I ended up by simply appending my new data to req.data, like this:

    const blueprint = await jsreport.render({
      template: {
        shortid: 'rJxZpXJYwB',
        recipe: 'chrome-pdf'
      },
      data: {
        ...res.data,
        test: 'TEST ABC'
      }
    });

What confuses me is that I don´t need to append any data if no new data is needed.

Is the following true?

• If no data is supplied to jsreport.render, jsreport will autmatically use res.data.

I need to append a report with differend page size (A3), so I need to use append. First I'm rendering the repor and then I'm appending it with pdfUtils.

I know that I can add data to the render command, but this doesn't seem to be necessary. The original req.data seems to be available in the rendered report anyway.

I need to add some specific information to the appended report. How would I insert this information via the render command and how would I reach it from within the rendered report?

Here is how I do it right now, without inserting any new data:

const jsreport = require('jsreport-proxy')

async function afterRender(req, res) {
  const blueprint = await jsreport.render({
      template: {
          shortid: 'rJxZpXJYwB',
          recipe: 'chrome-pdf'
      },
  })

  res.content = await jsreport.pdfUtils.append(res.content, blueprint.content)
}

I discovered the culprit.

I'm on an M2 Mac, so I have to use FROM --platform=linux/amd64 jsreport/jsreport:3.11.2 in my Dockerfile to be able to run the image at AWS.

When I build a local image for running in my dev-environment I need to use FROM --platform=linux/arm64 jsreport/jsreport:3.11.2.

I'll have to come up with a solution for differentiating my local builds. Does anyone know if there are ARM containers available at AWS Fargate? It would be best to run exacrly the same image locally and in production at AWS.

Today my Chrome-PDF reports stoped working. They work fine if I switch to HTML rendering.

I run jsreport from a local Docker image while developing. Does the local Chrome client influence the rendering, or is everything done with Puppeteer from inside of the container?

The last message in the log is mostly:
Executing recipe chrome-pdf

Sometimes I get one more row:
Converting with chrome HeadlessChrome/102.0.5005.182 using chrome-pool strategy rootId=0p0two50psg6i71, id=0p0two50psg6i71

I'm on v3.9.0, but I also tried v3.11.2 with the same result.

All reports worked yesterday. The only thing I can think of as changed would be:

• maybe Chrome is updated, but I don't recall accepting an update
• I rebuilt my Docker image, but I have done this frequently without any problems

I have also tried 3.9.0-full. I have never used this image before, but I thought it was worth a shot if this is a Puppeteer dependency problem. With this image I got into even more trouble:

I discovered the problem on our old production server with jsreport 2.11.
pdf-A is not supported in V2.

Do you mean that enabling pdf-a in V3 would solve the color problem?

What color profile conversions takes place when a PDF is rendered? Do I need to change something in order for my images to get correct colors?

The image below shows the original jpg to the left and the same image sitting in a pdf from jsreport to the right. The original is saved with colorspace sRGB.

Thanks again.
Maybe some of this should be added to the asstes docs (https://jsreport.net/learn/assets#wrap).

This would have helped me:

• Assets loads faster than components, but must be static
• Assets cannot be nested
• If the content is not static, use components

Thanks for claifying.
I love component, and I use them alot.

Could you clarify a bit more so I can choose the right tool.

• I suppose I could use component for almost everything except binary assets?
• When should I use assets, besides for loading binary files?
• Are assets faster to load than components if the content is static?

I separated the static content (pure CSS) from the part where I need to load binary assets.
The font-part is now a component:
https://playground.jsreport.net/w/msageryd/ZWNLbcpR

Is there any reason why I shouldn't simplify this and put the whole CSS in a component?
Like this:
https://playground.jsreport.net/w/msageryd/PaFIRXBa

Here you go.
https://playground.jsreport.net/w/msageryd/P9rTrXEF

The CSS loads the same font twice, once with {#asset..} and once with {{asset ..}}.
Two classes are defined to use each of the fonts.

Only "legacy" works as I want.

Edit:
I'm not sure why I need url(). That came from very old code. I cannot find any documentation about this function either. Is it really needed?

I'm cyrrently loading fonts like this in my CSS.

@font-face {
    font-family: "pt-regular";
    src: url({#asset Myriad-Pro-Regular.ttf @encoding=dataURI});
    format("ttf");
}

How should the {#asset ..} syntax look like if I want to use the new {{asset ..}} helper?
I tried this, but it didn't work:

@font-face {
    font-family: "pt-regular";
    src: url({{asset "Myriad-Pro-Regular.ttf" "dataURI"}});
    format("ttf");
}