Support for translating PDF files

[dietrichson]

First go at [#84]

[MarkEdmondson1234]

Looking good so far!

[MarkEdmondson1234]

Is it ready for review?

[dietrichson]

By all means. You'll notice in the code that I am using v3 of the API whose syntax is somewhat different from v2 on which gl_translate is based.

…

On Fri, Jun 2, 2023, 02:18 Mark ***@***.***> wrote: Is it ready for review? — Reply to this email directly, view it on GitHub <#85 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAQ6UTPBBOMPN5V57LOFKIDXJFZS5ANCNFSM6AAAAAAYXKP6AM> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[MarkEdmondson1234]

Looking good! May I suggest a test is added using the test PDF you have added?

[dietrichson]

@MarkEdmondson1234, I added a test. It runs (and passes) manually on my local system, however, there may be an issue regarding v2 and v3 of the API that I am not accounting for. Please check/test on your side as well to verify.

[dietrichson]

Two commits for pdf-based testing

[LukasWallrich]

Thanks for working on this! I wanted to try it - three comments:

• I needed to manually load library(googleAuthR), otherwise I get Error in gar_token() : could not find function "gar_token"
• It would be helpful if the defaults for source and target match gl_translate()
• Also, I ran into the 20-page limit because by default IS_NATIVE is false, so that the PDF is treated as a scanned PDF. That is probably a sensible default, but it might be worth it to expose that option, so that 'real' PDFs can be translated even if they are longer.

[LukasWallrich]

Also, some files require shadow text removal for the translation to be of any use - that would be a good option to expose as well (or set as a default, based on my understanding it shouldn't do any harm)

[MarkEdmondson1234]

I'm sorry I missed the pull, merged it