This is how it currently looks after using the script. As you can see with for example "Tverreggi" is matched against "eggi" and "egg". With the new method, it would only match against "eggi" since it wouldnt count the right side of the text as wildcard.
This is what i want(below), where as in the first example "Tverrdalstind" matched against both "dal" and "tind", it will now only match against "tind", since "dal" is not the end of the word and the new search will only check for wildcards on the left side of the text in "Ordbok", so if it has text on the right, it won't match. The other example shows text from "Ordbok" with capital letter at the start and therefore matches against the first part of the text "Nordrelandet" and "Nordre" since it only will search for the right side wildcards. This will give me fewer and more precise matches that I can use.
| SNAVN | OrdbokMatch |
|---|---|
| Tverrelvmyra | myra |
| Tverrdalstind | tind |
| Nordrelandet | Nordre |
Tverrdalstjønna | tjønna |
| Tverrdalselva | elva |
I made some changes to the code and this is what I get as result:
The code I used is posted below:
# -*- coding: utf-8 -*-
def main():
import arcpy
# settings (first uncompress geodatabase data!)
fc_anno = r'C:\GeoNet\SearchWildcard\gdb\TestBase.gdb\Navn_org'
fld_anno = 'SNAVN' # field to match with dictionary
fld_out = 'OrdbokMatch' # output field name
tbl = r'C:\GeoNet\SearchWildcard\gdb\TestBase.gdb\Ordbok'
fld_tbl = 'field' # dictionary data
# create dictionary list
lst_dct = [stripText2list(r[0]) for r in arcpy.da.SearchCursor(tbl, (fld_tbl))]
# Add output field to annotation
AddField(fc_anno, fld_out, "TEXT", 100)
# loop through annotation and update output field
cnt = 0
# dct_hits = {}
with arcpy.da.UpdateCursor(fc_anno, (fld_anno, fld_out)) as curs:
for row in curs:
cnt += 1
if cnt % 5000 == 0:
print "Processing row:", cnt
snavn = row[0]
lst_items = stripText2list(snavn)
result = FindMatches(lst_items, lst_dct)
row[1] = result
curs.updateRow(row)
def FindMatches(lst_items, lst_dct):
result = None
data = []
for lst_validate in lst_dct:
for validate in lst_validate:
for item in lst_items:
# if validate.upper() in item.upper():
if validate.istitle():
# first character is upper case, check start of string
if item.startswith(validate):
data.append(validate)
else:
# lower case, check end of string
if item.endswith(validate):
data.append(validate)
if len(data) > 0:
data = list(set(data))
result = data[0]
for data_item in data[1:]:
result += ", " + data_item
return result
def AddField(fc, fld_name, fld_type, fld_length=None):
if len(arcpy.ListFields(fc, fld_name)) == 0:
arcpy.AddField_management(fc, fld_name, fld_type, None, None, fld_length)
def stripText2list(text):
text_ori = text
lst_ok = []
if text is None:
return lst_ok
else:
s = text.find('(')
if s != -1:
e = text.find(')')
if e == -1:
e = len(text)-1
text = text[0:s] + text[e:len(text)-1]
text = text.replace('"','')
text = text.replace(';',' ')
text = text.replace('-',' ')
text = text.replace(',',' ')
while ' ' in text:
text = text.replace(' ',' ')
lst = text.split(' ')
for a in lst:
a = a.strip()
if a != '':
if a[0] != '(':
if a[-1:] != ')':
lst_ok.append(a)
return lst_ok
if __name__ == '__main__':
main()Modifications mainly on line 40 - 48 and I changed the 72 to 78 (due to something we discussed earlier)
