Document reader in 2.0?

benzhen

进士
Hi. i've seen little discussion on this feature, which is most important to me. (Right now i use two other products in addition to Pleodict to read on Palm). Just wondering the specifics of the loading files and tapping characters. i.e. how do multi-character phrases work? Thanks for your help.

Ben
 

mikelove

皇帝
Staff member
Some of those questions aren't actually going to be resolved until we've gotten feedback from the beta version, as we've never released a document reader before and hence there hasn't really been any customer input at all into the design of it. But as it stands right now, when you tap on a character the software checks first to see if it's the first character in a multi-character compound. If it is, it brings up the longest matching compound, if not, if looks back to see if it's the second, third, or fourth character in a compound and brings up the longest of those compounds if so. If there's no compound match in either direction it brings up an entry for the character by itself. The big arrows to the left and right of the headword in reader screenshots are an easy way to change the highlight if you want to look at a different word/character, the arrows on the left side move the start of the highlight and the arrows on the right move the end. All of this can also be controlled from the 5-way navigator pad; the middle button toggles between adjusting the highlight (each of the 4 directions mapped to one of those four buttons) and moving it around like a cursor (so up moves up one line in the document, down down one line, etc).
 

ipsi

状元
Formatting for the Reader?

How do we format documents for the reader? Does it like HTML? XML? LaTeX? Lots of random n-byte codes all over the place? MS Word? ODT?

I've got a fair few HTML documents at the moment, as I've been using Blazer to read Chinese stuff on my Treo, so having the document reader understand HTML would be nice. But I can probably write something to change them to something else should the need arise.
 

gato

状元
Hi Mike, are you going to be able to make the dictionary panel closeable in the reader in time for the beta? I think that's pretty important since space is at a premium on smaller-screen PDAs.
 

mikelove

皇帝
Staff member
ipsi - it's not going to support anything as complicated as HTML or LaTeX, unfortunately, at least not in 2.0; there's a very simple XML-based system that lets you do basic text formatting and some simple chapter indexing, but since we're still working out the standard for that it probably won't even be supported in the beta, so for that at least you're pretty much going to be working with text files.

gato - not sure, anything that we haven't already finished might potentially not make it; better to release a beta without it on October 1st than a beta with it and dozens of other little tweaks like it on November 1st.
 

ipsi

状元
That should be fine. I'm not expecting anything complex. Though a way to link from one document to another would be good, as I have ebooks of Dream of the Red Chamber, Romance of the Three Kingdoms, and a couple of others, and each chapter is a seperate HTML file. I'm not sure how much my Treo would like it if they turned into one big 120-chapter HTML file... Or text file.

Could I also recommend that you include an XML tag for the charset? It's one of the better features of HTML I feel (I have some in GB2312, some in UTF-8, etc).

Text files will be ok. Just so long as it works :). And there's an easy way to specify the encoding of a file.
 

gato

状元
. I'm not sure how much my Treo would like it if they turned into one big 120-chapter HTML file... Or text file.
It's simple to just copy the text in the HTML into a text file -- 10 chapter per file, 10 files in total in the case above. It sounds like the developers have quite a lot to do already. I don't think recognizing HMTL or XML is all that important. Having automatic recognition of GB, BIG5, and UTF-8 format, on the other hand, would be a plus and would help to minimize user support issues.
 

mikelove

皇帝
Staff member
ipsi - we're not 100% sure which character sets we're supporting for XML documents yet, but since we're probably going to allow UTF-16 as well as UTF-8 the character set will probably have to be chosen in the file loading screen.

gato - we actually have our own proprietary format for large documents, mainly since we need something we can lock/encrypt licensed ones with (no announcements on that front yet but hopefully not too too far off), and for larger documents that will probably be the recommended format for end users too; maybe with some sort of MakePlecoDoc utility which takes in an XML file and packages it up to work nice and quickly regardless of how big it is.
 
Top