From: John Snow <jsnow@redhat.com>
To: Markus Armbruster <armbru@redhat.com>
Cc: "Marc-André Lureau" <marcandre.lureau@redhat.com>,
"Cleber Rosa" <crosa@redhat.com>,
qemu-devel@nongnu.org, "Eduardo Habkost" <ehabkost@redhat.com>
Subject: Re: [PATCH v3 10/17] qapi/gen: Combine ._add_[user|system]_module
Date: Wed, 20 Jan 2021 11:10:52 -0500 [thread overview]
Message-ID: <941acfa1-940a-b264-7da3-6df4e08661c2@redhat.com> (raw)
In-Reply-To: <87y2gnpu4m.fsf@dusky.pond.sub.org>
On 1/20/21 9:20 AM, Markus Armbruster wrote:
> John Snow <jsnow@redhat.com> writes:
>
>> From: Markus Armbruster <armbru@redhat.com>
>>
>> QAPISchemaModularCVisitor attempts to encapsulate the way it splits
>> the module name space between user modules (name can't start with
>> './') and system modules (name is None or starts with './') by
>
> Is this still accurate?
>
Ah, not exactly; sorry I mangled your patches here. I had written my own
version of what this patch used to be, but your patch went one step
further, but the ordering got weird.
>> providing separate ._add_user_module() and ._add_system_module(),
>> where the latter prepends './' to names other than None.
>>
>> Not worthwhile. Dumb down to a single ._add_module().
>>
>> Signed-off-by: Markus Armbruster <armbru@redhat.com>
>> Signed-off-by: John Snow <jsnow@redhat.com>
>> ---
>> scripts/qapi/commands.py | 2 +-
>> scripts/qapi/events.py | 2 +-
>> scripts/qapi/gen.py | 20 +++++++-------------
>> 3 files changed, 9 insertions(+), 15 deletions(-)
>>
>> diff --git a/scripts/qapi/commands.py b/scripts/qapi/commands.py
>> index fc5fe27c472..49111663394 100644
>> --- a/scripts/qapi/commands.py
>> +++ b/scripts/qapi/commands.py
>> @@ -286,7 +286,7 @@ def _begin_user_module(self, name: str) -> None:
>> types=types))
>>
>> def visit_end(self) -> None:
>> - self._add_system_module('./init', ' * QAPI Commands initialization')
>> + self._add_module('./init', ' * QAPI Commands initialization')
>> self._genh.add(mcgen('''
>> #include "qapi/qmp/dispatch.h"
>>
>> diff --git a/scripts/qapi/events.py b/scripts/qapi/events.py
>> index 26faa829898..079c666ec69 100644
>> --- a/scripts/qapi/events.py
>> +++ b/scripts/qapi/events.py
>> @@ -191,7 +191,7 @@ def _begin_user_module(self, name: str) -> None:
>> types=types))
>>
>> def visit_end(self) -> None:
>> - self._add_system_module('./emit', ' * QAPI Events emission')
>> + self._add_module('./emit', ' * QAPI Events emission')
>> self._genc.preamble_add(mcgen('''
>> #include "qemu/osdep.h"
>> #include "%(prefix)sqapi-emit-events.h"
>> diff --git a/scripts/qapi/gen.py b/scripts/qapi/gen.py
>> index 55acd7e080d..b5505685e6e 100644
>> --- a/scripts/qapi/gen.py
>> +++ b/scripts/qapi/gen.py
>> @@ -272,22 +272,15 @@ def _module_filename(self, what: str, name: str) -> str:
>> self._module_basename(what, name))
>>
>> def _add_module(self, name: str, blurb: str) -> None:
>> + if QAPISchemaModule.is_user_module(name):
>> + if self._main_module is None:
>> + self._main_module = name
>> basename = self._module_filename(self._what, name)
>> genc = QAPIGenC(basename + '.c', blurb, self._pydoc)
>> genh = QAPIGenH(basename + '.h', blurb, self._pydoc)
>> self._module[name] = (genc, genh)
>> self._genc, self._genh = self._module[name]
>>
>> - def _add_user_module(self, name: str, blurb: str) -> None:
>> - assert QAPISchemaModule.is_user_module(name)
>> - if self._main_module is None:
>> - self._main_module = name
>> - self._add_module(name, blurb)
>> -
>> - def _add_system_module(self, name: str, blurb: str) -> None:
>> - assert QAPISchemaModule.is_system_module(name)
>> - self._add_module(name, blurb)
>> -
>> def write(self, output_dir: str, opt_builtins: bool = False) -> None:
>> for name in self._module:
>> if QAPISchemaModule.is_builtin_module(name) and not opt_builtins:
>> @@ -303,9 +296,9 @@ def _begin_user_module(self, name: str) -> None:
>> pass
>>
>> def visit_module(self, module: QAPISchemaModule) -> None:
>> - if module.system_module:
>> + if module.builtin_module:
>
> Looks like you're fixing a slip-up in PATCH 06. If yes, squash. If no,
> I'm confused.
>
Or patch 7, actually -- where we clarify that we are checking for the
built-in module and not any old system module.
>> if self._builtin_blurb:
>> - self._add_system_module(module.name, self._builtin_blurb)
>> + self._add_module(module.name, self._builtin_blurb)
>> self._begin_builtin_module()
>> else:
>> # The built-in module has not been created. No code may
>> @@ -313,7 +306,8 @@ def visit_module(self, module: QAPISchemaModule) -> None:
>> self._genc = None
>> self._genh = None
>> else:
>> - self._add_user_module(module.name, self._user_blurb)
>> + assert module.user_module, "Unexpected system module"
>
> The string provides no value.
>
That's just, like, your opinion, man. It has value to me.
Compare:
```
#!/usr/bin/env python3
def main():
assert False
if __name__ == '__main__':
main()
```
```
# ./foo.py
Traceback (most recent call last):
File "./foo.py", line 7, in <module>
main()
File "./foo.py", line 4, in main
assert False
AssertionError
```
With:
```
#!/usr/bin/env python3
def main():
assert False, "Unexpected system module"
if __name__ == '__main__':
main()
```
```
# ./foo.py
Traceback (most recent call last):
File "./foo.py", line 7, in <module>
main()
File "./foo.py", line 4, in main
assert False, "Unexpected system module"
AssertionError: Unexpected system module
```
I like the extra string for giving some semantic context as to
specifically what broke (We don't expect to see system modules here)
instead of just a stack trace.
>> + self._add_module(module.name, self._user_blurb)
>> self._begin_user_module(module.name)
>>
>> def visit_include(self, name: str, info: QAPISourceInfo) -> None:
next prev parent reply other threads:[~2021-01-20 16:19 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-19 18:02 [PATCH v3 00/17] qapi: static typing conversion, pt1.5 John Snow
2021-01-19 18:02 ` [PATCH v3 01/17] qapi/commands: assert arg_type is not None John Snow
2021-01-19 18:02 ` [PATCH v3 02/17] qapi/events: fix visit_event typing John Snow
2021-01-19 18:02 ` [PATCH v3 03/17] qapi/main: handle theoretical None-return from re.match() John Snow
2021-01-19 18:02 ` [PATCH v3 04/17] qapi/gen: inline _wrap_ifcond into end_if() John Snow
2021-01-19 18:02 ` [PATCH v3 05/17] qapi: pass QAPISchemaModule to visit_module instead of str John Snow
2021-01-20 12:07 ` Markus Armbruster
2021-01-20 15:51 ` John Snow
2021-01-21 7:22 ` Markus Armbruster
2021-01-20 16:02 ` Eric Blake
2021-01-20 16:16 ` John Snow
2021-01-21 7:23 ` Markus Armbruster
2021-01-19 18:02 ` [PATCH v3 06/17] qapi: centralize is_[user|system|builtin]_module methods John Snow
2021-01-20 13:56 ` Markus Armbruster
2021-01-19 18:02 ` [PATCH v3 07/17] qapi/gen: Replace ._begin_system_module() John Snow
2021-01-19 18:02 ` [PATCH v3 08/17] qapi: use explicitly internal module names John Snow
2021-01-19 18:02 ` [PATCH v3 09/17] qapi: use './builtin' as the built-in module name John Snow
2021-01-20 14:14 ` Markus Armbruster
2021-01-19 18:02 ` [PATCH v3 10/17] qapi/gen: Combine ._add_[user|system]_module John Snow
2021-01-20 14:20 ` Markus Armbruster
2021-01-20 16:10 ` John Snow [this message]
2021-01-21 7:40 ` Markus Armbruster
2021-01-19 18:02 ` [PATCH v3 11/17] qapi: centralize the built-in module name definition John Snow
2021-01-19 18:02 ` [PATCH v3 12/17] qapi/gen: write _genc/_genh access shims John Snow
2021-01-19 18:02 ` [PATCH v3 13/17] qapi/gen: Support for switching to another module temporarily John Snow
2021-01-19 18:02 ` [PATCH v3 14/17] qapi/commands: Simplify command registry generation John Snow
2021-01-19 18:02 ` [PATCH v3 15/17] qapi/gen: Drop support for QAPIGen without a file name John Snow
2021-01-19 18:02 ` [PATCH v3 16/17] qapi: type 'info' as Optional[QAPISourceInfo] John Snow
2021-01-19 18:02 ` [PATCH v3 17/17] qapi: enable strict-optional checks John Snow
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=941acfa1-940a-b264-7da3-6df4e08661c2@redhat.com \
--to=jsnow@redhat.com \
--cc=armbru@redhat.com \
--cc=crosa@redhat.com \
--cc=ehabkost@redhat.com \
--cc=marcandre.lureau@redhat.com \
--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).