Latest Contributions by JohannesLindner

Using Calculate Field in a Python script is a little tricky. As already said, you have to enclose field names with exclamation marks. What @dgiersz_cuyahoga did in his answer but didn't actually say: You have to supply the expression as a string. Your expression is (almost) correct when you use it in the Python window. For it to work in the tool, you have to turn it into a string. Just surround it with quotes (the type you didn't use in the expression, or triple quotes): # this works in a Python console, assuming LINKID is a defined variable 'https://msc.fema.gov/portal/downloadProduct?productID=' + str(LINKID) # or with newer syntax f'https://msc.fema.gov/portal/downloadProduct?productID={LINKID}' # for Calculate field, you have to supply the expression as string "'https://msc.fema.gov/portal/downloadProduct?productID=' + str(!LINKID!)" "f'https://msc.fema.gov/portal/downloadProduct?productID={!LINKID!}'"   ... View more

Seems like you have empty attachments. Are there entries with DATA_SIZE = 0 ? Try to include a where clause in the cursor: with da.SearchCursor(inTable, ["DATA", "ATT_NAME"], "DATA IS NOT NULL") as cursor: #... ... View more

Instead of getting the first row of the filtered featureset, you need a second for loop: // Populate feature array for (var site in sites) { var id = site.siteID // get all samples for this site, ordered by date var samples = OrderBy(Filter(fs, 'siteID = @ID'), 'sampleDate DESC') var i = 0 var prevDate // loop through the samples for(var sample in samples) { // abort after second feature i += 1 if(i > 2) { break } // abort if second feature has different date if(i == 2 && Text(sample.sampleDate, 'Y-MM-DD') != Text(prevDate, 'Y-MM-DD')) { break } // store date prevDate = sample.sampleDate // create the new feature and push it into the output fs var feat = { 'attributes': { //... }} Push(dowDict.features, feat); } };   ... View more

It's probably your symbology. Let's take BC_1 = 'CI20' as example.  The pink areas are polygons from the shapefile that aren't covered by your symbology:   These features are still in the layer, you can select them in the table and with Select Layer by Attribute, but they are not drawn on the map, so you can't select them with the selection tool and you can't see the selection outline and flash. If you load the shapefile (not the layers) into the map, you can select normally. This compares selecting in your layers and the "pure" shapefile. Without symbology, all 30 features are selected:   The problem are the GS seasons. The old shapefile lists these seasons as "GS 01", "GS 02", etc. You changed that to be consistent with the DS seasons. Now, they are called "GS_01", "GS_02", etc. You forgot to change the symbology. It doesn't draw the GS seasons earlier than GS_20, because it looks for spaces, but they are now stored with underscores:   ... View more

You have to do it in two steps: Use GroupBy to get the MAX date for each site Use a loop to fill your output featureset, filtering the input data I hate filtering with date values with a passion, because any number of things can go wrong, and the syntax depends on your database. I would use Distinct() instead: var p = Portal('https://arcgis.com'); // don't assign to names of functions! var fs = FeatureSetByPortalItem( p, 'XXX', 0, [ 'sys_estFreqReads_DeltaB16172', 'sys_estFreqReads_OmicronBA1', 'sys_estFreqReads_OmicronBA2', 'sys_estFreqReads_OmicronBA4', 'sys_estFreqReads_OmicronBA5', 'sys_estFreqReads_OmicronBA275', 'sys_sewershedName', 'sys_phu', 'sampleDate', 'siteID', 'lowQuality' ], false ); // create output fs var dowDict = { 'fields': [ { 'name': 'sampleDate', 'type': 'esriFieldTypeString'}, { 'name': 'siteID', 'type': 'esriFieldTypeString'}, {'name': 'sys_phu', 'type': 'esriFieldTypeString'}, {'name': 'sys_sewershedName', 'type': 'esriFieldTypeString'}, { 'name': 'sys_estFreqReads_DeltaB16172', 'type': 'esriFieldTypeDouble'}, { 'name': 'sys_estFreqReads_OmicronBA1', 'type': 'esriFieldTypeDouble'}, { 'name': 'sys_estFreqReads_OmicronBA2', 'type': 'esriFieldTypeDouble'}, { 'name': 'sys_estFreqReads_OmicronBA4', 'type': 'esriFieldTypeDouble'}, { 'name': 'sys_estFreqReads_OmicronBA5', 'type': 'esriFieldTypeDouble'}, { 'name': 'sys_estFreqReads_OmicronBA275', 'type': 'esriFieldTypeDouble'}, { 'name': 'dow_num', 'type': 'esriFieldTypeInteger'}, {'name': 'dow', 'type': 'esriFieldTypeString'} ], 'geometryType': '', 'features': [] }; // get all distinct combinations of ['siteID','sys_phu', 'sys_sewershedName'] var sites = Distinct(fs, ['siteID','sys_phu', 'sys_sewershedName']) // Populate feature array for (var site in sites) { var id = site.siteID var phu = site.sys_phu var sewershed = site.sys_sewershedName // get all samples for this site var samples = Filter(fs, 'siteID = @ID AND sys_phu = @phu AND sys_sewershedName = @sewershed') // get the latest sample var latest_sample = First(OrderBy(samples, 'sampleDate DESC')) // null check to avoid errors if(latest_sample == null) { continue } // create the new feature and push it into the output fs feat = { 'attributes': { 'sampleDate': Number(latest_sample['sampleDate']), 'siteID': latest_sample['siteID'], 'sys_phu': Text(latest_sample['sys_phu']), 'sys_sewershedName': Text(latest_sample['sys_sewershedName']), 'sys_estFreqReads_DeltaB16172': (latest_sample['sys_estFreqReads_DeltaB16172']), 'sys_estFreqReads_OmicronBA1': (latest_sample['sys_estFreqReads_OmicronBA1']), 'sys_estFreqReads_OmicronBA2': (latest_sample['sys_estFreqReads_OmicronBA2']), 'sys_estFreqReads_OmicronBA4': (latest_sample['sys_estFreqReads_OmicronBA4']), 'sys_estFreqReads_OmicronBA5': (latest_sample['sys_estFreqReads_OmicronBA5']), 'sys_estFreqReads_OmicronBA275': (latest_sample['sys_estFreqReads_OmicronBA275']), 'dow_num': Weekday(latest_sample['sampleDate']), 'dow': Text(latest_sample['sampleDate'], 'dddd') }} Push(dowDict.features, feat); }; // Convert dictionary to feature set. //var fs_dict = FeatureSet(Text(dowDict)); return FeatureSet(Text(dowDict)); ... View more

First Point I didn't think about timezones when I tested... The feature class stores date values in UTC. According to the documentation, the value returned by Date() is considered to be UTC, but the behavior you (and me, too) see suggests that it treats it as local time. You're in CET (UTC+1), so it subtracts the hour. In the summer months, you're in CEST (UTC+2), you should see values with a time of 22:00:00. I haven't found a way to change that behavior. ToUTC() and ToLocal() don't have any effect. Setting the hour to 1 (or adding an hour) works, but that still leaves the problem with summer time. I think in this case it's best to do it with Python. In the Field Calculation, switch to Python and use this expression: datetime.datetime.strptime(!zz_ff_jdatat!, "%d/%m/%Y")     Second Point This seems to be a bug. For validation, ArcGIS runs the expression on a row of your table. It apparently doesn't honor definition queries of the layer. You probably have a definition query set to "TextField IS NOT NULL" or something similar. When the tool runs, it doesn't encounter an error, but for validation, it grabs a feature that should be excluded from the layer. I just posted an idea about this, please lend your support to it, so that this will hopefully be fixed in the future: Honor Definition Query in Arcade Validation - Esri Community ... View more