lneves

Back in the day, there was this thing called the "Yellow Pages"! :-)

Can you give us some detail on what are those command-line switches?

I've been using postgres for a few years now, I also use MSSQL, MonetDB and MySQL.

Some of the things I miss most frequently are:

- Easy table partitioning.

- Support for "phrase search" in fts.

- MERGE/UPSERT support.

- Columnar storage (and related optimizations) for "OLAP like" workloads.

- EXPLAIN ANALYSE that traces functions.

  find . -type d

edit: for the same output as your command you can use:

  dir -l

I do not think that this is equivalent, it looks like querying complex JSON objects and not "simple" arrays.

From the "Array Element by Position" example:

  db.blogposts.find( { "comments.0.by" : "Abe" } )

PostgreSQL doesn't have JSON support but it does have XML/Xpath support.

If you stored XML documents in PostgreSQL the similar query to the MongoDB one would be something like this:

  select * from blogposts where (xpath('/comments[0]/@by', doc))::text[] = array['Abe']

Yes it's a little more verbose but not terrible so... at least in my opinion.

Yes I've looked at them. I use them every day. The example you posted its difficult to parse because it deals with multidimensional arrays and also populating one in the inner query. The simple, and much more common case of one dimension arrays is very straightforward. Can you show me the equivalent in MongoDB?

About the performance/scalability warning; I don't deal with very large arrays, a couple hundred items max, and when using a GIN index over the array field, search queries are screamingly fast.