsimem/vrtmisc_rat.tif',
'data/byte.tif',
format='MEM')
rat = gdal.RasterAttributeTable()
rat.CreateColumn("Ints", gdal.GFT_Integer, gdal.GFU_Generic)
ds.GetRasterBand(1).SetDefaultRAT(rat)
vrt_ds = gdal.GetDriverByName('VRT').CreateCopy('/vsimem/vrtmisc_rat.vrt',
ds)
xml_vrt = vrt_ds.GetMetadata('xml:VRT')[0]
if gdal.GetLastErrorMsg() != '':
gdaltest.post_reason('fail')
return 'fail'
vrt_ds = None
if xml_vrt.find('<GDALRasterAttributeTable>') < 0:
gdaltest.post_reason('fail')
print(xml_vrt)
return 'fail'
vrt_ds = gdal.Translate('/vsimem/vrtmisc_rat.vrt',
ds,
format='VRT',
srcWin=[0, 0, 1, 1])
xml_vrt = vrt_ds.GetMetadata('xml:VRT')[0]
if gdal.GetLastErrorMsg() != '':
gdaltest.post_reason('fail')
return 'fail'
vrt_ds = None
if xml_vrt.find('<GDALRasterAttributeTable>') < 0:
gdaltest.post_reason('fail')
print(xml_vrt)
return 'fail'
ds = None
vrt_ds = gdal.Open('/vsimem/vrtmisc_rat.vrt', gdal.GA_Update)
rat = vrt_ds.GetRasterBand(1).GetDefaultRAT()
if rat is None or rat.GetColumnCount() != 1:
gdaltest.post_reason('fail')
return 'fail'
vrt_ds.GetRasterBand(1).SetDefaultRAT(None)
if vrt_ds.GetRasterBand(1).GetDefaultRAT() is not None:
gdaltest.post_reason('fail')
return 'fail'
vrt_ds = None
ds = None
gdal.Unlink('/vsimem/vrtmisc_rat.vrt')
gdal.Unlink('/vsimem/vrtmisc_rat.tif')
return "success"
def rat_1():
gdaltest.saved_rat = None
try:
rat = gdal.RasterAttributeTable()
except:
return 'skip'
rat.CreateColumn('Value', gdal.GFT_Integer, gdal.GFU_MinMax)
rat.CreateColumn('Count', gdal.GFT_Integer, gdal.GFU_PixelCount)
rat.SetRowCount(3)
rat.SetValueAsInt(0, 0, 10)
rat.SetValueAsInt(0, 1, 100)
rat.SetValueAsInt(1, 0, 11)
rat.SetValueAsInt(1, 1, 200)
rat.SetValueAsInt(2, 0, 12)
rat.SetValueAsInt(2, 1, 90)
rat2 = rat.Clone()
if rat2.GetColumnCount() != 2:
gdaltest.post_reason('wrong column count')
return 'fail'
if rat2.GetRowCount() != 3:
gdaltest.post_reason('wrong row count')
return 'fail'
if rat2.GetNameOfCol(1) != 'Count':
gdaltest.post_reason('wrong column name')
return 'fail'
if rat2.GetUsageOfCol(1) != gdal.GFU_PixelCount:
gdaltest.post_reason('wrong column usage')
return 'fail'
if rat2.GetTypeOfCol(1) != gdal.GFT_Integer:
gdaltest.post_reason('wrong column type')
return 'fail'
if rat2.GetRowOfValue(11.0) != 1:
gdaltest.post_reason('wrong row for value')
return 'fail'
if rat2.GetValueAsInt(1, 1) != 200:
gdaltest.post_reason('wrong field value.')
return 'fail'
gdaltest.saved_rat = rat
return 'success'
def ehdr_rat():
tmpfile = '/vsimem/rat.bil'
gdal.Translate(tmpfile, 'data/int16_rat.bil', format='EHdr')
ds = gdal.Open(tmpfile)
rat = ds.GetRasterBand(1).GetDefaultRAT()
if rat is None:
gdaltest.post_reason('fail')
return 'fail'
if rat.GetColumnCount() != 4:
gdaltest.post_reason('fail')
print(rat.GetColumnCount())
return 'fail'
if rat.GetRowCount() != 25:
gdaltest.post_reason('fail')
print(rat.GetRowCount())
return 'fail'
for (idx, val) in [(0, -500), (1, 127), (2, 40), (3, 65)]:
if rat.GetValueAsInt(0, idx) != val:
gdaltest.post_reason('fail')
print(idx, rat.GetValueAsInt(0, idx))
return 'fail'
for (idx, val) in [(0, 2000), (1, 145), (2, 97), (3, 47)]:
if rat.GetValueAsInt(24, idx) != val:
gdaltest.post_reason('fail')
print(idx, rat.GetValueAsInt(24, idx))
return 'fail'
if ds.GetRasterBand(1).GetColorTable() is None:
gdaltest.post_reason('fail')
return 'fail'
ds = None
ds = gdal.Open(tmpfile, gdal.GA_Update)
ds.GetRasterBand(1).SetDefaultRAT(None)
ds.GetRasterBand(1).SetColorTable(None)
ds = None
ds = gdal.Open(tmpfile, gdal.GA_Update)
if ds.GetRasterBand(1).GetDefaultRAT() or ds.GetRasterBand(
1).GetColorTable():
gdaltest.post_reason('fail')
return 'fail'
with gdaltest.error_handler():
ret = ds.GetRasterBand(1).SetDefaultRAT(gdal.RasterAttributeTable())
if ret == 0:
gdaltest.post_reason('fail')
return 'fail'
ds = None
gdal.GetDriverByName('EHDR').Delete(tmpfile)
return 'success'
def mem_rat():
ds = gdal.GetDriverByName('MEM').Create('', 1, 1)
ds.GetRasterBand(1).SetDefaultRAT(gdal.RasterAttributeTable())
if ds.GetRasterBand(1).GetDefaultRAT() is None:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetDefaultRAT(None)
if ds.GetRasterBand(1).GetDefaultRAT() is not None:
gdaltest.post_reason('fail')
return 'fail'
return 'success'
def test_rat_4():
# Create test RAT
ds = gdal.GetDriverByName('GTiff').Create('/vsimem/rat_4.tif', 1, 1)
rat = gdal.RasterAttributeTable()
rat.CreateColumn('VALUE', gdal.GFT_Integer, gdal.GFU_MinMax)
rat.CreateColumn('CLASS', gdal.GFT_String, gdal.GFU_Name)
rat.SetValueAsInt(0, 0, 111)
rat.SetValueAsString(0, 1, 'Class1')
ds.GetRasterBand(1).SetDefaultRAT(rat)
ds = None
# Verify
ds = gdal.OpenEx('/vsimem/rat_4.tif')
gdal_band = ds.GetRasterBand(1)
rat = gdal_band.GetDefaultRAT()
assert rat.GetValueAsInt(0, 0) == 111
ds = None
# Replace existing RAT
rat = gdal.RasterAttributeTable()
rat.CreateColumn('VALUE', gdal.GFT_Integer, gdal.GFU_MinMax)
rat.CreateColumn('CLASS', gdal.GFT_String, gdal.GFU_Name)
rat.SetValueAsInt(0, 0, 222)
rat.SetValueAsString(0, 1, 'Class1')
ds = gdal.OpenEx('/vsimem/rat_4.tif', gdal.OF_RASTER | gdal.OF_UPDATE)
gdal_band = ds.GetRasterBand(1)
gdal_band.SetDefaultRAT(rat)
ds = None
# Verify
ds = gdal.OpenEx('/vsimem/rat_4.tif')
gdal_band = ds.GetRasterBand(1)
rat = gdal_band.GetDefaultRAT()
assert rat is not None
assert rat.GetValueAsInt(0, 0) == 222
ds = None
gdal.GetDriverByName('GTiff').Delete('/vsimem/rat_4.tif')
def df_to_gdal_rat(df):
df = df.copy()
if 'ClassNumber' not in df.columns:
df['ClassNumber'] = df.index
rat = gdal.RasterAttributeTable()
rat.SetRowCount(len(df))
for num, col in enumerate(df.columns):
gftype = dtype_map(df[col].dtype)
if col in f_names:
usetype = f_use_d[col]
else:
usetype = gdal.GFU_Generic
# have to call str(col) because CreateColumn can't take unicode
rat.CreateColumn(str(col), gftype, usetype)
rat.WriteArray(df[col].tolist(), num)
return rat
def hfa_unique_values_hist():
try:
gdal.RasterAttributeTable()
except:
return 'skip'
ds = gdal.Open('data/i8u_c_i.img')
md = ds.GetRasterBand(1).GetMetadata()
expected = '12603|1|0|0|45|1|0|0|0|0|656|177|0|0|5026|1062|0|0|2|0|0|0|0|0|0|0|0|0|0|0|0|0|75|1|0|0|207|158|0|0|8|34|0|0|0|0|538|57|0|10|214|20|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|1|31|0|0|9|625|67|0|0|118|738|117|3004|1499|491|187|1272|513|1|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|16|3|0|0|283|123|5|1931|835|357|332|944|451|80|40|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|12|5|0|0|535|1029|118|0|33|246|342|0|0|10|8|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|169|439|0|0|6|990|329|0|0|120|295|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|164|42|0|0|570|966|0|0|18|152|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|45|106|0|0|16|16517|'
if md['STATISTICS_HISTOBINVALUES'] != expected:
print(md['STATISTICS_HISTOBINVALUES'])
gdaltest.post_reason('Unexpected HISTOBINVALUES.')
return 'fail'
if md['STATISTICS_HISTOMIN'] != '0' \
or md['STATISTICS_HISTOMAX'] != '255':
print(md)
gdaltest.post_reason("unexpected histomin/histomax value.")
return 'fail'
# lets also check the RAT to ensure it has the BinValues column added.
rat = ds.GetRasterBand(1).GetDefaultRAT()
if rat.GetColumnCount() != 6 \
or rat.GetTypeOfCol(0) != gdal.GFT_Real \
or rat.GetUsageOfCol(0) != gdal.GFU_MinMax:
print(rat.GetColumnCount())
print(rat.GetTypeOfCol(0))
print(rat.GetUsageOfCol(0))
gdaltest.post_reason('BinValues column wrong.')
return 'fail'
if rat.GetValueAsInt(2, 0) != 4:
print(rat.GetValueAsInt(2, 0))
gdaltest.post_reason('BinValues value wrong.')
return 'fail'
rat = None
ds = None
return 'success'
def test_vrtmisc_rat():
ds = gdal.Translate('/vsimem/vrtmisc_rat.tif',
'data/byte.tif',
format='MEM')
rat = gdal.RasterAttributeTable()
rat.CreateColumn("Ints", gdal.GFT_Integer, gdal.GFU_Generic)
ds.GetRasterBand(1).SetDefaultRAT(rat)
vrt_ds = gdal.GetDriverByName('VRT').CreateCopy('/vsimem/vrtmisc_rat.vrt',
ds)
xml_vrt = vrt_ds.GetMetadata('xml:VRT')[0]
assert gdal.GetLastErrorMsg() == ''
vrt_ds = None
assert '<GDALRasterAttributeTable tableType="thematic">' in xml_vrt
vrt_ds = gdal.Translate('/vsimem/vrtmisc_rat.vrt',
ds,
format='VRT',
srcWin=[0, 0, 1, 1])
xml_vrt = vrt_ds.GetMetadata('xml:VRT')[0]
assert gdal.GetLastErrorMsg() == ''
vrt_ds = None
assert '<GDALRasterAttributeTable tableType="thematic">' in xml_vrt
ds = None
vrt_ds = gdal.Open('/vsimem/vrtmisc_rat.vrt', gdal.GA_Update)
rat = vrt_ds.GetRasterBand(1).GetDefaultRAT()
assert rat is not None and rat.GetColumnCount() == 1
vrt_ds.GetRasterBand(1).SetDefaultRAT(None)
assert vrt_ds.GetRasterBand(1).GetDefaultRAT() is None
vrt_ds = None
ds = None
gdal.Unlink('/vsimem/vrtmisc_rat.vrt')
gdal.Unlink('/vsimem/vrtmisc_rat.tif')
def main(rin, data):
rat = gdal.RasterAttributeTable()
rat.GetRowCount()
rat.GetColumnCount()
# safe_makedirs(os.path.dirname(rout))
data_ext = os.path.splitext(data)[1]
if data_ext == '.dbf':
dbf_process(rat, data)
elif data_ext == '.csv':
csv_process(rat, data)
else:
raise Exception('Datafile must be DBF or CSV')
driver = gdal.GetDriverByName("GTiff")
ds_in = gdal.Open(rin)
# ds_out = driver.CreateCopy(rout, ds_in, options=["TILED=YES", "COMPRESS=LZW"])
ds_in.GetRasterBand(1).SetDefaultRAT(rat)
ds_in = None
# ds_out = None
print('done')
def test_rat_1():
gdaltest.saved_rat = None
try:
rat = gdal.RasterAttributeTable()
except:
pytest.skip()
rat.CreateColumn('Value', gdal.GFT_Integer, gdal.GFU_MinMax)
rat.CreateColumn('Count', gdal.GFT_Integer, gdal.GFU_PixelCount)
rat.SetRowCount(3)
rat.SetValueAsInt(0, 0, 10)
rat.SetValueAsInt(0, 1, 100)
rat.SetValueAsInt(1, 0, 11)
rat.SetValueAsInt(1, 1, 200)
rat.SetValueAsInt(2, 0, 12)
rat.SetValueAsInt(2, 1, 90)
rat2 = rat.Clone()
assert rat2.GetColumnCount() == 2, 'wrong column count'
assert rat2.GetRowCount() == 3, 'wrong row count'
assert rat2.GetNameOfCol(1) == 'Count', 'wrong column name'
assert rat2.GetUsageOfCol(1) == gdal.GFU_PixelCount, 'wrong column usage'
assert rat2.GetTypeOfCol(1) == gdal.GFT_Integer, 'wrong column type'
assert rat2.GetRowOfValue(11.0) == 1, 'wrong row for value'
assert rat2.GetValueAsInt(1, 1) == 200, 'wrong field value.'
gdaltest.saved_rat = rat
def createRAT():
ds = gdal.Open(r'D:\cag_poultry\test\data\result\test_review5.tif')
rb = ds.GetRasterBand(1)
#print (rb)
u = numpy.unique(rb.ReadAsArray())
#print (len(u))
#print (u.size)
r = numpy.random.uniform(0,1000, size=u.size)
rat = gdal.RasterAttributeTable()
rat.CreateColumn("Value", gdal.GFT_Real, gdal.GFU_Generic)
rat.CreateColumn("RANDOM", gdal.GFT_Real, gdal.GFU_Generic)
for i in range(u.size):
#print (float(r[i]))
#print (float(u[i]))
rat.SetValueAsDouble(i,0,float(u[i]))
rat.SetValueAsDouble(i,1,float(r[i]))
rb.SetDefaultRAT(rat)
ds = None
from osgeo import gdal
# Don't forget to change the folder.
os.chdir(r'D:\osgeopy-data\Switzerland')
# Open the output from listing 9.3 and get the band.
ds = gdal.Open('dem_class2.tif')
band = ds.GetRasterBand(1)
# Change the NoData value to -1 so that the histogram will be computed
# using 0 values.
band.SetNoDataValue(-1)
# Create the raster attribute table and add 3 columns for the pixel value,
# number of pixels with that value, and elevation label.
rat = gdal.RasterAttributeTable()
rat.CreateColumn('Value', gdal.GFT_Integer, gdal.GFU_Name)
rat.CreateColumn('Count', gdal.GFT_Integer, gdal.GFU_PixelCount)
rat.CreateColumn('Elevation', gdal.GFT_String, gdal.GFU_Generic)
# Add 6 rows to the table, for values 0-5.
rat.SetRowCount(6)
# Write the values 0-5 (using range) to the first column (pixel value).
rat.WriteArray(range(6), 0)
# Get the histogram and write the results to the second column (count).
rat.WriteArray(band.GetHistogram(-0.5, 5.5, 6, False, False), 1)
# Add the labels for each pixel value to the third column.
rat.SetValueAsString(1, 2, '0 - 800')
def _gdal_api_proxy_sub():
src_ds = gdal.Open('data/byte.tif')
src_cs = src_ds.GetRasterBand(1).Checksum()
src_gt = src_ds.GetGeoTransform()
src_prj = src_ds.GetProjectionRef()
src_data = src_ds.ReadRaster(0, 0, 20, 20)
src_md = src_ds.GetMetadata()
src_ds = None
drv = gdal.IdentifyDriver('data/byte.tif')
assert drv.GetDescription() == 'API_PROXY'
ds = gdal.GetDriverByName('GTiff').Create('tmp/byte.tif', 1, 1, 3)
ds = None
src_ds = gdal.Open('data/byte.tif')
ds = gdal.GetDriverByName('GTiff').CreateCopy('tmp/byte.tif',
src_ds,
options=['TILED=YES'])
got_cs = ds.GetRasterBand(1).Checksum()
assert src_cs == got_cs
ds = None
ds = gdal.Open('tmp/byte.tif', gdal.GA_Update)
ds.SetGeoTransform([1, 2, 3, 4, 5, 6])
got_gt = ds.GetGeoTransform()
assert src_gt != got_gt
ds.SetGeoTransform(src_gt)
got_gt = ds.GetGeoTransform()
assert src_gt == got_gt
assert ds.GetGCPCount() == 0
assert ds.GetGCPProjection() == '', ds.GetGCPProjection()
assert not ds.GetGCPs()
gcps = [gdal.GCP(0, 1, 2, 3, 4)]
sr = osr.SpatialReference()
sr.ImportFromEPSG(4326)
wkt = sr.ExportToWkt()
assert ds.SetGCPs(gcps, wkt) == 0
got_gcps = ds.GetGCPs()
assert len(got_gcps) == 1
assert (got_gcps[0].GCPLine == gcps[0].GCPLine and \
got_gcps[0].GCPPixel == gcps[0].GCPPixel and \
got_gcps[0].GCPX == gcps[0].GCPX and \
got_gcps[0].GCPY == gcps[0].GCPY)
assert ds.GetGCPProjection() == wkt
ds.SetGCPs([], "")
assert not ds.GetGCPs()
ds.SetProjection('')
got_prj = ds.GetProjectionRef()
assert src_prj != got_prj
ds.SetProjection(src_prj)
got_prj = ds.GetProjectionRef()
assert src_prj == got_prj
ds.GetRasterBand(1).Fill(0)
got_cs = ds.GetRasterBand(1).Checksum()
assert got_cs == 0
ds.GetRasterBand(1).WriteRaster(0, 0, 20, 20, src_data)
got_cs = ds.GetRasterBand(1).Checksum()
assert src_cs == got_cs
ds.GetRasterBand(1).Fill(0)
got_cs = ds.GetRasterBand(1).Checksum()
assert got_cs == 0
ds.WriteRaster(0, 0, 20, 20, src_data)
got_cs = ds.GetRasterBand(1).Checksum()
assert src_cs == got_cs
# Not bound to SWIG
# ds.AdviseRead(0,0,20,20,20,20)
got_data = ds.ReadRaster(0, 0, 20, 20)
assert src_data == got_data
got_data = ds.GetRasterBand(1).ReadRaster(0, 0, 20, 20)
assert src_data == got_data
got_data_weird_spacing = ds.ReadRaster(0,
0,
20,
20,
buf_pixel_space=1,
buf_line_space=32)
assert len(got_data_weird_spacing) == 32 * (20 - 1) + 20
assert got_data[20:20 + 20] == got_data_weird_spacing[32:32 + 20]
got_data_weird_spacing = ds.GetRasterBand(1).ReadRaster(0,
0,
20,
20,
buf_pixel_space=1,
buf_line_space=32)
assert len(got_data_weird_spacing) == 32 * (20 - 1) + 20
assert got_data[20:20 + 20] == got_data_weird_spacing[32:32 + 20]
got_block = ds.GetRasterBand(1).ReadBlock(0, 0)
assert len(got_block) == 256 * 256
assert got_data[20:20 + 20] == got_block[256:256 + 20]
ds.FlushCache()
ds.GetRasterBand(1).FlushCache()
got_data = ds.GetRasterBand(1).ReadRaster(0, 0, 20, 20)
assert src_data == got_data
assert len(ds.GetFileList()) == 1
assert ds.AddBand(gdal.GDT_Byte) != 0
got_md = ds.GetMetadata()
assert src_md == got_md
assert ds.GetMetadataItem('AREA_OR_POINT') == 'Area'
assert ds.GetMetadataItem('foo') is None
ds.SetMetadataItem('foo', 'bar')
assert ds.GetMetadataItem('foo') == 'bar'
ds.SetMetadata({'foo': 'baz'}, 'OTHER')
assert ds.GetMetadataItem('foo', 'OTHER') == 'baz'
ds.GetRasterBand(1).SetMetadata({'foo': 'baw'}, 'OTHER')
assert ds.GetRasterBand(1).GetMetadataItem('foo', 'OTHER') == 'baw'
assert ds.GetMetadataItem('INTERLEAVE', 'IMAGE_STRUCTURE') == 'BAND'
assert not ds.GetRasterBand(1).GetMetadata()
assert ds.GetRasterBand(1).GetMetadataItem('foo') is None
ds.GetRasterBand(1).SetMetadataItem('foo', 'baz')
assert ds.GetRasterBand(1).GetMetadataItem('foo') == 'baz'
ds.GetRasterBand(1).SetMetadata({'foo': 'baw'})
assert ds.GetRasterBand(1).GetMetadataItem('foo') == 'baw'
assert ds.GetRasterBand(1).GetColorInterpretation() == gdal.GCI_GrayIndex
ds.GetRasterBand(1).SetColorInterpretation(gdal.GCI_Undefined)
ct = ds.GetRasterBand(1).GetColorTable()
assert ct is None
ct = gdal.ColorTable()
ct.SetColorEntry(0, (1, 2, 3))
assert ds.GetRasterBand(1).SetColorTable(ct) == 0
ct = ds.GetRasterBand(1).GetColorTable()
assert ct is not None
assert ct.GetColorEntry(0) == (1, 2, 3, 255)
ct = ds.GetRasterBand(1).GetColorTable()
assert ct is not None
assert ds.GetRasterBand(1).SetColorTable(None) == 0
ct = ds.GetRasterBand(1).GetColorTable()
assert ct is None
rat = ds.GetRasterBand(1).GetDefaultRAT()
assert rat is None
assert ds.GetRasterBand(1).SetDefaultRAT(None) == 0
ref_rat = gdal.RasterAttributeTable()
assert ds.GetRasterBand(1).SetDefaultRAT(ref_rat) == 0
rat = ds.GetRasterBand(1).GetDefaultRAT()
assert rat is None
assert ds.GetRasterBand(1).SetDefaultRAT(None) == 0
rat = ds.GetRasterBand(1).GetDefaultRAT()
assert rat is None
assert ds.GetRasterBand(1).GetMinimum() is None
got_stats = ds.GetRasterBand(1).GetStatistics(0, 0)
assert got_stats[3] < 0.0
got_stats = ds.GetRasterBand(1).GetStatistics(1, 1)
assert got_stats[0] == 74.0
assert ds.GetRasterBand(1).GetMinimum() == 74.0
assert ds.GetRasterBand(1).GetMaximum() == 255.0
ds.GetRasterBand(1).SetStatistics(1, 2, 3, 4)
got_stats = ds.GetRasterBand(1).GetStatistics(1, 1)
assert got_stats == [1, 2, 3, 4]
ds.GetRasterBand(1).ComputeStatistics(0)
got_stats = ds.GetRasterBand(1).GetStatistics(1, 1)
assert got_stats[0] == 74.0
minmax = ds.GetRasterBand(1).ComputeRasterMinMax()
assert minmax == (74.0, 255.0)
assert ds.GetRasterBand(1).GetOffset() == 0.0
assert ds.GetRasterBand(1).GetScale() == 1.0
ds.GetRasterBand(1).SetOffset(10.0)
assert ds.GetRasterBand(1).GetOffset() == 10.0
ds.GetRasterBand(1).SetScale(2.0)
assert ds.GetRasterBand(1).GetScale() == 2.0
ds.BuildOverviews('NEAR', [2])
assert ds.GetRasterBand(1).GetOverviewCount() == 1
assert ds.GetRasterBand(1).GetOverview(-1) is None
assert ds.GetRasterBand(1).GetOverview(0) is not None
assert ds.GetRasterBand(1).GetOverview(0) is not None
got_hist = ds.GetRasterBand(1).GetHistogram()
assert len(got_hist) == 256
(minval, maxval, nitems,
got_hist2) = ds.GetRasterBand(1).GetDefaultHistogram()
assert minval == -0.5
assert maxval == 255.5
assert nitems == 256
assert got_hist == got_hist2
ds.GetRasterBand(1).SetDefaultHistogram(1, 2, [3])
(minval, maxval, nitems,
got_hist3) = ds.GetRasterBand(1).GetDefaultHistogram()
assert minval == 1
assert maxval == 2
assert nitems == 1
assert got_hist3[0] == 3
got_nodatavalue = ds.GetRasterBand(1).GetNoDataValue()
assert got_nodatavalue is None
ds.GetRasterBand(1).SetNoDataValue(123)
got_nodatavalue = ds.GetRasterBand(1).GetNoDataValue()
assert got_nodatavalue == 123
assert ds.GetRasterBand(1).GetMaskFlags() == 8
assert ds.GetRasterBand(1).GetMaskBand() is not None
ret = ds.GetRasterBand(1).DeleteNoDataValue()
assert ret == 0
got_nodatavalue = ds.GetRasterBand(1).GetNoDataValue()
assert got_nodatavalue is None
ds.CreateMaskBand(0)
assert ds.GetRasterBand(1).GetMaskFlags() == 2
assert ds.GetRasterBand(1).GetMaskBand() is not None
ds.GetRasterBand(1).CreateMaskBand(0)
assert ds.GetRasterBand(1).HasArbitraryOverviews() == 0
ds.GetRasterBand(1).SetUnitType('foo')
assert ds.GetRasterBand(1).GetUnitType() == 'foo'
assert ds.GetRasterBand(1).GetCategoryNames() is None
ds.GetRasterBand(1).SetCategoryNames(['foo'])
assert ds.GetRasterBand(1).GetCategoryNames() == ['foo']
ds.GetRasterBand(1).SetDescription('bar')
ds = None
gdal.GetDriverByName('GTiff').Delete('tmp/byte.tif')
def findBurnScars(self,
bp_image,
seed_prob_thresh=97.5,
seed_size_thresh=5,
flood_fill_prob_thresh=75,
log_handler=None):
"""Identify the seeds for burn scars from the input burn probabilities.
Description: routine to find burn scars using the flood-fill approach.
Seed pixels are found by using the seed threshold. Any pixels with
a probability higher than the seed threshold are identified as seed
pixels. Any area with more than the seed size threshold is used as
a seed area for growing the burn extent using the flood fill
process. Areas with fewer than the seed size are ignored and not
flagged as burn areas.
History:
Created in 2013 by Jodi Riegle and Todd Hawbaker, USGS Rocky Mountain
Geographic Science Center
Updated on Nov. 26, 2013 by Gail Schmidt, USGS/EROS LSRD Project
Modified to use int32 arrays vs. int64 arrays for the burn
classification images
Updated on Feb. 13, 2015 by Gail Schmidt, USGS/EROS LSRD Project
Modified the skimage.measure.regionprops call to not use the
deprecated 'properties' parameter. Also changed the properties
values to match the correct names of the dynamic list of props
which is now created.
Args:
bp_image - input image of burn probabilities
seed_prob_thresh - threshold to be used to identify burn pixels
in the burn probability image as seed pixels for the burn area;
default is 97.5%
seed_size_thresh - threshold to be used to identify burn areas in the
probability image. If the count of burn probability seed pixels is
greater than this threshold for a certain area then the burn area
is left and flood-filled. If the count is less than the seed
threshold, then this is a false positive and the area is not used
in the burn classification. Default is 5 pixels.
flood_fill_prob_thresh - threshold to be used to add burn pixels
from the burn probability image to the burn classification via
flood filling; default is 75%
log_handler - file handler for the log file; if this is None then
informational/error messages will be written to stdout
Returns:
nFill - number of pixels that were flood filled
Notes:
1. The default lower threshold for flood filling is 75% burn
probability.
"""
# set up array to hold filled region labels, initialize to 0s which
# means unburned
bp_regions = numpy.zeros_like(bp_image, dtype=numpy.int32)
# find regions to start the flood fill from; these regions are seed
# pixels that are greater than the seed probability threshold
bp_seeds = bp_image >= seed_prob_thresh
# group the seed pixels into regions of connected components; these
# regions will be the start of the flood-fill algorithm
bp_seed_regions = numpy.zeros_like(bp_seeds, dtype=numpy.int32)
n_seed_labels = scipy.ndimage.label(bp_seeds, output=bp_seed_regions)
msg = 'Found %d seeds to use for flood fill' % n_seed_labels
logIt(msg, log_handler)
# get list of region pixel coordinates, use the first pixel from each
# as the seed for the region
bp_region_coords = skimage.measure.regionprops( \
label_image=bp_seed_regions)
# loop through regions and flood fill to expand them where they are of
# an appropriate size
for i in range(0, len(bp_region_coords)):
temp_label = bp_region_coords[i]['label']
temp_area = bp_region_coords[i]['area']
temp_coords = bp_region_coords[i]['coords'][0]
# if the number of pixels in this region exceeds the seed size
# threshold then process the region by flood-filling to grow the
# burn area
if temp_area >= seed_size_thresh:
col = temp_coords[1]
row = temp_coords[0]
nFilled = self.floodFill(input_image=bp_image, row=row, \
col=col, output_image=bp_regions, output_label=temp_label, \
local_threshold=flood_fill_prob_thresh, nodata=-9999)
if False:
print '#############################################' \
'###############'
print 'Seed region label:', temp_label
print 'Number of pixels in region:', temp_area
print 'First coordinate:', temp_coords
print 'Filled pixels:', nFilled
# find region properties for the flood filled burn areas
bc2 = bp_regions > 0
bp_regions2 = numpy.zeros_like(bc2, dtype=numpy.int32)
n_labels = scipy.ndimage.label(bc2, output=bp_regions2)
prop_names = ['area','filled_area','max_intensity','mean_intensity', \
'min_intensity']
bp_region2_props = skimage.measure.regionprops( \
label_image=bp_regions2, intensity_image=bp_image)
# define the RAT (raster attribute table)
#print 'Creating raster attribute table...'
label_rat = gdal.RasterAttributeTable()
label_rat.CreateColumn("Value", gdalconst.GFT_Integer, \
gdalconst.GFU_MinMax)
for prop in prop_names:
label_rat.CreateColumn(prop, gdalconst.GFT_Real, \
gdalconst.GFU_MinMax)
# resize the RAT
label_rat.SetRowCount(n_labels)
# set values in the RAT
#print 'Populating raster attribute table...'
for i in range(0, n_labels):
# label id
label_rat.SetValueAsInt(i, 0, bp_region2_props[i]['label'])
for j in range(0, len(prop_names)):
temp_prop = prop_names[j]
label_rat.SetValueAsDouble(i, j+1, \
float(bp_region2_props[i][temp_prop]))
return ([bp_regions2, label_rat])
def make_pastures_mask(self, raster_fn, ranch, dst_fn, nodata=-9999):
"""
:param raster_fn: utm raster
:param ranches:
:param dst_fn:
:param nodata:
:return:
"""
assert _exists(_split(dst_fn)[0])
assert dst_fn.endswith('.tif')
ds = rasterio.open(raster_fn)
ds_proj4 = ds.crs.to_proj4()
loc_path = self.loc_path
_d = self._d
sf_fn = _join(loc_path, _d['sf_fn'])
sf_feature_properties_key = _d['sf_feature_properties_key']
sf_fn = os.path.abspath(sf_fn)
sf = fiona.open(sf_fn, 'r')
reverse_key = self.reverse_key
pastures = {}
pastures_mask = np.zeros(ds.shape, dtype=np.uint16)
for feature in sf:
properties = feature['properties']
key = properties[sf_feature_properties_key].replace(' ', '_')
if not reverse_key:
_pasture, _ranch = key.split(self.key_delimiter)
else:
_ranch, _pasture = key.split(self.key_delimiter)
if _ranch.lower() != ranch.lower():
continue
if _pasture not in pastures:
pastures[_pasture] = len(pastures) + 1
# true where valid
_features = transform_geom(sf.crs_wkt, ds_proj4,
feature['geometry'])
_mask, _, _ = raster_geometry_mask(ds, [_features])
k = pastures[_pasture]
# update pastures_mask
pastures_mask[np.where(_mask == False)] = k
utm_dst_fn = ''
try:
head, tail = _split(dst_fn)
utm_dst_fn = _join(head, tail.replace('.tif', '.utm.tif'))
dst_vrt_fn = _join(head, tail.replace('.tif', '.wgs.vrt'))
dst_wgs_fn = _join(head, tail.replace('.tif', '.wgs.tif'))
with rasterio.Env():
profile = ds.profile
dtype = rasterio.uint16
profile.update(count=1,
dtype=rasterio.uint16,
nodata=nodata,
compress='lzw')
with rasterio.open(utm_dst_fn, 'w', **profile) as dst:
dst.write(pastures_mask.astype(dtype), 1)
assert _exists(utm_dst_fn)
except:
raise
try:
if _exists(dst_vrt_fn):
os.remove(dst_vrt_fn)
cmd = [
'gdalwarp', '-t_srs', 'EPSG:4326', '-of', 'vrt', utm_dst_fn,
dst_vrt_fn
]
p = Popen(cmd)
p.wait()
assert _exists(dst_vrt_fn)
except:
if _exists(dst_vrt_fn):
os.remove(dst_vrt_fn)
raise
try:
if _exists(dst_fn):
os.remove(dst_fn)
cmd = [
'gdal_translate', '-co', 'COMPRESS=LZW', '-of', 'GTiff',
dst_vrt_fn, dst_wgs_fn
]
p = Popen(cmd)
p.wait()
assert _exists(dst_wgs_fn)
except:
if _exists(dst_wgs_fn):
os.remove(dst_wgs_fn)
raise
if _exists(dst_vrt_fn):
os.remove(dst_vrt_fn)
for OUTPUT_RASTER in (dst_wgs_fn, utm_dst_fn):
# https://gdal.org/python/osgeo.gdal.RasterAttributeTable-class.html
# https://gdal.org/python/osgeo.gdalconst-module.html
ds = gdal.Open(OUTPUT_RASTER)
rb = ds.GetRasterBand(1)
# Create and populate the RAT
rat = gdal.RasterAttributeTable()
rat.CreateColumn('VALUE', gdal.GFT_Integer, gdal.GFU_Generic)
rat.CreateColumn('PASTURE', gdal.GFT_String, gdal.GFU_Generic)
for i, (pasture, key) in enumerate(pastures.items()):
rat.SetValueAsInt(i, 0, key)
rat.SetValueAsString(i, 1, pasture)
# Associate with the band
rb.SetDefaultRAT(rat)
# Close the dataset and persist the RAT
ds = None
def writeColumnToBand(gdalBand,
colName,
sequence,
colType=None,
colUsage=gdal.GFU_Generic):
"""
Given a GDAL band, Writes the data specified in sequence
(can be list, tuple or array etc)
to the named column in the attribute table assocated with the
gdalBand. colType must be one of gdal.GFT_Integer,gdal.GFT_Real,gdal.GFT_String.
can specify one of the gdal.GFU_* constants for colUsage - default is 'generic'
GDAL dataset must have been created, or opened with GA_Update
"""
if colType is None:
colType = inferColumnType(sequence)
if colType is None:
msg = "Can't infer type of column for sequence of %s" % type(
sequence[0])
raise rioserrors.AttributeTableTypeError(msg)
# check it is acually a valid type
elif colType not in (gdal.GFT_Integer, gdal.GFT_Real, gdal.GFT_String):
msg = "coltype must be a valid gdal column type"
raise rioserrors.AttributeTableTypeError(msg)
attrTbl = gdalBand.GetDefaultRAT()
if attrTbl is None:
# some formats eg ENVI return None
# here so we need to be able to cope
attrTbl = gdal.RasterAttributeTable()
isFileRAT = False
else:
isFileRAT = True
# but if it doesn't support dynamic writing
# we still ahve to call SetDefaultRAT
if not attrTbl.ChangesAreWrittenToFile():
isFileRAT = False
# We need to ensure colname doesn't already exist
colExists = False
for n in range(attrTbl.GetColumnCount()):
if attrTbl.GetNameOfCol(n) == colName:
colExists = True
colNum = n
break
if not colExists:
# preserve usage
attrTbl.CreateColumn(colName, colType, colUsage)
colNum = attrTbl.GetColumnCount() - 1
rowsToAdd = len(sequence)
# Imagine has trouble if not 256 items for byte
if gdalBand.DataType == gdal.GDT_Byte:
rowsToAdd = 256
# another hack to hide float (0-1) and int (0-255)
# color table handling.
# we assume that the column has already been created
# of the right type appropriate for the format (maybe by calcstats)
usage = attrTbl.GetUsageOfCol(colNum)
if (isColorColFromUsage(usage)
and attrTbl.GetTypeOfCol(colNum) == gdal.GFT_Real
and colType == gdal.GFT_Integer):
sequence = numpy.array(sequence, dtype=numpy.float)
sequence = sequence / 255.0
attrTbl.SetRowCount(rowsToAdd)
attrTbl.WriteArray(sequence, colNum)
if not isFileRAT:
# assume existing bands re-written
# Use GDAL's exceptions to trap the error message which arises when
# writing to a format which does not support it
usingExceptions = gdal.GetUseExceptions()
gdal.UseExceptions()
try:
gdalBand.SetDefaultRAT(attrTbl)
except Exception:
pass
if not usingExceptions:
gdal.DontUseExceptions()
def run(self):
"""Run method that performs all the real work"""
# show the dialog
self.dlg.show()
# add all raster layers in current session to UI as potential inputs
layers = QgsMapLayerRegistry.instance().mapLayers().values()
for layer in layers:
if layer.type() == QgsMapLayer.RasterLayer:
self.dlg.rasterBox.addItem(layer.name(), layer)
# Run the dialog event loop
result = self.dlg.exec_()
# TODO: add logic to auto-detect band and sensor using input_raster
# See if OK was pressed
if result:
# get variable names from input
input_raster = str(self.dlg.rasterBox.currentText())
band = str(self.dlg.bandBox.currentText())
sensor = str(self.dlg.sensorBox.currentText())
rm_low = self.dlg.rmLowBox.isChecked()
# use gdal to get unique values
ds = gdal.Open(input_raster)
rb = ds.GetRasterBand(1)
values = sorted(list(np.unique(np.array(rb.ReadAsArray()))))
#ds = None
# define lookup table
bit_flags = lookup_dict.bit_flags
#qa_values = lookup_dict.qa_values
# convert input_sensor to sensor values used in qa_values
if sensor == "Landsat 4-5, 7":
sens = "L47"
elif sensor == "Landsat 8":
sens = "L8"
else:
sys.exit("Incorrect sensor provided. Input: {0}; Potential "
"options: Landsat 4-5, 7; Landsat 8".format(sensor))
# get all possible bit values for sensor and band combination
bit_values = sorted(bit_flags[band][sens].values())
qa_labels = []
for row in values:
bit_bool = []
for bv in bit_values:
if len(bv) == 1: # single bit
bit_bool.append(row & 1 << bv[0] > 0)
elif len(bv) > 1: # 2+ bits
bits = []
for b in bv:
bits.append(row & 1 << b > 0)
if all(item == True for item in bits):
bit_bool.append(True)
else:
bit_bool.append(False)
else:
sys.exit("No valid bits found for target band.")
'''
NEW logic for getting labels using bit wise dictionary
'''
# create description of each value based upon all possible bits
true_bits = [i for (i, bb) in zip(bit_values, bit_bool) if bb]
# if double bits exist, eliminate single bit descriptions,
# otherwise, the descriptions will duplicate themselves.
bb_double = [len(i) > 1 for i in true_bits]
if any(bb_double):
# get only the double bits
dbit_nest = [
i for (i, db) in zip(true_bits, bb_double) if db
]
# collapse the bits into a single list
dbits = [item for sublist in dbit_nest for item in sublist]
# remove matching single bits out of true_bits list
tbo = []
for t in true_bits:
tb_out = []
for d in dbits:
if t[0] != d or len(t) > 1:
tb_out.append(True)
else:
tb_out.append(False)
if all(tb_out):
tbo.append(t)
# replace true_bits with filtered list
true_bits = tbo
def get_label(bits):
"""
Generate label for value in attribute table.
:param bits: <list> List of True or False for bit position
:return: <str> Attribute label
"""
if len(bits) == 0:
if band == 'radsat_qa':
return 'No Saturation'
elif band == 'sr_cloud_qa' or band == 'sr_aerosol':
return 'None'
elif band == 'BQA':
return 'Not Determined'
# build description from all bits represented in value
desc = []
for tb in bits:
k = next(
key
for key, value in bit_flags[band][sens].items()
if value == tb)
# if 'low' labels are disabled, do not add them here
if rm_low and band != 'BQA' and 'low' in k.lower():
continue
# if last check, and not radiometric sat, set to 'clear'
elif rm_low and band == 'BQA' and 'low' in k.lower() \
and tb == bits[-1] and \
'radiometric' not in k.lower() and \
not desc:
k = 'Clear'
# if BQA and bit is low radiometric sat, keep it
elif rm_low and band == 'BQA' and 'low' in k.lower():
if 'radiometric' not in k.lower():
continue
# if radsat_qa, handle differently to display cleaner
if band == 'radsat_qa':
if not desc:
desc = "Band {0} Data Saturation".format(tb[0])
else:
desc = "{0},{1} Data Saturation".format(
desc[:desc.find('Data') - 1], tb[0])
# string creation for all other bands
else:
if not desc:
desc = "{0}".format(k)
else:
desc += ", {0}".format(k)
# final check to make sure something was set
if not desc:
desc = 'ERROR: bit set incorrectly'
return desc
# add desc to row description
qa_labels.append(get_label(true_bits))
'''
OLD logic for getting lookup values
# use unique raster values (and sensor+band pair) to get defs
if band == 'radsat_qa':
qa_labels = {i:qa_values[band][i] for i in qa_values[band] if i
in list(values)}
elif band == 'pixel_qa' and sens == 'L8': # terrain occl. check
qa_labels = {}
for i in qa_values[band]:
if i >= 1024:
qa_labels[i] = 'Terrain occlusion'
else:
qa_labels[i] = qa_values[band][sens][i]
else:
qa_labels = {i:qa_values[band][sens][i] for i in
qa_values[band][sens] if i in list(values)}
'''
'''
Use gdal.RasterAttributeTable to embed qa values in raster
'''
# create table
rat = gdal.RasterAttributeTable()
# get column count (for indexing columns)
rat_cc = rat.GetColumnCount()
# add 'value' and 'descr' columns to table
rat.CreateColumn("Value", gdalconst.GFT_Integer,
gdalconst.GFU_MinMax)
rat.CreateColumn("Descr", gdalconst.GFT_String,
gdalconst.GFU_MinMax)
# populate table with contents of 'qa_labels'
uid = 0
for val, lab in zip(values, qa_labels):
# 'value' column
rat.SetValueAsInt(uid, rat_cc, int(val))
# 'descr' column
rat.SetValueAsString(uid, rat_cc + 1, lab)
uid += 1
# set raster attribute table to raster
rb.SetDefaultRAT(rat)
'''
METHOD 1: use RasterAttributeTable to display values.
QGIS' UI does not currently support reading Attribute Tables
embedded in raster datasets. Instead, we'll assign labels and
random colors to the raster's color palette in the QGIS UI.
Feature request: https://issues.qgis.org/issues/4321
# open raster with QGIS API
q_raster = QgsRasterLayer(input_raster,
os.path.basename(input_raster))
# make sure the raster is valid
if not q_raster.isValid():
sys.exit("Layer {0} not valid!".format(input_raster))
# save changes and close raster
ds = None
# add raster to QGIS interface
QgsMapLayerRegistry.instance().addMapLayer(q_raster)
'''
'''
METHOD 2: re-assign colors in QGIS
'''
# open raster
q_raster = QgsRasterLayer(input_raster,
os.path.basename(input_raster))
if not q_raster.isValid():
sys.exit("Layer {0} not valid!".format(input_raster))
# define color shader
shader = QgsRasterShader()
# define ramp for color shader
c_ramp_shader = QgsColorRampShader()
c_ramp_shader.setColorRampType(QgsColorRampShader.EXACT)
# assign a random color to each value, and apply label
c_ramp_vals = []
for val, lab in zip(values, qa_labels):
c_ramp_vals.append(
QgsColorRampShader.ColorRampItem(
float(val), QColor('#%06x' % randint(0, 2**24)), lab))
# apply new color/label combo to color ramps
c_ramp_shader.setColorRampItemList(c_ramp_vals)
shader.setRasterShaderFunction(c_ramp_shader)
# apply color ramps to raster
ps_ramp = QgsSingleBandPseudoColorRenderer(q_raster.dataProvider(),
1, shader)
q_raster.setRenderer(ps_ramp)
# add raster to QGIS interface
QgsMapLayerRegistry.instance().addMapLayer(q_raster)
def gdal_api_proxy_sub():
src_ds = gdal.Open('data/byte.tif')
src_cs = src_ds.GetRasterBand(1).Checksum()
src_gt = src_ds.GetGeoTransform()
src_prj = src_ds.GetProjectionRef()
src_data = src_ds.ReadRaster(0, 0, 20, 20)
src_md = src_ds.GetMetadata()
src_ds = None
drv = gdal.IdentifyDriver('data/byte.tif')
if drv.GetDescription() != 'API_PROXY':
gdaltest.post_reason('fail')
return 'fail'
ds = gdal.GetDriverByName('GTiff').Create('tmp/byte.tif', 1, 1, 3)
ds = None
src_ds = gdal.Open('data/byte.tif')
ds = gdal.GetDriverByName('GTiff').CreateCopy('tmp/byte.tif',
src_ds,
options=['TILED=YES'])
got_cs = ds.GetRasterBand(1).Checksum()
if src_cs != got_cs:
gdaltest.post_reason('fail')
return 'fail'
ds = None
ds = gdal.Open('tmp/byte.tif', gdal.GA_Update)
ds.SetGeoTransform([1, 2, 3, 4, 5, 6])
got_gt = ds.GetGeoTransform()
if src_gt == got_gt:
gdaltest.post_reason('fail')
return 'fail'
ds.SetGeoTransform(src_gt)
got_gt = ds.GetGeoTransform()
if src_gt != got_gt:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetGCPCount() != 0:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetGCPProjection() != '':
print(ds.GetGCPProjection())
gdaltest.post_reason('fail')
return 'fail'
if len(ds.GetGCPs()) != 0:
gdaltest.post_reason('fail')
return 'fail'
gcps = [gdal.GCP(0, 1, 2, 3, 4)]
ds.SetGCPs(gcps, "foo")
got_gcps = ds.GetGCPs()
if len(got_gcps) != 1:
gdaltest.post_reason('fail')
return 'fail'
if got_gcps[0].GCPLine != gcps[0].GCPLine or \
got_gcps[0].GCPPixel != gcps[0].GCPPixel or \
got_gcps[0].GCPX != gcps[0].GCPX or \
got_gcps[0].GCPY != gcps[0].GCPY:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetGCPProjection() != 'foo':
gdaltest.post_reason('fail')
print(ds.GetGCPProjection())
return 'fail'
ds.SetGCPs([], "")
if len(ds.GetGCPs()) != 0:
gdaltest.post_reason('fail')
return 'fail'
ds.SetProjection('')
got_prj = ds.GetProjectionRef()
if src_prj == got_prj:
gdaltest.post_reason('fail')
return 'fail'
ds.SetProjection(src_prj)
got_prj = ds.GetProjectionRef()
if src_prj != got_prj:
gdaltest.post_reason('fail')
print(src_prj)
print(got_prj)
return 'fail'
ds.GetRasterBand(1).Fill(0)
got_cs = ds.GetRasterBand(1).Checksum()
if 0 != got_cs:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).WriteRaster(0, 0, 20, 20, src_data)
got_cs = ds.GetRasterBand(1).Checksum()
if src_cs != got_cs:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).Fill(0)
got_cs = ds.GetRasterBand(1).Checksum()
if 0 != got_cs:
gdaltest.post_reason('fail')
return 'fail'
ds.WriteRaster(0, 0, 20, 20, src_data)
got_cs = ds.GetRasterBand(1).Checksum()
if src_cs != got_cs:
gdaltest.post_reason('fail')
return 'fail'
# Not bound to SWIG
# ds.AdviseRead(0,0,20,20,20,20)
got_data = ds.ReadRaster(0, 0, 20, 20)
if src_data != got_data:
gdaltest.post_reason('fail')
return 'fail'
got_data = ds.GetRasterBand(1).ReadRaster(0, 0, 20, 20)
if src_data != got_data:
gdaltest.post_reason('fail')
return 'fail'
got_data_weird_spacing = ds.ReadRaster(0,
0,
20,
20,
buf_pixel_space=1,
buf_line_space=32)
if len(got_data_weird_spacing) != 32 * (20 - 1) + 20:
gdaltest.post_reason('fail')
print(len(got_data_weird_spacing))
return 'fail'
if got_data[20:20 + 20] != got_data_weird_spacing[32:32 + 20]:
gdaltest.post_reason('fail')
return 'fail'
got_data_weird_spacing = ds.GetRasterBand(1).ReadRaster(0,
0,
20,
20,
buf_pixel_space=1,
buf_line_space=32)
if len(got_data_weird_spacing) != 32 * (20 - 1) + 20:
gdaltest.post_reason('fail')
print(len(got_data_weird_spacing))
return 'fail'
if got_data[20:20 + 20] != got_data_weird_spacing[32:32 + 20]:
gdaltest.post_reason('fail')
return 'fail'
got_block = ds.GetRasterBand(1).ReadBlock(0, 0)
if len(got_block) != 256 * 256:
gdaltest.post_reason('fail')
return 'fail'
if got_data[20:20 + 20] != got_block[256:256 + 20]:
gdaltest.post_reason('fail')
return 'fail'
ds.FlushCache()
ds.GetRasterBand(1).FlushCache()
got_data = ds.GetRasterBand(1).ReadRaster(0, 0, 20, 20)
if src_data != got_data:
gdaltest.post_reason('fail')
return 'fail'
if len(ds.GetFileList()) != 1:
gdaltest.post_reason('fail')
return 'fail'
if ds.AddBand(gdal.GDT_Byte) == 0:
gdaltest.post_reason('fail')
return 'fail'
got_md = ds.GetMetadata()
if src_md != got_md:
gdaltest.post_reason('fail')
print(src_md)
print(got_md)
return 'fail'
if ds.GetMetadataItem('AREA_OR_POINT') != 'Area':
gdaltest.post_reason('fail')
return 'fail'
if ds.GetMetadataItem('foo') is not None:
gdaltest.post_reason('fail')
return 'fail'
ds.SetMetadataItem('foo', 'bar')
if ds.GetMetadataItem('foo') != 'bar':
gdaltest.post_reason('fail')
return 'fail'
ds.SetMetadata({'foo': 'baz'}, 'OTHER')
if ds.GetMetadataItem('foo', 'OTHER') != 'baz':
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetMetadata({'foo': 'baw'}, 'OTHER')
if ds.GetRasterBand(1).GetMetadataItem('foo', 'OTHER') != 'baw':
gdaltest.post_reason('fail')
return 'fail'
if ds.GetMetadataItem('INTERLEAVE', 'IMAGE_STRUCTURE') != 'BAND':
gdaltest.post_reason('fail')
return 'fail'
if len(ds.GetRasterBand(1).GetMetadata()) != 0:
gdaltest.post_reason('fail')
print(ds.GetRasterBand(1).GetMetadata())
return 'fail'
if ds.GetRasterBand(1).GetMetadataItem('foo') is not None:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetMetadataItem('foo', 'baz')
if ds.GetRasterBand(1).GetMetadataItem('foo') != 'baz':
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetMetadata({'foo': 'baw'})
if ds.GetRasterBand(1).GetMetadataItem('foo') != 'baw':
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetColorInterpretation() != gdal.GCI_GrayIndex:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetColorInterpretation(gdal.GCI_Undefined)
ct = ds.GetRasterBand(1).GetColorTable()
if ct is not None:
gdaltest.post_reason('fail')
return 'fail'
ct = gdal.ColorTable()
ct.SetColorEntry(0, (1, 2, 3))
if ds.GetRasterBand(1).SetColorTable(ct) != 0:
gdaltest.post_reason('fail')
return 'fail'
ct = ds.GetRasterBand(1).GetColorTable()
if ct is None:
gdaltest.post_reason('fail')
return 'fail'
if ct.GetColorEntry(0) != (1, 2, 3, 255):
gdaltest.post_reason('fail')
return 'fail'
ct = ds.GetRasterBand(1).GetColorTable()
if ct is None:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).SetColorTable(None) != 0:
gdaltest.post_reason('fail')
return 'fail'
ct = ds.GetRasterBand(1).GetColorTable()
if ct is not None:
gdaltest.post_reason('fail')
return 'fail'
rat = ds.GetRasterBand(1).GetDefaultRAT()
if rat is not None:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).SetDefaultRAT(None) != 0:
gdaltest.post_reason('fail')
return 'fail'
ref_rat = gdal.RasterAttributeTable()
if ds.GetRasterBand(1).SetDefaultRAT(ref_rat) != 0:
gdaltest.post_reason('fail')
return 'fail'
rat = ds.GetRasterBand(1).GetDefaultRAT()
if rat is not None:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).SetDefaultRAT(None) != 0:
gdaltest.post_reason('fail')
return 'fail'
rat = ds.GetRasterBand(1).GetDefaultRAT()
if rat is not None:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetMinimum() is not None:
gdaltest.post_reason('fail')
return 'fail'
got_stats = ds.GetRasterBand(1).GetStatistics(0, 0)
if got_stats[3] >= 0.0:
gdaltest.post_reason('fail')
return 'fail'
got_stats = ds.GetRasterBand(1).GetStatistics(1, 1)
if got_stats[0] != 74.0:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetMinimum() != 74.0:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetMaximum() != 255.0:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetStatistics(1, 2, 3, 4)
got_stats = ds.GetRasterBand(1).GetStatistics(1, 1)
if got_stats != [1, 2, 3, 4]:
print(got_stats)
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).ComputeStatistics(0)
got_stats = ds.GetRasterBand(1).GetStatistics(1, 1)
if got_stats[0] != 74.0:
gdaltest.post_reason('fail')
return 'fail'
minmax = ds.GetRasterBand(1).ComputeRasterMinMax()
if minmax != (74.0, 255.0):
gdaltest.post_reason('fail')
print(minmax)
return 'fail'
if ds.GetRasterBand(1).GetOffset() != 0.0:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetScale() != 1.0:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetOffset(10.0)
if ds.GetRasterBand(1).GetOffset() != 10.0:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetScale(2.0)
if ds.GetRasterBand(1).GetScale() != 2.0:
gdaltest.post_reason('fail')
return 'fail'
ds.BuildOverviews('NEAR', [2])
if ds.GetRasterBand(1).GetOverviewCount() != 1:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetOverview(-1) is not None:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetOverview(0) is None:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetOverview(0) is None:
gdaltest.post_reason('fail')
return 'fail'
got_hist = ds.GetRasterBand(1).GetHistogram()
if len(got_hist) != 256:
gdaltest.post_reason('fail')
return 'fail'
(minval, maxval, nitems,
got_hist2) = ds.GetRasterBand(1).GetDefaultHistogram()
if minval != -0.5:
gdaltest.post_reason('fail')
return 'fail'
if maxval != 255.5:
gdaltest.post_reason('fail')
return 'fail'
if nitems != 256:
gdaltest.post_reason('fail')
return 'fail'
if got_hist != got_hist2:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetDefaultHistogram(1, 2, [3])
(minval, maxval, nitems,
got_hist3) = ds.GetRasterBand(1).GetDefaultHistogram()
if minval != 1:
gdaltest.post_reason('fail')
return 'fail'
if maxval != 2:
gdaltest.post_reason('fail')
return 'fail'
if nitems != 1:
gdaltest.post_reason('fail')
return 'fail'
if got_hist3[0] != 3:
gdaltest.post_reason('fail')
return 'fail'
got_nodatavalue = ds.GetRasterBand(1).GetNoDataValue()
if got_nodatavalue is not None:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetNoDataValue(123)
got_nodatavalue = ds.GetRasterBand(1).GetNoDataValue()
if got_nodatavalue != 123:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetMaskFlags() != 8:
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetMaskBand() is None:
gdaltest.post_reason('fail')
return 'fail'
ret = ds.GetRasterBand(1).DeleteNoDataValue()
if ret != 0:
gdaltest.post_reason('fail')
return 'fail'
got_nodatavalue = ds.GetRasterBand(1).GetNoDataValue()
if got_nodatavalue is not None:
gdaltest.post_reason('fail')
return 'fail'
ds.CreateMaskBand(0)
if ds.GetRasterBand(1).GetMaskFlags() != 2:
print(ds.GetRasterBand(1).GetMaskFlags())
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetMaskBand() is None:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).CreateMaskBand(0)
if ds.GetRasterBand(1).HasArbitraryOverviews() != 0:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetUnitType('foo')
if ds.GetRasterBand(1).GetUnitType() != 'foo':
gdaltest.post_reason('fail')
return 'fail'
if ds.GetRasterBand(1).GetCategoryNames() is not None:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetCategoryNames(['foo'])
if ds.GetRasterBand(1).GetCategoryNames() != ['foo']:
gdaltest.post_reason('fail')
return 'fail'
ds.GetRasterBand(1).SetDescription('bar')
ds = None
gdal.GetDriverByName('GTiff').Delete('tmp/byte.tif')
return 'success'
def writeColumnToBand(gdalBand,
colName,
sequence,
colType=None,
colUsage=gdal.GFU_Generic):
"""
Given a GDAL band, Writes the data specified in sequence
(can be list, tuple or array etc)
to the named column in the attribute table assocated with the
gdalBand. colType must be one of gdal.GFT_Integer,gdal.GFT_Real,gdal.GFT_String.
can specify one of the gdal.GFU_* constants for colUsage - default is 'generic'
GDAL dataset must have been created, or opened with GA_Update
"""
if colType is None:
colType = inferColumnType(sequence)
if colType is None:
msg = "Can't infer type of column for sequence of %s" % type(
sequence[0])
raise rioserrors.AttributeTableTypeError(msg)
# check it is acually a valid type
elif colType not in (gdal.GFT_Integer, gdal.GFT_Real, gdal.GFT_String):
msg = "coltype must be a valid gdal column type"
raise rioserrors.AttributeTableTypeError(msg)
# things get a bit weird here as we need different
# behaviour depending on whether we have an RFC40
# RAT or not.
if hasattr(gdal.RasterAttributeTable, "WriteArray"):
# new behaviour
attrTbl = gdalBand.GetDefaultRAT()
if attrTbl is None:
# some formats eg ENVI return None
# here so we need to be able to cope
attrTbl = gdal.RasterAttributeTable()
isFileRAT = False
else:
isFileRAT = True
# but if it doesn't support dynamic writing
# we still ahve to call SetDefaultRAT
if not attrTbl.ChangesAreWrittenToFile():
isFileRAT = False
else:
# old behaviour
attrTbl = gdal.RasterAttributeTable()
isFileRAT = False
# thanks to RFC40 we need to ensure colname doesn't already exist
colExists = False
for n in range(attrTbl.GetColumnCount()):
if attrTbl.GetNameOfCol(n) == colName:
colExists = True
colNum = n
break
if not colExists:
# preserve usage
attrTbl.CreateColumn(colName, colType, colUsage)
colNum = attrTbl.GetColumnCount() - 1
rowsToAdd = len(sequence)
# Imagine has trouble if not 256 items for byte
if gdalBand.DataType == gdal.GDT_Byte:
rowsToAdd = 256
# another hack to hide float (0-1) and int (0-255)
# color table handling.
# we assume that the column has already been created
# of the right type appropriate for the format (maybe by calcstats)
# Note: this only works post RFC40 when we have an actual reference
# to the RAT rather than a new one so we can ask GetTypeOfCol
usage = attrTbl.GetUsageOfCol(colNum)
if (isColorColFromUsage(usage)
and attrTbl.GetTypeOfCol(colNum) == gdal.GFT_Real
and colType == gdal.GFT_Integer):
sequence = numpy.array(sequence, dtype=numpy.float)
sequence = sequence / 255.0
if hasattr(attrTbl, "WriteArray"):
# if GDAL > 1.10 has these functions
# thanks to RFC40
attrTbl.SetRowCount(rowsToAdd)
attrTbl.WriteArray(sequence, colNum)
elif HAVE_TURBORAT:
# use turborat to write values to RAT if available
if not isinstance(sequence, numpy.ndarray):
# turborat.writeColumn needs an array
sequence = numpy.array(sequence)
# If the dtype of the array is some unicode type, then convert to simple string type,
# as turborat does not cope with the unicode variant.
if 'U' in str(sequence.dtype):
sequence = sequence.astype(numpy.character)
turborat.writeColumn(attrTbl, colNum, sequence, rowsToAdd)
else:
defaultValues = {
gdal.GFT_Integer: 0,
gdal.GFT_Real: 0.0,
gdal.GFT_String: ''
}
# go thru and set each value into the RAT
for rowNum in range(rowsToAdd):
if rowNum >= len(sequence):
# they haven't given us enough values - fill in with default
val = defaultValues[colType]
else:
val = sequence[rowNum]
if colType == gdal.GFT_Integer:
# appears that swig cannot convert numpy.int64
# to the int type required by SetValueAsInt
# so we need to cast.
# This is a problem as readColumn returns numpy.int64
# for integer columns.
# Seems fine converting numpy.float64 to
# float however for SetValueAsDouble.
attrTbl.SetValueAsInt(rowNum, colNum, int(val))
elif colType == gdal.GFT_Real:
attrTbl.SetValueAsDouble(rowNum, colNum, float(val))
else:
attrTbl.SetValueAsString(rowNum, colNum, val)
if not isFileRAT:
# assume existing bands re-written
# Use GDAL's exceptions to trap the error message which arises when
# writing to a format which does not support it
usingExceptions = gdal.GetUseExceptions()
gdal.UseExceptions()
try:
gdalBand.SetDefaultRAT(attrTbl)
except Exception:
pass
if not usingExceptions:
gdal.DontUseExceptions()
def save_as_xml(self, raster_source, band) -> bool:
"""Saves .aux.xml RAT using GDAL
:param raster_source: path of of the raster data file
:type raster_source: str
:param band: band number
:type band: int
:return: TRUE on success
:rtype: bool
"""
ds = gdal.OpenEx(raster_source, gdal.OF_RASTER | gdal.OF_UPDATE)
if ds:
self.band = band
gdal_band = ds.GetRasterBand(band)
if gdal_band:
rat = gdal.RasterAttributeTable()
rat.SetTableType(self.thematic_type)
for field in list(self.fields.values()):
rat.CreateColumn(field.name, field.type, field.usage)
type_map = {gdal.GFT_Integer: 'Int',
gdal.GFT_Real: 'Double', gdal.GFT_String: 'String'}
column_index = 0
for field_name, field in self.fields.items():
values = self.data[field_name]
func = getattr(rat, 'SetValueAs%s' % type_map[field.type])
for row_index in range(len(values)):
rat_log('Writing RAT value as %s, (%s, %s) %s' %
(type_map[field.type], row_index, column_index, values[row_index]))
value = html.escape(values[row_index]) if field.type == gdal.GFT_String else values[row_index]
func(row_index, column_index, value)
column_index += 1
assert rat.GetColumnCount() == len(self.fields)
assert rat.GetRowCount() == len(self.values[0])
# Ugly hack because GDAL does not know about the newly created RAT
for layer in [l for l in QgsProject.instance().mapLayers().values() if l.source() == raster_source]:
RAT._dirty_xml_rats["%s|%s" %
(self.band, self.path)] = self
if layer.id() not in RAT._dirty_xml_layer_ids:
RAT._dirty_xml_layer_ids.append(layer.id())
layer.destroyed.connect(self._restore_xml_rats)
gdal_band.SetDefaultRAT(rat)
ds.FlushCache()
# I don't know why but seems like you need to call this twice or
# the RAT is not really saved into the XML
gdal_band.SetDefaultRAT(rat)
ds.FlushCache()
rat_log('RAT saved as XML for layer %s' % raster_source)
return True
return False