Update Cursor.

RajP · ‎10-30-2015

Hello,

Can any one suggest me to the following code is correct or not?

Note: I need to update GISID from HT_GISID.

Code:

workspace1 = r"D:\ADC\Tool\Mumbai\PGDB For Reference\MVVNL_MAHMUDABAD\MVVNL_MAHMUDABAD.mdb"
workspace2 = r"D:\ADC\Tool\Mumbai\PGDB For Reference\MVVNL_MAHMUDABAD\Output.mdb"

searchFeatures = os.path.join(workspace2, "LTOVerHead_DTGISID")
updateFeatures = os.path.join(workspace1, "SecOHElectricLineSegment")

with arcpy.da.UpdateCursor(updateFeatures, ["OBJECTID", "GISID"]) as cursor1:
    for row1 in cursor1:
        gisiddict={}
        with arcpy.da.SearchCursor(searchFeatures, ["TARGET_FID", "HT_GISID"]) as cursor:
            for row in cursor:
                gisiddict[row[0]] = row[1]
            del row, cursor
        oid = row1[0]
        if(gisiddict.has_key(oid)):
            row1[1] = gisiddict[oid]
        cursor1.updateRow(row1)
    del row1, cursor1

Regards,

Raj P

Message was edited by: Dan Patterson Syntax Highlighting add to make code readable. Tags added to make it useful for searches

DanPatterson_Retired · ‎10-30-2015

Can you post the results showing the inputs and the outputs?

RajP · ‎10-30-2015

Hello,

Thanks for replying:

My Inputs are:

Table1:

OBJECTID *	Shape *	TARGET_FID *	FeederID	HT GIS ID	HT_T Code	SHAPE_Length
2814	Polyline	1564808	33232210034TO1	03003405A8613	NA	0.518247
2878	Polyline	1567587	33232210034SDR	03003405A8612	NA	0.499987
2287	Polyline	1569532	33232210034SDR	03003405A8610	NA	28.59185
171	Polyline	1567609	33232210034SDR	03003405A8610	NA	0.50004
2288	Polyline	1569533	33232210034SDR	03003405A8610	NA	58.23881
780	Polyline	1563761	33232210034SDR	03003405A8609	NA	0.282843
2295	Polyline	1569535	33232210034SDR	03003405A8606	NA	0.5255
113	Polyline	1567608	33232210034TO2	03003405A8602	NA	0.500025
37	Polyline	1569534	33232210034TO2	03003405A8601	NA	0.499996
2877	Polyline	1567586	33232210034TO2	03003405A8600	NA	0.500021

Table2:

OBJECTID *	FeederID	Gis ID	Survey Building ID	Substation Code	DTRGISID *	SHAPE_Length
1564808	33232210034TO1	0	NOTAP	3.32E+10	3003405A8613	0.518247
1564807	33232210034TO2	0	NOTAP	3.32E+10	300340528	0.068149
1564806	33232210034TO2	0	NOTAP	3.32E+10	300340528	0.043821
1564805	33232210034TO2	0	NOTAP	3.32E+10	300340531	50.01936
1564804	33232210034TO5	0	NOTAP	3.32E+10	300340553	41.90007
1564803	33232210034TO5	0	NOTAP	3.32E+10	300340553	30.91885
1564802	33232210034TO5	0	NOTAP	3.32E+10	300340553	36.66272
1564801	33232210034TO5	0	NOTAP	3.32E+10	300340553	44.50375
1564800	33232210034TO5	0	NOTAP	3.32E+10	300340553	54.88846
1564799	33232210034SDR	0	NOTAP	3.32E+10	300340548	2.978194

I tried to update Table2 "GIS ID" from Table1 "HT GIS ID" using Common field
from Table1 as "Target_FID" and Table 2 as "ObjectID"

Regards,

Raj P

JoshuaBixby · ‎10-30-2015

The first change I would make is to pull your search cursor out of your update cursor. The current code is creating a new search cursor for every record in the update table, and the same dictionary is being created over and over again.

Give this a try:

workspace1 = r"D:\ADC\Tool\Mumbai\PGDB For Reference\MVVNL_MAHMUDABAD\MVVNL_MAHMUDABAD.mdb"  
workspace2 = r"D:\ADC\Tool\Mumbai\PGDB For Reference\MVVNL_MAHMUDABAD\Output.mdb"  

searchFeatures = os.path.join(workspace2, "LTOVerHead_DTGISID")  
updateFeatures = os.path.join(workspace1, "SecOHElectricLineSegment") 

with arcpy.da.SearchCursor(searchFeatures, ["TARGET_FID", "HT_GISID"]) as cursor:
    gisiddict={} 
    for fid, gisid in cursor:  
        gisiddict[fid] = gisid  
del gisid, fid, cursor

with arcpy.da.UpdateCursor(updateFeatures, ["OBJECTID", "GISID"]) as cursor:  
    for oid, gisid in cursor:    
        if oid in gisiddict:
            cursor.updateRow([oid, gisiddict[oid]])
del gisid, oid, cursor

curtvprice · ‎10-30-2015

Just to mention, as I keep seeing people unnecessarily using del when Python takes care of this on its own...

No need to "del cursor" -- that variable is deleted the cursor is closed, and the table file lock is released when you leave the with block. That's kind of the point of the with block, to delete the variable/close the cursor no matter what happens on exiting the block.

The variables gisid, fid, also don't need to be deleted as they aren't attached to an object (like or layer or cursor would be) and are auto-deleted by python garbage collection when the script ends.

UPDATE: Thank you Joshua Bixby I stand corrected. See strikethrough and bold above.

JoshuaBixby · ‎10-31-2015

Python 2.7.10 documentation > 7. Compound Statements > 7.5. The with statement:

The with statement is used to wrap the execution of a block with methods defined by a context manager (see section With Statement Context Managers). This allows common try...except...finally usage patterns to be encapsulated for convenient reuse.

The with statement encapsulates code, but it encapsulates specific types of code within a class. For someone using a class and not developing one, the with statement is more about wrapping code within a context manager.

The variables defined within a with statement, including the target variable of the with statement, are wrapped and persist after the with statement:

>>> with arcpy.da.SearchCursor(r'Default.gdb\StateProvinceBoundaries_ESRI',"OID@") as cur:
...    for row in cur:
...        pass
...       
>>> dir()
['__builtins__', '__doc__', '__file__', '__name__', '__package__', 'arcgis', 'arcpy', 'cur', 'datetime', 'math', 'row', 'sys', 'time']
>>> type(cur)
<type 'da.SearchCursor'>
>>> type(row)
<type 'tuple'>

The with statement doesn't delete variables and the variables don't go out of scope after the with statement. For this specific case, the with statement ensures the arcpy.da.SearchCursor.__enter__() and arcpy.da.SearchCursor.__exit__() methods are called, thus freeing some of the locks the cursor holds on the data source.

Regardless of whether a with statement is used or not, the variables will all be cleaned up when the script ends.

DanPatterson_Retired · ‎10-31-2015

They really have to work on their help topics arcpy.da.SearchCursor.__enter__() so secretive

JoshuaBixby · ‎11-01-2015

Dan Patterson, given how much you like NumPy (who can blame you, it is pretty slick), you might also have fun with arcpy.da.SearchCursor._dtype and arcpy.da.SearchCursor._as_narray().

>>> cur = arcpy.da.SearchCursor(r'StateProvinceBoundaries_ESRI',
                                ["OID@", "SHAPE@XY"],
                                where_clause="NAME1 = 'Minnesota'")
>>> cur._dtype
dtype([('OID@', '<i4'), ('SHAPE@XY', '<f8', (2,))])
>>> cur._as_narray()
array([(139, [205923.5657559298, 5264242.273746617]),
       (140, [544608.9951297314, 5365824.304238412]),
       (141, [221359.03653639136, 4966096.278488112]),
       (142, [580541.5304663723, 5054285.671024574])], 
      dtype=[('OID@', '<i4'), ('SHAPE@XY', '<f8', (2,))])
>>>
>>> help(arcpy.da.SearchCursor._as_narray)
Help on method_descriptor:
_as_narray(...)
    _as_narray() -> numpy.record.
    Return snapshot of the current state as NumPy array.

That part of the API might not be published but Python is for consenting adults.

DanPatterson_Retired · ‎11-01-2015

Nicccceee so much for relaxing... have several more things to check... the pesky _method and _property etc things

>>> import arcpy
>>> src = r"F:\Testing\AOI_mtm9.shp"
>>> arcpy.env.workspace = "in_memory"
>>> #
>>> #cur = arcpy.da.SearchCursor(src, *, where_clause=None, spatial_reference) 
>>> SR = arcpy.Describe(src).spatialReference
>>> SR.name
'NAD_1983_CSRS_MTM_9'
>>> 
>>> #-----------------------
>>> # fun time
>>> cur = arcpy.da.SearchCursor(src, "*", where_clause=None, spatial_reference=SR, explode_to_points=True)
>>> cur._dtype
dtype([('FID', '<i4'), ('Shape', '<f8', (2,)), ('Id', '<i4'), ('X_c', '<f8'), ('Y_c', '<f8'), ('area', '<f8')])
>>> cur.fields
('FID', 'Shape', 'Id', 'X_c', 'Y_c', 'area')
>>> a = cur._as_narray()
>>> a
array([(0, [340000.0, 5022000.0], 0, 342000.0, 5024000.0, 16000000.0),
       (0, [340000.0, 5026000.0], 0, 342000.0, 5024000.0, 16000000.0),
       (0, [344000.0, 5026000.0], 0, 342000.0, 5024000.0, 16000000.0),
       (0, [344000.0, 5022000.0], 0, 342000.0, 5024000.0, 16000000.0),
       (0, [340000.0, 5022000.0], 0, 342000.0, 5024000.0, 16000000.0)], 
      dtype=[('FID', '<i4'), ('Shape', '<f8', (2,)), ('Id', '<i4'), ('X_c', '<f8'), ('Y_c', '<f8'), ('area', '<f8')])
>>> import numpy as np
>>> a_clone = a.copy()
>>> a_clone = a_clone.view(type=np.recarray)
>>> a_clone.Shape
array([[  340000.,  5022000.],
       [  340000.,  5026000.],
       [  344000.,  5026000.],
       [  344000.,  5022000.],
       [  340000.,  5022000.]])
>>> a_clone.Shape[1:].mean(axis=0) # skip the duplicate point
array([  342000.,  5024000.])
>>> # oooooo nice!
‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

hmmmm... me thinks the hole hasn't been dug deep enought... thanks Joshua Bixby

Visit the future.... returned to this topic