diff mbox series

Avoid spurious breakpoint-setting failure in DAP

Message ID 20230905125957.1879149-1-tromey@adacore.com
State New
Headers
Series Avoid spurious breakpoint-setting failure in DAP |

Checks

Context Check Description
linaro-tcwg-bot/tcwg_gdb_build--master-aarch64 success Testing passed
linaro-tcwg-bot/tcwg_gdb_build--master-arm success Testing passed
linaro-tcwg-bot/tcwg_gdb_check--master-aarch64 success Testing passed
linaro-tcwg-bot/tcwg_gdb_check--master-arm success Testing passed

Commit Message

Tom Tromey Sept. 5, 2023, 12:59 p.m. UTC 
  A user pointed out that if a DAP setBreakpoints request has a 'source'
field in a SourceBreakpoint object, then the gdb DAP implementation
will throw an exception.

While SourceBreakpoint does not allow 'source' in the spec, it seems
better to me to accept it.  I don't think we should fully go down the
"Postel's Law" path -- after all, we have the type-checker -- but at
the same time, if we do send errors, they should be intentional and
not artifacts of the implementation.

Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30820
---
 gdb/python/lib/gdb/dap/breakpoint.py | 8 +++++++-
 gdb/testsuite/gdb.dap/basic-dap.exp  | 9 +++++++--
 2 files changed, 14 insertions(+), 3 deletions(-)

Comments

Tom Tromey Sept. 12, 2023, 4:31 p.m. UTC | #1 
>>>>> "Tom" == Tom Tromey via Gdb-patches <gdb-patches@sourceware.org> writes:

Tom> A user pointed out that if a DAP setBreakpoints request has a 'source'
Tom> field in a SourceBreakpoint object, then the gdb DAP implementation
Tom> will throw an exception.

Tom> While SourceBreakpoint does not allow 'source' in the spec, it seems
Tom> better to me to accept it.  I don't think we should fully go down the
Tom> "Postel's Law" path -- after all, we have the type-checker -- but at
Tom> the same time, if we do send errors, they should be intentional and
Tom> not artifacts of the implementation.

Tom> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30820

This one is just a simple DAP-specific bug fix, so I'm checking it in.

Tom
diff mbox series

Patch

diff --git a/gdb/python/lib/gdb/dap/breakpoint.py b/gdb/python/lib/gdb/dap/breakpoint.py
index 76ff129ddbc..1622ae69023 100644
--- a/gdb/python/lib/gdb/dap/breakpoint.py
+++ b/gdb/python/lib/gdb/dap/breakpoint.py
@@ -273,7 +273,13 @@  def set_breakpoint(*, source, breakpoints: Sequence = (), **args):
     if "path" not in source:
         result = []
     else:
-        specs = [_rewrite_src_breakpoint(source=source, **bp) for bp in breakpoints]
+        # Setting 'source' in BP avoids any Python error if BP already
+        # has a 'source' parameter.  Setting this isn't in the spec,
+        # but it is better to be safe.  See PR dap/30820.
+        specs = []
+        for bp in breakpoints:
+            bp["source"] = source
+            specs.append(_rewrite_src_breakpoint(**bp))
         # Be sure to include the path in the key, so that we only
         # clear out breakpoints coming from this same source.
         key = "source:" + source["path"]
diff --git a/gdb/testsuite/gdb.dap/basic-dap.exp b/gdb/testsuite/gdb.dap/basic-dap.exp
index 7dfc759baac..98ca739ccab 100644
--- a/gdb/testsuite/gdb.dap/basic-dap.exp
+++ b/gdb/testsuite/gdb.dap/basic-dap.exp
@@ -87,10 +87,15 @@  foreach entry [dict get [lindex $obj 0] body breakpoints] {
     incr first_line
 }
 
+# Note that in this request, we add a 'source' field to the
+# SourceBreakpoint object.  This isn't in the spec but it once caused
+# an incorrect exception in the Python code.  See PR dap/30820.
 set obj [dap_check_request_and_response "reset breakpoint by line number" \
 	     setBreakpoints \
-	     [format {o source [o path [%s]] breakpoints [a [o line [i %d]]]} \
-		  [list s $srcfile] $line]]
+	     [format {o source [o path [%s]] \
+			  breakpoints [a [o source [o path [%s]] \
+					      line [i %d]]]} \
+		  [list s $srcfile] [list s $srcfile] $line]]
 set new_line_bpno [dap_get_breakpoint_number $obj]
 gdb_assert {$new_line_bpno == $line_bpno} "re-setting kept same breakpoint number"