47 lines
4.1 KiB
Plaintext
47 lines
4.1 KiB
Plaintext
[video member=pervognsen stream_platform=twitch project=bitwise title="Functions & Statements" vod_platform=youtube id=VRMxHYuW2BY annotator=Miblo]
|
||
[0:07][Recap and set the stage for the day, reviewing the code from the previous (extra) stream][:speech]
|
||
[1:16][Review the work on resolve_expr_compound(), resolving components of compound literals, and on implementing support for named and index initialisers][:parsing :research]
|
||
[9:36][Review the work on integer binary operators, including constant folding][:parsing :research]
|
||
[11:19][Review the work on type alignment][:memory :parsing :research]
|
||
[14:21][Mention the hope to get to code generation, and determine to finish type-checking and inference for statements][:parsing :speech]
|
||
[15:18][Perform some cleanup, renaming the Entity struct to Sym][:parsing]
|
||
[17:52][Enable sym_get() to perform a linear search over symbols in nested scopes][:parsing :searching]
|
||
[23:17][:Run it to see that the existing stuff still works][:parsing]
|
||
[23:43][Introduce sym_push(), sym_enter() and sym_leave(), to operate on symbols within scopes][:parsing]
|
||
[27:38][Introduce resolve_func(), sym_var(), resolve_stmt_block(), resolve_stmt() and resolve_cond_expr(), to parse function bodies][:parsing]
|
||
[48:55][Explain the logic behind our for-loop :parsing][:speech]
|
||
[50:40][Continue to work through our statement cases – switch, assign and init – in resolve_stmt()][:parsing]
|
||
[1:02:28][:Run it to verify we don't crash][:parsing]
|
||
[1:02:47][Add a ++ test in resolve_test() and fix a crash bug in the STMT_ASSIGN case in resolve_stmt()][:parsing]
|
||
[1:04:24][:Run it to see it doesn't crash][:parsing]
|
||
[1:05:02][Add further tests of our new code in resolve_test(), running to see how it works][:parsing :programming :run]
|
||
[1:07:38][Hit a "Non-existent name" error][:parsing :run]
|
||
[1:07:48][Add a sym_install_type() for "float" in resolve_test()][:parsing]
|
||
[1:07:59][:Run it and hit our expected "Return type mismatch" error][:parsing]
|
||
[1:08:11][Add further tests, of switch(), in resolve_test(), running to see how it works][:parsing :programming :run]
|
||
[1:10:15][Crash in resolve_expected_expr()][:parsing :run]
|
||
[1:10:37][Fix the iterator in the STMT_SWITCH case in resolve_stmt()][:parsing]
|
||
[1:10:55][:Run it to see that it works][:parsing]
|
||
[1:11:03][Add further tests, of while(), in resolve_test(), running to see how it works][:parsing :programming :run]
|
||
[1:13:09][Add a type-checking test in resolve_test(), running to see how it works][:parsing :programming :run]
|
||
[1:13:53][Note that it pretty much all works, and that statements aren't difficult to type-check][:parsing :speech]
|
||
[1:15:19][Enable resolve_stmt() to handle both the presence and absence of a return value, adding a test, and running to see that it works][:parsing :run]
|
||
[1:17:32][Q&A][:speech]
|
||
[1:17:58][@0b0000000000000][You have duplicate f5 function names][:parsing]
|
||
[1:18:00][Fix the func test names in resolve_test()][:parsing]
|
||
[1:18:23][@0b0000000000000][Will there be function overloading?][:language]
|
||
[1:18:31][@chrisburtbrown][You might have answered this already but can functions contain other functions?][:parsing]
|
||
[1:20:00][@nothings2][This is why you use signed ints for everything and fuck size_t's unsignedness]
|
||
[1:21:51][@nothings2][Do compound literals require filling out all fields, or are they 0-inited?][:language]
|
||
[1:23:32][@recursivechat][Are you making a compiler? Or your own :language?]
|
||
[1:25:08][@buzzfizz64][@pervognsen Sorry if I missed it, but are you going to change the syntax of function pointers?][:language]
|
||
[1:26:17][@nothings2][@pervognsen I've forgotten the reason why you're writing a :language in the first place. Is there no gcc or clang for RISC-V?][:isa]
|
||
[1:27:15][@0b0000000000000][You'll also get a lot of coverage of the :language that way]
|
||
[1:27:51][@godling72][@pervognsen Debugger too?]
|
||
[1:29:35][@naysayer88][chocotDecent]
|
||
[1:29:41][@0b0000000000000][Will you be writing an interpreter for Ion in C?][:language]
|
||
[1:31:27][@0b0000000000000][How will you be confident that your Ion compilation is correct?][:language]
|
||
[1:33:11][@buzzfizz64][@pervognsen Are you going to do some sort of formal verification at some point?][:language]
|
||
[1:34:44][Stop recording][:speech]
|
||
[/video]
|