[squeak-dev] FormInspector, or also: Text>>#= and its consequences

Eliot Miranda eliot.miranda at gmail.com
Wed Jan 11 16:40:19 UTC 2023



> On Jan 11, 2023, at 5:37 AM, Taeumel, Marcel <Marcel.Taeumel at hpi.de> wrote:
> 
> 
> Hi Christoph --
> 
> At this point, let's not fiddle with Text >> #=.

+1

> 
> I think that comparing sets of attributes with each other can be as tricky as comparing morphs, unless you restrict yourself to very simple emphasis (bold/italic/...).
> 
> Identity-vs-state will bite you for "non-literal" attributes:
> - PluggableTextAttribute (i.e., compiled code, bindings, complex objects, ...)
> - TextFontReference (i.e., various font properties, form-set fonts, pixel comparison?, ...)
> - TextAnchor (i.e., morphs ...)
> 
> Thus, the very mechanism of text properties is so flexible that implementing a useful comparison should be done on a case-by-case basis. For example, compare #runs if necessary and sufficient, as suggested in the comment in #=.
> 
> Just comparing the identity of text attributes is not worth breaking backwards compatibility.

+1

I faced a similar issue when defining CompiledMethod >> #=. CompiledMethods hold onto their class via the method class association.  So if the same exact sequence of bytecodes and literals occurs in two methods but on different classes, comparing the method class associations will show them as different. But this kind of code duplication is often exactly what we want to identify. So CompiledMethod >> #= treats the method class associations comparison carefully; each method should have one or neither should.

As a thought experiment let’s imagine one wanted to compare two Smalltalk images to find out where they differ. One could progress through the Smalltalk dictionary in alphabetic order and compare classes and globals.  But classes refer to subclasses and superclasses, and methods, so the transitive closure from any class will include the Smalltalk dictionary.  That’s not a serious issue; we can add a visited set to avoid looping in the comparison. But if equality does look deep into the structure, and not just at a surface string (eg of methods and inst var names & class vars) then *any* difference, say in the value of a global in Smalltalk (release id?) will mark any class as different, and that’s not very useful.  Instead we probably want to compare classes “skin deep” (do they have the same superclass name, inst var names, class vars, methods and organization) and then we have a chance of finding the much more specific difference.

I know the text comparison issue is a little different. But the current definition is useful (after all the visual representation of a text is affected by the font and that’s not defined by the text itself (in our model)). And as Marcel says, one can always define a different one for special cases. And we have lots of precedence for that (isSameSequenceAs: et al).

Equality is tricky. A given definition must be useful “in a general context” (hand waving I know), and must match #hash, and must be reasonably efficient to be usable.  All that implies that in non-trivial implements a good comment should be written :-)

Happy ‘23!

> 
> Best,
> Marcel
> 
>> Am 16.09.2020 16:44:07 schrieb Thiede, Christoph <christoph.thiede at student.hpi.uni-potsdam.de>:
>> 
>> Hi Levente,
>> 
>> 
>> 
>> hm, I think #= should be always commutative and transitive, everything else is at least confusing ...
>> 
>> 
>> 
>> Can't we move that "attribute invariant" comparison rather to something like Text >> #sameAs:?
>> 
>> 
>> 
>> Best,
>> 
>> Christoph
>> 
>> Von: Squeak-dev <squeak-dev-bounces at lists.squeakfoundation.org> im Auftrag von Levente Uzonyi <leves at caesar.elte.hu>
>> Gesendet: Mittwoch, 16. September 2020 15:00:28
>> An: The general-purpose Squeak developers list
>> Betreff: Re: [squeak-dev] FormInspector, or also: Text>>#= and its consequences
>>  
>> Hi Christoph,
>> 
>> 
>> On Wed, 16 Sep 2020, Thiede, Christoph wrote:
>> 
>> > 
>> > Interesting, I would not have assumed that this would be only about performance, sounded like a more profound design decision to me.
>> 
>> If you have a look at the comment of Text >> #=, you'll find that it's a 
>> design decision (no reasoning though):
>> 
>> = other
>>          "Am I equal to the other Text or String?
>>          ***** Warning ***** Two Texts are considered equal if they have 
>> the same characters in them.  They might have completely different 
>> emphasis, fonts, sizes, text actions, or embedded morphs.  If you need to 
>> find out if one is a true copy of the other, you must do (text1 = text2 
>> and: [text1 runs = text2 runs])."
>> 
>> 
>> Though equality with Strings is not symmetric;
>> 
>> 'foo' asText = 'foo'. "==> true"
>> 'foo' = 'foo' asText. "==> false"
>> 
>> I don't know what relies on Text-String equality, but probably many things 
>> assume that Texts and Strings are somewhat interchangable (remember when 
>> you changed SHParserST80 >> #initializeVariablesFromContext, and Shout ran 
>> into errors because it expected source to be a String but got a Text?)
>> 
>> You can't keep equality with Strings if you change #= because you'll lose 
>> transitivity:
>> 
>> 'foo' asText allBold = 'foo'. "==> true"
>> 'foo' asText = 'foo'. "==> true"
>> 'foo' asText allBold = 'foo' asText "==> false"
>> 
>> 
>> Should you decide to change #=, remember to change #hash as well, and 
>> rehash all hashed collections with text keys.
>> 
>> 
>> Levente
>> 
>> > 
>> > 
>> > If no one sees a problem in changing this behavior, I can try my luck. :-)
>> > 
>> > 
>> > Best,
>> > 
>> > Christoph
>> > 
>> > __________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
>> > Von: Squeak-dev <squeak-dev-bounces at lists.squeakfoundation.org> im Auftrag von Taeumel, Marcel
>> > Gesendet: Dienstag, 15. September 2020 16:42:11
>> > An: squeak-dev
>> > Betreff: Re: [squeak-dev] FormInspector, or also: Text>>#= and its consequences  
>> > Hi Christoph.
>> > Performance. Change it, bench it, post the results here. :-) Please specify you machine and try it on a slow RaspPi, too.
>> > 
>> > Best,
>> > Marcel
>> >
>> >       Am 10.09.2020 20:32:34 schrieb Thiede, Christoph <christoph.thiede at student.hpi.uni-potsdam.de>:
>> >
>> >       Hi all,
>> > 
>> >
>> >       is there any old thread about the design discussion of how Text>>#= works? (It does not consider attributes for quality.) Has this decision ever been questioned?
>> > 
>> >
>> >       Naively and without an overview of any existing components that could rely on this implementation, I would like to question it.
>> >
>> >       Why should 'foo' asText allBold be equal to 'foo' asText addAttribute: TextURL new? With the same logic, we could also say that two dictionaries are equal iff they have got the same keys ...
>> > 
>> >
>> >       There is even a concrete client in the Trunk suffering from this design decision: Marcel's new FormInspector (and analogously, MorphInspector). It uses 
>> >
>> >       TextFontReference with a FormSetFont to display a screenshot right in the inspector pane. Unfortunately, the pane is never updated automatically because even if the screenshot changes, the text morph thinks the old text would equal the new one. I'd like to fix that without hacking any workaround into
>> >       the inspectors.
>> > Even though this inspector implementation is a bit unusual, in my opinion, it shows that the current Text >> #= implementation might not be a perfect solution.
>> > 
>> > I'm looking forward to your opinions.
>> > 
>> > Best,
>> > Christoph
>> > 
>> > 
>> >
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20230111/ed070a5c/attachment.html>


More information about the Squeak-dev mailing list