get rid of some false negatives in rustdoc::broken_intra_doc_links

rustdoc will not try to do intra-doc linking if the "path"
of a link looks too much like a "real url".

however, only inline links ([text](url)) can actually contain
a url, other types of links (reference links, shortcut links)
contain a *reference* which is later resolved to an actual url.

the "path" in this case cannot be a url, and therefore it should
not be skipped due to looking like a url.

Co-authored-by: Michael Howell <[email protected]>

Author: lolbinarycat

src/librustdoc/passes/collect_intra_doc_links.rs

@@ -941,13 +941,20 @@ fn preprocess_link(
     ori_link: &MarkdownLink,
     dox: &str,
 ) -> Option<Result<PreprocessingInfo, PreprocessingError>> {
+    // certain link kinds cannot have their path be urls,
+    // so they should not be ignored, no matter how much they look like urls.
+    // e.g. [https://example.com/] is not a link to example.com.
+    let can_be_url = ori_link.kind != LinkType::ShortcutUnknown
+        && ori_link.kind != LinkType::CollapsedUnknown
+        && ori_link.kind != LinkType::ReferenceUnknown;
+
     // [] is mostly likely not supposed to be a link
     if ori_link.link.is_empty() {
         return None;
     }
 
     // Bail early for real links.
-    if ori_link.link.contains('/') {
+    if can_be_url && ori_link.link.contains('/') {
         return None;
     }
 
@@ -972,7 +979,7 @@ fn preprocess_link(
         Ok(None) => (None, link, link),
         Err((err_msg, relative_range)) => {
             // Only report error if we would not have ignored this link. See issue #83859.
-            if !should_ignore_link_with_disambiguators(link) {
+            if !(can_be_url && should_ignore_link_with_disambiguators(link)) {
                 let disambiguator_range = match range_between_backticks(&ori_link.range, dox) {
                     MarkdownLinkRange::Destination(no_backticks_range) => {
                         MarkdownLinkRange::Destination(
@@ -989,7 +996,7 @@ fn preprocess_link(
         }
     };
 
-    if should_ignore_link(path_str) {
+    if can_be_url && should_ignore_link(path_str) {
         return None;
     }
 

tests/rustdoc-ui/bad-intra-doc.rs

@@ -0,0 +1,13 @@
+#![no_std]
+#![deny(rustdoc::broken_intra_doc_links)]
+
+// regression test for https://github.com/rust-lang/rust/issues/54191
+
+/// this is not a link to [`example.com`] //~ERROR unresolved link
+///
+/// this link [`has spaces in it`].
+///
+/// attempted link to method: [`Foo.bar()`] //~ERROR unresolved link
+///
+/// classic broken intra-doc link: [`Bar`] //~ERROR unresolved link
+pub struct Foo;

tests/rustdoc-ui/bad-intra-doc.stderr

@@ -0,0 +1,31 @@
+error: unresolved link to `example.com`
+  --> $DIR/bad-intra-doc.rs:6:29
+   |
+LL | /// this is not a link to [`example.com`]
+   |                             ^^^^^^^^^^^ no item named `example.com` in scope
+   |
+   = help: to escape `[` and `]` characters, add '\' before them like `\[` or `\]`
+note: the lint level is defined here
+  --> $DIR/bad-intra-doc.rs:2:9
+   |
+LL | #![deny(rustdoc::broken_intra_doc_links)]
+   |         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+error: unresolved link to `Foo.bar`
+  --> $DIR/bad-intra-doc.rs:10:33
+   |
+LL | /// attempted link to method: [`Foo.bar()`]
+   |                                 ^^^^^^^^^ no item named `Foo.bar` in scope
+   |
+   = help: to escape `[` and `]` characters, add '\' before them like `\[` or `\]`
+
+error: unresolved link to `Bar`
+  --> $DIR/bad-intra-doc.rs:12:38
+   |
+LL | /// classic broken intra-doc link: [`Bar`]
+   |                                      ^^^ no item named `Bar` in scope
+   |
+   = help: to escape `[` and `]` characters, add '\' before them like `\[` or `\]`
+
+error: aborting due to 3 previous errors
+