Notices


Reply
Thread Tools
Posts: 452 | Thanked: 522 times | Joined on Nov 2007
#111
Originally Posted by Hogwash View Post
Well done Nathan! I daresay that a solution will be found to the ARMEL build crash - it's not the sort of problem that can possibly be ignored for long
I sure hope so. The problem was known in Aug. So, I just hope now that more people are using the SDK, that it will become something that has to be fixed.

Have you built the 'release' target of these projects, and if so, how much space do they occupy?
LOL, sorry, I have no idea on size yet -- I just built debug versions of everything to be able to diagnose issues.

---

Their are some "gotchas" if you want to duplicate what I've done:

CLucene 0.9.23 no longer has the CLBackwards.h header file.
So the sword code has to have that line commented out in one of the files. It doesn't affect anything it was apparently a code cleanup.

CLucene now compiles to libclucene-core and sword expects just libclucene. I had to manually add a symlink. (I need to make sure my "package" will make that symlink )

Several version of sword I had diffeerent issues with -- Sword 1.60 the python bindings will work. 1.6. SVN version bindings didn't work. Also didn't realize Sword required ICU, so until I added it it was a bit wonky. But ICU was simple to get and compile.

I was fighting a stupid rapier/python/hildon bug forever, and I finally just decided to remove the progress bar dialog and change it to a information dialog and bam, I had it working. I was also pleased that I was able to download the Bibles from inside rapier restart it; and it was working fine. So their isn't really any more python code changes "needed" to make it function. (I still need to report this bug to the pymaemo people)

I think I'm going to put this on hold for the moment until we can get the armel building fixed. I might trying compiling Ziphos and Bibletime to see how easy it is to compile in the scratchbox.

I also need to find out about "opt"ifing stuff; I understand libraries are supposed to be optified but I'm not sure what is needed to make that process work. Be nice to actually have a device to test everything on. ;-D

Nathan.
 
Posts: 452 | Thanked: 522 times | Joined on Nov 2007
#112
Originally Posted by VulcanRidr View Post
The one thing that I miss from Sword is the fact that Rapier can't do parallel translations or a translation and commentaries, where the secondary sources are slaved to the translation.

I know it would be difficult with the screen size, but it would be nice..
Actually, I happen to love that feature on my desktop bible program I use. I think if I (and/or Hogwash) works on a new port of this using these libraries this is a feature that will show up somehow . ;-)

Nathan
 
Flandry's Avatar
Posts: 1,559 | Thanked: 1,786 times | Joined on Oct 2009 @ Boston
#113
Excellent. I guess it's worth hacking up Rapier to do what we want, since it's working. Good job!

I wish i had more time to spend on this. The learning curve is proving to be strenuous for this noob.
 
Posts: 369 | Thanked: 191 times | Joined on Sep 2009 @ Virginia
#114
Originally Posted by Nathan View Post
I also need to find out about "opt"ifing stuff; I understand libraries are supposed to be optified but I'm not sure what is needed to make that process work.
I had to put my noob hat on and ask what the heck 'optify' meant

You know about the "maemo-optify" tool, right?

Be nice to actually have a device to test everything on. ;-D
Really? Whatever gives you that crazy idea?
 
Posts: 452 | Thanked: 522 times | Joined on Nov 2007
#115
Originally Posted by Flandry View Post
Excellent. I guess it's worth hacking up Rapier to do what we want, since it's working. Good job!

I wish i had more time to spend on this. The learning curve is proving to be strenuous for this noob.
Actually, I probably WON'T be hacking on Rapier. I just wanted to make sure everything "worked". The easiest test to make sure the whole "Chain" of dependencies was working was to get Rapier to run. If I submit Rapier to the repository it will be pretty much as-is in its current state.

I don't know is Hogwash is still interested; but I'm still interested in working up a QT version of a bible version using some of the ideas presented in this thread and to fix it so it works more like what I want in a Bible program. I need to still evaluate BibleTime and GnomeSword/Ziphor(sp?) and then we can decide where to go from their.

Nathan
 
Posts: 452 | Thanked: 522 times | Joined on Nov 2007
#116
Originally Posted by Hogwash View Post
I had to put my noob hat on and ask what the heck 'optify' meant

You know about the "maemo-optify" tool, right?
Yeah, but I've been reading that python and optify have some serious issues. So, we need to find out can I optify the base libraries (sword, icu, clucene) and then we don't' optify the python bindings and rapier, or do we have to not optify any of it... Not sure yet. That will need to be found out.


Be nice to actually have a device to test everything on. ;-D
Really? Whatever gives you that crazy idea?
LOL. I don't know something about not being able to run Armel code in scratchbox to even see if the code is compiling properly...

Nathan.

Nathan.
 
Posts: 369 | Thanked: 191 times | Joined on Sep 2009 @ Virginia
#117
Sure I'm interested in the QT route....barring the ARMEL build glitch, between us we've shown that the 'building blocks' are there and it's a feasible project.

Given that we know Qt4.6 is on its way, it does seem kinda pointless to invest any more time in gtk stuff....I'm going to get my head around the current Qt4 dev API for Maemo5
 
Posts: 369 | Thanked: 191 times | Joined on Sep 2009 @ Virginia
#118
Originally Posted by Nathan View Post
Yeah, but I've been reading that python and optify have some serious issues. So, we need to find out can I optify the base libraries (sword, icu, clucene) and then we don't' optify the python bindings and rapier, or do we have to not optify any of it... Not sure yet. That will need to be found out.
Of course. The learning curve continues....

LOL. I don't know something about not being able to run Armel code in scratchbox to even see if the code is compiling properly...
What, you mean you can't debug by smell?
 
Posts: 452 | Thanked: 522 times | Joined on Nov 2007
#119
Originally Posted by Hogwash View Post
Sure I'm interested in the QT route....barring the ARMEL build glitch, between us we've shown that the 'building blocks' are there and it's a feasible project.

Given that we know Qt4.6 is on its way, it does seem kinda pointless to invest any more time in gtk stuff....I'm going to get my head around the current Qt4 dev API for Maemo5
Agreed, I have to wrap my head around Qt4 also. Another thing that I'm a "newbie" at. But even newbies can preserver. We got everything compiling in scratchbox and Rapier running -- woohoo! I hadn't much of a clue what was required when I started working on it. (especially from a scratchbox/maemo perspective) I've learned A LOT in the last couple days;-D

Nathan.
 
Posts: 452 | Thanked: 522 times | Joined on Nov 2007
#120
LOL. I don't know something about not being able to run Armel code in scratchbox to even see if the code is compiling properly...
What, you mean you can't debug by smell?
Wow, a new skill to learn! Sniff, Sniff, Sniff. Nope. Rats... Guess I will have to settle for a device sometime in the future. Hopefully I can qualify for a n900 developer device.

Nathan.
 
Reply

Tags
bible, maemo 5, rapier, reference browser, religious apps, scripture reader, sword


 
Forum Jump


All times are GMT. The time now is 11:55.