Django dumpdata fails on special characters
Question:
I’m trying to dump my entire DB to a json. When I run python manage.py dumpdata > data.json I get an error:
(env) PS C:devwatch_something> python manage.py dumpdata > data.json
CommandError: Unable to serialize database: 'charmap' codec can't encode character 'u0130' in position 1: character maps to <undefined>
Exception ignored in: <generator object cursor_iter at 0x0460C140>
Traceback (most recent call last):
File "C:devwatch_somethingenvlibsite-packagesdjangodbmodelssqlcompiler.py", line 1602, in cursor_iter
cursor.close()
sqlite3.ProgrammingError: Cannot operate on a closed database.
It’s because one of the characters in my DB is a sepcial character. How can I dump the DB correctly?
FYI, all other DB functionalities work fine
Answers:
To save json data in django the TextIOWrapper is used:
The default encoding is now locale.getpreferredencoding(False) (…)
In documentation of locale.getpreferredencoding fuction we can read:
Return the encoding used for text data, according to user preferences. User preferences are expressed differently on different systems, and might not be available programmatically on some systems, so this function only returns a guess.
Here I found "hacky" but working method to overwrite these settings:
In file settings.py of your django project add these lines:
import _locale
_locale._getdefaultlocale = (lambda *args: ['en_US', 'utf8'])
Here is the solution from djangoproject.com
You go to Settings there’s a "Use Unicode UTF-8 for worldwide language support", box in "Language" – "Administrative Language Settings" – "Change system locale" – "Region Settings".
If we apply that, and reboot, then we get a sensible, modern, default encoding from Python.
djangoproject.com
the settings box looks like this. Enable the check and restart your system
One solution is to use ./manage.py dumpdata -o data.json instead of ./manage.py dumpdata > data.json.
Another solution is to use Python’s UTF-8 mode, run:
python -Xutf8 ./manage.py dumpdata > data.json
If you have multiple Python environments, before applying workarounds, it’s worth checking that you issue python manage.py against the correct environment. That was my case where I met the same error: the database was created under containerized environment on Linux, with a higher Python version, but Django and other packages were present also in the legacy local environment on Windows. As well, the project directory was attached to the container as a volume, and the content was identical locally and in the container. So, I just mix up and ran manage.py locally instead of being attached to the container.
On windows the way i solved mine was
Add to your settings
import _locale
_locale._getdefaultlocale = (lambda *args: ['en_US', 'utf8'])
run this on shell only on windows
python -Xutf8 manage.py dumpdata --exclude auth.permission --exclude contenttypes > db.json
I actually did this it worked
python -Xutf8 manage.py dumpdata -o data.json
But wasnt showing data from my installed apps
I think this is the current best way: In the directory Python311Libsite-packagesdjangocoreserializers, you open the json.py file. In the json.py file, find the section:
class Serializer(PythonSerializer):
.................
def _init_options(self):
.............
# Default is False
self.json_kwargs.setdefault("ensure_ascii", False)
# You fix False to True
self.json_kwargs.setdefault("ensure_ascii", True)
I’m trying to dump my entire DB to a json. When I run python manage.py dumpdata > data.json I get an error:
(env) PS C:devwatch_something> python manage.py dumpdata > data.json
CommandError: Unable to serialize database: 'charmap' codec can't encode character 'u0130' in position 1: character maps to <undefined>
Exception ignored in: <generator object cursor_iter at 0x0460C140>
Traceback (most recent call last):
File "C:devwatch_somethingenvlibsite-packagesdjangodbmodelssqlcompiler.py", line 1602, in cursor_iter
cursor.close()
sqlite3.ProgrammingError: Cannot operate on a closed database.
It’s because one of the characters in my DB is a sepcial character. How can I dump the DB correctly?
FYI, all other DB functionalities work fine
To save json data in django the TextIOWrapper is used:
The default encoding is now
locale.getpreferredencoding(False)(…)
In documentation of locale.getpreferredencoding fuction we can read:
Return the encoding used for text data, according to user preferences. User preferences are expressed differently on different systems, and might not be available programmatically on some systems, so this function only returns a guess.
Here I found "hacky" but working method to overwrite these settings:
In file settings.py of your django project add these lines:
import _locale
_locale._getdefaultlocale = (lambda *args: ['en_US', 'utf8'])
Here is the solution from djangoproject.com
You go to Settings there’s a "Use Unicode UTF-8 for worldwide language support", box in "Language" – "Administrative Language Settings" – "Change system locale" – "Region Settings".
If we apply that, and reboot, then we get a sensible, modern, default encoding from Python.
djangoproject.com
the settings box looks like this. Enable the check and restart your system
One solution is to use ./manage.py dumpdata -o data.json instead of ./manage.py dumpdata > data.json.
Another solution is to use Python’s UTF-8 mode, run:
python -Xutf8 ./manage.py dumpdata > data.json
If you have multiple Python environments, before applying workarounds, it’s worth checking that you issue python manage.py against the correct environment. That was my case where I met the same error: the database was created under containerized environment on Linux, with a higher Python version, but Django and other packages were present also in the legacy local environment on Windows. As well, the project directory was attached to the container as a volume, and the content was identical locally and in the container. So, I just mix up and ran manage.py locally instead of being attached to the container.
On windows the way i solved mine was
Add to your settings
import _locale
_locale._getdefaultlocale = (lambda *args: ['en_US', 'utf8'])
run this on shell only on windows
python -Xutf8 manage.py dumpdata --exclude auth.permission --exclude contenttypes > db.json
I actually did this it worked
python -Xutf8 manage.py dumpdata -o data.json
But wasnt showing data from my installed apps
I think this is the current best way: In the directory Python311Libsite-packagesdjangocoreserializers, you open the json.py file. In the json.py file, find the section:
class Serializer(PythonSerializer):
.................
def _init_options(self):
.............
# Default is False
self.json_kwargs.setdefault("ensure_ascii", False)
# You fix False to True
self.json_kwargs.setdefault("ensure_ascii", True)