Powerful New Page Method insert_htmlbox

[JorjMcKie]

Many have waited for this functionality:
Method Page.insert_htmlbox writes HTML/CSS-enriched text into a rectangle of a PDF page.
It has many similarities with methods Page.insert_textbox and TextWriter.fill_textbox.

More important are the significant differences:

• Support of HTML tags and CSS styling instructions being part of the text: all syntax elements supported by Story are possible. As a consequence, all of the following is supported:
  • multiple fonts
  • user-supplied fonts, auto-fallback fonts
  • multiple font sizes and text colors
  • embedded hyperlinks
  • images
  • tables
  • multiple languages with automatic font selection
  • automatic "text shaping" (using HarfBuzz), finally yielding correct output of Hindi, Bengali, Tamil, Sanskrit etc. - text using complex scripting systems like Devanagari, right-to-left languages and many more.
• No more worries whether the boundary box is large enough: The content will automatically be scaled down to fit. The extent of acceptable down scaling can be specified.
  • As before: if no downscaling was necessary, a float is returned indicating the unused part of the rectangle.
• Content can be rotated by multiples of 90° as with the other methods.
• Text alignment and other effects, including font specification can be (and must be) achieved via HTML or CSS syntax - there are no method parameters for this.
• As an alternative to a text string, a Story object can be given. This allows for all the preparative adjustments achievable by manipulating the Xml object Story.body prior to executing page.insert_htmlbox(bbox, story).

Please refer to a collection of simple examples in the HOW TO GUIDE.

[ericpias]

I am very interested in using insert_htmlbox. I upgraded to 1.23.8 and fitz.version now reports 1.23.8. The docs say that page.insert_htmlbox is new for 1.23.8. However, the insert_htmlbox is not available on page in 1.23.8. Any ideas?

[JorjMcKie]

This method is available for the new, "rebased" version of PyMuPDF.
For a short period of time, you must use the import statement import fitz_new as fitz. During the course of this month, we will enter phase 2 and switch versions, which will make rebased the default fitz and keep the classic version available as fitz_old as a fallback.

[ericpias]

Thanks.

[lexxwork]

How to remove/clear/erase all content made by insert_htmlbox leaving background image/color int the defined rect?
Maybe i should ask this question on mupdf page?

[lexxwork]

seems they have fixed it

https://bugs.ghostscript.com/show_bug.cgi?id=707560#c2

[JorjMcKie]

Yes, I know they have 😎.

[lexxwork]

@JorjMcKie, i did update to latest version PyMuPDF-1.23.26, the issue is still there, so it means you did not update mupdf?

[JorjMcKie]

It is in the next version ...

[JB2001216]

Hi @JorjMcKie , could you please let me know when the next version will be published with this issue resolved?