When I use the in-app dictionary for the word currently being studied, I can click on individual words to go to their entries, but on their entries If I click on their sub components I can't "jump" any deeper.

Sometimes I really want to drill down into the structure of the components. You could argue Pleco already offers this, but in Pleco this functionality takes three or four taps for each jump. Further, you already offer this component analysis functionality, you just seem to stop at one jump arbitrarily (superficially the entries look exactly the same, except they don't allow recursing on another definition).

I'm surprised this functionality isn't in there. Maybe you have a good reason (maybe this data has to be cached or something?). Anyway, I'd appreciate your explanation.

FWIW, it's going to be a lot fewer taps in the next version of Pleco.

Right now it's four taps in the worst case (tap on word in search results to select it, tap on character to select that, tap on 字 to bring up character info, tap on "Words") and two in the best case (Settings / Dictionary / Entry Display / Head tap action -> "Char Info", then you just tap on word in search results / tap on character to bring up character info / tap on "Words" but since it remembers your last tab selection you're probably already there).

But our current Skritter integration was added to a minor bug-fix update in a very crude way - "put this text in the search box and display the results"; we're also going to support jumping directly to a definition by specifying a character plus its traditional and Pinyin versions. (needed to disambiguate in a lot of cases) In fact they've already done the necessary work for this on Skritter's end, they're just waiting on us to support it on ours. Additionally, our new iOS update looks a lot like our Android version (http://www.pleco.com/anmanual/dict.html#switchentrydict) in that there's now a "Words" tab right at the top of the dictionary entry screen, removing another tap. So we'll potentially be going from worst case 4 taps / best case 2 to requiring just 1 tap in all cases.

Agreed. I've often been frustrated by this limitation.

Your guess is correct: the data has to be cached, and we don't have arbitrarily deep downloading of component structures: it only goes one level deep into the components of characters you have downloaded for study.

On the web version, this limitation is not present, but it can also store our maprillions of data entries for more expansive queries.

could it potentially call in to the web if you're connected to keep digging down into component structures? I too really miss this when I'm on my ipad (that, and sample sentences for context)

Explanation of the limitation makes perfect sense. Sure wish there were a way to keep drilling though, for me this is something I want to do virtually every time I encounter a new character.