fix(codecs) csv encoding quoting bug

[scMarkus]

Closes #17261 and fixes a csv quoting bug where a closing double quote was not written in the last field if needed.

This extends #18149

@pront could you pleas have a look at this one? I opted to create a new branch since this exceeds the initial scope of adding csv configuration options.
What I tried here is implementing the suggestion of my issue in the rust-csv repository. My primary intention was to see if I can make this work. I am still willing to expand on #18149 if you deem this more reasonable.

[netlify]

✅ Deploy Preview for vrl-playground ready!

Name	Link
🔨 Latest commit	75a55fd
🔍 Latest deploy log	https://app.netlify.com/sites/vrl-playground/deploys/64ec5ec1c4a65000081c0859
😎 Deploy Preview	https://deploy-preview-18320--vrl-playground.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[netlify]

✅ Deploy Preview for vector-project ready!

Name	Link
🔨 Latest commit	75a55fd
🔍 Latest deploy log	https://app.netlify.com/sites/vector-project/deploys/64ec5ec1aa074e000867c636
😎 Deploy Preview	https://deploy-preview-18320--vector-project.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[pront]

Hello @scMarkus, is this meant to replace #18149? I will take a look shortly.

[scMarkus]

@pront yes, this implements #18149 and additionally fixes the bug. When you looking into the changes I would kindly ask you to

1. look into how I write to buffer since I have not implemented the more elaborated flushing logic present in rust-csv (mostly because of not grasping the concept) found here:
   https://github.com/BurntSushi/rust-csv/blob/574ae1ff64693b42ae0ce153926d9a0a5d546936/src/writer.rs#L1054-L1069

2. I think I am allocating the internal buffer for each call to encode() at
   https://github.com/vectordotdev/vector/blob/19fab882ecfcf716c70467d2e81795e65a3cc595/lib/codecs/src/encoding/format/csv.rs#L218C1-L218C58
   I was trying to make this part of the CsvSerializer struct but I was not able to convince the borrow checker since I would mutably borrow inside the loop iterating the fields.

[pront]

Sounds good. Should we close #18149?

[scMarkus]

yes, I just did it

[scMarkus]

I think I have found a solution to 2. of my earlier post in 2cf48fb

@pront yes, this implements #18149 and additionally fixes the bug. When you looking into the changes I would kindly ask you to

1. look into how I write to `buffer` since I have not implemented the more elaborated flushing logic present in rust-csv (mostly because of not grasping the concept) found here:
   https://github.com/BurntSushi/rust-csv/blob/574ae1ff64693b42ae0ce153926d9a0a5d546936/src/writer.rs#L1054-L1069

2. I think I am allocating the internal buffer for each call to `encode()` at
   https://github.com/vectordotdev/vector/blob/19fab882ecfcf716c70467d2e81795e65a3cc595/lib/codecs/src/encoding/format/csv.rs#L218C1-L218C58
   I was trying to make this part of the `CsvSerializer` struct but I was not able to convince the borrow checker since I would mutably borrow inside the loop iterating the fields.

[pront]

I think I have found a solution to 2. of my earlier post in 2cf48fb

Nice, I will take a look when you push a new commit.

[scMarkus]

@pront what I wanted to say is that in the referred commit my idea is implemented already. So pleas feel free to take a look at any time. I am finished with my changes :)

[pront]

Thank you for this contribution @scMarkus!

Since this PR introduces a lot of low level details in encode() and there are a lot of moving parts, I would like to strengthen the tests as much as possible.

Regarding the points you brought up, we can discuss further later. Let's focus on correctness for now.

[pront]

This is starting to look great, almost there.

[scMarkus]

@pront thanks for all your time so far. I have just pushed my latest state. I would kindly ask you to look over it again and point me towards any more to dos you may come across.

[pront]

Thank you @scMarkus, LGTM!

[scMarkus]

Question: would I be needed to change the docs to fix the test suits checks? My naive guess thus far has been by adding the respective comments to the new options in the code base the documentation would magically be handled somehow?

[pront]

Question: would I be needed to change the docs to fix the test suits checks? My naive guess thus far has been by adding the respective comments to the new options in the code base the documentation would magically be handled somehow?

Here are some instructions:

brew install cue
cd path/to/vector/repo/root
git rebase origin/master
make generate-component-docs
find . -name '*.cue' -exec cue fmt {} \; # for macOS
make check-docs

If challenges come up, I can do this locally for you. However, I would also need perms to push to your remote branch. Let me know if you have any questions.

[scMarkus]

@pront thanks for your hints. I managed to create the documentation in a container. hopefully this works now.

[github-actions]

Regression Detector Results

Run ID: 5a70a818-8192-48ae-92f5-46eb943b3bd9
Baseline: fe431c3
Comparison: c83be0d
Total vector CPUs: 7

Explanation

A regression test is an integrated performance test for vector in a repeatable rig, with varying configuration for vector. What follows is a statistical summary of a brief vector run for each configuration across SHAs given above. The goal of these tests are to determine quickly if vector performance is changed and to what degree by a pull request.

Because a target's optimization goal performance in each experiment will vary somewhat each time it is run, we can only estimate mean differences in optimization goal relative to the baseline target. We express these differences as a percentage change relative to the baseline target, denoted "Δ mean %". These estimates are made to a precision that balances accuracy and cost control. We represent this precision as a 90.00% confidence interval denoted "Δ mean % CI": there is a 90.00% chance that the true value of "Δ mean %" is in that interval.

We decide whether a change in performance is a "regression" -- a change worth investigating further -- if both of the following two criteria are true:

1. The estimated |Δ mean %| ≥ 5.00%. This criterion intends to answer the question "Does the estimated change in mean optimization goal performance have a meaningful impact on your customers?". We assume that when |Δ mean %| < 5.00%, the impact on your customers is not meaningful. We also assume that a performance change in optimization goal is worth investigating whether it is an increase or decrease, so long as the magnitude of the change is sufficiently large.

2. Zero is not in the 90.00% confidence interval "Δ mean % CI" about "Δ mean %". This statement is equivalent to saying that there is at least a 90.00% chance that the mean difference in optimization goal is not zero. This criterion intends to answer the question, "Is there a statistically significant difference in mean optimization goal performance?". It also means there is no more than a 10.00% chance this criterion reports a statistically significant difference when the true difference in mean optimization goal is zero -- a "false positive". We assume you are willing to accept a 10.00% chance of inaccurately detecting a change in performance when no true difference exists.

The table below, if present, lists those experiments that have experienced a statistically significant change in mean optimization goal performance between baseline and comparison SHAs with 90.00% confidence OR have been detected as newly erratic. Negative values of "Δ mean %" mean that baseline is faster, whereas positive values of "Δ mean %" mean that comparison is faster. Results that do not exhibit more than a ±5.00% change in their mean optimization goal are discarded. An experiment is erratic if its coefficient of variation is greater than 0.1. The abbreviated table will be omitted if no interesting change is observed.

No interesting changes in experiment optimization goals with confidence ≥ 90.00% and |Δ mean %| ≥ 5.00%.

Fine details of change detection per experiment.

experiment	goal	Δ mean %	Δ mean % CI	confidence
syslog_log2metric_humio_metrics	ingress throughput	+2.31	[+2.20, +2.42]	100.00%
socket_to_socket_blackhole	ingress throughput	+1.88	[+1.85, +1.91]	100.00%
http_to_http_acks	ingress throughput	+1.09	[+0.96, +1.21]	100.00%
otlp_http_to_blackhole	ingress throughput	+0.57	[+0.41, +0.74]	100.00%
otlp_grpc_to_blackhole	ingress throughput	+0.24	[+0.17, +0.32]	100.00%
http_to_s3	ingress throughput	+0.05	[-0.15, +0.25]	25.05%
splunk_hec_indexer_ack_blackhole	ingress throughput	+0.01	[-0.11, +0.14]	8.50%
http_to_http_json	ingress throughput	+0.00	[-0.04, +0.04]	0.48%
splunk_hec_to_splunk_hec_logs_noack	ingress throughput	-0.00	[-0.11, +0.10]	3.74%
splunk_hec_to_splunk_hec_logs_acks	ingress throughput	-0.02	[-0.13, +0.10]	16.37%
enterprise_http_to_http	ingress throughput	-0.14	[-0.26, -0.02]	85.81%
http_to_http_noack	ingress throughput	-0.23	[-0.37, -0.10]	97.58%
file_to_blackhole	egress throughput	-0.32	[-1.19, +0.55]	36.69%
datadog_agent_remap_blackhole_acks	ingress throughput	-0.32	[-0.40, -0.24]	100.00%
fluent_elasticsearch	ingress throughput	-0.35	[-0.59, -0.12]	94.52%
splunk_hec_route_s3	ingress throughput	-0.44	[-0.84, -0.03]	83.38%
datadog_agent_remap_datadog_logs_acks	ingress throughput	-0.71	[-0.83, -0.60]	100.00%
syslog_log2metric_splunk_hec_metrics	ingress throughput	-1.11	[-1.23, -0.99]	100.00%
datadog_agent_remap_blackhole	ingress throughput	-1.38	[-1.47, -1.30]	100.00%
datadog_agent_remap_datadog_logs	ingress throughput	-1.59	[-1.70, -1.48]	100.00%
syslog_splunk_hec_logs	ingress throughput	-1.64	[-1.70, -1.57]	100.00%
syslog_regex_logs2metric_ddmetrics	ingress throughput	-2.34	[-2.56, -2.12]	100.00%
syslog_loki	ingress throughput	-2.76	[-2.82, -2.69]	100.00%
http_text_to_http_json	ingress throughput	-3.06	[-3.19, -2.93]	100.00%
syslog_humio_logs	ingress throughput	-3.68	[-3.76, -3.59]	100.00%

[github-actions]

Regression Detector Results

Run ID: 745c9f84-d182-4dab-9b5d-ebc68aa33594
Baseline: e39d9b3
Comparison: 93ad80d
Total vector CPUs: 7

Explanation

A regression test is an integrated performance test for vector in a repeatable rig, with varying configuration for vector. What follows is a statistical summary of a brief vector run for each configuration across SHAs given above. The goal of these tests are to determine quickly if vector performance is changed and to what degree by a pull request.

Because a target's optimization goal performance in each experiment will vary somewhat each time it is run, we can only estimate mean differences in optimization goal relative to the baseline target. We express these differences as a percentage change relative to the baseline target, denoted "Δ mean %". These estimates are made to a precision that balances accuracy and cost control. We represent this precision as a 90.00% confidence interval denoted "Δ mean % CI": there is a 90.00% chance that the true value of "Δ mean %" is in that interval.

We decide whether a change in performance is a "regression" -- a change worth investigating further -- if both of the following two criteria are true:

1. The estimated |Δ mean %| ≥ 5.00%. This criterion intends to answer the question "Does the estimated change in mean optimization goal performance have a meaningful impact on your customers?". We assume that when |Δ mean %| < 5.00%, the impact on your customers is not meaningful. We also assume that a performance change in optimization goal is worth investigating whether it is an increase or decrease, so long as the magnitude of the change is sufficiently large.

2. Zero is not in the 90.00% confidence interval "Δ mean % CI" about "Δ mean %". This statement is equivalent to saying that there is at least a 90.00% chance that the mean difference in optimization goal is not zero. This criterion intends to answer the question, "Is there a statistically significant difference in mean optimization goal performance?". It also means there is no more than a 10.00% chance this criterion reports a statistically significant difference when the true difference in mean optimization goal is zero -- a "false positive". We assume you are willing to accept a 10.00% chance of inaccurately detecting a change in performance when no true difference exists.

The table below, if present, lists those experiments that have experienced a statistically significant change in mean optimization goal performance between baseline and comparison SHAs with 90.00% confidence OR have been detected as newly erratic. Negative values of "Δ mean %" mean that baseline is faster, whereas positive values of "Δ mean %" mean that comparison is faster. Results that do not exhibit more than a ±5.00% change in their mean optimization goal are discarded. An experiment is erratic if its coefficient of variation is greater than 0.1. The abbreviated table will be omitted if no interesting change is observed.

No interesting changes in experiment optimization goals with confidence ≥ 90.00% and |Δ mean %| ≥ 5.00%.

Fine details of change detection per experiment.

experiment	goal	Δ mean %	Δ mean % CI	confidence
file_to_blackhole	egress throughput	+2.41	[+1.54, +3.27]	99.96%
datadog_agent_remap_datadog_logs	ingress throughput	+2.29	[+2.17, +2.41]	100.00%
socket_to_socket_blackhole	ingress throughput	+1.72	[+1.70, +1.74]	100.00%
splunk_hec_route_s3	ingress throughput	+1.56	[+1.15, +1.97]	100.00%
http_to_http_acks	ingress throughput	+1.43	[+1.31, +1.54]	100.00%
datadog_agent_remap_blackhole	ingress throughput	+0.98	[+0.90, +1.06]	100.00%
http_to_s3	ingress throughput	+0.97	[+0.76, +1.18]	100.00%
otlp_http_to_blackhole	ingress throughput	+0.78	[+0.60, +0.96]	100.00%
otlp_grpc_to_blackhole	ingress throughput	+0.58	[+0.49, +0.66]	100.00%
syslog_log2metric_humio_metrics	ingress throughput	+0.18	[+0.07, +0.28]	96.92%
http_text_to_http_json	ingress throughput	+0.04	[-0.05, +0.13]	45.64%
splunk_hec_indexer_ack_blackhole	ingress throughput	+0.01	[-0.12, +0.13]	4.60%
http_to_http_json	ingress throughput	-0.00	[-0.05, +0.04]	12.01%
splunk_hec_to_splunk_hec_logs_noack	ingress throughput	-0.01	[-0.11, +0.10]	7.69%
splunk_hec_to_splunk_hec_logs_acks	ingress throughput	-0.03	[-0.15, +0.09]	27.73%
http_to_http_noack	ingress throughput	-0.28	[-0.42, -0.15]	99.26%
enterprise_http_to_http	ingress throughput	-0.81	[-0.93, -0.69]	100.00%
fluent_elasticsearch	ingress throughput	-1.22	[-1.46, -0.98]	100.00%
datadog_agent_remap_blackhole_acks	ingress throughput	-1.73	[-1.81, -1.64]	100.00%
datadog_agent_remap_datadog_logs_acks	ingress throughput	-1.79	[-1.89, -1.69]	100.00%
syslog_regex_logs2metric_ddmetrics	ingress throughput	-1.97	[-2.12, -1.81]	100.00%
syslog_loki	ingress throughput	-3.46	[-3.51, -3.41]	100.00%
syslog_log2metric_splunk_hec_metrics	ingress throughput	-3.69	[-3.82, -3.57]	100.00%
syslog_humio_logs	ingress throughput	-3.78	[-3.87, -3.69]	100.00%
syslog_splunk_hec_logs	ingress throughput	-4.42	[-4.49, -4.35]	100.00%

[pront]

Thank you @scMarkus! Happy to see this work merged.

[scMarkus]

Hey many thanks from my side as well. looking forward to the next release :)