May I please ask a stupid but polite question? What is the fascination with trying to find what appear to be obscure and unrealistic ways to "break" SQLite? I understand that identifying areas of potential errors is an important part of computer science, especially in software used for things in which an error could result in the loss of life. But does testing SQLite really transfer to testing those types of software?

I've been watching the forum for only the past four years or so. I used to feel that I could learn new things from reading the posts; and error reports were more along the lines of "I tried to do something and received an unexpected result; was my expectation wrong, am I doing something stupid, or is SQLite wrong?" Now they are more like "I put sugar in SQLite's gas tank and now it won't run correctly. Ha, ha; aint I brilliant."

I may be being very unfair due to my ignorance in the area of software testing; but it feels like the focus among participants in the forum has moved in a direction that is less productive toward developing SQLite further for those that want to actually employ it and learn it.

I realize that one of the major contributors to the forum--and one that often took an interest in questions involving SQL queries and getting real work done in SQLite--passed away, leaving a bit of a void. And I realize that the developers want real bug reports. My question is, what is the desired focus and content of the forum; and am I the only one that feels that the forum focus has changed?

Thank you.

I think that its to try and make SQLite as robust as possible. Some are obscure edge-cases that will never be replicated in the wild .. like select this high number, join it to a table that has exactly two billion rows in it, and if one of those rows has got an index value of '26', and you then say 'Banana' three times, SQLite throws an exception.

However, SQLite is used more and more in embedded systems. One day, that weird counter could be used in a point-of-sale system used to count how many fish-fingers have been sold. People won't care that SQLite has errored, but the resulting application will crash - which people will care about. So if SQLite doesn't crash, the resulting application won't crash.

Maybe. Thats just my assumption. Make SQLite bullet-proof, and the application its used in will also be more reliable.

In answer to your second point, I think the main core of the forum are now people that are bending SQLite into all weird angles. People who just don't know how 'select' works are scared off and don't want to ask 'silly' questions like 'how do I get this working?', which is a shame. SQLite knowledge is like a conveyor belt - there are always new people starting at the beginning wondering how 'select' works...

may I ask a follow-up stupid but polite question? (thanks "anonymous" for getting the ball rolling)

In answer to your second point, I think the main core of the forum are now people that are bending SQLite into all weird angles

I am curious as to who is doing all this weird angles bending? Most all of these "select this high number, join it to a table that has exactly two billion rows in it, and if one of those rows has got an index value of '26', and you then say 'Banana' three times" reports seem to come from users with IDs "8452349323". And, many of them report an error, and then immediately report where that error was fixed. Are these real people either contracted by SQLite to make the software more bulletproof, or a server farm of bots that are running fuzzers and reporting each breakage? There really does seem to be a spike of these kinds of reports in the recent past. I must say, they are kinda fascinating to watch even though I don't learn anything from them (kinda like watching a car wreck while driving by and causing a gaper's block).

Still here with unsteady legs on the conveyor belt

I'd like to apologize for my current actions. In fact, I'm using a large language model to systematically read through and test all the code in SQLite3, but I'm still working through the shell.c file at this point. Perhaps the forum would benefit more from discussions about SQL statement usage? Any vulnerabilities I discover in sqlite3.c moving forward will be sent directly to the author's email to avoid disturbing everyone. My apologies once again.

Hi there! Quick heads up – I used to get a personal thank you + fix note every time I submitted a report, but figured it wasn’t necessary to ping the author each time. So I started noting the fix myself to prevent confusion for future readers. The SQLite contributors are truly outstanding developers who consistently fixes reported issues within 24 hours. My bad for overstepping – won’t happen again! (And just to clarify: I’m human, not a robot! )

The complaint here seems to be that users violin (yuelinwang) and Lanbigking (257342004) are filling up this Forum

For the record, my own post was not a complaint. It was a question that sprang from genuine curiosity. I am certainly grateful to the bug reporters in the belief that their reports make the software better. And, please, no user "owns" the forum, so it is not my place to dictate what should or should not be posted here. I have the choice to glaze over topics that don't interest me. In that spirit, keep it coming per the forum guidelines, of course.

My bad for overstepping

SQLite forum hug

No problem. Don't take it personal. I don't believe you have 'overstepped' at all. :)

my 112 dominican cents ($1 = 56 pesos): I enjoyed (and learned a few things) reading those. And by the way, someone that is anonymous should not have any type of request in this forum (my own personal believe, of course). If you can't identify yourself, you should not complaint. Thanks.

josé

Fortunately, the powers that be disagree and will cheerfully fix anonymously reported bugs.

Yes, that is the beauty of this forum. :-)

I'm the anonymous that submitted this topic. I have an account and one reason (not the only) for posting anonymously was to have the post pass moderation, such that if it was deemed inappropriate or erroneous then it would never appear.

I'm not trying to start an argument or write one of those long posts that sometimes appear about ones "feelings." I just wanted to be a bit more clear about what I was trying to ask.

You'll note that I only asked a couple questions, never named any posters or mentioned excessive anything, and made no requests that anything be done. However, in review, my phrasing could be interpreted as unkind.

My observation, whether accurate or not, has been that the forum focus has changed quite a bit after losing Keith and the participation of Larry. For me, at times at least, this forum was sort of like working in the wood shop with Norm Abram. I can figure a lot of things out on my own and may not need the forum, but sometimes I see superior methods here and think, "I'd have never thought of that in a 100 years." Programming is a secondary area for me and one which I do sort of in isolation; so, it's nice to have a glimpse now and then into the ways professionals work and think.

The kinds of posts that the new forum rules state should no longer be posted have not been directly blocking the kinds of discussions to which I'm referring; but they just don't seem to be appearing as frequently and I wondered why. Is it because they do not receive answers and stopped asking, because the forum no longer looks like the appropriate place to ask, or because this isn't the place to ask?

I'm glad you enjoyed "reading those" but am a bit surprised by your response because you were the first one to inquire concerning Keith's absence from the forum and, I think, part of the group that wanted to ensure that his extensions were captured from the internet before his site was taken down. I don't mean to keep referring to Keith (I didn't know him) but his posts are a good example of the kinds of discussions to which I've been referring.

There are people who really know a lot and enjoy sharing it. This forum and some of the people that have answered questions on Tcl for me appear to be in a class of knowledge and skill level that I haven't seen on other forums. Without the questions, they can't share it; and I was asking if others felt that there have been fewer questions and why that might be. I wasn't attacking bug reports; but I admit that I find it strange that many people appear to spend more time on trying to break SQLite than employing it.

Thanks and I'll shut up now.

the participation of Larry

oh yes, I miss his erudite prose… whither art thou?

The kinds of posts that the new forum rules state should no longer be posted have not been directly blocking the kinds of discussions to which I'm referring; but they just don't seem to be appearing as frequently and I wondered why. Is it because they do not receive answers and stopped asking, because the forum no longer looks like the appropriate place to ask, or because this isn't the place to ask?

From my side, I started with SQLite back in 2006. This used to be a mail service list. I still have emails saved. I even remember when Dr. Hipp asked Keith to please consider joining the forum as he didn't want to. Anyway, there were so many questions from me back then, but as I grew in knowledge, these came less and less. I am still very much a novice, but most of my questions now I am now able to answer myself, SQLite-wise. I've learned so much from everyone here, specially Igor, Keith, Larry, and others. However, when these c programming questions came, I would go and try to understand them (because I want to learn more and more c). So, it was nice to see things get fixed whether or not they would ever come up in real-world scenarios.

Regarding the anonymous, I've always had problems with people that would post questions anonymously. Simply because I never get to know who they are. In the back of my mind I think they are hiding something, but this is just me. Plus, it's nice to have a community that we can say, "like Larry, Keith, Igor and...". That's all. For example, I won't be able to say, "Hey guys, do you remember anonymous? I miss him/her!" :-) Anyway, I still stand that it's nice to know the name of the person.

Thanks and I'll shut up now.

No need to. We just won't know who you are. :-)

I’ve never minded the reports of obscure bugs. I’ve been around for a few years, and I have decades of experience with SQL, but I consider myself a newbie on this forum. I find the easiest way to follow it is to click on the link in the daily emails:

This is an automated email sent by the Fossil repository at https://sqlite.org/forum to report changes.

I can then select the topics that interest me and ignore old stuff.