arthurscchan commented on code in PR #54912:
URL: https://github.com/apache/arrow/pull/54912#discussion_r4629856575


##########
cpp/build-support/fuzzing/dicts/arrow-csv-fuzz.dict:
##########
@@ -0,0 +1,60 @@
+# Tokens for arrow-csv-fuzz: CSV dialect characters, null/bool spellings,
+# numeric edge cases and timestamp formats understood by the Arrow CSV reader.

Review Comment:
   In general, dictionaries are used to help random data mutation construct 
inputs that can get past certain magic numbers and checks. This applies when 
some of the logic contains a check requiring specific patterns or values in 
order to get past it and reach deeper logic. In general, a fuzzer mutates data 
at random, mostly guided by coverage. But with a dictionary, the mutation 
engine tries specific input combinations or characters, making fuzzing more 
likely to generate an input that passes a magic check or reaches deeper logic.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to