Mark, that might be a huge step forward. Here is a brain dump of hypotheses and thoughts from someone who doesn't really know what he is talking about. You will, for sure, think "what is this fool twaddling about?" but while my ignorance is typically a weakness it occasionally brings an out-of-the-box perspective that is fruitful. I used to think most people were too shy or polite to express ignorant ideas but it turned out they just barely have any ideas to begin with - so, for what it's worth:
- I'm assuming that filtering works by saving the step output in a variable that progressively changes for each new step. However if, to our surprise, the position of the tag operator in the filter doesn't affect the time it takes then maybe it happens to not use the latest instance of the variable but an earlier one? A question here is if the variable status is saved somewhere and the tag op somehow accesses this earlier filter step output instead of the latest instance? - I believe the tags field is treated differently from other fields. Matching a tag, means iterating over all items in this field. Just maybe the system is made to deal with this in some parallel (asynched?) process and this messes things up when one of the parallel processes needs the input from the other one? - Maybe the repeated iteration in the tag operator gets stuck in some recursive loop? Are the edge cases properly investigated - a tiddler with no tag? A tiddler with tag fooX when using tag[foo]? Might some tag fields overflow a magic max number of allowed tags? - And are tiddlers that DO fulfill the filter properly dealt with? For example "tag[foo]" will get the tiddler with "tags: foo bar baz" but will it effectively avoid iterating over the rest of the tags when "foo" is already found? (This possible behaviour would not explain Marks wikis behaviour but maybe it is relevant anyway) - Is there some magic number of parallel processes where the *browser* switches methods for performing calculations? Might specifically the tag operator initiate an atypical number of processes which triggers this? - Does the tag op - or filters in general - treat pure tags (i.e tags without existing such-named tiddlers) differently from TagTiddlers? - Does the system do any preprocessing on the filter before the steps are processes? Perhaps it cleans up the syntax, e.g trimming white spaces or some such, which affects specifically the tag op? Again, just a brain dump to trigger ideas. I'm not expecting anyone to *answer* these . <:-) -- 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 [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/30626d7b-13c4-4b92-8d4e-fd1bda2b5f4d%40googlegroups.com.
