Conversion of OOXML to CiceroMark and vice versa

[algomaster99]

Issue #247

Add package for converting OOXML <-> CiceroMark

Changes

• initialize package
• add function for converting OOXML to CiceroMark with proper tests
• add function for converting CiceroMark to OOXML with proper tests

[jeromesimeon]

Why is this a different package from docx ?

[algomaster99]

Oh, I just thought it was supposed to be a different package. One reason I can think of is that markdown-docx uses a path to a DOCX file which I think will install dependencies which might only work on Node and not on the browser. Mammoth works on the web too though.

[jeromesimeon]

Oh, I just thought it was supposed to be a different package. One reason I can think of is that markdown-docx uses a path to a DOCX file which I think will install dependencies which might only work on Node and not on the browser. Mammoth works on the web too though.

I think it would make sense for all transforms to work on input data in memory and not depend on the file system? Maybe I'm missing something there.

Also, it would help to figure out if there are any differences between ooxml or docx before creating a whole new package.

@DianaLease @irmerk @dselman help here please?

[dselman]

Why is this a different package from docx ?

I think making this OOXML is more descriptive - as technically that is the (open) data format, vs docx which is the Microsoft proprietary file wrapper.

[algomaster99]

@dselman But we will only be able to transform OOXML from Word Documents to CiceroMark. Other OOXML (for eg. from PowerPoint, Excel, etc.) will fail as it makes no sense.

[jeromesimeon]

Why is this a different package from docx ?

I think making this OOXML is more descriptive - as technically that is the (open) data format, vs docx which is the Microsoft proprietary file wrapper.

Sounds like an argument for renaming the current package, not creating a new one?

Also, maybe I'm wrong but isn't a docx containing several xml files? (some of which may or may not be ooxml?).

[algomaster99]

Also, maybe I'm wrong but isn't a docx containing several xml files?

@jeromesimeon OOXML file is just a zipped collection of all the XML files the Word contains as written here.

[jolanglinais]

After discussing on a call, I think it would likely make the most sense to keep the code in markdown-docx instead of making a new package. In the future we can change the name if that makes sense... But based on @algomaster99's findings it seems like docx is the most descriptive at the moment.

This will get rid of mammoth and replace the current functionality with the implementation Aman creates.

[jolanglinais]

@algomaster99 I think we wanted to keep it markdown-docx instead of markdown-ooxml

[algomaster99]

@irmerk cool, I will try to integrate it there. Once I will do that, I will close this PR.

[jolanglinais]

One thought I had going through the code:

[jolanglinais]

Is there a case where getId() will not return anything, and you'd want to have a base case after the for loop?

[algomaster99]

@irmerk I think every variable will have a corresponding ID so it ought to return something. And I am only calling this function when I parse w:sdt which are node types for content controls.