topic Re: loop terminates in Python Questions

loop terminates

TomaszChojnacki — Thu, 21 Sep 2017 10:28:49 GMT

Hello,

I am trying to perform some editing on features that have the same attribute value as features of a different feature class. So I tried this:

field0 = 'SKMS_tor'
field1 = 'SKMS'
cursor0 = arcpy.da.SearchCursor(fc0, field0)
cursor1 = arcpy.da.SearchCursor(fc1, field1)

for row0 in cursor0:
for row1 in cursor1:

if row0 == row1:

do do editing

else:

skip editing

I works, but only for the first round. Then it stops and does not proceed to the second feature in the feature class. Not sure what is wrong with this script.

I tested the same syntax on a different example:

q = ['a1', 'a2', 'a3']
w = ['a3', 'a1', 'b2']
for x in q:
    for y in w:
        if x == y:
            print x, " YES ", y
        else:
            print x, " NO ", y

and it works fine.

To my novice eye there is no difference in both scripts except the use arcpy.da.SearchCursor(). Can this be a problem? if not, what else?

Thanks for help.

Re: loop terminates

JoeBorgione — Thu, 21 Sep 2017 13:08:17 GMT

Don't have an answer for you Tomasz, but I do have a question; which of the cursors stop? Cursor0 or Cursor1? Just yesterday I was faced with a similar problem where I want to edit one table based on it's interaction with another feature class.

Let's call in the guys that have helped me in the not so distant past with cursors:

curtvprice‌, dkwiens‌, Dan_Patterson‌, jskinner-esristaff‌

and see what they might say....

Re: loop terminates

DanPatterson_Retired — Thu, 21 Sep 2017 13:17:08 GMT

sounds like nested cursor issues.

Don't you cursor types use 'while' loops? or do you have to call a next()? I am sure the others have examples, but here is the docs If it data were not in the millions of records, I would just read both into arrays/lists, do the comparisons and keep an index of what to change for processing after, rather than on the go

Re: loop terminates

JoshuaBixby — Thu, 21 Sep 2017 13:37:32 GMT

Overall, the approach is not very efficient, but I only have time now to speak to the specific error you are seeing. You need to reset cursor1 before going to the next row of cursor0. I don't know what is entailed in "do do editing," but you won't be able to do any updates to either data set using search cursors.

Re: loop terminates

JoeBorgione — Thu, 21 Sep 2017 13:59:24 GMT

Darren Wiens‌ provided this link yesterday: /blogs/richard_fairhurst/2014/11/08/turbo-charging-data-manipulation-with-python-cursors-and-dictionaries

I'm about to get into it....

Re: loop terminates

JoshuaBixby — Sat, 11 Dec 2021 21:05:23 GMT

Have you tried resetting the cursor yet?

field0 = 'SKMS_tor'
field1 = 'SKMS'
cursor0 = arcpy.da.SearchCursor(fc0, field0)
cursor1 = arcpy.da.SearchCursor(fc1, field1)

for row0 in cursor0:
    for row1 in cursor1:

        if row0 == row1:

            do do editing

        else:

            skip editing

     cursor1.reset()‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

Re: loop terminates

RandyBurton — Fri, 22 Sep 2017 16:24:49 GMT

I would also suggest Turbo Charging Data Manipulation with Python Cursors and Dictionaries (see Example 1).

Using a search cursor, you would load a dictionary with values contained in the SKMS_for field along with the objectID or globalID for that dataset.

Then using an update cursor, if the field value is in the dictionary, you can do your editing or mark the row for later action.

Re: loop terminates

TomaszChojnacki — Mon, 25 Sep 2017 19:13:14 GMT

This is what I needed. Thanks Joshua!

Re: loop terminates

TomaszChojnacki — Mon, 25 Sep 2017 19:19:30 GMT

Thank you Randy,

Usefull stuff and I need to get into this. Lots of learning for me as a complete novice!

Re: loop terminates

JoshuaBixby — Mon, 25 Sep 2017 19:51:46 GMT

Please mark a response as the answer to close the question out. Also, if you haven't already, give a helpful to other responses that were, well, helpful.

Re: loop terminates

JoeBorgione — Tue, 26 Sep 2017 13:52:36 GMT

bixb0012‌ ; thanks for providing the lesson in nested cursors. That just went into my PythonSnipetss folder for future reference.

Re: loop terminates

JoshuaBixby — Tue, 26 Sep 2017 20:31:55 GMT

I was just focused on making the error go away. As others pointed out in the comments, nesting cursors can lead to some big performance problems, and there may be other, better ways of accomplishing the work.

Re: loop terminates

JoeBorgione — Tue, 26 Sep 2017 22:56:29 GMT

Yep- just like it as an example; I need to figure out the whole dictionary thing.....

Re: loop terminates

ClintonDow1 — Wed, 27 Sep 2017 00:01:31 GMT

For a bit of context, the reason this happens is because the data access (.da) cursors are implemented as iterators which by design only allow one 'pass' and will then raise StopIteration exceptions on subsequent calls. If you want to repeatedly access the collection, or access items within the collection via an index, you can cast the iterator into a list. The .reset() method also works for multiple passes as Joshua has suggested but does not allow for indexing. The list approach is only viable if the data set you are working with is smaller than the amount of available RAM, so for larger data sets the iterator is necessary to feed the data 'piece by piece' into the script, so depending on the circumstances either approach has strengths and weaknesses.