feat(SpoonifierVisitor): a visitor that generate the spoon code to recreate a spoon model

[nharrand]

…red to recreate a spoon model

[nharrand]

There are still some issues.
Works on src/test/java/spoon/test/spoonifier/testclasses/ArrayRealVector.java but not on src/test/java/spoon/test/prettyprinter/testclasses/ArrayRealVector.java

[monperrus]

Really cool. What about adding an assertEquals(code, spoonify(elem)) so that we have a feeling of the generated code?

[nharrand]

@monperrus Do you mean an assertion on the element itself like in src/test/java/spoon/test/spoonifier/SpoonifierTest.java:81 ? or do you mean an assertion on the code pretty printed?

[monperrus]

I mean an assertion on the code pretty printed.

[nharrand]

I was about to say that assertEquals on elements should be stronger then on the prettyprinted code, and that obviously two elements that are equal should be printed identically by construction. But of course I have tried...

[nharrand]

Ok it does not necessarily leads to the same pretty printing. If I understand correctly, one of the issue is containers that are map. Some elements contained might not be in the same order. (But the order does not carry semantic in this case.)
Example:

@spoon.test.annotation.testclasses.AnnotParamTypes(
integer = 42, 
integers = { 42 }, 
string = "Hello World!", 
strings = { "Hello", "World" }, 
clazz = java.lang.Integer.class, 
classes = { java.lang.Integer.class, java.lang.String.class }, 
b = true, 
byt = 42, 
c = 'c', 
s = ((short) (42)), 
l = 42, 
f = 3.14F, 
d = 3.14159, 
e = AnnotParamTypeEnum.G, 
ia = @InnerAnnot("dd")
)

becomes:

@spoon.test.annotation.testclasses.AnnotParamTypes(
strings = { "Hello", "World" }, 
string = "Hello World!", 
s = ((short) (42)), 
l = 42, 
integers = { 42 }, 
integer = 42, 
ia = @InnerAnnot("dd"), 
f = 3.14F, 
e = AnnotParamTypeEnum.G, 
d = 3.14159, 
clazz = java.lang.Integer.class, 
classes = { java.lang.Integer.class, java.lang.String.class }, 
c = 'c', byt = 42, 
b = true
)

I am not sure if we want to fix this, as I am not sure it is a problem. WDYT?

[monperrus]

Some elements contained might not be in the same order.

yes,

But I meant the generated code sorry!

[nharrand]

Ok I have added one. (I am not completely sure that the order is deterministic though. I just obtained different result by running it inside and outside of my debugger. (The issue was on the order of modifiers (PUBLIC and STATIC) addition))

[monperrus]

LGTM, will merge.

[nharrand]

@monperrus CI is finally happy!

[monperrus]

Cool, thanks!

[nharrand]

@arnobl Here is the feature we discussed in #3107 . It does not generate super concise code yet, but it is a start!