Skip to content

feat: python native JSON format parser #53

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
Apr 17, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .github/workflows/test.yml
Original file line number Diff line number Diff line change
Expand Up @@ -32,4 +32,4 @@ jobs:
python -m pip install ".[test]"
- name: Run tests
run: |
python -m pytest
python -m pytest tests
3 changes: 3 additions & 0 deletions .gitmodules
Original file line number Diff line number Diff line change
@@ -1,3 +1,6 @@
[submodule "third_party/substrait"]
path = third_party/substrait
url = https://github.com/substrait-io/substrait
[submodule "third_party/substrait-cpp"]
path = third_party/substrait-cpp
url = https://github.com/substrait-io/substrait-cpp
4 changes: 2 additions & 2 deletions CONTRIBUTING.md
Original file line number Diff line number Diff line change
Expand Up @@ -21,14 +21,14 @@ git submodule update --init --recursive
```


# Upgrade the substrait submodule
# Upgrade the substrait protocol definition

## a) Use the upgrade script

Run the upgrade script to upgrade the submodule and regenerate the protobuf stubs.

```
./upgrade.sh <version>
./update_proto.sh <version>
```

## b) Manual upgrade
Expand Down
80 changes: 79 additions & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -135,6 +135,84 @@ relations {
}
```

## Load a Substrait Plan from JSON
A substrait plan can be loaded from its JSON representation
using the ``substrait.json.load_json`` and ``substrait.json.parse_json``
functions:

```
>>> import substrait.json
>>> jsontext = """{
... "relations":[
... {
... "root":{
... "input":{
... "read":{
... "baseSchema":{
... "names":[
... "first_name",
... "surname"
... ],
... "struct":{
... "types":[
... {
... "string":{
... "nullability":"NULLABILITY_REQUIRED"
... }
... },
... {
... "string":{
... "nullability":"NULLABILITY_REQUIRED"
... }
... }
... ]
... }
... },
... "namedTable":{
... "names":[
... "people"
... ]
... }
... }
... },
... "names":[
... "first_name"
... ]
... }
... }
... ]
... }"""
>>> substrait.json.parse_json(jsontext)
relations {
root {
input {
read {
base_schema {
names: "first_name"
names: "surname"
struct {
types {
string {
nullability: NULLABILITY_REQUIRED
}
}
types {
string {
nullability: NULLABILITY_REQUIRED
}
}
}
}
named_table {
names: "people"
}
}
}
names: "first_name"
}
}
```

## Produce a Substrait Plan with Ibis
Let's use an existing Substrait producer, [Ibis](https://ibis-project.org),
to provide an example using Python Substrait as the consumer.
Expand Down Expand Up @@ -280,4 +358,4 @@ version {
minor_number: 24
producer: "ibis-substrait"
}
```
```
1 change: 1 addition & 0 deletions pyproject.toml
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@ test = ["pytest >= 7.0.0"]

[tool.pytest.ini_options]
pythonpath = "src"
testpaths = "tests"

[build-system]
requires = ["setuptools>=61.0.0", "setuptools_scm[toml]>=6.2.0"]
Expand Down
25 changes: 25 additions & 0 deletions src/substrait/json.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,25 @@
from google.protobuf import json_format
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There are three Google ways of loading/saving protobuffers:

binary
JSON format
text format

It's worth considering how you want to handle the four different types (including the Substrait text format). The C++ package implements methods to read/write any.


from substrait.proto import Plan


def load_json(filename):
"""Load a Substrait Plan from a json file"""
with open(filename, encoding="utf-8") as f:
return parse_json(f.read())


def parse_json(text):
"""Generate a Substrait Plan from its JSON definition"""
return json_format.Parse(text=text, message=Plan())


def write_json(plan, filename):
"""Write a Substrait Plan to a json file"""
with open(filename, "w+") as f:
f.write(dump_json(plan))


def dump_json(plan):
"""Dump a Substrait Plan to a string in JSON format"""
return json_format.MessageToJson(plan)
72 changes: 72 additions & 0 deletions tests/test_json.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,72 @@
import os
import pathlib
import tempfile
import json

from substrait.proto import Plan
from substrait.json import load_json, parse_json, dump_json, write_json

import pytest


JSON_FIXTURES = (
pathlib.Path(os.path.dirname(__file__))
/ ".."
/ "third_party"
/ "substrait-cpp"
/ "src"
/ "substrait"
/ "textplan"
/ "data"
)
JSON_TEST_FILE = sorted(JSON_FIXTURES.glob("*.json"))
JSON_TEST_FILENAMES = [path.name for path in JSON_TEST_FILE]


@pytest.mark.parametrize("jsonfile", JSON_TEST_FILE, ids=JSON_TEST_FILENAMES)
def test_json_load(jsonfile):
with open(jsonfile) as f:
jsondata = _strip_json_comments(f)
parsed_plan = parse_json(jsondata)

# Save to a temporary file so we can test load_json
# on content stripped of comments.
with tempfile.TemporaryDirectory() as tmpdir:
# We use a TemporaryDirectory as on Windows NamedTemporaryFile
# doesn't allow for easy reopening of the file.
with open(pathlib.Path(tmpdir) / "jsonfile.json", "w+") as stripped_file:
stripped_file.write(jsondata)
loaded_plan = load_json(stripped_file.name)

# The Plan constructor itself will throw an exception
# in case there is anything wrong in parsing the JSON
# so we can take for granted that if the plan was created
# it is a valid plan in terms of protobuf definition.
assert type(loaded_plan) is Plan

# Ensure that when loading from file or from string
# the outcome is the same
assert parsed_plan == loaded_plan


@pytest.mark.parametrize("jsonfile", JSON_TEST_FILE, ids=JSON_TEST_FILENAMES)
def test_json_roundtrip(jsonfile):
with open(jsonfile) as f:
jsondata = _strip_json_comments(f)

parsed_plan = parse_json(jsondata)
assert parse_json(dump_json(parsed_plan)) == parsed_plan

# Test with write/load
with tempfile.TemporaryDirectory() as tmpdir:
filename = pathlib.Path(tmpdir) / "jsonfile.json"
write_json(parsed_plan, filename)
assert load_json(filename) == parsed_plan


def _strip_json_comments(jsonfile):
# The JSON files in the cpp testsuite are prefixed with
# a comment containing the SQL that matches the json plan.
# As Python JSON parser doesn't support comments,
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Technically none of the parsers support the comments including the json_format library.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Well, there are parsers that support comments, usually in the C/C++ style, but yes in general the specification had avoided comments to make sure they couldn't be abused to implement parser directives.

# we have to strip them to make the content readable
return "\n".join(l for l in jsonfile.readlines() if l[0] != "#")
1 change: 1 addition & 0 deletions third_party/substrait-cpp
Submodule substrait-cpp added at cc8d08
5 changes: 5 additions & 0 deletions update_cpp.sh
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
#!/bin/bash

echo "Updating substrait-cpp submodule..."
git submodule update --remote third_party/substrait-cpp

File renamed without changes.