nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

MediaBin Deployment Agent

maheshd

Hi,

I have created project (Project Properties, Derivatives, and Targets) and Schedules. In schedules I have selected user defined project derivatives as well as <>.I tested the schedule by using (run/test) option. Schedule has been completed successfully and created metadata.exp along with mediabin assets on the target machine. This metadata .exp file contains all the unnecessary fields like Ascii Comment, Asset Type, Audio Avg Bit Rate, Audio Bits per Sample, Audio Channels, Audio Codec etc... here my questions are

1) how to define schedule or project that create metadata.exp
With few specified metadata fields
Example: generate medtadata.exp with following 4 fields (name, Author, Resolution, date) i.e. I have to filter all remaining metadata fields.

2) How to generate metadata.exp for each asset instead of common metadata.exp file.

Environment:

MediaBin asset server 4.5.2

Thanks,
mahesh

Find more posts tagged with

Comments

Migrateduser

Specification of which metadata fields you want to include in the metadata.exp file is not supported. All metadata fields are included.
Also there is no way to cause a separate metadata export file to be generated for each asset. The file contains metadata for all target assets.

If these enhancements are essential to your workflow, you might consider using the Interwoven Customer Services Organization (CSO) to help you customize Deployment Agent.

lyman

Some thoughts from the server side. Would it be possible in your use case to remove read permission for the unwanted fields for the user DA is running as (this implies that DA is not running as an MBAdmin...not positive if that is possible)? If possible this would be my first idea. Also, could you add a "metadata remover" primitive to the task being used to deploy the assets?

Lyman Hurd

Migrateduser

Permissioning metadata for a special DA project user would enable you to limit the metadata the server makes visible to a client app (DA). That is a way to reduce the number of columns in the metadata export file. Tasks would need to be permissioned so this DA project user can use them.

Currently, DA includes all displayable metadata for the project user in the metadata export file (metadata.exp). This file was designed to be imported into a spreadsheet rendering program like Excel. CSO has developed a "canned" customization to both the DA UI and DA service to allow a user to choose which metadata fields they want to include in the file. From the UI, the field separation characters and intrafield separators can also be specified. Currently field separators are specified for all projects globally via the registry.

Aryan

Hi All,

Thanks for the Info.

Does this mean that there can be only one metadata file for all the assets on the delivery system and currently there is no way of getting mediabin to publish one metadata file for each asset on the delivery system?

Thanks,
Aryan.

Migrateduser

Possibly... see post further on in thread.... Otherwise, you have the option of writing your own Deployment Agent processing step or working with CSO to accomplish that.

Aryan

Thanks a lot for the info.

Aryan

Could you be more clear as to what "writing your own Deployment Agent processing step" means?

Thanks,
Aryan.

Migrateduser

Normally, this function is used by CSO only, but some customers have been asking about the "extensibility" of Deployment Agent (DA).
I'm attaching a document on Deployment Agent. It is also on the Interwoven Knowledge base and I believe I've posted it before, but you should read it to understand how Deployment Agent is architected.
You can extend the function of Deployment Agent by inserting your own processing step. For example, if you wanted to break the metadata export file up into individual files, you could add a P3.5 step to do that. Deployment Agent's processing steps can be dictated using the registry. Each processing step must expose a COM interface containing the following method signature: Execute(objInfo as Object). The parameter passed in is global information and supplied by the DA executable, so you don't need to supply that. You could write your code using .NET, VB6, C++, or use a scriptlet (VBScript). The code would need to create each metadata file and place it in the proper location. This location is under a temporary folder whose name is randomly generated by DA. You should be able to get the path from the DerivativeQueue table. The code would also need to update the DA database to add an additional derivative row in the DeriviateQueue table for each metadata file created. The count of derivatives and targets would also need to be updated in the ScheduleQueue table. I think that's it. You may need to read the DerivativeQueue table to get the path for each derivative... or it may be in the metadata.exp file along with the derivative name. The name in the metadata.exp file is the "derived" name using rules you may have specified for a derivative.
The normal DA step P4 reads the DerivativeQueue table and populates the TargetQueue table.
If you look at the stored procedures for DA, you'll see a naming convention that includes the processing step in the name (p1, p2,...). If you run a schedule with a target that fails, you'll be able to inspect the contents of the transient tables (those named with Queue suffix). These tables contain working information for a schedule and are deleted if a schedule runs successfully. You have the ability of adding a post P5 step and inspect these tables even when a schedule does complete successfully, because the deletion is delayed until the next DA polling interval.
If you're interested in pursuing extending DA on your own, I can provide direct assistance.

Migrateduser

I was posting a reply to one of your other questions about getting metadata and there may be a way to extract metadata for an asset in an xml file, one per asset. You could specify two derivatives for your schedule. One derivative would cause the asset to be downloaded and the other derivative would use the XML encoder task to download a separate XML file per asset. I'm guessing you would want to place the download asset content file and the asset metadata file at the same location. If you have a static set of metadata you want to include for each asset, you can set up a special DA user as described in a previous post to force only that metadata to be exported.

I just tried this in version 4.5.3 but there seems to be a bug in MediaBin server. It is including metadata that the user doesn't have access to in the export file. Therefore, the concept of limiting metadata in the file based on user permissions won't work here.

Aryan

Thanks a lot for the info.

I shall try the same and get back to you.

Thanks a lot,
Aryan.

Aryan

Thanks a lot.

Aryan.

Aryan

It works. Thanks a lot.

But it does not seem to produce the XML in the format I require.

Attached are the formats in XML is generated when ECM Connector imports the images on to its repository and the format in which the Deployment agent generates the same.

Is there a way of making these two consistent.

I want the Deployment agent to generate metadata in a format ECM connector does it.

Thanks,
Aryan.

Migrateduser

The XML file is not produced by Deployment Agent. It is produced by MediaBin server, so there's no way to change it without rewriting that function within the server.
You would need to write your own script to reformat the XML file to conform to the requirements of the ECM connector. That could be added as a processing step in Deployment Agent or you could do it with a script running at your target location.

Aryan

Thanks for that info.

So the bottomline is that I cannot generate metadata XML using Deployment agent in the format ECM Connector does, without customization, right?

But, is there a way to remove unwanted things like "preview" metadata fields from the XML generated by MediaBin Deployment agent? Metadata remover does not seem to be of much use as far as this is concerned.

Also, the format of the image seems to be in number like 0,1,2 etc. How do I map it real format types like jpg, bmp etc?

PS: The "delivery system" I talk about in my posts is the J2EE server which acts as a frontend for the CM system.

Thanks,
Aryan.

Aryan

--------------------------------------------------------------------------------

Thanks for that info.

So the bottomline is that I cannot generate metadata XML using Deployment agent in the format ECM Connector does, without customization, right?

Is there a way to add all the default metadata like keywords, author etc to the metadata XML generated during the Deployment agent deployment step. When I say default metadata, I want all the asset metadata to be put in the XML. Metadata adder does not seem to be of much use as far as this is concerned.

But, is there a way to remove unwanted things like "preview" metadata fields from the XML generated by MediaBin Deployment agent? Metadata remover does not seem to be of much use as far as this is concerned.

Also, the format of the image seems to be in number like 0,1,2 etc. How do I map it real format types like jpg, bmp etc?

PS: The "delivery system" I talk about in my posts is the J2EE server which acts as a frontend for the CM system.

Thanks,
Aryan.

Aryan

--------------------------------------------------------------------------------

Thanks for that info.

So the bottomline is that I cannot generate metadata XML using Deployment agent in the format ECM Connector does, without customization, right?

Is there a way to add all the default metadata(read only metadata) like keywords, author etc to the metadata XML generated during the Deployment agent deployment step. When I say default metadata, I want all the asset metadata(including all the readonly metadata) to be put in the XML. Metadata adder does not seem to be of much use as far as this is concerned.

But, is there a way to remove unwanted things like "preview" metadata fields from the XML generated by MediaBin Deployment agent? Metadata remover does not seem to be of much use as far as this is concerned.

Also, the format of the image seems to be in number like 0,1,2 etc. How do I map it real format types like jpg, bmp etc?

PS: The "delivery system" I talk about in my posts is the J2EE server which acts as a frontend for the CM system.

Thanks,
Aryan.

maheshd

MediaBin_MTS,

You have mentioned that 'Deployment Agent's processing steps can be dictated using the registry'.I have checked MediaBin registry entries but i can not find anything related to processing steps.It would be great if you can give some more details on this.

Thanks,
Mahesh

nestedTable_report.rptdesign

doubleLine.JPG

Migrateduser

I believe you're talking about the metadata file produce by MediaBin server... not metadata.exp, which is produced by Deployment Agent.
There is no way to customize the metadata xml file produced by MediaBin server (which is one xml file per asset).

You should be able to use the XML Encoder to export Editable metadata. That should include keywords.

The enumerated values for asset type and others aren't published. We do publish a list of them for the API, but the API maintains its own constant list and internally maps those to what is used by the server. That way the server can change the values without interfering with the COM interface.

If you provide a list of enumerated values you need resolved, I can post the mappings from enumerated value to strings.

Aryan

Hello MediaBin_MTS,

Editable metadata does not seem to include the "Keywords" metadata.

Also the metadata such as file format seems to be in number such as 0,1,2 etc.

Can you please enlighten me as to what these are?

I shall supply the list of values by EOD today.

Thanks,
Aryan.

Aryan

Also, Do you mean to say that the keywords are included in the metadata file only if the "Metatagger" task is present as a part of the derivatives added to the deployment agent steps?

Aryan.

-----------------------------------------------------------------------

Hello MediaBin_MTS,

Editable metadata does not seem to include the "Keywords" metadata.

Also the metadata such as file format seems to be in number such as 0,1,2 etc.

Can you please enlighten me as to what these are?

I shall supply the list of values by EOD today.

Thanks,
Aryan.

Migrateduser

I just tested and keywords appeared in an xml file created by MediaBin server when Deployment Agent requested retrieval of an asset using a task that includes the XMLEncoder. The XMLEncoder is configured to export editable metadata. Also, keywords must be assigned to the asset. If this metadata is not assigned to the asset, keywords are not included in the exported xml file that MediaBin server creates.
I've attached the exported xml file.

Migrateduser

Here is a list of the asset type enumerated values and the associated asset type string. We don't publish this relationship because it can be retrieved using the COM API. A "choice list" of enumerated values and their associated strings is returned by the COM API function named GetChoices. This is a method on the MBMetadataDefinition object. An array is returned with the odd numbered elements being the enumerated value and the even numbered elements the associated string.
Here's a dump of that information:
0 - Unknown
5 - BMP
20 - Corel Draw
18 - Flash
4 - GIF
14 - ICC Profile
1 - JPEG
21 - Macromedia Freehand
15 - MS Office Document
16 - Multimedia File
12 - Non-Image
11 - PhotoCD
13 - PDF
7 - Photoshop EPS
22 - Placeholder
10 - PNG
2 - Photoshop
24 - QuickTime
3 - STING
9 - TARGA
6 - TIFF
8 - Vector EPS
2000 - MrSID
2001 - JPEG2000
1000 - MS Word Document
1001 - MS Excel Document
1002 - MS PowerPoint Document
1003 - MS Visio Document
17 - Quark