Make original path available to render hooks

[simonw]

https://github.com/simonw/datasette-export-notebook/blob/0.1/datasette_export_notebook/__init__.py

async def render_notebook(datasette, request):
    return Response.html(
        await datasette.render_template(
            "export_notebook.html",
            {
                "csv_stream_url": datasette.absolute_url(
                    request,
                    path_with_format(
                        request=request, format="csv", extra_qs={"_stream": "on"}
                    ),
                ),
                "json_url": datasette.absolute_url(
                    request,
                    path_with_format(
                        request=request, format="json", extra_qs={"_shape": "array"}
                    ),
                ),
                "json": json,
            },
        )
    )

This results in https://latest-with-plugins.datasette.io/github/issue_comments.Notebook showing http://latest-with-plugins.datasette.io/github/issue_comments.Notebook?_format=json&_shape=array

[simonw]

An optional path argument to https://docs.datasette.io/en/stable/plugin_hooks.html#register-output-renderer-datasette which shows the path WITHOUT the .Notebook extension would be useful here.

[simonw]

This parameter will return the URL path, with querystring arguments, to the HTML version of the page - e.g. /github/issue_comments or /github/issue_comments?_sort_desc=created_at

Open questions:

• What should it be called? path could be misleading since it also includes the querystring.
• Should I provide a url or full_url version which includes https://blah.com/...?

[simonw]

Django calls this HttpRequest.get_full_path() - for the path plus the querystring.

[simonw]

I think I'll call this full_path for consistency with Django.

[simonw]

In that case maybe there are three new arguments: path, full_path and url.

I'll also add request.full_path for consistency with these:

datasette/datasette/utils/asgi.py

Lines 77 to 90 in 97fb10c

	@property
	def path(self):
	if self.scope.get("raw_path") is not None:
	return self.scope["raw_path"].decode("latin-1")
	else:
	path = self.scope["path"]
	if isinstance(path, str):
	return path
	else:
	return path.decode("utf-8")
	@property
	def query_string(self):
	return (self.scope.get("query_string") or b"").decode("latin-1")

[simonw]

I'm just going to do path and full_path (which includes the querystring). The datasette.absolute_url()` method can be used by plugins that need the full URL.

[simonw]

The challenge here is figuring out what the original path, without the .format, actually was - while taking into account that Datasette has a special case for tables that themselves end in a .something.

The path_with_format() function nearly does what we need here:

datasette/datasette/utils/__init__.py

Lines 710 to 729 in b6a7b58

	def path_with_format(
	*, request=None, path=None, format=None, extra_qs=None, replace_format=None
	):
	qs = extra_qs or {}
	path = request.path if request else path
	if replace_format and path.endswith(f".{replace_format}"):
	path = path[: -(1 + len(replace_format))]
	if "." in path:
	qs["_format"] = format
	else:
	path = f"{path}.{format}"
	if qs:
	extra = urllib.parse.urlencode(sorted(qs.items()))
	if request and request.query_string:
	path = f"{path}?{request.query_string}&{extra}"
	else:
	path = f"{path}?{extra}"
	elif request and request.query_string:
	path = f"{path}?{request.query_string}"
	return path

It can be called with replace_format="csv" to REMOVE the .csv format and replace it with something else.

Problem is, we want to use it to get rid of the format entirely.

We could update path_with_format() to accept format='' to mean "remove the format entirely", but it's a bit messy. It may be better to reconsider the design of path_with_format() and related utility functions entirely.

[simonw]

Relevant existing tests:

datasette/tests/test_utils.py

Lines 365 to 398 in 461670a

	@pytest.mark.parametrize(
	"path,format,extra_qs,expected",
	[
	("/foo?sql=select+1", "csv", {}, "/foo.csv?sql=select+1"),
	("/foo?sql=select+1", "json", {}, "/foo.json?sql=select+1"),
	("/foo/bar", "json", {}, "/foo/bar.json"),
	("/foo/bar", "csv", {}, "/foo/bar.csv"),
	("/foo/bar.csv", "json", {}, "/foo/bar.csv?_format=json"),
	("/foo/bar", "csv", {"_dl": 1}, "/foo/bar.csv?_dl=1"),
	("/foo/b.csv", "json", {"_dl": 1}, "/foo/b.csv?_dl=1&_format=json"),
	(
	"/sf-trees/Street_Tree_List?_search=cherry&_size=1000",
	"csv",
	{"_dl": 1},
	"/sf-trees/Street_Tree_List.csv?_search=cherry&_size=1000&_dl=1",
	),
	],
	)
	def test_path_with_format(path, format, extra_qs, expected):
	request = Request.fake(path)
	actual = utils.path_with_format(request=request, format=format, extra_qs=extra_qs)
	assert expected == actual
	def test_path_with_format_replace_format():
	request = Request.fake("/foo/bar.csv")
	assert (
	utils.path_with_format(request=request, format="blob")
	== "/foo/bar.csv?_format=blob"
	)
	assert (
	utils.path_with_format(request=request, format="blob", replace_format="csv")
	== "/foo/bar.blob"
	)