Parse Accept #100
Parse Accept #100
Conversation
examples/accept/accept.zig
Outdated
| }){}; | ||
|
|
||
| fn on_request_verbose(r: zap.Request) void { | ||
| const allocator = gpa.allocator(); |
There was a problem hiding this comment.
Wouldn't it make sense to use a FixedBufferAllocator here? Maybe with a 1kB buffer?
src/request.zig
Outdated
| }; | ||
|
|
||
| /// Parses `Accept:` http header into `list`, ordered from highest q factor to lowest | ||
| pub fn parseAccept(self: *const Self, list: *std.ArrayList(AcceptItem)) !void { |
There was a problem hiding this comment.
Maybe it makes sense to make a type alias for the ArrayList?
So that users can write zap.AcceptList.init() instead of std.ArrayList(zap.AcceptItem).init()?
There was a problem hiding this comment.
And taking it one step further: instead of passing the ArrayList, let the user just pass an allocator and do the init within parseAccept. I think that would look very zig-like, so I propose:
pub fn parseAccept(self: *const Self, allocator: std.mem.Allocator) !std.ArrayList(AcceptItem)There was a problem hiding this comment.
Now, regarding naming: What do you think of the more verbose parseAcceptHeaders()?
pub fn parseAcceptHeaders(self: *const Self, allocator: std.mem.Allocator) !std.ArrayList(AcceptItem)
|
BTW really nice work! I like it! I especially like the catch break stuff! |
|
^^^ I just pushed a suggestion commit to this PR |
|
^^^ that last commit was meant for master 🤣 |
|
I liked your changes, I've done a little more work on top of them. I count the number of commas up front and allocate the array list. The comma to item relationship should hold for all valid Accept headers I think. |
|
Nice! Ready to merge if you are. |
|
Lets get it merged then :) |
Parses the Accept field from the header to return the ranked list of content types the client expects. I'd like feedback on the API - it doesn't seem quite right to me yet.