AnsweredAssumed Answered

Appending data to an empty dictionary

Question asked by vshah6 on Mar 27, 2018
Latest reply on Mar 29, 2018 by blake.terhune

Like • Show 0 Likes0
Comment • 15

So I am currently working on a tool that the user needs to apply a query in order to the run the tool for the queried data. I am trying to get around that by creating a blank dictionary in the script and then appending data to it. The data comes from a field in a feature class. I have never done this before but in theory it should work the same as appending to lists correct?

hubFeatures = r'hub_locations'

hubDict = {}

These are my working variables for this. The field that I want the data from to append to the dictionary is called 'NAME' in the hubFeatures FC. I was thinking of using cursors to achieve this. Unless there is an alternative approach. The idea is to then use the items in the dictionary to apply a query to the necessary layers and complete the functionality of the tool using a for loop. That part I can do. Need some assistance in setting up the dictionary.

blake.terhune Mar 29, 2018 8:06 AM

Correct Answer

If I'm understanding your request, I think this will get you what you need. I learned it's easy to enumerate a cursor for an index or key field. Here's the expanded code so you can see what's happening:

valueDict = {}
with arcpy.da.SearchCursor(sourceFC, ["NAME"]) as cursor:
    for enum, row in enumerate(cursor):
        valueDict[enum] = row[0]

This will build a dictionary like this:

{
    0: "Some Name",
    1: "Another Name",
    2: "Name Example",
    3: "And Another"
}

See the reply in context

No one else had this question

Outcomes

15 Replies

rvburton Mar 27, 2018 3:29 PM
Mark Correct
Correct Answer
For an example of populating a dictionary, see Turbo Charging Data Manipulation with Python Cursors and Dictionaries (starting with example 1). You will see code like this:
```
# Use list comprehension to build a dictionary from a da SearchCursor  
valueDict = {r[0]:(r[1:]) for r in arcpy.da.SearchCursor(sourceFC, sourceFieldsList)}  
```
You mention using the NAME field. If it is to be used as the key, what will you use as the value?
2 people found this helpful
Like • Show 0 Likes0
Actions
- Dan_Patterson @ rvburton on Mar 27, 2018 6:41 PM
  Mark Correct
  Correct Answer
  a 'dictionary comprehension' or just 'dict comprehension'
  dictionary order is now preserved in python 3.6 (ie if you are using PRO)
  2 people found this helpful
  Like • Show 0 Likes0
  Actions
- vshah6 @ rvburton on Mar 29, 2018 6:14 AM
  Mark Correct
  Correct Answer
  I want to use the NAME field's values to be the value in the dictionary and a numeric value starting at 0 as the key if that is possible?
  
  I am unfortunately still using ArcGIS for Desktop and thus do not have access to python 3.6.
  Like • Show 0 Likes0
  Actions
  - Dan_Patterson @ vshah6 on Mar 29, 2018 7:43 AM
    Mark Correct
    Correct Answer
    Did you try Randy's suggestion? It should work without problem
    Like • Show 0 Likes0
    Actions
    - vshah6 @ Dan_Patterson on Mar 29, 2018 8:00 AM
      Mark Correct
      Correct Answer
      I did. While it does create a dictionary and populates it with a value, the NAME field value becomes the key and the value portion is empty. I'm trying to make it where the key is a numeric integer and the NAME field value is the value in the dictionary.
      
      hubFeatures = r'hub_locations' fieldList = "NAME" hubDict = {r[0]:(r[1:]) for r in arcpy.da.SearchCursor(hubFeatures, fieldList)}
      
      using:
      
      print hubDict
      
      I get:
      
      {u'Hub Name': ()}
      
      My desired output:
      {1: 'Hub Name'}
      
      How do I get to that? Reason I want to get that to be the output is because I want to use this for applying queries before doing the core function of the tool, which is conducting a network analysis.
      Like • Show 0 Likes0
      Actions
  - blake.terhune @ vshah6 on Mar 29, 2018 8:06 AM
    Unmark Correct
    Correct Answer
    If I'm understanding your request, I think this will get you what you need. I learned it's easy to enumerate a cursor for an index or key field. Here's the expanded code so you can see what's happening:
    valueDict = {} with arcpy.da.SearchCursor(sourceFC, ["NAME"]) as cursor: for enum, row in enumerate(cursor): valueDict[enum] = row[0]
    This will build a dictionary like this:
    { 0: "Some Name", 1: "Another Name", 2: "Name Example", 3: "And Another" }
    1 person found this helpful
    Like • Show 0 Likes0
    Actions
    - vshah6 @ blake.terhune on Mar 29, 2018 8:14 AM
      Mark Correct
      Correct Answer
      Blake that seems to get it in the format I want it in.
      
      Do you know why the u appears before the value from the NAME field?
      
      {0: u'Hub Name'}
      Like • Show 0 Likes0
      Actions
      - Dan_Patterson @ vshah6 on Mar 29, 2018 8:19 AM
        Mark Correct
        Correct Answer
        Unicode ...
        1 person found this helpful
        Like • Show 1 Like1
        Actions
        blake.terhune @ Dan_Patterson on Mar 29, 2018 8:24 AM
        Mark Correct
        Correct Answer
        Indeed.
        The 'u' in front of the string values means the string has been represented as unicode. Letters before strings here are called "String Encoding declarations". Unicode is a way to represent more characters than normal ascii can manage.
        python - What does the 'u' symbol mean in front of string values? - Stack Overflow
        1 person found this helpful
        Like • Show 0 Likes0
        Actions
        vshah6 @ blake.terhune on Mar 29, 2018 8:38 AM
        Mark Correct
        Correct Answer
        Awesome and that doesn't interfere with my definition query that I am applying afterwards. Thank you so much for your help Blake!
        Like • Show 0 Likes0
        Actions
        blake.terhune @ vshah6 on Mar 29, 2018 9:13 AM
        Mark Correct
        Correct Answer
        Typically, no, it does not interfere. However, it can depend on how you're creating your query expression. Assuming my example valueDict above:
        for i in valueDict:     expression = "NAME = '{}'".format(valueDict[i])     print expression
        Output:
        NAME = 'Some Name' NAME = 'Another Name' NAME = 'Name Example' NAME = 'And Another'
        This makes perfectly valid SQL.
        
        However, If you do something like this:
        names = tuple(valueDict.values()) expression = "NAME in{}".format(names) print expression
        Output:
        NAME in(u'Some Name', u'Another Name', u'Name Example', u'And Another')
        It will keep the u designation and it won't be valid SQL.
        
        You would have to do something like:
        names = tuple(str(i) for i in valueDict.values()) expression = "NAME in{}".format(names) print expression
        Output:
        NAME in('Some Name', 'Another Name', 'Name Example', 'And Another')
        But if you have exactly one name record in the dictionary, a single item tuple still has a comma at the end:
        NAME in('Some Name',)
        Which is not valid SQL.
        
        The best would be something like:
        names = ",".join("'{}'".format(i) for i in valueDict.values()) expression = "NAME in({})".format(names) print expression
        Output:
        NAME in('Some Name','Another Name','Name Example','And Another')
        
        EDIT:
        Or, you could get rid of the u designation by converting row[0] to a plain string when you create the dictionary:
        valueDict = {} with arcpy.da.SearchCursor(sourceFC, ["NAME"]) as cursor:     for enum, row in enumerate(cursor):         valueDict[enum] = str(row[0])
        1 person found this helpful
        Like • Show 1 Like1
        Actions
        vshah6 @ blake.terhune on Mar 29, 2018 9:57 AM
        Mark Correct
        Correct Answer
        I didn't even think of the first way you had it set up.
        
        What I used was this:
        
        hubFeatures = r'hub_locations' hubDict = {} with arcpy.da.SearchCursor(hubFeatures, 'HubName') as cursor:      for enum, row in enumerate(cursor):          hubDict[enum] = row[0] for hub in hubDict:      valueHub = dict.get(hubDict, hub)      hubQuery = "NAME = " + "'" + str(valueHub) + "'"      hubLayer = arcpy.mapping.ListLayers(mxd, 'hub_locations')[0]      hubLayer.definitionQuery = hubQuery
        Like • Show 0 Likes0
        Actions
        blake.terhune @ vshah6 on Mar 29, 2018 1:26 PM
        Mark Correct
        Correct Answer
        On line 10, you've got the wrong syntax for get(). Should be:
        valueHub = hubDict.get(hub)
        But since you are iterating over the keys in the dictionary (as hub), you know the key exists so you don't need to use get()
        Like • Show 0 Likes0
        Actions
        vshah6 @ Dan_Patterson on Mar 29, 2018 8:38 AM
        Mark Correct
        Correct Answer
        Thank you so much Dan!
        Like • Show 0 Likes0
        Actions

Dan_Patterson

Mar 29, 2018 8:09 AM

do you have them in the right order ?

r = [[1, 'a'], [2, 'b'], [3, 'c']]

valueDict = {v[0]:(v[1:]) for v in r}

valueDict
{1: ['a'], 2: ['b'], 3: ['c']}

# ---- switch it up

valueDict2 = {v[1]:(v[0]) for v in r}

valueDict2
{'a': 1, 'b': 2, 'c': 3}

1 person found this helpful

Actions

GeoNet

The Esri Community

Appending data to an empty dictionary

Outcomes

Related Content

Recommended Content