Merge pull request #201 from networktocode/release-v2.2.1

Release v2.2.1

Author: chadell

.github/workflows/ci.yml

@@ -149,6 +149,8 @@ jobs:
         run: "echo RELEASE_VERSION=${GITHUB_REF:10} >> $GITHUB_ENV"
       - name: "Run Poetry Version"
         run: "poetry version $RELEASE_VERSION"
+      - name: "Run Poetry Build"
+        run: "poetry build"
       - name: "Upload binaries to release"
         uses: "svenstaro/upload-release-action@v2"
         with:
@@ -176,6 +178,8 @@ jobs:
         run: "echo RELEASE_VERSION=${GITHUB_REF:10} >> $GITHUB_ENV"
       - name: "Run Poetry Version"
         run: "poetry version $RELEASE_VERSION"
+      - name: "Run Poetry Build"
+        run: "poetry build"
       - name: "Push to PyPI"
         uses: "pypa/gh-action-pypi-publish@release/v1"
         with:

CHANGELOG.md

@@ -1,5 +1,18 @@
 # Changelog
 
+## v2.2.1 - 2023-01-17
+
+### Changed
+
+- #197 - Updated Equinix parser: Adding support for additional impact statement and notification types.
+- #192 - Updated Cogent parser: Adding subject and text parser.
+- #186 - Updated Telia Carrier as Arelion (while keeping Telia for backwards compatibility).
+
+### Fixed
+
+- #198 - Fixed Verizon parser: use European-style day-first date parsing
+- #187 - Fixed Zayo parser: adds chardet.detect method before decoding data_part.content.
+
 ## v2.2.0 - 2022-10-25
 
 ### Added

README.md

@@ -58,10 +58,10 @@ By default, there is a `GenericProvider` that support a `SimpleProcessor` using
 
 #### Supported providers using the BCOP standard
 
+- Arelion (previously Telia)
 - EuNetworks
 - NTT
 - PacketFabric
-- Telia
 - Telstra
 
 #### Supported providers based on other parsers

circuit_maintenance_parser/parsers/cogent.py

@@ -6,13 +6,145 @@
 from pytz import timezone, UTC
 from bs4.element import ResultSet  # type: ignore
 
-from circuit_maintenance_parser.parser import Html, Impact, CircuitImpact, Status
+from circuit_maintenance_parser.parser import CircuitImpact, EmailSubjectParser, Html, Impact, Status, Text
 
 logger = logging.getLogger(__name__)
 
 # pylint: disable=too-many-branches
 
 
+class SubjectParserCogent1(EmailSubjectParser):
+    """Subject parser for Cogent nofifications."""
+
+    def parse_subject(self, subject: str):
+        """Parse subject.
+
+        Example:
+        11/19/2022 Circuit Provider Maintenance - Edina, MN 1-300123456
+        Correction 06/11/2021 AB987654321-1 Planned Network Maintenance - San Jose, CA 1-123456789
+        """
+        data: Dict = {"circuits": []}
+
+        subject = subject.lower()
+
+        if subject.startswith("correction") or "rescheduled" in subject:
+            data["status"] = Status("RE-SCHEDULED")
+        elif "cancellation" in subject:
+            data["status"] = Status("CANCELLED")
+        elif "planned" in subject or "provider" in subject or "emergency" in subject:
+            data["status"] = Status("CONFIRMED")
+        elif "completed" in subject:
+            data["status"] = Status("COMPLETED")
+        else:
+            data["status"] = Status("NO-CHANGE")
+
+        match = re.search(r".* ([\d-]+)", subject)
+        if match:
+            circuit_id = match.group(1)
+            data["circuits"].append(CircuitImpact(impact=Impact("OUTAGE"), circuit_id=circuit_id.strip()))
+
+        return [data]
+
+
+class TextParserCogent1(Text):
+    """Parse text body of Cogent emails."""
+
+    def parse_text(self, text):
+        """Execute parsing of text.
+
+        Example:
+            CIRCUIT PROVIDER MAINTENANCE
+
+            Dear Cogent Customer,
+
+            As a valued customer, Cogent is committed to keeping you informed about any changes in the status of your service with us. This email is to alert you regarding a circuit provider maintenance which will affect your connection to Cogent:
+
+            Start time: 10:00pm CT 11/19/2022
+            End time: 5:00am CT 11/20/2022
+            Work order number: VN16123
+            Order ID(s) impacted: 1-300123456
+            Expected Outage/Downtime: 7 hours
+
+            Cogent customers receiving service in Edina, MN will be affected by this outage. This outage has been scheduled by Zayo. The purpose of this maintenance is to repair damaged fiber. Only the Cogent Order ID(s) above will be impacted.
+
+            During this maintenance window, you will experience an interruption in service while Zayo completes the maintenance activities; the interruption is expected to be less than 7 hours; however, due to the complexity of the work, your downtime may be longer.
+
+            Our network operations engineers closely monitor the work and will do everything possible to minimize any inconvenience to you.  If you have any problems with your connection after this time, or if you have any questions regarding the maintenance at any point, please call Customer Support at 1-877-7-COGENT and refer to this Maintenance Ticket: VN16123.
+
+        """
+        data = {
+            # "circuits": [],
+            "summary": "Cogent circuit maintenance",
+        }
+
+        lines = text.splitlines()
+
+        for line in lines:
+            if line.startswith("Dear"):
+                match = re.search(r"Dear (.*),", line)
+                if match:
+                    data["account"] = match.group(1)
+            elif line.startswith("Start time:"):
+                match = re.search(r"Start time: ([A-Za-z\d: ]*) [()A-Za-z\s]+ (\d+/\d+/\d+)", line)
+                if match:
+                    start_str = " ".join(match.groups())
+            elif line.startswith("End time:"):
+                match = re.search(r"End time: ([A-Za-z\d: ]*) [()A-Za-z\s]+ (\d+/\d+/\d+)", line)
+                if match:
+                    end_str = " ".join(match.groups())
+            elif line.startswith("Cogent customers receiving service"):
+                data["summary"] = line
+                match = re.search(r"[^Cogent].*?((\b[A-Z][a-z\s-]+)+, ([A-Za-z-]+[\s-]))", line)
+                if match:
+                    local_timezone = timezone(self._geolocator.city_timezone(match.group(1).strip()))
+
+                    # set start time using the local city timezone
+                    try:
+                        start = datetime.strptime(start_str, "%I:%M %p %d/%m/%Y")
+                    except ValueError:
+                        start = datetime.strptime(start_str, "%I:%M%p %d/%m/%Y")
+                    local_time = local_timezone.localize(start)
+                    # set start time to UTC
+                    utc_start = local_time.astimezone(UTC)
+                    data["start"] = self.dt2ts(utc_start)
+                    logger.info(
+                        "Mapped start time %s at %s (%s), to %s (UTC)",
+                        start_str,
+                        match.group(1).strip(),
+                        local_timezone,
+                        utc_start,
+                    )
+                    # set end time using the local city timezone
+                    try:
+                        end = datetime.strptime(end_str, "%I:%M %p %d/%m/%Y")
+                    except ValueError:
+                        end = datetime.strptime(end_str, "%I:%M%p %d/%m/%Y")
+                    local_time = local_timezone.localize(end)
+                    # set end time to UTC
+                    utc_end = local_time.astimezone(UTC)
+                    data["end"] = self.dt2ts(utc_end)
+                    logger.info(
+                        "Mapped end time %s at %s (%s), to %s (UTC)",
+                        end_str,
+                        match.group(1).strip(),
+                        local_timezone,
+                        utc_end,
+                    )
+            elif line.startswith("Work order number:"):
+                match = re.search("Work order number: (.*)", line)
+                if match:
+                    data["maintenance_id"] = match.group(1)
+            elif line.startswith("Order ID(s) impacted:"):
+                data["circuits"] = []
+                match = re.search(r"Order ID\(s\) impacted: (.*)", line)
+                if match:
+                    for circuit_id in match.group(1).split(","):
+                        data["circuits"].append(CircuitImpact(impact=Impact("OUTAGE"), circuit_id=circuit_id.strip()))
+            elif line.startswith("During this maintenance"):
+                data["summary"] = line
+        return [data]
+
+
 class HtmlParserCogent1(Html):
     """Notifications Parser for Cogent notifications."""
 

circuit_maintenance_parser/parsers/equinix.py

@@ -86,6 +86,8 @@ def _parse_b(self, b_elements, data):
                     impact = Impact.OUTAGE
                 elif "Loss of redundancy" in impact_line:
                     impact = Impact.REDUCED_REDUNDANCY
+                elif "Traffic will be re-routed" in impact_line:
+                    impact = Impact.REDUCED_REDUNDANCY
         return impact
 
     def _parse_table(self, theader_elements, data, impact):  # pylint: disable=no-self-use
@@ -97,7 +99,31 @@ def _parse_table(self, theader_elements, data, impact):  # pylint: disable=no-se
                         continue
                     circuit_info = list(tr_elem.find_all("td"))
                     if circuit_info:
-                        account, _, circuit = circuit_info  # pylint: disable=unused-variable
+                        if len(circuit_info) == 4:
+                            # Equinix Connect notifications contain the IBX name
+                            account, _, _, circuit = circuit_info  # pylint: disable=unused-variable
+                        elif len(circuit_info) == 14:
+                            # Equinix Fabric notifications include a lot of additional detail on seller and subscriber ID's
+                            (
+                                account,
+                                _,
+                                _,
+                                circuit,
+                                _,
+                                _,
+                                _,
+                                _,
+                                _,
+                                _,
+                                _,
+                                _,
+                                _,
+                                _,
+                            ) = circuit_info  # pylint: disable=unused-variable
+                        elif len(circuit_info) == 3:
+                            account, _, circuit = circuit_info  # pylint: disable=unused-variable
+                        else:
+                            return
                         data["circuits"].append(
                             {
                                 "circuit_id": circuit.text,

circuit_maintenance_parser/parsers/verizon.py

@@ -57,7 +57,7 @@ def parse_tables(self, tables: ResultSet, data: Dict):  # pylint: disable=too-ma
                 data["start"] = self.dt2ts(start)
                 data["end"] = self.dt2ts(end)
 
-        for row in circuit_table.find("tbody").find_all("tr"):
+        for row in circuit_table.find_all("tr"):
             cells = row.find_all("td")
             cells_text = [cell.string.strip() for cell in cells if cell.string]
             if not cells_text or cells_text[0].startswith("Company Name"):

circuit_maintenance_parser/provider.py

@@ -4,6 +4,7 @@
 import traceback
 
 from typing import Iterable, List, Dict
+import chardet
 
 from pydantic import BaseModel
 
@@ -19,7 +20,7 @@
 from circuit_maintenance_parser.parsers.aquacomms import HtmlParserAquaComms1, SubjectParserAquaComms1
 from circuit_maintenance_parser.parsers.aws import SubjectParserAWS1, TextParserAWS1
 from circuit_maintenance_parser.parsers.bso import HtmlParserBSO1
-from circuit_maintenance_parser.parsers.cogent import HtmlParserCogent1
+from circuit_maintenance_parser.parsers.cogent import HtmlParserCogent1, TextParserCogent1, SubjectParserCogent1
 from circuit_maintenance_parser.parsers.colt import CsvParserColt1, SubjectParserColt1, SubjectParserColt2
 from circuit_maintenance_parser.parsers.equinix import HtmlParserEquinix, SubjectParserEquinix
 from circuit_maintenance_parser.parsers.gtt import HtmlParserGTT1
@@ -95,7 +96,8 @@ def filter_check(filter_dict: Dict, data: NotificationData, filter_type: str) ->
             if filter_data_type not in filter_dict:
                 continue
 
-            data_part_content = data_part.content.decode().replace("\r", "").replace("\n", "")
+            data_part_encoding = chardet.detect(data_part.content).get("encoding", "utf-8")
+            data_part_content = data_part.content.decode(data_part_encoding).replace("\r", "").replace("\n", "")
             if any(re.search(filter_re, data_part_content) for filter_re in filter_dict[filter_data_type]):
                 logger.debug("Matching %s filter expression for %s.", filter_type, data_part_content)
                 return True
@@ -172,6 +174,14 @@ class AquaComms(GenericProvider):
     _default_organizer = "[email protected]"
 
 
+class Arelion(GenericProvider):
+    """Arelion (formerly Telia Carrier) provider custom class."""
+
+    _exclude_filter = {EMAIL_HEADER_SUBJECT: ["Disturbance Information"]}
+
+    _default_organizer = "[email protected]"
+
+
 class AWS(GenericProvider):
     """AWS provider custom class."""
 
@@ -195,6 +205,7 @@ class Cogent(GenericProvider):
 
     _processors: List[GenericProcessor] = [
         CombinedProcessor(data_parsers=[EmailDateParser, HtmlParserCogent1]),
+        CombinedProcessor(data_parsers=[EmailDateParser, TextParserCogent1, SubjectParserCogent1]),
     ]
     _default_organizer = "[email protected]"
 
@@ -308,12 +319,10 @@ class Sparkle(GenericProvider):
     _default_organizer = "[email protected]"
 
 
-class Telia(GenericProvider):
+class Telia(Arelion):
     """Telia provider custom class."""
 
-    _exclude_filter = {EMAIL_HEADER_SUBJECT: ["Disturbance Information"]}
-
-    _default_organizer = "[email protected]"
+    # Kept for compatibility purposes, but Telia is renamed Arelion
 
 
 class Telstra(GenericProvider):