qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Stefan Hajnoczi <stefanha@redhat.com>
To: Mads Ynddal <mads@ynddal.dk>
Cc: qemu-devel@nongnu.org, Cleber Rosa <crosa@redhat.com>,
	John Snow <jsnow@redhat.com>, Mads Ynddal <m.ynddal@samsung.com>
Subject: Re: [PATCH v2 05/12] simpletrace: Changed Analyzer class to become context-manager
Date: Tue, 9 May 2023 10:40:04 -0400	[thread overview]
Message-ID: <20230509144004.GH1008478@fedora> (raw)
In-Reply-To: <20230502092339.27341-6-mads@ynddal.dk>

[-- Attachment #1: Type: text/plain, Size: 3282 bytes --]

On Tue, May 02, 2023 at 11:23:32AM +0200, Mads Ynddal wrote:
> From: Mads Ynddal <m.ynddal@samsung.com>
> 
> Instead of explicitly calling `begin` and `end`, we can change the class
> to use the context-manager paradigm. This is mostly a styling choice,
> used in modern Python code. But it also allows for more advanced analyzers
> to handle exceptions gracefully in the `__exit__` method (not
> demonstrated here).
> 
> Signed-off-by: Mads Ynddal <m.ynddal@samsung.com>
> ---
>  scripts/simpletrace.py | 40 +++++++++++++++++++++++-----------------
>  1 file changed, 23 insertions(+), 17 deletions(-)
> 
> diff --git a/scripts/simpletrace.py b/scripts/simpletrace.py
> index 7444a6e090..10ca093046 100755
> --- a/scripts/simpletrace.py
> +++ b/scripts/simpletrace.py
> @@ -121,12 +121,12 @@ def read_trace_records(event_mapping, event_id_to_name, fobj):
>  
>              yield rec
>  
> -class Analyzer(object):
> +class Analyzer:
>      """A trace file analyzer which processes trace records.
>  
> -    An analyzer can be passed to run() or process().  The begin() method is
> -    invoked, then each trace record is processed, and finally the end() method
> -    is invoked.
> +    An analyzer can be passed to run() or process().  The __enter__() method is
> +    invoked when opening the analyzer using the `with` statement, then each trace
> +    record is processed, and finally the __exit__() method is invoked.

Bearing in mind compatibility with existing simpletrace analysis
scripts, how about the following default method implementations?

  def __enter__(self):
      self.begin()

  def __exit__(self, exc_type, exc_val, exc_tb):
      if exc_type is None:
          self.end()
      return False

Now simpletrace.py can switch to using the context manager and new
scripts can implement __enter__()/__exit__(), while old scripts continue
to work.

>  
>      If a method matching a trace event name exists, it is invoked to process
>      that trace record.  Otherwise the catchall() method is invoked.
> @@ -152,19 +152,25 @@ def runstate_set(self, timestamp, pid, new_state):
>            ...
>      """
>  
> -    def begin(self):
> +    def __enter__(self):
>          """Called at the start of the trace."""
> -        pass
> +        return self
>  
>      def catchall(self, event, rec):
>          """Called if no specific method for processing a trace event has been found."""
>          pass
>  
> -    def end(self):
> +    def __exit__(self, _type, value, traceback):
>          """Called at the end of the trace."""
>          pass
>  
> -def process(events, log, analyzer, read_header=True):
> +    def __call__(self):
> +        """Fix for legacy use without context manager.
> +        We call the provided object in `process` regardless of it being the object-type or instance.
> +        With this function, it will work in both cases."""
> +        return self
> +
> +def process(events, log, analyzer_class, read_header=True):

Please don't change the function signature since this is a public method
and we should avoid breaking existing callers when possible.

Instead of:

  with analyzer_class() as analyzer:

we can use:

  with analyzer:
      ...

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

  reply	other threads:[~2023-05-09 14:40 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-02  9:23 [PATCH v2 00/12] simpletrace: refactor and general improvements Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 01/12] simpletrace: Improve parsing of sys.argv; fix files never closed Mads Ynddal
2023-05-04 18:03   ` Stefan Hajnoczi
2023-05-08 13:18     ` Mads Ynddal
2023-05-08 15:08       ` Stefan Hajnoczi
2023-05-02  9:23 ` [PATCH v2 02/12] simpletrace: Annotate magic constants from QEMU code Mads Ynddal
2023-05-09 14:34   ` Stefan Hajnoczi
2023-05-15  6:51     ` Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 03/12] simpletrace: changed naming of edict and idtoname to improve readability Mads Ynddal
2023-05-09 14:36   ` Stefan Hajnoczi
2023-05-02  9:23 ` [PATCH v2 04/12] simpletrace: update code for Python 3.11 Mads Ynddal
2023-05-09 14:38   ` Stefan Hajnoczi
2023-05-15  6:47     ` Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 05/12] simpletrace: Changed Analyzer class to become context-manager Mads Ynddal
2023-05-09 14:40   ` Stefan Hajnoczi [this message]
2023-05-15  7:48     ` Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 06/12] simpletrace: Simplify construction of tracing methods Mads Ynddal
2023-05-09 14:40   ` Stefan Hajnoczi
2023-05-15  8:11     ` Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 07/12] simpletrace: Improved error handling on struct unpack Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 08/12] simpletrace: define exception and add handling Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 09/12] simpletrace: Refactor to separate responsibilities Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 10/12] MAINTAINERS: add maintainer of simpletrace.py Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 11/12] scripts/analyse-locks-simpletrace.py: changed iteritems() to items() Mads Ynddal
2023-05-02  9:23 ` [PATCH v2 12/12] scripts/analyse-locks-simpletrace.py: reflect changes to process in simpletrace.py Mads Ynddal
2023-05-03 13:55 ` [PATCH v2 00/12] simpletrace: refactor and general improvements John Snow
2023-05-08 13:28   ` Mads Ynddal
2023-05-08 15:16     ` Stefan Hajnoczi
2023-05-10 19:14       ` John Snow
2023-05-04 17:48 ` Stefan Hajnoczi
2023-05-04 17:53   ` John Snow
2023-05-08 15:07     ` Stefan Hajnoczi
2023-05-08 16:50       ` Mads Ynddal
2023-05-09 14:33         ` Stefan Hajnoczi

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=20230509144004.GH1008478@fedora \
    --to=stefanha@redhat.com \
    --cc=crosa@redhat.com \
    --cc=jsnow@redhat.com \
    --cc=m.ynddal@samsung.com \
    --cc=mads@ynddal.dk \
    --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).