docxtpl_evm

This package is a quick and dirty way to make vertical merge in nested table possible in docxtpl.
While docxtpl_dynamic_vertical_merging exists, its approach messed up the document xml and make all chars that need to escape (>, <, &) dissappear.

How to use

In docx template, use extended vertical merge tag: {% evm <start_row_index_list> <counter> %} in the table cell.

• <start_row_index_list> is the list of row indexes merged cell starts.
• <counter> is the row counter.

In python, add items for <start_row_index_list> and <counter> into your document context, for example:

context = {
    "A": [f"a{i+1}" for i in range(2)],
    "B": [f"b{i+1}" for i in range(3)],
    "C": [f"c{i+1}" for i in range(4)],
    "index_a": [i*3*4 for i in range(2)], # row index list for A
    "index_b": [i*4 for i in range(6)],   # row index list for B
    "count_a": [0],                       # counter for A
    "count_b": [0],                       # counter for B
    "escape": "<>&"
}

Notice that the counter should be a list contain the counter, this approach is due to jinja2 scoping rules, check this for more information.

Replace docxtpl DocxTemplate with CustomDocxTemplate and use it as usual.

from docxtplevm.CustomDocxTemplate import CustomDocxTemplate

tpl = CustomDocxTemplate("tests/test_file/test.docx")

# Add your context here...

tpl.render(context)
tpl.save("tests/test_file/test_output.docx")

Example

• context

context = {
    "A": [f"a{i+1}" for i in range(2)],
    "B": [f"b{i+1}" for i in range(3)],
    "C": [f"c{i+1}" for i in range(4)],
    "index_a": [i*3*4 for i in range(2)],
    "index_b": [i*4 for i in range(6)],
    "count_a": [0],
    "count_b": [0],
    "escape": "<>&"
}

• template

• output