[GitHub] [nifi-minifi-cpp] fgerlits commented on a change in pull request #1188: MINIFICPP-1651: Added DefragmentText processor

[GitBox <gi...@apache.org>]

fgerlits commented on a change in pull request #1188:
URL: https://github.com/apache/nifi-minifi-cpp/pull/1188#discussion_r735408467



##########
File path: PROCESSORS.md
##########
@@ -295,6 +296,29 @@ In the list below, the names of required properties appear in bold. Any other pr
 | - | - |
 |success|FlowFiles that are sent successfully to the destination are transferred to this relationship|
 
+## DefragmentText
+
+### Description
+
+DefragmentText splits and merges incoming flowfiles so cohesive messages are not split between them
+### Properties
+
+In the list below, the names of required properties appear in bold. Any other properties (not in bold) are considered optional. The table also indicates any default values, and whether a property supports the NiFi Expression Language.
+
+| Name | Default Value | Allowable Values | Description |
+| - | - | - | - |
+|**Pattern**|||A regular expression to match at the start or end of messages.|
+|Pattern Location|Start of Message|Start of Message<br>End of Message|Whether the pattern is located at the start or at the end of the messages.|
+|MaxBufferAge||&lt;duration&gt; &lt;time unit&gt;|The maximum age of a buffer after which the buffer will be transferred to failure.|
+|MaxBufferSize||&lt;size&gt; &lt;size unit&gt;|The maximum buffer size, if the buffer exceed this, it will be transferred to failure.|
+
+### Relationships
+
+| Name | Description |
+| - | - |
+|success|Flowfiles that have no fragmented messages in them|

Review comment:
       "... or have been successfully defragmented"?

##########
File path: docker/test/integration/minifi/processors/DefragmentText.py
##########
@@ -0,0 +1,9 @@
+from ..core.Processor import Processor
+
+
+class DefragmentText(Processor):
+    def __init__(self, delimiter="<[0-9]+>", schedule={'scheduling period': '2 sec'}):

Review comment:
       I think it would make more sense to make it event driven by default:
   ```suggestion
       def __init__(self, delimiter="<[0-9]+>", schedule={'scheduling strategy': 'EVENT_DRIVEN'}):
   ```

##########
File path: extensions/standard-processors/processors/DefragmentText.cpp
##########
@@ -0,0 +1,340 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "DefragmentText.h"
+
+#include <vector>
+
+#include "core/Resource.h"
+#include "serialization/PayloadSerializer.h"
+#include "TextFragmentUtils.h"
+#include "utils/gsl.h"
+
+namespace org::apache::nifi::minifi::processors {
+
+const core::Relationship DefragmentText::Success("success", "Flowfiles that have no fragmented messages in them");
+const core::Relationship DefragmentText::Failure("failure", "Flowfiles that failed the defragmentation process");
+const core::Relationship DefragmentText::Self("__self__", "Marks the FlowFile to be owned by this processor");
+
+const core::Property DefragmentText::Pattern(
+    core::PropertyBuilder::createProperty("Pattern")
+        ->withDescription("A regular expression to match at the start or end of messages.")
+        ->isRequired(true)->build());
+
+const core::Property DefragmentText::PatternLoc(
+    core::PropertyBuilder::createProperty("Pattern Location")->withDescription("Whether the pattern is located at the start or at the end of the messages.")
+        ->withAllowableValues(PatternLocation::values())
+        ->withDefaultValue(toString(PatternLocation::START_OF_MESSAGE))->build());
+
+
+const core::Property DefragmentText::MaxBufferSize(
+    core::PropertyBuilder::createProperty("Max Buffer Size")
+        ->withDescription("The maximum buffer size, if the buffer exceeds this, it will be transferred to failure. Expected format is <size> <data unit>")
+        ->withType(core::StandardValidators::get().DATA_SIZE_VALIDATOR)->build());
+
+const core::Property DefragmentText::MaxBufferAge(
+    core::PropertyBuilder::createProperty("Max Buffer Age")->
+        withDescription("The maximum age of a buffer after which the buffer will be transferred to failure. Expected format is <duration> <time unit>")->build());
+
+void DefragmentText::initialize() {
+  setSupportedRelationships({Success, Failure});
+  setSupportedProperties({Pattern, PatternLoc, MaxBufferAge, MaxBufferSize});
+}
+
+void DefragmentText::onSchedule(core::ProcessContext* context, core::ProcessSessionFactory*) {
+  gsl_Expects(context);
+
+  std::string max_buffer_age_str;
+  if (context->getProperty(MaxBufferAge.getName(), max_buffer_age_str)) {
+    core::TimeUnit unit;
+    uint64_t max_buffer_age;
+    if (core::Property::StringToTime(max_buffer_age_str, max_buffer_age, unit) && core::Property::ConvertTimeUnitToMS(max_buffer_age, unit, max_buffer_age)) {
+      buffer_.setMaxAge(std::chrono::milliseconds(max_buffer_age));
+      logger_->log_trace("The Buffer maximum age is configured to be %" PRIu64 " ms", max_buffer_age);
+    }
+  }
+
+  std::string max_buffer_size_str;
+  if (context->getProperty(MaxBufferSize.getName(), max_buffer_size_str)) {
+    uint64_t max_buffer_size = core::DataSizeValue(max_buffer_size_str).getValue();
+    if (max_buffer_size > 0) {
+      buffer_.setMaxSize(max_buffer_size);
+      logger_->log_trace("The Buffer maximum size is configured to be %" PRIu64 " B", max_buffer_size);
+    }
+  }
+
+  context->getProperty(PatternLoc.getName(), pattern_location_);
+
+  std::string pattern_str;
+  if (context->getProperty(Pattern.getName(), pattern_str) && !pattern_str.empty()) {
+    pattern_ = std::regex(pattern_str);
+    logger_->log_trace("The Pattern is configured to be %s", pattern_str);
+  } else {
+    throw Exception(PROCESS_SCHEDULE_EXCEPTION, "Pattern property missing or invalid");
+  }
+}
+
+void DefragmentText::onTrigger(core::ProcessContext*, core::ProcessSession* session) {
+  gsl_Expects(session);
+  auto flowFiles = flow_file_store_.getNewFlowFiles();
+  for (auto& file : flowFiles) {
+    processNextFragment(session, file);
+  }
+  std::shared_ptr<core::FlowFile> original_flow_file = session->get();
+  processNextFragment(session, original_flow_file);
+  if (buffer_.maxAgeReached() || buffer_.maxSizeReached()) {
+    buffer_.flushAndReplace(session, Failure, nullptr);
+  }
+}
+
+void DefragmentText::processNextFragment(core::ProcessSession *session, const std::shared_ptr<core::FlowFile>& next_fragment) {
+  if (!next_fragment)
+    return;
+  if (!buffer_.isCompatible(next_fragment)) {
+    buffer_.flushAndReplace(session, Failure, nullptr);
+    session->transfer(next_fragment, Failure);
+    return;
+  }
+  std::shared_ptr<core::FlowFile> split_before_last_pattern;
+  std::shared_ptr<core::FlowFile> split_after_last_pattern;
+  bool found_pattern = splitFlowFileAtLastPattern(session, next_fragment, split_before_last_pattern,
+                                                  split_after_last_pattern);
+  buffer_.append(session, split_before_last_pattern);
+  if (found_pattern) {
+    buffer_.flushAndReplace(session, Success, split_after_last_pattern);
+  }
+  session->remove(next_fragment);
+}
+
+
+void DefragmentText::updateAttributesForSplitFiles(const std::shared_ptr<const core::FlowFile> &original_flow_file,
+                                                   const std::shared_ptr<core::FlowFile> &split_before_last_pattern,
+                                                   const std::shared_ptr<core::FlowFile> &split_after_last_pattern,
+                                                   const size_t split_position) const {
+  std::string base_name, post_name, offset_str;
+  if (!original_flow_file->getAttribute(textfragmentutils::BASE_NAME_ATTRIBUTE, base_name))
+    return;
+  if (!original_flow_file->getAttribute(textfragmentutils::POST_NAME_ATTRIBUTE, post_name))
+    return;
+  if (!original_flow_file->getAttribute(textfragmentutils::OFFSET_ATTRIBUTE, offset_str))
+    return;
+
+  size_t fragment_offset = std::stoi(offset_str);
+
+  if (split_before_last_pattern) {
+    std::string first_part_name = textfragmentutils::createFileName(base_name, post_name, fragment_offset, split_before_last_pattern->getSize());
+    split_before_last_pattern->setAttribute(core::SpecialFlowAttribute::FILENAME, first_part_name);
+  }
+  if (split_after_last_pattern) {
+    std::string second_part_name = textfragmentutils::createFileName(base_name, post_name, fragment_offset + split_position, split_after_last_pattern->getSize());
+    split_after_last_pattern->setAttribute(core::SpecialFlowAttribute::FILENAME, second_part_name);
+    split_after_last_pattern->setAttribute(textfragmentutils::OFFSET_ATTRIBUTE, std::to_string(fragment_offset + split_position));
+  }
+}
+
+namespace {
+class AppendFlowFileToFlowFile : public OutputStreamCallback {
+ public:
+  explicit AppendFlowFileToFlowFile(const std::shared_ptr<core::FlowFile>& flow_file_to_append, PayloadSerializer& serializer)
+      : flow_file_to_append_(flow_file_to_append), serializer_(serializer) {}
+
+  int64_t process(const std::shared_ptr<io::BaseStream> &stream) override {
+    return serializer_.serialize(flow_file_to_append_, stream);
+  }
+ private:
+  const std::shared_ptr<core::FlowFile>& flow_file_to_append_;
+  PayloadSerializer& serializer_;
+};
+
+void updateAppendedAttributes(const std::shared_ptr<core::FlowFile>& buffered_ff) {
+  std::string base_name, post_name, offset_str;
+  if (!buffered_ff->getAttribute(textfragmentutils::BASE_NAME_ATTRIBUTE, base_name))
+    return;
+  if (!buffered_ff->getAttribute(textfragmentutils::POST_NAME_ATTRIBUTE, post_name))
+    return;
+  if (!buffered_ff->getAttribute(textfragmentutils::OFFSET_ATTRIBUTE, offset_str))
+    return;
+  size_t fragment_offset = std::stoi(offset_str);
+
+  std::string buffer_new_name = textfragmentutils::createFileName(base_name, post_name, fragment_offset, buffered_ff->getSize());
+  buffered_ff->setAttribute(core::SpecialFlowAttribute::FILENAME, buffer_new_name);
+}
+
+class LastPatternFinder : public InputStreamCallback {
+ public:
+  LastPatternFinder(const std::regex &pattern, DefragmentText::PatternLocation pattern_location) : pattern_(pattern), pattern_location_(pattern_location) {}
+
+  ~LastPatternFinder() override = default;
+
+  int64_t process(const std::shared_ptr<io::BaseStream> &stream) override {
+    if (nullptr == stream)
+      return 0;
+    std::string content;
+    content.resize(stream->size());
+    const auto ret = stream->read(reinterpret_cast<uint8_t *>(content.data()), stream->size());
+    if (io::isError(ret))
+      return -1;
+    searchContent(content);
+
+    return 0;
+  }
+
+  const std::optional<size_t> &getLastPatternPosition() const { return last_pattern_location; }
+
+ protected:
+  void searchContent(const std::string &content) {
+    auto matches_begin = std::sregex_iterator(content.begin(), content.end(), pattern_);
+    auto number_of_matches = std::distance(matches_begin, std::sregex_iterator());
+    if (number_of_matches > 0) {
+      auto last_match = std::next(matches_begin, number_of_matches - 1);

Review comment:
       This will do the searching twice: first for `std::distance`, then for `std::next`.  It would be more efficient to iterate once in a loop and store the latest `match_result` object.

##########
File path: extensions/standard-processors/processors/DefragmentText.cpp
##########
@@ -0,0 +1,340 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "DefragmentText.h"
+
+#include <vector>
+
+#include "core/Resource.h"
+#include "serialization/PayloadSerializer.h"
+#include "TextFragmentUtils.h"
+#include "utils/gsl.h"
+
+namespace org::apache::nifi::minifi::processors {
+
+const core::Relationship DefragmentText::Success("success", "Flowfiles that have no fragmented messages in them");
+const core::Relationship DefragmentText::Failure("failure", "Flowfiles that failed the defragmentation process");
+const core::Relationship DefragmentText::Self("__self__", "Marks the FlowFile to be owned by this processor");
+
+const core::Property DefragmentText::Pattern(
+    core::PropertyBuilder::createProperty("Pattern")
+        ->withDescription("A regular expression to match at the start or end of messages.")
+        ->isRequired(true)->build());
+
+const core::Property DefragmentText::PatternLoc(
+    core::PropertyBuilder::createProperty("Pattern Location")->withDescription("Whether the pattern is located at the start or at the end of the messages.")
+        ->withAllowableValues(PatternLocation::values())
+        ->withDefaultValue(toString(PatternLocation::START_OF_MESSAGE))->build());
+
+
+const core::Property DefragmentText::MaxBufferSize(
+    core::PropertyBuilder::createProperty("Max Buffer Size")
+        ->withDescription("The maximum buffer size, if the buffer exceeds this, it will be transferred to failure. Expected format is <size> <data unit>")
+        ->withType(core::StandardValidators::get().DATA_SIZE_VALIDATOR)->build());
+
+const core::Property DefragmentText::MaxBufferAge(
+    core::PropertyBuilder::createProperty("Max Buffer Age")->
+        withDescription("The maximum age of a buffer after which the buffer will be transferred to failure. Expected format is <duration> <time unit>")->build());
+
+void DefragmentText::initialize() {
+  setSupportedRelationships({Success, Failure});
+  setSupportedProperties({Pattern, PatternLoc, MaxBufferAge, MaxBufferSize});
+}
+
+void DefragmentText::onSchedule(core::ProcessContext* context, core::ProcessSessionFactory*) {
+  gsl_Expects(context);
+
+  std::string max_buffer_age_str;
+  if (context->getProperty(MaxBufferAge.getName(), max_buffer_age_str)) {
+    core::TimeUnit unit;
+    uint64_t max_buffer_age;
+    if (core::Property::StringToTime(max_buffer_age_str, max_buffer_age, unit) && core::Property::ConvertTimeUnitToMS(max_buffer_age, unit, max_buffer_age)) {
+      buffer_.setMaxAge(std::chrono::milliseconds(max_buffer_age));
+      logger_->log_trace("The Buffer maximum age is configured to be %" PRIu64 " ms", max_buffer_age);
+    }
+  }
+
+  std::string max_buffer_size_str;
+  if (context->getProperty(MaxBufferSize.getName(), max_buffer_size_str)) {
+    uint64_t max_buffer_size = core::DataSizeValue(max_buffer_size_str).getValue();
+    if (max_buffer_size > 0) {
+      buffer_.setMaxSize(max_buffer_size);
+      logger_->log_trace("The Buffer maximum size is configured to be %" PRIu64 " B", max_buffer_size);
+    }
+  }
+
+  context->getProperty(PatternLoc.getName(), pattern_location_);
+
+  std::string pattern_str;
+  if (context->getProperty(Pattern.getName(), pattern_str) && !pattern_str.empty()) {
+    pattern_ = std::regex(pattern_str);
+    logger_->log_trace("The Pattern is configured to be %s", pattern_str);
+  } else {
+    throw Exception(PROCESS_SCHEDULE_EXCEPTION, "Pattern property missing or invalid");
+  }
+}
+
+void DefragmentText::onTrigger(core::ProcessContext*, core::ProcessSession* session) {
+  gsl_Expects(session);
+  auto flowFiles = flow_file_store_.getNewFlowFiles();
+  for (auto& file : flowFiles) {
+    processNextFragment(session, file);
+  }
+  std::shared_ptr<core::FlowFile> original_flow_file = session->get();
+  processNextFragment(session, original_flow_file);
+  if (buffer_.maxAgeReached() || buffer_.maxSizeReached()) {
+    buffer_.flushAndReplace(session, Failure, nullptr);
+  }
+}
+
+void DefragmentText::processNextFragment(core::ProcessSession *session, const std::shared_ptr<core::FlowFile>& next_fragment) {
+  if (!next_fragment)
+    return;
+  if (!buffer_.isCompatible(next_fragment)) {
+    buffer_.flushAndReplace(session, Failure, nullptr);
+    session->transfer(next_fragment, Failure);
+    return;
+  }
+  std::shared_ptr<core::FlowFile> split_before_last_pattern;
+  std::shared_ptr<core::FlowFile> split_after_last_pattern;
+  bool found_pattern = splitFlowFileAtLastPattern(session, next_fragment, split_before_last_pattern,
+                                                  split_after_last_pattern);
+  buffer_.append(session, split_before_last_pattern);
+  if (found_pattern) {
+    buffer_.flushAndReplace(session, Success, split_after_last_pattern);
+  }
+  session->remove(next_fragment);
+}
+
+
+void DefragmentText::updateAttributesForSplitFiles(const std::shared_ptr<const core::FlowFile> &original_flow_file,
+                                                   const std::shared_ptr<core::FlowFile> &split_before_last_pattern,
+                                                   const std::shared_ptr<core::FlowFile> &split_after_last_pattern,
+                                                   const size_t split_position) const {
+  std::string base_name, post_name, offset_str;
+  if (!original_flow_file->getAttribute(textfragmentutils::BASE_NAME_ATTRIBUTE, base_name))
+    return;
+  if (!original_flow_file->getAttribute(textfragmentutils::POST_NAME_ATTRIBUTE, post_name))
+    return;
+  if (!original_flow_file->getAttribute(textfragmentutils::OFFSET_ATTRIBUTE, offset_str))
+    return;
+
+  size_t fragment_offset = std::stoi(offset_str);
+
+  if (split_before_last_pattern) {
+    std::string first_part_name = textfragmentutils::createFileName(base_name, post_name, fragment_offset, split_before_last_pattern->getSize());
+    split_before_last_pattern->setAttribute(core::SpecialFlowAttribute::FILENAME, first_part_name);
+  }
+  if (split_after_last_pattern) {
+    std::string second_part_name = textfragmentutils::createFileName(base_name, post_name, fragment_offset + split_position, split_after_last_pattern->getSize());
+    split_after_last_pattern->setAttribute(core::SpecialFlowAttribute::FILENAME, second_part_name);
+    split_after_last_pattern->setAttribute(textfragmentutils::OFFSET_ATTRIBUTE, std::to_string(fragment_offset + split_position));
+  }
+}
+
+namespace {
+class AppendFlowFileToFlowFile : public OutputStreamCallback {
+ public:
+  explicit AppendFlowFileToFlowFile(const std::shared_ptr<core::FlowFile>& flow_file_to_append, PayloadSerializer& serializer)
+      : flow_file_to_append_(flow_file_to_append), serializer_(serializer) {}
+
+  int64_t process(const std::shared_ptr<io::BaseStream> &stream) override {
+    return serializer_.serialize(flow_file_to_append_, stream);
+  }
+ private:
+  const std::shared_ptr<core::FlowFile>& flow_file_to_append_;
+  PayloadSerializer& serializer_;
+};
+
+void updateAppendedAttributes(const std::shared_ptr<core::FlowFile>& buffered_ff) {
+  std::string base_name, post_name, offset_str;
+  if (!buffered_ff->getAttribute(textfragmentutils::BASE_NAME_ATTRIBUTE, base_name))
+    return;
+  if (!buffered_ff->getAttribute(textfragmentutils::POST_NAME_ATTRIBUTE, post_name))
+    return;
+  if (!buffered_ff->getAttribute(textfragmentutils::OFFSET_ATTRIBUTE, offset_str))
+    return;
+  size_t fragment_offset = std::stoi(offset_str);
+
+  std::string buffer_new_name = textfragmentutils::createFileName(base_name, post_name, fragment_offset, buffered_ff->getSize());
+  buffered_ff->setAttribute(core::SpecialFlowAttribute::FILENAME, buffer_new_name);
+}
+
+class LastPatternFinder : public InputStreamCallback {

Review comment:
       `LastPatternFinder` does 3 things, and I think it would be better to separate them:
   1. reads the flow file into a string -- this could be done with existing callback types
   2. searches for the last occurrence of a pattern in the string -- this could be a reusable utility function in `StringUtils`
   3. computes either `match.position()` or `match.position() + match.length()`, depending on the value of `pattern_location_` -- this logic could be inlined into `splitFlowFileAtLastPattern()`.
   
   This is just a suggestion, feel free to ignore it, but I think these changes would improve the structure of the code.




-- 
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: issues-unsubscribe@nifi.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org