This repository was archived by the owner on Mar 4, 2026. It is now read-only.
This repository was archived by the owner on Mar 4, 2026. It is now read-only.
runStream returning incomplete and out of order rows #180
Description
Environment details
- OS: Ubuntu 16.04.4 LTS
- Node.js version: 6.14.1
- npm version: 5.8.0
@google-cloud/spannerversion: 1.4.1
Steps to reproduce
I have a table with this structure:
CREATE TABLE accounts (
recordId STRING(36) NOT NULL,
account_CID STRING(255),
account_created_on STRING(1024),
account_created_on_unixTimestamp FLOAT64,
account_dashboard_tabs STRING(255),
account_ok_tabs STRING(255),
account_summary_content_tabs STRING(255),
account_to_creator STRING(36),
account_to_customer_admin STRING(36),
account_view_tabs STRING(255),
Address STRING(1024),
Address1 STRING(1024),
Address1_city STRING(255),
Address1_postalcode STRING(255),
Address1_provinceterritory STRING(255),
Address_city STRING(255),
Address_postalcode STRING(255),
Address_provinceterritory STRING(255),
current_document_number FLOAT64,
customer_content_tabs STRING(255),
Email STRING(255),
fund_to_account STRING(36),
FundsTab_AccountSubNavigation STRING(255),
Number FLOAT64,
Number1 FLOAT64,
Number_of_Payments FLOAT64,
payment_plan_to_account STRING(36),
Phone_Number STRING(1024),
Phone_Number1 STRING(1024),
requested_payment_plan_to_account STRING(36),
RequestRelationshipCID STRING(255),
RequestRelationshipContentTab STRING(255),
Selection_Field___Static STRING(MAX),
Short_Text STRING(255),
Short_Text1 STRING(255),
Short_Text4 STRING(255),
upload_customer_csv STRING(1024),
upload_customer_csv_name STRING(255),
user_to_accounts STRING(36),
) PRIMARY KEY (recordId)
When I run the following query:
SELECT `root`.`account_created_on` as `field0`, `root`.`Short_Text` as `field1`, `root`.`Short_Text1` as `field2`, `root`.`account_CID` as `field3`, `root`.`recordId` as `recordId` FROM `accounts` AS root
My table has 17349 records
Occasionally (1 in 5) when I run this query using:
database.runStream(query)
.on('error', function (err) {
console.error(err);
return reject(err);
})
.on('data', function (row) {
let rowJSON = row.toJSON();
if (!rowJSON.recordId || rowJSON.recordId.match(/^\d+$/)) {
console.log('Worker '+process.pid+' Row: '+JSON.stringify(rowJSON));
console.log(row);
}
records[rowJSON.recordId] = rowJSON;
})
.on('end', function () {
return resolve(records);
});I will get rows that are missing the recordId and the values will be the wrong labels for example field1 will contain the data for field2