Assignment operator and multi-line

[Vivek Shankar]

Hi,

I have been using CEL in my product as an engine to execute user-authored "code" and it works great. I am working on a multi-line capability, but I was wondering if there were any thoughts on adding two things to the parser -

1. An end-of-line terminator: Like semi-colon in certain languages
2. An assignment operator: I know this opens up scoping of variables etc. but it might be possible to define certain caveats. For example, you can't define the same var name in a sub-frame.

This would save me the need to write my own parser.

[Tristan Swadell]

Hi Vivek,

If you're looking for an example of assignments, I would recommend looking over https://github.com/google/cel-policy-templates-go for how we combined CEL + YAML for making assignments and more complex logical constructs. CEL will forever remain small so that it more easily be embedded into larger frameworks. The moment we introduce assignments, we'll increase the analytical complexity of the core language in ways that would be undesirable.

If you're looking for tips on how to embed CEL into a larger ANTLR grammar, I can provide some examples there as well.

-Tristan

--
You received this message because you are subscribed to the Google Groups "CEL Go Discussion Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cel-go-discus...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cel-go-discuss/b3579d68-e1c0-49e4-9771-964dd1393cfao%40googlegroups.com.

[Michael Leonard]

Just in case it's helpful... you might want to look into cue as well (in go, also by Google).

https://cuelang.org/

To view this discussion on the web visit https://groups.google.com/d/msgid/cel-go-discuss/CAHx9AioAtkfwXm4C5%3DfcXWSGR8GEMFtJ%2BLwK9nAsz2PfbjEV9Q%40mail.gmail.com.

[Vivek Shankar]

Hi Tristan,

Thanks for the reply. Makes sense.

I have managed to implement something similar. Here's an example -

stmts:

- init: userExists = has(user.name)

- init: givenNameExists = context.userExists && has(user.name.givenName) && user.name.givenName != ""

- init: familyNameExists = context.userExists && has(user.name.familyName) && user.name.familyName != ""

- init: formattedExists = context.userExists && has(user.name.formatted) && user.name.formatted != ""

- if:

match: context.formattedExists

block:

- ret: user.name.formatted

eifs:

- match: context.givenNameExists

block:

- if:

match: context.familyNameExists

block:

- ret: user.name.familyName + ", " + user.name.givenName

else:

- ret: user.name.givenName

- match: context.familyNameExists

block:

- ret: user.name.familyName

The "context" var implements traits.Indexer and contains a list of maps representing stacked state, so assigned variables can be inherited in nested blocks (like in nested if conditions etc.).

On Tuesday, 14 July 2020 01:40:43 UTC+8, Tristan Swadell wrote:

Hi Vivek,

If you're looking for an example of assignments, I would recommend looking over https://github.com/google/cel-policy-templates-go for how we combined CEL + YAML for making assignments and more complex logical constructs. CEL will forever remain small so that it more easily be embedded into larger frameworks. The moment we introduce assignments, we'll increase the analytical complexity of the core language in ways that would be undesirable.

If you're looking for tips on how to embed CEL into a larger ANTLR grammar, I can provide some examples there as well.

-Tristan

On Fri, Jul 10, 2020 at 7:24 PM Vivek Shankar <viv.s...@gmail.com> wrote:

Hi,

I have been using CEL in my product as an engine to execute user-authored "code" and it works great. I am working on a multi-line capability, but I was wondering if there were any thoughts on adding two things to the parser -

1. An end-of-line terminator: Like semi-colon in certain languages
2. An assignment operator: I know this opens up scoping of variables etc. but it might be possible to define certain caveats. For example, you can't define the same var name in a sub-frame.

This would save me the need to write my own parser.

--
You received this message because you are subscribed to the Google Groups "CEL Go Discussion Forum" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cel-go-...@googlegroups.com.

[Vivek Shankar]

Thanks! I will check it out.

On Tuesday, 14 July 2020 15:45:46 UTC+8, Michael Leonard wrote:

Just in case it's helpful... you might want to look into cue as well (in go, also by Google).

https://cuelang.org/

On Mon, 13 Jul 2020, 18:40 'Tristan Swadell' via CEL Go Discussion Forum, <cel-go-...@googlegroups.com> wrote:

Hi Vivek,

If you're looking for an example of assignments, I would recommend looking over https://github.com/google/cel-policy-templates-go for how we combined CEL + YAML for making assignments and more complex logical constructs. CEL will forever remain small so that it more easily be embedded into larger frameworks. The moment we introduce assignments, we'll increase the analytical complexity of the core language in ways that would be undesirable.

If you're looking for tips on how to embed CEL into a larger ANTLR grammar, I can provide some examples there as well.

-Tristan

On Fri, Jul 10, 2020 at 7:24 PM Vivek Shankar <viv.s...@gmail.com> wrote:

Hi,

I have been using CEL in my product as an engine to execute user-authored "code" and it works great. I am working on a multi-line capability, but I was wondering if there were any thoughts on adding two things to the parser -

1. An end-of-line terminator: Like semi-colon in certain languages
2. An assignment operator: I know this opens up scoping of variables etc. but it might be possible to define certain caveats. For example, you can't define the same var name in a sub-frame.

This would save me the need to write my own parser.

--
You received this message because you are subscribed to the Google Groups "CEL Go Discussion Forum" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cel-go-...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cel-go-discuss/b3579d68-e1c0-49e4-9771-964dd1393cfao%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "CEL Go Discussion Forum" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cel-go-...@googlegroups.com.