decode functions PER

[ezor1]

Dear eerimoq,

I'd like to use your tool for PER encoding/decoding. Do you know when the corresponding functions will be implemented?
Are you still working on it?

Thanks in advance

[eerimoq]

Hello!

asn1tools is a side project of mine that doesn't have any planning or schedule. Features are added whenever I find the project interesting, and a user request for additional functionality is certainly a good motivator. I wont promise anything, but I might have a look at it soon.

Is there any particular feature (as in ASN1 type) that you need?

[ezor1]

Thanks for your answer.
I've tryed to do it by myself but I can't work out how functions works (I'm quite new in Python)
However I could probably find out a solution if you implement one of them (let say the class Enumerate in per.py file for instance). Of course it's up to you.
Anyway, I don't want to be rude by putting pressure on you.

Thanks again,

Regards,

[eerimoq]

There are already a few functions implemented in https://github.com/eerimoq/asn1tools/blob/master/asn1tools/codecs/per.py with tests in https://github.com/eerimoq/asn1tools/blob/master/tests/test_per.py.

[eerimoq]

I implemented decode of the enumerated type in per.py. Not very well tested, but might work =)

[ezor1]

Many thanks errimoq. I'll try to use it and let you know if I find anything wrong (even if I'm sure it would work perfectly !)

[ezor1]

Hi,
Could you tell me wether the PER codec is the aligned or unaligned version? Cause I'm using asn1c and what your tools return does not match what I'm supposed to get (even if this is not completely different).

Thanks in advance

[eerimoq]

It's aligned PER. Unaligned PER may be implemented eventually, but first I'll finalize aligned PER.

[eerimoq]

Okay, I couldn't resist, I started on the UPER codec as well.

See https://github.com/eerimoq/asn1tools/blob/master/tests/test_uper.py for tests.

[eerimoq]

The UPER decoder now successfully decodes a couple ASN.1 types; SEQUENCE, CHOICE, BIT STRING, OCTET STRING, ENUMERATION, BOOLEAN and INTEGER. The implementation is by no means complete, but should generally work.

Would be nice if you can create a pull request with one or more test cases of your ASN.1 specification. It's expected that some of the test cases will fail, and that's fine.

Adding test cases will ensure that your types and be encoded and decoded. Add test cases in https://github.com/eerimoq/asn1tools/blob/master/tests/test_uper.py.

[ezor1]

I'll see what I can do

[ezor1]

Dear Eerimoq,

I've push a very simple script to test multiple modules case (I'll try to make something more comprehensive soon). It seems that it works here
Maybe it will be better/easier to allow compile_file() to use as first argument a list of files, each one containing a module and store all of them temporarily to deals with dependencies (IMPORTS/EXPORT types).

EDIT : my bad, it's quite easy to do what I described above thanks to compile_string().

The others tests I've performed on all types let me think your tools does the job pretty well though ;)

PS : I'm not very skilled in github, please let me know if the pull request you have is correct (especially if the syntax in Spec.asn is ok or not) cause I made two requests but I'm not sure if the second one has been taken into account.

[eerimoq]

Hello,

I think it's a good idea to accept either a single file or a list of files as first argument of compile_file(). It's probably a common use case.

I think the pull request (PR) includes your latest changes. I'll make a code review of the PR and eventually merge it.

[ezor1]

I'll make another PR to improve compile_file().

[eerimoq]

I already implemented compile_file() to allow one or more files. I also renamed the function to compile_files(). :)

[eerimoq]

Okay, I merged the first PR and styled the code to my liking. Anything else, or should we close this issue and open new issues if there are other features missing?