String strangely turned into mako expression

hi

In this environment setup:

1. A py3.4 virtual-env
2. Created a Pyramid application (www.pylonsproject.org) using sqlalchemy starterkit.
3. Manually replaced chamelion templating system with mako (everything works fine so far and the page is rendering in browser with no problem).
4. Converted PyHAML using 2to3 (everything went fine).
5. Added "mako.preprocessor = haml.preprocessor" to application's .ini file which will be picked up by mako later.
6. Running (in virtual-env) pserve development.ini
7. Visiting application's web page will raise a mako exception.

I put a open('/tmp/x', 'w').write(generate_mako(parse_string(source))) in init.py file to investigate it more, looking at the file 'x' I can see this comment which is by default in pyramid's page has turned to Mako expression:

Original in mytemplate.mako:

<!-- Bootstrap core JavaScript
================================================== -->
<!-- Placed at the end of the document so the pages load faster -->

Output by PyHAML:

<!-- Bootstrap core JavaScript -->
${================================================== -->}
<!-- Placed at the end of the document so the pages load faster -->

Simply removing the offending line like this:

fixed the problem (for now).

is this really an issue or did I miss something in the docs? That 2to3 thing seems fishy though.
best wishes

[mikeboers]

What is happening is the first = is signalling that the rest of the line is a Python expression that should be inserted. See the ruby docs.

The question then becomes if we are being too accepting in taking =expr vs = expr (note the space).

I am also surprised at the 2to3 being necessary; it has been a while since I worked on the Python 3 aspect of this package, but I remember getting it to work seamlessly in both.

maybe 2to3 wasn't necessary. It was ran before using/running PyHAML under py3, it found some py2 print statements though. Sorry about my carelessness.

And about the question, I guess decision is all yours! Thanks for the quick response + I leave this issue open so you can close it if you think it's not a major problem.

[mikeboers]

No apologies necessary. I'm not even sure it doesn't need 2to3 honestly. I'll take a look in the future.

The thing for me to check is if the Ruby HAML requires that space or not, and then do that.

I'll leave this open until I check that out.

Cheers,

Mike

On Jul 30, 2014, at 11:40 AM, Koosha Hosseiny notifications@github.com wrote:

maybe 2to3 wasn't necessary. It was ran before using/running PyHAML under py3, it found some py2 print statements though. Sorry about my carelessness.

And about the question, I guess decision is all yours! Thanks for the quick response + I leave this issue open so you can close it if you think it's not a major problem.

—
Reply to this email directly or view it on GitHub.