I can only assume it happened during normal operations. In this environment (test), we only build new reports and export them to import in to our production environment.

The only thing i can think of is that we utilize AWS ECS (docker) and there are times it will spin up/down new instances. I wouldn't think this would be a problem and has not happened in the past until this new version (3.7.1). The only other difference is the number of items that we have. We are constantly adding more folders and templates (normally we have 4-5 files per folder). But compared to what you guys run on your public side, it's nothing.

Is there a way i can recreate a folder that goes missing (with the same shortId)? Is this something that could get lost in S3 or is a folder metadata only?

We have been having an issue with files or folders being lost. Thankfully we have a backup of them, but we would like to understand how this could happen. Right now we have an issue where it is erroring with "Error: Unable to find entity with shortid bGvzATI". This happens when I try to click on a file that is showing at the root level (we don't normally have files at the root level).

This is now stopping us from being able to find while folder it was to recreate. I can pull logs from AWS ECS if it would be helpful, but i don't know the exact time it happened.

Tech details:
version 3.7.1
Running as a docker on AWS
FileStore config:
"fs-store": {
"persistence": {
"provider": "aws-s3"
},
"sync": {
"provider": "aws-sns"
},
"compactionInterval": 300000
},

Any info on how to stop this or how to find out what's going on would be appreciated.

I don't know i did wrong the other day, but adding Encoding to it worked!

Thanks...we can now use JSReports for document conversion of Docx to PDF!!

I will give that a try again today. I tried it last time and i got an error about a Set not being allowed on a read only or something like that. I will get you the exact error message though.

To save setting up a new docker to handle file conversion, we are wondering if we can use JSReports to handle the conversion from docx to pdf. We have already proved that we can use unoconv to take the docx and get the PDF that we want. Now we are trying to do it from our .NET app by passing in the docx file to the template and getting it to do the unoconv for us. We are failing at getting it to pass in the whole template. We have verified that the byte array and memory stream are valid. We made the assumption that we need to base64 encode it. When trying to pass the encoding type, it errors that it's a read-only property.

Here is the pseudo-code for what we do:
Create a template in JSReports but don't put set the Docx, enabled Unoconv and output to PDF
Load the template via OData to .NET Code
Load the Docx from external source and set it using the follwoing:

	testDocx = new Docx() {
		TemplateAsset = new Asset()
		{
			Content = {MyBase64String}
		}
	};
	conversionTemplate.Docx = testDocx;

Then call RenderAsync() passing in the template and NULL for data with the correct options.

I get this error:
Report "word" render failed.

Error while executing docx recipe. end of central directory record signature not found
Error: Error while executing docx recipe. end of central directory record signature not found
at module.exports (/app-rprt/node_modules/@jsreport/jsreport-core/lib/shared/createError.js:10:13)
at WorkerReporter.createError (/app-rprt/node_modules/@jsreport/jsreport-core/lib/shared/reporter.js:47:12)
at /app-rprt/node_modules/@jsreport/jsreport-docx/lib/processDocx.js:93:20
at async module.exports (/app-rprt/node_modules/@jsreport/jsreport-docx/lib/recipe.js:38:28)
at async invokeRender (/app-rprt/node_modules/@jsreport/jsreport-core/lib/worker/render/render.js:95:5)
at async WorkerReporter._render (/app-rprt/node_modules/@jsreport/jsreport-core/lib/worker/render/render.js:147:7)
at async /app-rprt/node_modules/@jsreport/jsreport-core/lib/worker/reporter.js:179:19
at async Domain.<anonymous> (/app-rprt/node_modules/@jsreport/advanced-workers/lib/workerHandler.js:141:19)
caused by: Error: end of central directory record signature not found
at /app-rprt/node_modules/yauzl/index.js:187:14
at Immediate.<anonymous> (/app-rprt/node_modules/yauzl/index.js:624:38)
at processImmediate (node:internal/timers:464:21)
at process.topLevelDomainCallback (node:domain:152:15)
at process.callbackTrampoline (node:internal/async_hooks:128:24)

Any assistance in being able to pass a full template content over and have it unoconv to pdf would be much appreciated. It seems that if it doesn't have a ShortId to work with, it fails.

I know this is a nit-pick request, but it's a pain that i have to resize my areas in the Studio every time i reload the page. Do you think it would be possible to store the div sizes in a cookie or something and then when we go back to the page we don't have to resize each panel to where we prefer it? When you have lots of folders on the left and the folder names are long, it's just a bit annoying.

Just a suggestion.

Thanks

You solved my issue!!!

The reason I thought it was odd was because I actually had the wrong syntax

In my example i was doing {{#xlsxAdd "xl/drawings/drawing1.xml" "xdr:wsDr.twoCellAnchor"}}

In your example you did {{#xlsxAdd "xl/drawings/drawing1.xml" "xdr:wsDr.xdr:twoCellAnchor"}}

I completely missed the 2nd namespace which is what caused things to not work at all. This is why things looked odd to me but I couldn't figure out why.

As always...thanks for everything! It's a great product!

You should consider bankers rounding.
https://stackoverflow.com/questions/3108986/gaussian-bankers-rounding-in-javascript

I have a scenario where i want to reuse an image within my Excel spreadsheet. In the sheet I have an image (red/yellow/green circle for status). Rather than use xlsxAddImage which will create a copy of the image X number of times, I have created an Excel Template that has the image in there once and then I am using the twoCellAnchor concept and referencing the same image by rID#. This works fine when working with OpenXML and verifying everything in the Office productivity tool. However, since this is in a different namespace than the original x: used in Row, it seems that the xlsxAdd has a big issue here. Not sure why, but it's not working. here's what I am seeing.

First, I would normally thing i would reference it by doing {{#xlsxAdd "xl/drawings/drawing1.xml" "wsDr.twoCellAnchor"}}, but instead I have to do {{#xlsxAdd "xl/drawings/drawing1.xml" "xdr:wsDr.twoCellAnchor"}} otherwise it creates another XML node without the namespace. This is not an issue, but is slightly odd.

Second (and the real issue) is that my content should be something like
<xdr:twoCellAnchor editAs="oneCell" xmlns:xdr="http://schemas.openxmlformats.org/drawingml/2006/spreadsheetDrawing">
...
</xdr:twoCellAnchor>

However, the result in the file generated strips out the xdr: from the node. This is causing Excel to not render anything (as well as the productivity tool not knowing what type of node it is). I have also tried
<xdr:twoCellAnchor editAs="oneCell">
and that didn't help anything.

Any help I can get on handling the namespace appropriately would be appreciated.

I have about 300 reports that I am converting from SSRS to JSReport. In the Excel rendering, many of the reports reuse the same functionality with xslxReplace or Merge to do the same thing per report (such as build a header or create a footer or add Properties in Excel). I was hoping there would be a way i could put these reusable xslxReplace or merge pieces in a component or childtemplate in case i need to change it in the future. Then i could update that functionality in 1 place and not have to change the copy/paste in each Excel portion.

Is there a way to do this? I have tried with Component and ChildTemplate and can't seem to do it anywhere. A super simple reused piece would be setting the SheetName to a value in the data.

{{!-- Adding the name to sheet1 of the workbook --}}
{{#xlsxMerge "xl/workbook.xml" "workbook.sheets[0].sheet[0]"}}
<sheet name="{{sheetName}}"/>
{{/xlsxMerge}}