Kristian Rickert created OPENNLP-1935:
-----------------------------------------
Summary: RegEx removal part 7: checkstyle guard for
java.util.regex in production code
Key: OPENNLP-1935
URL: https://issues.apache.org/jira/browse/OPENNLP-1935
Project: OpenNLP
Issue Type: Improvement
Components: Build, Packaging and Test
Reporter: Kristian Rickert
Assignee: Kristian Rickert
Two checkstyle checks keep the policy in place after the epic is merged:
* {{NoRegexImport}}: {{IllegalImport}} of {{java.util.regex}}.
* {{NoRegexStringCall}}: {{RegexpSinglelineJava}} for {{String.replaceAll}},
{{replaceFirst}}, {{matches}} with a string argument, and {{split}} with a
backslash-escaped argument, since each compiles a Pattern per call.
Test sources are excluded. The documented exceptions live in a new
{{checkstyle-suppressions.xml}}, wired through {{suppressionsLocation}} in the
parent pom, each with its reason: {{RegexNameFinder}} and its factory,
{{AncoraSpanishHeadRules}}, {{Parse}}, {{NameSample}},
{{SpellCheckingCharSequenceNormalizer}}, and the tokenizer alphanumeric classes
from part 6.
Run on parts 1 to 6 alone the checks report exactly the sites those parts fix;
with all parts applied the whole reactor validates clean. This ticket merges
last; its build fails until the other parts are in.
For the epic description, add a "Tickets" section listing the seven keys in
merge order 2, 3, 4, 5, bug, 6, 7, move the tokenizer row from "still
determining" to exempt, add the spellcheck normalizer and the part 5 rows to
the tables, and note that part 1 also added the four StringUtil helpers the
later parts reuse. Once the keys exist, give them to me in that order and I
reword the commits, rename the branches, and push.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
