Optimize function and interface object length computation in bindings generator

Optimize function and interface object length computation in bindings generator

Introduce new GetFunctionLength() function that efficiently compute the length
of a function (i.e. its number of mandatory parameters).

We now call GetFunctionLength() instead of GenerateFunctionParametersCheck()
whenever we care only interested in the function length and not the actual
expression for checking the parameters. This is much more efficient as
GenerateFunctionParametersCheck() does a lot more processing than we need in
this case.

NOTRY=true

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=149196

[haraken, 2013-04-26 11:32:16]

Maybe can you kill the first argument of GenerateFunctionParametersCheck()?

[do-not-use, 2013-04-26 11:37:46]

On 2013/04/26 11:32:16, haraken wrote:
> Maybe can you kill the first argument of GenerateFunctionParametersCheck()?

Do you mean the first return value (numMandatoryParams)?

If so, I'm not sure this is a good idea. All callers of
GenerateFunctionParametersCheck() are interested in both the number of mandatory
arguments and the parameter check expression.

If I remove the numMandatoryParams return value, the caller will need to call
both GetFunctionLength() and GenerateFunctionParametersCheck() which would be
less efficient (because it would iterate over the parameters twice).

Also, I think it would decrease readability a bit considering those caller are
really interested in the number of mandatory arguments, not the function length.
Sure, they are the same in latest Web IDL spec but this is not as obvious.

[haraken, 2013-04-26 11:48:40]

Thanks for the clarification. Makes sense. I don't think the efficiency matters
here, but readability matters.

LGTM.

[commit-bot: I haz the power, 2013-04-26 11:57:48]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ch.dumez@sisa.samsung.com/14002020/1

[commit-bot: I haz the power, 2013-04-26 12:28:42]

Retried try job too often on linux_layout_rel for step(s) webkit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_layo...

[commit-bot: I haz the power, 2013-04-26 12:32:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ch.dumez@sisa.samsung.com/14002020/1

[haraken, 2013-04-26 12:32:57]

Landing with NOTRY=true

[commit-bot: I haz the power, 2013-04-26 12:33:10]

Change committed as 149196