Fix Map

[Li-Xiang-Ideal]

CHANGES

• Fixes Map does not automatically map a function over Association values #1690 . Now Map automatically maps function over Association values
• Add operator forms for several functions, including Apply, Map, MapIndexed, MapThread

[Li-Xiang-Ideal]

A note on my absence and SparseArray:

I want to apologize for being silent for the past two years. Back in 2024, I underestimated the complexity of enhancing SparseArray, and honestly, my research workload became quite heavy, so I had to put open-source contributions on hold.

However, as part of my research, I’ve been working on a C++ based sparse tensor project SparseRREF during this time. Through this, I’ve gained more experience in implementing sparse tensor calculations. I remember @rocky once mentioned that "we are not experts in how to implement sparse arrays." While I'm not an expert yet, I definitely feel more confident about implementing sparse arrays now!

I can't promise a strict timeline yet, but I do plan to pick up the unfinished SparseArray work with my new experience when my schedule permits.

[Copilot]

Pull request overview

This pull request fixes issue #1690 where Map does not automatically map a function over Association values, and adds operator forms for several functional programming functions.

Changes:

• Modified Map to detect when the expression is an Association and apply the function to values only (preserving keys in Rule elements)
• Added operator form support for Apply, Map, MapIndexed, and MapThread using transformation rules

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The implementation only handles Rule but not RuleDelayed. Associations can contain both Rule and RuleDelayed elements (as seen in the codebase convention at mathics/builtin/list/associations.py:63, 91, 112, 153, 223, 324). The condition should be level.has_form(("Rule", "RuleDelayed"), 2) to handle both cases, and when creating the new expression, preserve the original head type using level.get_head() instead of always using SymbolRule.

Suggested change

	if is_association and level.has_form("Rule", 2):
	return Expression(
	SymbolRule,
	if is_association and level.has_form(("Rule", "RuleDelayed"), 2):
	return Expression(
	level.get_head(),

[mmatera]

This was a good idea. We can include it in another round.

[rocky]

LGTM

[rocky]

A note on my absence and SparseArray:

I want to apologize for being silent for the past two years.

No apology needed. This kind of thing happens a lot in open-source projects. We are grateful for any time you can devote here.

Back in 2024, I underestimated the complexity of enhancing SparseArray, and honestly, my research workload became quite heavy, so I had to put open-source contributions on hold.

However, as part of my research, I’ve been working on a C++ based sparse tensor project SparseRREF during this time. Through this, I’ve gained more experience in implementing sparse tensor calculations. I remember @rocky once mentioned that "we are not experts in how to implement sparse arrays." While I'm not an expert yet, I definitely feel more confident about implementing sparse arrays now!

Recent work by @bdlucas1 on better-performing plotting has revealed a few things.

See this discussion.

I can't promise a strict timeline yet, but I do plan to pick up the unfinished SparseArray work with my new experience when my schedule permits.

Again, no problem. This is perfectly understandable. Again, we are grateful for any time you can devote here.

[mmatera]

A note on my absence and SparseArray:

I want to apologize for being silent for the past two years. Back in 2024, I underestimated the complexity of enhancing SparseArray, and honestly, my research workload became quite heavy, so I had to put open-source contributions on hold.

@Li-Xiang-Ideal, I also have been interested in making progress on that stuff for a much longer time, and the same happened to me, so let's keep faith.

However, as part of my research, I’ve been working on a C++ based sparse tensor project SparseRREF during this time. Through this, I’ve gained more experience in implementing sparse tensor calculations. I remember @rocky once mentioned that "we are not experts in how to implement sparse arrays." While I'm not an expert yet, I definitely feel more confident about implementing sparse arrays now!

I can't promise a strict timeline yet, but I do plan to pick up the unfinished SparseArray work with my new experience when my schedule permits.

If you want, I started this repo https://github.com/Mathics3/Mathics3-Module-sparse, to start building SparseArray functions as a separate module. I cannot promise to make many contributions there until July, but if at some point you are up for that, I can try to help you start implementing some basic Builtins there.