[WIP] Implement Git Wire Protocol v2 #420
Draft
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ulugbekna
force-pushed
the
protocol-v2
branch
from
88b5037
to
8b9ec8b
Compare
dinosaure
reviewed
Nov 29, 2020
dinosaure
reviewed
Nov 29, 2020
dinosaure
reviewed
Nov 29, 2020
Merged
ulugbekna
force-pushed
the
protocol-v2
branch
2 times, most recently
from
97a1d3b
to
aebe313
Compare
ulugbekna
force-pushed
the
protocol-v2
branch
4 times, most recently
from
6754219
to
02079f8
Compare
* make client, server capabilities management more explicit
* rename 'decoder_from' to more conventional 'of_string' * add safer version of peek pkt that also support git wire protocol v2 pkts: delim-pkt and response-end-pkt * move 'prompt_pkt' to 'decoder.mli' to reuse it in git wire proto v2 * add 'read_pkt' * add 'junk_chars' fn to 'Decoder' to increase 'decoder.pos' by 'n' * move 'bind' and '>>=' from nss/protocol.ml to pkt_line.decoder * [wip] support wire proto v2 capabilities * add 'Ls_refs', 'Fetch_command' modules to represent commands 'ls-refs' and 'fetch' respectively; * add 'Encoder' module to wire-proto-v2 with support for encoding command requests and copy-paste NSS's 'encode_proto_request' * add some comments to better define parts of a packet line: specific names for 4 bytes that encode packet length, the bytes that follow the length bytes, etc. * rename length calculating function 'pkt_len' to 'encoded_pkt_len' that returns the value hex-encoded in the first 4 bytes of the packet line and 'pkt_len_at_least_4' returns 'max 4 (encoded_pkt_len pkt)' * copy-paste 'Proto_request' module from NSS * update 'response' type in proto-v2 'Protocol' * add 'Extended_pkt_line_decoder' that provides more functionality than 'Pkt_line.Decoder' but not specific to the protocol * add decoding for all commands of wire proto v2 * reflect changes after 'mimic' lib introduction * make 'smart_flow' more understandable * reduce dup code, e.g., (>>=) * reorganize stuff closer to its use * rename stuff for more clarity * move smart protocol (wire proto v1)-based 'fetch' code to separate modules * functorize 'Smart_flow' * rename 'Smart_flow' to 'State_flow' * add mli file to 'State_flow' * improve 'nss/state.ml' API: - it improves cases when we want to "open" the module to get infix/syntax operators - it also make the API more uniform and rich, e.g., adds "map" fn * rename "fail" to "io_raise": 1) avoid clash with "fail" from "smart"/"wire_proto_v2" 2) to highlight that it causes "exception"al behavior * add support for "ls-refs" command (without args) * fix log.debug use and its message
a module that translates state read-write monad into 'flow' operations, to be usable both by 'Smart' and 'Wire_proto_v2'; hence, they shouldn't have own copies of Context
ulugbekna
force-pushed
the
protocol-v2
branch
from
02079f8
to
285dce3
Compare
dinosaure
reviewed
Feb 8, 2021
| let rec prompt_pkt ?strict k decoder = | ||
| if at_least_one_pkt decoder then k decoder | ||
| else prompt ?strict (prompt_pkt ?strict k) decoder | ||
|
|
There was a problem hiding this comment.
Not sure about such deletion, I know that usually, I re-bind an OCaml value with the same name several such as:
let prompt_pkt = ...
let prompt_pkt = ... prompt_pkt ...
let prompt_pkt = ... prompt_pkt ...
...So, in my mind, such deletion will disturb behaviors of functions defined then. Why you deleted this code?
There was a problem hiding this comment.
I moved prompt_pkt from protocol.ml to decoder.ml because
protocol.mlcontainsDecodermodule, which implements decoding for values of the pack protocol. Butprompt_pktis a function that is not specific to protocol and could be defined on a lower and more general layer such asdecoder.ml.- I also needed it in
proto_vals_v2.ml, so it made sense to move this general and low-level function todecoder.mlwhich specifically deals with pkt-line decoding - Moving
prompt_pkttodecoder.mlshould be safe because all references toprompt_pktinprotocol.mlstill refer to that value but "imported" fromdecoder.ml.
|
The pull-request has some style updates which should be out of that. I don't have strong opinion about style but they should be done with some others PRs (for example, I saw some update about |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is a WIP implementation of git wire protocol v2 (spec available here).
It's not really ready for a review, but if you still want to see how things are going, I'd recommend going from the oldest commit to the newest. Things should make sense.