crashlog.py: Improve regular expressions

[aprantl]

Summary

This is yet another change to the regular expressions in crashlog.py that fix a few edge cases, and attempt to improve the readability quite a bit in the process. My last change to support spaces in filenames introduced a bug that caused the version/archspec field to be parsed as part of the image name.

So in 0x1111111 - 0x22222 +MyApp Pro arm64 <01234>, the name of the image was recognized as MyApp Pro arm64 instead of MyApp Pro with a "version" of arm64.

The bugfix makes the space following an optional field mandatory *inside* the optional group.

rdar://problem/56883435

[jasonmolenda]

I'll be honest I'm having trouble thinking of all the possible problems these regexes might have just by looking at them, but overall this looks good to me.

Should we add a couple of extra file-check tests for variations in the version number part of the binary list in a crash report? Looking around some random crash reports, I see number-number "(6.9 - 1674.102)" I see one where the framework name is in there "(1.11 - FrameworkName 1.11)" and hah a random hand built lldb crash I have on my system has "(7.0.0svn - 7.0.0svn)".

[aprantl]

Added some more tests as suggested by Jason and made the regex slightly more robust by requiring a / at the beginning of img_path. All crashlogs I found were meeting this requirement.

lldb/examples/python/crashlog.py

 
 class CrashLog(symbolication.Symbolicator):
     """Class that does parses darwin crash logs"""
     parent_process_regex = re.compile('^Parent Process:\s*(.*)\[(\d+)\]')
     thread_state_regex = re.compile('^Thread ([0-9]+) crashed with')
     thread_regex = re.compile('^Thread ([0-9]+)([^:]*):(.*)')
     app_backtrace_regex = re.compile(
         '^Application Specific Backtrace ([0-9]+)([^:]*):(.*)')
-    frame_regex = re.compile('^([0-9]+)\s+(.+?)\s+(0x[0-9a-fA-F]{7}[0-9a-fA-F]+) +(.*)')
-    null_frame_regex = re.compile('^([0-9]+)\s+\?\?\?\s+(0{7}0+) +(.*)')
-    image_regex_uuid = re.compile(
-        '(0x[0-9a-fA-F]+)[-\s]+(0x[0-9a-fA-F]+)\s+[+]?(.+?)\s+(\(.+\))?\s?(<([-0-9a-fA-F]+)>)? (.*)')
+    version = r'(\(.+\)|(arm|x86_)[0-9a-z]+)\s+'
+    frame_regex = re.compile(r'^([0-9]+)' r'\s'                # id
+                             r'+(.+?)'    r'\s+'               # img_name
+                             r'(' +version+ r')?'              # img_version
+                             r'(0x[0-9a-fA-F]{7}[0-9a-fA-F]+)' # addr
+                             r' +(.*)'                         # offs
+                            )
+    null_frame_regex = re.compile(r'^([0-9]+)\s+\?\?\?\s+(0{7}0+) +(.*)')
+    image_regex_uuid = re.compile(r'(0x[0-9a-fA-F]+)'            # img_lo
+                                  r'\s+' '-' r'\s+'              #   -
+                                  r'(0x[0-9a-fA-F]+)'     r'\s+' # img_hi
+                                  r'[+]?(.+?)'            r'\s+' # img_name
+                                  r'(' +version+ ')?'            # img_version
+                                  r'(<([-0-9a-fA-F]+)>\s+)?'     # img_uuid
+                                  r'(/.*)'                       # img_path
+                                 )
     empty_line_regex = re.compile('^$')
 
     class Thread:
         """Class that represents a thread in a darwin crash log"""
 
         def __init__(self, index, app_specific_backtrace):
             self.index = index
             self.frames = list()
def __init__(self, path, verbose):
             elif parse_mode == PARSE_MODE_THREAD:
                 if line.startswith('Thread'):
                     continue
                 if self.null_frame_regex.search(line):
                     print('warning: thread parser ignored null-frame: "%s"' % line)
                     continue
                 frame_match = self.frame_regex.search(line)
                 if frame_match:
-                    ident = frame_match.group(2)
+                    (frame_id, frame_img_name, _, frame_img_version, _,
+                     frame_addr, frame_ofs) = frame_match.groups()
+                    ident = frame_img_name
                     thread.add_ident(ident)
                     if ident not in self.idents:
                         self.idents.append(ident)
-                    thread.frames.append(CrashLog.Frame(int(frame_match.group(1)), int(
-                        frame_match.group(3), 0), frame_match.group(4)))
+                    thread.frames.append(CrashLog.Frame(int(frame_id), int(
+                        frame_addr, 0), frame_ofs))
                 else:
                     print('error: frame regex failed for line: "%s"' % line)
             elif parse_mode == PARSE_MODE_IMAGES:
                 image_match = self.image_regex_uuid.search(line)
                 if image_match:
-                    (img_lo, img_hi, img_name, img_version,
+                    (img_lo, img_hi, img_name, _, img_version, _,
                      _, img_uuid, img_path) = image_match.groups()
                     image = CrashLog.DarwinImage(int(img_lo, 0), int(img_hi, 0),
                                                  img_name.strip(),
                                                  img_version.strip()
                                                  if img_version else "",
                                                  uuid.UUID(img_uuid), img_path,
                                                  self.verbose)
                     self.images.append(image)

lldb/test/Shell/Python/crashlog.test

 #                                                                 -*- python -*-
 # REQUIRES: system-darwin
+# DEBUG: cd %S/../../../examples/python && cat %s | %lldb && false
 # RUN: cd %S/../../../examples/python && cat %s | %lldb | FileCheck %s
 # CHECK-LABEL: {{S}}KIP BEYOND CHECKS
 script
 import crashlog
 cl = crashlog.CrashLog
 images = [
 "0x10b60b000 - 0x10f707fff com.apple.LLDB.framework (1.1000.11.38.2 - 1000.11.38.2) <96E36F5C-1A83-39A1-8713-5FDD9701C3F1> /Applications/Xcode.app/Contents/SharedFrameworks/LLDB.framework/LLDB",
 # CHECK: 0x10b60b000

 # CHECK: libc++.1.dylib
 # CHECK: (400.9.4)
 # CHECK: D4AB366F-48A9-3C7D-91BD-41198F69DD57
 # CHECK: /usr/lib/libc++.1.dylib
 
 "0x1111111 - 0x22222 +MyApp Pro arm64  <01234> /tmp/MyApp Pro.app/MyApp Pro",
 # CHECK: 0x1111111
 # CHECK: 0x22222
-# CHECK: MyApp Pro arm64
-# CHECK: None
+# CHECK: MyApp Pro
+# CHECK: arm64
 # CHECK: 01234
 # CHECK: /tmp/MyApp Pro.app/MyApp Pro
 
 "0x1111111 - 0x22222 +MyApp Pro (0) <01234> /tmp/MyApp Pro.app/MyApp Pro",
 # CHECK: 0x1111111
 # CHECK: 0x22222
 # CHECK: MyApp Pro
 # CHECK: (0)
 # CHECK: 01234
 # CHECK: /tmp/MyApp Pro.app/MyApp Pro
 
-"0x7fff63f20000 - 0x7fff63f77ff7  libc++.1.dylib (400.9.4) /usr/lib/libc++.1.dylib"
+"0x1111111 - 0x2222222 MyFramework Plus.dylib (1.11 - MyFramework 1.11) <01234> /tmp/MyFramework Plus.dylib",
+# CHECK: 0x1111111
+# CHECK: 0x2222222
+# CHECK: MyFramework Plus.dylib
+# CHECK: ({{.*}}
+# CHECK: 1.11 - MyFramework 1.11
+# CHECK: <{{.*}}
+# CHECK: 01234
+# CHECK: /tmp/MyFramework Plus.dylib
+
+"0x1111111 - 0x2222222 MyFramework-dev.dylib (1.0.0svn - 1.0.0svn) <01234> /MyFramework-dev.dylib",
+# CHECK: 0x1111111
+# CHECK: 0x2222222
+# CHECK: MyFramework-dev.dylib
+# CHECK: ({{.*}}
+# CHECK: 1.0.0svn - 1.0.0svn
+# CHECK: <{{.*}}
+# CHECK: 01234
+# CHECK: /MyFramework-dev.dylib
+
+"0x7fff63f20000 - 0x7fff63f77ff7  libc++.1.dylib (400.9.4) /usr/lib/libc++.1.dylib",
 # CHECK: 0x7fff63f20000
 # CHECK: 0x7fff63f77ff7
 # CHECK: libc++.1.dylib
-# CHECK: (400.9.4)
+# CHECK: ({{.*}}
+# CHECK: 400.9.4
+# CHECK: None
 # CHECK: None
 # CHECK: /usr/lib/libc++.1.dylib
+
+"0x1047b8000 - 0x10481ffff dyld arm64e  <cfa789d10da63f9a8996daf84ed9d04f> /usr/lib/dyld"
+# CHECK: 0x1047b8000
+# CHECK: 0x10481ffff
+# CHECK: dyld
+# CHECK: {{.*}}
+# CHECK: arm64e
+# CHECK: <{{.*}}
+# CHECK: cfa789d10da63f9a8996daf84ed9d04f
+# CHECK: /usr/lib/dyld
 ]
 # CHECK-LABEL: FRAMES
 frames = [
 "0 libsystem_kernel.dylib          0x00007fff684b78a6 read + 10",
 # CHECK: 0
 # CHECK: libsystem_kernel.dylib
 # CHECK: 0x00007fff684b78a6
 # CHECK: read + 10
 "1   com.apple.LLDB.framework        0x000000010f7954af lldb_private::HostNativeThreadBase::ThreadCreateTrampoline(void*) + 105",
 # CHECK: 1
 # CHECK: com.apple.LLDB.framework
 # CHECK: 0x000000010f7954af
 # CHECK: lldb_private{{.*}} + 105
 "2   MyApp Pro arm64     0x000000019b0db3a8 foo + 72",
 # CHECK: 2
-# CHECK: MyApp Pro arm64
+# CHECK: MyApp Pro
+# CHECK: a
+# CHECK: arm64
+# CHECK: a
 # CHECK: 0x000000019b0db3a8
 # CHECK: foo + 72
 "3   He 0x1      0x000000019b0db3a8 foo + 72"
 # CHECK: 3
 # CHECK: He 0x1
 # CHECK: 0x000000019b0db3a8
 # CHECK: foo + 72
 ]