topic Re: Use numpy array for a quercy? in Python Questions

Use numpy array for a quercy?

JohannesBierer — Sun, 12 Dec 2021 05:27:14 GMT

I would like to use numpy array from a excel file for a selection of features. Do I have to convert the array to a list first, and how to do that? With the following code I get the following error? Or some better ideas, many files to select?

#!/usr/bin/env python
# -*- coding: iso-8859-1 -*-
# 

import numpy as np
import arcpy
import pandas as pd
import sys, os

# Encoding der Standardausgabe herausfinden
stdout_encoding = sys.stdout.encoding or sys.getfilesystemencoding()

inTable = r"R:\Karto\Bierer2019\FFH_MEKA_FACT\Tabellen\wk13.xlsx"
outP = r"R:\Karto\Bierer2019\FFH_MEKA_FACT\Shape"
fc = r"\\RPKSF001\GIS_DATA\ArcUISTools15\Data\UIS_ALk.gdb\UIS_0100000047100001"
fl = "ALK"

# Tabelle
Flrstkz = pd.read_excel(inTable, index_col=None, na_values=['NA'], parse_cols = "A")

x = np.array(Flrstkz)

arcpy.MakeFeatureLayer_management(fc, fl)

field = 'FLURSTUECKSKENNZEICHEN'

for i in x:
    print i
        
    sc = "%s = + %s"(field, i)
    print sc

    arcpy.SelectLayerByAttribute_management (fl, "NEW_SELECTION", sc)

    outFc = os.path.join(outP, "WK13_"+ i + ".shp")

    arcpy.CopyFeatures_management(fl, outFc)‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

[8334100002890000384]

Traceback (most recent call last):
File "R:/Karto/zGIS/Python_2019/NumPy/NumPY_MEKA_FAKT.py", line 30, in <module>
sc = "%s = + %s"(field, i)
TypeError: 'str' object is not callable

Better would be the "quercy" in (np.array)?

Re: Use numpy array for a quercy?

JoeBorgione — Tue, 02 Apr 2019 12:12:55 GMT

Try posting your code again: it didn't make the trip.

Re: Use numpy array for a quercy?

JoshuaBixby — Tue, 02 Apr 2019 13:41:29 GMT

You missed a critical part of string formatting on line 30:

sc = "%s = + %s" % (field, i)‍

Re: Use numpy array for a quercy?

JohannesBierer — Tue, 02 Apr 2019 14:16:24 GMT

Sorry, my mistake.

Re: Use numpy array for a quercy?

JoshuaBixby — Tue, 02 Apr 2019 14:18:45 GMT

My code snippet will address your TypeError, is the rest of the script working now?

Re: Use numpy array for a quercy?

JohannesBierer — Sun, 12 Dec 2021 05:27:17 GMT

I changed the code in this, but have problems to code the selection quercy properly.

#!/usr/bin/env python
# -*- coding: iso-8859-1 -*-
# 

import numpy as np
import arcpy
import pandas as pd
import sys, os

# Encoding der Standardausgabe herausfinden
stdout_encoding = sys.stdout.encoding or sys.getfilesystemencoding()
arcpy.env.overwriteOutput = True

inTable = r"R:\Karto\Bierer2019\FFH_MEKA_FACT\Tabellen\wk13.xlsx"
outP = r"R:\Karto\Bierer2019\FFH_MEKA_FACT\Shape"
fc = r"\\RPKSF001\GIS_DATA\ArcUISTools15\Data\UIS_ALk.gdb\UIS_0100000047100001"
fl = "ALK"

# Tabelle
Flrstkz = pd.read_excel(inTable, index_col=None, na_values=['NA'], parse_cols = "A")

x = np.array(Flrstkz)
x1 = np.array(map(str, x))
x2 = x1.tolist()

field = 'FLURSTUECKSKENNZEICHEN'

sc = "'" + field + " IN ({})".format(x2) + "'"

sc1 = sc.replace("[", "")
sc2 = sc1.replace("]", "")

print sc2

arcpy.MakeFeatureLayer_management(fc, fl)

arcpy.SelectLayerByAttribute_management (fl, "NEW_SELECTION", sc2)

outFc = os.path.join(outP, "ALK_WK13.shp")

arcpy.CopyFeatures_management(fl, outFc)‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

Line 28 is somehow wrong. The print of sc2 looks like this:

'FLURSTUECKSKENNZEICHEN IN ('8334100002890000384', '8334100002890000384', '8334100002890000384', '8334100002890000384')'

but gives a error in Line 37? Invalid expression?

Re: Use numpy array for a quercy?

JohannesBierer — Sun, 12 Dec 2021 05:27:20 GMT

With the right code formatting:

#!/usr/bin/env python
# -*- coding: iso-8859-1 -*-
# 

import numpy as np
import arcpy
import pandas as pd
import sys, os

# Encoding der Standardausgabe herausfinden
stdout_encoding = sys.stdout.encoding or sys.getfilesystemencoding()
arcpy.env.overwriteOutput = True

inTable = r"R:\Karto\Bierer2019\FFH_MEKA_FACT\Tabellen\wk13.xlsx"
outP = r"R:\Karto\Bierer2019\FFH_MEKA_FACT\Shape"
fc = r"\\RPKSF001\GIS_DATA\ArcUISTools15\Data\UIS_ALk.gdb\UIS_0100000047100001"
fl = "ALK"

# Tabelle
Flrstkz = pd.read_excel(inTable, index_col=None, na_values=['NA'], parse_cols = "A")

x = np.array(Flrstkz)
x1 = np.array(map(str, x))
x2 = x1.tolist()

field = 'FLURSTUECKSKENNZEICHEN'

sc = "'" + field + " IN ({})".format(x2) + "'"

sc1 = sc.replace("[", "")
sc2 = sc1.replace("]", "")

print sc2

arcpy.MakeFeatureLayer_management(fc, fl)

arcpy.SelectLayerByAttribute_management (fl, "NEW_SELECTION", sc2)

outFc = os.path.join(outP, "ALK_WK13.shp")

arcpy.CopyFeatures_management(fl, outFc)‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

Re: Use numpy array for a quercy?

JoshuaBixby — Tue, 02 Apr 2019 15:40:54 GMT

Is the field text or numeric, i.e., are those IN values supposed to be strings or numbers?

Re: Use numpy array for a quercy?

JohannesBierer — Wed, 03 Apr 2019 06:06:06 GMT

not in the office. Can answer next monday.

Re: Use numpy array for a quercy?

JohannesBierer — Mon, 08 Apr 2019 11:35:27 GMT

field is a text field and the in values should be text as well.

Re: Use numpy array for a quercy?

LukeWebb — Sun, 12 Dec 2021 05:27:22 GMT

Here is a function I use to build where cluases from lists, I use it like this.

values = ['test', 'test2']
field_name = 'query_field'
sql_where = build_where_clause_from_list (field_name, values)

values = [11, 27]
field_name = 'query_field'
sql_where = build_where_clause_from_list (field_name, values, field_type='Number')



def build_where_clause_from_list(field_name, value_list, field_type="String"):
 """Build a SQL In Where clause using a list

 e.g. field_name = 'Cheese', value_list=['Edam', 'Cheddar', 'Stinky Bishop']

 output: "Cheese" in ('Edam', 'Cheddar', 'Stinky Bishop')
 """
 where = """ "%s" in (""" % field_name
 if field_type == 'String':
 where+= "'"

 if field_type == 'String':
 joinTxt = "','"
 else:
 joinTxt = ","

 where += joinTxt.join(map(str, value_list))
 if field_type == 'String':
 where += "'"
 where += ")"
 return where

Re: Use numpy array for a quercy?

JoshuaBixby — Mon, 08 Apr 2019 16:02:51 GMT

Try changing Lines #28-#31 with:

sc2 = "{} IN ({})".format(field, ",".join("'" + str(i) + "'" for i in x2))‍

Re: Use numpy array for a quercy?

JoshuaBixby — Sun, 12 Dec 2021 05:27:25 GMT

Luke Webb‌, below is a slight modification you might want to consider:

def build_where_clause_from_list(field_name, value_list, force_string=False):
    import sys
    
    string_type = (str, bytes) if sys.version_info >= (3,0,0) else basestring
    
    fmt = lambda x: (
        "'{}'".format(x) 
        if force_string or isinstance(x, string_type) else
        "{}".format(x)
    )
    sql = '"{}" IN ({})'.format(field_name, ",".join(fmt(i) for i in value_list))
    
    return sql‍‍‍‍‍‍‍‍‍‍‍‍‍

The code above works in Python 2 and Python 3. By default, it will enclose strings in single quotes and leave numbers as is, unless force_string is set.