I can't convert to pdf or pptx on Windows Subsystem for Linux

[oren]

I am using an Ubuntu shell inside Windows (Windows Subsystem for Linux). Running --server and --html works but when I try --pdf or --pptx the command continues forever and I never get the cursor back.

Any idea how to troubleshoot this?

[yhatt]

I'll test later in same environment to try reproduce. Some advice I can tell now:

• PDF/PPTX conversion in WSL requires Chrome for Windows like /mnt/c/Program Files/Google/Chrome/Application/chrome.exe.
  https://github.com/marp-team/marp-cli/blob/master/src/utils/find-chrome.ts#L11-L12
• The latest Marp CLI is using Puppeteer v1.20.0 to convert. If you are using outdated Chrome <= 76, it may not return the output because of difference of PDF generation protocol.
  Upgrade Marp Core to v0.10.2 #106 (comment)

[yhatt]

If you want troubleshooting in myself environment, DEBUG=* env would help to check working some dependent modules. (Please careful it might output very huge log)

DEBUG=* marp xxxxxx.md

[oren]

awesome advice. I'll try it out in a few days and will report back

[yhatt]

I've confirmed the same issue on Marp for VS Code + Remote WSL. (marp-team/marp-vscode#89)

According to debug log (DEBUG=*), Puppeteer seems to fail execute installed Chrome in Windows host.

[  INFO ] Converting 1 markdown...
  puppeteer:protocol SEND ► {"method":"Target.setDiscoverTargets","params":{"discover":true},"id":1} +0ms
  puppeteer:protocol ◀ RECV {"method":"Target.targetCreated","params":{"targetInfo":{"targetId":"170035E506EBF2213D1473583EEC98A7","type":"page","title":"","url":"about:blank","attached":false,"browserContextId":"E9E0AB510BF8FD88BC02217A79F4C278"}}} +4ms
  puppeteer:protocol ◀ RECV {"method":"Target.targetCreated","params":{"targetInfo":{"targetId":"26db2069-c3c7-4a29-9e15-20b5f25cc0b6","type":"browser","title":"","url":"","attached":false}}} +0ms
  puppeteer:protocol ◀ RECV {"method":"Target.targetCreated","params":{"targetInfo":{"targetId":"720114e9-3491-4b37-84a6-ff9968c0057f","type":"browser","title":"","url":"","attached":true}}} +1ms
  puppeteer:protocol ◀ RECV {"id":1,"result":{}} +0ms
  puppeteer:protocol SEND ► {"method":"Target.createTarget","params":{"url":"about:blank"},"id":2} +1ms
  puppeteer:protocol ◀ RECV {"method":"Target.targetCreated","params":{"targetInfo":{"targetId":"5ECF9FB590AA1CDCDCF48B81A23FA7D6","type":"page","title":"","url":"","attached":false,"browserContextId":"E9E0AB510BF8FD88BC02217A79F4C278"}}} +2ms
  puppeteer:protocol ◀ RECV {"id":2,"result":{"targetId":"5ECF9FB590AA1CDCDCF48B81A23FA7D6"}} +1ms
  puppeteer:protocol ◀ RECV {"method":"Target.targetCrashed","params":{"targetId":"5ECF9FB590AA1CDCDCF48B81A23FA7D6","status":"failed to launch","errorCode":2}} +1ms

[yhatt]

@oren I've confirmed a patched v0.15.1 is working PDF conversion in the WSL environment correctly.

Please report again if still failing conversion in the latest version of CLI.