fix log values with non-utf8 characters and remove index_log (#20173)

* remove index_log

* handle non-unicode characters in values

* dump.py better handling of printing non-unicode characters

* only replace errors in main

.dockerignore

@@ -26,10 +26,8 @@ chffr/backend/env
 selfdrive/nav
 selfdrive/baseui
 chffr/lib/vidindex/vidindex
-chffr/lib/index_log/index_log
 selfdrive/test/simulator2
 **/cache_data
-xx/chffr/lib/index_log/index_log
 xx/chffr/lib/vidindex/vidindex
 xx/plus
 xx/community

SConstruct

@@ -392,7 +392,6 @@ if arch != "Darwin":
 
 if real_arch == "x86_64":
   SConscript(['tools/nui/SConscript'])
-  SConscript(['tools/lib/index_log/SConscript'])
 
 external_sconscript = GetOption('external_sconscript')
 if external_sconscript:

selfdrive/debug/dump.py

@@ -4,6 +4,8 @@ import sys
 import argparse
 import json
 from hexdump import hexdump
+import codecs
+codecs.register_error("strict", codecs.backslashreplace_errors)
 
 from cereal import log
 import cereal.messaging as messaging

tools/lib/index_log/.gitignore

@@ -1 +0,0 @@
-index_log

tools/lib/index_log/Makefile

@@ -1,9 +0,0 @@
-CC := gcc
-CXX := g++
-
-index_log: index_log.cc
-	$(eval $@_TMP := $(shell mktemp))
-	$(CXX) -std=c++1z -o $($@_TMP) \
-    index_log.cc \
-    -lcapnp -lkj
-	mv $($@_TMP) $@

tools/lib/index_log/SConscript

@@ -1,3 +0,0 @@
-Import('env', 'arch', 'messaging')
-
-env.Program('index_log', ['index_log.cc'], LIBS=[messaging, 'kj', 'capnp'])

tools/lib/index_log/index_log.cc

@@ -1,63 +0,0 @@
-#include <cstdio>
-#include <cstdlib>
-#include <cassert>
-#include <string>
-
-#include <unistd.h>
-#include <fcntl.h>
-
-#include <sys/stat.h>
-#include <sys/mman.h>
-
-#include <kj/io.h>
-#include <capnp/serialize.h>
-
-int main(int argc, char** argv) {
-
-  if (argc != 3) {
-    printf("usage: %s <log_path> <index_output_path>\n", argv[0]);
-    return 1;
-  }
-
-  const std::string log_fn = argv[1];
-  const std::string index_fn = argv[2];
-
-  int log_fd = open(log_fn.c_str(), O_RDONLY, 0);
-  assert(log_fd >= 0);
-
-  off_t log_size = lseek(log_fd, 0, SEEK_END);
-  lseek(log_fd, 0, SEEK_SET);
-
-  FILE* index_f = NULL;
-  if (index_fn == "-") {
-    index_f = stdout;
-  } else {
-    index_f = fopen(index_fn.c_str(), "wb");
-  }
-  assert(index_f);
-
-  void* log_data = mmap(NULL, log_size, PROT_READ, MAP_PRIVATE, log_fd, 0);
-  assert(log_data);
-
-  auto words = kj::arrayPtr((const capnp::word*)log_data, log_size/sizeof(capnp::word));
-  while (words.size() > 0) {
-    uint64_t idx = ((uintptr_t)words.begin() - (uintptr_t)log_data);
-    // printf("%llu - %ld\n", idx, words.size());
-    const char* idx_bytes = (const char*)&idx;
-    fwrite(idx_bytes, 8, 1, index_f);
-    try {
-      capnp::FlatArrayMessageReader reader(words);
-      words = kj::arrayPtr(reader.getEnd(), words.end());
-    } catch (const kj::Exception& exc) {
-      break;
-    }
-  }
-
-  munmap(log_data, log_size);
-
-  fclose(index_f);
-
-  close(log_fd);
-
-  return 0;
-}

tools/lib/logreader.py

@@ -2,48 +2,15 @@
 import os
 import sys
 import bz2
-import tempfile
-import subprocess
 import urllib.parse
 import capnp
-import numpy as np
 
-from tools.lib.exceptions import DataUnreadableError
 try:
   from xx.chffr.lib.filereader import FileReader
 except ImportError:
   from tools.lib.filereader import FileReader
 from cereal import log as capnp_log
 
-OP_PATH = os.path.dirname(os.path.dirname(capnp_log.__file__))
-
-def index_log(fn):
-  index_log_dir = os.path.join(os.path.dirname(os.path.realpath(__file__)), "index_log")
-  index_log = os.path.join(index_log_dir, "index_log")
-
-  if not os.path.exists(index_log):
-    phonelibs_dir = os.path.join(OP_PATH, 'phonelibs')
-    subprocess.check_call(["make", "PHONELIBS=" + phonelibs_dir], cwd=index_log_dir, stdout=subprocess.DEVNULL)
-
-  try:
-    dat = subprocess.check_output([index_log, fn, "-"])
-  except subprocess.CalledProcessError as e:
-    raise DataUnreadableError("%s capnp is corrupted/truncated" % fn) from e
-  return np.frombuffer(dat, dtype=np.uint64)
-
-def event_read_multiple_bytes(dat):
-  with tempfile.NamedTemporaryFile() as dat_f:
-    dat_f.write(dat)
-    dat_f.flush()
-    idx = index_log(dat_f.name)
-
-  end_idx = np.uint64(len(dat))
-  idx = np.append(idx, end_idx)
-
-  return [capnp_log.Event.from_bytes(dat[idx[i]:idx[i+1]])
-          for i in range(len(idx)-1)]
-
-
 # this is an iterator itself, and uses private variables from LogReader
 class MultiLogIterator(object):
   def __init__(self, log_paths, wraparound=True):
@@ -117,17 +84,17 @@ class LogReader(object):
 
     if ext == "":
       # old rlogs weren't bz2 compressed
-      ents = event_read_multiple_bytes(dat)
+      ents = capnp_log.Event.read_multiple_bytes(dat)
     elif ext == ".bz2":
       dat = bz2.decompress(dat)
-      ents = event_read_multiple_bytes(dat)
+      ents = capnp_log.Event.read_multiple_bytes(dat)
     else:
       raise Exception(f"unknown extension {ext}")
 
-    self._ts = [x.logMonoTime for x in ents]
+    self._ents = list(ents)
+    self._ts = [x.logMonoTime for x in self._ents]
     self.data_version = data_version
     self._only_union_types = only_union_types
-    self._ents = ents
 
   def __iter__(self):
     for ent in self._ents:
@@ -141,6 +108,10 @@ class LogReader(object):
         yield ent
 
 if __name__ == "__main__":
+  import codecs
+  # capnproto <= 0.8.0 throws errors converting byte data to string
+  # below line catches those errors and replaces the bytes with \x__
+  codecs.register_error("strict", codecs.backslashreplace_errors)
   log_path = sys.argv[1]
   lr = LogReader(log_path)
   for msg in lr: