Import the metadata parser workflow

Author: ArtyomVancyan

.github/workflows/update.yml

@@ -0,0 +1,40 @@
+name: Update Validation Patterns
+
+on:
+  schedule:
+    - cron: "0 0 * * 0"
+
+jobs:
+  update:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v2
+
+      - name: Setup Python
+        uses: actions/setup-python@v3
+
+      - name: Fetch third-party metadata
+        run: |
+          curl -o resources/metadata.xml https://raw.githubusercontent.com/google/libphonenumber/master/resources/PhoneNumberMetadata.xml
+
+      - name: Update metadata
+        run: |
+          python scripts/prepare-metadata
+
+      - name: Compare metadata
+        id: compare
+        run: |
+          git diff --exit-code resources/metadata.xml || echo "::set-output name=differs::true"
+
+      - name: Create Pull Request
+        if: steps.compare.outputs.differs == 'true'
+        uses: peter-evans/create-pull-request@v5
+        with:
+          token: ${{ secrets.GH_TOKEN }}
+          branch-suffix: short-commit-hash
+          branch: update-validation-patterns
+          title: Update the validation patterns
+          commit-message: Update the validation patterns
+          body: This PR updates the validation patterns i.e. metadata.xml to keep the repository up-to-date with the upstream changes.

scripts/prepare-metadata/__main__.py

@@ -0,0 +1,61 @@
+import json
+import re
+from pathlib import Path
+from xml.etree import ElementTree
+
+project_root = Path(__file__).parent.parent.parent
+metadata_path = project_root / "resources" / "metadata.xml"
+patterns_path = project_root / "src" / "metadata" / "validations.json"
+countries_path = project_root / "src" / "metadata" / "countries.json"
+
+tree = ElementTree.parse(metadata_path)
+territories = tree.find("territories")
+
+
+def update_mask(mask, length):
+    length_without_cc = len(re.findall(r"[\d.]", re.sub(r"\+\d+\s", "", mask)))
+    if length_without_cc != length:
+        cc_mask = re.match(r"(\+\d+)\s", mask).group(1)
+        ac_mask = re.match(r"\+\d+\s(?:(\S+)\s)?", mask).group(1) or ""
+        ac_mask_length = len(re.findall(r"[\d.]", ac_mask))
+        pn_mask_expected_length = length - ac_mask_length
+        pn_mask_prefix = re.sub(r"\+\d+\s(?:\S+\s)?", "", mask).split()[0]
+        pn_mask_parts = [pn_mask_prefix]
+        while len("".join(pn_mask_parts)) < pn_mask_expected_length:
+            pn_mask_parts.append("." * len(pn_mask_prefix))
+        pn_mask = " ".join(pn_mask_parts)
+        while len(pn_mask.replace(" ", "")) > pn_mask_expected_length:
+            pn_mask = pn_mask[:-1]
+        mask = f"{cc_mask} {ac_mask} {pn_mask}" if ac_mask else f"{cc_mask} {pn_mask}"
+    return re.sub(r"\s(\.{1,2})$", r"\1", mask)
+
+
+with open(patterns_path) as fp:
+    patterns = json.load(fp)
+
+with open(countries_path) as fp:
+    countries = json.load(fp)
+
+for territory in territories:
+    # Regenerate masks based on possible maximum lengths
+    possible_lengths = map(lambda e: territory.find(f"{e.tag}/possibleLengths"), territory.iter())
+    possible_lengths = map(lambda e: e.get("national"), filter(lambda e: e is not None, possible_lengths))
+    possible_lengths = list(map(int, re.findall(r"\d+", ",".join(possible_lengths))))
+    min_length, max_length = min(possible_lengths), max(possible_lengths)
+    for country in [c for c in countries if c[0] == territory.get("id").lower()]:
+        country[3] = update_mask(country[3], max_length)
+
+    # Update phone validation patterns
+    general_desc = territory.find("generalDesc")
+    national_number_pattern = general_desc.find("nationalNumberPattern").text
+    national_number_pattern = re.sub(r"[\s\n]", "", national_number_pattern)
+    patterns[territory.get("id").lower()] = [
+        f"^\\d{{{min_length},{max_length}}}$" if min_length != max_length else f"^\\d{{{max_length}}}$",
+        f"^{national_number_pattern}$"
+    ]
+
+with open(patterns_path, "w") as fp:
+    json.dump(patterns, fp, indent=2)
+
+with open(countries_path, "w") as fp:
+    json.dump(countries, fp, indent=2)