Extract all the variables from the word document

[cesarve77]

First thanks and congratulation for this great package

What is the best way to extract all the variables (in dot notation for example) from the word document.

My idea is build the query based in the variables asked in the docx.

Thanks in advance

[mathe42]

Interesting question... Maybe use https://www.npmjs.com/package/textract anderen then search with regex

[wildhart]

I've come up with an inefficient solution to this, but imo there should be an easier way.

Since the data object can be a promise-returning function, I would expect this to work:

report = await createReport({
	template,
	cmdDelimiter: ['[', ']'],
	data: function(query) {
		console.log('field:', query);
		return new Promise((resolve, reject) => resolve('hello'));
	},
});

But in the data function query is always undefined. My templates have fields like [ProjectNumber]

Why doesn't this work?

Here's my solution which does work, by catching the error in a loop:

const data = {};
let report = null;
do {
	try {
		report = await createReport({
			template,
			cmdDelimiter: ['[', ']'],
			data,
			// USE ONLY IN THE BROWSER, AND WITH TRUSTED TEMPLATES
			noSandbox: true, // WARNING: INSECURE
		});
		import('file-saver').then(FileSaver => FileSaver.saveAs(new Blob([report]), res.name+'.done.docx'));
	} catch(e) {
		const m = e.toString().match(/INS (\w+)/);
		if (m) {
			data[m[1]] = m[1];
			console.log('new field:', m[1]);
		}
		else throw e;
	}
} while (!report);
console.log(Object.keys(data)); // this is a list of the fields!

PS - @cesarve77 please can you edit the title of this issue to something like [question] get list of fields so people have more chance of finding it.

[mathe42]

Maybe you also can use a Proxy object in a first run. And getting the requestet fields...

[wildhart]

I tried a Proxy object but that didn't work. I think internally docx-templates is using something like Object.keys(data) to get a list of fields in the data object because it doesn't even try proxy.get() on fields which don't already exist.

[wildhart]

FYI, the alternative module docxtemplater handles this really easily with its nullgetter setting: https://docxtemplater.readthedocs.io/en/latest/configuration.html#nullgetter in which part.value is the field name.

[bulolo]

i also need to extract all the variables from template which has follow the +++=XXX+++ by template-docx'way
is there any good way to parse it self.

[jjhbw]

I was planning on implementing something like this. I'll let you know when i've gotten around to it.

[bulolo]

any progress？

[bernardohmr]

I hope someone shows up with this solution soon also (=

[jjhbw]

Added this feature in https://github.com/guigrpa/docx-templates/releases/tag/v4.3.0

[Arend-dev]

Hi there,

When trying to use listCommands, I always get an error:

"...\node_modules\jszip\lib\utils.js:454
new Error("Can't read the data of '" + name + "'. Is it " +
^

Error: Can't read the data of 'the loaded zip file'. Is it in a supported JavaScript type (String, Blob, ArrayBuffer, etc) ?"

Do not know what I am doing wrong. I start node with "--unhandled-rejections=strict ".

[jjhbw]

@Arend-dev Please make a new issue for stuff like this.

I can't do much without seeing your code. First off, can you verify that you are using it correctly by checking the suggestion in the error message (see below)?
...Is it in a supported JavaScript type (String, Blob, ArrayBuffer, etc) ?

The unhandled rejections flag should have little to do with it.