The Perl 6 design team met by phone on 02 September 2009. Larry, Allison,
Patrick, Jerry, Will, Nicholas, and chromatic attended.
Allison:
- mainly working on Foundations and non-code obligations
- as well as travel
- hope to be in the UK tomorrow and have a chance to work on bugs
- not that productive fixing bugs in airports
Will:
Jerry:
- down the rabbit hole, trying to get the SSL certificate for parrot.org
- expired over the weekend
- need access to the machine, can't remember my password, et cetera
Patrick:
- worked on grant things here and there last week
- did overloading for built-in operators
- can override
infix:<+> now
- moved the operators into the setting by writing them as Perl 6 (possibly with inline PIR for the bodies)
- colomon put in several hundred tests for that
- writing more of the builtins in Perl 6 has been a goal for a while
- glad to see we're making good progress on this
- did an initial implementation of the
Rat type for someone else to take over
- we don't seem to take a significant speed hit in the spectests
by doing this
- a big jump in the number of passing tests (and the size of the test suite)
- trying to support other people working on Rakudo
- making final changes for PGE and NQP plan
- made a bunch of t-shirt designs last night
- there are butterflies on them
- see http://rakudo.spreadshirt.com/, http://zazzle.com/rakudo, and http://cafepress.com/rakudo
Larry:
- tidied the definitions of division and modulus
- stole
::= to read-only binding, not compile time
- the
$* variables are now implicitly contextual (use $* in a declarator to create contextuals, where previously it was mainly is context)
- documented how declarations work inside of regular expression syntax, as suggested by Patrick
- we now forbid lists and ranges as the endpoints of range constructors
- that catches a bunch of errors
- documented the literal forms of
Rat and Complex
- decided that a .p6 extension is an acceptable way of identifying a Perl 6 source file
- tried to bring sanity to the paintball contest surrounding the range bikeshed
- clarified some of my positions in the discussion of perl6.org
- Camelia is only the logo for the language, not necessarily for distributions or implementations
- we can also have multiple websites with different cultural biases
- trying to discourage people from using the "Patches welcome" argument with respect to Camelia, because patches aren't actually welcome with respect to that
- at least not in the sense that most people are intending their patches
- did my share of rolling out the welcome mat for people on #perl6 and making sure that people didn't feel neglected
- did a little troll hugging, sometimes rather violently
- participated in the discussion of how the numeric roles relate to each other
- the current version of STD defaults everything to
my declarations, not our
- except for package-like declarations
- now uses the new forms for contextuals and read-only binding
gimme5 supports that- doesn't complain about Perl 5 ternary unless it finds a colon before the next semicolon
- no longer guesses about two terms in a row
- now just checks and looks ahead for the second term
- that lookahead could generate confusing error messages
- now does that in a
try to give a better error message
- the message recognizing use of statement control as a function was sometimes incorrect
- two different ways to get that -- explicit parens, in which the message is accurate
- can also use control where it doesn't expect a statement
- now suggests using
do for that
- writing a talk about the ecology of open source
c:
- reviewing the profiling code, preparing for a merge
- trying to slim down some structs
- thinking about Lorito and more GC changes
