issue_comments

Would it be useful to allow this mechanism to alias columns, for example supporting one of the following:

• ?_col=airline as name_of_airline
• ?_col=airline:name_of_airline

This could be handy for renaming columns to match a specific expected JSON output.

Interesting side-effect of this implementation is that you can both control column order and request the same column multiple times:

/fivethirtyeight/airline-safety%2Fairline-safety?_col=fatal_accidents_00_14&_col=fatalities_00_14&_col=airline&_col=airline

A better interface for this would be a full list of columns each with a checkbox for making it visible on invisible - this could then be used to apply a bulk change (rather than refreshing the interface after every removed column) and it could also be easily designed to work on narrow mobile screens where the cog icon is not visible.

Here's a bug: removing the rowid column returns an error.

I've changed my mind about forbidding ?_col= and ?_nocol= from being applied at the same time - I'm going to come up with a simple resolution rule instead.

Natalie suggests a quick way to implement "undo" would be to add a "Show all columns" item to that menu which only appears when at least one column is hidden.

Here's that prototype of a "Hide this column" cog menu button:

Need a way to undo that once you've hidden a column - maybe a list of currently hidden columns that lets you un-hide them.

I started looking at this again, inspired by #1326. I have a new diff that works against the latest main branch.

```diff diff --git a/datasette/views/table.py b/datasette/views/table.py index 4879228..f4b2ee2 100644 --- a/datasette/views/table.py +++ b/datasette/views/table.py @@ -64,6 +64,36 @@ class Row:

class RowTableShared(DataView): + async def columns_to_select(self, db, table, request): + table_columns = await db.table_columns(table) + if "_col" in request.args and "_nocol" in request.args: + raise DatasetteError("Cannot use _col and _nocol at the same time") + if "_col" in request.args: + new_columns = [] + for column in request.args.getlist("_col"): + if column not in table_columns: + raise DatasetteError("_col={} is an invalid column".format(column)) + new_columns.append(column) + return new_columns + elif "_nocol" in request.args: + # Return all columns EXCEPT these + bad_columns = [ + column + for column in request.args.getlist("_nocol") + if column not in table_columns + ] + if bad_columns: + raise DatasetteError( + "_nocol={} - invalid columns".format(", ".join(bad_columns)) + ) + return [ + column + for column in table_columns + if column not in request.args.getlist("_nocol") + ] + else: + return table_columns + async def sortable_columns_for_table(self, database, table, use_rowid): db = self.ds.databases[database] table_metadata = self.ds.table_metadata(database, table) @@ -321,18 +351,16 @@ class TableView(RowTableShared): )

     pks = await db.primary_keys(table)

• table_column_details = await db.table_column_details(table)

• table_columns = [column.name for column in table_column_details]

• select_columns = ", ".join(escape_sqlite(t) for t in table_columns)
• table_columns = await self.columns_to_select(db, table, request)
• select_clause = ", ".join(escape_sqlite(t) for t in table_columns)

   use_rowid = not pks and not is_view
   if use_rowid:
  

  • select = f"rowid, {select_columns}"
  • select = f"rowid, {select_clause}" order_by = "rowid" order_by_pks = "rowid" else:
  • select = select_columns
  • select = select_clause order_by_pks = ", ".join([escape_sqlite(pk) for pk in pks]) order_by = order_by_pks

@@ -715,6 +743,8 @@ class TableView(RowTableShared): column = fk["column"] if column not in columns_to_expand: continue + if column not in columns: + continue expanded_columns.append(column) # Gather the values column_index = columns.index(column) ```

Currently you can do this with a custom SQL query, but as you've noticed this means you don't get automatic pagination: https://latest.datasette.io/fixtures.json?sql=select+pk%2C+created%2C+state+from+facetable+order+by+pk+limit+10&_shape=array

I have an open issue to allow selecting specific columns on the table page: #615

A workaround you can use today is to define a SQL view that selects just the tables you want. View pages behave exactly like tables in that they provide pagination by default.

Here's an alpha release with updated dependencies: https://pypi.org/project/datasette/0.57a0/

I'm stuck also because datasette wants itsdangerous~=1.1 instead of allowing itsdangerous-2.0.0

Bumped that dependency in b64d87204612a84663616e075f542499a5d82a03

I'm going to push an alpha release with the new dependencies shortly.

The tricky thing about this is that ideally I'd like this to link to a paginated set of results - but arbitrary SQL queries do not yet support pagination, see #856.

Documentation:

• https://docs.datasette.io/en/latest/facets.html#facets-in-query-strings at the bottom of the section
• https://docs.datasette.io/en/latest/json_api.html#special-table-arguments at the bottom of the list of arguments

Here's the code in question: https://github.com/simonw/datasette/blob/5e9672c9bb33e41686472db4aa427168f9e67dbe/datasette/facets.py#L187-L210

It has access to the current request object as self.request.

?_facet_size=50 seems like a very reasonable thing to support - I'll cap it at whatever the max_returned_rows setting is (which defaults to 1000).

A URL parameter to modify that facet size is a really good idea.

I thought I had an issue open for "..." linking to more results but I can't find it now.