Last active
January 23, 2017 14:11
-
-
Save j-coll/cb523ba937a9efb47f9a4f4acd23f82f to your computer and use it in GitHub Desktop.
Repair opencga-storage-mongodb variants database due to opencb/opencga#504
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
// Generic migration utility method | |
function migrateCollection(collection, query, projection, migrateFunc) { | |
var bulk = db.getCollection(collection).initializeOrderedBulkOp(); | |
var count = 0; | |
var bulkSize = 500; | |
db.getCollection(collection).find(query,projection).forEach(function(doc) { | |
migrateFunc(bulk, doc); | |
if ( bulk.nUpdateOps > bulkSize ) { | |
count += bulk.nUpdateOps; | |
print("Execute bulk! " + count); | |
bulk.execute(); | |
bulk = db.getCollection(collection).initializeOrderedBulkOp(); | |
} | |
}); | |
if ( bulk.nUpdateOps > 0 ) { | |
count += bulk.nUpdateOps; | |
print("Execute bulk! " + count); | |
bulk.execute(); | |
bulk = db.getCollection(collection).initializeOrderedBulkOp(); | |
} | |
if (count == 0) { | |
print("Nothing to do!"); | |
} | |
} | |
// Remove extra secondary alternate and fix type on variants with reference.length == 50 | |
migrateCollection("variants", {reference:{$regex:"^.{50}$"}, type:"SV"}, {stats:0, annotation:0}, function (bulk, doc) { | |
var ref = doc.reference; | |
var id = doc._id; | |
print("Updating variant: _id : '" + id + "'"); | |
for (var i in doc.studies) { | |
var study = doc.studies[i]; | |
bulk.find({_id: id, "studies.sid": study.sid}).update({$set: {type: "INDEL"}, $pull: {"studies.$.alts": {ref: ref}}}); | |
} | |
}); | |
// Change type of SV secondary alternates with reference.length = 50 | |
migrateCollection("variants", {"studies.alts":{$elemMatch:{type:"SV", ref:{$regex:"^.{50}$"}}}}, {stats:0, annotation:0}, function (bulk, doc) { | |
var id = doc._id; | |
print("Updating variant: _id : '" + id + "'"); | |
for (var i in doc.studies) { | |
var study = doc.studies[i]; | |
if (study.hasOwnProperty("alts")) { | |
var alts = study.alts; | |
var modified = false; | |
for (var i2 in alts) { | |
var alt = alts[i2]; | |
if (alt.ref.length == 50 && alt.type == "SV") { | |
alt.type = "INDEL"; | |
modified = true; | |
} | |
} | |
if (modified) { | |
bulk.find({_id: id, "studies.sid": study.sid}).update({$set: {"studies.$.alts": alts}}); | |
} | |
} | |
} | |
}); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment