lstrip_blocks seems dead and doesn’t work

[avibrazil]

The lstrip_blocks doesn’t work as documented and seems to have no effect at all. I need it to help me attain more readable code and templates while keeping full control on rendered text.

Here is a fully functional sample code:

import textwrap
import jinja2

data=[
    dict(a=1,b=2),
    dict(a=3,b=4),
]

template = jinja2.Template(
    textwrap.dedent("""\
        Preamble

        {% for item in data %}
            Block start
            This is a:{{ item.a }}
            This is b:{{ item.b }}
            Block end
        {% endfor %}

        Epilogue
    """),
    lstrip_blocks=True
)

print(template.render(dict(data=data)))

Actual result:

Preamble


    Block start
    This is a:1
    This is b:2
    Block end

    Block start
    This is a:3
    This is b:4
    Block end


Epilogue

Expected and desired result:

Preamble


Block start
This is a:1
This is b:2
Block end

Block start
This is a:3
This is b:4
Block end


Epilogue

Environment:

• Python version: 3.12.8 on Fedora Linux
• Jinja version: 3.1.5

[avibrazil]

Why this bug report was converted into a discussion?
Any chance this is not bug and I'm doing something wrong in the code?

[spectereye]

met the same issue, lstrip_blocks not work as expected and which caused jinja template much less readability - surprise this bug exist so long for such mature lib ... or we did some config wrong?

[plgruener]

lstrip_blocks only works on those lines that begin with a block, the text lines inbetween are unaffected. from the docs:

The lstrip_blocks option can also be set to strip tabs and spaces from the beginning of a line to the start of a block. (Nothing will be stripped if there are other characters before the start of the block.)

This is only really visible if you have oneline-blocks like {%for item in data%}{{item}}{%endfor%} (because I think the leading spaces are stripped either way for multiline-blocks).

There's currently no way to have recursive indentation, see eg. #178