Add path suffix lookup #16326

pnkfelix · 2014-08-07T17:11:54Z

Extended ast_map::Map with an iterator over all node id's that match a path suffix.

Extended pretty printer to let users choose particular items to pretty print, either by indicating an integer node-id, or by providing a path suffix.

Example 1: the suffix typeck::check::check_struct matches the item with the path rustc::middle::typeck::check::check_struct when compiling the rustc crate.
Example 2: the suffix and matches core::option::Option::and and core::result::Result::and when compiling the core crate.

Refactored pprust slightly to support the pretty printer changes.

(See individual commits for more description.)

pnkfelix · 2014-08-07T17:14:36Z

This enhancement was inspired during my work on for non-zeroing drop.

It makes working with --pretty flowgraph so so so much better, for two reasons:

you can use path suffixes instead of node-id's as input to flowgraph=... when the suffix uniquely identifies the node-id, and
you can use --pretty expanded,identified=<path-suffix> to find out all of the node-id's for a given path suffix, for cases when the desired suffix does not uniquely identify the node-id.

pnkfelix · 2014-08-07T17:14:52Z

r? anyone

pnkfelix · 2014-08-07T21:28:52Z

src/librustc/driver/driver.rs

+
+trait AstMapCarrier {
+    /// Provides a uniform interface for re-extracting a reference to an
+    ///  from a value that now owns it.


typo: missing ast_map::Map here

huonw · 2014-08-07T22:37:47Z

Will review sometime in the next day or so (if someone else doesn't get to it first).

huonw · 2014-08-08T13:01:32Z

src/libsyntax/ast_map/mod.rs

+    /// match `parts`.  (Requires `parts` is non-empty.)
+    ///
+    /// For example, if given `parts` equal to `["bar", "quux"]`, then
+    /// the iterator will produce node id's such as items with paths


I can't parse this sentence? (Is there an extra 'such as'?)

I think I meant "will produce node-ids for items with paths such as ..."

huonw · 2014-08-08T13:28:57Z

There's probably now enough pretty-printing code to warrant a driver::pretty module.

pnkfelix · 2014-08-08T15:21:06Z

There's probably now enough pretty-printing code to warrant a driver::pretty module.

Probably true, but I am not planning to throw that refactoring into this PR. :)

pnkfelix · 2014-08-08T15:21:33Z

r? @huonw

huonw · 2014-08-09T00:11:33Z

r=me with some tests, e.g. checking that running --pretty normal=foo on fn foo() {} fn bar() {} will only print one of them (and not crash the compiler etc.).

pnkfelix · 2014-08-09T07:35:52Z

@huonw heh, hoped to slip the lack of tests by you. :)

But now that I think about it, it will be easy to just change/adapt one of the existing run-make/graphviz-flowgraph tests (or all of them) that have that hacky FIND_LAST_BLOCK in the Makefile.

(Though I suppose you asked for --pretty normal, not --pretty flowgraph. Still, easy enough to add a run-make test, either way.)

This is useful e.g. for tools need a node-id, such as the flowgraph pretty printer, since it can avoids the need to first pretty-print the whole expanded,identified input in order to find out what the node-id actually is. It currently only supports path suffixes thst are made up of module names (e.g. you cannot use the type instantiation form `a::<int>::b` or `option::Option::unwrap_or` as a path suffix for this tool, though the tool will produce paths that have non-modulues in the portion of the path that is not included in the suffix). (addressed review feedback too)

…crate`. (Groundwork for pretty-printing only selected items in an input crate.)

…:Node`.

With this change: * `--pretty variant=<node-id>` will print the item associated with `<node-id>` (where `<node-id>` is an integer for some node-id in the AST, and `variant` means one of {`normal`,`expanded`,...}). * `--pretty variant=<path-suffix>` will print all of the items that match the `<path-suffix>` (where `<path-suffix>` is a suffix of a path, and `variant` again means one of {`normal`,`expanded`,...}). Example 1: the suffix `typeck::check::check_struct` matches the item with the path `rustc::middle::typeck::check::check_struct` when compiling the `rustc` crate. Example 2: the suffix `and` matches `core::option::Option::and` and `core::result::Result::and` when compiling the `core` crate. Both of the `--pretty variant=...` modes will include the full path to the item in a comment that follows the item. Note that when multiple paths match, then either: 1. all matching items are printed, in series; this is what happens in the usual pretty-print variants, or 2. the compiler signals an error; this is what happens in flowgraph printing. ---- Some drive-by improvements: Heavily refactored the pretty-printing glue in driver.rs, introducing a couple local traits to avoid cut-and-pasting very code segments that differed only in how they accessed the `Session` or the `ast_map::Map`. (Note the previous code had three similar calls to `print_crate` which have all been unified in this revision; the addition of printing individual node-ids exacerbated the situation beyond tolerance.) We may want to consider promoting some of these traits, e.g. `SessionCarrier`, for use more generally elsewhere in the compiler; right now I have to double check how to access the `Session` depending on what context I am hacking in. Refactored `PpMode` to make the data directly reflect the fundamental difference in the categories (in terms of printing source-code with various annotations, versus printing a control-flow graph). (also, addressed review feedback.)

…ctionality.

Extended `ast_map::Map` with an iterator over all node id's that match a path suffix. Extended pretty printer to let users choose particular items to pretty print, either by indicating an integer node-id, or by providing a path suffix. * Example 1: the suffix `typeck::check::check_struct` matches the item with the path `rustc::middle::typeck::check::check_struct` when compiling the `rustc` crate. * Example 2: the suffix `and` matches `core::option::Option::and` and `core::result::Result::and` when compiling the `core` crate. Refactored `pprust` slightly to support the pretty printer changes. (See individual commits for more description.)

pnkfelix reviewed Aug 7, 2014
View reviewed changes

huonw reviewed Aug 8, 2014
View reviewed changes

pnkfelix added 5 commits August 9, 2014 10:17

refactored pprust::State constructor methods out from `pprust::print_…

a9b1a3b

…crate`. (Groundwork for pretty-printing only selected items in an input crate.)

Helper method for pprust::State for printing instances of `ast_map:…

8a80e0f

…:Node`.

pretty printer: Added some run-make tests of path-suffix lookup fun…

db0e71f

…ctionality.

bors closed this Aug 9, 2014

bors merged commit db0e71f into rust-lang:master Aug 9, 2014

petrochenkov mentioned this pull request Nov 20, 2019

Remove pretty printing of specific nodes in AST #66575

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add path suffix lookup #16326

Add path suffix lookup #16326

pnkfelix commented Aug 7, 2014

pnkfelix commented Aug 7, 2014

pnkfelix commented Aug 7, 2014

pnkfelix Aug 7, 2014

huonw commented Aug 7, 2014

huonw Aug 8, 2014

pnkfelix Aug 8, 2014

huonw commented Aug 8, 2014

pnkfelix commented Aug 8, 2014

pnkfelix commented Aug 8, 2014

huonw commented Aug 9, 2014

pnkfelix commented Aug 9, 2014

Add path suffix lookup #16326

Add path suffix lookup #16326

Conversation

pnkfelix commented Aug 7, 2014

pnkfelix commented Aug 7, 2014

pnkfelix commented Aug 7, 2014

pnkfelix Aug 7, 2014

Choose a reason for hiding this comment

huonw commented Aug 7, 2014

huonw Aug 8, 2014

Choose a reason for hiding this comment

pnkfelix Aug 8, 2014

Choose a reason for hiding this comment

huonw commented Aug 8, 2014

pnkfelix commented Aug 8, 2014

pnkfelix commented Aug 8, 2014

huonw commented Aug 9, 2014

pnkfelix commented Aug 9, 2014