Silence ply errors about slimit HOT 4 OPEN

This can be worked around by removing those *tab.py files under the slimit directory (along with the related .pyc there or inside the __pycache__ directory); you will likely need to be the root user (or sudo) to remove them and regenerate them (again as root) using the current version of ply that was also installed (by running python -c 'import slimit;slimit.minify("")', and as root for a system level installation).

Really though, this package doesn't look maintained anymore so I ended up forking this, took the good bits as a base for my own.

from slimit.

I get the same verbose dreck spewed out to stderr. This exceedingly ugly one-liner anywhere after import slimit, but before calling minify() silences the logging:

slimit.lexer.ply.lex.PlyLogger =  \
slimit.parser.ply.yacc.PlyLogger = \
  type('_NullLogger', (slimit.lexer.ply.lex.NullLogger,),
       dict(__init__=lambda s, *_, **__: (None, s.super().__init__())[0]))

This works for me in Python 3.6 with slimit 0.8.1 and ply 3.11 -- maybe others, but YMMV.

All this does is monkey-patches the PlyLogger class with the NullLogger class the ply package defines itself, adapting for the differences in initializer signatures between the two classes (shame, shame). If ply has a configuration setting or optional initializer parameter somewhere that effectively does this -- and it should -- I couldn't find it. (Hint, hint, ply maintainers.) Maybe in lieu of that, slimit should do this internally instead of putting the onus on the user?

from slimit.

@rod-app The problem actually goes deeper than that - if those messages are generated every time minify is called, it means that the ply tab modules are being regenerated (which takes as long as a second), and if there are a number of JavaScript files being processed that extra delay and thus performance hit to the program will accumulate rather quickly. Removing the message simply hides the problem but not actually fixing it - the only way to fix it is to manually remove the problematic tab files and regenerate them correct as I had suggested.

The fork that I made (calmjs.parse) does not track the tab files for the source repository, but they are package/generated for every supported (and released) ply versions, such that they are not needlessly being regenerated for the end-users, and in the case a newer version of ply is out there is a specific command (python -m calmjs.parse.parsers.optimize) with the reasoning behind it explained in the README for the package.

Also please note that slimit has not received a new release for nearly 6 years now, so I made that fork as an attempt to rectify the situation (the slimit minify equivalent is available via from calmjs.parse import es5 and then call es5.minify_print).

from slimit.

from slimit.