Page 1 of 1

Feature Request: strip punctuation prior to lookup

PostPosted: Mon Oct 22, 2012 7:24 pm
by jeromecarney
Major Annoyance: PDF Readers that integrate with GoldenDict invariably send adjacent punctuation when passing text to an intent

I'm a big fan of GoldenDict ... in fact, since I work in a bilingual context and I'm an avid language learner (I'm up to four, and looking to add a fifth), the primary reason I bough an Android tablet rather than an iPad was the availability of apps that allow for single tap dictionary lookup, combined with brilliant dictionary apps like GoldenDict.

There is one ongoing annoyance, though: the two most popular PDF readers offering GoldenDict integration — ezPDF and RepliGo — both pick up adjacent punctuation (periods, commas, quotes, colons, etc.) when sending text for lookup. Because I look up scores of words each and every day, the constant tapping / editing / correcting became a big enough pain that I purchased Tasker and figured out basic Python script so that I could write a work around.

It's not, however, the most elegant solution, and it slows things down a bit, so I'm begging: please add a feature to allow automatic punctuation stripping from dictionary entry lookups.

Thanks — and thanks for a great Android app !

Re: Feature Request: strip punctuation prior to lookup

PostPosted: Tue Oct 30, 2012 12:30 pm
by geniuses
Motion seconded!
Indeed it would render the program much more user-friendly.

Re: Feature Request: strip punctuation prior to lookup

PostPosted: Sat Nov 17, 2012 6:23 pm
by cousin_itt

However, exPDF have finally implemented this after ignoring my suggestion many months ago (obviously I nicked the idea from fbReader).
The latest Repligo doesn't seem to pick up the punctuation either.

Re: Feature Request: strip punctuation prior to lookup

PostPosted: Sat Nov 17, 2012 6:54 pm
by jeromecarney
I noticed that myself in this week's ezPDF upgrade — this is phenomenal, and it definitely makes things a lot easier! There's still no filtering of double punctuation, such as the last word in a sentence ending with a period and a quote, but that's a minor quibble.

My biggest wish would be for PDF readers to give users the option to set a default action for long taps, the same way epub readers do. After all, that's how most people work with PDF documents: some of us read in foreign languages and only select words to find the definition. Other people, though, may use the select text function exclusively to highlight words or to add annotations. I know that when I'm reading in epub readers having one less tap makes a world of difference! If the GoldenDict team could pass a word along to the PDF apps, it would be greatly appreciated.

Second on my list of usability enhancements would be to have look ups handle hyphenated words ... for example, if I highlight the word in red below:

This is a hyphen-

GoldenDict reads it as "hyphen- ated". It would be really nice if ezPDF or Repligo or GoldenDict would filter out the hyphen and space and jump right to the word — or even better, if the PDF readers would select hyphenated words with a long tap, in the same way they select other words.

Re: Feature Request: strip punctuation prior to lookup

PostPosted: Sat Nov 17, 2012 7:13 pm
by cousin_itt
Agreed, it's a bit clunky in ezPDF to have to long tap and then tap on colorDict. Lord knows how many users of it actually add annotations, I suspect only a small percent. (That said, ezPDF have been doing a cheaper, non-annotation version for a while which I've not tried).
Another feature ezPDF added (which I suggested and was ignored) a little while ago was selected text read by text-to-speech. This I needed for some French books and it works well.

I guess they need they ability to add tools to the long tap and if only one tool exists then run that.

And yes, whole-heartedly agree with readers handling hyphenated words better.
If this hypen on the long tapped word is the last character on a line, grab the contents of the next line up to the first space or punctuation mark and add it. Then send it off to the dictionary.

Good luck getting the ezPDF devs to do this. They seem side-tracked making their pdf reader handle ePub. Most odd considering there is the free and wonderful fbReader that is almost perfect and has an active dev team.