when handling a returned error, if the expression/block does not try/return error, reset error return trace index to 0 #1923
Labels
Milestone
Comments
andrewrk
added
the
bug
|
I think there are 2 components to this issue:
|
|
|
|
That's correct - and intentional. The example in the docs here shows how this is useful: https://ziglang.org/documentation/master/#Error-Return-Traces If we wanted to avoid giving this up, what we could do is "reset" the error return trace when a function returns a non-error, by setting the index to 0. This would solve the problem, but require an extra memory store for every |
|
I think that would be insufficient, in my example as the executable is never found, it would still keep all of Lines 613 to 616 in 7843c96 How about this : in a |
I like this idea a lot! And I think it will be straightforward to implement too. 💯 from me |
|
BTW I believe this is a duplicate of #1810 |
|
Closing as a duplicate of #1810 but with the note that @Sahnvour's proposed fix in #1923 (comment) is planned. |
|
I don't know why I thought this was a duplicate of #1810. The issues are entirely unrelated. |
andrewrk
changed the title
andrewrk
added
proposal
|
I'm implementing this (have working code that resets the index), but it's not clear to me where to put that code in the catch/else block. There are a few options:
loop.init() catch |err| {
std.log.err("{}", .{@errorName(err)});
if (@errorReturnTrace()) |trace| {
std.debug.dumpStackTrace(trace.*);
}
return 1;
};To solve that we might want to check for a
I'm not sure what would be the best solution. |
|
I think we could emit a single control flow block to do the reset to 0 (it's only a constant write to a static variable right ?), and have all returns and such jump to it. It shouldn't need any stack or registers so that's fine ? However I don't know if that is possible as is with LLVM. If not, I would go for 3. and put a redundant block before any return/break/end of zig block. The optimizer should be able to get away with it, and at least that is a simple implementation we can still reevaluate in the future if need be. |
kristoff-it
pushed a commit
that referenced
this issue
Aug 23, 2022
This reverts commit 1a32f2a. Sorry, this workaround is not welcome. Instead, please solve the actual issue by doing the accepted behavior in the compiler itself: > in a catch or else (handling a returned error), if the block does not > try or return error.xyz, set the index to 0 This also applies to if statements, such as the one that test runner is doing just above this hack.
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 14, 2022
This implement trace "popping" for correctly handled errors within
`catch { ... }` and `else { ... }` blocks.
When breaking from these blocks with any non-error, we pop the error
trace frames corresponding to the operand. When breaking with an error,
we preserve the frames so that error traces "chain" together as usual.
```zig
fn foo(cond1: bool, cond2: bool) !void {
bar() catch {
if (cond1) {
// If baz() result is a non-error, pop the error trace frames from bar()
// If baz() result is an error, leave the bar() frames on the error trace
return baz();
} else if (cond2) {
// If we break/return an error, then leave the error frames from bar() on the error trace
return error.Foo;
}
};
// An error returned from here does not include bar()'s error frames in the trace
return error.Bar;
}
```
Notice that if foo() does not return an error it, it leaves no extra
frames on the error trace.
This is piece (1/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 14, 2022
This allows for errors to be "re-thrown" by yielding any error as the
result of a catch block. For example:
```zig
fn errorable() !void {
return error.FallingOutOfPlane;
}
fn foo(have_parachute: bool) !void {
return errorable() catch |err| b: {
if (have_parachute) {
// error trace will include the call to errorable()
break :b error.NoParachute;
} else {
return;
}
};
}
pub fn main() !void {
// Anything that returns a non-error does not pollute the error trace.
try foo(true);
// This error trace will still include errorable(), whose error was "re-thrown" by foo()
try foo(false);
}
```
This is piece (2/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 14, 2022
In order to enforce a strict stack discipline for error return traces,
we cannot track error return traces that are stored in variables:
```zig
const x = errorable(); // errorable()'s error return trace is killed here
// v-- error trace starts here instead
return x catch error.UnknownError;
```
In order to propagate error return traces, function calls need to be passed
directly to an error-handling expression (`if`, `catch`, `try` or `return`):
```zig
// When passed directly to `catch`, the return trace is propagated
return errorable() catch error.UnknownError;
// Using a break also works
return blk: {
// code here
break :blk errorable();
} catch error.UnknownError;
```
Why do we need this restriction? Without it, multiple errors can co-exist
with their own error traces. Handling that situation correctly means either:
a. Dynamically allocating trace memory and tracking lifetimes, OR
b. Allowing the production of one error to interfere with the trace of another
(which is the current status quo)
This is piece (3/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 14, 2022
This allows for errors to be "re-thrown" by yielding any error as the
result of a catch block. For example:
```zig
fn errorable() !void {
return error.FallingOutOfPlane;
}
fn foo(have_parachute: bool) !void {
return errorable() catch |err| b: {
if (have_parachute) {
// error trace will include the call to errorable()
break :b error.NoParachute;
} else {
return;
}
};
}
pub fn main() !void {
// Anything that returns a non-error does not pollute the error trace.
try foo(true);
// This error trace will still include errorable(), whose error was "re-thrown" by foo()
try foo(false);
}
```
This is piece (2/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 14, 2022
In order to enforce a strict stack discipline for error return traces,
we cannot track error return traces that are stored in variables:
```zig
const x = errorable(); // errorable()'s error return trace is killed here
// v-- error trace starts here instead
return x catch error.UnknownError;
```
In order to propagate error return traces, function calls need to be passed
directly to an error-handling expression (`if`, `catch`, `try` or `return`):
```zig
// When passed directly to `catch`, the return trace is propagated
return errorable() catch error.UnknownError;
// Using a break also works
return blk: {
// code here
break :blk errorable();
} catch error.UnknownError;
```
Why do we need this restriction? Without it, multiple errors can co-exist
with their own error traces. Handling that situation correctly means either:
a. Dynamically allocating trace memory and tracking lifetimes, OR
b. Allowing the production of one error to interfere with the trace of another
(which is the current status quo)
This is piece (3/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 14, 2022
This allows for errors to be "re-thrown" by yielding any error as the
result of a catch block. For example:
```zig
fn errorable() !void {
return error.FallingOutOfPlane;
}
fn foo(have_parachute: bool) !void {
return errorable() catch |err| b: {
if (have_parachute) {
// error trace will include the call to errorable()
break :b error.NoParachute;
} else {
return;
}
};
}
pub fn main() !void {
// Anything that returns a non-error does not pollute the error trace.
try foo(true);
// This error trace will still include errorable(), whose error was "re-thrown" by foo()
try foo(false);
}
```
This is piece (2/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 14, 2022
In order to enforce a strict stack discipline for error return traces,
we cannot track error return traces that are stored in variables:
```zig
const x = errorable(); // errorable()'s error return trace is killed here
// v-- error trace starts here instead
return x catch error.UnknownError;
```
In order to propagate error return traces, function calls need to be passed
directly to an error-handling expression (`if`, `catch`, `try` or `return`):
```zig
// When passed directly to `catch`, the return trace is propagated
return errorable() catch error.UnknownError;
// Using a break also works
return blk: {
// code here
break :blk errorable();
} catch error.UnknownError;
```
Why do we need this restriction? Without it, multiple errors can co-exist
with their own error traces. Handling that situation correctly means either:
a. Dynamically allocating trace memory and tracking lifetimes, OR
b. Allowing the production of one error to interfere with the trace of another
(which is the current status quo)
This is piece (3/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 23, 2022
This implement trace "popping" for correctly handled errors within
`catch { ... }` and `else { ... }` blocks.
When breaking from these blocks with any non-error, we pop the error
trace frames corresponding to the operand. When breaking with an error,
we preserve the frames so that error traces "chain" together as usual.
```zig
fn foo(cond1: bool, cond2: bool) !void {
bar() catch {
if (cond1) {
// If baz() result is a non-error, pop the error trace frames from bar()
// If baz() result is an error, leave the bar() frames on the error trace
return baz();
} else if (cond2) {
// If we break/return an error, then leave the error frames from bar() on the error trace
return error.Foo;
}
};
// An error returned from here does not include bar()'s error frames in the trace
return error.Bar;
}
```
Notice that if foo() does not return an error it, it leaves no extra
frames on the error trace.
This is piece (1/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 23, 2022
This allows for errors to be "re-thrown" by yielding any error as the
result of a catch block. For example:
```zig
fn errorable() !void {
return error.FallingOutOfPlane;
}
fn foo(have_parachute: bool) !void {
return errorable() catch |err| b: {
if (have_parachute) {
// error trace will include the call to errorable()
break :b error.NoParachute;
} else {
return;
}
};
}
pub fn main() !void {
// Anything that returns a non-error does not pollute the error trace.
try foo(true);
// This error trace will still include errorable(), whose error was "re-thrown" by foo()
try foo(false);
}
```
This is piece (2/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 23, 2022
In order to enforce a strict stack discipline for error return traces,
we cannot track error return traces that are stored in variables:
```zig
const x = errorable(); // errorable()'s error return trace is killed here
// v-- error trace starts here instead
return x catch error.UnknownError;
```
In order to propagate error return traces, function calls need to be passed
directly to an error-handling expression (`if`, `catch`, `try` or `return`):
```zig
// When passed directly to `catch`, the return trace is propagated
return errorable() catch error.UnknownError;
// Using a break also works
return blk: {
// code here
break :blk errorable();
} catch error.UnknownError;
```
Why do we need this restriction? Without it, multiple errors can co-exist
with their own error traces. Handling that situation correctly means either:
a. Dynamically allocating trace memory and tracking lifetimes, OR
b. Allowing the production of one error to interfere with the trace of another
(which is the current status quo)
This is piece (3/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 26, 2022
This implement trace "popping" for correctly handled errors within
`catch { ... }` and `else { ... }` blocks.
When breaking from these blocks with any non-error, we pop the error
trace frames corresponding to the operand. When breaking with an error,
we preserve the frames so that error traces "chain" together as usual.
```zig
fn foo(cond1: bool, cond2: bool) !void {
bar() catch {
if (cond1) {
// If baz() result is a non-error, pop the error trace frames from bar()
// If baz() result is an error, leave the bar() frames on the error trace
return baz();
} else if (cond2) {
// If we break/return an error, then leave the error frames from bar() on the error trace
return error.Foo;
}
};
// An error returned from here does not include bar()'s error frames in the trace
return error.Bar;
}
```
Notice that if foo() does not return an error it, it leaves no extra
frames on the error trace.
This is piece (1/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 26, 2022
This allows for errors to be "re-thrown" by yielding any error as the
result of a catch block. For example:
```zig
fn errorable() !void {
return error.FallingOutOfPlane;
}
fn foo(have_parachute: bool) !void {
return errorable() catch |err| b: {
if (have_parachute) {
// error trace will include the call to errorable()
break :b error.NoParachute;
} else {
return;
}
};
}
pub fn main() !void {
// Anything that returns a non-error does not pollute the error trace.
try foo(true);
// This error trace will still include errorable(), whose error was "re-thrown" by foo()
try foo(false);
}
```
This is piece (2/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Sep 26, 2022
In order to enforce a strict stack discipline for error return traces,
we cannot track error return traces that are stored in variables:
```zig
const x = errorable(); // errorable()'s error return trace is killed here
// v-- error trace starts here instead
return x catch error.UnknownError;
```
In order to propagate error return traces, function calls need to be passed
directly to an error-handling expression (`if`, `catch`, `try` or `return`):
```zig
// When passed directly to `catch`, the return trace is propagated
return errorable() catch error.UnknownError;
// Using a break also works
return blk: {
// code here
break :blk errorable();
} catch error.UnknownError;
```
Why do we need this restriction? Without it, multiple errors can co-exist
with their own error traces. Handling that situation correctly means either:
a. Dynamically allocating trace memory and tracking lifetimes, OR
b. Allowing the production of one error to interfere with the trace of another
(which is the current status quo)
This is piece (3/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 15, 2022
This implement trace "popping" for correctly handled errors within
`catch { ... }` and `else { ... }` blocks.
When breaking from these blocks with any non-error, we pop the error
trace frames corresponding to the operand. When breaking with an error,
we preserve the frames so that error traces "chain" together as usual.
```zig
fn foo(cond1: bool, cond2: bool) !void {
bar() catch {
if (cond1) {
// If baz() result is a non-error, pop the error trace frames from bar()
// If baz() result is an error, leave the bar() frames on the error trace
return baz();
} else if (cond2) {
// If we break/return an error, then leave the error frames from bar() on the error trace
return error.Foo;
}
};
// An error returned from here does not include bar()'s error frames in the trace
return error.Bar;
}
```
Notice that if foo() does not return an error it, it leaves no extra
frames on the error trace.
This is piece (1/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 15, 2022
This allows for errors to be "re-thrown" by yielding any error as the
result of a catch block. For example:
```zig
fn errorable() !void {
return error.FallingOutOfPlane;
}
fn foo(have_parachute: bool) !void {
return errorable() catch |err| b: {
if (have_parachute) {
// error trace will include the call to errorable()
break :b error.NoParachute;
} else {
return;
}
};
}
pub fn main() !void {
// Anything that returns a non-error does not pollute the error trace.
try foo(true);
// This error trace will still include errorable(), whose error was "re-thrown" by foo()
try foo(false);
}
```
This is piece (2/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 15, 2022
In order to enforce a strict stack discipline for error return traces,
we cannot track error return traces that are stored in variables:
```zig
const x = errorable(); // errorable()'s error return trace is killed here
// v-- error trace starts here instead
return x catch error.UnknownError;
```
In order to propagate error return traces, function calls need to be passed
directly to an error-handling expression (`if`, `catch`, `try` or `return`):
```zig
// When passed directly to `catch`, the return trace is propagated
return errorable() catch error.UnknownError;
// Using a break also works
return blk: {
// code here
break :blk errorable();
} catch error.UnknownError;
```
Why do we need this restriction? Without it, multiple errors can co-exist
with their own error traces. Handling that situation correctly means either:
a. Dynamically allocating trace memory and tracking lifetimes, OR
b. Allowing the production of one error to interfere with the trace of another
(which is the current status quo)
This is piece (3/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 20, 2022
This implement trace "popping" for correctly handled errors within
`catch { ... }` and `else { ... }` blocks.
When breaking from these blocks with any non-error, we pop the error
trace frames corresponding to the operand. When breaking with an error,
we preserve the frames so that error traces "chain" together as usual.
```zig
fn foo(cond1: bool, cond2: bool) !void {
bar() catch {
if (cond1) {
// If baz() result is a non-error, pop the error trace frames from bar()
// If baz() result is an error, leave the bar() frames on the error trace
return baz();
} else if (cond2) {
// If we break/return an error, then leave the error frames from bar() on the error trace
return error.Foo;
}
};
// An error returned from here does not include bar()'s error frames in the trace
return error.Bar;
}
```
Notice that if foo() does not return an error it, it leaves no extra
frames on the error trace.
This is piece (1/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 20, 2022
This allows for errors to be "re-thrown" by yielding any error as the
result of a catch block. For example:
```zig
fn errorable() !void {
return error.FallingOutOfPlane;
}
fn foo(have_parachute: bool) !void {
return errorable() catch |err| b: {
if (have_parachute) {
// error trace will include the call to errorable()
break :b error.NoParachute;
} else {
return;
}
};
}
pub fn main() !void {
// Anything that returns a non-error does not pollute the error trace.
try foo(true);
// This error trace will still include errorable(), whose error was "re-thrown" by foo()
try foo(false);
}
```
This is piece (2/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 20, 2022
In order to enforce a strict stack discipline for error return traces,
we cannot track error return traces that are stored in variables:
```zig
const x = errorable(); // errorable()'s error return trace is killed here
// v-- error trace starts here instead
return x catch error.UnknownError;
```
In order to propagate error return traces, function calls need to be passed
directly to an error-handling expression (`if`, `catch`, `try` or `return`):
```zig
// When passed directly to `catch`, the return trace is propagated
return errorable() catch error.UnknownError;
// Using a break also works
return blk: {
// code here
break :blk errorable();
} catch error.UnknownError;
```
Why do we need this restriction? Without it, multiple errors can co-exist
with their own error traces. Handling that situation correctly means either:
a. Dynamically allocating trace memory and tracking lifetimes, OR
b. Allowing the production of one error to interfere with the trace of another
(which is the current status quo)
This is piece (3/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 21, 2022
This implement trace "popping" for correctly handled errors within
`catch { ... }` and `else { ... }` blocks.
When breaking from these blocks with any non-error, we pop the error
trace frames corresponding to the operand. When breaking with an error,
we preserve the frames so that error traces "chain" together as usual.
```zig
fn foo(cond1: bool, cond2: bool) !void {
bar() catch {
if (cond1) {
// If baz() result is a non-error, pop the error trace frames from bar()
// If baz() result is an error, leave the bar() frames on the error trace
return baz();
} else if (cond2) {
// If we break/return an error, then leave the error frames from bar() on the error trace
return error.Foo;
}
};
// An error returned from here does not include bar()'s error frames in the trace
return error.Bar;
}
```
Notice that if foo() does not return an error it, it leaves no extra
frames on the error trace.
This is piece (1/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 21, 2022
This allows for errors to be "re-thrown" by yielding any error as the
result of a catch block. For example:
```zig
fn errorable() !void {
return error.FallingOutOfPlane;
}
fn foo(have_parachute: bool) !void {
return errorable() catch |err| b: {
if (have_parachute) {
// error trace will include the call to errorable()
break :b error.NoParachute;
} else {
return;
}
};
}
pub fn main() !void {
// Anything that returns a non-error does not pollute the error trace.
try foo(true);
// This error trace will still include errorable(), whose error was "re-thrown" by foo()
try foo(false);
}
```
This is piece (2/3) of ziglang#1923 (comment)
topolarity
added a commit
to topolarity/zig
that referenced
this issue
Oct 21, 2022
In order to enforce a strict stack discipline for error return traces,
we cannot track error return traces that are stored in variables:
```zig
const x = errorable(); // errorable()'s error return trace is killed here
// v-- error trace starts here instead
return x catch error.UnknownError;
```
In order to propagate error return traces, function calls need to be passed
directly to an error-handling expression (`if`, `catch`, `try` or `return`):
```zig
// When passed directly to `catch`, the return trace is propagated
return errorable() catch error.UnknownError;
// Using a break also works
return blk: {
// code here
break :blk errorable();
} catch error.UnknownError;
```
Why do we need this restriction? Without it, multiple errors can co-exist
with their own error traces. Handling that situation correctly means either:
a. Dynamically allocating trace memory and tracking lifetimes, OR
b. Allowing the production of one error to interfere with the trace of another
(which is the current status quo)
This is piece (3/3) of ziglang#1923 (comment)
andrewrk
added a commit
that referenced
this issue
Oct 22, 2022
…1923 stage2: Pop error trace frames for handled errors (#1923)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
When trying to run the first example code from #1813 with an executable that can't be found, the error trace is weird. It looks like the return address array is filled with addresses from
windowsCreateProcess.I believe this is due to the loop in
spawnWindowsthat tries every directory fromPATHand returns a lot of errors because the file doesn't exist.As for why it is sometimes
PATH_NOT_FOUNDand othersFILE_NOT_FOUND, I assume myPATHcontains some old directories that don't exist anymore.Example output:
The text was updated successfully, but these errors were encountered: