Insert sequential row and column values automatically into a fishnet dataset

Question

I have a grid that has about 30 rows with approx 15 columns each.

I want to label them as A1,A2,B1,B2 etc (Alphabet row, integer is column). Is there any way to do this programatically?

There is no fixed number of columns in a row, so we can not say that after every x columns the next item is B etc.

If necessary I guess we can compare the y on the top right of each grid and when it changes Row A becomes B but I am hoping that there is an easier way of doing this in SQL (Mapinfo) or QGIS.

Unfortunately, I don't have access to ArcGIS grid tool that does this.

I used "QGIS--Vector--Research--Vector Grid" to make the grid.

Another way of asking this is, can this QGIS tool be modified to include grid numbering like in the ArcGIS equivalent Grid Index Features (Cartography) and also ideally the surrounding map numbers.

I am also happy to use a mapbasic program to do this.

Answer 1

You can use pyqgis with the string and collections.defaultdict libraries.

Add a field named label of string/text type, change Grid to the name of your layer below and execute:

import string
from collections import defaultdict as dd

layer = QgsProject.instance().mapLayersByName('Grid')[0]
features = [f for f in layer.getFeatures()] #List all features
features.sort(key=lambda x: x.geometry().centroid().asPoint().y(), reverse=True) #Sort by y coordinate

rows = dd(list) #Create a defaultdict list which will have y coordinate (/the row identifier) as key and all features in that row as a list

for feature in features: 
    rowid = round(feature.geometry().centroid().asPoint().y())
    rows[rowid].append(feature)
# {6901627: [<qgis._core.QgsFeature object at 0x7f7804a1a160>,..., 
# 6896627: [<qgis._core.QgsFeature object at 0x7f7804a1a550>, <qgis._core.QgsFeature object at 0x7f7804a11af0>,

for rowid, featurelist in rows.items(): #Now sort each feature list by x coordinate
    rows[rowid] = sorted(rows[rowid], key=lambda x: x.geometry().centroid().asPoint().x())

fieldindex = layer.fields().indexFromName('label')
attributemap = {}
for row, letter, (_, features) in zip(rows, string.ascii_uppercase, rows.items()): #For each row
    for e, feature in enumerate(features, 1): #For each feature in the rows list of ids
        attributemap[feature.id()] = {fieldindex:letter+str(e)} #Store feature id and row letter and number as value
#attributemap is now 0: {1: 'A1'}, 3: {1: 'A2'}, 11: {1: 'A3'}, ...
layer.dataProvider().changeAttributeValues(attributemap)

(The code became complicated because I removed some grid cells):