From: John Snow <jsnow@redhat.com>
To: Markus Armbruster <armbru@redhat.com>
Cc: "Peter Maydell" <peter.maydell@linaro.org>,
"Cleber Rosa" <crosa@redhat.com>,
"Alex Bennée" <alex.bennee@linaro.org>,
qemu-devel@nongnu.org, "Eduardo Habkost" <ehabkost@redhat.com>
Subject: Re: [PATCH 09/37] qapi/common.py: Add indent manager
Date: Wed, 16 Sep 2020 18:25:04 -0400 [thread overview]
Message-ID: <37ea889c-746e-bea9-a719-6bee9e86f1a8@redhat.com> (raw)
In-Reply-To: <87k0wtiwlb.fsf@dusky.pond.sub.org>
On 9/16/20 11:13 AM, Markus Armbruster wrote:
> John Snow <jsnow@redhat.com> writes:
>
>> Code style tools really dislike the use of global keywords, because it
>> generally involves re-binding the name at runtime which can have strange
>> effects depending on when and how that global name is referenced in
>> other modules.
>>
>> Make a little indent level manager instead.
>>
>> Signed-off-by: John Snow <jsnow@redhat.com>
>> ---
>> scripts/qapi/common.py | 50 +++++++++++++++++++++++++++++-------------
>> scripts/qapi/visit.py | 7 +++---
>> 2 files changed, 38 insertions(+), 19 deletions(-)
>>
>> diff --git a/scripts/qapi/common.py b/scripts/qapi/common.py
>> index 4fb265a8bf..87d87b95e5 100644
>> --- a/scripts/qapi/common.py
>> +++ b/scripts/qapi/common.py
>> @@ -93,33 +93,53 @@ def c_name(name, protect=True):
>> pointer_suffix = ' *' + eatspace
>>
>>
>> -def genindent(count):
>> - ret = ''
>> - for _ in range(count):
>> - ret += ' '
>> - return ret
>> +class Indent:
>
> Since this class name will be used just once... Indentation or
> IndentationManager?
>
Indentation is fine, if you'd like. IndentationManager makes me feel
ashamed for writing this patch, like I am punishing myself with Java.
>> + """
>> + Indent-level management.
>>
>> + :param initial: Initial value, default 0.
>
> The only caller passes 0.
>
> Let's drop the parameter, and hardcode the default initial value until
> we have an actual use for another initial value.
>
The parameter is nice because it gives meaning to the output of repr(),
see below.
>> + """
>> + def __init__(self, initial: int = 0) -> None:
>> + self._level = initial
>>
>> -indent_level = 0
>> + def __int__(self) -> int:
>> + """Return the indent as an integer."""
>
> Isn't "as an integer" obvious?
Just a compulsion to write complete sentences that got lost in the sea
of many patches. I'll nix this one, because dunder methods do not need
to be documented.
>
> Let's replace "the indent" by "the indentation" globally.
>
They're both cromulent as nouns and one is shorter.
I'll switch in good faith; do you prefer "Indentation" as a noun?
>> + return self._level
>>
>> + def __repr__(self) -> str:
>> + return "{}({:d})".format(type(self).__name__, self._level)
>
> Is __repr__ needed?
>
Yes; it's used in the underflow exception , and it is nice when using
the python shell interactively.
repr(Indent(4)) == "Indent(4)"
>>
>> -def push_indent(indent_amount=4):
>> - global indent_level
>> - indent_level += indent_amount
>> + def __str__(self) -> str:
>> + """Return the indent as a string."""
>> + return ' ' * self._level
>>
>> + def __bool__(self) -> bool:
>> + """True when there is a non-zero indent."""
>> + return bool(self._level)
>
> This one lets us shorten
>
> if int(INDENT):
>
> to
>
> if INDENT:
>
> There's just one instance. Marginal. I'm not rejecting it.
>
Yep, it's trivial in either direction. Still, because it's a custom
type, I thought I'd be courteous and have it support bool().
>> -def pop_indent(indent_amount=4):
>> - global indent_level
>> - indent_level -= indent_amount
>> + def push(self, amount: int = 4) -> int:
>> + """Push `amount` spaces onto the indent, default four."""
>> + self._level += amount
>> + return self._level
>> +
>> + def pop(self, amount: int = 4) -> int:
>> + """Pop `amount` spaces off of the indent, default four."""
>> + if self._level < amount:
>> + raise ArithmeticError(
>> + "Can't pop {:d} spaces from {:s}".format(amount, repr(self)))
>> + self._level -= amount
>> + return self._level
>
> The push / pop names never made sense. The functions don't push onto /
> pop from a stack, they increment / decrement a counter, and so do the
> methods. Good opportunity to fix the naming.
>
OK.
I briefly thought about using __isub__ and __iadd__ to support e.g.
indent += 4, but actually it'd be annoying to have to specify 4 everywhere.
Since you didn't suggest anything, I am going to change it to 'increase'
and 'decrease'.
> The @amount parameter has never been used. I don't mind keeping it.
>
I'll keep it, because I like having the default amount show up in the
signature instead of as a magic constant in the function body.
>> +
>> +
>> +INDENT = Indent(0)
>
> Uh, does this really qualify as a constant? It looks quite variable to
> me...
>
Ever make a mistake? I thought I did once, but I was mistaken.
>> # Generate @code with @kwds interpolated.
>> # Obey indent_level, and strip eatspace.
>> def cgen(code, **kwds):
>> raw = code % kwds
>> - if indent_level:
>> - indent = genindent(indent_level)
>> - raw, _ = re.subn(r'^(?!(#|$))', indent, raw, flags=re.MULTILINE)
>> + if INDENT:
>> + raw, _ = re.subn(r'^(?!(#|$))', str(INDENT), raw, flags=re.MULTILINE)
>> return re.sub(re.escape(eatspace) + r' *', '', raw)
>>
>>
>> diff --git a/scripts/qapi/visit.py b/scripts/qapi/visit.py
>> index 31bf46e7f7..66ce3dc52e 100644
>> --- a/scripts/qapi/visit.py
>> +++ b/scripts/qapi/visit.py
>> @@ -19,8 +19,7 @@
>> gen_endif,
>> gen_if,
>> mcgen,
>> - pop_indent,
>> - push_indent,
>> + INDENT,
>> )
>> from .gen import QAPISchemaModularCVisitor, ifcontext
>> from .schema import QAPISchemaObjectType
>> @@ -68,7 +67,7 @@ def gen_visit_object_members(name, base, members, variants):
>> if (visit_optional(v, "%(name)s", &obj->has_%(c_name)s)) {
>> ''',
>> name=memb.name, c_name=c_name(memb.name))
>> - push_indent()
>> + INDENT.push()
>> ret += mcgen('''
>> if (!visit_type_%(c_type)s(v, "%(name)s", &obj->%(c_name)s, errp)) {
>> return false;
>> @@ -77,7 +76,7 @@ def gen_visit_object_members(name, base, members, variants):
>> c_type=memb.type.c_name(), name=memb.name,
>> c_name=c_name(memb.name))
>> if memb.optional:
>> - pop_indent()
>> + INDENT.pop()
>> ret += mcgen('''
>> }
>> ''')
next prev parent reply other threads:[~2020-09-16 22:26 UTC|newest]
Thread overview: 98+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-15 22:39 [PATCH 00/37] qapi: static typing conversion, pt1 John Snow
2020-09-15 22:39 ` [PATCH 01/37] python: Require 3.6+ John Snow
2020-09-16 8:42 ` Markus Armbruster
2020-09-16 8:51 ` Daniel P. Berrangé
2020-09-15 22:39 ` [PATCH 02/37] [DO-NOT-MERGE] qapi: add debugging tools John Snow
2020-09-15 22:39 ` [PATCH 03/37] qapi-gen: Separate arg-parsing from generation John Snow
2020-09-15 22:39 ` [PATCH 04/37] qapi: move generator entrypoint into module John Snow
2020-09-16 11:54 ` Markus Armbruster
2020-09-16 14:24 ` John Snow
2020-09-17 7:46 ` Markus Armbruster
2020-09-15 22:39 ` [PATCH 05/37] qapi: Remove wildcard includes John Snow
2020-09-15 22:39 ` [PATCH 06/37] qapi: delint using flake8 John Snow
2020-09-16 12:12 ` Markus Armbruster
2020-09-16 14:29 ` John Snow
2020-09-17 7:54 ` Markus Armbruster
2020-09-17 16:57 ` John Snow
2020-09-18 10:33 ` Markus Armbruster
2020-09-18 18:13 ` John Snow
2020-09-21 7:31 ` Markus Armbruster
2020-09-21 14:50 ` John Snow
2020-09-15 22:39 ` [PATCH 07/37] qapi: add pylintrc John Snow
2020-09-16 12:30 ` Markus Armbruster
2020-09-16 14:37 ` John Snow
2020-09-17 7:58 ` Markus Armbruster
2020-09-17 17:06 ` John Snow
2020-09-15 22:39 ` [PATCH 08/37] qapi/common.py: Remove python compatibility workaround John Snow
2020-09-16 12:34 ` Markus Armbruster
2020-09-16 14:38 ` John Snow
2020-09-15 22:39 ` [PATCH 09/37] qapi/common.py: Add indent manager John Snow
2020-09-16 15:13 ` Markus Armbruster
2020-09-16 22:25 ` John Snow [this message]
2020-09-17 8:07 ` Markus Armbruster
2020-09-17 17:18 ` John Snow
2020-09-18 10:55 ` Markus Armbruster
2020-09-18 16:08 ` John Snow
2020-09-21 7:43 ` Markus Armbruster
2020-09-15 22:40 ` [PATCH 10/37] qapi/common.py: delint with pylint John Snow
2020-09-17 14:15 ` Markus Armbruster
2020-09-17 17:48 ` John Snow
2020-09-18 11:03 ` Markus Armbruster
2020-09-15 22:40 ` [PATCH 11/37] qapi/common.py: Replace one-letter 'c' variable John Snow
2020-09-17 14:17 ` Markus Armbruster
2020-09-17 17:51 ` John Snow
2020-09-15 22:40 ` [PATCH 12/37] qapi/common.py: check with pylint John Snow
2020-09-15 22:40 ` [PATCH 13/37] qapi/common.py: add notational type hints John Snow
2020-09-17 14:32 ` Markus Armbruster
2020-09-17 18:18 ` John Snow
2020-09-17 20:06 ` John Snow
2020-09-18 11:14 ` Markus Armbruster
2020-09-18 15:24 ` John Snow
2020-09-15 22:40 ` [PATCH 14/37] qapi/common.py: Move comments into docstrings John Snow
2020-09-17 14:37 ` Markus Armbruster
2020-09-17 18:44 ` John Snow
2020-09-17 19:14 ` Eduardo Habkost
2020-09-17 19:31 ` John Snow
2020-09-24 15:06 ` Markus Armbruster
2020-09-24 16:31 ` John Snow
2020-09-25 7:49 ` Markus Armbruster
2020-09-25 14:07 ` John Snow
2020-09-15 22:40 ` [PATCH 15/37] qapi/common.py: split build_params into new file John Snow
2020-09-17 14:42 ` Markus Armbruster
2020-09-17 18:53 ` John Snow
2020-09-17 19:40 ` John Snow
2020-09-15 22:40 ` [PATCH 16/37] qapi: establish mypy type-checking baseline John Snow
2020-09-18 11:55 ` Markus Armbruster
2020-09-18 14:27 ` John Snow
2020-09-21 8:05 ` Markus Armbruster
2020-09-21 14:41 ` John Snow
2020-09-25 1:18 ` Eduardo Habkost
2020-09-18 19:03 ` John Snow
2020-09-21 8:05 ` Markus Armbruster
2020-09-21 14:46 ` John Snow
2020-09-15 22:40 ` [PATCH 17/37] qapi/events.py: add notational type hints John Snow
2020-09-15 22:40 ` [PATCH 18/37] qapi/events.py: Move comments into docstrings John Snow
2020-09-15 22:40 ` [PATCH 19/37] qapi/commands.py: Don't re-bind to variable of different type John Snow
2020-09-15 22:40 ` [PATCH 20/37] qapi/commands.py: add notational type hints John Snow
2020-09-15 22:40 ` [PATCH 21/37] qapi/commands.py: enable checking with mypy John Snow
2020-09-15 22:40 ` [PATCH 22/37] qapi/source.py: add notational type hints John Snow
2020-09-15 22:40 ` [PATCH 23/37] qapi/source.py: delint with pylint John Snow
2020-09-15 22:40 ` [PATCH 24/37] qapi/gen.py: Fix edge-case of _is_user_module John Snow
2020-09-15 22:40 ` [PATCH 25/37] qapi/gen.py: add notational type hints John Snow
2020-09-15 22:40 ` [PATCH 26/37] qapi/gen.py: Enable checking with mypy John Snow
2020-09-15 22:40 ` [PATCH 27/37] qapi/gen.py: Remove unused parameter John Snow
2020-09-15 22:40 ` [PATCH 28/37] qapi/gen.py: update write() to be more idiomatic John Snow
2020-09-15 22:40 ` [PATCH 29/37] qapi/gen.py: delint with pylint John Snow
2020-09-15 22:40 ` [PATCH 30/37] qapi/introspect.py: Add a typed 'extra' structure John Snow
2020-09-15 22:40 ` [PATCH 31/37] qapi/introspect.py: add _gen_features helper John Snow
2020-09-15 22:40 ` [PATCH 32/37] qapi/introspect.py: create a typed 'Node' data structure John Snow
2020-09-15 22:40 ` [PATCH 33/37] qapi/introspect.py: add notational type hints John Snow
2020-09-15 22:40 ` [PATCH 34/37] qapi/types.py: " John Snow
2020-09-15 22:40 ` [PATCH 35/37] qapi/types.py: remove one-letter variables John Snow
2020-09-15 22:40 ` [PATCH 36/37] qapi/visit.py: remove unused parameters from gen_visit_object John Snow
2020-09-15 22:40 ` [PATCH 37/37] qapi/visit.py: add notational type hints John Snow
2020-09-16 22:33 ` [PATCH 00/37] qapi: static typing conversion, pt1 John Snow
2020-09-17 20:22 ` John Snow
2020-09-18 7:50 ` Markus Armbruster
2020-09-18 13:07 ` Philippe Mathieu-Daudé
2020-09-18 14:30 ` 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=37ea889c-746e-bea9-a719-6bee9e86f1a8@redhat.com \
--to=jsnow@redhat.com \
--cc=alex.bennee@linaro.org \
--cc=armbru@redhat.com \
--cc=crosa@redhat.com \
--cc=ehabkost@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).