mirror of
https://github.com/roc-lang/roc.git
synced 2025-10-03 08:34:33 +00:00
Remove examples/task for now
This commit is contained in:
parent
e0c89e1208
commit
37e67e58a3
8 changed files with 9 additions and 468 deletions
|
@ -202,15 +202,15 @@ mod cli_run {
|
|||
// use_valgrind: true,
|
||||
// }
|
||||
// ],
|
||||
"task" => vec![
|
||||
Example {
|
||||
filename: "Main.roc",
|
||||
executable_filename: "task-example",
|
||||
stdin: &[],
|
||||
expected_ending: "successfully wrote to file\n",
|
||||
use_valgrind: true,
|
||||
}
|
||||
],
|
||||
// "task" => vec![
|
||||
// Example {
|
||||
// filename: "Main.roc",
|
||||
// executable_filename: "task-example",
|
||||
// stdin: &[],
|
||||
// expected_ending: "successfully wrote to file\n",
|
||||
// use_valgrind: true,
|
||||
// }
|
||||
// ],
|
||||
"benchmarks" => vec![
|
||||
Example {
|
||||
filename: "NQueens.roc",
|
||||
|
|
2
examples/task/.gitignore
vendored
2
examples/task/.gitignore
vendored
|
@ -1,2 +0,0 @@
|
|||
task-example
|
||||
thing.txt
|
|
@ -1,19 +0,0 @@
|
|||
app "task-example"
|
||||
packages { base: "platform" }
|
||||
imports [ base.Task.{ Task }, base.File, base.Path ]
|
||||
provides [ main ] to base
|
||||
|
||||
main : Task.Task {} []
|
||||
main =
|
||||
when Path.fromStr "examples/task/thing.txt" is
|
||||
Ok path ->
|
||||
{} <- Task.await (Task.putLine "Writing to file")
|
||||
|
||||
result <- Task.attempt (File.writeUtf8 path "zig is awesome")
|
||||
|
||||
when result is
|
||||
Ok _ -> Task.putLine "successfully wrote to file"
|
||||
Err BadThing -> Task.putLine "error writing to file"
|
||||
Err _ -> Task.putLine "something worse"
|
||||
|
||||
_ -> Task.putLine "invalid path"
|
|
@ -1,97 +0,0 @@
|
|||
interface File
|
||||
exposes [ FileReadErr, FileOpenErr, FileWriteErr, DirReadErr, readUtf8, writeUtf8 ]
|
||||
imports [ Task.{ Task }, fx.Effect.{ after }, Path ]
|
||||
|
||||
# TODO FIXME should be able to import this as Path.{ Path }, but there's a bug.
|
||||
Path : Path.Path
|
||||
|
||||
# These various file errors come from the POSIX errno values - see
|
||||
# http://www.virtsync.com/c-error-codes-include-errno for the actual codes, and
|
||||
# https://www.gnu.org/software/libc/manual/html_node/Error-Codes.html for documentation
|
||||
#
|
||||
# The goal of this design is:
|
||||
# * Whenever a function returns a `Task`, that task's error type represents all the errors that could happen.
|
||||
# * The errors are union-friendly; if I run a task that reads, and then another that writes, I should get all the read *and* write errors.
|
||||
# * To make the errors friendlier to chaining, they should always include the `Path` of the attempted operation. This way it's possible to tell which one failed after the fact.
|
||||
|
||||
|
||||
## These errors can happen when opening a file, before attempting to read from
|
||||
## it or write to it. The #FileReadErr and #FileWriteErr tag unions begin with
|
||||
## these tags and then add more specific ones.
|
||||
FileOpenErr a :
|
||||
[
|
||||
FileNotFound Path,
|
||||
PermissionDenied Path,
|
||||
SymLinkLoop Path,
|
||||
TooManyOpenFiles Path,
|
||||
IoError Path,
|
||||
UnknownError I64 Path,
|
||||
]a
|
||||
|
||||
## Errors when attempting to read a non-directory file.
|
||||
FileReadErr a :
|
||||
FileOpenErr
|
||||
[
|
||||
FileWasDir Path,
|
||||
InvalidSeek Path,
|
||||
IllegalByteSequence Path,
|
||||
FileBusy Path,
|
||||
]a
|
||||
|
||||
## Errors when attempting to read a directory.
|
||||
DirReadErr a :
|
||||
FileOpenErr
|
||||
[
|
||||
FileWasNotDir Path,
|
||||
]a
|
||||
|
||||
## Errors when attempting to write a non-directory file.
|
||||
FileWriteErr a :
|
||||
FileOpenErr
|
||||
[
|
||||
FileWasDir Path,
|
||||
ReadOnlyFileSystem Path,
|
||||
]a
|
||||
|
||||
|
||||
## Read a file's raw bytes
|
||||
#readBytes : Path -> Task (List U8) (FileReadErr *)
|
||||
#readBytes = \path ->
|
||||
# Effect.readBytes (Path.toStr path)
|
||||
|
||||
## Read a file's bytes and interpret them as UTF-8 encoded text.
|
||||
readUtf8 : Path -> Task.Task Str (FileReadErr [ BadUtf8 Str.Utf8ByteProblem Nat ]*)
|
||||
readUtf8 = \path ->
|
||||
Effect.map (Effect.readAllUtf8 (Path.toStr path)) \answer ->
|
||||
# errno values - see
|
||||
# https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/errno.h.html
|
||||
when answer.errno is
|
||||
0 ->
|
||||
when Str.fromUtf8 answer.bytes is
|
||||
Ok str -> Ok str
|
||||
Err (BadUtf8 problem index) -> Err (BadUtf8 problem index)
|
||||
1 -> Err (PermissionDenied path)
|
||||
2 -> Err (FileNotFound path)
|
||||
19 -> Err (FileWasDir path)
|
||||
# TODO handle other errno scenarios that could come up
|
||||
_ -> Err (UnknownError answer.errno path)
|
||||
|
||||
writeUtf8 : Path, Str -> Task.Task {} (FileWriteErr [ BadThing ]*)
|
||||
writeUtf8 = \path, data ->
|
||||
path
|
||||
|> Path.toStr
|
||||
|> Effect.writeAllUtf8 data
|
||||
|> Effect.map \res ->
|
||||
when res.errno is
|
||||
0 -> Ok {}
|
||||
_ -> Err BadThing
|
||||
|
||||
## Read a file's bytes, one chunk at a time, and use it to build up a state.
|
||||
##
|
||||
## After each chunk is read, it gets passed to a callback which builds up a
|
||||
## state - optionally while running other tasks.
|
||||
#readChunks : Path, U64, state, (state, List U8 -> Task state []err) -> Task state (FileReadErr err)
|
||||
|
||||
## Like #readChunks except after each chunk you can either `Continue`,
|
||||
## specifying how many bytes you'd like to read next, or `Stop` early.
|
||||
#readChunksOrStop : Path, U64, state, (state, List U8 -> [ Continue U64 (Task state []err), Stop (Task state []err) ]) -> Task state (FileReadErr err)
|
|
@ -1,16 +0,0 @@
|
|||
platform folkertdev/foo
|
||||
requires {}{ main : Task {} [] }
|
||||
exposes []
|
||||
packages {}
|
||||
imports [ Task ]
|
||||
provides [ mainForHost ]
|
||||
effects fx.Effect
|
||||
{
|
||||
# TODO change errno to I32
|
||||
readAllUtf8 : Str -> Effect { errno : I64, bytes : List U8 },
|
||||
writeAllUtf8 : Str, Str -> Effect { errno: I64 },
|
||||
putLine : Str -> Effect {}
|
||||
}
|
||||
|
||||
mainForHost : Task.Task {} [] as Fx
|
||||
mainForHost = main
|
|
@ -1,16 +0,0 @@
|
|||
interface Path
|
||||
exposes [ Path, fromStr, toStr ]
|
||||
imports []
|
||||
|
||||
|
||||
Path : [ @Path Str ]
|
||||
|
||||
|
||||
fromStr : Str -> Result Path [ MalformedPath ]*
|
||||
fromStr = \str ->
|
||||
# TODO actually validate the path - may want a Parser for this!
|
||||
Ok (@Path str)
|
||||
|
||||
toStr : Path -> Str
|
||||
toStr = \@Path str ->
|
||||
str
|
|
@ -1,41 +0,0 @@
|
|||
interface Task
|
||||
exposes [ Task, succeed, fail, await, map, putLine, attempt ]
|
||||
imports [ fx.Effect ]
|
||||
|
||||
|
||||
Task ok err : Effect.Effect (Result ok err)
|
||||
|
||||
|
||||
succeed : val -> Task val *
|
||||
succeed = \val ->
|
||||
Effect.always (Ok val)
|
||||
|
||||
|
||||
fail : err -> Task * err
|
||||
fail = \val ->
|
||||
Effect.always (Err val)
|
||||
|
||||
|
||||
await : Task a err, (a -> Task b err) -> Task b err
|
||||
await = \effect, transform ->
|
||||
Effect.after effect \result ->
|
||||
when result is
|
||||
Ok a -> transform a
|
||||
Err err -> Task.fail err
|
||||
|
||||
attempt : Task a b, (Result a b -> Task c d) -> Task c d
|
||||
attempt = \effect, transform ->
|
||||
Effect.after effect \result ->
|
||||
when result is
|
||||
Ok ok -> transform (Ok ok)
|
||||
Err err -> transform (Err err)
|
||||
|
||||
map : Task a err, (a -> b) -> Task b err
|
||||
map = \effect, transform ->
|
||||
Effect.after effect \result ->
|
||||
when result is
|
||||
Ok a -> Task.succeed (transform a)
|
||||
Err err -> Task.fail err
|
||||
|
||||
putLine : Str -> Task {} *
|
||||
putLine = \line -> Effect.map (Effect.putLine line) (\_ -> Ok {})
|
|
@ -1,268 +0,0 @@
|
|||
const std = @import("std");
|
||||
const str = @import("str");
|
||||
const RocStr = str.RocStr;
|
||||
const testing = std.testing;
|
||||
const expectEqual = testing.expectEqual;
|
||||
const expect = testing.expect;
|
||||
|
||||
comptime {
|
||||
// This is a workaround for https://github.com/ziglang/zig/issues/8218
|
||||
// which is only necessary on macOS.
|
||||
//
|
||||
// Once that issue is fixed, we can undo the changes in
|
||||
// 177cf12e0555147faa4d436e52fc15175c2c4ff0 and go back to passing
|
||||
// -fcompiler-rt in link.rs instead of doing this. Note that this
|
||||
// workaround is present in many host.zig files, so make sure to undo
|
||||
// it everywhere!
|
||||
if (std.builtin.os.tag == .macos) {
|
||||
_ = @import("compiler_rt");
|
||||
}
|
||||
}
|
||||
|
||||
const mem = std.mem;
|
||||
const Allocator = mem.Allocator;
|
||||
|
||||
extern fn roc__mainForHost_1_exposed([*]u8) void;
|
||||
extern fn roc__mainForHost_1_size() i64;
|
||||
extern fn roc__mainForHost_1_Fx_caller(*const u8, *const u8, [*]u8, [*]u8) void;
|
||||
extern fn roc__mainForHost_1_Fx_size() i64;
|
||||
extern fn roc__mainForHost_1_Fx_result_size() i64;
|
||||
|
||||
extern fn malloc(size: usize) callconv(.C) ?*c_void;
|
||||
extern fn realloc(c_ptr: [*]align(@alignOf(u128)) u8, size: usize) callconv(.C) ?*c_void;
|
||||
extern fn free(c_ptr: [*]align(@alignOf(u128)) u8) callconv(.C) void;
|
||||
|
||||
export fn roc_alloc(size: usize, alignment: u32) callconv(.C) ?*c_void {
|
||||
return malloc(size);
|
||||
}
|
||||
|
||||
export fn roc_realloc(c_ptr: *c_void, new_size: usize, old_size: usize, alignment: u32) callconv(.C) ?*c_void {
|
||||
return realloc(@alignCast(16, @ptrCast([*]u8, c_ptr)), new_size);
|
||||
}
|
||||
|
||||
export fn roc_dealloc(c_ptr: *c_void, alignment: u32) callconv(.C) void {
|
||||
free(@alignCast(16, @ptrCast([*]u8, c_ptr)));
|
||||
}
|
||||
|
||||
const Unit = extern struct {};
|
||||
|
||||
pub export fn main() u8 {
|
||||
const stdout = std.io.getStdOut().writer();
|
||||
|
||||
const size = @intCast(usize, roc__mainForHost_1_size());
|
||||
const raw_output = std.heap.c_allocator.alloc(u8, size) catch unreachable;
|
||||
var output = @ptrCast([*]u8, raw_output);
|
||||
|
||||
defer {
|
||||
std.heap.c_allocator.free(raw_output);
|
||||
}
|
||||
|
||||
roc__mainForHost_1_exposed(output);
|
||||
|
||||
const elements = @ptrCast([*]u64, @alignCast(8, output));
|
||||
|
||||
var flag = elements[0];
|
||||
|
||||
if (flag == 0) {
|
||||
// all is well
|
||||
const function_pointer = @intToPtr(*const u8, elements[1]);
|
||||
const closure_data_pointer = @ptrCast([*]u8, output[16..size]);
|
||||
|
||||
call_the_closure(function_pointer, closure_data_pointer);
|
||||
} else {
|
||||
unreachable;
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
fn call_the_closure(function_pointer: *const u8, closure_data_pointer: [*]u8) void {
|
||||
const size = roc__mainForHost_1_Fx_result_size();
|
||||
const raw_output = std.heap.c_allocator.alloc(u8, @intCast(usize, size)) catch unreachable;
|
||||
var output = @ptrCast([*]u8, raw_output);
|
||||
|
||||
defer {
|
||||
std.heap.c_allocator.free(raw_output);
|
||||
}
|
||||
|
||||
const flags: u8 = 0;
|
||||
|
||||
roc__mainForHost_1_Fx_caller(&flags, function_pointer, closure_data_pointer, output);
|
||||
|
||||
const elements = @ptrCast([*]u64, @alignCast(8, output));
|
||||
|
||||
var flag = elements[0];
|
||||
|
||||
if (flag == 0) {
|
||||
return;
|
||||
} else {
|
||||
unreachable;
|
||||
}
|
||||
}
|
||||
|
||||
pub export fn roc_fx_putLine(rocPath: str.RocStr) i64 {
|
||||
const stdout = std.io.getStdOut().writer();
|
||||
|
||||
const u8_ptr = rocPath.asU8ptr();
|
||||
|
||||
var i: usize = 0;
|
||||
while (i < rocPath.len()) {
|
||||
stdout.print("{c}", .{u8_ptr[i]}) catch unreachable;
|
||||
|
||||
i += 1;
|
||||
}
|
||||
|
||||
stdout.print("\n", .{}) catch unreachable;
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
pub const ReadResult = extern struct {
|
||||
bytes: RocStr, // TODO RocList<u8> once Roc supports U8
|
||||
errno: i64, // TODO i32 when Roc supports I32
|
||||
};
|
||||
|
||||
pub const WriteResult = extern struct {
|
||||
errno: i64,
|
||||
};
|
||||
|
||||
pub export fn roc_fx_readAllUtf8(rocPath: RocStr) callconv(.C) ReadResult {
|
||||
var dir = std.fs.cwd();
|
||||
|
||||
var content = dir.readFileAlloc(testing.allocator, rocPath.asSlice(), 1024) catch |e| switch (e) {
|
||||
error.FileNotFound => return .{ .bytes = RocStr.empty(), .errno = 2 },
|
||||
error.IsDir => return .{ .bytes = RocStr.empty(), .errno = 19 },
|
||||
else => return .{ .bytes = RocStr.empty(), .errno = 9999 },
|
||||
};
|
||||
|
||||
var str_ptr = @ptrCast([*]u8, content);
|
||||
var roc_str3 = RocStr.init(str_ptr, content.len);
|
||||
|
||||
return .{ .bytes = roc_str3, .errno = 0 };
|
||||
}
|
||||
|
||||
pub export fn roc_fx_writeAllUtf8(filePath: RocStr, content: RocStr) callconv(.C) WriteResult {
|
||||
var dir = std.fs.cwd();
|
||||
|
||||
dir.writeFile(filePath.asSlice(), content.asSlice()) catch |e| switch (e) {
|
||||
else => return .{ .errno = 1 },
|
||||
};
|
||||
|
||||
return .{ .errno = 0 };
|
||||
}
|
||||
|
||||
pub fn roc_fx_readAllUtf8_that_does_not_work(rocPath: *RocStr) ReadResult {
|
||||
const allocator = std.heap.c_allocator;
|
||||
|
||||
// fopen wants a C string, so stack-allocate one using rocPath's contents
|
||||
const len = rocPath.len() + 1;
|
||||
|
||||
var raw = allocator.alloc(u8, len) catch unreachable;
|
||||
var path: [*:0]u8 = @ptrCast([*:0]u8, raw);
|
||||
rocPath.memcpy(path, len);
|
||||
path[len] = 0; // nul-terminate the path, since it's a C string
|
||||
|
||||
// Open the file
|
||||
const file = fopen(path, "r") orelse {
|
||||
return ReadResult{ .bytes = RocStr.empty(), .errno = errno };
|
||||
};
|
||||
|
||||
// Now that the file has been opened, make sure we always (try to) close it
|
||||
// before returning, even if something went wrong while reading it.
|
||||
defer {
|
||||
if (fclose(file) != 0) {
|
||||
return ReadResult{ .bytes = RocStr.empty(), .errno = errno };
|
||||
}
|
||||
}
|
||||
|
||||
// Next we'll count the total number of bytes in the file, which we need
|
||||
// to know so we can allocate a correctly-sized buffer to read into.
|
||||
|
||||
// First, seek to the end of the file
|
||||
if (fseek(file, 0, SEEK_END) != 0) {
|
||||
return ReadResult{ .bytes = RocStr.empty(), .errno = errno };
|
||||
}
|
||||
|
||||
// Now the current file position (which ftell returns) will be the end of
|
||||
// the file - which will be equal to the total number of bytes in the file.
|
||||
const totalBytes: c_long = ftell(file);
|
||||
|
||||
// In the highly unusal case that there are no bytes to read, return early.
|
||||
if (totalBytes <= 0) {
|
||||
// If the file was empty, return an empty list.
|
||||
if (totalBytes == 0) {
|
||||
return ReadResult{ .bytes = RocStr.empty(), .errno = 0 };
|
||||
}
|
||||
|
||||
// ftell returns -1 on error, so return an error here
|
||||
return ReadResult{ .bytes = RocStr.empty(), .errno = errno };
|
||||
}
|
||||
|
||||
// Rewind to the beginning of the file, so we can start actually reading.
|
||||
if (fseek(file, 0, SEEK_SET) != 0) {
|
||||
return ReadResult{ .bytes = RocStr.empty(), .errno = errno };
|
||||
}
|
||||
|
||||
// Allocate enough bytes for the contents of the file, plus the refcount.
|
||||
const refcountBytes = @sizeOf(usize);
|
||||
var buffer: [*]u8 = malloc(totalBytes + refcountBytes) orelse {
|
||||
// If allocation failed, throw a runtime exception for Roc to catch.
|
||||
|
||||
// fclose the file explicitly before throwing, because libunwind
|
||||
// will disregard our defer block. (TODO verify this!)
|
||||
//
|
||||
// Silently ignore fclose errors here, because we're about to throw an
|
||||
// allocation failure exception; fclose failures won't affect that.
|
||||
fclose(file);
|
||||
|
||||
// TODO use libunwind to throw an exception here
|
||||
// TODO set an "allocation failed" exception object for `catch` to receive
|
||||
// TODO write a test for this which simulates allocation failure
|
||||
};
|
||||
|
||||
// Initialize the refcount to a positive number - meaning it's actually
|
||||
// a capacity value, which is appropriate since we return a Unique value.
|
||||
@ptrCast(buffer, [*]usize)[0] = totalBytes;
|
||||
|
||||
// The buffer pointer should point to the first byte *after* the refcount
|
||||
buffer += refcountBytes;
|
||||
|
||||
// Read the bytes into the buffer.
|
||||
const bytesRead = fread(buffer, 1, totalBytes, file);
|
||||
|
||||
// fread indicates an error by returning a number that's different from
|
||||
// the number of elements we requested to read
|
||||
if (bytesRead != totalBytes) {
|
||||
return ReadResult{ .bytes = RocStr.empty(), .errno = errno };
|
||||
}
|
||||
|
||||
// Explicitly return errno = 0 to indicate there was no error.
|
||||
//
|
||||
// (We don't want to read from the errno global here because it might have
|
||||
// a nonzero value leftover from previous unrelated operations.)
|
||||
return ReadResult{ .bytes = RocStr.init(buffer, totalBytes), .errno = 0 };
|
||||
}
|
||||
|
||||
// const c = @cImport({
|
||||
// @cInclude("stdio.h");
|
||||
// @cInclude("stdlib.h");
|
||||
// });
|
||||
//
|
||||
// extern var errno: c_int;
|
||||
//
|
||||
// const FILE = extern struct {
|
||||
// unused: u8,
|
||||
// };
|
||||
|
||||
// extern "c" fn fopen(filename: [*:0]const u8, modes: [*:0]const u8) ?*FILE;
|
||||
//extern "c" fn fopen(filename: [*:0]const u8, modes: [*:0]const u8) ?*FILE;
|
||||
//extern "c" fn fclose(stream: *FILE) c_int;
|
||||
//extern "c" fn fseek(stream: *FILE, offset: c_long, origin: c_int) c_int;
|
||||
|
||||
// extern fn fopen([*:0]const u8, [*:0]const u8) ?*FILE;
|
||||
// extern fn fseek(*FILE, c_long, c_int) c_int;
|
||||
|
||||
//extern fn fopen([*c]const u8, [*c]const u8) [*c]FILE;
|
||||
// extern fn ftell([*c]FILE) c_long;
|
||||
// extern fn fread([*c]u8, size_t, size_t, [*c]FILE) size_t;
|
||||
// extern fn fclose([*c]FILE) c_int;
|
Loading…
Add table
Add a link
Reference in a new issue