14

To the best of my memory, and in reading my past queries- some of which I re-run on a regular basis- the format of the Tags column of the Posts table in SEDE used to be like <starball><is><so><heckin><awesome>.

However, the format seems to have recently changed to |starball|is|so|heckin|awesome|.

I suspect this to be in relation to changes with Shifting the data dump schedule: A proposal (which has already had some unintended format changes).

There are two problems with this:

Broken queries

As a result of this change, several of my queries have broken. For example, (and don't judge my pattern-matching naivity- it works well enough), detecting if a question is tagged with a git-related tag, using %<git% and %git>% (which prevents undesirable matches of tags that contain the substring "digital").

That fact alone doesn't bother me much. I can just rewrite my queries for the new format. BUT there are a lot of popular queries that may have relied on this delimitation format. If this change is kept, those will all have to be rewritten as well. I imagine that would be initially confusing to people who don't notice. I'm not sure if that would be disruptive to their rankings as they are rewritten, but I imagine it could be.

Broken display

https://data.stackexchange.com has a feature to give tag rendering for columns named [Tags] and [TagName] (see help docs). This is now broken, and would need to be updated for the format change if the format change is kept.


Assuming that this was not an accident, I for one would suggest the simplest option of reverting the format.

Outside the context of history and practicalities of history, I actually think the new format is not bad. If it had been like that from the start, I'd have no qualms with it. But... history.

2
  • 2
    That are some great tags you've found ...
    – rene
    Apr 8 at 7:40
  • @rene not as great as me (joking) Apr 8 at 7:41

1 Answer 1

18
+50

This is fixed; sorry for any inconvenience.

What happened

An inefficient scalar function that took a flattened list of tags and swapped it from one denormalized form (|tag1|tag2|) to another (<tag1><tag2>) was inadvertently skipped when running the April 14th & 21st SEDE and data dump processes.

What we're doing about it

  • For SEDE
    I've replaced the existing module with a much more efficient inline table-valued function, and the data has been corrected across the network (Stack Overflow finished just before 06:30 UTC) for Posts. Work still in progress for PostHistory, the remaining item being PostHistoryTypeId = 9. This should be corrected by the refresh on April 28th, 2024.
  • For the data dump
    This issue has already been resolved for the next data dump, but the Q1 data that's out there right now |looks|like|this|. That data set may get refreshed (and therefore corrected) sooner, as we'll be testing more improvements before then, but I don't have exact timelines.

While it may feel like we're creating more problems than we're solving, I promise that is not the case.

0

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .