Right-aligned HTML paragraph breaks inline font styling

[bgdncz]

Bold/italic/underline text in p tags works for default parameters, but setting align="right" breaks the layout:

[Lucas-C]

Thanks for the report @boghison!

Could you please provide your source code for this test?

[bgdncz]

@Lucas-C I got it from the write_html documentation. I also tried it with the example here by changing the first paragraph to <p align="right"><b>Hello</b> world. <u>I am</u> <i>tired</i>.</p>:

PS Thank you for your work!

[bgdncz]

I think it’s because aligned paragraphs get written with multi_cell, so this might be related to #108

[Lucas-C]

You're totally right, implementing #108 would help solving this!

I'll try to get back to this other issue in the next weeks.

[Lucas-C]

Quick update: now pending on #171

[Lucas-C]

Quick status update: solving this would require rendering the whole <p> block with FPDF.multi_cell().
But how to combine that with parsing of HTML tags like <b> or <u> inside the paragraph?
A solution could be to convert those to Markdown markers, but that's not an elegant solution...

#91 is also closely related

[gmischler]

The reason for this problem is that the html.py module is currently rather a dirty hack than a correct implementation.

Among other shortcomings, it just copies all the whitespace fromt the HTML data to the output, instead of condensing each run to a single space, which very often results in a very weird layout.
Using multi_cell() to align text is also an extremely quick-and-dirty approach, since as demonstrated above, it will necessarily start a new line each time the character formatting changes. Fixing this part will become a lot easier, once phase 2 of the proposed additions to write() are implemented (see #339).

The HTML parsing, and several other shortcomings, will have to be fixed seperately.

[gmischler]

I've just checked in #897, which should solve this problem.