SDK tutorial¶

I. Login authentication¶

render_para = {
    "domain": "jop.foxrenderfarm.com",  # If it doesn't work, you can use "task.foxrenderfarm.com"
    "platform": "2",
    "access_id": "xxxx",
    "access_key": "xxxx",
}

api = RayvisionAPI(
        access_id=render_para['access_id'],
        access_key=render_para['access_key'],
        domain=render_para['domain'],
        platform=render_para['platform']
    )

RayvisionAPI Parameters:

Parameters	Type	Required	Default	Description
domain	string	Y	task.renderbus.com	China user：task.renderbus.com，Foreign user：jop.foxrenderfarm.com
platform	string	Y	2	platform ID，example: W2:"2", W6/qingyun:"6", GPU Region 1:"21"
access_id	string	Y		user authorization id
access_key	string	Y		user authorization key

II. Analysis of the scene¶

Analysis is independent(Maya / Houdini / Clarisse)

Example for Houdini:

from rayvision_houdini.analyze_houdini import AnalyzeHoudini

analyze_info = {
    "cg_file": r"D:\files\CG FILE\flip_test_slice4.hip",
    "workspace": "c:/workspace",
    "software_version": "17.5.293",
    "project_name": "Project1",
    "plugin_config": {
        'renderman': '22.6'
    }
}
analyze_obj = AnalyzeHoudini(**analyze_info)
analyze_obj.analyse()

Instructions：

“workspace” is used to control the location of the generated json file. If workspace is not set, the default location is generated:
```
windows : os.environ["USERPROFILE"] + "renderfarm_sdk"  
Linux：os.environ["HOME"] + “renderfarm_sdk”
```
Analytically generated task.json is no “task_id”、“user_id”、”project_id” parameters，Users can choose to write the three parameters themselves, or to write the three parameters automatically when check.

AnalyzeHoudini Parameters:

Parameters	Type	Required	Default	Description
cg_file	string	Y		Scenario files to analyze
software_version	string	Y		The version of the software
project_name	string	N		The project name
plugin_config	dict	N		The plug-in configuration，example {'renderman': '22.6'}
workspace	string	N	None	Analyze the location of the generated json file (avoiding duplication automatically adds a timestamp folder)

III. Add special fields and update the json file interface¶

Only updates and modifications to the parameters of task.json and upload.json files are supported.

1. Modify the task.json¶

update_task_info(update_info, task_path)

task_info

from rayvision_api.utils import update_task_info, append_to_task, append_to_upload
update_task = {
    "pre_frames": "100",  # Sets the priority to render the first frame
    "stop_after_test": "1"  # Stop rendering after rendering the priority frame
}
update_task_info(update_task, analyze_obj.task_json)

2. task.json add custom parameters¶

The added custom parameters will be integrated into dictionary for key is “additional_info”. 【Warning】：Custom parameters will not take effect immediately. If you have this requirement, please contact our customer service。

additional_info

custom_info_to_task = {
    "env": "houdini_env"
}
append_to_task(custom_info_to_task, analyze_obj.task_json)

3. user custom upload.json¶

Support custom add file path to upload.json, will automatically deduplicate append_to_upload(files_paths, upload_path)

upload

custom_info_to_upload = [
    r"D:\houdini\CG file\Cam003\cam.abc",
]
append_to_upload(custom_info_to_upload, analyze_obj.upload_json)

IV. Setting hardware Configuration and Validate json file¶

The hardware configuration is controlled by the parameter “hardwareconfigid”, which can be obtained through the API interface.(”API interfaces use methods” –> “Get platform hardware configuration information”)

Specify the hardware configuration by setting model, ram, gpuNum of hardware_config

hardware_config = {
    "model": "Default",  # Platform CPU: Default or Platform GPU: 1080Ti or 2080Ti
    "ram": "64GB",  # memory: 64GB or 128GB
    "gpuNum": None  # GPU platform requires input like 2*GPU, if CPU platform it is None
}

When validating,we set the hardware configuration based on the hardware_config and check whether there isuser_id, project_id, task_id in task.json, If not, the interface is called to fetch the relevant parameters from the server and write task.json

check_obj = RayvisionCheck(api, analyze_obj)
task_id = check_obj.execute(hardware_config, analyze_obj.task_json, analyze_obj.upload_json)

V. Upload¶

Now there are two ways:

1.Upload the json file first and then upload the resource file according to “upload.json”:¶

Four json files are uploaded, and the transfer engine automatically starts uploading the scene, resource, and other files based on the upload.json file.

upload(self,task_id,task_json_path,tips_json_path,asset_json_path,upload_json_path, max_speed=None)

CONFIG_PATH = [
    r"C:\workspace\work\tips.json",
    r"C:\workspace\work\task.json",
    r"C:\workspace\work\asset.json",
    r"C:\workspace\work\upload.json",
]
upload_obj = RayvisionUpload(api)
upload_obj.upload(str(task_id), *CONFIG_PATH)

2.Uploading json files is completely independent of the user resources:¶

Uploading resources：upload_asset(self, upload_json_path, max_speed=None, is_db=True)

Upload json file: upload_config(self, task_id, config_file_list, max_speed=None)

CONFIG_PATH = [
    r"C:\workspace\work\tips.json",
    r"C:\workspace\work\task.json",
    r"C:\workspace\work\asset.json",
]
UPLOAD = RayvisionUpload(api)
UPLOAD.upload_asset(r"C:\workspace\work\upload.json", is_db=False)
UPLOAD.upload_config("5165465", CONFIG_PATH)

【Warning】:You need a task ID to upload a json file, but you don’t need a task ID to upload a resource file;
The ‘is_db’ parameter in upload_asset is used to control whether or not a local database is needed. By default, a local database is used;

VI. Submit Task¶

api.submit(int(task_id))

VIII. Download¶

Download now provides 3 ways:

1. Supports custom downloads of hierarchical directory structures under each rendering task.¶

download(self, task_id_list=None, max_speed=None, print_log=True, download_filename_format="true",local_path=None, server_path=None)

download = RayvisionDownload(api)
download.download(download_filename_format="true", server_path="18164087_muti_layer_test/l_ayer2")

Warning：This method needs to provide the task ID if the “server_path” is not empty, and the task ID does not take effect if there is a custom “server_path”.

2. Realtime download, that is, the task rendering completed a frame began to download¶

auto_download(self, task_id_list=None, max_speed=None, print_log=False, sleep_time=10, download_filename_format="true", local_path=None)

download = RayvisionDownload(api)
download.auto_download([18164087], download_filename_format="false")

Warning：The method task ID cannot be empty

3. The task is not downloaded until all frames are rendered¶

auto_download_after_task_completed(self, task_id_list=None, max_speed=None, print_log=True, sleep_time=10, download_filename_format="true", local_path=None):

download = RayvisionDownload(api)
download.auto_download_after_task_completed([18164087], download_filename_format="false")

Warning: The method task ID cannot be empty

IX. Attachment: transfer configuration file¶

1. Transport configuration Settings include:

Select database type, database file path Settings, transfer log path Settings

2. The transport configuration file used by default: db_config.ini， The following figure

db_config.ini

db_config.ini:

[TRANSFER_LOG_PATH]
transfer_log_path =

[DATABASE_CONFIG]
on = true
type = sqlite
db_path =D:\test\upload

[REDIS]
host = 127.0.0.1
port = 6379
password =
table_index = 0
timeout = 5000

[SQLITE]
temporary = false

The user can also modify the configuration for the template based on the default configuration and specify the database configuration file location, specifying a custom configuration file as follows.

from rayvision_sync.upload import RayvisionUpload

UPLOAD = RayvisionUpload(api, db_config_path=r"D:\test\upload\db_config.ini")

3. db_config.ini Parameters:

Parameters	instructions	Default
transfer_log_path	Transfer engine log file path
on	Whether to use a local database, true / false: Yes / No	true
type	Select database,currently only supports "redis" and "sqlite"	sqlite
db_path	The database file saves the path
host	redis database host	127.0.0.1
port	redis database port	6379
password	redis database password
table_index	Redis Repositories，cannot be empty
timeout	Redis client connection timeout,unit ms	5000
temporary	When using sqlite database, if the uploaded record data is deleted after the completion of the upload, the default "false" will not be deleted	false

4. transfer_log_path and db_path The priority rule for values is as follows:

db_config.ini custom paths are preferred if they are set；
No custom path is as follows:

transfer_log_path
- Use environment variables first ‘RAYVISION_LOG’
- Second use:
  
  window: The environment variable “USERPROFILE”/<renderfarm_sdk>
  Linux: The environment variable “HOME” /<renderfarm_sdk>
db_path
- Use environment variables first ‘RAYVISION_LOG’
- Second use:
  
  window: The environment variable “USERPROFILE”/<renderfarm_sdk>
  Linux：The environment variable “HOME” /<renderfarm_sdk>