4 ms·
Indeed. > why isn't it a standard feature of SQL to just compare two tables? Not enough people have complained about needing it (it doesn't hit the desired RO
by RajT88 3y ago
Indeed.
> why isn't it a standard feature of SQL to just compare two tables?
Not enough people have complained about needing it (it doesn't hit the desired ROI for a PM to prioritize the feature request).
SQLite's SQL Diff I first came across years ago - super useful. It's the perfect example of the industries which pop up to fill in the gaps that huge software vendors like Microsoft leave open. I used to work at a company valued at hundreds of millions of dollars, which made nothing but such products filling gaps.
- murki 3y agoisn't that what MS SQL Server's tablediff does? https://learn.microsoft.com/en-us/sql/tools/tablediff-utility?view=sql-server-ver16 https://learn.microsoft.com/en-us/sql/tools/tablediff-utilit...
- hatware 3y ago[dead]
- TheRealPomax 3y agoSQL isn't any specific product for a PM to greenlight work on or not, it's a programming language (the L in SQL =) that an structured (the S in SQL =) data application can support as either "the" or "one of the" programming interfaces to query (the Q in SQL =) that data. So it's really a question of why SQL, the language, doesn't come with a table comparison command so that you can write statements akin to "SELECT * FROM (COMPARE table1 table2) AS diff WHERE ..."
- tuwtuwtuwtuw 3y agoWouldn't the process be that some PM wants a feature to be added and sends it of to the representative in the appropriate committe who then push for it to be added? I mean it's not the SQL language itself which sits in the committe specifying new features, it would be representatives from Oracle, Microsoft, IBM and similar.. Right?
- tremon 3y agoIf some PM wants to add a debugger to C, do they write a language extension proposal or do they focus on putting their requirements in a separate tool? This is roughly the same problem. The feature exists, but it's rare enough that the only people that need it are programmers/DBAs that need to deep-dive into a specific issue. Regular people/applications will never need this feature, so why should it be baked in the core language? The specialists already have specialist tooling, so it makes much more sense to implement this as a toolkit feature than a language feature.