Add support for item access from Const data-structures

[kmr-srbh]

Fixes #2578
The following changes were made:

• Handle subscripts for mutable types dict and list. Return the corresponding value.
• Throw an error for immutable types stating that Const is not required.

Fix

Dictionary

d: Const[dict[str, i32]] = {"a": 1, "b": 2, "c": 3}
print(d["a"])

(base) saurabh-kumar@Awadh:~/Projects/System/lpython$ ./src/bin/lpython ./examples/example.py
1

List

l: Const[list[i32]] = [1, 2, 3, 4, 5]
print(l[0])

(base) saurabh-kumar@Awadh:~/Projects/System/lpython$ ./src/bin/lpython ./examples/example.py
1

Tuple

t: Const[tuple[i32, i32, i32, i32, i32]] = (1, 2, 3, 4, 5)
print(t[0])

(base) saurabh-kumar@Awadh:~/Projects/System/lpython$ ./src/bin/lpython ./examples/example.py
semantic error: 'Const' not required as tuples are already immutable
  --> ./examples/example.py:10:4
   |
10 | t: Const[tuple[i32, i32, i32, i32, i32]] = (1, 2, 3, 4, 5)
   |    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 


Note: Please report unclear or confusing messages as bugs at
https://github.com/lcompilers/lpython/issues.

String

s: Const[str] = "Hello, LPython!"
print(s[0])

(base) saurabh-kumar@Awadh:~/Projects/System/lpython$ ./src/bin/lpython ./examples/example.py
semantic error: 'Const' not required as strings are already immutable
  --> ./examples/example.py:13:4
   |
13 | s: Const[str] = "Hello, LPython!"
   |    ^^^^^^^^^^ 


Note: Please report unclear or confusing messages as bugs at
https://github.com/lcompilers/lpython/issues.

[Thirumalai-Shaktivel]

Let's add test to intregration_tests and register in CMakeLists

[Thirumalai-Shaktivel]

Please mark this PR ready for review once it is ready.

[kmr-srbh]

@Shaikh-Ubaid @Thirumalai-Shaktivel there is a considerable degree of redundancy here, like for dictionary and lists. I have to implement slicing on Const list and the problem will arise there too. What is the fix? Functions?

[Thirumalai-Shaktivel]

What is the status of this PR?

The CI seems to fail, please mark this PR ready for review once it is ready.

[kmr-srbh]

What is the status of this PR?

The CI seems to fail, please mark this PR ready for review once it is ready.

@Thirumalai-Shaktivel this PR is ready. The only blocker is the issue stated here - #2567 (comment). Could you please look into it?

[kmr-srbh]

@Shaikh-Ubaid , type_get_past_const was removed in the new additions from LFortran. What do we do here now? Use get_contained_type()?

[Shaikh-Ubaid]

@Shaikh-Ubaid , type_get_past_const was removed in the new additions from LFortran. What do we do here now? Use get_contained_type()?

Simply remove the call to type_get_past_const(). Cont_t node no more exists. You need not use anything.

[kmr-srbh]

@Shaikh-Ubaid as the Const node was removed, this works out-of-the-box now. I have reverted all the previous unnecessary changes. The tests and semantic error for creating Const str and Const tuple have been retained. This is ready.

A related issue is the failing slicing on Const list and Const str. I will open an issue for it.

[Shaikh-Ubaid]

@kmr-srbh For this PR, you can simply add an integration test that verifies that const list and dict and operation on them are supported by lpython. And it will be good to merge.

[Shaikh-Ubaid]

Looks good. Minimal changes that just do the task. Thanks for this.