From: Markus Armbruster <armbru@redhat.com>
To: qemu-devel@nongnu.org
Cc: peter.maydell@linaro.org, John Snow <jsnow@redhat.com>
Subject: [PULL 01/16] qapi/commands: assert arg_type is not None
Date: Mon, 8 Feb 2021 14:58:31 +0100 [thread overview]
Message-ID: <20210208135846.3707170-2-armbru@redhat.com> (raw)
In-Reply-To: <20210208135846.3707170-1-armbru@redhat.com>
From: John Snow <jsnow@redhat.com>
When boxed is True, expr.py asserts that we must have
arguments. Ultimately, this should mean that if boxed is True that
arg_type should be defined. Mypy cannot infer this, and does not support
'stateful' type inference, e.g.:
```
if x:
assert y is not None
...
if x:
y.etc()
```
does not work, because mypy does not statefully remember the conditional
assertion in the second block. Help mypy out by creating a new local
that it can track more easily.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-Id: <20210201193747.2169670-2-jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
---
scripts/qapi/commands.py | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/scripts/qapi/commands.py b/scripts/qapi/commands.py
index 50978090b4..71744f48a3 100644
--- a/scripts/qapi/commands.py
+++ b/scripts/qapi/commands.py
@@ -126,6 +126,9 @@ def gen_marshal(name: str,
boxed: bool,
ret_type: Optional[QAPISchemaType]) -> str:
have_args = boxed or (arg_type and not arg_type.is_empty())
+ if have_args:
+ assert arg_type is not None
+ arg_type_c_name = arg_type.c_name()
ret = mcgen('''
@@ -147,7 +150,7 @@ def gen_marshal(name: str,
ret += mcgen('''
%(c_name)s arg = {0};
''',
- c_name=arg_type.c_name())
+ c_name=arg_type_c_name)
ret += mcgen('''
@@ -163,7 +166,7 @@ def gen_marshal(name: str,
ok = visit_check_struct(v, errp);
}
''',
- c_arg_type=arg_type.c_name())
+ c_arg_type=arg_type_c_name)
else:
ret += mcgen('''
ok = visit_check_struct(v, errp);
@@ -193,7 +196,7 @@ out:
ret += mcgen('''
visit_type_%(c_arg_type)s_members(v, &arg, NULL);
''',
- c_arg_type=arg_type.c_name())
+ c_arg_type=arg_type_c_name)
ret += mcgen('''
visit_end_struct(v, NULL);
--
2.26.2
next prev parent reply other threads:[~2021-02-08 20:44 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-08 13:58 [PULL 00/16] QAPI patches patches for 2021-02-08 Markus Armbruster
2021-02-08 13:58 ` Markus Armbruster [this message]
2021-02-08 13:58 ` [PULL 02/16] qapi/events: fix visit_event typing Markus Armbruster
2021-02-08 13:58 ` [PULL 03/16] qapi/main: handle theoretical None-return from re.match() Markus Armbruster
2021-02-08 13:58 ` [PULL 04/16] qapi/gen: inline _wrap_ifcond into end_if() Markus Armbruster
2021-02-08 13:58 ` [PULL 05/16] qapi: centralize is_[user|system|builtin]_module methods Markus Armbruster
2021-02-08 13:58 ` [PULL 06/16] qapi/gen: Replace ._begin_system_module() Markus Armbruster
2021-02-08 13:58 ` [PULL 07/16] qapi: use explicitly internal module names Markus Armbruster
2021-02-08 13:58 ` [PULL 08/16] qapi: use './builtin' as the built-in module name Markus Armbruster
2021-02-08 13:58 ` [PULL 09/16] qapi/gen: Combine ._add_[user|system]_module Markus Armbruster
2021-02-08 13:58 ` [PULL 10/16] qapi: centralize the built-in module name definition Markus Armbruster
2021-02-08 13:58 ` [PULL 11/16] qapi/gen: write _genc/_genh access shims Markus Armbruster
2021-02-08 13:58 ` [PULL 12/16] qapi/gen: Support switching to another module temporarily Markus Armbruster
2021-02-08 13:58 ` [PULL 13/16] qapi/commands: Simplify command registry generation Markus Armbruster
2021-02-08 13:58 ` [PULL 14/16] qapi/gen: Drop support for QAPIGen without a file name Markus Armbruster
2021-02-08 13:58 ` [PULL 15/16] qapi: type 'info' as Optional[QAPISourceInfo] Markus Armbruster
2021-02-08 13:58 ` [PULL 16/16] qapi: enable strict-optional checks Markus Armbruster
2021-02-08 17:32 ` [PULL 00/16] QAPI patches patches for 2021-02-08 Peter Maydell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210208135846.3707170-2-armbru@redhat.com \
--to=armbru@redhat.com \
--cc=jsnow@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).