Dynamic Document Generator

Hi Ilan,

Would like to mention that we have encountered the below error while re-deploying the plugin to version 1.7.5 in one of the environments.

We had saved XSLParameter CDT initially. Then to upgrade the plugin, i tried to redeploy the same but got the error as mentioned below.

Hence, have deleted the CDT from Appian Designer and tried to redeploy the plugin. It seems like the old version of CDT still exists in the environment and is not allowing to deploy the plugin.

Please let us know if you have the resolution.

2019-12-01 15:22:01,697 [ajp-nio-8009-exec-4] INFO  com.appiancorp.plugins.AppianPluginManager - Updating plugin 'DynamicDocumentGen:1.7.5 There was a problem loading the module descriptor: TextDocFromXslt.<br/>com.appiancorp.suiteapi.type.exceptions.InvalidTypeException: com.appiancorp.suiteapi.common.exceptions.AppianException: Invalid Type: Unsupported type [Lcom.appiancorp.ps.xmlfo.types.XslParameter; (APNX-1-4047-000)' to 'DynamicDocumentGen'
2019-12-01 15:22:01,902 [ajp-nio-8009-exec-4] ERROR com.appiancorp.type.config.plugin.PojoDatatypesConfig - The data type [uuid={urn:appian:ps:dynamicdocgen}XslParameter] in plug-in [key=DynamicDocumentGen, module=XslParameter] was not imported because it or another data type in the plug-in module was already present on the system with a different structure, and the version number of this plug-in is not greater than the deployed plug-in that last updated this data type. You must increment the plug-in version number in the appian-plugin.xml file and deploy this plug-in again. If the problem persists after incrementing the version number, check that all data type precedents of the data type are declared in the same plug-in module or are listed before this data type in the appian-plug.xml. (APNX-1-4165-004)
2019-12-01 15:22:01,922 [ajp-nio-8009-exec-4] INFO  com.appiancorp.process.admin.LoadACSchemas - Attempted to skip registration of [com.appiancorp.ps.xmlfo.ss.PdfFromDocx3] as it is current [1.7.5], but the ActivityClassSchema was not registered, so re-registering
2019-12-01 15:22:01,958 [ajp-nio-8009-exec-4] INFO  com.appiancorp.process.admin.LoadACSchemas - Attempted to skip registration of [com.appiancorp.ps.xmlfo.ss.DocxFromXhtmlWithStyling] as it is current [1.7.5], but the ActivityClassSchema was not registered, so re-registering
2019-12-01 15:22:01,987 [ajp-nio-8009-exec-4] INFO  com.appiancorp.process.admin.LoadACSchemas - Attempted to skip registration of [com.appiancorp.ps.xmlfo.ss.DocxMergeWithSeparator] as it is current [1.7.5], but the ActivityClassSchema was not registered, so re-registering
2019-12-01 15:22:02,016 [ajp-nio-8009-exec-4] ERROR com.atlassian.plugin.manager.DefaultPluginManager - There was an error loading the descriptor 'TextDocFromXslt' of plugin 'DynamicDocumentGen'. Disabling.
com.atlassian.plugin.PluginException: com.appiancorp.suiteapi.type.exceptions.InvalidTypeException: com.appiancorp.suiteapi.common.exceptions.AppianException: Invalid Type: Unsupported type [Lcom.appiancorp.ps.xmlfo.types.XslParameter; (APNX-1-4047-000)

I am using the PDF from Docx smart service included in this plugin. One of the inputs is "P D F Options". I cannot find any documentation or comments explaining how the Options are configured. Does anyone have more information on this? 

DOCX from dynamic template seems to be turning single spaces into double spaces randomly. Any advice on how to fix this. Our template is very simple.

Hey Illan, your post was probably the most helpful among anything else I found on the web. So thanks for that! I do have one question though: Do you know a way to insert a page break into docx so that the first element on the new page is a Heading1 element? The problems I've encountered so far is that if I wrap a page break around <p> tags, the first line of the new page is empty. And if I put in multiple classes inside a <p> tag, only the first class renders. Let me know if you know of any workarounds!

Sorry - Community apparently doesn't auto-notify about replies here, so I never saw your followup questions

Hi, while we upgraded from 18.4 to 19.1, the PDF from HTLM stopped working. We recently deployed the upgraded plugin version (1.7.2), however this smart service still does not provide the required output. The process is stuck at the node and then errors out after some time. The only details we find in logs are "2019-07-01 13:02:14,509 [ajp-nio-8009-exec-2232] ERROR com.appiancorp.common.struts.BaseAction - Error executing action: Could not initialize class javax.swing.RepaintManager java.lang.NoClassDefFoundError: Could not initialize class javax.swing.RepaintManager "
Can you help with this issue?

For anyone using DOCX from XHTML a few useful notes for you: 

• To access MS Word standard styles (like Title, Heading1, Normal, etc.) format your HTML body as so
  • Title Example: <p class="Title">Your Title Here</p>
  • Heading1 Example: <p class="Heading1">Heading Text</p>
  • Basically the point being is that you should us <p> tags for all of your text and specify the formatting class as the class
    • This will allow MS Word to register the style for the navigation pane
• It's worth noting that although MS Word identifies the style applied to text, it will not use the default style of MS Word (this tripped me up for a while)
  • You should create a document in word with some styles applied and save as an HTML and grab the relevant CSS style definitions out of the CSS and put it in a constant or something that you add to the <head> of your HTML by default
  • Make sure that you set your CSS class names to p.<msWordStyle> (eg: p.Title, p.Heading1, etc.)
    • This will allow your class to both be recognized by MS Word and will allow you to apply a style to the text so that it looks decent on openning the document
  • It's also worth noting that the CSS created by MS Word has non-standard tags in it generally of the format "mso-style..."
    • You can remove these as they won't do anything for you and will just take up space
• Also remember the main difference between XHTML and HTML is that all attributes must have "" around them!!!

Happy document generation everyone!

The plugin does not do anything special to handle fonts. It uses the Apache FOP classes to initiate the parsing and transformation. Please look at the class XmlFoConverter inside the plugin for more details. 

I would suggest to do some research on how to handle fonts with Apache FOP transformation and either update the XSLT stylesheet or update the plugin code.

Hi, using PDF from XSL-FO Transformation gives us good resulting outputs but there is issue with fonts where the plugin does not react to font set inside XSLT (Georgia) and output is each time in Times font. Is there some solution / workaround? 

It appears that it was an issue with the plugin framework, not the plugin itself. Our engineering department released a hotfix to address this.

Please create a support case to request the latest hotfix to be installed.