| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
I'll let Gerrit call out who owns what --- apologies for letting this refactor CL get so unwieldy.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
nullptr, // Null-terminated array.It's not new in this CL but I'm curious why this one is needed.
std::to_array<const char* const>({Can it be `std::string_view`?
// Null-terminated list of first party extension IDs. We need to have thisAs the span has its size, maybe it doesn't have to be NULL terminated anymore?
RedactionTool(base::span<const char* const> first_party_extension_ids,Same question
base::span<const char* const> first_party_extension_ids =Can we switch to `string_view`?
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
nullptr, // Null-terminated array.It's not new in this CL but I'm curious why this one is needed.
This is a sentinel value to stop iteration (if you look down in the impl of `RedacitionTool`). This is also addressed in the downstream CL, where we no longer need a sentinel.
std::to_array<const char* const>({Can it be `std::string_view`?
This is done in the downstream CL. I noticed the refactor could be a bit more self-contained if we kept it as `const char* const` for now. This change involves more code owners than usual, so I prefer to minimize changes wherever possible.
If you really prefer, I could squash that CL into this one. That would smuggle the change to `std::string_view` here along with the business logic change, which marginally increases the chance I made a silly mistake that causes the whole CL to get reverted...
// Null-terminated list of first party extension IDs. We need to have thisAs the span has its size, maybe it doesn't have to be NULL terminated anymore?
Downstream CL
RedactionTool(base::span<const char* const> first_party_extension_ids,Kalvin LeeSame question
Downstream CL
base::span<const char* const> first_party_extension_ids =Can we switch to `string_view`?
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Code-Review | +1 |
whops forgot to hit the +1 button
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Spanify `RedactionTool` (1/2)
Broadly,
1. Convert `kBuiltInFirstPartyExtensionIds` to a `std::array` and
inline it directly in the associated header. This exposes its size
everywhere and makes it compatible with `base::span`.
2. Move away from trafficking `const char* const*`. Use `base::span`
where possible, and default it to the empty span where `nullptr`
would be passed.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
merge
so did it get wiped out ?? the tree wont open so ill take it by your neasyness of bluntly ignoring me while begging for help means I got got,,,,