[OE-core] [PATCH] oeqa/logparser: Fix performance issues with ptest log parsing

[Richard Purdie]

On the autobuilder a ptest log with 2.1 million lines took around 18 hours to
process. This is clearly crazy.

We can tweak the processing code to:
a) Stop repeatedly joining large strings together (append to a list instead)
b) Use one startswith expression instead of multiple re.search() operations

With this change it takes 18 hours down to around 12s.

Signed-off-by: Richard Purdie <richard.purdie at linuxfoundation.org>
---
 meta/lib/oeqa/utils/logparser.py | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/meta/lib/oeqa/utils/logparser.py b/meta/lib/oeqa/utils/logparser.py
index 7313df8ec35..54037210736 100644
--- a/meta/lib/oeqa/utils/logparser.py
+++ b/meta/lib/oeqa/utils/logparser.py
@@ -25,13 +25,20 @@ class PtestParser(object):
         section_regex['exitcode'] = re.compile(r"^ERROR: Exit status is (.+)")
         section_regex['timeout'] = re.compile(r"^TIMEOUT: .*/(.+)/ptest")
 
+        # Cache markers so we don't take the re.search() hit all the time.
+        markers = ("PASSED", "FAILED", "SKIPPED", "BEGIN:", "END:", "DURATION:", "ERROR: Exit", "TIMEOUT:")
+
         def newsection():
-            return { 'name': "No-section", 'log': "" }
+            return { 'name': "No-section", 'log': [] }
 
         current_section = newsection()
 
         with open(logfile, errors='replace') as f:
             for line in f:
+                if not line.startswith(markers):
+                    current_section['log'].append(line)
+                    continue
+
                 result = section_regex['begin'].search(line)
                 if result:
                     current_section['name'] = result.group(1)
@@ -61,7 +68,7 @@ class PtestParser(object):
                         current_section[t] = result.group(1)
                         continue
 
-                current_section['log'] = current_section['log'] + line 
+                current_section['log'].append(line)
 
                 for t in test_regex:
                     result = test_regex[t].search(line)
@@ -70,6 +77,11 @@ class PtestParser(object):
                             self.results[current_section['name']] = {}
                         self.results[current_section['name']][result.group(1).strip()] = t
 
+        # Python performance for repeatedly joining long strings is poor, do it all at once at the end.
+        # For 2.1 million lines in a log this reduces 18 hours to 12s.
+        for section in self.sections:
+            self.sections[section]['log'] = "".join(self.sections[section]['log'])
+
         return self.results, self.sections
 
     # Log the results as files. The file name is the section name and the contents are the tests in that section.
-- 
2.25.0