I am working on a project where I have to capture residential address of multiple customers by opening different application forms in PDF format and enter or select that residential address on the property valuation website.
I am facing following challenges:
Format and length of residential address differ from customer to customer on respective application form. How to select/ capture every parameter (Zone, District, Estate/ Building name, Block Name, Floor, Flat).
I have to capture these details for every customer on each PDF and select the right parameter onHSBC, Hang Seng Bank and Bank of China property valuation websites by using the drop down list.
Every bank has defined the details in its drop down menu differently. In layman terms for example:
if the zone in the residential address (in PDF) is “HK” -
then on the bank’s website when you try to select the zone -
HSBC shows - “Hong Kong”
Hang Seng Bank shows - “Hong Kong Island”
BOC shows - “Hong Kong”
in their drop down menus.
How should I match the zone name (and this also applies to other parameters in the address - District, Building Name, Block, Floor & Flat) to the one in the drop down menu of respective bank accurately?
Please refer to the following websites and the addresses in the pdf samples here.
With the PDF forms, is there a pattern to their format, ie all PDFs from source A are the same as sample 1? If so you could take the approach where you standardise all addresses first, and create a data table, either in UiPath or external in Excel, then work from there?
As for the banks relying on different codes, HK/HKI etc, one option you could consider would be to create a reference file/matrix in an excel which your process could read to check which website you’re on, and therefore take the related value which you could bring in as a variable? For example, you’re on HSBC, so it checks which version of HONG KONG is required on their site?
@Alex_Cross,
I didn’t really understand what you suggested.
I don’t think the addresses in pdfs have any pattern. In that case, can you show me how can I pull out the exact data for different customers against different fields on banking application i.e. Zone, District, Building Name, Street name, Block, Room?
Sorry, I misunderstood the source of the PDF files. What I was trying to suggest, and still likely the best way, is to try and normalise the data from the PDFs, so you would be able to have a generic dataset of addresses. Perhaps you could check if the details given contain multiple lines (split by newline - How to split a variable value with respect to each each line?) and if this returns a value of >1, in other words there are multiple lines THEN split as such and record each entry either in an excel you create (write cell), or internally on a UiPath database (How to add new value (data row) in existing Datatable?). ELSE you could split by comma and then input into datatable the same way.
An advanced step for splitting would be to use regex checks to ensure the info you’re getting from each item in the array you’ve split (each line of address) matches what you’re expecting (Use regex in contain) as this could ensure you’re results aren’t unexpected, and if they are, it could email them to you to check for example.
When you have a normalised set of data you could then input this into the websites based on which one it is, for example, a flowchart could be used here - “is it HSBC?” - yes (flow to HSBC input), no - flow to “is it BOC?” etc… and in here you could convert HK as required etc.
Sorry, I don’t have time to create a workflow for you right now, I’m swamped myself, but I hope I’ve made sense and linked to appropriate forum threads to help guide you?
Hi @Alex_Cross, I apprehend the information you furnished me with. I also know you are swamped but I shall highly appreciate if you could demonstrate it to me anytime this week.
Anil_G
ashokkarale
jinal.shah
Gautham_Pattabiraman
postwick
chandreshsinh.jadeja
vrdabberu
Ajay_Mishra
sven.wullum1
Vyshnavi_Nalumachu