Add offsets file
This commit is contained in:
parent
d8aba7da0b
commit
a0ae982202
64
src/offsets.zig
Normal file
64
src/offsets.zig
Normal file
@ -0,0 +1,64 @@
|
||||
const std = @import("std");
|
||||
const types = @import("types.zig");
|
||||
|
||||
fn utf16leCharSequenceLength(first_char: u16) !u2 {
|
||||
const c0: u21 = first_char;
|
||||
if (first_char & ~@as(u21, 0x03ff) == 0xd800) {
|
||||
return 2;
|
||||
} else if (c0 & ~@as(u21, 0x03ff) == 0xdc00) {
|
||||
return error.UnexpectedSecondSurrogateHalf;
|
||||
}
|
||||
return 1;
|
||||
}
|
||||
|
||||
pub const Encoding = enum {
|
||||
utf8,
|
||||
utf16,
|
||||
};
|
||||
|
||||
pub const DocumentPosition = struct {
|
||||
line: []const u8,
|
||||
line_index: usize,
|
||||
absolute_index: usize,
|
||||
};
|
||||
|
||||
pub fn documentPosition(doc: types.TextDocument, position: types.Position, encoding: Encoding) !DocumentPosition {
|
||||
var split_iterator = std.mem.split(doc.text, "\n");
|
||||
|
||||
var line_idx: i64 = 0;
|
||||
var line: []const u8 = "";
|
||||
while (line_idx < position.line) : (line_idx += 1) {
|
||||
line = split_iterator.next() orelse return error.InvalidParams;
|
||||
}
|
||||
|
||||
const line_start_idx = split_iterator.index.?;
|
||||
line = split_iterator.next() orelse return error.InvalidParams;
|
||||
|
||||
if (encoding == .utf8) {
|
||||
const index = @intCast(i64, line_start_idx) + position.character;
|
||||
if (index < 0 or index > @intCast(i64, doc.text.len)) {
|
||||
return error.InvalidParams;
|
||||
}
|
||||
return DocumentPosition{ .line = line, .absolute_index = @intCast(usize, index), .line_index = @intCast(usize, position.character) };
|
||||
} else {
|
||||
const utf8 = doc.text[line_start_idx..];
|
||||
var utf8_idx: usize = 0;
|
||||
var utf16_idx: usize = 0;
|
||||
while (utf16_idx < position.character) {
|
||||
if (utf8_idx > utf8.len) {
|
||||
return error.InvalidParams;
|
||||
}
|
||||
|
||||
const n = try std.unicode.utf8ByteSequenceLength(utf8[utf8_idx]);
|
||||
const next_utf8_idx = utf8_idx + n;
|
||||
const codepoint = try std.unicode.utf8Decode(utf8[utf8_idx..next_utf8_idx]);
|
||||
if (codepoint < 0x10000) {
|
||||
utf16_idx += 1;
|
||||
} else {
|
||||
utf16_idx += 2;
|
||||
}
|
||||
utf8_idx = next_utf8_idx;
|
||||
}
|
||||
return DocumentPosition{ .line = line, .absolute_index = line_start_idx + utf8_idx, .line_index = utf8_idx };
|
||||
}
|
||||
}
|
Loading…
Reference in New Issue
Block a user