René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"PYTHONSTARTUP file with non-ascii codes generates error","articleBody":"I have this code set up on `.pythonrc`:\r\n\r\n```\r\ntry:\r\n    import django\r\n    from django.conf import settings\r\n    if not settings.configured:\r\n        django.setup()\r\n        print('Imported django settings.')\r\n        try:\r\n            exec_strs = [\r\n                (app, 'from {}.models import *'.format(app))\r\n                for app in settings.INSTALLED_APPS\r\n            ]\r\n            for app, x in exec_strs:\r\n                try:\r\n                    exec(x)\r\n                except:\r\n                    pass\r\n                else:\r\n                    print('Imported models for app {}'.format(app))\r\n        except:\r\n            pass\r\nexcept django.core.exceptions.ImproperlyConfigured as msg:\r\n    print('Erro na configuração do django ao importar módulo: {}'.format(msg))\r\n```\r\n\r\nOn latest bpython, it prints an ascii decode error whenever I start bpython (I'm running Python 3.9.2).\r\n\r\nThe issue is this code:\r\n\r\nhttps://github.com/bpython/bpython/blob/34f0e700082bf79223689ff120dddf4e46442129/bpython/inspection.py#L332\r\n\r\nIt assumes old style comment block `coding: utf-8`, and if it fails, then assumes ascii encoding.\r\n\r\nYou could assume `utf-8` encoding if there is no block, or try to decode the string and catch a `UnicodeDecodeError`.","author":{"url":"https://github.com/niloct","@type":"Person","name":"niloct"},"datePublished":"2021-03-11T22:35:27.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":0},"url":"https://github.com/888/bpython/issues/888"}