lineHeightFactor is ignored if specified as a text option

[novsstation]

Currently lineHeightFactor is ignored, if it is specified as text option.
It correctly checked for calculating lineHeight variable.
But this variable is not used for calculating descent value (still used global lineHeightFactor from document).
As result, those two options produces the different PDF files

https://github.com/MrRio/jsPDF/blob/cef97fb34eda41a8704c9f3983e680919a328ce4/src/jspdf.js#L3612

Steps to reproduce:

1. Open the sample https://codepen.io/SNovikov/pen/VwbJOzG?editors=0010
2. Export PDF with specified document option

   doc.setLineHeightFactor(1.5);
   doc.text('Some text', 10, 10, { baseline: 'middle' });

3. Export PDF with specified text option

   doc.text('Some text', 10, 10, { lineHeightFactor: 1.5, baseline: 'middle' });

As you can see, PDF files are different

[HackbrettXXX]

Thanks for reporting this. Could you prepare a pull request?

[Megharth]

I have added a test that checks output of pdf generated using setLineHeightFactor() method and one where we pass lineHeightFactor as an option to the text() method. The tests are passing locally by running npm run test-unit and npm run test-local but the test that I included is failing on saucelab. I tried debugging it, but locally it provides the expected results. I have no idea how to debug on saucelab. If anyone can guide me, I am happy to resolve the issue with failing tests on saucelab.

[Megharth]

I saw the karma.conf.js for saucelab and it was taking the jspdf from the dist folder. I created a new build and now all the tests passed. I saw in contribution guide that we should not update files in the dist folder. So, how should I go about this?

[HackbrettXXX]

On CI, the files in dist are built before running the tests. For local testing you can npm run build && npm run test-local. We could/should probably either improve the contribution guide or add a pretest-local npm script similar to pretest.