Latest Contributions by VinceE

Hey @DrewFlater, a follow up to this. I just tried to use arcpy's AddField with a "BIGINTEGER" data type, and ran into a similar issue as above: ERROR 000800: The value is not a member of TEXT | FLOAT | DOUBLE | SHORT | LONG | DATE | BLOB | RASTER | GUID. Failed to execute (AddField). I'm not going to open a Pro Idea because... this issue of new data types clearly hasn't been addressed in a lot of places, so the "idea" would extremely general. I'll leave the prioritization decisions to you all. Thanks again for the initial reply. ... View more

Fair enough, I appreciate the insights. I'll go ahead and play around with replacing the string in FeatureSetByName. Thanks again! ... View more

Thanks for the inputs @RPGIS, but the code I have provided already "works"--even without specifying this line below. Maybe that's because the Arcade "environment" knows that all feature classes/tables will have an ObjectID, so you don't need to specify? Regardless, it seems inconsistent. But I don't have it listed in my version, and things work as expected. var OID = $feature.OBJECTID My issue is that I don't want to have to specify the field name three different times in the same code. I'm not saying Python is without problems, but coming from Python there would never be a case where I would have to hard-code a string three different times in the same extremely small block of code. I would assign the string field name to a variable, and then use that variable in every instance where it is needed--if the field name changes, I change the single instance of that variable assignment, and it updates everywhere. In my example, and both of your examples, we are specifying the same field name "Name_Text" three different times in the same tiny script. If this is a bug, fine, but if this is how the language was designed... why? ... View more

Appreciate the information Drew, thank you. ... View more

Well at least the problem is repeatable. Thanks for testing. ... View more

Appreciate the input, thanks! This option will have to do for now. Here's what my implementation looks like: HHMM is a TEXT field that the user types in, HH is a FLOAT field that gets calculated. Figured reading HH would be more intuitive to understand than MM, but the concept is the same. I try to use the same "template" for Attribute Rules, so this is probably overly verbose, but here's the automatic calculation from HH:MM text to HH float:   /* Converts a text HH:MM format field and converts it to a float HH field. */ //---------------------------------------------------------------------// // FUNCTIONS function calculateHH(hhmm) { // Set up results dict. If all checks pass, return this as-is. var result = Dictionary("VALUE", null, "ERROR", null) if (IsEmpty(hhmm)) { return result } Console(hhmm) var timeArr = Split(hhmm, ":") var hh = Number(timeArr[0]) + (Number(timeArr[1]) / 60) Console(`${hhmm} --> ${hh}`) result["VALUE"] = Round(hh, 2) return result } //---------------------------------------------------------------------// // INPUTS var hhmm = $feature.Hiking_HHMM //---------------------------------------------------------------------// // MAIN var fldUpdates = Dictionary() // Convert HHMM text to MM float ---------------- var result = calculateHH(hhmm) if (!IsEmpty(result["ERROR"])) {return {"errorMessage": `ERROR: ${result}`}} else {fldUpdates["Hiking_HH"] = result["VALUE"]} Console({"result": {"attributes": fldUpdates}}) return {"result": {"attributes": fldUpdates}}     And my solution to verifying the initial HH:MM inputs:   /* Script provides validation for a TEXT field that should contain a time duration in the format H?H?:MM. There are several checks, since RegEx is not available. 1) Blank strings are not allowed; null values are fine 2) Format must contain a colon--as in 3:30, 12:04, :37 3) Values on either side of the colon must be numeric 4) Hours are optional, but must be 0 - 23 5) Minutes must always be two characters, and numeric values 0 - 59 This script is meant to be used as Calculation type AR, rather than a Constaint AR. */ //---------------------------------------------------------------------// // FUNCTIONS function timeValidate (time) { // Set up results dict. If all checks pass, return this as-is. var result = Dictionary("VALUE", time, "ERROR", null) // First check empty, then null. Can't do both simultaneously nicely. if ((TypeOf(time) == "String") && Trim(time) == "") { result["ERROR"] = "Blank input is not allowed." return result } else if (IsEmpty(time)) { return result } // Console(timeArray, Count(timeArray)); // Must contain one ":", yields two elements with Split(). if (Count(timeArray) != 2) { result["ERROR"] = "Format must contain one colon." return result } // Check if both values in the array are numbers. for (var i in timeArray) { var tPart = timeArray[i] Console(`PART: ${i}, VAL: ${tPart}, ASNUM: ${Number(tPart)}, ISNAN: ${IsNan(Number(tPart))}`) // If Number returns NaN, "Casting a non-numeric text or undefined to a number". if (IsNan(Number(tPart))) { result["ERROR"] = "At least one value is not numeric." return result } // Hours validation, the first array item. if (i == 0) { Console(`VALIDATING HOURS: ${tPart}`) // Comparison operators coerce Text to Number for comparison. if (tPart > 23 || tPart < 0) { result["ERROR"] = "Hours must be 0-23." return result } } // Minutes validation, the second array item. else if (i == 1) { Console(`VALIDATING MINUTES: ${tPart}`) if (Count(tPart) != 2) { result["ERROR"] = "Minutes must be two digits." return result } else if (tPart > 59 || tPart < 0) { result["ERROR"] = "Minutes must be 0-59." return result } } } return result } //---------------------------------------------------------------------// // INPUTS // Expects($feature, "Shape_Length", "MPH", "Miles") var time = $feature.Hiking_HHMM // DERIVED var timeArray = Split(time, ":") Console(`TIME ARR: ${timeArray}`) //---------------------------------------------------------------------// // MAIN var fldUpdates = Dictionary() // result = {"VALUE": time, "ERROR": error string OR null} var result = timeValidate(time) // If an error is returned in results dict, show and make no changes. if (!IsEmpty(result["ERROR"])) {return {"errorMessage": `ERROR: ${result} Please use H?H?:MM format.`}} // Else, add the field update information to the running updates log. else {fldUpdates["Hiking_HHMM"] = time} Console({"result": {"attributes": fldUpdates}}) return {"result": {"attributes": fldUpdates}}     ... View more

For "data" reasons, a great solution--at the resolution I require, even minutes would work. My issue is that if a user is transcribing a record from elsewhere, say "7:37", they have to do the math to convert those hours into minutes (or hours and minutes into seconds, if seconds are my unit). Not a huge deal, but an extra step, and possibly error-prone. The other thing is quickly reading the table. I want to be able to read these records at a glance. Number of seconds (beyond a minute or two) means virtually nothing to me, and even minutes can get difficult: 407 minutes takes me a little while to process. So I'd really want that to show as HH:MM. And I don't know of a table "display" setting that would allow the math required to calculate that from seconds/minutes alone, and an Attribute Rule would... change the actual underlying data value? Plus, an Attribute Rule returning a value with the colon would be problematic with a DOUBLE field type. Storing as minutes is probably going to be the only solution here, but it's annoying that it can't be displayed in an easily readable way. My best bet might be to use an Attribute Rule to calculate the minutes into a SHORT/DOUBLE field from a TEXT field. So the user can type in 5:49 or whatever in the TEXT field, and then an Attribute Rule calculates the raw number of minutes into a different numeric field, and then valid sorting can take place on the MINUTES field. Appreciate the comment, thanks! ... View more

Your point about data types (String vs. NoneType) is an interesting one I hadn't considered, thanks for that. I can't replicate Pro (3.X) throwing an error regarding "nulling" a non-nullable field though, unless you're talking about using Field Calculator to "<Null>" a field, in which case it does yield an error. Thanks again for the detailed response. ... View more

I think I understand what you're after. I reworked some of the code internals to simplify things, so I'd try copying this version, rather than updating anything you're currently using (unless you're comfortable doing that). The outputs should be the same, with the addition of the new label field. I'm using a label expression to show the "point_id" + "quadrant" labels. This takes the angle of the line (say, 45 degrees) and then assigns the quadrant (NE). It then inverts that to get the opposing quadrant (SE). The code is set up so that if the line is perfectly north-south or east-west (90 degrees, 180 degrees) then those labels will be "N" and "S" or "W" and "E", without the secondary direction. Let me know if something doesn't work! Here's an W-E example, which I suppose would only occur... ~1% of the time?   """ RANDOM ANGLE TRANSECT LINES CENTERED ON SAMPLING POINTS """ #-----------------------------------------------------------------------------# # IMPORTS import math import os import random import sys from typing import Literal, Union, Tuple import arcpy #-----------------------------------------------------------------------------# # INPUTS # Full path to the geodatabase where the sampling points are, and where outputs will go. workspace = r"your_full_path\geodatabase_name.gdb" # Name of the point feature class in the above geodatabase that houses the # sampling points. in_pnt_fc = "_SAMPLE_POINT" # Distance between sampling points (along x and y, not diagonals). # Units match coordinate system. point_spacing = 400 # feet, in my case box_width = 80 # NORTH_BOX: squares will be oriented to grid-north # NORTH_DIAMOND: diamonds will be oriented to grid-north # ANGLE_BOX: square sides oriented to the tranect lines # ANGLE_DIAMOND: diamonds angled to the transect lines box_rotation = 'ANGLE_BOX' # OPTIONS: (NORTH_BOX, NORTH_DIAMOND, ANGLE_BOX, ANGLE_DIAMOND) # Name of the output feature class to be created. fc_line = "TRANSECT_LINE" fc_endpoints = "TRANSECT_LINE_ENDPOINTS" fc_box = f"TRANSECT_{box_rotation}" # -----------------------------------------------------------------------------# # SETTINGS # This just overwrites previous outputs to help with testing. arcpy.env.overwriteOutput = True #-----------------------------------------------------------------------------# # FUNCTIONS def rotate_point(coordinate: tuple[int, int], pivot_point: tuple[int, int], angle: Union[int, float], angle_format: Literal['DEGREES', 'RADIANS']='DEGREES' ) -> tuple[float, float]: """Rotates coordinate values around another point. :coordinate: (x, y) coordinate pair to be rotated :pivot_point: (x, y) coordinate pair around which point will be rotated :angle: angle to rotate the coordinate :angle_format: angle format, RADIANS or DEGREES |(x_prime, y_prime)| rotated point coordinates""" if angle_format == 'DEGREES': angle = math.radians(angle) x, y = coordinate X, Y = pivot_point x_trans = x - X y_trans = y - Y sin = math.sin(angle) cos = math.cos(angle) # Counter-clockwise rotation: x_transprime = cos * x_trans - sin * y_trans y_transprime = sin * x_trans + cos * y_trans x_prime = x_transprime + X y_prime = y_transprime + Y # print(f"XY TRANS: {x_trans:,.0f}, {y_trans:,.0f}\n" # f"XY TRANSPRIME: {x_transprime:,.0f}, {y_transprime:,.0f}\n" # f"XY PRIME: {x_prime:,.0f}, {y_prime:,.0f}\n") return (x_prime, y_prime) def create_box(method: Literal['NORTH_BOX', 'NORTH_DIAMOND', 'ANGLE_BOX', 'ANGLE_DIAMOND'], center_pnt: arcpy.Point, width: Union[int, float], angle: Union[int, float], spat_ref: int) -> arcpy.Polygon: """Create a square box around a given point. Optionally rotate it. :center_pnt: centroid of the square box :width: width of side of box :angle: angle the box should be rotated |box_geom| geometry object representing the box""" angle_dict = {'NORTH_BOX': 45, 'NORTH_DIAMOND': 0, 'ANGLE_BOX': angle + 45, 'ANGLE_DIAMOND': angle} # Distance to move point based on known hypotenuse (1/2 width of box). offset = math.sqrt(2 * ((width / 2) ** 2)) x, y = center_pnt.X, center_pnt.Y Nx, Ny = x, y + offset Ex, Ey = x + offset, y Sx, Sy = x, y - offset Wx, Wy = x - offset, y box_corners = [] for coordinate in [(Nx, Ny), (Ex, Ey), (Sx, Sy), (Wx, Wy), (Nx, Ny)]: box_corners.append(rotate_point(coordinate=coordinate, pivot_point=(x, y), angle=angle_dict[method], angle_format='DEGREES')) pnt_array = arcpy.Array([arcpy.Point(*corner) for corner in box_corners]) box_geom = arcpy.Polygon(pnt_array, spatial_reference=spat_ref) return box_geom def get_angle_quadrant(a): """As previously defined, angle resctricted [1-360].""" # Check for 90 degree cardinal directions. if a == 90: return "N" elif a == 180: return "W" elif a == 270: return "S" elif a == 360: return "E" # Determine non-90 quadrants. if 0 < a < 90: return "NE" elif 90 < a < 180: return "NW" elif 180 < a < 270: return "SW" elif 270 < a < 360: return "SE" else: raise ValueError("Angle value must be numeric and between 1-360, inclusive.") #-----------------------------------------------------------------------------# # MAIN in_points = os.path.join(workspace, in_pnt_fc) transect_lines = [] sample_box_dict = {} pnt_desc = arcpy.da.Describe(in_points) pnt_oid_fld = pnt_desc['OIDFieldName'] pnt_spat_ref = pnt_desc['spatialReference'].factoryCode with arcpy.da.SearchCursor(in_points, [pnt_oid_fld, 'SHAPE@XY']) as scurs: for oid, (x, y) in scurs: # Random angle in degrees, 1-360; convert to radians. angle_deg = random.randint(1, 360) angle_rad = math.radians(angle_deg) print(f"POINT {oid}, ANGLE: {angle_deg}") # Get x/y offset of random point on the imaginary circle around each point. start_x = (point_spacing/2) * math.cos(angle_rad) start_y = (point_spacing/2) * math.sin(angle_rad) # Using the sample point coordinate as the starting point, # calculate real-world x and y of both start and end of transect line. start_point = arcpy.Point(x + start_x, y + start_y) end_point = arcpy.Point(x - start_x, y - start_y) # -S/-E suffixes differentiate start/end boxes. Not currently used, but # documented just in case (also, unique keys are required). sample_box_dict[f'{oid}-S'] = create_box(box_rotation, start_point, box_width, angle_deg, pnt_spat_ref) sample_box_dict[f'{oid}-E'] = create_box(box_rotation, end_point, box_width, angle_deg, pnt_spat_ref) # Create a Polyline Geometry Object. Append dict entry to list. new_line = arcpy.Polyline(arcpy.Array([start_point, end_point])) transect_lines.append({'OID': oid, 'GEOM': new_line, 'ANGLE': angle_deg}) # Create three feature classes. for fc, geom in [(fc_line, 'POLYLINE'), (fc_endpoints, 'POINT'), (fc_box, 'POLYGON')]: arcpy.management.CreateFeatureclass(out_path=workspace, out_name=fc, geometry_type=geom, spatial_reference=pnt_spat_ref) print(f"CREATED FC {fc}") # FC paths. ln_fc = os.path.join(workspace, fc_line) pt_fc = os.path.join(workspace, fc_endpoints) box_fc = os.path.join(workspace, fc_box) # Add and ID field to the points, the lines, and box FCs. for fc in (ln_fc, pt_fc, box_fc): arcpy.management.AddField(fc, 'POINT_OID', 'SHORT') # Add angle field to the line FC. arcpy.management.AddField(ln_fc, 'LINE_ANGLE', 'SHORT') # Add the point type and the point angle to the point FC. arcpy.management.AddField(pt_fc, 'POINT_TYPE', 'TEXT', field_length=5) arcpy.management.AddField(pt_fc, 'POINT_QUAD', 'TEXT', field_length=5) print("ALL FIELDS ADDED") # Write transect line features----------------------------------# with arcpy.da.InsertCursor(ln_fc, ['SHAPE@', 'POINT_OID', 'LINE_ANGLE']) as icurs: for feature_dict in transect_lines: icurs.insertRow([feature_dict['GEOM'], feature_dict['OID'], feature_dict['ANGLE']]) print("TRANSECTS WRITTEN") # Write endpoint features----------------------------------# invert_quad = {"NE": "SW", "SE": "NW", "SW": "NE", "NW": "SE", "N": "S", "S": "N", "E": "W", "W": "E", } with arcpy.da.InsertCursor(pt_fc, ['SHAPE@', 'POINT_OID', 'POINT_TYPE', 'POINT_QUAD']) as icurs: for feature_dict in transect_lines: start_pnt = feature_dict['GEOM'].firstPoint end_pnt = feature_dict['GEOM'].lastPoint angle = feature_dict['ANGLE'] quadrant_1 = get_angle_quadrant(angle) quadrant_2 = invert_quad[quadrant_1] icurs.insertRow([start_pnt, feature_dict['OID'], "START", quadrant_1]) icurs.insertRow([end_pnt, feature_dict['OID'], "END", quadrant_2]) print("ENDPOINTS WRITTEN") # Create Sampling Zones----------------------------------# # Transect Line Oriented with arcpy.da.InsertCursor(box_fc, ['SHAPE@', 'POINT_OID']) as icurs: for oid, geom in sample_box_dict.items(): icurs.insertRow([geom, oid.split('-')[0]]) print("BOXES WRITTEN")   ... View more

I tested several methods of adding fields to a blank feature class in an attempt to overcome the sluggish process that is "AddField" in a for loop. Posting all the code might make this unreadable, so here is a summary. All tests result in the same output**--field names are the same, domains are the same, etc. In this example, I am adding 500 fields to a newly created feature class. There is no data involved here, this is  schema setup only. Methods & Times: Add fields in a for loop to a feature class already saved to file – 6:19 Add fields in a for loop to a memory feature class, then use CreateFeatureclass with the memory copy as the template – 0:56 Using AddFields (batch version with the “s”) – 4:56 FieldMappings() to create fields, ExportFeatures to file, AssignDomainToField in a loop – 2:09 FieldMappings() to create fields, ExportFeatures to memory, AssignDomainToField, CreateFeatureclass to file using memory as template – 0:19 So, in my experiments anyway, using a FieldMappings object and working in memory for as long as possible is the absolute fastest. BUT, simply adding fields to a memory feature class and using that as a template during the export to file is probably the simplest solution in terms of lines of code and maintenance.  Results from a more reasonable 40 fields, in the same order as above: 0:25 0:02 0:14 0:09 0:02 So even when generating a handful of feature classes with am more modest number of fields, it might best to avoid using AddField in a for loop for a feature class that is already saved to file. **AddFields does not allow the user to specify Scale and Precision, and I didn’t see a quick way to modify those attributes after the fact. In a FGDB, it is irrelevant, but this might matter in an enterprise environment. Here's an example of the randomly generated fields for three of the five output methods (all three should be the same).   ... View more