moritzs / jinja2-django-tags Goto Github PK

In Django 4.0+ an ImportError exception is raised on from django.utils.encoding import force_text as it is renamed to forse_str

First of all, this project is awesome! I don't understand why it doesn't have more traction. It's certainly more convenient than https://github.com/niwinz/django-jinja

Anyhow, for more Django tag compatibility it would help a lot to have a load tag that does absolutely nothing,
The code itself can easily be borrowed from coffin: https://github.com/cdleary/coffin/blob/05ab3fc79fe0c1b2aa4103f67520e109d660b999/coffin/template/defaulttags.py#L9-L24

Hi.
I'm not understanding, how date region dependent localization works.

1. Is it like django {{ date|localize }} ? I always get isoformat now, my FORMAT_MODULE_PATH is ignored,
2. If I want to get localized date, what should I activate? translation.activate('ru')? or timezone.activate(...) or both?
   Thanks

Hi.
When I updated jinja2 verion 3, templates using blocktrans started failing:

from django.template import engines
env = engines.all()[0].env
env.from_string("{% blocktrans %}test.key{% endblocktrans %}")

raises an error

Error
Traceback (most recent call last):
  File "/opt/project/tests/test_helpers.py", line 203, in test_without_nulls
    template = self.env.from_string("{% blocktrans %}test.key{% endblocktrans %}")
  File "/app/pip/lib/python3.10/site-packages/jinja2/environment.py", line 1092, in from_string
    return cls.from_code(self, self.compile(source), gs, None)
  File "/app/pip/lib/python3.10/site-packages/jinja2/environment.py", line 757, in compile
    self.handle_exception(source=source_hint)
  File "/app/pip/lib/python3.10/site-packages/jinja2/environment.py", line 925, in handle_exception
    raise rewrite_traceback_stack(source=source)
  File "<unknown>", line 1, in template
jinja2.exceptions.TemplateSyntaxError: unexpected end of template, expected 'end of statement block'.

Template:

{% blocktrans count candy_num=4 %}
  {{ candy_num }} candy
{% plural %}
  {{ candy_num }} candies
{% endblocktrans %}

Result:

'candy_num' is undefined

In extensions.py, things seem to get wrong at line 266:

            if context is None:
                return ugettext(force_text(singular)) % trans_vars
            else:
                return pgettext(force_text(context), force_text(singular)) % trans_vars
        else:
            if context is None:
                return ungettext(
                    force_text(singular), force_text(plural), trans_vars[count_var] # errorneous
                ) % trans_vars
            else:
                return npgettext(
                    force_text(context), force_text(singular), force_text(plural),
                    trans_vars[count_var]
                ) % trans_vars

Local vars:

Hi!

I'm a django-jinja developer and maintainer.

First of all, great work. I thinking that it make sense join forces and integrate that into django-jinja directly. If you are interested we can talk how we can colaborate.

Making this extensions easy available for all versions of django, not only django 1.8.

That do you thing about?

You should ping coffin to use those. Special thanks for the fact that you've implemented the pgettext with blocktrans context.

I'm now replacing my jinja2 trans tags with this because jinja2 doesn't support context...

Hi.
I use gunicorn server served by supervisor (in venv). When I reboot supervisor code is updated.
But when I cleanly install django-jinja on the server via pip and release project version with django-jinja included libs, I get an error:
("expected token 'name', got 'string'",)
in jinja2 templates on statements like:
{% trans 'panel.emails.password_reset.link_valid_interval' %}
After some time, the error is gone (some sort of caching?) and never occures again. The error can at the same time appear on one template and don't appear on another one.
I've tried deleting pycache directory in all project directories, but it didn't resolve the issue.

Hi.
I may have misunderstood something, but I can't make it working.
In russian language nouns are pluralized into 4 forms, not 2. Example:

{% blocktrans count counter=stats['users_count'] %}Уникальный посетитель{% plural %}Уникальные посетители{% endblocktrans %}

And .po file:

#: apps/jinja2/emails/app_stats_email.jinja2:348
msgid "Уникальный посетитель"
msgid_plural "Уникальные посетители"
msgstr[0] "Уникальный посетитель"
msgstr[1] "Уникальных посетителя"
msgstr[2] "Уникальных посетителей"

If I understood correctly, django's template pluralization should output one of msgstr[0] - msgstr[2] depending on the counter. But it always outputs msgid_plural.

Hi! I am switching from Django templates to Jinja and trying to write my own Extensions, by studying your code in extensions.py. Thank you for providing this guide.

I have a tag that takes 1 required positional arg and 1 optional kwarg:

{% form_field 'age' label='how old are you?' %}.

I know that in Jinja this should be a macro like {% form_field('age', label='how old are you?') %}, but for compatibility I need to preserve the old Django-style syntax.

I am wondering about the simplest way to implement this? Because this template tag is just like a python function with 2 args (no special syntax or custom keywords like with, as, etc).

Or do I need to parse the tokens one by one, check if each is TOKEN_STRING, TOKEN_NAME, etc.? Although I read your code, I am not sure how to deal with the case where the args are context variables, e.g.:

context = {'field_name': 'age', 'label': 'How old are you?'}

{% form_field field_name label=label %}.

Any advice? Thank you so much.