A new load method returning a tuple (an instance of current class and byte length)

[ladaan]

No description provided.

[wbond]

Looks like I need to tweak the dependency resolver now that projects are starting to drop Python 2 support.

Could you describe your use case, perhaps with some example code?

I'm somewhat inclined to create a new method rather than having a param change the type of the return value.

[codecov]

Codecov Report

Merging #176 into master will decrease coverage by 0.06%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##           master     #176      +/-   ##
==========================================
- Coverage   87.16%   87.09%   -0.07%     
==========================================
  Files          21       21              
  Lines        5243     5246       +3     
==========================================
- Hits         4570     4569       -1     
- Misses        673      677       +4

Impacted Files	Coverage Δ
asn1crypto/core.py	82.24% <66.66%> (-0.12%)	⬇️
asn1crypto/_int.py	80% <0%> (-20%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c09b87c...afd05cb. Read the comment docs.

[ladaan]

Here is an use case:

cdr_file = open(filename, 'rb')
data = cdr_file.read()

offset = 0
bytes_left = len(data)
while bytes_left > 0:
    asrecord, length = spec_class.load_p(data[offset:])
    records.append(asrecord)
    bytes_left = bytes_left - length
    offset = offset + length

Data on the input are ASN.1 encoded CDR records (several fields per record).
They need to be decoded based on the "spec_class" that is defined separately along with other subclasses (definition of data types of each field etc.)

Possibly you can suggest the method name if you like.

[joernheissler]

What is "CDR"?
The problem you're trying to solve looks similar to #152.

[ladaan]

Thanks I will have a look. CDR is a Call Detail Record. It is from telecommunication terminology. CDRs keeps the history of all dialogues that includes calling and called party and further important fields.

[ladaan]

Joern is right. I can use parser.parse method
Is there any more effective use of buffers where I would use pointers instead of a new object (slice of data)?
Such as memoryview (but it is not possible here as the parse method expects a byte string)

Here is the code I used:

data = cdr_file.read()
offset = 0
bytes_left = len(data) 
while bytes_left: 
    parsed_data = parser.parse(data[offset:])
    record = parsed_data[3] + parsed_data[4] + parsed_data[5]
    length = len(record) 
    asrecord = spec_class.load(record)
    records.append(asrecord)
    bytes_left = bytes_left - length
    offset = offset + length

cdr_file.close()

Otherwise my pull request can be closed as a solution exists.

[joernheissler]

Is there any more effective use of buffers where I would use pointers instead of a new object

Python brings a bytearray object which has efficient append/remove operations on both ends:

>>> from os import urandom
>>> buf = bytearray(urandom(1_000_000_000))
>>> del buf[:567] # finishes near instantly

[joernheissler]

Otherwise my pull request can be closed as a solution exists.

I don't think there's a solution yet.