email-forward-parser 0.2.0__tar.gz → 0.2.1__tar.gz

email_forward_parser-0.2.1/.githooks/pre-push

@@ -0,0 +1,93 @@
+#!/bin/sh
+# Style and type checks on the commits being pushed; GitHub Actions runs the tests.
+# Turn it on once per clone: git config core.hooksPath .githooks
+
+fix='uv run black . && uv run isort . && uv run ruff check --fix .'
+
+# The same install as CI's "Install locked dependencies".
+setup() {
+    uv sync --frozen || return
+}
+
+checks() {
+    check uv run black --check .
+    check uv run isort --check-only .
+    check uv run ruff check .
+    check uv run mypy
+}
+
+# --- the same in every repo from here ---
+
+set -u
+unset GIT_DIR GIT_WORK_TREE GIT_INDEX_FILE
+common=$(cd "$(git rev-parse --git-common-dir)" && pwd)
+checked="$common/pre-push-checked"
+# A commit passes for this version of the hook: changing the checks checks it again.
+version=$(cksum <"$0" | cut -d' ' -f1)
+# One checkout per clone, kept between pushes so its dependencies and tool caches stay installed.
+cache="${XDG_CACHE_HOME:-$HOME/.cache}/pre-push"
+dir="$cache/$(basename "$(dirname "$common")")-$(printf %s "$common" | cksum | cut -d' ' -f1)"
+lock="$dir.lock"
+
+# Run one check and carry on, so a push shows every failure at once.
+check() {
+    "$@" || failed=1
+}
+
+# Another push in this clone may be using the checkout: wait for it, or take over a lock its push left behind.
+take_lock() {
+    mkdir -p "$cache"
+    until mkdir "$lock" 2>/dev/null; do
+        owner=$(cat "$lock/pid" 2>/dev/null)
+        if [ -n "$owner" ] && ! kill -0 "$owner" 2>/dev/null; then
+            rm -rf "$lock"
+            continue
+        fi
+        echo "pre-push: waiting for another push's checks to finish" >&2
+        sleep 5
+    done
+    echo $$ >"$lock/pid"
+}
+trap '[ "$(cat "$lock/pid" 2>/dev/null)" = $$ ] && rm -rf "$lock"' EXIT
+trap 'exit 1' INT TERM
+
+# Check out the commit itself, not the working tree: uncommitted files don't go out.
+checkout() {
+    if [ -f "$dir/.git" ] && git -C "$dir" checkout --quiet --force --detach "$1" 2>/dev/null; then
+        git -C "$dir" clean -fdq
+        return
+    fi
+    # First push, or the folder or its registration went missing: start it again.
+    rm -rf "$dir"
+    git worktree add --quiet --force --detach "$dir" "$1"
+}
+
+status=0
+while read -r _ sha _ _; do
+    # A deleted ref sends all zeros, which isn't a commit.
+    commit=$(git rev-parse --verify --quiet "$sha^{commit}") || continue
+    # Pushing the same commit to the next remote doesn't check it again.
+    grep -qsx "$commit $version" "$checked" && continue
+
+    [ -d "$lock" ] && [ "$(cat "$lock/pid" 2>/dev/null)" = $$ ] || take_lock
+    short=$(git rev-parse --short "$commit")
+    if ! checkout "$commit"; then
+        echo "pre-push: couldn't check out $short in $dir" >&2
+        status=1
+        continue
+    fi
+    if ! (cd "$dir" && setup) </dev/null >&2; then
+        echo "pre-push: installing the dependencies of $short failed" >&2
+        status=1
+        continue
+    fi
+    if (cd "$dir" && failed=0 && checks && exit $failed) </dev/null; then
+        echo "$commit $version" >>"$checked"
+    else
+        echo >&2
+        echo "pre-push: checks failed on $short. Style fixes itself with: $fix" >&2
+        echo "pre-push: type errors need a code change. Commit the fix and push again (git push --no-verify skips the checks)." >&2
+        status=1
+    fi
+done
+exit $status

{email_forward_parser-0.2.0 → email_forward_parser-0.2.1}/.github/dependabot.yml

@@ -12,8 +12,14 @@ updates:
       - "dependencies"
       - "python"
 
+  # The actions in .github/workflows are pinned to commit SHAs; this keeps the SHAs and their version comments
+  # current, in one pull request a week.
   - package-ecosystem: "github-actions"
     directory: "/"
+    groups:
+      github-actions:
+        patterns:
+          - "*"
     schedule:
       interval: "weekly"
       day: "monday"

{email_forward_parser-0.2.0 → email_forward_parser-0.2.1}/.github/workflows/ci.yml

@@ -11,26 +11,21 @@ concurrency:
   group: ci-${{ github.workflow }}-${{ github.ref }}
   cancel-in-progress: true
 
+# Style and type checks run before each push, in .githooks/pre-push; this workflow runs the rest.
+# Runners are pinned to ubuntu-24.04 (ubuntu-latest moves to Ubuntu 26 from 2026-10-19) and actions to commit
+# SHAs, which Dependabot keeps current.
 jobs:
   verify:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
     timeout-minutes: 30
     steps:
-      - uses: actions/checkout@v7
+      - uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
       - name: Install uv and Python
-        uses: astral-sh/setup-uv@v10.0.1
+        uses: astral-sh/setup-uv@1c37ad07a6a961277cf70c0d37d6f313000f5884 # v10.3.0
         with:
           python-version: "3.14"
           enable-cache: true
       - name: Install locked dependencies
         run: uv sync --frozen
-      - name: Check formatting
-        run: |
-          uv run black --check .
-          uv run isort --check-only .
-      - name: Lint
-        run: uv run ruff check .
-      - name: Type check
-        run: uv run mypy
       - name: Test with configured coverage gate
         run: uv run pytest

{email_forward_parser-0.2.0 → email_forward_parser-0.2.1}/.github/workflows/workflow.yml

@@ -11,35 +11,36 @@ permissions:
 jobs:
   verify:
     name: Verify package
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
+    timeout-minutes: 30
     steps:
-      - uses: actions/checkout@v7
-      - uses: astral-sh/setup-uv@v10.0.1
+      - uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
+      - name: Install uv and Python
+        uses: astral-sh/setup-uv@1c37ad07a6a961277cf70c0d37d6f313000f5884 # v10.3.0
         with:
           python-version: "3.14"
           enable-cache: true
-      - run: uv sync --frozen
+      - name: Install locked dependencies
+        run: uv sync --frozen
       - name: Verify tag matches package version
         run: |
           PACKAGE_VERSION=$(python -c 'import tomllib; print(tomllib.load(open("pyproject.toml", "rb"))["project"]["version"])')
           test "v${PACKAGE_VERSION}" = "${GITHUB_REF_NAME}"
-      - run: uv run black --check .
-      - run: uv run isort --check-only .
-      - run: uv run ruff check .
-      - run: uv run mypy
-      - run: uv run pytest
+      - name: Test with configured coverage gate
+        run: uv run pytest
 
   build:
     name: Build distribution
     needs: verify
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
+    timeout-minutes: 10
 
     steps:
       - name: Checkout
-        uses: actions/checkout@v7
+        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
 
       - name: Set up Python
-        uses: actions/setup-python@v7
+        uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
         with:
           python-version: "3.14"
 
@@ -53,7 +54,7 @@ jobs:
         run: python -m twine check dist/*
 
       - name: Upload distribution artifacts
-        uses: actions/upload-artifact@v7
+        uses: actions/upload-artifact@cf430e030ddbb5b0abf93d22962f4752f3646cd9 # v7.0.2
         with:
           name: python-package-distributions
           path: dist/
@@ -61,7 +62,8 @@ jobs:
   publish:
     name: Publish distribution to PyPI
     needs: build
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
+    timeout-minutes: 10
     environment: pypi
 
     permissions:
@@ -70,32 +72,33 @@ jobs:
 
     steps:
       - name: Download distribution artifacts
-        uses: actions/download-artifact@v8
+        uses: actions/download-artifact@9000827ccba6bdab643e8b6fd33ac0654aef8333 # v8.0.2
         with:
           name: python-package-distributions
           path: dist/
 
       - name: Publish package
-        uses: pypa/gh-action-pypi-publish@release/v1
+        uses: pypa/gh-action-pypi-publish@dc37677b2e1c63e2034f94d8a5b11f265b73ba33 # v1.14.2
 
   release:
     name: Publish GitHub release
     needs: publish
     if: startsWith(github.ref, 'refs/tags/')
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
+    timeout-minutes: 10
 
     permissions:
       contents: write
 
     steps:
       - name: Download distribution artifacts
-        uses: actions/download-artifact@v8
+        uses: actions/download-artifact@9000827ccba6bdab643e8b6fd33ac0654aef8333 # v8.0.2
         with:
           name: python-package-distributions
           path: dist/
 
       - name: Publish release
-        uses: softprops/action-gh-release@v3
+        uses: softprops/action-gh-release@efb35369e0ad2afab669f228072c1b0d510eae64 # v3.0.3
         with:
           files: dist/*
           generate_release_notes: true

{email_forward_parser-0.2.0 → email_forward_parser-0.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: email-forward-parser
-Version: 0.2.0
+Version: 0.2.1
 Summary: Python Library to parse forwarded email messages
 Project-URL: Homepage, https://github.com/felipehertzer/email-forward-parser-py
 Project-URL: Repository, https://github.com/felipehertzer/email-forward-parser-py

{email_forward_parser-0.2.0 → email_forward_parser-0.2.1}/emailforwardparser/client.py

@@ -4,6 +4,8 @@ import base64
 import binascii
 import logging
 from email import policy
+from email.errors import HeaderParseError
+from email.header import decode_header, make_header
 from email.message import EmailMessage, Message
 from email.mime.multipart import MIMEMultipart
 from email.mime.text import MIMEText
@@ -38,7 +40,7 @@ class EmailParserClient:
         :rtype: dict
         """
         msg = self._parse_message(email)
-        send_to = self._first_header_address(msg.get("From", ""))
+        send_to = self._first_header_address(self._header(msg, "From"))
         eml = self._get_eml_attachment(msg)
         if eml:
             return self._get_dict(eml, fp.OriginalMetadata(), False, send_to=send_to)
@@ -107,9 +109,20 @@ class EmailParserClient:
             )
             eml = self._build_original_email(email, source_message).as_string()
         else:
-            eml = message.as_string() if isinstance(message, Message) else message
+            eml = self._as_string(message) if isinstance(message, Message) else message
         return {"forward": forwarded, "eml": eml, "Send-To": send_to}
 
+    @staticmethod
+    def _as_string(message: Message) -> str:
+        """The message as text. A header longer than a line is folded again as it is written,
+        which builds its mailboxes again and refuses (ValueError) one whose encoded-word name
+        decodes to a line break: the message is then written with its headers as they came.
+        """
+        try:
+            return message.as_string()
+        except ValueError:
+            return message.as_string(policy=policy.default.clone(refold_source="none"))
+
     def _build_original_email(
         self, metadata: fp.OriginalMetadata, message: Message
     ) -> EmailMessage | MIMEMultipart:
@@ -176,8 +189,7 @@ class EmailParserClient:
 
     def _get_forwarded_metadata(self, message: Message) -> fp.ForwardMetadata:
         body = self._get_body(message)
-        subject = message.get("Subject")
-        subject = str(subject) if subject is not None else ""
+        subject = self._header(message, "Subject")
         if subject:
             return fp.get_forwarded_metadata(body.strip(), subject.strip())
         return fp.get_forwarded_metadata(body.strip())
@@ -233,16 +245,39 @@ class EmailParserClient:
         return Parser(policy=policy.default).parsestr(email)
 
     def _get_message_metadata(self, message: Message) -> fp.OriginalMetadata:
-        from_mailboxes = self._mailboxes_from_header(message.get("From", ""))
+        from_mailboxes = self._mailboxes_from_header(self._header(message, "From"))
         return fp.OriginalMetadata(
-            date=str(message.get("Date", "")).strip(),
-            subject=str(message.get("Subject", "")).strip(),
+            date=self._header(message, "Date").strip(),
+            subject=self._header(message, "Subject").strip(),
             body=self._get_body(message).strip(),
             from_=from_mailboxes[0] if from_mailboxes else fp.MailboxResult(),
-            to=self._mailboxes_from_header(message.get("To", "")),
-            cc=self._mailboxes_from_header(message.get("Cc", "")),
+            to=self._mailboxes_from_header(self._header(message, "To")),
+            cc=self._mailboxes_from_header(self._header(message, "Cc")),
         )
 
+    @staticmethod
+    def _header(message: Message, name: str) -> str:
+        """The header's decoded value, or "" when the message has none.
+
+        The email package builds an address header's mailboxes as it reads it, and refuses
+        (ValueError) one whose encoded-word name decodes to a line break, so reading such a
+        From raised out of the parser. Its mailboxes are read from the raw value instead,
+        each name decoded with its line breaks as spaces.
+        """
+        try:
+            value = message.get(name)
+        except ValueError:
+            raw = next(str(raw) for key, raw in message.raw_items() if key.lower() == name.lower())
+            mailboxes = []
+            for encoded_name, address in getaddresses([raw]):
+                try:
+                    decoded = str(make_header(decode_header(encoded_name)))
+                except HeaderParseError, LookupError, UnicodeError:
+                    decoded = encoded_name
+                mailboxes.append(formataddr((" ".join(decoded.split()), address)))
+            return ", ".join(mailboxes)
+        return str(value) if value is not None else ""
+
     def _first_header_address(self, header: object) -> str:
         mailboxes = self._mailboxes_from_header(header)
         if not mailboxes:

{email_forward_parser-0.2.0 → email_forward_parser-0.2.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "email-forward-parser"
-version = "0.2.0"
+version = "0.2.1"
 description = "Python Library to parse forwarded email messages"
 readme = "README.md"
 requires-python = ">=3.14"

{email_forward_parser-0.2.0 → email_forward_parser-0.2.1}/tests/regressions_test.py

@@ -193,3 +193,26 @@ def test_loop_regexes_match_keeps_earliest_then_first_listed() -> None:
     assert loop.loop_regexes_match([earlier, later], "abc") == (["b"], earlier)
     assert loop.loop_regexes_match([later, earlier], "abc") == (["b"], earlier)
     assert loop.loop_regexes_match([earlier, tie], "bc") == (["b"], earlier)
+
+
+def test_a_from_whose_encoded_name_decodes_to_a_line_break_is_read_without_raising() -> None:
+    """The email package refuses to build an address with a line break in it (ValueError),
+    and raised it out of get_original_metadata, failing whoever parsed the email."""
+    raw = (
+        "From: =?utf-8?q?Jane=0D=0ABcc=3A_evil=40example=2Ecom?= <jane@example.com>\n"
+        "To: =?utf-8?q?Bob=0A?= <bob@example.com>, =?x-unknown?q?Ann?= <ann@example.com>\n"
+        "Subject: Hello\n\nNo forward here.\n"
+    )
+    client = EmailParserClient()
+
+    metadata = client.get_original_metadata(raw)
+    assert metadata.forwarded is False
+    assert metadata.email.from_ == MailboxResult("Jane Bcc: evil@example.com", "jane@example.com")
+    # A name in a charset Python has no codec for stays as it was written.
+    assert metadata.email.to == [
+        MailboxResult("Bob", "bob@example.com"),
+        MailboxResult("=?x-unknown?q?Ann?=", "ann@example.com"),
+    ]
+    original = client.get_original_eml(raw)
+    assert original["Send-To"] == "jane@example.com"
+    assert original["eml"].startswith(raw.splitlines()[0])