Hi,

You can't. It is not implemented in web services API.

Suppose you have a publication:

• identifier: GUID-75CD4383-F94E-49AA-8182-9776535F8522
• version: 1
• format: PDF (A5 Booklet)

You can query the MetaDataAssist2.0 > GetLOVValues [deprecated], the list you are looking for is normally called DOUTPUTFORMATNAME.

You can query the metadata of this object with PublicationOutput2.5 > GetMetaData, and you'd get back something like:

<ishobjects>
   <ishobject
      ishref="GUID-75CD4383-F94E-49AA-8182-9776535F8522"
      ishtype="ISHPublication" ishlogicalref="7025"
      ishversionref="7028" ishlngref="7029"/>
</ishobjects>

Grab the ishlngref attribute's value and use PublicationOutput2.5 > GetDataObjectInfoByIshLngRef to get information on that. You get back something like:

<ishdataobjects>
   <ishdataobject ishlngref="7029" ishdataref="7030"
      ed="GUID-450AE758-34F0-4F6A-8A34-23FDCADF931C"
      edt="EDTPDF" size="3195659"
      mimetype="application/pdf" fileextension="pdf"/>
</ishdataobjects>

I guess the ed attribute is the same as psEdGUID in PublicationOutput2.5 > GetDataObjectByIshLngRef, which could be used to fetch the blob. Except this function call is billed as "internal" and it clearly says that it is " not implemented in the InfoShare webservices".

Sorry.

HTH

Joakim

There is example code available in C# online. A couple of months ago we released a PowerShell business automation library on SDL's open source corner, see https://sdl.github.io/#dita It is called ISHRemote where the classic ISH prefix refers to Knowledge Center Content Manager's code name 'InfoShare'.

As you can read on github.com/.../, ISHRemote is a PowerShell module on SDL Knowledge Center Content Manager. Its goal is business automation on top of the Component Content Management System (InfoShare). This library is constructed close to the "Web Services API" to:

• allow business logic automation ranging from triggering publishing into the continuous integration pipeline over legacy data correction up to provisioning
• show case code examples and API best practices

So the library allows you to automate using the PowerShell language and is available on www.powershellgallery.com/.../ Most of it is written in C# where probably the following implementations of cmdlet's ProcessRecord function could inspire you for this thread
A) github.com/.../PublishIshPublicationOutput.cs
B) github.com/.../GetIshPublicationOutputData.cs

Unfortunately, Python is not my strong suit. I program mostly in java and C#. The java code is fairly extensive in detail. I can send some code samples directly to you if needed, but I've listed the basic logic below. The basic API calls, btw, are the same regardless of asmx vs. svc, though obviously the authentication bits are different.

1.) Call PublicationOutput25.GetMetaData to get metadata for the publication.
(You will need to send the authorization context, publication GUID, language combination, output format name, publication version, and requested metadata (can be empty))

docs.sdl.com/.../pub.xql

2.) Parse the xml returned from the GetMetaData call

Get the value of ishobject[@ishlngref]

3.) Call PublicationOutput25.GetDataObjectInfoByIshLngRef
(You will need to send the authorization context, and the lngref that you obtained in step 2 above)

https://docs.sdl.com/LiveContent/web/pub.xql?action=home&pub=SDL%20Knowledge%20Center%20full%20documentation-v2.1.2&lang=en-US#docid=GUID-B3A0D845-D992-444D-B1CE-4267CDA0A865&addHistory=true&query=GetDataObjectInfoByIshLngRef&scope=&tid=b6741009-bdf0-49d7-b047-35445d0af6a2&filename=GUID-B3A0D845-D992-444D-B1CE-4267CDA0A865.xml&resource=&inner_id=&toc=false&eventType=lcContent.loadDocGUID-B3A0D845-D992-444D-B1CE-4267CDA0A865&url=/LiveContent/web/search.xql%3Fc%3Dt%26pub%3DSDL+Knowledge+Center+full+documentation-v2.1.2%26lang%3Den-US%26action%3Dsearch%26query%3DGetDataObjectInfoByIshLngRef&sid=lcSearch.runSearch1481907555400&currentQuery=GetDataObjectInfoByIshLngRef&currentScope=

4.) Parse the xml returned from the GetDataObjectInfoByIshLngRef call

Get the following values:

lngref = ishdataobject[@ishlngref]
outputGUID = ishdataobject[@ed]
size = ishdataobject[@size]

5.) Make successive calls to getNextDataObjectChunkByIshLngRef to get the data from the blob

docs.sdl.com/.../pub.xql

This is the tricky part, and hard to explain. In my java code, it looks like this:

int offsetInt = 0;
int maxBytesInt = 200000;
ByteArray myByteArray = new ByteArray(); // This is a special class I created to handle progressively expanding byte arrays
while (offsetInt < size) { // The size is what you got in step 4 above
if (offsetInt + maxBytesInt > size) {
maxBytesInt = size - offsetInt;
maxBytesInt++;
}
byte[] receivedBytes = getNextDataObjectChunkByIshLngRef(authid, lngRef, outputGUID, offsetInt, maxBytesInt); // lngRef is what you got in step 2, outputGUID is what you got in step 4
myByteArray.add(receivedBytes, receivedBytes.length);
offsetInt = offsetInt + receivedBytes.length;
}

That's pretty much it, and you can then do whatever you want with the byte stream.
Note that sometimes the PDF byte streams can contain extra nulls that need to be trimmed off before saving out to a file.

Unfortunately, Python is not my strong suit. I program mostly in java and C#. The java code is fairly extensive in detail. I can send some code samples directly to you if needed, but I've listed the basic logic below. The basic API calls, btw, are the same regardless of asmx vs. svc, though obviously the authentication bits are different.

1.) Call PublicationOutput25.GetMetaData to get metadata for the publication.
(You will need to send the authorization context, publication GUID, language combination, output format name, publication version, and requested metadata (can be empty))

docs.sdl.com/.../pub.xql

2.) Parse the xml returned from the GetMetaData call

Get the value of ishobject[@ishlngref]

3.) Call PublicationOutput25.GetDataObjectInfoByIshLngRef
(You will need to send the authorization context, and the lngref that you obtained in step 2 above)

https://docs.sdl.com/LiveContent/web/pub.xql?action=home&pub=SDL%20Knowledge%20Center%20full%20documentation-v2.1.2&lang=en-US#docid=GUID-B3A0D845-D992-444D-B1CE-4267CDA0A865&addHistory=true&query=GetDataObjectInfoByIshLngRef&scope=&tid=b6741009-bdf0-49d7-b047-35445d0af6a2&filename=GUID-B3A0D845-D992-444D-B1CE-4267CDA0A865.xml&resource=&inner_id=&toc=false&eventType=lcContent.loadDocGUID-B3A0D845-D992-444D-B1CE-4267CDA0A865&url=/LiveContent/web/search.xql%3Fc%3Dt%26pub%3DSDL+Knowledge+Center+full+documentation-v2.1.2%26lang%3Den-US%26action%3Dsearch%26query%3DGetDataObjectInfoByIshLngRef&sid=lcSearch.runSearch1481907555400&currentQuery=GetDataObjectInfoByIshLngRef&currentScope=

4.) Parse the xml returned from the GetDataObjectInfoByIshLngRef call

Get the following values:

lngref = ishdataobject[@ishlngref]
outputGUID = ishdataobject[@ed]
size = ishdataobject[@size]

5.) Make successive calls to getNextDataObjectChunkByIshLngRef to get the data from the blob

docs.sdl.com/.../pub.xql

This is the tricky part, and hard to explain. In my java code, it looks like this:

int offsetInt = 0;
int maxBytesInt = 200000;
ByteArray myByteArray = new ByteArray(); // This is a special class I created to handle progressively expanding byte arrays
while (offsetInt < size) { // The size is what you got in step 4 above
if (offsetInt + maxBytesInt > size) {
maxBytesInt = size - offsetInt;
maxBytesInt++;
}
byte[] receivedBytes = getNextDataObjectChunkByIshLngRef(authid, lngRef, outputGUID, offsetInt, maxBytesInt); // lngRef is what you got in step 2, outputGUID is what you got in step 4
myByteArray.add(receivedBytes, receivedBytes.length);
offsetInt = offsetInt + receivedBytes.length;
}

That's pretty much it, and you can then do whatever you want with the byte stream.
Note that sometimes the PDF byte streams can contain extra nulls that need to be trimmed off before saving out to a file.

Hi,

Thanks! This was easy to port to Python. I created a Github gist in case someone wants to take a look.

https://gist.github.com/jockster74/f6f0795a204575b8414ddc4fb05d84ff