Too short variable names

[olexandr-konovalov]

After loading, the following too short names starting with capitals are created:

[ "Fix", "TGS", "YB" ] 

This is not a very good practice, see https://www.gap-system.org/Manuals/doc/ref/chap76.html#X7DEACD9786DE29F1 - especially with such a trivial name like Fix.

[fingolfin]

@vendramin even after your PR to fix the documentation for Fix, it is not quite clear to me why it is called like that?

[vendramin]

This is the standard terminology that appears in the literature. It is indeed the set of fixed points of the action given by the map lambda. Did I make a mistake?

…

On Tue, Aug 2, 2022 at 7:24 AM Max Horn ***@***.***> wrote: @vendramin <https://github.com/vendramin> even after your PR to fix the documentation for Fix, it is not quite clear to me why it is called like that? — Reply to this email directly, view it on GitHub <#71 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACC7PG74FQFSG5TT4TKMY3DVXDZMLANCNFSM4LE4DVXQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[vendramin]

We can use a different name, but I don't think FixLeftIdeal is a better choice. If Fix and YB are not Okay, what would you do? Maybe GAP has rules for names like those... On Tue, Aug 2, 2022 at 7:25 AM Leandro Vendramin ***@***.***> wrote:

…

This is the standard terminology that appears in the literature. It is indeed the set of fixed points of the action given by the map lambda. Did I make a mistake? On Tue, Aug 2, 2022 at 7:24 AM Max Horn ***@***.***> wrote: > @vendramin <https://github.com/vendramin> even after your PR to fix the > documentation for Fix, it is not quite clear to me why it is called like > that? > > — > Reply to this email directly, view it on GitHub > <#71 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/ACC7PG74FQFSG5TT4TKMY3DVXDZMLANCNFSM4LE4DVXQ> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >

[olexandr-konovalov]

@vendramin GAP has some rules (or rather suggestions) indeed: https://docs.gap-system.org/doc/ref/chap76_mj.html#X7DEACD9786DE29F1. We need to think of some alternatives...

[olexandr-konovalov]

@vendramin let's come back to this!

1. Trying to understand what 'Fix' is actually returning. The manual says it returns an ideal, but it seems like returns a list. You said above that it's a set of of fixed points. Maybe FixedPointsOfSkewBrace?

2. TGS is undocumented and defined here

   YangBaxter/data/TGsmall.g

   Line 1 in 3958563

   TGS:=[];

   
   Seems to me a global variable to store some data - could be called YangBaxterTGS or something. Could be also made write protected.

3. YB is undocumented and defined here

   YangBaxter/lib/ybe.gi

   Lines 4 to 21 in 3958563

   	### This function returns the set-theoretic solution given by the permutations <l_actions> and <r_actions>
   	### <l_action> and <r_action> are matrices!
   	InstallMethod(YB, "for a two lists of permutations", [ IsList, IsList ],
   	function(l, r)
   	local obj;
   	if not IS_YB(l, r) then
   	Error("this is not a solution of the YBE");
   	fi;
   	obj := Objectify(YBType, rec());
   	SetSize(obj, Size(l));
   	SetLMatrix(obj, List(l, x->List(x, y->y)));
   	SetRMatrix(obj, List(r, x->List(x, y->y)));
   	return obj;
   	end);

   
   Maybe YBSolutionByPerms?

[vendramin]

1. Fix should return a left ideal, as this is not an ideal in general. If you don’t like Fix as the function's name (for example, because it is dangerous), we can go with your suggestion. 2. If I am not wrong, this function belongs to a file that is just experimental. Is this inside tau.gi? We could remove the files tau.gi and tau.gd, I think. 3. If we need to change the name, again because it goes against the rules, we can use your suggesstion. Other possible names are YBFromPermutations and essentially every other variant. Feel free to use your experience and choose. I tend to use very short names for functions and I am even worse with variable names :)

…

On 18 May 2023, at 15:00, Olexandr Konovalov ***@***.***> wrote: @vendramin let's come back to this! • Trying to understand what 'Fix' is actually returning. The manual says it returns an ideal, but it seems like returns a list. You said above that it's a set of of fixed points. Maybe FixedPointsOfSkewBrace? • TGS is undocumented and defined here https://github.com/gap-packages/YangBaxter/blob/3958563edddb606b4a2c284c430fe3fb72e47558/data/TGsmall.g#L1 Seems to me a global variable to store some data - could be called YangBaxterTGS or something. Could be also made write protected. • YB is undocumented and defined here https://github.com/gap-packages/YangBaxter/blob/3958563edddb606b4a2c284c430fe3fb72e47558/lib/ybe.gi#L4-L21 Maybe YBSolutionByPerms? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[olexandr-konovalov]

1. FixedPointsOfSkewBrace is ok then. My concerns is that it returns a list - so a different type of object, not the one promised in the manual (neither ideal nor left ideal)

2. Yes, it's a data file which is read from tau.gd. So we can delete tau.* and data/TGsmall.g then?

3. I would still go with YBSolutionByPerms - By convention is documented in GAP conventions, and Perms is used in various places as an abbreviation.

[vendramin]

So 1 should return a left ideal. We can remove these files, but I would like to keep them in another folder, just in case. Okay with 3. 

[olexandr-konovalov]

1. So then one should not just rename 1, but also fix the output - generate a left ideal by the list currently returns?
2. If you remove them, they will still be recorded in the repository's revisions history though. If you have doubts, we can keep them, or move them to a branch instead of deleting... no hurry needed for that.
3. ok, thanks!

[vendramin]

For 1. One has a function to compute the left series and it returns a list of left ideals. The same behavior is expected when computed the fix, I think.