[Refactor] Remove TinyDB (#468)

<!--
⚠️ If you do not respect this template, your pull request will be
closed.
⚠️ Your pull request title should be short detailed and understandable
for all.
⚠️ Also, please add a release note file using reno if the change needs
to be
  documented in the release notes.
⚠️ If your pull request fixes an open issue, please link to the issue.

- [ ] I have added the tests to cover my changes.
- [ ] I have updated the documentation accordingly.
- [ ] I have read the CONTRIBUTING document.
-->

### Summary

Removes TinyDB in favour of reading/writing from TOML files directly.

docs/project_overview.md

@@ -107,7 +107,7 @@ We store each member of the ecosystem as a TOML file under
 [`ecosystem/resources/members`](https://github.com/qiskit-community/ecosystem/blob/main/ecosystem/resources/members);
 these are the files you should edit when adding / updating members to the
 ecosystem. Access to this file is handled through the
-[`JsonDAO`](https://github.com/qiskit-community/ecosystem/blob/main/ecosystem/daos/jsondao.py)
+[`DAO`](https://github.com/qiskit-community/ecosystem/blob/main/ecosystem/daos/dao.py)
 class.
 
 The qiskit.org page pulls information from the compiled

ecosystem/daos/__init__.py

@@ -1,3 +1,3 @@
 """DAOs for ecosystem."""
 
-from .jsondao import JsonDAO
+from .dao import DAO

ecosystem/daos/dao.py

@@ -0,0 +1,257 @@
+"""
+DAO for json db.
+
+File structure:
+
+    root_path
+    ├── members.json  # compiled file; don't edit manually
+    └── members
+        └── repo-name.toml
+"""
+import json
+from pathlib import Path
+import shutil
+import toml
+
+from ecosystem.models import TestResult, StyleResult, CoverageResult, TestType
+from ecosystem.models.repository import Repository
+
+
+class TomlStorage:
+    """
+    Read / write TOML files from a dict where keys are repo URLs, and values
+    are Repository objects.
+
+    Can use as a context manager like so:
+
+    with TomlStorage() as data:  # Data is read from TOML files
+        data[url] = new_repo     # Mutate the data
+                                 # Changes are saved on exit
+    """
+
+    def __init__(self, root_path: str):
+        self.toml_dir = Path(root_path, "members")
+        self._data = None  # for use with context manager
+
+    def _url_to_path(self, url):
+        repo_name = url.strip("/").split("/")[-1]
+        return self.toml_dir / f"{repo_name}.toml"
+
+    def read(self) -> dict:
+        """
+        Search for TOML files and read into dict with types:
+        { url (str): repo (Repository) }
+        """
+        data = {}
+        for path in self.toml_dir.glob("*"):
+            repo = Repository.from_dict(toml.load(path))
+            data[repo.url] = repo
+        return data
+
+    def write(self, data: dict):
+        """
+        Dump everything to TOML files from dict of types
+        { key (any): repo (Repository) }
+        """
+        # Erase existing TOML files
+        # (we erase everything to clean up any deleted repos)
+        if self.toml_dir.exists():
+            shutil.rmtree(self.toml_dir)
+
+        # Write to human-readable TOML
+        self.toml_dir.mkdir()
+        for repo in data.values():
+            with open(self._url_to_path(repo.url), "w") as file:
+                toml.dump(repo.to_dict(), file)
+
+    def __enter__(self) -> dict:
+        self._data = self.read()
+        return self._data
+
+    def __exit__(self, _type, _value, exception):
+        if exception is not None:
+            raise exception
+        self.write(self._data)
+
+
+class DAO:
+    """
+    Data access object for repository database.
+    """
+
+    def __init__(self, path: str):
+        """
+        Args:
+            path: path to store database in
+        """
+        self.storage = TomlStorage(path)
+        self.labels_json_path = Path(path, "labels.json")
+        self.compiled_json_path = Path(path, "members.json")
+
+    def write(self, repo: Repository):
+        """
+        Update or insert repo (identified by URL).
+        """
+        self.update_labels(repo.labels)
+        with self.storage as data:
+            data[repo.url] = repo
+
+    def get_repos_by_tier(self, tier: str) -> list[Repository]:
+        """
+        Returns all repositories in specified tier.
+
+        Args:
+            tier: tier of the repo (MAIN, COMMUNITY, ...)
+        """
+        matches = [repo for repo in self.storage.read().values() if repo.tier == tier]
+        return matches
+
+    def delete(self, repo_url: str):
+        """Deletes repository from tier.
+
+        Args:
+            repo_url: repository url
+        """
+        with self.storage as data:
+            del data[repo_url]
+
+    def get_by_url(self, url: str) -> Repository:
+        """
+        Returns repository by URL.
+        """
+        data = self.storage.read()
+        if url not in data:
+            raise KeyError(f"No repo with URL '{url}'")
+        return self.storage.read()[url]
+
+    def update(self, repo_url: str, **kwargs):
+        """
+        Update attributes of repository.
+
+        Args:
+            repo_url (str): URL of repo
+            kwargs: Names of attributes and new values
+
+        Example usage:
+            update("github.com/qiskit/qiskit, name="qiskit", stars=300)
+        """
+        with self.storage as data:
+            for arg, value in kwargs.items():
+                data[repo_url].__dict__[arg] = value
+
+    def update_labels(self, labels: list[str]):
+        """
+        Updates labels file for consumption by qiskit.org.
+        """
+        with open(self.labels_json_path, "r") as labels_file:
+            existing_labels = {
+                label["name"]: label["description"] for label in json.load(labels_file)
+            }
+
+        merged = {**{l: "" for l in labels}, **existing_labels}
+        new_label_list = [
+            {"name": name, "description": dsc} for name, dsc in merged.items()
+        ]
+        with open(self.labels_json_path, "w") as labels_file:
+            json.dump(
+                sorted(new_label_list, key=lambda x: x["name"]), labels_file, indent=4
+            )
+
+    def compile_json(self):
+        """
+        Dump database to JSON file for consumption by qiskit.org
+        Needs this structure:
+
+        { tier: {  # e.g. Main, Community
+            index: repo  # `repo` is data from repo-name.toml
+        }}
+
+        """
+        data = self.storage.read()
+
+        out = {}
+        for repo in data.values():
+            if repo.tier not in out:
+                out[repo.tier] = {}
+            index = str(len(out[repo.tier]))
+            out[repo.tier][index] = repo.to_dict()
+
+        with open(self.compiled_json_path, "w") as file:
+            json.dump(out, file, indent=4)
+
+    def add_repo_test_result(self, repo_url: str, test_result: TestResult):
+        """
+        Adds test result to repository.
+        Overwrites the latest test results and adds to historical test results.
+
+        Args:
+            repo_url: url of the repo
+            test_result: TestResult from the tox -epy3.x
+        """
+        repo = self.get_by_url(repo_url)
+
+        # add new result and remove old from list
+        new_test_results = [
+            tr for tr in repo.tests_results if tr.test_type != test_result.test_type
+        ] + [test_result]
+
+        # add last working version
+        if test_result.test_type == TestType.STABLE_COMPATIBLE and test_result.passed:
+            last_stable_test_result = TestResult(
+                passed=True,
+                test_type=TestType.LAST_WORKING_VERSION,
+                package=test_result.package,
+                package_version=test_result.package_version,
+                logs_link=test_result.logs_link,
+            )
+            new_test_results_with_latest = [
+                tr
+                for tr in new_test_results
+                if tr.test_type != last_stable_test_result.test_type
+            ] + [last_stable_test_result]
+            new_test_results = new_test_results_with_latest
+
+        repo.tests_results = sorted(new_test_results, key=lambda r: r.test_type)
+
+        new_historical_test_results = [
+            tr
+            for tr in repo.historical_test_results
+            if tr.test_type != test_result.test_type
+            or tr.qiskit_version != test_result.qiskit_version
+        ] + [test_result]
+        repo.historical_test_results = new_historical_test_results
+        self.write(repo)
+
+    def add_repo_style_result(self, repo_url: str, style_result: StyleResult):
+        """
+        Adds style result for repository.
+
+        Args:
+            repo_url: url of the repo
+            style_result: StyleResult from the tox -elint
+        """
+        repo = self.get_by_url(repo_url)
+
+        new_style_results = [
+            tr for tr in repo.styles_results if tr.style_type != style_result.style_type
+        ] + [style_result]
+        repo.styles_results = new_style_results
+        self.write(repo)
+
+    def add_repo_coverage_result(self, repo_url: str, coverage_result: CoverageResult):
+        """
+        Adds coverage result for repository.
+
+        Args:
+            repo_url: url of the repo
+            coverage_result: CoverageResult from the tox -ecoverage
+        """
+        repo = self.get_by_url(repo_url)
+
+        new_coverage_results = [
+            tr
+            for tr in repo.coverages_results
+            if tr.coverage_type != coverage_result.coverage_type
+        ] + [coverage_result]
+        repo.coverages_results = new_coverage_results
+        self.write(repo)