Hello Brian, El miércoles, 24 de agosto de 2016, 4:31:51 (UTC+2), Brian Theado escribió: > > That was news to me. Thanks for explaining it so clearly. >> > You're welcome!
> > I now have some evidence there is more to the story than what you > explained. I was stepping through the code in the debugger to help me learn > about the syncadapter functionality in general and the pouchdb adapter in > particular. I had a breakpoint set in PouchAdaptor.prototype.loadTiddler > and when I hit the save button, I saw the load tiddler was triggered for > every tiddler I had stored in pouchdb. Even the ones which were not open > and I hadn't opened. But before the asynchronous loading could take place, > I saw the html file appear in the downloads status bar at the bottom of my > chrome window. I guessed that the tiddlers got loaded, but only after the > download file was generated. And I figured if they all loaded, then if I > hit the save button a second time, then the download would this time > contain the bodies of the tiddlers. I did a diff of the first download > against the second and found that I was right! > Very good catch! You have spotted a very interesting behavior. That happens because the lazy-loading mechanism is asynchronous, and the save mechanism is synchronous. When you click save the saver selects a template ($:/core/saveAll in this case) and tries to render it. Rendering the template triggers the lazy-loading mechanism for all the tiddlers, but because the synchronous nature of the rendering mechanism it does not waits the lazy-loading mechanism to complete and triggers the download with the available data at that moment. The lazy-loading mechanism continues working in the background and eventually completes, that's why when you click the saving button a second time you get a complete wiki with all the tiddlers. If I'm not missing anything, then a full download can be had by clicking > save, then waiting long enough for the tiddler bodies to be asynchronously > loaded, then clicking save again. Not the most user friendly, but it seems > to work. Yes, that is definitely a workaround, but there is no way (currently) to know how long you have to wait. It can vary from the order of milliseconds to a couple of seconds, depending on the database size. To be honest, probably waiting a second will be enough for 99% of the times. Here's how you can duplicate my test: > > 1. open a new noteself database instance > 2. create a few tiddlers with some body text > 3. close all tiddlers > 4. refresh the page (no open tiddlers = no tiddler body loading) > 5. click the save button > 6. look in the console for the "PouchAdaptor: Retrieving tiddler" > message > 7. click the save button a 2nd time > 8. compare the two html files, noticing that the 2nd one has the > tiddler bodies and the first does not. > > Brian > Thank you very much for your report Brian. I will think in ways around this. I have opened a bug on Tiddlywiki repository: https://github.com/Jermolene/TiddlyWiki5/issues/2552 -- You received this message because you are subscribed to the Google Groups "TiddlyWiki" group. To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+unsubscr...@googlegroups.com. To post to this group, send email to tiddlywiki@googlegroups.com. Visit this group at https://groups.google.com/group/tiddlywiki. To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/8b2f9165-43df-47b2-ad42-7b028988184d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
