You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
These belong in V&U, rather - the concepts are already well-defined in terms of the CSS parser's output, and in any case the parser isn't aware of specific grammars when it produces its structures, until the very last step when it :handwaves: turns blocks and declarations into rules and properties.
I've retagged, but otherwise I agree that we should upstream these.
Shouldn't the parser at least have the necessary context for that last step though? It seems misleading to suggest the parser only needs bytes. Or am I missing something?
The CSS Working Group just discussed Upstream parser quirks to syntax, and agreed to the following:
RESOLVED: Add quicky hex color and quirky lengths to V&U
RESOLVED: Add a quirks mode flag to parsing algo and syntax which is passed to property specific parsers
The full IRC log of that discussion
<dael> Topic: Upstream parser quirks to syntax
<dael> github: https://github.com//issues/6100
<dael> TabAtkins: 2 simple bits
<dael> TabAtkins: Ana pointed out there's parser quirks for hashless and unitless
<dael> TabAtkins: In quirks spec
<dael> TabAtkins: agreed in past we should try and upstream to relevant specs
<dael> TabAtkins: agree can move in. Don't need to do anything in syntax b/c don't parse normally so should be in V&U where we define hashless hex and unitless length production
<dael> TabAtkins: Notes about when in quirks mode we accept these
<florian> q+
<dael> TabAtkins: One syntax change is let you pass in that you're parsing in quirks mode into syntax parse. Parsing algo gneerically evokes do what you need to parse and that should get quirks mode flag explicitly so it knows to accept quirky hex and length
<dael> TabAtkins: Asking for 2 resoltions. Add quicky hex color and quirky lengths to V&U
<dael> TabAtkins: Add an explicit quirk flag for parsing algo in syntax that is invoked at point syntax spec asks you to parse these
<astearns> ack florian
<dael> florian: I believe I know answer, none of these quirks could disappear?
<dael> TabAtkins: No, have been around for forever. As long as quirks mode exists, they will
<dael> fantasai: Fine, but shove into an appendix
<dael> TabAtkins: Happy to do that
<dael> astearns: Other comments?
<dael> RESOLVED: Add quicky hex color and quirky lengths to V&U
<dael> TabAtkins: Add a quirks mode flag to parsing algo and syntax which is passed to property specific parsers
<dael> astearns: Obj?
<dael> florian: Questions - what passes it to parsing algo?
<dael> TabAtkins: Collars. Things don't generally invoke those, but now they exist they can if necessary pass a quirks mode flag
<dael> astearns: Objections
<TabAtkins> s/Collars/Callers/
<dael> RESOLVED: Add a quirks mode flag to parsing algo and syntax which is passed to property specific parsers
Activity
tabatkins commentedon Mar 15, 2021
These belong in V&U, rather - the concepts are already well-defined in terms of the CSS parser's output, and in any case the parser isn't aware of specific grammars when it produces its structures, until the very last step when it :handwaves: turns blocks and declarations into rules and properties.
I've retagged, but otherwise I agree that we should upstream these.
annevk commentedon Mar 16, 2021
Shouldn't the parser at least have the necessary context for that last step though? It seems misleading to suggest the parser only needs bytes. Or am I missing something?
tabatkins commentedon Mar 16, 2021
Hm, yes, I suppose passing in the fact that you're doing quirks-mode parsing is relevant.
css-meeting-bot commentedon Mar 24, 2021
The CSS Working Group just discussed
Upstream parser quirks to syntax, and agreed to the following:RESOLVED: Add quicky hex color and quirky lengths to V&URESOLVED: Add a quirks mode flag to parsing algo and syntax which is passed to property specific parsersThe full IRC log of that discussion
<dael> Topic: Upstream parser quirks to syntax<dael> github: https://github.com//issues/6100
<dael> TabAtkins: 2 simple bits
<dael> TabAtkins: Ana pointed out there's parser quirks for hashless and unitless
<dael> TabAtkins: In quirks spec
<dael> TabAtkins: agreed in past we should try and upstream to relevant specs
<dael> TabAtkins: agree can move in. Don't need to do anything in syntax b/c don't parse normally so should be in V&U where we define hashless hex and unitless length production
<dael> TabAtkins: Notes about when in quirks mode we accept these
<florian> q+
<dael> TabAtkins: One syntax change is let you pass in that you're parsing in quirks mode into syntax parse. Parsing algo gneerically evokes do what you need to parse and that should get quirks mode flag explicitly so it knows to accept quirky hex and length
<dael> TabAtkins: Asking for 2 resoltions. Add quicky hex color and quirky lengths to V&U
<dael> TabAtkins: Add an explicit quirk flag for parsing algo in syntax that is invoked at point syntax spec asks you to parse these
<astearns> ack florian
<dael> florian: I believe I know answer, none of these quirks could disappear?
<dael> TabAtkins: No, have been around for forever. As long as quirks mode exists, they will
<dael> fantasai: Fine, but shove into an appendix
<dael> TabAtkins: Happy to do that
<dael> astearns: Other comments?
<dael> RESOLVED: Add quicky hex color and quirky lengths to V&U
<dael> TabAtkins: Add a quirks mode flag to parsing algo and syntax which is passed to property specific parsers
<dael> astearns: Obj?
<dael> florian: Questions - what passes it to parsing algo?
<dael> TabAtkins: Collars. Things don't generally invoke those, but now they exist they can if necessary pass a quirks mode flag
<dael> astearns: Objections
<TabAtkins> s/Collars/Callers/
<dael> RESOLVED: Add a quirks mode flag to parsing algo and syntax which is passed to property specific parsers
[css-color-4][css-values-4] Per WG resolution, adopt the quirky-lengt…