I solved this by implementing a new helper method that simply checks for the
presence of a specific book within a given module.
This faster than doing the look-up based on a complete book list.
bool ModuleHelper::moduleHasBook(sword::SWModule* module, std::string bookCode)
{
bool hasBook;
stringstream key;
key << bookCode;
key << " 1:1";
module->setKey(key.str().c_str());
hasBook = module->hasEntry(module->getKey());
return hasBook;
}
Having 18 modules installed locally … calling this method for all 18 modules
takes between 20-30 ms on the JavaScript side … and that’s sufficient :-).
Best regards,
Tobias
> Am 13.05.2021 um 11:33 schrieb Tobias Klein <cont...@tklein.info>:
>
> Hi all,
>
> I am wondering whether it is possible to optimize the performance for getting
> a module’s book list.
>
> The function I am currently using looks like this:
>
> vector<string> bookList;
> string currentBookName = "";
> VerseKey *vk = (VerseKey *)module->getKey();
>
> for ((*vk) = TOP; !vk->popError(); vk->setBook(vk->getBook()+1)) {
> if (module->hasEntry(vk)) {
> currentBookName = vk->getOSISBookName();
> bookList.push_back(currentBookName);
> }
> }
>
> In my node-sword-interface JavaScript world the wrapped function returns in a
> time between 10-15ms depending on the module.
>
> When calling this once for the currently used Bible module … that’s fine.
> But I now have a use case where I need to retrieve the book lists of all
> installed modules at once. In Ezra Bible App, the user has a possibility to
> switch the Bible module for the currently opened text.
> In the respective selection box I would like to disable all Bible modules
> that do not contain the currently opened book. This could be the case for NT
> only, OT only or other modules of that sort.
> In order to implement this dynamic disabling/enabling logic I need the book
> lists of all available SWORD modules at once. And suddenly we are looking at
> 200+ ms for the above mentioned function executed for let’s say 15 modules.
>
> I am testing on a Mac mini to get performance figures for a lower end machine
> …
>
> Now, having said all that … is there a way of optimizing the performance of
> the function above? Or have I overlooked something and need to approach this
> use case differently?
>
> Best regards,
> Tobias
> _______________________________________________
> sword-devel mailing list: sword-devel@crosswire.org
> http://crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
_______________________________________________
sword-devel mailing list: sword-devel@crosswire.org
http://crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
