Regen and simplify

pablogsal · Dec 22, 2025 · Dec 15, 2025 · Dec 15, 2025 · Dec 15, 2025 · Dec 15, 2025
commit 1e2400b53f694fb7757d3138c552d451f4268288
diff --git a/Include/internal/pycore_global_objects_fini_generated.h b/Include/internal/pycore_global_objects_fini_generated.h
diff --git a/Include/internal/pycore_global_strings.h b/Include/internal/pycore_global_strings.h
@@ -376,9 +376,11 @@ struct _Py_global_strings {
         STRUCT_FOR_ID(co_varnames)
         STRUCT_FOR_ID(code)
         STRUCT_FOR_ID(col_offset)
+        STRUCT_FOR_ID(collector)
         STRUCT_FOR_ID(command)
         STRUCT_FOR_ID(comment_factory)
         STRUCT_FOR_ID(compile_mode)
+        STRUCT_FOR_ID(compression)
         STRUCT_FOR_ID(config)
         STRUCT_FOR_ID(consts)
         STRUCT_FOR_ID(context)
@@ -441,7 +443,9 @@ struct _Py_global_strings {
         STRUCT_FOR_ID(event)
         STRUCT_FOR_ID(eventmask)
         STRUCT_FOR_ID(exc)
+        STRUCT_FOR_ID(exc_tb)
         STRUCT_FOR_ID(exc_type)
+        STRUCT_FOR_ID(exc_val)
         STRUCT_FOR_ID(exc_value)
         STRUCT_FOR_ID(excepthook)
         STRUCT_FOR_ID(exception)
@@ -697,6 +701,7 @@ struct _Py_global_strings {
         STRUCT_FOR_ID(print_file_and_line)
         STRUCT_FOR_ID(priority)
         STRUCT_FOR_ID(progress)
+        STRUCT_FOR_ID(progress_callback)
         STRUCT_FOR_ID(progress_routine)
         STRUCT_FOR_ID(proto)
         STRUCT_FOR_ID(protocol)
@@ -736,6 +741,7 @@ struct _Py_global_strings {
         STRUCT_FOR_ID(reversed)
         STRUCT_FOR_ID(rounding)
         STRUCT_FOR_ID(salt)
+        STRUCT_FOR_ID(sample_interval_us)
         STRUCT_FOR_ID(sched_priority)
         STRUCT_FOR_ID(scheduler)
         STRUCT_FOR_ID(script)
@@ -775,8 +781,10 @@ struct _Py_global_strings {
         STRUCT_FOR_ID(spam)
         STRUCT_FOR_ID(src)
         STRUCT_FOR_ID(src_dir_fd)
+        STRUCT_FOR_ID(stack_frames)
         STRUCT_FOR_ID(stacklevel)
         STRUCT_FOR_ID(start)
+        STRUCT_FOR_ID(start_time_us)
         STRUCT_FOR_ID(statement)
         STRUCT_FOR_ID(stats)
         STRUCT_FOR_ID(status)
@@ -817,6 +825,7 @@ struct _Py_global_strings {
         STRUCT_FOR_ID(times)
         STRUCT_FOR_ID(timespec)
         STRUCT_FOR_ID(timestamp)
+        STRUCT_FOR_ID(timestamp_us)
         STRUCT_FOR_ID(timetuple)
         STRUCT_FOR_ID(timeunit)
         STRUCT_FOR_ID(top)

diff --git a/Include/internal/pycore_runtime_init_generated.h b/Include/internal/pycore_runtime_init_generated.h
diff --git a/Include/internal/pycore_unicodeobject_generated.h b/Include/internal/pycore_unicodeobject_generated.h
@@ -174,20 +174,6 @@ frames from the previous stack are kept, and new frames are prepended.
 Used when the code path changed: some frames were popped (function returns)
 and new frames were pushed (different function calls).
 
-### Compression Benefits
-
-This delta encoding provides massive savings for typical profiling workloads:
-
-- **CPU-bound code**: Hot loops produce many identical samples. RLE encoding
-  compresses 100 identical samples to just 2-3 bytes of overhead plus the
-  timestamp/status data.
-
-- **I/O-bound code**: Alternating between wait and work produces similar
-  stacks with small variations. SUFFIX encoding captures this efficiently.
-
-- **Call-heavy code**: Functions calling other functions share common stack
-  prefixes. POP_PUSH encoding only stores the changed frames.
-
 ### Thread and Interpreter Identification
 
 Thread IDs are 64-bit values that can be large (memory addresses on some

@@ -98,7 +98,6 @@ def export(self, filename=None):
 
     @property
     def total_samples(self):
-        """Total number of samples written."""
         return self._writer.total_samples
 
     def get_stats(self):
@@ -111,7 +110,6 @@ def get_stats(self):
         return self._writer.get_stats()
 
     def __enter__(self):
-        """Context manager entry."""
         return self
 
     def __exit__(self, exc_type, exc_val, exc_tb):

@@ -23,13 +23,11 @@ def __init__(self, filename):
         self._reader = None
 
     def __enter__(self):
-        """Open the binary file for reading."""
         import _remote_debugging
         self._reader = _remote_debugging.BinaryReader(self.filename)
         return self
 
     def __exit__(self, exc_type, exc_val, exc_tb):
-        """Close the binary file."""
         if self._reader is not None:
             self._reader.close()
             self._reader = None
@@ -71,7 +69,6 @@ def replay_samples(self, collector, progress_callback=None):
 
     @property
     def sample_count(self):
-        """Number of samples in the file."""
         if self._reader is None:
             raise RuntimeError("Reader not open. Use as context manager.")
         return self._reader.get_info()['sample_count']

@@ -102,7 +102,7 @@ def _parse_mode(mode_string):
 def _check_process_died(process):
     """Check if process died and raise an error with stderr if available."""
     if process.poll() is None:
-        return  # Process still running
+        return
 
     # Process died - try to get stderr for error message
     stderr_msg = ""
@@ -471,7 +471,6 @@ def _handle_output(collector, args, pid, mode):
                 sort_mode, limit, not args.no_summary, mode
             )
     else:
-        # Export to file
         filename = args.outfile or _generate_output_filename(args.format, pid)
         collector.export(filename)
 
@@ -726,7 +725,6 @@ def _handle_attach(args):
         mode != PROFILING_MODE_WALL if mode != PROFILING_MODE_ALL else False
     )
 
-    # For binary format, determine output file before creating collector
     output_file = None
     if args.format == "binary":
         output_file = args.outfile or _generate_output_filename(args.format, args.pid)
@@ -806,7 +804,6 @@ def _handle_run(args):
         mode != PROFILING_MODE_WALL if mode != PROFILING_MODE_ALL else False
     )
 
-    # For binary format, determine output file before creating collector
     output_file = None
     if args.format == "binary":
         output_file = args.outfile or _generate_output_filename(args.format, process.pid)
@@ -940,7 +937,6 @@ def _handle_replay(args):
     """Handle the 'replay' command - convert binary profile to another format."""
     import os
 
-    # Check input file exists
     if not os.path.exists(args.input_file):
         sys.exit(f"Error: Input file not found: {args.input_file}")
 
@@ -956,10 +952,8 @@ def _handle_replay(args):
         print(f"  Sample interval: {interval} us")
         print(f"  Compression: {'zstd' if info.get('compression_type', 0) == 1 else 'none'}")
 
-        # Create appropriate collector
         collector = _create_collector(args.format, interval, skip_idle=False)
 
-        # Replay with progress bar
         def progress_callback(current, total):
             if total > 0:
                 pct = current / total
@@ -969,20 +963,17 @@ def progress_callback(current, total):
                 print(f"\r  [{bar}] {pct*100:5.1f}% ({current:,}/{total:,})", end="", flush=True)
 
         count = reader.replay_samples(collector, progress_callback)
-        print()  # Newline after progress bar
+        print()
 
-        # Handle output similar to other formats
         if args.format == "pstats":
             if args.outfile:
                 collector.export(args.outfile)
             else:
-                # Print to stdout with defaults applied
                 sort_choice = args.sort if args.sort is not None else "nsamples"
                 limit = args.limit if args.limit is not None else 15
                 sort_mode = _sort_to_mode(sort_choice)
                 collector.print_stats(sort_mode, limit, not args.no_summary, PROFILING_MODE_WALL)
         else:
-            # Export to file
             filename = args.outfile or _generate_output_filename(args.format, os.getpid())
             collector.export(filename)