Ukwakha i-Crash Report Automation ye-iOS kunye ne-Android

Hi, ndiye Vilian Iaumbaev! Ndivela ngexesha elidlulileyo isisombululo sokukhusela ezintsha kwi-iOS kunye ne-Android - yenza ukucacisa kunye nokuguqulwa kwimibuzo kakhulu kakhulu. Yintoni siyenze oku. Ukuhanjiswa kwama-crash kwi-developer eyenziwe ngempumelelo yaba ngempumelelo kwaye engabonakaliweyo. Kukho lula ukufumana into, ukufikelela kwimeko ze-edge, okanye ukufikelela kwimeko ezininzi ezininzi. Ufuna ukwenza le nkqubo enokufanelekileyo kunye ne-systematic - ukuze akukho umntu kwi-team akufuneka ukuchitha ixesha kwi-crashes, kwaye akukho ingxaki ebalulekileyo iyafumaneka kwi-crackers. Ukucaciswa Ukuqala, uya kufuneka izixhobo ze-Google ezifana Ukucinga Kwaye Jira. Ukucinga I-Google Cloud Platform Ukucinga I-Google Cloud Platform Emva kokufaka iiprojekthi yakho kwiinkonzo ze-Google, ukuguqulela ukuhambisa idatha kwi-Crashlytics kwi-GCP usebenzisa i-Google Emva koko, zonke iingxelo ze-crash ziyafumaneka kwi-BigQuery tables. inkxaso page inkxaso page I-crash data structures ye-iOS kunye ne-Android ziyafumaneka, kunye neentlungu ezincinane ezincinane - nto leyo kuthetha ukuba usebenzise i-script eyodwa ukuxhaswa ngamnye. Kwaye ngoku unayo iingcaciso zakho kwi-BigQuery leyo kuthetha ukuba unako ukuvelisa umsebenzi kwiinkcukacha zayo. Unako ukhangela zonke iinkcukacha zeengcaciso kwaye usebenzise njengoko ufuna kwi-akhawunti yakho. Okokuqala, kufuneka ukufumana zonke iinkcukacha i-crashes ukuba i-analyzied kodwa ukuba unayo iinkcukacha ezininzi kwiiyunithi ezingaphezu kwama-2000,000 abasebenzisi kufuneka uqhagamshelane zonke iinkcukacha kwi-Google, ukwenza ezinye iinkcukacha. Ukucinga Ukufundisa ezinye i-SQL ezisemgangathweni ukufumana iinkcukacha ze-crash kwi-BigQuery Qhubeka iinkcukacha nge-Python Yenza zonke i-committers kwi-repository kunye ne-fuse duplicates Ukubhalisa zonke iingxaki kwi-repo file kunye neengcali yayo Yenza umsebenzi ye-Jira ye-file owner, ukuba isicelo esikho kwakhona Ukufundisa ezinye iziseko ze-SQL ukufumana idatha evela kwi-BigQuery I-BigQuery isetyenziselwa i-SQL ye-dialect yayo, efana ne-SQL ye-standard kodwa inikeza ukuncedisa kwedatha. Ukusetyenziswa kwethu, sisebenza kunye ne-crasch data set epheleleyo, kodwa kwindlela ye-aggregated. Ngokutsho, sisetyenziswe i-crasch reports ezilodwa kwi-crasch signatures kwaye isetyenziswe iinkcukacha ezinxulumeneyo kwinqanaba ngamnye - njenge-number of occurrences, user count affected, version breakdown, kunye nezinye iinkcukacha. Uyakwazi ukufumana i-SQL script apha kwaye utshintshe kwi-environment yakho nge-link elandelayo: https://console.cloud.google.com/bigquery https://console.cloud.google.com/bigquery WITH pre as( SELECT issue_id, ARRAY_AGG(DISTINCT issue_title IGNORE NULLS) as issue_titles, ARRAY_AGG(DISTINCT blame_frame.file IGNORE NULLS) as blame_files, ARRAY_AGG(DISTINCT blame_frame.library IGNORE NULLS) as blame_libraries, ARRAY_AGG(DISTINCT blame_frame.symbol IGNORE NULLS) as blame_symbols, COUNT(DISTINCT event_id) as total_events, COUNT(DISTINCT installation_uuid) as total_users, '{"version":"' || application.display_version || '","events":' || COUNT(DISTINCT event_id) || ',"users":' || COUNT(DISTINCT installation_uuid) || '}' AS events_info FROM `YOUR TABLE NAME` WHERE 1=1 AND event_timestamp >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 24 HOUR) AND event_timestamp < CURRENT_TIMESTAMP() AND error_type = "FATAL" GROUP BY issue_id, application.display_version ) SELECT issue_id, ARRAY_CONCAT_AGG(issue_titles) as issue_titles, ARRAY_CONCAT_AGG(blame_files) as blame_files, ARRAY_CONCAT_AGG(blame_libraries) as blame_libraries, ARRAY_CONCAT_AGG(blame_symbols) as blame_symbols, SUM(total_events) as total_events, SUM(total_users) as total_users, '[' || STRING_AGG(events_info, ",") || ']' as events_info FROM pre WHERE 1=1 AND issue_id IS NOT NULL AND events_info IS NOT NULL GROUP BY issue_id ORDER BY total_users DESC Ngenxa yoko, uya kufumana umgca omnye ngalinye ye- issue_id eyodwa, kunye nemimeko ezilandelayo ezihlangeneyo: issue_titles — umxholo we zonke iidilesi ze-crash. Le mveliso ifumaneka iimeko apho iidilesi ezininzi eziqhelekileyo ziquka kwimeko efanayo. Kwi-scripting, siyaziqhelekileyo. i-blame_files — umxholo weefayile ze-stacktrace eyinxalenye kwi-crash. Oku kunokwenzeka ukuba i-crash ifumaneka kwi-codebase yakho (ukuba kwi-system libraries). i-blame_libraries — umxholo weibhayisikili ezihambelana ne-crash. Oku kubandakanyeka kwisiqingatha, eyakhelwe ngenxa yeemeko efana ne- issue_titles. i-blame_symbols — umxholo wesixhobo (i-functions/methods) apho i-crash iye ifumaneka. Njengezinto ezininzi ezidlulileyo, i-array. total_events – Inani lwezinto zokusetyenziswa ngexesha elidlulileyo. total_users — inani lomsebenzisi eyenziwe. Kwiimeko i-crash kunokwenzeka kuphela kwigumbi elifanayo lomsebenzisi. events_info — i-array ye-JSON (njenge-string) ebandakanya i-total_events kunye ne-total_users ezahlukileyo ngexesha le-app. [ { "version": "1.0.1", "events": 131, "users": 110 }, { "version": "1.2.1", "events": 489, "users": 426 } ] Qinisekisa iinkcukacha kwi-BigQuery usebenzisa iPython Ukuqala ukufaka i-BigQuery Python client library Emva kokufaka, ukwakha ifayile BigQueryExecutor.py - le module uya kusebenza yonke ukuxhumana kunye Google Cloud BigQuery. Ukucinga Ukucinga import os import json import tempfile from google.oauth2 import service_account from google.cloud import bigquery from collections import Counter class BigQueryExecutor: def __init__(self, credentialsJson: str, bqProjectId: str = ''): temp_file_path='' with tempfile.NamedTemporaryFile(mode='w', delete=False) as temp_file: json.dump(json.loads(credentialsJson), temp_file, indent=4) temp_file_path = temp_file.name credentials = service_account.Credentials.from_service_account_file(temp_file_path) os.remove(temp_file_path) self.client = bigquery.Client(project=bqProjectId, credentials=credentials) Ukuqala ukusetyenziswa kwe-script, kufuneka nje izinto ezimbini: I-Google Service Account ifayile ye-JSON credentials igama (okanye id) yeprojekthi yakho ye-BigQuery Xa unayo ezi, unokufumana kwaye uqala ukuqhuba iingxaki nge-script. Umxholo we-Google JSON credential Ukuze ufake i-akhawunti ye-Service, faka kwi Nceda ushiye umdlalo I-Google Cloud Console BigQuery Data Editor I-Google Cloud Console Emva kokufunda i-akhawunti, ufake, uqhagamshelane iimveliso Ukukhetha Ukusebenza oku kuvelisa kunye nokulanda ifayile ye-JSON credentials ye-akhawunti ye-service. “Keys” “Add key” “JSON” I-akhawunti yeenkonzo ye-JSON ikhona oku: { "type": "service_account", "project_id": YOUR_PROJECT, "private_key_id": private_key_id, "private_key": GCP_PRIVATE_KEY, "client_email": "email", "client_id": "id", "auth_uri": "auth_uri", "token_uri": "token_uri", "auth_provider_x509_cert_url": "auth_provider_x509_cert_url", "client_x509_cert_url": "url", "universe_domain": "universe_domain" } Ukusetyenziswa kwimeko, unako ukuguqula iziqinisekiso ze-JSON kwi-single-line string kwaye zihlanganisa ngqo kwi-script yakho. Nangona kunjalo, le ndlela — usebenzisa i-Secrets Manager ukugcina kwaye usebenzise i-credentials yakho ngokhuseleko. not recommended for production Uyakwazi ukufumana i-bqProjectId yakho kwi-project_id ye-field kwi-credentials ye-JSON. Iimodeli Ukuze usebenzise idatha ye-BigQuery ngempumelelo, kubalulekile ukucacisa iimodeli zeendatha ezinxulumene ne-structure yeemiphumo yeengxaki. Oku kukuvumela ukucacisa ikhowudi evumileyo, evumileyo kwaye enokufanelekileyo. Okulandelayo na umzekelo weemveliso efanayo: class BQCrashlyticsVersionsModel: def __init__(self, version: str, events: int, users: int ): self.version = version self.events = events self.users = users class BQCrashlyticsIssueModel: def __init__(self, issue_id: str, issue_title: str, blame_file: str, blame_library: str, blame_symbol: str, total_events: int, total_users: int, versions: list[BQCrashlyticsVersionsModel] ): self.issue_id = issue_id self.issue_title = issue_title self.blame_file = blame_file self.blame_library = blame_library self.blame_symbol = blame_symbol self.total_events = total_events self.total_users = total_users self.versions = versions iimveliso iimveliso iimveliso Kwaye lokugqibela, sinokufumana idatha kwi-BigQuery. Yongeza i-methode elandelayo kwi-class yakho ye-BigQueryExecutor eyenza i-SQL query ebhalwe ngexesha elandelayo kwi isixeko kunye nokuguqulwa imiphumela eyenziwe kwiimeko ze-model. Ukucinga kwe-SQL def getCrashlyticsIssues(self, lastHoursCount: int, tableName: str) -> list[BQCrashlyticsIssueModel]: firstEventsInfo = """'[' || STRING_AGG(events_info, ",") || ']' as events_info""" asVersions = """ '{"version":"' || application.display_version || '","events":' || COUNT(DISTINCT event_id) || ',"users":' || COUNT(DISTINCT installation_uuid) || '}' AS events_info """ query = f""" WITH pre as( SELECT issue_id, ARRAY_AGG(DISTINCT issue_title IGNORE NULLS) as issue_titles, ARRAY_AGG(DISTINCT blame_frame.file IGNORE NULLS) as blame_files, ARRAY_AGG(DISTINCT blame_frame.library IGNORE NULLS) as blame_libraries, ARRAY_AGG(DISTINCT blame_frame.symbol IGNORE NULLS) as blame_symbols, COUNT(DISTINCT event_id) as total_events, COUNT(DISTINCT installation_uuid) as total_users, {asVersions} FROM `{tableName}` WHERE 1=1 AND event_timestamp >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL {lastHoursCount} HOUR) AND event_timestamp < CURRENT_TIMESTAMP() AND error_type = "FATAL" GROUP BY issue_id, application.display_version ) SELECT issue_id, ARRAY_CONCAT_AGG(issue_titles) as issue_titles, ARRAY_CONCAT_AGG(blame_files) as blame_files, ARRAY_CONCAT_AGG(blame_libraries) as blame_libraries, ARRAY_CONCAT_AGG(blame_symbols) as blame_symbols, SUM(total_events) as total_events, SUM(total_users) as total_users, {firstEventsInfo} FROM pre WHERE 1=1 AND issue_id IS NOT NULL AND events_info IS NOT NULL GROUP BY issue_id ORDER BY total_users DESC """ bqRows = self.client.query(query).result() rows: list[BQCrashlyticsIssueModel] = [] def mergeArray(array: list[str]) -> str: if not array: return '' counter = Counter(array) most_common = counter.most_common(1) return most_common[0][0] if most_common else '' for row in bqRows: issueModel = BQCrashlyticsIssueModel( issue_id=row.issue_id, issue_title=mergeArray(row.issue_titles), blame_file=mergeArray(row.blame_files), blame_library=mergeArray(row.blame_libraries), blame_symbol=mergeArray(row.blame_symbols), total_events=row.total_events, total_users=row.total_users, versions=[BQCrashlyticsVersionsModel(version=jj['version'], events=jj['events'], users=jj['users']) for jj in json.loads(row.events_info)] ) rows.append(issueModel) return rows Ngoku kunokufumana isicelo yethu ye-SQL kwi-BigQuery ngqo kwi-Python. Nazi isibonelo epheleleyo yokuhamba kunye nokusebenza kunye nemiphumo: executor = BigQueryExecutor(credentialsJson=_jsonToken, bqProjectId=_bqProjectId) allCrashes = executor.getCrashlyticsIssues(lastHoursCount=24, tableName="tableNAME_IOS_REALTIME") for i in allCrashes: print(i.issue_title) [libswift_Concurrency.dylib] swift::swift_Concurrency_fatalError(unsigned int, char const*, ...) [SwiftUI] static DisplayList.ViewUpdater.Platform.updateClipShapesAsync(layer:oldState:newState:) Foundation [SwiftUI] __swift_memcpy112_8 [libswift_Concurrency.dylib] swift::AsyncTask::waitFuture(swift::AsyncTask*, swift::AsyncContext*, void (swift::AsyncContext* swift_async_context) swiftasynccall*, swift::AsyncContext*, swift::OpaqueValue*) [libobjc.A.dylib] objc_opt_respondsToSelector [VectorKit] void md::COverlayRenderLayer::layoutRibbon (std::__1::unique_ptr , std::__1::default_delete > > const&, ggl::CommandBuffer*, md::PolylineOverlayLayoutContext&, unsigned int, unsigned long long, bool, bool, float) [libswiftCore.dylib] _swift_release_dealloc [libobjc.A.dylib] objc_msgSend Hooray! 🎉 Ngokuqhelekanga ukufumana iinkcukacha ze-crash evela kwi-BigQuery, sinokufutshane ngexesha elandelayo - ukuthatha i-5 iingxaki ezininzi ezininzi kunye nokwenza iingxaki ze-Jira ngokuzenzakalelayo. Thola zonke iingcali ze-repository kwaye zihlanganisa Kwixesha lokufikelela iingxaki ze-crash kwi-developer, kufuneka kuqala ukucacisa abathengi ezinokufanelekileyo ngamnye i-crash. Ukwenza oku, siza kuqala ngokufaka zonke abathengi be-commit kwi-repository. Xa usebenzisa i-GitHub, kufuneka siphinde ezinye iinkcukacha ezizodwa: Zonke abadlali bangasebenzisa iidilesi ze-imeyile ezininzi kwi-commit, ngoko kufuneka uqhagamshelane iidilesi apho kufanelekileyo. I-GitHub isetyenziselwa i-imeyili ezininzi (isib. username@users.noreply.github.com), ngoko siza kuthatha iimeko ezininzi ngokufanayo. Indawo yokuqala kule step yi-extract and normalize the list of Git authors with their names and emails, usebenzisa isicelo esilandelayo: nqakraza nqakraza nqakraza nqakraza nqakraza import re class GitUserModel: def __init__(self, nicknames: set[str], emails: set[str], gitLogins: set[str] ): self.nicknames = nicknames self.emails = emails self.gitLogins = gitLogins def returnPossibleNicknames(text: str) -> set[str]: res = [findEmail(text), loginFromEmail(text), findGitNoreplyLogin(text)] return set(list(filter(None, res))) def findEmail(text: str) -> str: e = re.match(r"(([A-Za-z0-9+\.\_\-]*@[A-Za-z0-9+]*\.[A-Za-z0-9+]*))", text) if e: return e.group(1) def loginFromEmail(text: str) -> str: e = re.match(r"(([A-Za-z0-9+\.\_\-]*))@[A-Za-z0-9+]*\.[A-Za-z0-9+]*", text) if e: return e.group(1) def findGitNoreplyLogin(text: str) -> str: gu = re.match(r"\d+\+(([A-Za-z0-9+\.\_\-]*))@users\.noreply\.github\.com", text) if gu: return gu.group(1) else: gu = re.match(r"(([A-Za-z0-9+\.\_\-]*))@users\.noreply\.github\.com", text) if gu: return gu.group(1) class GitBlamer: def getAllRepoUsersMap(self, projectRootPath: str) -> list[GitUserModel]: users: list[GitUserModel] = [] allGitLog = os.popen("cd {}; git log | grep '^Author' | sort | uniq -c".format(projectRootPath)).read() for line in allGitLog.split('\n'): user = self._createUserFromBlameLine(line) if user: users.append(user) self._enrichUsersNicknames(users=users) users = self._mergeSameUsers(users) users = sorted(users, key=lambda x: list(x.emails)[0] if x.emails else list(x.gitLogins)[0] if x.gitLogins else "") return users def _createUserFromBlameLine(self, line): m = re.match(r".* Author: (.*) ", line) user = GitUserModel(nicknames=set(), emails=set(), gitLogins=set()) if m: val=set() if m.group(1): val.add(m.group(1)) if m.group(2): val.add(m.group(2)) user.nicknames = val else: return return user def _enrichUsersNicknames(self, users: list[GitUserModel]): for user in users: possibleNicknames = set() for nick in user.nicknames: possibleNicknames = possibleNicknames.union(returnPossibleNicknames(text=nick)) e = findEmail(text=nick) if e: user.emails.add(e) gu = findGitNoreplyLogin(text=nick) if gu: user.gitLogins.add(gu) user.nicknames = user.nicknames.union(possibleNicknames) def _mergeSameUsers(self, users: list[GitUserModel]): for i in range(0, len(users)): if i >= len(users): break for j in range(i+1, len(users)): if j >= len(users): break setLoweredJNicknames=set([u.lower() for u in users[j].nicknames]) for k in range(0, j): if k >= j: break setLoweredKNicknames=set([u.lower() for u in users[k].nicknames]) isSameNickname=len(setLoweredKNicknames.intersection(setLoweredJNicknames)) > 0 if isSameNickname: users[j].gitLogins = users[j].gitLogins.union(users[k].gitLogins) users[j].emails = users[j].emails.union(users[k].emails) users.pop(k) break return users Kwi-code elandelayo, sincoma ukuqhagamshela iindidi ezahlukileyo ze-commit ezininzi ezininzi ezinxulumene nomntu efanayo - umzekelo, user@gmail.com kunye ne-user@users.noreply.github.com. Kwakhona sincoma kunye namaxwebhu zabo kunye namaxwebhu ze-GitHub (kuye ziyafumaneka) kumgangatho. Nge-script elandelayo, unako ukuqalisa le nkqubo kwaye ufumane isithombe esifutshane, eyahlukileyo ze-committers zonke kwi-repository: projectRootPath="/IOS_project_path" blamer = GitBlamer() allUsers = blamer.getAllRepoUsersMap(projectRootPath=projectRootPath) for user in allUsers: print(", ".join(user.nicknames)) Ukubhalisa yonke imibuzo kwiifayile ye-repository kunye ne-file owner Okwangoku, sinazo ulwazi oluthe ngexabiso malunga neengxaki zethu kunye nabasebenzisi abalandeli. Oku kuthetha ukuba sinxulumene ngxaki elifanelekileyo ngexabiso elifanelekileyo ngexabiso elifanelekileyo kunye nabasebenzisi elifanelekileyo kwaye sinikeza ngokuzenzakalelayo umsebenzi ye-Jira elifanelekileyo. Kwixesha lokuvelisa i-crash-to-user mapping logic, siqonde iinkqubo zokusebenza ze-iOS kunye ne-Android. Lezi zindlela zisebenzisa iifomati ze-symbol ezahlukileyo, kwaye iingcebiso zokuhambisa iifayile ze-crash kwiimeko zihlukile. Ukusebenza oku ngokucacileyo, sinikezela i-class ye-abstract kunye ne-platform-specific implementations, okuvumela ukuba i-encapsulate iingcebiso kunye nokuphendula inkqubo ngokucacileyo. class AbstractFileToIssueMapper: def isPathBelongsToIssue(self, file: str, filePath: str, issue: BQCrashlyticsIssueModel) -> bool: raise Exception('Not implemented method AbstractFileToIssueMapper') class AndroidFileToIssueMapper(AbstractFileToIssueMapper): def __init__(self): self.libraryName = 'inDrive' def isPathBelongsToIssue(self, file: str, filePath: str, issue: BQCrashlyticsIssueModel) -> bool: if file != issue.blame_file or not issue.blame_symbol.startswith(self.libraryName): return False fileNameNoExtension = file.split('.')[0] fileNameIndexInSymbol = issue.blame_symbol.find(fileNameNoExtension) if fileNameIndexInSymbol < 0: return False relativeFilePathFromSymbol = issue.blame_symbol[0:fileNameIndexInSymbol].replace('.', '/') relativeFilePathFromSymbol = relativeFilePathFromSymbol + file return filePath.endswith(relativeFilePathFromSymbol) class IosFileToIssueMapper(AbstractFileToIssueMapper): def __init__(self): self.indriveLibraryName = 'inDrive' self.indriveFolderName = 'inDrive' self.modulesFolderName = 'Modules' def isPathBelongsToIssue(self, file: str, filePath: str, issue: BQCrashlyticsIssueModel) -> bool: if file != issue.blame_file: return False isMatchFolder = False if issue.blame_library == self.indriveLibraryName: isMatchFolder = filePath.startswith('{}/'.format(self.indriveFolderName)) else: isMatchFolder = filePath.startswith('{}/{}/'.format(self.modulesFolderName, issue.blame_library)) return isMatchFolder Ukuqhagamshelwano olufanelekileyo kunokufumaneka kwiprojekthi yakho, kodwa ingxaki yokuqala le-class kuyinto ukuhlaziywa ukuba ingxaki elifanelekileyo kwifayile elifanelekileyo. Xa le logic iyatholakala, singakwazi ukuqhagamshela ifayile kwimeko kunye nokuthintela kubathengi ifayile ezifanelekileyo. import subprocess class MappedIssueFileModel: def __init__(self, fileGitLink: str, filePath: str, issue: BQCrashlyticsIssueModel, fileOwner: GitUserModel ): self.fileGitLink = fileGitLink self.filePath = filePath self.issue = issue self.fileOwner = fileOwner class BigQueryCrashesFilesMapper: def getBlameOut(self, filePath: str, projectRootPath: str) -> list[str]: dangerousChars = re.compile(r'[;|&\r\n]|\.\.') if dangerousChars.search(filePath) or dangerousChars.search(projectRootPath): return None if not subprocess.check_output(['git', 'ls-files', filePath], cwd=projectRootPath, text=True): return None blameProc = subprocess.Popen(['git', 'blame', filePath, '-cwe'], cwd=projectRootPath, stdout=subprocess.PIPE, text=True) blameRegex=r' ' grepProc = subprocess.Popen(['grep', '-o', blameRegex], stdin=blameProc.stdout, stdout=subprocess.PIPE, text=True) blameProc.stdout.close() sortProc = subprocess.Popen(['sort'], stdin=grepProc.stdout, stdout=subprocess.PIPE, text=True) grepProc.stdout.close() uniqProc = subprocess.Popen(['uniq', '-c'], stdin=sortProc.stdout, stdout=subprocess.PIPE, text=True) sortProc.stdout.close() finalProc = subprocess.Popen(['sort', '-bgr'], stdin=uniqProc.stdout, stdout=subprocess.PIPE, text=True) uniqProc.stdout.close() blameOut, _ = finalProc.communicate() blameArray=list(filter(len, blameOut.split('\n'))) return blameArray def findFileOwner(self, filePath: str, gitFileOwners: list[GitUserModel], projectRootPath: str) -> GitUserModel: blameArray = self.getBlameOut(filePath=filePath, projectRootPath=projectRootPath) if not blameArray: return foundAuthor = None for blameI in range(0, len(blameArray)): author = re.match(r".*\d+ ", blameArray[blameI]) if author: possibleNicknames = returnPossibleNicknames(text=author.group(1)) for gitFileOwner in gitFileOwners: if len(gitFileOwner.nicknames.intersection(possibleNicknames)) > 0: foundAuthor = gitFileOwner break if foundAuthor: break return foundAuthor def mapBQResultsWithFiles(self, fileToIssueMapper: AbstractFileToIssueMapper, issues: list[BQCrashlyticsIssueModel], gitFileOwners: list[GitUserModel], projectRootPath: str ) -> list[MappedIssueFileModel]: mappedArray: list[MappedIssueFileModel] = [] githubMainBranch = "https://github.com/inDriver/UDF/blob/master" for root, dirs, files in os.walk(projectRootPath): for file in files: filePath = os.path.join(root, file).removeprefix(projectRootPath).strip('/') gitFileOwner = None for issue in issues: if fileToIssueMapper.isPathBelongsToIssue(file, filePath, issue): if not gitFileOwner: gitFileOwner = self.findFileOwner(filePath=filePath, gitFileOwners=gitFileOwners, projectRootPath=projectRootPath) mappedIssue = MappedIssueFileModel( fileGitLink='{}/{}'.format(githubMainBranch, filePath.strip('/')), filePath=filePath, issue=issue, fileOwner=gitFileOwner ) mappedArray.append(mappedIssue) mappedArray.sort(key=lambda x: x.issue.total_users, reverse=True) return mappedArray Konke kufuneka uyenze ngexesha elandelayo i-update ye-githubMainBranch kunye ne-link kwi-repository yakho. Okulandelayo, sinika iingxaki kunye neengcali zefayela, iifayile ze-map ngokufanelekileyo usebenzisa ikhodi elandelayo, kwaye sinike imiphumo yokugqibela - umxholo weengxaki eyenziwe ngexesha le-total_users ngexesha elandelayo. mapper = BigQueryCrashesFilesMapper() mappedIssues = mapper.mapBQResultsWithFiles( fileToIssueMapper=IosFileToIssueMapper(), issues=allCrashes, gitFileOwners=allUsers, projectRootPath=projectRootPath ) for issue in mappedIssues: if issue.fileOwner: print(list(issue.fileOwner.nicknames)[0], issue.issue.total_users, issue.issue.issue_title) else: print('no owner', issue.issue.total_users, issue.issue.issue_title) Yenza umsebenzi ye-Jira ye-file owner ye-crash Okwangoku, sinayo yonke into efunekayo ukuqala ukuvelisa iingxaki ze-Jira kubathengi be-crash. Nangona kunjalo, nceda ushiye ukuba iingxaki ze-Jira ziquka kumazwe ngamazwe ngamazwe - iingxaki ze-custom, iinkqubo zokusebenza kunye neengxaki ziyafumaneka. Nceda uqhagamshelane neendokumentation ye-API ye-Jira yaye usebenzisa i-Python ye-client yayo yokusebenza kunye ne-setup yakho. Nazi ezinye iingcebiso ezisebenzayo ezisekelwe ngexesha lethu: Ngaba unayo iingxaki kuzo zonke iingxaki. Qhagamshelana neengxaki ze-5-10 ezidlulileyo ezisekelwe kwinqanaba yabasetyenziswa ngabasetyhini okanye i-impact threshold. I-Task metadata ifumaneka. I-Storage information on created tasks in a persistent storage. I-BigQuery isetyenziselwa ukugcina idatha kwi-table eyahlukileyo kwaye isebenze ngalinye i-script run. Ukulungiselela izicelo ezivulekileyo ukuba ingxaki ifumaneka kwizilwanyana ezintsha ze-app - oku kuqinisekisa ukuba i-regression ayidluliswa. Ukuqhagamshela umsebenzi kwimeko efanayo ukunceda ukuyifaka kwizifundo ezidlulileyo kwaye ukunceda ukuxhaswa. Zifumaneke iinkcukacha ezininzi iinkcukacha zokusebenza. Faka iinkcukacha ze-crash, iinkcukacha zokusetyenziswa kwabasebenzisi, iinkcukacha njl. Ukuqhagamshelana ne-link ifumaneka ukuba zithunyelwe kwifayile elifanayo - oku inikeza inkqubo eyongezelelweyo. Qinisekisa i-team yakho kwi-Slack (okanye kwinkqubo ye-messaging) xa izicelo ezintsha ziye ziye ziye ziye zithunyelwe okanye ziye ziye zithunyelwe. Zihlanganisa izixhumanisi ezisebenzayo kwi-crash report, task, iifayile ze-GitHub ezinxulumene, njl. Ukongeza ukwelashwa kweengxaki kwi-script yakho. Ukusebenzisa i-try/except blocks kunye nokuthumela iingcebiso ze-Slack xa kukho ingxaki. I-Cache isebenza ngokushesha ngexesha lokuvelisa. Umzekelo, i-Cache ye-BigQuery i-crash retrievals kwi-locally yokushesha i-iteration. Kwimeko ezininzi iingxaki ziquka iibhayisikili ezahlukahlukeneyo okanye zangaphakathi. Kwiimeko ezininzi, kufuneka uqhagamshelane isebenzo ngamanzi, kodwa kunokuncedisa ukwenza i-Jira problem ngokuzenzekelayo kunye ne-context epheleleyo yeengxaki. Ukucinga Le nkqubo inikezela usebenzise iindidi ezininzi zeengxelo zeengxelo ngosuku, kwaye zithunyelwe kwi-developer efanelekileyo kwimizuzu embalwa - ngaphandle kokusebenza okuzenzakalelayo. Ukuba iqela lakho lihlala kwiinkcubeko ze-crash engabonakaliweyo - uqhagamshelane.

Ukwakha i-Crash Report Automation ye-iOS kunye ne-Android

About Author

IZIMVO

ZIJONGE IIMPAWU

ELI NQAKU LINIKEZELWE KU

Related Stories

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps