diff options
author | RaNaN <Mast3rRaNaN@hotmail.de> | 2012-01-07 20:11:16 +0100 |
---|---|---|
committer | RaNaN <Mast3rRaNaN@hotmail.de> | 2012-01-07 20:11:16 +0100 |
commit | 1bb6ebf544b43cacf7c0755c5a8608b79b95e2d6 (patch) | |
tree | c09084a44bcde0528db1007c026ef2eb6a349c05 /docs/api | |
parent | closed #486 (diff) | |
download | pyload-1bb6ebf544b43cacf7c0755c5a8608b79b95e2d6.tar.xz |
MultiHoster plugin type, some fixes, new documentation structure
Diffstat (limited to 'docs/api')
-rw-r--r-- | docs/api/access_api.rst | 122 | ||||
-rw-r--r-- | docs/api/datatypes.rst | 352 | ||||
-rw-r--r-- | docs/api/overview.rst | 35 |
3 files changed, 509 insertions, 0 deletions
diff --git a/docs/api/access_api.rst b/docs/api/access_api.rst new file mode 100644 index 000000000..efa1ae3fc --- /dev/null +++ b/docs/api/access_api.rst @@ -0,0 +1,122 @@ +.. _access_api: + +********************* +How to access the API +********************* + +pyLoad has a very powerfull API with can be accessed in several ways. + +Overview +-------- + +First of all, you need to know what you can do with our API. It lets you do all common task like +retrieving download status, manage queue, manage accounts, modify config and so on. + +This document is not intended to explain every function in detail, for a complete listing +see :class:`Api <module.Api.Api>`. + +Of course its possible to access the ``core.api`` attribute in plugins and hooks, but much more +interesting is the possibillity to call function from different programs written in many different languages. + +pyLoad uses thrift as backend and provides its :class:`Api <module.Api.Api>` as service. +More information about thrift can be found here http://wiki.apache.org/thrift/. + + +Using Thrift +------------ + +Every thrift service has to define all data structures and declare every method which should be usable via rpc. +This file is located at :file:`module/remote/thriftbackend/pyload.thrift`, its very helpful to inform about +arguments and detailed structure of return types. However it does not contain any information about what the functions does. +You can also look at it :doc:`here <datatypes>` + +Assuming you want to use the API in any other language than python than check if it is +supported here http://wiki.apache.org/thrift/LibraryFeatures?action=show&redirect=LanguageSupport. + +Now install thrift, for instructions see http://wiki.apache.org/thrift/ThriftInstallation. +If every thing went fine you are ready to generate the method stubs, the command basically looks like this. :: + + $ thrift --gen (language) pyload.thrift + +You find now a directory named :file:`gen-(language)`. For instruction how to use the generated files consider the docs +at the thrift wiki and the examples here http://wiki.apache.org/thrift/ThriftUsage. + + +======= +Example +======= +In case you want to use python, pyload has already all files included to access the api over rpc. + +A basic script that prints out some information: :: + + from module.remote.thriftbackend.ThriftClient import ThriftClient, WrongLogin + + try: + client = ThriftClient(host="127.0.0.1", port=7227, user="User", password="yourpw") + except: + print "Login was wrong" + exit() + + print "Server version:", client.getServerVersion() + print client.statusDownloads() + q = client.getQueue() + for p in q: + data = client.getPackageData(p.pid) + print "Package Name: ", data.name + +That's all for now, pretty easy isn't it? +If you still have open questions come around in irc or post them at our pyload forum. + + +Using HTTP/JSON +--------------- + +Another maybe easier way, which does not require much setup is to access the JSON Api via HTTP. +For this reason the webinterface must be enabled. + +===== +Login +===== + +First you need to authenticate, if you using this within the webinterface and the user is logged the API is also accessible, +since they share the same cookie/session. + +However, if you are building a external client and want to authenticate manually +you have to send your credentials ``username`` and ``password`` as +POST parameter to ``http://pyload-core/api/login``. + +The result will be your session id. If you are using cookies, it will be set and you can use the API now. +In case you dont't have cookies enabled you can pass the session id as ``session`` POST parameter +so pyLoad can authenticate you. + +=============== +Calling Methods +=============== + +In general you can use any method listed at the :class:`Api <module.Api.Api>` documentation, which is also available to +the thriftbackend. + +Access works simply via ``http://pyload-core/api/methodName``, where ``pyload-core`` is the ip address +or hostname including the webinterface port. By default on local access this would be `localhost:8000`. + +The return value will be formatted in JSON, complex data types as dictionaries. +As mentionted above for a documentation about the return types look at the thrift specification file :file:`module/remote/thriftbackend/pyload.thrift`. + +================== +Passing parameters +================== + +To pass arguments you have two choices. +Either use positional arguments, eg ``http://pyload-core/api/getFileData/1``, where 1 is the FileID, or use keyword arguments +supplied via GET or POST ``http://pyload-core/api/getFileData?fid=1``. You can find the argument names in the :class:`Api <module.Api.Api>` +documentation. + +It is important that *all* arguments are in JSON format. So ``http://pyload-core/api/getFileData/1`` is valid because +1 represents an integer in json format. On the other hand if the method is expecting strings, this would be correct: +``http://pyload-core/api/getUserData/"username"/"password"``. + +Strings are wrapped in double qoutes, because `"username"` represents a string in json format. It's not limited to strings and intergers, +every container type like lists and dicts are possible. You usually don't have to convert them. just use a json encoder before using them +in the HTTP request. + +Please note that the data have to be urlencoded at last. (Most libaries will do that automatically)
\ No newline at end of file diff --git a/docs/api/datatypes.rst b/docs/api/datatypes.rst new file mode 100644 index 000000000..02e3209cc --- /dev/null +++ b/docs/api/datatypes.rst @@ -0,0 +1,352 @@ +.. _api_datatypes: + +*********************** +API Datatype Definition +*********************** + +Below you find a copy of :file:`module/remote/thriftbackend/pyload.thrift`, which is used to generate the data structs +for various languages. It is also a good overview of avaible methods and return data. + +.. code-block:: c + + .. [[[cog cog.out(open('module/remote/thriftbackend/pyload.thrift', 'rb').read()) ]]] + namespace java org.pyload.thrift + + typedef i32 FileID + typedef i32 PackageID + typedef i32 ResultID + typedef i32 InteractionID + typedef list<string> LinkList + typedef string PluginName + typedef byte Progress + typedef byte Priority + + + enum DownloadStatus { + Finished + Offline, + Online, + Queued, + Skipped, + Waiting, + TempOffline, + Starting, + Failed, + Aborted, + Decrypting, + Custom, + Downloading, + Processing, + Unknown + } + + enum Destination { + Collector, + Queue + } + + // types for user interaction + // some may only be place holder currently not supported + // also all input - output combination are not reasonable, see InteractionManager for further info + enum Input { + NONE, + TEXT, + TEXTBOX, + PASSWORD, + BOOL, // confirm like, yes or no dialog + CLICK, // for positional captchas + CHOICE, // choice from list + MULTIPLE, // multiple choice from list of elements + LIST, // arbitary list of elements + TABLE // table like data structure + } + // more can be implemented by need + + // this describes the type of the outgoing interaction + // ensure they can be logcial or'ed + enum Output { + CAPTCHA = 1, + QUESTION = 2, + NOTIFICATION = 4, + } + + struct DownloadInfo { + 1: FileID fid, + 2: string name, + 3: i64 speed, + 4: i32 eta, + 5: string format_eta, + 6: i64 bleft, + 7: i64 size, + 8: string format_size, + 9: Progress percent, + 10: DownloadStatus status, + 11: string statusmsg, + 12: string format_wait, + 13: i64 wait_until, + 14: PackageID packageID, + 15: string packageName, + 16: PluginName plugin, + } + + struct ServerStatus { + 1: bool pause, + 2: i16 active, + 3: i16 queue, + 4: i16 total, + 5: i64 speed, + 6: bool download, + 7: bool reconnect + } + + struct FileData { + 1: FileID fid, + 2: string url, + 3: string name, + 4: PluginName plugin, + 5: i64 size, + 6: string format_size, + 7: DownloadStatus status, + 8: string statusmsg, + 9: PackageID packageID, + 10: string error, + 11: i16 order + } + + struct PackageData { + 1: PackageID pid, + 2: string name, + 3: string folder, + 4: string site, + 5: string password, + 6: Destination dest, + 7: i16 order, + 8: optional i16 linksdone, + 9: optional i64 sizedone, + 10: optional i64 sizetotal, + 11: optional i16 linkstotal, + 12: optional list<FileData> links, + 13: optional list<FileID> fids + } + + struct InteractionTask { + 1: InteractionID iid, + 2: Input input, + 3: list<string> structure, + 4: list<string> preset, + 5: Output output, + 6: list<string> data, + 7: string title, + 8: string description, + 9: string plugin, + } + + struct ConfigItem { + 1: string name, + 2: string long_name, + 3: string description, + 4: string type, + 5: string default_value, + 6: string value, + } + + struct ConfigSection { + 1: string name, + 2: string long_name, + 3: string description, + 4: string long_description, + 5: optional list<ConfigItem> items, + 6: optional map<string, InteractionTask> handler, + } + + struct CaptchaTask { + 1: i16 tid, + 2: binary data, + 3: string type, + 4: string resultType + } + + struct EventInfo { + 1: string eventname, + 2: list<string> event_args, + } + + struct UserData { + 1: string name, + 2: string email, + 3: i32 role, + 4: i32 permission, + 5: string templateName + } + + struct AccountInfo { + 1: PluginName plugin, + 2: string loginname, + 3: bool valid, + 4: i64 validuntil, + 5: i64 trafficleft, + 6: i64 maxtraffic, + 7: bool premium, + 8: bool activated, + 9: map<string, string> options, + } + + struct ServiceCall { + 1: PluginName plugin, + 2: string func, + 3: string arguments, // empty string or json encoded list + } + + struct OnlineStatus { + 1: string name, + 2: PluginName plugin, + 3: string packagename, + 4: DownloadStatus status, + 5: i64 size, // size <= 0 : unknown + } + + struct OnlineCheck { + 1: ResultID rid, // -1 -> nothing more to get + 2: map<string, OnlineStatus> data, //url to result + } + + + // exceptions + + exception PackageDoesNotExists{ + 1: PackageID pid + } + + exception FileDoesNotExists{ + 1: FileID fid + } + + exception UserDoesNotExists{ + 1: string user + } + + exception ServiceDoesNotExists{ + 1: string plugin + 2: string func + } + + exception ServiceException{ + 1: string msg + } + + service Pyload { + + //config + string getConfigValue(1: string section, 2: string option), + void setConfigValue(1: string section, 2: string option, 3: string value), + map<string, ConfigSection> getConfig(), + map<PluginName, ConfigSection> getPluginConfig(), + ConfigSection configureSection(1: string section), + + // server status + void pauseServer(), + void unpauseServer(), + bool togglePause(), + ServerStatus statusServer(), + i64 freeSpace(), + string getServerVersion(), + void kill(), + void restart(), + list<string> getLog(1: i32 offset), + bool isTimeDownload(), + bool isTimeReconnect(), + bool toggleReconnect(), + + // download preparing + + // packagename - urls + map<string, LinkList> generatePackages(1: LinkList links), + map<PluginName, LinkList> checkURLs(1: LinkList urls), + map<PluginName, LinkList> parseURLs(1: string html, 2: string url), + + // parses results and generates packages + OnlineCheck checkOnlineStatus(1: LinkList urls), + OnlineCheck checkOnlineStatusContainer(1: LinkList urls, 2: string filename, 3: binary data) + + // poll results from previosly started online check + OnlineCheck pollResults(1: ResultID rid), + + // downloads - information + list<DownloadInfo> statusDownloads(), + PackageData getPackageData(1: PackageID pid) throws (1: PackageDoesNotExists e), + PackageData getPackageInfo(1: PackageID pid) throws (1: PackageDoesNotExists e), + FileData getFileData(1: FileID fid) throws (1: FileDoesNotExists e), + list<PackageData> getQueue(), + list<PackageData> getCollector(), + list<PackageData> getQueueData(), + list<PackageData> getCollectorData(), + map<i16, PackageID> getPackageOrder(1: Destination destination), + map<i16, FileID> getFileOrder(1: PackageID pid) + + // downloads - adding/deleting + list<PackageID> generateAndAddPackages(1: LinkList links, 2: Destination dest), + PackageID addPackage(1: string name, 2: LinkList links, 3: Destination dest, 4: string password), + void addFiles(1: PackageID pid, 2: LinkList links), + void uploadContainer(1: string filename, 2: binary data), + void deleteFiles(1: list<FileID> fids), + void deletePackages(1: list<PackageID> pids), + + // downloads - modifying + void pushToQueue(1: PackageID pid), + void pullFromQueue(1: PackageID pid), + void restartPackage(1: PackageID pid), + void restartFile(1: FileID fid), + void recheckPackage(1: PackageID pid), + void stopAllDownloads(), + void stopDownloads(1: list<FileID> fids), + void setPackageName(1: PackageID pid, 2: string name), + void movePackage(1: Destination destination, 2: PackageID pid), + void moveFiles(1: list<FileID> fids, 2: PackageID pid), + void orderPackage(1: PackageID pid, 2: i16 position), + void orderFile(1: FileID fid, 2: i16 position), + void setPackageData(1: PackageID pid, 2: map<string, string> data) throws (1: PackageDoesNotExists e), + list<PackageID> deleteFinished(), + void restartFailed(), + + //events + list<EventInfo> getEvents(1: string uuid) + + //accounts + list<AccountInfo> getAccounts(1: bool refresh), + list<string> getAccountTypes() + void updateAccount(1: PluginName plugin, 2: string account, 3: string password, 4: map<string, string> options), + void removeAccount(1: PluginName plugin, 2: string account), + + //auth + bool login(1: string username, 2: string password), + UserData getUserData(1: string username, 2:string password) throws (1: UserDoesNotExists ex), + map<string, UserData> getAllUserData(), + + //services + + // servicename : description + map<PluginName, map<string, string>> getServices(), + bool hasService(1: PluginName plugin, 2: string func), + string call(1: ServiceCall info) throws (1: ServiceDoesNotExists ex, 2: ServiceException e), + + + //info + // {plugin: {name: value}} + map<PluginName, map<string,string>> getAllInfo(), + map<string, string> getInfoByPlugin(1: PluginName plugin), + + //scheduler + + // TODO + + + // User interaction + + //captcha + bool isCaptchaWaiting(), + CaptchaTask getCaptchaTask(1: bool exclusive), + string getCaptchaTaskStatus(1: InteractionID tid), + void setCaptchaResult(1: InteractionID tid, 2: string result), + } + .. [[[end]]] + diff --git a/docs/api/overview.rst b/docs/api/overview.rst new file mode 100644 index 000000000..02cee3e0d --- /dev/null +++ b/docs/api/overview.rst @@ -0,0 +1,35 @@ +.. _overview: + +======================================= +API - Application Programming Interface +======================================= + +From Wikipedia, the free encyclopedia [1]_: + + An application programming interface (API) is a source code based specification intended to be used as an interface + by software components to communicate with each other. An API may include specifications for routines, + data structures, object classes, and variables. + +.. rubric:: Motivation + +The idea of the centralized pyLoad :class:`Api <module.Api.Api>` is to give uniform access to all integral parts +and plugins in pyLoad, and furthermore to other clients, written in arbitrary programming languages. +Most of the :class:`Api <module.Api.Api>` functionality is exposed via RPC [2]_ and accessible via thrift [3]_ or +simple JSON objects [4]_. In conclusion the :class:`Api <module.Api.Api>` is accessible via many programming language, +over network from remote maschines and over browser with javascript. + + +.. rubric:: Contents + +.. toctree:: + + access_api.rst + datatypes.rst + + +.. rubric:: Footnotes + +.. [1] http://en.wikipedia.org/wiki/Application_programming_interface +.. [2] http://en.wikipedia.org/wiki/Remote_procedure_call +.. [3] http://en.wikipedia.org/wiki/Thrift_(protocol) +.. [4] http://en.wikipedia.org/wiki/Json
\ No newline at end of file |