-
Notifications
You must be signed in to change notification settings - Fork 193
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix quoting of string columns in csv #1379
Changes from 4 commits
0c8390c
c10da7f
fe7eb07
aa270e5
a3a0513
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -341,31 +341,66 @@ export class QueryEvaluationInfo { | |
/** | ||
* Creates the CSV file containing the results of this query. This will only be called if the query | ||
* does not have interpreted results and the CSV file does not already exist. | ||
* | ||
* @return Promise<true> if the operation creates the file. Promise<false> if the operation does | ||
* not create the file. | ||
* | ||
* @throws Error if the operation fails. | ||
*/ | ||
async exportCsvResults(qs: qsClient.QueryServerClient, csvPath: string, onFinish: () => void): Promise<void> { | ||
async exportCsvResults(qs: qsClient.QueryServerClient, csvPath: string): Promise<boolean> { | ||
const resultSet = await this.chooseResultSet(qs); | ||
if (!resultSet) { | ||
void showAndLogWarningMessage('Query has no result set.'); | ||
return false; | ||
} | ||
let stopDecoding = false; | ||
const out = fs.createWriteStream(csvPath); | ||
out.on('finish', onFinish); | ||
out.on('error', () => { | ||
if (!stopDecoding) { | ||
stopDecoding = true; | ||
void showAndLogErrorMessage(`Failed to write CSV results to ${csvPath}`); | ||
} | ||
|
||
const promise: Promise<boolean> = new Promise((resolve, reject) => { | ||
out.on('finish', () => resolve(true)); | ||
out.on('error', () => { | ||
if (!stopDecoding) { | ||
stopDecoding = true; | ||
reject(new Error(`Failed to write CSV results to ${csvPath}`)); | ||
} | ||
}); | ||
}); | ||
|
||
let nextOffset: number | undefined = 0; | ||
while (nextOffset !== undefined && !stopDecoding) { | ||
const chunk: DecodedBqrsChunk = await qs.cliServer.bqrsDecode(this.resultsPaths.resultsPath, SELECT_QUERY_NAME, { | ||
do { | ||
const chunk: DecodedBqrsChunk = await qs.cliServer.bqrsDecode(this.resultsPaths.resultsPath, resultSet, { | ||
pageSize: 100, | ||
offset: nextOffset, | ||
}); | ||
for (const tuple of chunk.tuples) { | ||
out.write(tuple.join(',') + '\n'); | ||
} | ||
const quotes = chunk.columns.map(col => col.kind === 'String' ? '"' : ''); | ||
chunk.tuples.forEach((tuple) => { | ||
out.write(tuple.map((v, i) => { | ||
return `${quotes[i]}${v}${quotes[i]}`; | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I think we also need to be escaping any There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Good point. I'll do that. |
||
}).join(',') + '\n'); | ||
}); | ||
nextOffset = chunk.next; | ||
} | ||
} while (nextOffset && !stopDecoding); | ||
out.end(); | ||
|
||
return promise; | ||
} | ||
|
||
/** | ||
* Choose the name of the result set to run. If the `#select` set exists, use that. Otherwise, | ||
* arbitrarily choose the first set. Most of the time, this will be correct. | ||
* | ||
* If the query has no result sets, then return undefined. | ||
*/ | ||
async chooseResultSet(qs: qsClient.QueryServerClient) { | ||
const resultSets = (await qs.cliServer.bqrsInfo(this.resultsPaths.resultsPath, 0))['result-sets']; | ||
if (!resultSets.length) { | ||
return undefined; | ||
} | ||
if (resultSets.find(r => r.name === SELECT_QUERY_NAME)) { | ||
return SELECT_QUERY_NAME; | ||
} | ||
return resultSets[0].name; | ||
} | ||
/** | ||
* Returns the path to the CSV alerts interpretation of this query results. If CSV results have | ||
* not yet been produced, this will return first create the CSV results and then return the path. | ||
|
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,11 +1,16 @@ | ||
import { expect } from 'chai'; | ||
import * as path from 'path'; | ||
import * as fs from 'fs-extra'; | ||
import * as sinon from 'sinon'; | ||
import { Uri } from 'vscode'; | ||
|
||
import { QueryEvaluationInfo } from '../../run-queries'; | ||
import { Severity, compileQuery } from '../../pure/messages'; | ||
import * as config from '../../config'; | ||
import { tmpDir } from '../../helpers'; | ||
import { QueryServerClient } from '../../queryserver-client'; | ||
import { CodeQLCliServer } from '../../cli'; | ||
import { SELECT_QUERY_NAME } from '../../contextual/locationFinder'; | ||
|
||
describe('run-queries', () => { | ||
let sandbox: sinon.SinonSandbox; | ||
|
@@ -53,6 +58,51 @@ describe('run-queries', () => { | |
expect(info.canHaveInterpretedResults()).to.eq(true); | ||
}); | ||
|
||
[SELECT_QUERY_NAME, 'other'].forEach(resultSetName => { | ||
it(`should export csv results for result set ${resultSetName}`, async () => { | ||
const csvLocation = path.join(tmpDir.name, 'test.csv'); | ||
const qs = createMockQueryServerClient( | ||
createMockCliServer({ | ||
bqrsInfo: [{ 'result-sets': [{ name: resultSetName }, { name: 'hucairz' }] }], | ||
bqrsDecode: [{ | ||
columns: [{ kind: 'NotString' }, { kind: 'String' }], | ||
tuples: [['a', 'b'], ['c', 'd']], | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Let's expand the tuples here to include some quote characters and make sure those get escaped correctly. |
||
next: 1 | ||
}, { | ||
// just for fun, give a different set of columns here | ||
// this won't happen with the real CLI, but it's a good test | ||
columns: [{ kind: 'String' }, { kind: 'NotString' }, { kind: 'StillNotString' }], | ||
tuples: [['a', 'b', 'c']] | ||
}] | ||
}) | ||
); | ||
const info = createMockQueryInfo(); | ||
const promise = info.exportCsvResults(qs, csvLocation); | ||
|
||
const result = await promise; | ||
expect(result).to.eq(true); | ||
|
||
const csv = fs.readFileSync(csvLocation, 'utf8'); | ||
expect(csv).to.eq('a,"b"\nc,"d"\n"a",b,c\n'); | ||
|
||
// now verify that we are using the expected result set | ||
expect((qs.cliServer.bqrsDecode as sinon.SinonStub).callCount).to.eq(2); | ||
expect((qs.cliServer.bqrsDecode as sinon.SinonStub).getCall(0).args[1]).to.eq(resultSetName); | ||
}); | ||
}); | ||
|
||
it('should handle csv exports for a query with no result sets', async () => { | ||
const csvLocation = path.join(tmpDir.name, 'test.csv'); | ||
const qs = createMockQueryServerClient( | ||
createMockCliServer({ | ||
bqrsInfo: [{ 'result-sets': [] }] | ||
}) | ||
); | ||
const info = createMockQueryInfo(); | ||
const result = await info.exportCsvResults(qs, csvLocation); | ||
expect(result).to.eq(false); | ||
}); | ||
|
||
describe('compile', () => { | ||
it('should compile', async () => { | ||
const info = createMockQueryInfo(); | ||
|
@@ -116,7 +166,7 @@ describe('run-queries', () => { | |
); | ||
} | ||
|
||
function createMockQueryServerClient() { | ||
function createMockQueryServerClient(cliServer?: CodeQLCliServer): QueryServerClient { | ||
return { | ||
config: { | ||
timeoutSecs: 5 | ||
|
@@ -131,7 +181,20 @@ describe('run-queries', () => { | |
})), | ||
logger: { | ||
log: sandbox.spy() | ||
} | ||
}; | ||
}, | ||
cliServer | ||
} as unknown as QueryServerClient; | ||
} | ||
|
||
function createMockCliServer(mockOperations: Record<string, any[]>): CodeQLCliServer { | ||
const mockServer: Record<string, any> = {}; | ||
for (const [operation, returns] of Object.entries(mockOperations)) { | ||
mockServer[operation] = sandbox.stub(); | ||
returns.forEach((returnValue, i) => { | ||
mockServer[operation].onCall(i).resolves(returnValue); | ||
}); | ||
} | ||
|
||
return mockServer as unknown as CodeQLCliServer; | ||
} | ||
}); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should we be displaying an error in the
else
case here? I see wereject()
where we were previously usingshowAndLogErrorMessage()
but as far as I can tell we are then swallowing that error.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@aeisenberg I don't think you've addressed this comment. It's less blocking the other others, so I'm happy to approve without any changes here if you don't think any changes are needed, but just checking it wasn't missed accidentally.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Oh...sorry. I forgot to comment here. The error is not being swallowed. Take a look at
commandRunnerWithProgress
, you can see that there is an error barrier there that catches and logs all thrown and rejected errors. So, you don't ever need to explicitly catch and log an error when you're inside a command as long as that command is wrapped incommandRunnerWithProgress
.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
And, indeed, when I try this out, forcing an error inside of
exportCsvResults
, the error is logged appropriately.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Gotcha, thanks for the explanation 🙂