Issue with ignore_tags parameter

[roland71]

Hello! Running the deepl package with anvil.works (https://pypi.org/project/deepl/1.5.0/)
Example code: translation_output = translator.translate_text(translation_input target_lang="EN-US", ignore_tags=x)
Getting: NameError: name 'x' is not defined

[SamueLacombe]

Hello,

I believe your issue is that are providing an Int instead of a string:

Here is my request and it's working. I believe you just have to replace 'ignore' by 'x'
translation = client.translate_text(df['SourceTokenized'].tolist(), target_lang=targetLangDeeplCode, tag_handling='html', ignore_tags='ignore', split_sentences="0", preserve_formatting="1", formality=formality)

[daniel-jones-deepl]

Hi @roland71, yes @SamueLacombe is right (thank you Samuel!)

The ignore_tags argument expects either a string (..., ignore_tags='x', ...) or list of strings (..., ignore_tags=['x', 'another_tag_to_ignore'], ...).

[roland71]

@daniel-jones-deepl
but this only works if an opening tag is followed by a closing tag - right? if, for example, the string that is not to be translated is marked with an opening tag at the beginning and at the end, then this string will still be translated.

[SamueLacombe]

I believe it depend on your argument: tag_handling

if tag_handling='html' then you need opening and closing.
if tag_handling='xml' then it's just a tag.

[daniel-jones-deepl]

Hi @roland71, apologies that this reply was delayed.

The ignore_tags parameter is only used if the tag_handling parameter is set to 'xml'. If instead tag_handling is set to 'html', then the ignore_tags parameter is not used and instead DeepL uses predefined tags that suit HTML. However, we do provide a method to ignore HTML tags if necessary, explained here in our documentation.

In @SamueLacombe's example code above, as tag_handling is set to 'html', the ignore_tags parameter will be ignored. Sorry for the confusion.

Here is a corrected example using XML:

text = "This will be translated.<x>This will not be translated.</x>"
result = translator.translate_text(text, target_lang="de", tag_handling="xml", ignore_tags="x")
print(result.text)  # Dies wird übersetzt werden.<x>This will not be translated.</x>

Regarding opening & closing tags, for best results you should ensure you match opening tags with closing tags. We do use some heuristics to try to fix invalid XML/HTML, but the results may be unpredictable.