Read and Write Compatibility Issues

[simonvincenthk]

Hi,

I'd like to use this pandoc wrapper to convert complex *.docx reports into *.md (markdown). The fact that this wrapper preserves pandoc's (the core haskel library's) modular structure is helpful. Successful conversion will require I alter the AST after reading and before writing. (The other commonly used wrapper, pypandoc, does not preserve the modular structure and, therefore, would not allow me to alter the AST...)

I've run up against a few issues using this wrapper, through:

(1) When writing to markdown, there is a version incompatibility issue.

import pandoc
text = "Hello world!"
doc = pandoc.read(text)
md = pandoc.write(doc, file = 'HelloWorld.md', format = 'markdown')

>> JSON parse error: Error in $: Incompatible API versions: encoded with [1,22,2,1] but attempted to decode with [1,23,1].

(2) When configuring the wrapper to write with version 1.22.2.1, the writer does not recognize the output of the reader as a Pandoc object.

import pandoc
text = "Hello world!"
pandoc.configure(reset=True)
doc = pandoc.read(text)
pandoc.configure(pandoc_types_version='1.22.2.1') 
md = pandoc.write(doc, file = 'HelloWorld.md', format = 'markdown')

>> TypeError: Pandoc(Meta({}), [Para([Str('Hello'), Space(), Str('world!')])]) is not a Pandoc, Block or Inline instance.

Help is appreciated!

Best regards, Simon

[simonvincenthk]

I've managed to temporarily resolve the read and write incompatibility issues arising from pandoc version and pandoc_types_version by using the last supported version of pandoc (2.19.2) and configuring pandoc automatically.

Checking pandoc's version in the CLI:

pandoc --version

>> pandoc.exe 2.19.2

Automaticaly cofiguring pandoc:

pandoc.configure(auto = True)

[boisgera]

Hi @simonvincenthk !

Thanks for your feedback. Could you tell me what version of pandoc you were using initially?

The current version of pandoc-the-python-library is tested against pandoc-the-Haskell-CLI 3.1.1 so you should be ok up to this version.

Note that if you import pandoc.types, the automatic configuration happens automatically if no manual configuration was done beforehand.

Regards,

Sébastien