[Pkg-crosswire-devel] peaceful coexistence of libsword6 and libsword7

[Jonathan Morgan]

On Fri, Jan 30, 2009 at 3:40 PM, Matthew Talbert <ransom1982 at gmail.com> wrote:
> On Thu, Jan 29, 2009 at 11:21 PM, Jonathan Morgan <jonmmorgan at gmail.com> wrote:
>> On Fri, Jan 30, 2009 at 10:53 AM, Peter von Kaehne <refdoc at gmx.net> wrote:
>>> Jonathan Marsden wrote:
>>>> What will it take to get these patches into the upstream codebase?
>>>
>>> I gather that Bpbible is simply applying a couple of bug related and
>>> filter expanding patches which are already in SVN head, while avoiding
>>> the whole GenBook bible stuff (also in SVN head)
>>
>> Not quite: two of the three patches are bug-fixes, but from memory
>> they were made by Ben, and took ages to get into head (I don't think
>> they were even in head by the time we released 0.4).  The other one is
>> the compressed module speedup fix.
>>
>> We do not ever want to rely on SVN Sword, which is why we push for
>> more frequent Sword releases.
>>
>>> Gen book bibles: This is major API change currently moving through which
>>> will allow SWORD to use arbitrary versifications and canons - instead of
>>> the current fixed KJV versification.
>>>
>>> Large chunks of this are committed already, but it needs to be fleshed
>>> our further and also requires some front end changes. GS e.g compiles
>>> fine and works without problem on the new SVN head, but does not yet
>>> make use of any of the new functions.
>>>
>>>> Are the patches that BPBible needs in order to function already in the
>>>> upstream (Crosswire) development codebase, just not yet released?  Or
>>>> have they not been integrated/accepted into the main code tree at all so
>>>> far?  If the patches are applied, do they break any other applications
>>>> that use the libraries?
>>>
>>> I think I answered most above
>>>
>>> It appears there is no activity on the bindings yet and without these
>>> Bpbible can not move forward. But then the work is not completed in
>>> other parts of the library either. It should work though fine on the
>>> unpatched 1.5.11 I think, no, Jonathan?
>>
>> I do not remember the nature of the bug-fixes, but I have a feeling
>> that at least one of them was required by BPBible.  You can find them
>> on our Google Code site for download in conformance with the terms of
>> the GPL.
>>
>
> Not wanting to really contradict you here, as you certainly know more
> about BPBible than I. However, I had a little chat with Ben last night
> about the fixes. Here's a summary:

I don't see it as too much of a contradiction when I said I didn't
remember the nature of the bugfixes.  I thought Ben said one was
required, but I have been known to be wrong before.

> thml -> plain filter was outputting incorrect UTF-8 for accented
> characters (specified with HTML codes). This patch is in r2221 in
> trunk. This is equally a fix for all front-ends, but it's
> non-essential for stability.
>
> zverse.cpp and .h patch
> this patch dramatically speeds up zlib compressed modules. It has not
> been applied to trunk as far as I can tell. This would be a nice patch
> to have, but nothing depends on it.

This patch was taken from trunk when the speed issue was talked about
on sword-devel some months ago.  It was part of a larger change, from
memory.

> divine name patch
> Sword mangles divine names where the first character is composed of
> more than one byte. This occurs particularly in Chinese Bibles. Sword
> tries to uppercase only the first byte which results in some weird
> behavior (not sure exactly what). This would seem to be fairly
> important, but it is also not in trunk.
>
> None of the front-ends, including BPBible, "rely on" any of these for
> operation (ie, none cause crashes). It would be great to get all of
> these into trunk, but as far as I can see, only one is.

I think all changes done by Ben would have been submitted to
patches at crosswire.  As already stated, the speedup was taken from
trunk and so I believe 2/3 are in trunk.

Jon