function to match a native TIA portal export to a DB layout specification string

[Novecento99]

Hi,
as anticipated I was working on a function that could ingest a native TIA portal export of a non-optimized db to generate a string compatible with the DB class layout_specification

It also manage names duplicates by adding a suffix "_X" at the end of the variables, where "X" is a progressive numeration.

[nikteliy]

nit: If you replace the list with a set, the operation x in s will be faster

[Novecento99]

done

[nikteliy]

Is there a reason why the literals in valid_names are uppercase? If they were lowercase, you wouldn't need to call .upper() for every line

[Novecento99]

valid_names are list a copied from another part of this repo.
TIA portal exports variabiles types with the first letter in uppercase anyway

[nikteliy]

I don't know what the export file format looks like. Can there be empty lines or comments in it? This could cause an exception, something like 'list index out of range'

[Novecento99]

as far as I know: no.
this is an example of tia export:

PRESENT Bool 3968.0 false --- False True True True False PIECE PRESENT SCRAP Bool 3968.1 false --- False True True True False PIECE SCRAP READY Bool 3968.2 false --- False True True True False PIECE READY DRY_TEST Bool 3968.3 false --- False True True True False DRY TEST MOVING Bool 3968.4 false --- False True True True False PIECE MOVING SCANNED Bool 3968.5 false --- False True True True False QR_NOK Bool 3968.6 false --- False True True True False QR CODE NOT OK LONG_PROFILE Bool 3968.7 false --- False True True True False VB_ACK Bool 3969.0 false --- False True True True False

[nikteliy]

If var_type is an empty string, then var_type in valid_list is will be False. So there is no point to check var_type

[Novecento99]

okay thankyou

[Novecento99]

done

[nikteliy]

I think using a dictionary, where the key is var_name and the value is a counter, would be cleaner and more efficient than using a list. Something like this:

var_names: dict[str, int] = {}
to_add = var_names.setdefault(var_name, 0)
var_names[var_name] += 1
var_name = f'{var_name}_{to_add}'

And you won’t need an inner loop in this case.

[nikteliy]

nit: Perhaps appending "new_line" to a list and then using join() is more efficient. But I’m not sure it makes a significant difference

[Novecento99]

I doubt so, especially considering this function is meant to be run only one time for db....

[lupaulus]

@Novecento99 Nice work

[gijzelaerr]

Code Review

Thanks for contributing this TIA Portal export parser! This is a useful feature. However, there are several issues that need to be addressed before we can merge.

1. Bug: Duplicate name detection is broken for underscored variables

The current logic uses rsplit("_")[0] which fails for variable names containing underscores:

>>> "my_value_0".rsplit("_")
['my', 'value', '0']
>>> "my_value_0".rsplit("_")[0]
'my'  # Expected: 'my_value'

This means duplicates of my_value will all get _0 suffix instead of _0, _1, _2:

# Current behavior:
'my_value' -> 'my_value_0'
'my_value' -> 'my_value_0'  # BUG: should be 'my_value_1'

Fix: Use rsplit("_", 1) to split only on the last underscore:

if name.rsplit("_", 1)[0] == var_name:
    to_add = "_" + str(int(name.rsplit("_", 1)[-1]) + 1)

2. Remove debug print statement

Line with print(name.rsplit("_")[-1]) should be removed or replaced with logger.debug().

3. Docstring parameter mismatch

The docstring says tia_export but the parameter is named txt_path.

4. Missing error handling

What happens if:

• A line doesn't have 3 tab-separated fields?
• The file doesn't exist?
• The offset field isn't valid?

Consider adding try/except or validation.

5. Incomplete type list

The valid_list is missing types that are supported elsewhere in the codebase:

• WORD, UINT, UDINT, LREAL, SINT, USINT, TIME, DATE

Consider matching the types supported in Row.get_value() and Row.set_value().

6. Inefficient string concatenation

Using db_specification = db_specification + "\n" + new_line in a loop is O(n²). Use a list instead:

lines = []
for line in file:
    ...
    if var_type in valid_list:
        lines.append(f"{var_offset}\t{var_name}\t{var_type}")
return "\n".join(lines)

7. No tests

Please add unit tests. At minimum:

• Test with a sample TIA export file
• Test duplicate name handling (including names with underscores)
• Test with missing/invalid fields

8. Separate formatting changes

The PR includes many unrelated line-wrapping changes. Please split those into a separate PR or remove them to keep this PR focused.

---

Summary of required changes:

1. Fix the rsplit bug (use rsplit("_", 1))
2. Remove print() statement
3. Fix docstring
4. Add basic error handling
5. Add missing types to valid_list
6. Use list + join instead of string concatenation
7. Add unit tests
8. Remove unrelated formatting changes

Happy to help if you have questions!