Code

This script is useful for auditing metadata. The script can be run across multiple enterprise geodatabases, and will create an output CSV file with these columns: 'Geodatabase', 'Feature Dataset', 'Feature Class', 'Spatial Reference', 'Row Count', 'Column Count', 'Metadata Title', 'Metadata Description', 'Metadata Summary', 'Metadata Credits', 'Metadata Access Constraints'.

To use this script, you need to change Line 19 to point to your SDE connection files.

import csv
import os

from arcpy import Describe
from arcpy import env
from arcpy import GetCount_management
from arcpy import ListDatasets
from arcpy import ListFeatureClasses
from arcpy import ListFields
from arcpy import metadata as md

# the output CSV file name
output_csv_file = 'enterprise_gdbs_results.csv'

# or use this to make the output CSV file name the same as the script name
# output_csv_file = os.path.basename(__file__).replace('.py','.csv')

# change this path to use your SDE connection files
sde_connection_files = 'c:/gis/sde_connection_files'

with open(output_csv_file,'w',newline='') as csv_file:
    csv_writer = csv.writer(csv_file, delimiter=',')
    column_headers = [
        'Geodatabase', 'Feature Dataset', 'Feature Class', 
        'Spatial Reference', 'Row Count', 'Column Count', 
        'Metadata Title', 'Metadata Description', 'Metadata Summary',
        'Metadata Credits', 'Metadata Access Constraints'
    ]
    
    csv_writer.writerow(column_headers)

    for (folder, dirs, files) in os.walk(sde_connection_files):
        for file in files:
            if(not file.startswith('_') and file.endswith('.sde')):
                geodatabase = file.split('@')[1]

                wksp = os.path.join(folder,file)
                env.workspace = wksp

                datasets = ListDatasets(feature_type='feature')
                datasets = [''] + datasets if datasets is not None else []

                for dataset in datasets:
                    for feature_class in ListFeatureClasses(dataset):
                        desc = Describe(feature_class)
                        spatial_reference = desc.SpatialReference.factoryCode
                        count = GetCount_management(feature_class)[0]
                        column_count = len(ListFields(feature_class))
                        metadata = md.Metadata(feature_class)

                        row = [
                            geodatabase,dataset,feature_class,
                            spatial_reference,count,column_count,
                            metadata.title,metadata.description,
                            metadata.summary,metadata.credits,
                            metadata.accessConstraints    
                        ]

                        print(row)
                        
                        csv_writer.writerow(row)

This script is useful for auditing metadata. The script can be run across multiple file geodatabases, and will create an output CSV file with these columns: 'Last Modified','File Geodatabase', 'Feature Dataset', 'Feature Class', 'Spatial Reference', 'Row Count', 'Column Count', 'Metadata Title', 'Metadata Description', 'Metadata Summary', 'Metadata Credits', 'Metadata Access Constraints'.

import csv
import os
import time

from arcpy import Describe
from arcpy import env
from arcpy import GetCount_management
from arcpy import ListDatasets
from arcpy import ListFeatureClasses
from arcpy import ListFields
from arcpy import metadata as md

folder = r'...'
log = open('log.txt', 'w')
output_csv_file = os.path.basename(__file__).replace('.py','.csv')

with open(output_csv_file,'w',newline='') as csv_file:
    csv_writer = csv.writer(csv_file, delimiter=',')
    column_headers = [
        'Last Modified','File Geodatabase', 'Feature Dataset', 'Feature Class', 
        'Spatial Reference', 'Row Count', 'Column Count', 
        'Metadata Title', 'Metadata Description', 'Metadata Summary',
        'Metadata Credits', 'Metadata Access Constraints'
    ]
    
    csv_writer.writerow(column_headers)

    for (folder, dirs, files) in os.walk(folder):
        if(folder.endswith('.gdb')):
            print(folder)

            last_modified = '???'

            try:
                last_modified = time.ctime(max(os.stat(root).st_mtime for root,_,_ in os.walk(folder)))
            except:
                msg = 'Error getting last modified date for {}'.format(folder)
                print(msg)
                log.write('{}\n'.format(msg))
                pass

            fgdb = folder
            env.workspace = fgdb

            datasets = []

            try:
                datasets = ListDatasets(feature_type='feature')
                datasets = [''] + datasets if datasets is not None else []
            except:
                msg = 'Error getting datasets for {}'.format(folder)
                print(msg)
                log.write('{}\n'.format(msg))
                pass

            for dataset in datasets:
                for feature_class in ListFeatureClasses(dataset):
                    desc = Describe(feature_class)
                    spatial_reference = desc.SpatialReference.factoryCode
                    count = GetCount_management(feature_class)[0]
                    column_count = len(ListFields(feature_class))
                
                    metadata = None
                    row = []

                    try:
                        metadata = md.Metadata(feature_class)
                        row = [
                            last_modified,
                            fgdb,dataset,feature_class,
                            spatial_reference,count,column_count,
                            metadata.title,metadata.description,
                            metadata.summary,metadata.credits,
                            metadata.accessConstraints    
                        ]
                    except:
                        row = [
                            last_modified,
                            fgdb,dataset,feature_class,
                            spatial_reference,count,column_count,
                            'error','error','error','error','error'    
                        ]
                        pass

                    csv_writer.writerow(row)

log.close()

Sometimes when data is converted from shapefiles to feature classes, or if care has not been taken when running geoprocessing tools, feature classes can end up with redundant fields. For example, a feature class could have a field called "OBJECTID" and there may be another field called "OBJECTID_1" in the same feature class. One of these fields should be the OID field for the feature class, however, sometimes it also not clear which OBJECTID field is the correct one. This script below does not need to know which field is the OID field because Line 35 will fail if the field cannot be deleted, and the script will continue running. The script below shows two different ways the field name can be checked, i.e. the field name can be included in the list on Line 10, or the the field name can be prefixed with "OBJECTID" on Line 33. For this script, the deleted fields are output to a CSV file. The name of the output CSV file is the same name as the Python script - see Line 12. Before running the script, it is a good idea to comment out Line 35, run the script, and check the output CSV file to see which fields will be deleted. Also, once the field is deleted, it cannot be recovered, so if in doubt, the file geodatabases should be backed up first.

import csv
import os

from arcpy import env
from arcpy import ListDatasets
from arcpy import ListFeatureClasses
from arcpy import ListFields
from arcpy.management import DeleteField

fields_to_delete = ['objectid', 'objectid_1', 'objectid_12', 'shape_le_1', 'shape_leng', 'shape_length', 'shape__length', 'shape__area']
input_folder = r'...'
output_csv_file = os.path.basename(__file__).replace('.py','.csv')

with open(output_csv_file, 'w', newline='') as csv_file:
    csv_writer = csv.writer(csv_file)
    csv_writer.writerow(['File Geodatabase','Feature Dataset', 'Feature Class', 'Field'])

    for (folder, dirs, files) in os.walk(input_folder):
        if(folder.endswith('.gdb')):
            file_geodatabase = folder
            print('Processing {}'.format(file_geodatabase))

            env.workspace = file_geodatabase

            feature_datasets = ListDatasets(feature_type='feature')
            feature_datasets = [''] + feature_datasets if feature_datasets is not None else []

            for feature_dataset in feature_datasets:
                feature_classes = ListFeatureClasses(feature_dataset=feature_dataset)

                for feature_class in feature_classes:
                    for field in ListFields(feature_class):
                        if field.name.lower() in fields_to_delete or field.name.lower().startswith('objectid'):
                            try:
                                DeleteField(feature_class, field.name)
                                csv_writer.writerow([file_geodatabase, feature_dataset, feature_class, field.name])
                            
                            except:
                                pass

Before publishing map services or feature services, it is good practice to recalculate the feature class extents first. This is because occasionally the feature class extent can be wrong. There are several things that can cause the feature class extent to be wrong, including running geoprocessing tasks on the data that changes the extent of the data, and the feature class extent can also be wrong when importing data because the initial feature class extent may have been calculated based on a subset of the data that was imported. The first easy way to see the feature class extent is wrong is when clicking on the zoom-to-layer button, the map does not zoom to the correct extent. Another sign that the layer extent is incorrect is the layer does not render correctly. For example, when importing a large GeoJSON file, it may not render correctly in ArcGIS Pro until the extent has been recalculated for that feature class.

import os

from arcpy import env
from arcpy import ListDatasets
from arcpy import ListFeatureClasses
from arcpy.management import RecalculateFeatureClassExtent

input_folder = r'...'

for (folder, dirs, files) in os.walk(input_folder):
    if(folder.endswith('.gdb')):
        env.workspace = folder

        feature_datasets = ListDatasets(feature_type='feature')
        feature_datasets = [''] + feature_datasets if feature_datasets is not None else []

        for feature_dataset in feature_datasets:
            feature_classes = ListFeatureClasses(feature_dataset=feature_dataset)

            for feature_class in feature_classes:
                print('{}\t{}\t{}'.format(folder, feature_dataset, feature_class))
                RecalculateFeatureClassExtent(feature_class)

This script is for listing all shapefiles in a directory to help get an understanding of how many shapefiles you have, so they can be migrated to feature classes in file geodatabases or enterprise geodatabases. Note that shapefiles should generally be avoided for several reasons. The shapefile format is a very old format and there are much better formats for sharing spatial data. Shapefiles have poor compression and when data is exported to shapefiles, the field names get truncated to ten characters, which can make it difficult to guess what the original field name was. Use geodatabase feature classes, geopackages, geojson, or other formats before considering using shapefiles.

import csv
import os

# change the input folder here for the directory you want to search
input_folder = r'\\...\...'

# change the file name here to the output file you want
results_file = 'shapefiles.csv'

extns = ['.aih', '.ain', '.atx', '.cpg', '.dbf', '.fbn', '.fbx', '.ixs', '.mxs', '.prj', '.sbn', '.sbx', '.shp', '.shx', '.xml']

with open(results_file, 'w', newline='') as csvfile:
    csvwriter = csv.writer(csvfile)
    column_headers = ['Folder','Shapefile'] + extns
    csvwriter.writerow(column_headers)

    for (folder, dirs, files) in os.walk(input_folder):
        for file in files:
            if(file.endswith('.shp')):
                shapefile_name = file.replace('.shp','')

                line = [folder,shapefile_name]

                for extn in extns:
                    file_path = os.path.join(folder,'{}{}'.format(shapefile_name,extn))
                    file_exists = 'Y' if os.path.exists(file_path) else 'N'
                    line.append(file_exists)
                
                csvwriter.writerow(line)

This script lists the layer info for MXDs (ArcMap files) and APRXs (ArcGIS Pro files). The results are output to a CSV file and contain these columns:

Note that ArcMap uses Python 2 and ArcGIS Pro uses Python 3, so to get the script working with Python 3 for MXD files, we convert the MXD to an APRX in the script below. This APRX is only in memory and will not be saved to disk (but can be saved to disk if needed).

To run this below change lines 5 and 7 and manually create a blank APRX that can be cloned into memory (line 7).

import csv
import os

from arcpy.mp import ArcGISProject

directory = r'\\...\...'
blank_aprx = r'\\...\...\temp.aprx'
results_file = os.path.basename(__file__).replace('.py','.csv')

# create a CSV file
with open(results_file, 'w', newline='') as csvfile:
    csvwriter = csv.writer(csvfile)

    # write out the column headers in the CSV file
    csvwriter.writerow(['File Path','Folder','File Name','File Type','Map / Data Frame','Layer Name','Layer Source','Is Broken','Definition Query'])

    # loop through all of the folders and files in the input directory
    for folder, dirs, files in os.walk(directory):
        for file in files:
            file_path = os.path.join(folder,file)
            
            # check if it is an MXD file - we need to convert the MXD to APRX (in-memory) because we cannot use Python 3 with MXDs
            if file.endswith('.mxd'):     
                file_type = 'mxd'
                aprx = ArcGISProject(blank_aprx)
                aprx.importDocument(file_path)

            #  check if it is an APRX file
            elif file.endswith('.aprx'):
                file_type = 'aprx'
                aprx = ArcGISProject(file_path)
            
            else:
                # file is not an MXD or APRX, so skip the loop below
                continue

            # list all layers within the 'maps' (for APRXs), or 'data frames' (for MXDs)
            for map in aprx.listMaps():
                for layer in map.listLayers():
                    # if the layer is not a group layer and has a data source
                    if layer.supports('dataSource'):
                        # check if the layer is broken
                        isBroken = layer.isBroken if layer.supports('isBroken') else 'n/a'
                        # check if the layer has a definition query (the WHERE clause in a normal SQL statement - definition queries use SQL)
                        definitionQuery = layer.definitionQuery if layer.supports('definitionQuery') else 'n/a'
                        line = [file_path, folder, file, file_type, map.name, layer.name, layer.dataSource, isBroken, definitionQuery]

                        # write the results to CSV
                        csvwriter.writerow(line)

            # delete the in-memory aprx
            del aprx

print('Results file: {}'.format(results_file))

This script lists the service info for all services and outputs the results to CSV. The column headers in the CSV are listed below:

Before running the script, change the parameters from Lines 22 to 31.

import csv
import json
import requests

def getToken(url, params):
    response = requests.post(url, params)
    token = json.loads(response.text)['token']
    return token

def fetchUrl(url):
    params = {
        'f': 'json',
        'token': token
    }

    response = requests.post(url, params)
    data = json.loads(response.text)

    return data

web_adaptor = 'arcgis' # web adaptor or load balancer 
token_url = 'https://{}/portal/sharing/rest/generateToken'.format(map_server)

params = {
    'f': 'json',
    'referer': 'https://.../portal/sharing/rest/generateToken',
    'username': 'portaladmin',
    'password': '**************'
}

token = getToken(token_url, params)

''' 
    ArcGIS Admin URL - we are using the admin URL, not the 
    Services Directory URL because the Services Directory will 
    only list the services that are started/running.  We also
    want to know about the stopped services.
'''
admin_url = 'https://{}/{}/admin/services'.format(map_server,web_adaptor)

params = {
    'f': 'json',
    'token': token
}

response = requests.get(admin_url, params)
data = json.loads(response.text)

folders = data['folders']
foldersToExclude = ['System','Utilities']

for folder in foldersToExclude:
    folders.remove(folder)

services = data['services'] 

with open('map_service_info.csv', 'w', newline='') as csv_file:
    csv_writer = csv.writer(csv_file)

    column_names = ['Map Server', 'Folder', 'Service name', 'Service type', 'URL', 'Published By', 'Path', 'Service Description', 'Copyright Info']
    csv_writer.writerow(column_names)

    for service in services:
        folder = ''
        service_name = service['serviceName']
        service_type = service['type']
        url = '{}/{}.{}'.format(admin_url,service_name,service_type)
       
        service_info = fetchUrl(url)
        service_description = service_info['description']
        user_name = service_info['properties']['userName'] if 'userName' in service_info['properties'] else ''
        copyright_info = service_info['properties']['copyright'] if 'copyright' in service_info['properties'] else ''
        path = service_info['properties']['path'] if 'path' in service_info['properties'] else ''

        row = [map_server, folder, service_name, service_type, url, user_name, path, service_description, copyright_info]
        csv_writer.writerow(row)

    for folder in folders:
        url = '{}/{}'.format(admin_url,folder)
        folder_info = fetchUrl(url)

        for service in folder_info['services']:
            folder = folder_info['folderName']
            service_name = service['serviceName']
            service_type = service['type']
            url = '{}/{}/{}.{}'.format(admin_url, folder, service_name, service_type, url)

            service_info = fetchUrl(url)
            service_description = service_info['description']
            user_name = service_info['properties']['userName'] if 'userName' in service_info['properties'] else ''
            copyright_info = service_info['properties']['copyright'] if 'copyright' in service_info['properties'] else ''
            path = service_info['properties']['path'] if 'path' in service_info['properties'] else ''

            row = [map_server, folder, service_name, service_type, url, user_name, path, service_description, copyright_info]
            csv_writer.writerow(row)