Microsoft Word File Import/Export (Book)

Here the file: https://1drv.ms/w/s!AkHBEeE049R2hv8BvpsRK_o_0XOo-g?e=XJpt0t

Certainly! Below is a detailed description of the problem in English, which you can send to the creator of the plugin:
---
**Subject: Issue with XSLT Transformation Error When Importing Word Document into Moodle**
**Dear [Plugin Developer's Name],**
I hope this message finds you well. I am writing to seek your assistance regarding an issue I encountered while trying to import a Word document into Moodle. The import process fails with the following error message:
```
XSLT transformation failed ({$a})
```
**Context:**
- I am using a Python script to generate a Word document (`output.docx`) from an Excel file. Each page of the Word document contains an image and several lines of text.
- The generated Word document opens without any issues in Microsoft Word and seems to be properly formatted.
- However, when attempting to import this document into Moodle, the aforementioned error occurs.
**Details of the Process:**
1. **Excel Data:**
- The Excel file (`datenquelle.xlsx`) contains the following columns: `Bildpfad` (Image Path), `Textzeile_1` (Text Line 1), `Textzeile_2` (Text Line 2), `Textzeile_3` (Text Line 3), `Textzeile_4` (Text Line 4), and `Textzeile_5` (Text Line 5).
2. **Python Script:**
- The script reads the Excel file and creates a Word document where each page includes:
- `Textzeile_1` (formatted as bold)
- An image from the `Bildpfad`
- `Textzeile_2` to `Textzeile_5` (formatted as italic if `Textzeile_2`, normal otherwise)
Here is the Python script I am using:
```python
import pandas as pd
from docx import Document
from docx.shared import Inches
import os
# Reading the Excel file
excel_file = 'datenquelle.xlsx'
df = pd.read_excel(excel_file)
# Creating a new Word document
doc = Document()
# Function to add formatted paragraph
def add_formatted_paragraph(document, text, bold=False, italic=False):
paragraph = document.add_paragraph()
run = paragraph.add_run(text)
run.bold = bold
run.italic = italic
# Iterating through the data and adding to the Word document
for index, row in df.iterrows():
# Adding Textzeile_1
if pd.notna(row['Textzeile_1']):
add_formatted_paragraph(doc, row['Textzeile_1'], bold=True)
# Adding the image if the path is valid
bildpfad = row['Bildpfad']
if pd.notna(bildpfad) and isinstance(bildpfad, str) and os.path.exists(bildpfad):
doc.add_picture(bildpfad, width=Inches(4.0))
# Adding the remaining text lines
for i in range(2, 6):
text_key = f'Textzeile_{i}'
if pd.notna(row[text_key]):
if text_key == 'Textzeile_2':
add_formatted_paragraph(doc, row[text_key], italic=True)
else:
add_formatted_paragraph(doc, row[text_key])
# Adding a page break
doc.add_page_break()
# Saving the Word document
doc.save('output.docx')
```
**Issue:**
Despite the document appearing to be correctly formatted in Microsoft Word, Moodle throws an XSLT transformation error during the import process. We suspect that there might be some incompatibilities or non-standard formatting elements in the generated `.docx` file that Moodle's import process cannot handle.
**Request:**
Could you please provide guidance on how to troubleshoot and resolve this XSLT transformation error? Are there specific requirements or limitations for Word documents to be successfully imported into Moodle that we might be overlooking?
Any help or suggestions you could provide would be greatly appreciated.
Thank you for your time and assistance.
Best regards,
[Your Name]
---
You can send this message to the plugin developer to explain the issue you are encountering. If you need further assistance or modifications to the message, please let me know!

Hi Eliot, the plugin actually exports an XHTML file, not a native Word file. By using the ".doc" suffix, it fools Windows into opening the file using the Microsoft Word app, which then interprets the HTML content into native Word format. Saving as ".docx" or ".htm" would not achieve this effect.
I did investigate allowing students to download as Word a while ago, but it required a bit more work than I thought, so I put it aside. I might take another look.

Hi Eoin, we really appreciate this plugin. I've had some questions from our end users which would be great to get your responses to. Is there a reason this plugin exports to .doc instead of docx? Have you thought of making the plugin available to student, i.e. allow students to export Moodle books to word (as an improvement to the print to PDF they can currently do which doesn't allow them to edit the content for accessibility reasons)?

Hi Malcolm, the import works on all versions up to 4.1. I wonder have you formatted your Word file correctly to use the "Heading 2" style? You also need to have at least 1 "Heading 1" style before any "Heading 2" styles, or nothing gets imported.

Thanks Eoin. I have tried checking the box before importing the Word file but I'm still only seeing the topic title in the side menu. I'm using Moodle 4, is that version supported?

Hi Malcolm, you can create subheadings based on the Word "Heading 2" style (in addition to headings based on the "Heading 1" style), by checking the "Create subchapters based on subheadings" box when importing the Word file.

I would like to create a side index for my book based on H2 headings and from the documentation it seems that this should be possible but I'm not seeing it. Do I need to enable this is in the settings?

Hi Eoin, Thanks for the follow-up and suggestions. Have you been able to confirm whether this is actually an incompatibility with the newest Word release versions? I am unfortunately not able to test in older versions (My Word license/access is corporate, so is managed outside my control). Regarding A) The images should have been directly embedded. I replicated the problem using ATTO to simply browse for an image and upload it. I did not use any external repository and the same method worked for PNG and GIF formats, just not reliably for JPG/JPEG. For B) The file extensions used to test and replicate were standard. I tested with multiple common formats (PNG, JPG, JPEG and GIF). For C) no web-incompatible formats were used and D) at this time I am not at liberty to authorise the release of a backup containing the Book activities. The Moodle sites I administer and have tested within are not owned/maintained by myself.

Hi Alex, this looks like an incompatibility introduced in the newest version of Word. Have you been able to test with an older version? Regarding troubleshooting, things to check are a) that the image is actually embedded into the Book, and not a reference to an image stored somewhere else; b) the image file suffix is "standard", like png, jpg, jpeg, gif; and c) the image format is web-compatible (i.e. not BMP or WMF). You could try saving the book as an ePub (there's a "Download as ebook" option in the Book administration menu), and checking that the image appears when read in an ePub Reader, or by opening the ePub file in a Zip reader to check that the images are present and have the right suffixes. If you backup the Book to a Moodle course backup file and send it to me, I can take a closer look.

Hi Eoin,
I posted a comment to https://github.com/ecampbell/moodle-booktool_wordimport/issues/8 about a potential continuation of the same problem in v1.4.11 (2021083100) of this plugin. The exported file is showing broken images when exporting and opening in current (2022) desktop versions in Windows and Mac. The issue isn't occurring 100% of the time, but reliably enough for majority of embedded images in Book chapters. Any tips for troubleshooting this at all?
Thanks,
Alex

heading row elements (<table><thead><tr><th...</th><th>...</th></tr></thead><tbody>....</tbody></table>)

Hi Joshua, many thanks for your comments, most of which are true. However, regarding item 2) Tables, heading row elements (.......) should be included if you have set the "Repeat Header Rows" flag in the Table Tools Layout ribbon menu (visible when the cursor is inside a table). Regarding item 4) Bulleted/Numbered lists, list paragraphs must use the "List Bullet" or "List "Number" styles. By default, Word applies a bullet or number to a "Normal" or "Body Text" style paragraph, and the conversion tool isn't smart enough to figure this out. Note also that nested lists (using "List Bullet 2") are not supported. I will try to fix items 1) and 3) at some point. 3) is a bit tricky because I try to support some simple visual formatting like coloured text, but don't do a good enough job at stripping out font-size style directives.

Hi Eoin! Thanks for the plugin: it looks really useful! I've tested it out and have a few concerns around accessibility about the conversion to html.
1) The alt-text gets put in the longdesc attribute (which is depreciated) rather than alt. This has a few issues: users can't edit alt-text through the Atto editor (but must edit the html instead), accessibility checkers like Brickfield won't see this and some screenreaders (e.g. Chromevox) don't handle longdesc well.
2) Tables: these get imported without header row tags (even if this is set in the Word document originally).
3) tags everywhere! In Word, I use a 14pt font as my normal style (the default 11pt in Word is smaller than recommended for print accessibility) - this formatting gets applied to each paragraph in a tag, but sometimes for no discernible reason each word in a sentence gets it's own tag. This has a few potential issues:
a) with a per word it can make screenreader navigation cumbersome.
b) Sometimes my screenreader skips the image (potentially because there is so much nested formatting instructions centre, font size etc). I'm not really sure why this happens, but it is not an issue content authored directly with the Book Activity.
c) it makes the html very long and difficult to edit.
4) Bulleted/Numbered lists get removed. At least this issue is apparent to the author, but bulleted/numbered lists are important in supporting accessibility.

Hi Iván, this problem typically arises if the paragraph style definition specifies one format, but it is overridden in with a different format in a specific paragraph using that style. For example the style definition for a "Quotation" paragraph might be italic by default, but this is overridden with bold. Reset the text formatting in the paragraph by selecting all the text and pressing +to clear the formatting.