Add support for Level 3

[prologic]

Adding support for Level 3 would be nice. I plan to evaluate the use of your library in a production project at work. The spec seems fairly clear to me on Level 3 aspects. Not sure on "expressions" though.

[mwhooker]

Hi @therealprologic,

funny you should mention it. level 3 is something I plan on implementing this weekend.

@latestrevision implemented the level in go, which should be quite transferrable https://github.com/latestrevision/go-jsonselect

He also pointed me to this article which I hope should clear things up http://lloyd.io/jsonselect-grows-up

In the mean time, don't let me stop you from contributing any fixes you see

[prologic]

Thats fantastic news, Look forward to it!

Sent from my iPad

On 28 Nov 2013, at 15:53, Matthew Hooker [email protected] wrote:

Hi @therealprologic https://github.com/therealprologic,

funny you should mention it. level 3 is something I plan on implementing this weekend.

@latestrevision https://github.com/latestrevision implemented the level in go, which should be quite transferrable https://github.com/latestrevision/go-jsonselect

He also pointed me to this article which I hope should clear things up http://lloyd.io/jsonselect-grows-up

In the mean time, don't let me stop you from contributing any fixes you see

— Reply to this email directly or view it on GitHubhttps://github.com/mwhooker/jsonselect/issues/1#issuecomment-29441878 .

[prologic]

Hi @mwhooker just wondering if you are still working on this or have lost a bit of interest? Curious more than anything :)

[prologic]

FAIL tests/test_conformance.py::test[level_3_basic_multiple-has-with-strings]
FAIL tests/test_conformance.py::test[level_3_expr_greater-than]
FAIL tests/test_conformance.py::test[level_3_expr_less-than]
FAIL tests/test_conformance.py::test[level_3_expr_mod]
FAIL tests/test_conformance.py::test[level_3_expr_mult]
FAIL tests/test_conformance.py::test[level_3_expr_null-eq]
FAIL tests/test_conformance.py::test[level_3_expr_precedence-2]
FAIL tests/test_conformance.py::test[level_3_expr_simple-false]
FAIL tests/test_conformance.py::test[level_3_polykids_val]

So I've been trying to work on this but running into several issues:

• The context free grammar that jsonselect.org defines is probably more complex than what re.Scanner can provide... I'm tempted to use a proper parser such as rply, funcparserlib, pyparsing, etc.

[prologic]

ping @mwhooker

[danvk]

Since this project seems to have stalled, I've created a new one: pyjsonselect. It's based on the JS reference implementation and hence is fully conformant with all levels of the spec. If you need the level 3 selectors, you may wish to give it a try.

[mwhooker]

Hi All,

sorry for the radio silence. I'm no longer maintaining this, as there are far better solutions, such as http://www.jsoniq.org/ and https://github.com/boto/jmespath and now @danvk's pyjsonselect.

[prologic]

Awesome! Thanks guys :) I'll check these out!

James Mills / prologic

E: [email protected] W: prologic.shortcircuit.net.au

On Tue, Oct 7, 2014 at 10:43 AM, Matthew Hooker [email protected] wrote:

Hi All,

sorry for the radio silence. I'm no longer maintaining this, as there are far better solutions, such as http://www.jsoniq.org/ and https://github.com/boto/jmespath and now @danvk https://github.com/danvk's

pyjsonselect https://github.com/boto/jmespath.

— Reply to this email directly or view it on GitHub https://github.com/mwhooker/jsonselect/issues/1#issuecomment-58121860.