You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@druid.apache.org by "vogievetsky (via GitHub)" <gi...@apache.org> on 2023/08/08 22:44:36 UTC
[PR] Web console: adding format notice for CSV and TSV (druid)
vogievetsky opened a new pull request, #14783:
URL: https://github.com/apache/druid/pull/14783
Adds a warning that the CSV or TSV sample contains newlines in values which is allowed in the format but Druid will not accept it in the expected way. Adding this because I got really tripped up by this myself.
<img width="1792" alt="image" src="https://github.com/apache/druid/assets/177816/1bd4bc68-079c-473e-b43a-a561fe78abb6">
Try it with:
```csv
"a","b","c","d","e"
"hello
world",2,3,4,5
```
Feedback on copy very welcome.
Also changed the links from http://ndjson.org/ to https://jsonlines.org (due to https://github.com/ndjson/ndjson-spec/issues/35#issuecomment-1285673417)
--
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: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org
Re: [PR] Web console: adding format notice for CSV and TSV (druid)
Posted by "317brian (via GitHub)" <gi...@apache.org>.
317brian commented on code in PR #14783:
URL: https://github.com/apache/druid/pull/14783#discussion_r1287762585
##########
web-console/src/druid-models/ingestion-spec/ingestion-spec.tsx:
##########
@@ -2135,33 +2136,53 @@ export function updateIngestionType(
return newSpec;
}
+function findValueWithNewline(rows: string[][]): string | undefined {
+ return findMap(rows, row => findMap(row, value => (value.includes('\n') ? value : undefined)));
+}
+
export function issueWithSampleData(
- sampleData: SampleResponse,
- spec: Partial<IngestionSpec>,
+ sampleLines: string[],
+ isStreaming: boolean,
): JSX.Element | undefined {
- if (isStreamingSpec(spec)) return;
+ if (!sampleLines.length) return;
- const firstData: string = findMap(sampleData.data, l => l.input?.raw);
- if (firstData) return;
+ const firstLine = sampleLines[0];
+ if (!isStreaming) {
+ if (firstLine === '{') {
+ return (
+ <>
+ This data looks like multi-line formatted JSON object. For Druid to parse a text file it
+ must have one row per event. Consider reformatting your data as{' '}
+ <ExternalLink href="https://jsonlines.org">JSON Lines</ExternalLink>.
+ </>
+ );
+ }
- if (firstData === '{') {
- return (
- <>
- This data looks like multi-line formatted JSON object. For Druid to parse a text file it
- must have one row per event. Consider reformatting your data as{' '}
- <ExternalLink href="http://ndjson.org/">newline delimited JSON</ExternalLink>.
- </>
- );
+ if (oneOf(firstLine, '[', '[]')) {
+ return (
+ <>
+ This data looks like a multi-line JSON array. For Druid to parse a text file it must have
Review Comment:
```suggestion
This data looks like a multi-line JSON array. For Druid to parse a text file, it must have
```
##########
web-console/src/druid-models/ingestion-spec/ingestion-spec.tsx:
##########
@@ -2135,33 +2136,53 @@ export function updateIngestionType(
return newSpec;
}
+function findValueWithNewline(rows: string[][]): string | undefined {
+ return findMap(rows, row => findMap(row, value => (value.includes('\n') ? value : undefined)));
+}
+
export function issueWithSampleData(
- sampleData: SampleResponse,
- spec: Partial<IngestionSpec>,
+ sampleLines: string[],
+ isStreaming: boolean,
): JSX.Element | undefined {
- if (isStreamingSpec(spec)) return;
+ if (!sampleLines.length) return;
- const firstData: string = findMap(sampleData.data, l => l.input?.raw);
- if (firstData) return;
+ const firstLine = sampleLines[0];
+ if (!isStreaming) {
+ if (firstLine === '{') {
+ return (
+ <>
+ This data looks like multi-line formatted JSON object. For Druid to parse a text file it
+ must have one row per event. Consider reformatting your data as{' '}
+ <ExternalLink href="https://jsonlines.org">JSON Lines</ExternalLink>.
+ </>
+ );
+ }
- if (firstData === '{') {
- return (
- <>
- This data looks like multi-line formatted JSON object. For Druid to parse a text file it
- must have one row per event. Consider reformatting your data as{' '}
- <ExternalLink href="http://ndjson.org/">newline delimited JSON</ExternalLink>.
- </>
- );
+ if (oneOf(firstLine, '[', '[]')) {
+ return (
+ <>
+ This data looks like a multi-line JSON array. For Druid to parse a text file it must have
+ one row per event. Consider reformatting your data as{' '}
+ <ExternalLink href="https://jsonlines.org">JSON Lines</ExternalLink>.
+ </>
+ );
+ }
}
- if (oneOf(firstData, '[', '[]')) {
- return (
- <>
- This data looks like a multi-line JSON array. For Druid to parse a text file it must have
- one row per event. Consider reformatting your data as{' '}
- <ExternalLink href="http://ndjson.org/">newline delimited JSON</ExternalLink>.
- </>
+ const format = guessSimpleInputFormat(sampleLines, isStreaming);
+ const text = sampleLines.join('\n');
+ if (oneOf(format.type, 'csv', 'tsv')) {
+ const valueWithNewline = findValueWithNewline(
+ format.type === 'csv' ? csvParseRows(text) : tsvParseRows(text),
);
+ if (valueWithNewline) {
+ const formatLabel = format.type.toUpperCase();
+ return (
+ <>
+ {`This ${formatLabel} data has values that contain new lines. Druid requires ${formatLabel} files to have one event per line and thus ${formatLabel} values can not contain new lines. Consider encoding new lines in the values of your ${formatLabel} with some special delimiter.`}
Review Comment:
```suggestion
{`This ${formatLabel} data has values that contain new lines. Druid requires ${formatLabel} files to have one event per line, so ${formatLabel} values can not contain new lines. Consider encoding new lines in the values of your ${formatLabel} with some special delimiter.`}
```
##########
web-console/src/druid-models/ingestion-spec/ingestion-spec.tsx:
##########
@@ -2135,33 +2136,53 @@ export function updateIngestionType(
return newSpec;
}
+function findValueWithNewline(rows: string[][]): string | undefined {
+ return findMap(rows, row => findMap(row, value => (value.includes('\n') ? value : undefined)));
+}
+
export function issueWithSampleData(
- sampleData: SampleResponse,
- spec: Partial<IngestionSpec>,
+ sampleLines: string[],
+ isStreaming: boolean,
): JSX.Element | undefined {
- if (isStreamingSpec(spec)) return;
+ if (!sampleLines.length) return;
- const firstData: string = findMap(sampleData.data, l => l.input?.raw);
- if (firstData) return;
+ const firstLine = sampleLines[0];
+ if (!isStreaming) {
+ if (firstLine === '{') {
+ return (
+ <>
+ This data looks like multi-line formatted JSON object. For Druid to parse a text file it
Review Comment:
```suggestion
This data looks like multi-line formatted JSON object. For Druid to parse a text file, it
```
--
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: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org
Re: [PR] Web console: adding format notice for CSV and TSV (druid)
Posted by "ektravel (via GitHub)" <gi...@apache.org>.
ektravel commented on code in PR #14783:
URL: https://github.com/apache/druid/pull/14783#discussion_r1287773134
##########
web-console/src/druid-models/ingestion-spec/ingestion-spec.tsx:
##########
@@ -2135,33 +2136,53 @@ export function updateIngestionType(
return newSpec;
}
+function findValueWithNewline(rows: string[][]): string | undefined {
+ return findMap(rows, row => findMap(row, value => (value.includes('\n') ? value : undefined)));
+}
+
export function issueWithSampleData(
- sampleData: SampleResponse,
- spec: Partial<IngestionSpec>,
+ sampleLines: string[],
+ isStreaming: boolean,
): JSX.Element | undefined {
- if (isStreamingSpec(spec)) return;
+ if (!sampleLines.length) return;
- const firstData: string = findMap(sampleData.data, l => l.input?.raw);
- if (firstData) return;
+ const firstLine = sampleLines[0];
+ if (!isStreaming) {
+ if (firstLine === '{') {
+ return (
+ <>
+ This data looks like multi-line formatted JSON object. For Druid to parse a text file it
Review Comment:
```suggestion
This data looks like a multi-line formatted JSON object. For Druid to parse a text file it
```
--
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: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org
Re: [PR] Web console: adding format notice for CSV and TSV (druid)
Posted by "vogievetsky (via GitHub)" <gi...@apache.org>.
vogievetsky merged PR #14783:
URL: https://github.com/apache/druid/pull/14783
--
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: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org
Re: [PR] Web console: adding format notice for CSV and TSV (druid)
Posted by "ektravel (via GitHub)" <gi...@apache.org>.
ektravel commented on code in PR #14783:
URL: https://github.com/apache/druid/pull/14783#discussion_r1287773480
##########
web-console/src/druid-models/ingestion-spec/ingestion-spec.tsx:
##########
@@ -2135,33 +2136,53 @@ export function updateIngestionType(
return newSpec;
}
+function findValueWithNewline(rows: string[][]): string | undefined {
+ return findMap(rows, row => findMap(row, value => (value.includes('\n') ? value : undefined)));
+}
+
export function issueWithSampleData(
- sampleData: SampleResponse,
- spec: Partial<IngestionSpec>,
+ sampleLines: string[],
+ isStreaming: boolean,
): JSX.Element | undefined {
- if (isStreamingSpec(spec)) return;
+ if (!sampleLines.length) return;
- const firstData: string = findMap(sampleData.data, l => l.input?.raw);
- if (firstData) return;
+ const firstLine = sampleLines[0];
+ if (!isStreaming) {
+ if (firstLine === '{') {
+ return (
+ <>
+ This data looks like multi-line formatted JSON object. For Druid to parse a text file it
+ must have one row per event. Consider reformatting your data as{' '}
+ <ExternalLink href="https://jsonlines.org">JSON Lines</ExternalLink>.
+ </>
+ );
+ }
- if (firstData === '{') {
- return (
- <>
- This data looks like multi-line formatted JSON object. For Druid to parse a text file it
- must have one row per event. Consider reformatting your data as{' '}
- <ExternalLink href="http://ndjson.org/">newline delimited JSON</ExternalLink>.
- </>
- );
+ if (oneOf(firstLine, '[', '[]')) {
+ return (
+ <>
+ This data looks like a multi-line JSON array. For Druid to parse a text file it must have
+ one row per event. Consider reformatting your data as{' '}
+ <ExternalLink href="https://jsonlines.org">JSON Lines</ExternalLink>.
+ </>
+ );
+ }
}
- if (oneOf(firstData, '[', '[]')) {
- return (
- <>
- This data looks like a multi-line JSON array. For Druid to parse a text file it must have
- one row per event. Consider reformatting your data as{' '}
- <ExternalLink href="http://ndjson.org/">newline delimited JSON</ExternalLink>.
- </>
+ const format = guessSimpleInputFormat(sampleLines, isStreaming);
+ const text = sampleLines.join('\n');
+ if (oneOf(format.type, 'csv', 'tsv')) {
+ const valueWithNewline = findValueWithNewline(
+ format.type === 'csv' ? csvParseRows(text) : tsvParseRows(text),
);
+ if (valueWithNewline) {
+ const formatLabel = format.type.toUpperCase();
+ return (
+ <>
+ {`This ${formatLabel} data has values that contain new lines. Druid requires ${formatLabel} files to have one event per line and thus ${formatLabel} values can not contain new lines. Consider encoding new lines in the values of your ${formatLabel} with some special delimiter.`}
Review Comment:
```suggestion
{`This ${formatLabel} data has values that contain new lines. Druid requires ${formatLabel} files to have one event per line, so ${formatLabel} values cannot contain new lines. Consider encoding new lines in the values of your ${formatLabel} with some special delimiter.`}
```
--
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: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org