klut to AIG/XAG/MIG

[costamag]

Work done:

1. resynthesis function combining Disjoint Support Decomposition, shannon decomposition and NPN-based resynthesis.
   takes as input a k-LUT and returns a AIG, XAG or MIG network.
2. tests considering 4 cases for the three networks:
   case 1: the function is fully DS-decomposable;
   case 2: one step of DSD and then the NPN mapping is performed;
   case 3: No DSD. One step of shannon and then NPN mapping;
   case 4: One step of DSD, one of Shannon and then NPN mapping.

[lee30sonia]

Great job making so many tests! Just a few comments on little things...

[lee30sonia]

(Just to clarify) After resolving all the comments, please let me know (explicitly) that I should review it again; otherwise I will take it as still work-in-progress.

[costamag]

Dear Sonia, Sorry, I thought I had told you that I had re-submitted it some days ago but I was not sure about having done the documentation properly. Did you sent me other comments? All the best, Andrea Inviato da ProtonMail mobile

…

-------- Messaggio originale -------- On 30 Ott 2021, 00:05, Siang-Yun (Sonia) Lee ha scritto:

(Just to clarify) After resolving all the comments, please let me know (explicitly) that I should review it again; otherwise I will take it as still work-in-progress. — You are receiving this because you authored the thread. Reply to this email directly, [view it on GitHub](#502 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/AVN62RKLCFVTZLUK2MSTLD3UJMSCBANCNFSM5GVX7QPQ). Triage notifications on the go with GitHub Mobile for [iOS](https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675) or [Android](https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub).

[lee30sonia]

I saw there are some unresolved comments above so I thought you haven't completely finished. But now I see, they are outdated. Okay, I'll review again later.

[costamag]

I see, in the future I will resolve the comments explicitly. Thank you for pointing this out. All the best, Andrea Inviato da ProtonMail mobile

…

-------- Messaggio originale -------- On 30 Ott 2021, 08:25, Siang-Yun (Sonia) Lee ha scritto:

I saw there are some unresolved comments above so I thought you haven't completely finished. But now I see, they are outdated. Okay, I'll review again later. — You are receiving this because you authored the thread. Reply to this email directly, [view it on GitHub](#502 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/AVN62RMOK6OYNBCEU4DO7QDUJOMVBANCNFSM5GVX7QPQ). Triage notifications on the go with GitHub Mobile for [iOS](https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675) or [Android](https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub).

[codecov-commenter]

Codecov Report

Merging #502 (97671a6) into master (1d133e3) will increase coverage by 0.07%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #502      +/-   ##
==========================================
+ Coverage   84.17%   84.25%   +0.07%     
==========================================
  Files         140      142       +2     
  Lines       17336    17377      +41     
==========================================
+ Hits        14593    14641      +48     
+ Misses       2743     2736       -7     

Impacted Files	Coverage Δ
include/mockturtle/algorithms/klut_to_graph.hpp	100.00% <100.00%> (ø)
include/mockturtle/networks/mig.hpp	91.04% <0.00%> (-0.92%)	⬇️
...ude/mockturtle/algorithms/node_resynthesis/dsd.hpp	95.45% <0.00%> (ø)
...mockturtle/algorithms/node_resynthesis/xmg_npn.hpp	93.16% <0.00%> (+0.85%)	⬆️
include/mockturtle/networks/xmg.hpp	88.11% <0.00%> (+1.01%)	⬆️
...mockturtle/algorithms/node_resynthesis/xag_npn.hpp	91.17% <0.00%> (+4.41%)	⬆️
...mockturtle/algorithms/node_resynthesis/shannon.hpp	100.00% <0.00%> (+30.76%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 1d133e3...97671a6. Read the comment docs.

[lee30sonia]

To speed up the process, I fixed a few things directly by myself:

• I moved the file out of the node_resynthesis directory because it does not implement a new resynthesis method, but a wrapper to be used by higher-level users.
• Updates in the documentation: Some are simply writing/wording style, some are syntax issues (e.g. the :math: decoration, tparam for template parameters, insert an empty line to change line, etc.), and also some to enhance understandability for outside users and to reduce repetitions.
• Fixed inclusion style: Within mockturtle, we should use #include "<relative_path>.hpp". When including with #include <mockturtle/<some_algorithm>.hpp, we are using mockturtle as a system-level library. (See, e.g., here)
• I don't understand why you need the always_false<T> template class. [Edit: okay, I understand now that this is a work-around that people on the internet suggest to deal with the "ill-formed static_assert". After trying to understand why the standards prohibit this, I decided to do it differently in a way that is even more readable. That is, we check at the beginning of the function that the base type of NtkDest is one of the supported ones. Then, we don't have to worry about the "else" case in the if constexprs.]
• I removed the parameter and statistics objects because I suppose this function will be used by users who just want to quickly and easily convert the network and would not be interested in making it verbose or seeing the stats. This also simplifies the interface.
• Finally, I ran clang-format -i <file> to automatically correct spaces, intentations, etc.

[lee30sonia]

@costamag if you are fine with all the changes, this PR can be merged. Sorry for the delay.

[costamag]

dear @lee30sonia, everything is fine for me. Yesterday evening I saw that a typo in a header was giving troubles to the checks but if it is fixed I thank you for the adjustments and for the suggestions and I am fine for the merging.