repo stringlengths 5 53 | pr_number int32 1 321k | task_type stringclasses 2
values | issue_text stringlengths 0 81.2k | pr_title stringlengths 1 319 | pr_body stringlengths 0 105k | base_sha stringlengths 40 40 | head_sha stringlengths 40 40 | gold_diff stringlengths 0 202M | changed_files listlengths 0 100 | review_threads listlengths 0 100 | test_patch stringlengths 0 23.4M | merged bool 1
class |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
twitter/rsc | 381 | issue_to_patch | Document a known issue in current ExtModClassRef resolution | 08aaeb8cf03ca2ba3822cde2b54af52155e03dc4 | 2d1f5f2aa0919c97b174b89928731966f484c46e | diff --git a/rsc/src/main/scala/rsc/classpath/scalacp/Symbols.scala b/rsc/src/main/scala/rsc/classpath/scalacp/Symbols.scala
index 8d3fe3a7..4c38380a 100644
--- a/rsc/src/main/scala/rsc/classpath/scalacp/Symbols.scala
+++ b/rsc/src/main/scala/rsc/classpath/scalacp/Symbols.scala
@@ -115,6 +115,7 @@ trait Symbols {
... | [
"rsc/src/main/scala/rsc/classpath/scalacp/Symbols.scala"
] | [] | true | |||
twitter/rsc | 378 | issue_to_patch | Correctly read ExtRef symbols (fix #377) | d58ebbb42ec1ce559193689bd2940af6eb28be58 | 67e7cc217ee0f2fd869ecd43f259bae54e3417ce | diff --git a/examples/semantic/src/main/scala/377.scala b/examples/semantic/src/main/scala/377.scala
new file mode 100644
index 00000000..88edac1c
--- /dev/null
+++ b/examples/semantic/src/main/scala/377.scala
@@ -0,0 +1,5 @@
+package ticket377
+
+class C {
+ def m: Map.Map1[Int, Int] = ???
+}
diff --git a/rsc/src/mai... | [
"examples/semantic/src/main/scala/377.scala",
"rsc/src/main/scala/rsc/classpath/scalacp/Symbols.scala"
] | [] | true | |||
twitter/scalding | 1,985 | issue_to_patch | Try sbt-publish-ci | it handles the GPG part nicely (should work with the exiting ENV variable). It might take some iterations to get this to work properly as it's hard to test.
### testing
I created some dummy GPG creds that should simulate how to CI environment will look like:
```
# simulate keys that will exist in CI with a du... | ebbfeb71f97c544c7595c123aee31d84c053d147 | 8c1b62f5c1bed7fc0ce59691e81caf273b776a52 | diff --git a/.github/workflows/publish.yml b/.github/workflows/publish.yml
index 6d7dec1db..49c2064f2 100644
--- a/.github/workflows/publish.yml
+++ b/.github/workflows/publish.yml
@@ -3,9 +3,9 @@ name: Publish
on:
push:
branches:
- - 'develop'
+ - "develop"
release:
- types: [ created ]
+ ... | [
".github/workflows/publish.yml",
"README.md",
"build.sbt",
"project/plugins.sbt"
] | [
{
"comment": "sbt throws some annoying warnings without this I don't think it does anything functional though",
"path": "build.sbt",
"hunk": "@@ -63,6 +62,7 @@ val sharedSettings = Seq(\n crossScalaVersions := Seq(scalaVersion.value, \"2.12.14\"),\n javacOptions ++= Seq(\"-source\", \"1.8\", \"-targ... | true | ||
twitter/scalding | 1,762 | issue_to_patch | Merge develop into cascading3 | Putting this up to have CI kick off tests so that we can move on to fixing tests and cleaning up stuff I might have messed up as the merge was pretty large. | bb7e48c5f423703289260dd303a29b6dace4d7b7 | 3c350b900301a13ec378d49459a6137c900cc763 | diff --git a/.gitignore b/.gitignore
index ca3990a270..15a05753a0 100644
--- a/.gitignore
+++ b/.gitignore
@@ -35,3 +35,7 @@ tutorial/data/avrooutput0.avro
.scalding_repl
scalding-hadoop-test/NOTICE
NOTICE
+
+# Auto-copied by sbt-microsites
+docs/src/main/tut/contributing.md
+.DS_Store
diff --git a/.travis.yml b/.tr... | [
".gitignore",
".travis.yml",
"CHANGES.md",
"CONTRIBUTING.md",
"README.md",
"build.sbt",
"docs/src/main/resources/microsite/img/favicon.png",
"docs/src/main/resources/microsite/img/navbar_brand.png",
"docs/src/main/resources/microsite/img/navbar_brand2x.png",
"docs/src/main/resources/microsite/img/... | [] | diff --git a/scalding-commons/src/test/scala/com/twitter/scalding/commons/VersionedKeyValSourceTest.scala b/scalding-commons/src/test/scala/com/twitter/scalding/commons/VersionedKeyValSourceTest.scala
index da9bd0290e..93d38ed42b 100644
--- a/scalding-commons/src/test/scala/com/twitter/scalding/commons/VersionedKeyValS... | true | |
twitter/scalding | 2,003 | issue_to_patch | Typo fix: Execution.scala | a0516e0f43b966a04244e8e8bad80438d61773ea | 2cb05e763f68c003d5819d51324ce49f48bce910 | diff --git a/scalding-base/src/main/scala/com/twitter/scalding/Execution.scala b/scalding-base/src/main/scala/com/twitter/scalding/Execution.scala
index 0447e311d..d7fedf157 100644
--- a/scalding-base/src/main/scala/com/twitter/scalding/Execution.scala
+++ b/scalding-base/src/main/scala/com/twitter/scalding/Execution.s... | [
"scalding-base/src/main/scala/com/twitter/scalding/Execution.scala"
] | [] | true | |||
twitter/scalding | 550 | issue_to_patch | Transient Config. | Make the conf transient.
| 318a98be86395877ab48d2c8ea652e22d68b7594 | 1b3e3506d1b473004cf99273dc70d6086635bbc4 | diff --git a/scalding-core/src/main/scala/com/twitter/scalding/Mode.scala b/scalding-core/src/main/scala/com/twitter/scalding/Mode.scala
index f9f8d55c87..39e12eb80d 100644
--- a/scalding-core/src/main/scala/com/twitter/scalding/Mode.scala
+++ b/scalding-core/src/main/scala/com/twitter/scalding/Mode.scala
@@ -146,7 +14... | [
"scalding-core/src/main/scala/com/twitter/scalding/Mode.scala",
"version.sbt"
] | [] | true | ||
twitter/scalding | 1,996 | issue_to_patch | publish for scalding base missing | I noticed when pulling down the `scalding-core` snapshot the `scalding-base` module was not published. This can be confirmed by looking at https://oss.sonatype.org/content/repositories/snapshots/com/twitter/
I'm not 100% on how sbt works but I think it needs to be listed under the aggregation for
```
lazy val sca... | e96797b9c5bb8bc979e06a36cd0cfb813a65a5dd | 5be9ff09a54ffd5736a650d4b8d842b06140741f | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index 6c5a88459..ce4c88060 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -115,7 +115,7 @@ jobs:
- uses: actions/setup-java@v2
with:
distribution: 'adopt-openj9'
- java-version: 8
+ jav... | [
".github/workflows/CI.yml",
".github/workflows/publish.yml",
"build.sbt",
"scalding-base/src/main/scala/com/twitter/scalding/Config.scala",
"scalding-base/src/main/scala/com/twitter/scalding/Execution.scala",
"scalding-base/src/main/scala/com/twitter/scalding/ExecutionOptimizationRules.scala",
"scalding... | [
{
"comment": "I think we want tags here instead of release. The last tag push didn't trigger publish because of this.",
"path": ".github/workflows/publish.yml",
"hunk": "@@ -4,8 +4,8 @@ on:\n push:\n branches:\n - \"develop\"\n- release:\n- types: [created]\n+ tags:\n+ - \"v*... | true | ||
twitter/scalding | 1,994 | issue_to_patch | make codecov not suck
**Describe the bug**
So we have two main flakes that make us re-run the CI:
1. the optimization tests that are jank
2. codecov failing to upload
this concerns the second point. If you look at the codecov issue tracker this is a common issue. The solution appear to be:
1. adding a sec... | use backoff on codecov uploader to reduce CI failures | added exponential retry to uploader which seems to fail often.
closes #1993 | 110055879abf738e329176ef0471ce86ba9f74b0 | 39edd33dd48a9af854b9c2179c2cbe725465b8b7 | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index edfc956bb..6c5a88459 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -137,34 +137,3 @@ jobs:
run: |
mkdir -p /home/runner/.ivy2/cache/ # some scripts assume that this folder exists
${{ matrix.sc... | [
".github/workflows/CI.yml",
"scripts/run_test.sh"
] | [] | diff --git a/scripts/run_test.sh b/scripts/run_test.sh
index dc29a763c..7e016bd50 100755
--- a/scripts/run_test.sh
+++ b/scripts/run_test.sh
@@ -23,8 +23,37 @@ echo "Running mima checks ... "
time ./sbt -Dhttp.keepAlive=false -Dsbt.repository.secure=true ++$TRAVIS_SCALA_VERSION "$(withCmd mimaReportBinaryIssues)"
... | true |
twitter/scalding | 1,992 | issue_to_patch | spark-backend CounterPipe | so counterPipe looks to be good... the discussion point I think is the best way to make `addForce` a blocking call. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Op.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Op.scala
index 7caa0b406..9d4c4a768 100644
--- a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Op.scala
+++ b/scalding-spark/src/main/scala/com/twi... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Op.scala",
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkBackend.scala",
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"scalding-spark/src/main/scala/com/twitter/scalding/spark_... | [
{
"comment": "I am currently confused why getting the SparkCounters state is empty inside of this map scope but if I compute it lazily it has the correct values later when asserting in tests. ",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala",
"hunk": "@@ -21... | diff --git a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala b/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala
index f6638d1e2..62fbf91f9 100644
--- a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scal... | true | |
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | I am currently confused why getting the SparkCounters state is empty inside of this map scope but if I compute it lazily it has the correct values later when asserting in tests. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala
index 03bf84f61..6dec15840 100644
--- a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala
+++ b/scalding-s... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala"
] | [
{
"comment": "I am currently confused why getting the SparkCounters state is empty inside of this map scope but if I compute it lazily it has the correct values later when asserting in tests. ",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala",
"hunk": "@@ -21... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | perf nits:
1. access `counterAction._1` once
2. iterator is better since it is always inlined
```scala
resolve.map { x =>
val counts = x._2.iterator
while (counts.hasNext) {
val ((g, c), amt) = counts.next()
SparkCounters.addToCounter(g, c, amt)
}
x._1
}
``` | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkBackend.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkBackend.scala
index 94e9b6244..a8639cbd7 100644
--- a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkBackend.scala
+++ b/scaldin... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkBackend.scala"
] | [
{
"comment": "perf nits:\r\n\r\n1. access `counterAction._1` once\r\n2. iterator is better since it is always inlined\r\n\r\n```scala\r\nresolve.map { x =>\r\n val counts = x._2.iterator\r\n while (counts.hasNext) {\r\n val ((g, c), amt) = counts.next()\r\n SparkCounters.addToCounter(g, c, amt)\r\n }\r... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | 1. can we make this private?
2. I would probably instead do `@transient private lazy val accumulator = new AtomicReference[SparkCountersInternal](null)` and use atomic operations since I think you can have multithreading in spark. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "1. can we make this private?\r\n2. I would probably instead do `@transient private lazy val accumulator = new AtomicReference[SparkCountersInternal](null)` and use atomic operations since I think you can have multithreading in spark.",
"path": "scalding-spark/src/main/scala/com/twitter/scaldin... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | if you do this in a loop you are calling `getAccumulator` many times. I would also add a method that accepts `Iterable[((String, String), Long)]` or something so you can get once, then add everything. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "if you do this in a loop you are calling `getAccumulator` many times. I would also add a method that accepts `Iterable[((String, String), Long)]` or something so you can get once, then add everything.",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.sca... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | why make this a val? | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "why make this a val?",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0,0 +1,104 @@\n+package com.twitter.scalding.spark_backend\n+\n+import com.twitter.scalding.{ExecutionCounters, StatKey}\n+import org.apache.spark.sql.SparkSe... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | if you make this `AccumulatorV2[((String, String), Long), ...]` can you avoid reboxing the counters? | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "if you make this `AccumulatorV2[((String, String), Long), ...]` can you avoid reboxing the counters?",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0,0 +1,104 @@\n+package com.twitter.scalding.spark_backend\n+\n+import com.twi... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | `0L` would be better to be more specific. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "`0L` would be better to be more specific.",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0,0 +1,104 @@\n+package com.twitter.scalding.spark_backend\n+\n+import com.twitter.scalding.{ExecutionCounters, StatKey}\n+import org.apa... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | I think this is probably better done in `newWriter` in `SparkMode` or in the constructor here.
You can have many stages to an `Execution`. Do you want to do this on every stage? I don't think so.
Also, I think ideally we could have a cleanup where maybe we unregister when the Writer is closed? | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala
index 03bf84f61..6dec15840 100644
--- a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala
+++ b/scalding-s... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala"
] | [
{
"comment": "I think this is probably better done in `newWriter` in `SparkMode` or in the constructor here.\r\n\r\nYou can have many stages to an `Execution`. Do you want to do this on every stage? I don't think so.\r\n\r\nAlso, I think ideally we could have a cleanup where maybe we unregister when the Writer ... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | we are using `toIterableExecution` here, so the fact that the `toLocalIterator` is being done in the wrong place is likely your issue. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala b/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala
index f6638d1e2..62fbf91f9 100644
--- a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scal... | [
"scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala"
] | [
{
"comment": "we are using `toIterableExecution` here, so the fact that the `toLocalIterator` is being done in the wrong place is likely your issue.",
"path": "scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala",
"hunk": "@@ -67,6 +67,12 @@ class SparkBackendTests e... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | okay I currently have this hack to make the addForce method be blocking. I'm sure there is a better way but I currently can't think of one. Is this alright for now? | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala
index 03bf84f61..6dec15840 100644
--- a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala
+++ b/scalding-s... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala"
] | [
{
"comment": "okay I currently have this hack to make the addForce method be blocking. I'm sure there is a better way but I currently can't think of one. Is this alright for now? ",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala",
"hunk": "@@ -161,19 +168,32 ... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | added a more tricky test case | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala b/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala
index f6638d1e2..62fbf91f9 100644
--- a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scal... | [
"scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala"
] | [
{
"comment": "added a more tricky test case",
"path": "scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala",
"hunk": "@@ -145,6 +151,44 @@ class SparkBackendTests extends FunSuite with BeforeAndAfter {\n }\n }\n \n+ // note tallyAll takes (group, counter)\n+ ... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | let's remove the `{ }` please. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "let's remove the `{ }` please.",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0,0 +1,89 @@\n+package com.twitter.scalding.spark_backend\n+\n+import com.twitter.scalding.{ExecutionCounters, StatKey}\n+import org.apache.spark.sq... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | can this be private? Can you add a comment on why this mutable value is safe? Is it never used in a multi-threaded context? | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "can this be private? Can you add a comment on why this mutable value is safe? Is it never used in a multi-threaded context?",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0,0 +1,89 @@\n+package com.twitter.scalding.spark_backe... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | Can we remove the side-effecting default value. Instead can you make a method on the companion:
```scala
object SparkCounters {
def withRandomContextPrefix(sparkSession: SparkSession): SparkCounters =
new SparkCounters(sparkSession, java.util.UUID.randomUUID.toString)
}
``` | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "Can we remove the side-effecting default value. Instead can you make a method on the companion:\r\n\r\n```scala\r\nobject SparkCounters {\r\n def withRandomContextPrefix(sparkSession: SparkSession): SparkCounters =\r\n new SparkCounters(sparkSession, java.util.UUID.randomUUID.toString)\r\n}\r\... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | why doesn't `def keys = copyState.keySet` work here? copystate never changes after this method right? | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "why doesn't `def keys = copyState.keySet` work here? copystate never changes after this method right?",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0,0 +1,89 @@\n+package com.twitter.scalding.spark_backend\n+\n+import com.twi... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | unfortunately, `getOrElseUpdate` allocates since the `else` clause is lazy. Can we move to:
```scala
val prev =
mapState.get(count._1) match {
case Some(v) => v
case None => 0L
}
mapState(count._1) = prev + count._2
``` | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "unfortunately, `getOrElseUpdate` allocates since the `else` clause is lazy. Can we move to:\r\n\r\n```scala\r\nval prev =\r\n mapState.get(count._1) match {\r\n case Some(v) => v\r\n case None => 0L\r\n }\r\nmapState(count._1) = prev + count._2\r\n```",
"path": "scalding-spark/src/main... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | convention in scala is to use `()` for side-effecting things, like this: `val count = in.next()` | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "convention in scala is to use `()` for side-effecting things, like this: `val count = in.next()`",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0,0 +1,89 @@\n+package com.twitter.scalding.spark_backend\n+\n+import com.twitter.... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | same comment as above about using the match pattern. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "same comment as above about using the match pattern.",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0,0 +1,89 @@\n+package com.twitter.scalding.spark_backend\n+\n+import com.twitter.scalding.{ExecutionCounters, StatKey}\n+impo... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | mapState is already a map, so I don't think you need this `getOrElseUpdate`. I think:
```scala
val bldr = MutableMap.newBuilder[(String, String), Long]()
bldr ++= mapState
bldr.result()
```
is the more efficient way to do this. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "mapState is already a map, so I don't think you need this `getOrElseUpdate`. I think:\r\n\r\n```scala\r\nval bldr = MutableMap.newBuilder[(String, String), Long]()\r\nbldr ++= mapState\r\nbldr.result()\r\n```\r\nis the more efficient way to do this.",
"path": "scalding-spark/src/main/scala/com... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | I think it is better to do:
```scala
val prev = mapState.get(kvPair._1) {
case Some(v) => v
case None => 0L
}
mapState(kvPair._1) = prev + kvPair._2
``` | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "I think it is better to do:\r\n```scala\r\nval prev = mapState.get(kvPair._1) {\r\n case Some(v) => v\r\n case None => 0L\r\n}\r\nmapState(kvPair._1) = prev + kvPair._2\r\n```",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala",
"hunk": "@@ -0... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | shouldn't we get the `forcedRdd` from the `WorkVal` you have here? I think so. If not, please add a comment why not. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala
index 03bf84f61..6dec15840 100644
--- a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala
+++ b/scalding-s... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala"
] | [
{
"comment": "shouldn't we get the `forcedRdd` from the `WorkVal` you have here? I think so. If not, please add a comment why not.",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkWriter.scala",
"hunk": "@@ -63,9 +65,9 @@ class SparkWriter(val sparkMode: SparkMode) extend... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | do you actually need the `Ordering` here? I don't see why. | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala b/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala
index f6638d1e2..62fbf91f9 100644
--- a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scal... | [
"scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala"
] | [
{
"comment": "do you actually need the `Ordering` here? I don't see why.",
"path": "scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala",
"hunk": "@@ -67,6 +67,12 @@ class SparkBackendTests extends FunSuite with BeforeAndAfter {\n t.toIterableExecution.waitFor(... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | Can we also do a version where you write to a null-sink or text file or something and make sure the `writeExecution` version also works? | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala b/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala
index f6638d1e2..62fbf91f9 100644
--- a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scal... | [
"scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala"
] | [
{
"comment": "Can we also do a version where you write to a null-sink or text file or something and make sure the `writeExecution` version also works?",
"path": "scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBackendTests.scala",
"hunk": "@@ -145,6 +151,44 @@ class SparkBackendTes... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | this isn't thread-safe: mapState can change while iterating.
Can you comment why this is safe?
Alternatives:
1. use `mapState.synchronized { }` before reading/writing
2. use an `immutable.Map[(String, String), Long]` and an `AtomicReference` so taking a copy is very cheap (although updates will be maybe twice a... | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "this isn't thread-safe: mapState can change while iterating.\r\n\r\nCan you comment why this is safe?\r\n\r\nAlternatives:\r\n1. use `mapState.synchronized { }` before reading/writing\r\n2. use an `immutable.Map[(String, String), Long]` and an `AtomicReference` so taking a copy is very cheap (alth... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | I believe I located the problem with the failing test. We need to reset the counter state before we return from here. I think what's going on is that flapMapping the execution re-uses the same writer which duplicates the counter values.
I confirmed that the actual counter incrementation occurs only once by stepping ... | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "I believe I located the problem with the failing test. We need to reset the counter state before we return from here. I think what's going on is that flapMapping the execution re-uses the same writer which duplicates the counter values.\r\n\r\nI confirmed that the actual counter incrementation occ... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | nit:
```scala
// get this without holding the lock here, so we avoiding holding two locks simultaneously
val thatMapState = other.value
mapState.synchronized {
for (kvPair <- thatMapState) { ...
}
``` | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala
new file mode 100644
index 000000000..a4e802e86
--- /dev/null
+++ b/scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/main/scala/com/twitter/scalding/spark_backend/SparkCounters.scala"
] | [
{
"comment": "nit:\r\n\r\n```scala\r\n// get this without holding the lock here, so we avoiding holding two locks simultaneously\r\nval thatMapState = other.value\r\nmapState.synchronized {\r\n for (kvPair <- thatMapState) { ... \r\n}\r\n```",
"path": "scalding-spark/src/main/scala/com/twitter/scalding/spar... | true | ||
twitter/scalding | 1,992 | comment_to_fix | spark-backend CounterPipe | can we remove `[A]` and use `Execution[Any]` | 110055879abf738e329176ef0471ce86ba9f74b0 | 9ec0b7e52fbf708b31dee550381f1659ed47f744 | diff --git a/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBaseTest.scala b/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBaseTest.scala
new file mode 100644
index 000000000..ffb004c42
--- /dev/null
+++ b/scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/Sp... | [
"scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBaseTest.scala"
] | [
{
"comment": "can we remove `[A]` and use `Execution[Any]`",
"path": "scalding-spark/src/test/scala/com/twitter/scalding/spark_backend/SparkBaseTest.scala",
"hunk": "@@ -0,0 +1,80 @@\n+package com.twitter.scalding.spark_backend\n+\n+import com.twitter.scalding.{Config, Execution}\n+import com.twitter.sc... | true | ||
twitter/scalding | 1,991 | issue_to_patch | [bug/nit] code format enforcement lost
**Describe the bug**
When we migrated the CI, we lost the auto-formatting CI check: https://github.com/twitter/scalding/pull/1954
I prefer to not worry about formatting when working because it's a waste of time. Todo: add back format enforcement on CI, re-run format on PR. I ... | add format check | closes #1990
should first fail on develop. Once I push a format commit to the PR it should pass. | 1813d8f32904598d34b92ffd71010f85c18d58db | 70f0213e4d40d0410ee222825c1ac2cf75a2dd7d | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index c8d106f0a..edfc956bb 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -21,6 +21,11 @@ jobs:
# all combinations that we will run in parallel to increase throughput
matrix:
include:
+
+ - scala-ve... | [
".github/workflows/CI.yml",
".scalafmt.conf",
"build.sbt",
"project/plugins.sbt",
"project/scalding-dagon.scala",
"scalding-args/src/main/scala/com/twitter/scalding/Args.scala",
"scalding-avro/src/main/scala/com/twitter/scalding/avro/SchemaType.scala",
"scalding-beam/src/main/scala/com/twitter/scaldin... | [
{
"comment": "seems like autoformat can break some multiline commends maybe not a big deal?",
"path": "scalding-core/src/main/scala/com/twitter/package.scala",
"hunk": "@@ -32,8 +32,8 @@ package object scalding {\n type Grouped[K, +V] = com.twitter.scalding.typed.Grouped[K, V]\n \n /**\n- * scaldi... | diff --git a/scalding-commons/src/test/scala/com/twitter/scalding/PageRankTest.scala b/scalding-commons/src/test/scala/com/twitter/scalding/PageRankTest.scala
index f210fee45..f9998f1b6 100644
--- a/scalding-commons/src/test/scala/com/twitter/scalding/PageRankTest.scala
+++ b/scalding-commons/src/test/scala/com/twitter... | true |
twitter/scalding | 1,986 | issue_to_patch | scalding-hadoop3 backend OR spark backend upgrade to 3.X
**Is your feature request related to a problem? Please describe.**
For the use cases that @jonshea and I need, we require hadoop3 or spark3. It doesn't matter which one but he have a managed runtime (AWS EMR) that forces us to use newer versions of either.
... | cross build spark 2.X (2.11) and spark 3.X (2.12) on spark-backend | build spark 2.X on scala 2.11 and 3.X on scala 2.12. Tests pass cleanly without further changes:
```bash
./sbt +scalding-spark/test
```
If there's no objection to this from any current users of the spark backend it would be helpful for the use case that I need. Big thanks to the work done in https://github.com... | 64343489135111ca205e5f5e894367cdd545f765 | 3f52739fac2b85aa7cd525adaec5ca2a7c74ef1d | diff --git a/build.sbt b/build.sbt
index 054a19320..aa8624304 100644
--- a/build.sbt
+++ b/build.sbt
@@ -29,7 +29,6 @@ val scalameterVersion = "0.8.2"
val scalaCheckVersion = "1.13.4"
val scalaTestVersion = "3.0.1"
val scroogeVersion = "19.8.0"
-val sparkVersion = "2.4.0"
val beamVersion = "2.29.0"
val slf4jVersio... | [
"build.sbt"
] | [] | true | |
twitter/scalding | 1,984 | issue_to_patch | no op change to test publish and see if something breaks | added new line to readme @johnynek | e44f2f21fa90f669d6bf81af607c5d537f19e93c | d2ef66a7332589af749649e1597cd20de130f34e | diff --git a/README.md b/README.md
index 14d7896aa..afb0c18c0 100644
--- a/README.md
+++ b/README.md
@@ -156,3 +156,4 @@ A full list of [contributors](https://github.com/twitter/scalding/graphs/contrib
Copyright 2016 Twitter, Inc.
Licensed under the [Apache License, Version 2.0](http://www.apache.org/licenses/LICEN... | [
"README.md"
] | [] | true | ||
twitter/scalding | 1,972 | issue_to_patch | add scalding-base, minimal dep base module | The idea here is to make an *almost completely* source compatible change that moves as many of the main API types into a scalding-base module that does not depend on any compute substrate (e.g. Hadoop).
Most jobs should require no source change, but some library functions or tests may need to do:
```
import com.... | 3d631701bf04da92dbbff0400d2aa801b6048bbc | 8310a79a17b967b31497f8db64785926a67c97b3 | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index 3b15f0d65a..c8d106f0a9 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -23,11 +23,11 @@ jobs:
include:
- scala-version: "2.11.12"
BUILD: "base"
- TEST_TARGET: "scalding-args scaldin... | [
".github/workflows/CI.yml",
"build.sbt",
"scalding-avro/src/main/scala/com/twitter/scalding/avro/package.scala",
"scalding-base/src/main/scala/com/twitter/scalding/CFuture.scala",
"scalding-base/src/main/scala/com/twitter/scalding/CPromise.scala",
"scalding-base/src/main/scala/com/twitter/scalding/Cancell... | [
{
"comment": "this needs to be built up a bit more and there is a duplicated class with the same name in core.",
"path": "scalding-base/src/main/scala/com/twitter/scalding/Execution.scala",
"hunk": "@@ -40,6 +38,15 @@ import scala.concurrent.{\n import scala.util.{Failure, Success, Try}\n import scala.u... | diff --git a/scalding-core/src/test/scala/com/twitter/scalding/DistinctByTest.scala b/scalding-base/src/test/scala/com/twitter/scalding/DistinctByTest.scala
similarity index 100%
rename from scalding-core/src/test/scala/com/twitter/scalding/DistinctByTest.scala
rename to scalding-base/src/test/scala/com/twitter/scaldin... | true | |
twitter/scalding | 1,981 | issue_to_patch | Enable Codecoverage Upload (Jacoco/Codecov) | - [x] verify that it is actually running
- [x] codecov.yml file probably good idea
- [x] clean up for review
- [x] verify that codecov seems reasonable
basically moving changes over from https://github.com/twitter/scalding/pull/1922 based on https://github.com/twitter/scalding/issues/1977 | 525321c728ca79a7977f66a02d4b5962ad64ce29 | 5db2df2a50b4a92c0fbf934a5dc13b06229624e9 | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index 6ace454d6..3b15f0d65 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -132,3 +132,34 @@ jobs:
run: |
mkdir -p /home/runner/.ivy2/cache/ # some scripts assume that this folder exists
${{ matrix.sc... | [
".github/workflows/CI.yml",
"README.md",
"build.sbt",
"codecov.yml",
"project/plugins.sbt",
"scripts/run_test.sh"
] | [
{
"comment": "the documentation assures that no credentials are needed using the provided action. I'm not sure how the uploaded works and if it can auto-detect the generated coverage files",
"path": ".github/workflows/CI.yml",
"hunk": "@@ -130,3 +130,12 @@ jobs:\n run: |\n mkdir -p /ho... | diff --git a/scripts/run_test.sh b/scripts/run_test.sh
index bd85b4cf9..dc29a763c 100755
--- a/scripts/run_test.sh
+++ b/scripts/run_test.sh
@@ -4,41 +4,27 @@
BASE_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )"/.. && pwd )"
cd $BASE_DIR
-export JVM_OPTS="-XX:+CMSClassUnloadingEnabled -XX:+UseConcMarkSweepGC -XX:Rese... | true | |
twitter/scalding | 1,981 | comment_to_fix | Enable Codecoverage Upload (Jacoco/Codecov) | the documentation assures that no credentials are needed using the provided action. I'm not sure how the uploaded works and if it can auto-detect the generated coverage files | 525321c728ca79a7977f66a02d4b5962ad64ce29 | 5db2df2a50b4a92c0fbf934a5dc13b06229624e9 | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index 6ace454d6..3b15f0d65 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -132,3 +132,34 @@ jobs:
run: |
mkdir -p /home/runner/.ivy2/cache/ # some scripts assume that this folder exists
${{ matrix.sc... | [
".github/workflows/CI.yml"
] | [
{
"comment": "the documentation assures that no credentials are needed using the provided action. I'm not sure how the uploaded works and if it can auto-detect the generated coverage files",
"path": ".github/workflows/CI.yml",
"hunk": "@@ -130,3 +130,12 @@ jobs:\n run: |\n mkdir -p /ho... | true | ||
twitter/scalding | 1,982 | issue_to_patch | CI.yml still has some typos | Noticed these while working on a draft for test coverage upload.
@johnynek @nownikhil | 95ba97371ba6c519bd8fd9c1c460457bdbfaa7eb | 6c0786bb30c59518e77b5c04831215dbcfe4dc08 | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index f758f110e..6ace454d6 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -36,7 +36,7 @@ jobs:
script: './scripts/run_test.sh'
- scala-version: "2.12.14"
BUILD: "base"
- TEST_TAGET: ... | [
".github/workflows/CI.yml"
] | [
{
"comment": "typo here wasn't running these at all",
"path": ".github/workflows/CI.yml",
"hunk": "@@ -36,7 +36,7 @@ jobs:\n script: './scripts/run_test.sh'\n - scala-version: \"2.12.14\"\n BUILD: \"base\"\n- TEST_TAGET: \"scalding-avro scalding-hraven scaldi... | true | ||
twitter/scalding | 1,980 | issue_to_patch | [Ready for Review / Merge] Subtree/Submodule stripe/dagon | - [x] copy dagon src into scalding module
- [x] dagon tests pass in isolation
- [x] wire scalding-dagon to CI tests
- [x] re-write scalding imports against module
- [x] clean up PR
notes: dagon from the repo used scalatestplus that had conflicts with the `scalatest` and `scalacheck` versions of scalding default... | 71fd36cc0316efab14db0d42f7928d85f3116d99 | f223c8c0588f6229a983c95e6760448a471e30b5 | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index 33a2ab1e6..f758f110e 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -23,11 +23,11 @@ jobs:
include:
- scala-version: "2.11.12"
BUILD: "base"
- TEST_TARGET: "scalding-args scalding-... | [
".github/workflows/CI.yml",
"build.sbt",
"project/scalding-dagon.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWr... | [
{
"comment": "these type of changes are based on https://github.com/typelevel/kind-projector/pull/172",
"path": "scalding-dagon/src/main/scala/com/twitter/scalding/dagon/Dag.scala",
"hunk": "@@ -34,7 +34,7 @@ sealed abstract class Dag[N[_]] extends Serializable { self =>\n * the law that for all Expr... | diff --git a/scalding-core/src/test/scala/com/twitter/scalding/ExecutionOptimizationRulesTest.scala b/scalding-core/src/test/scala/com/twitter/scalding/ExecutionOptimizationRulesTest.scala
index 079b0b734..237746ff5 100644
--- a/scalding-core/src/test/scala/com/twitter/scalding/ExecutionOptimizationRulesTest.scala
+++ ... | true | |
twitter/scalding | 1,975 | issue_to_patch | [Ready for Review] Add Snapshot Workflow for Develop | I looked at some tools and ended up deciding that just re-using the existing publish workflow makes the most sense without having to change the existing CI workflow for tests or causing a lot of diffs in general.
| 354e9c0d0cb7b855da6f86543f3ff9d51c892483 | 01cd0acc8f9aa471babb3513cfc2221507185177 | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index 0d3b80c6e..33a2ab1e6 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -102,6 +102,8 @@ jobs:
steps:
- name: Checkout Repo
uses: actions/checkout@v2
+ with:
+ fetch-depth: 0 # fetch all tags... | [
".github/workflows/CI.yml",
".github/workflows/publish.yml",
".gitignore",
"build.sbt",
"project/plugins.sbt",
"scripts/scald.rb",
"scripts/test_execution_tutorial.sh",
"version.sbt"
] | [
{
"comment": "todo: I will make this depend conditionally on if we are pushing a snapshot or not",
"path": "build.sbt",
"hunk": "@@ -118,7 +118,6 @@ val sharedSettings = Seq(\n setNextVersion,\n commitNextVersion,\n ReleaseStep(action = Command.process(\"sonatypeReleaseAll\", _)),\n- push... | diff --git a/scripts/test_execution_tutorial.sh b/scripts/test_execution_tutorial.sh
index 0bee453e0..b5d3a296e 100755
--- a/scripts/test_execution_tutorial.sh
+++ b/scripts/test_execution_tutorial.sh
@@ -4,9 +4,7 @@
BASE_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )"/.. && pwd )"
source ${BASE_DIR}/scripts/common.sh
... | true | |
twitter/scalding | 1,975 | comment_to_fix | [Ready for Review] Add Snapshot Workflow for Develop | todo: I will make this depend conditionally on if we are pushing a snapshot or not | 354e9c0d0cb7b855da6f86543f3ff9d51c892483 | 01cd0acc8f9aa471babb3513cfc2221507185177 | diff --git a/build.sbt b/build.sbt
index 9fbdef8e3..27729cf90 100644
--- a/build.sbt
+++ b/build.sbt
@@ -40,7 +40,25 @@ val jlineVersion = "2.14.3"
val printDependencyClasspath = taskKey[Unit]("Prints location of the dependencies")
+// these are override functions for sbt-dynver (plugin for resolving project versi... | [
"build.sbt"
] | [
{
"comment": "todo: I will make this depend conditionally on if we are pushing a snapshot or not",
"path": "build.sbt",
"hunk": "@@ -118,7 +118,6 @@ val sharedSettings = Seq(\n setNextVersion,\n commitNextVersion,\n ReleaseStep(action = Command.process(\"sonatypeReleaseAll\", _)),\n- push... | true | ||
twitter/scalding | 1,975 | comment_to_fix | [Ready for Review] Add Snapshot Workflow for Develop | this will be in follow-up PR once we have snapshots working well off develop | 354e9c0d0cb7b855da6f86543f3ff9d51c892483 | 01cd0acc8f9aa471babb3513cfc2221507185177 | diff --git a/build.sbt b/build.sbt
index 9fbdef8e3..27729cf90 100644
--- a/build.sbt
+++ b/build.sbt
@@ -40,7 +40,25 @@ val jlineVersion = "2.14.3"
val printDependencyClasspath = taskKey[Unit]("Prints location of the dependencies")
+// these are override functions for sbt-dynver (plugin for resolving project versi... | [
"build.sbt"
] | [
{
"comment": "this will be in follow-up PR once we have snapshots working well off develop",
"path": "build.sbt",
"hunk": "@@ -106,19 +106,18 @@ val sharedSettings = Seq(\n publishMavenStyle := true,\n Test / publishArtifact := false,\n pomIncludeRepository := { x => false },\n- releaseProcess :=... | true | ||
twitter/scalding | 1,967 | issue_to_patch | Implement ForceToDisk for Beam runner | We are defining a `TempSource` which captures output path and coder. We then add a resolver from `TempSource` to corresponding `BeamSource`.
Added unit test for `forceToDiskExecution`. | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
index bb0518629..084dae04f 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
+++ b/scalding-beam/src/main/sc... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala",
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "Tested by adding println() statements and they got printed only once. Whats a good way to add a test which checks that logic is not computed twice?\r\nI tried by defining AtomicLong() and incrementing it in a map function but that didn't work",
"path": "scalding-beam/src/test/scala/com/twitter... | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 85fe1dc01..c7df08104 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | true | |
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | Tested by adding println() statements and they got printed only once. Whats a good way to add a test which checks that logic is not computed twice?
I tried by defining AtomicLong() and incrementing it in a map function but that didn't work | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 85fe1dc01..c7df08104 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | [
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "Tested by adding println() statements and they got printed only once. Whats a good way to add a test which checks that logic is not computed twice?\r\nI tried by defining AtomicLong() and incrementing it in a map function but that didn't work",
"path": "scalding-beam/src/test/scala/com/twitter... | true | ||
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | Can it have some kind of unexpected behavior? | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
index 9061c801d..b57811d42 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala"
] | [
{
"comment": "Can it have some kind of unexpected behavior?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala",
"hunk": "@@ -115,6 +114,11 @@ class BeamWriter(val beamMode: BeamMode) extends Writer {\n }\n \n object BeamWriter {\n+ // We create BeamWriter per Ex... | true | ||
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | can you add a comment to the code here? It looks to me like everything in the `Iterator` is going to be read in a single thread.
I can't really see how that's not the case? That seems like it is going to really hard performance since you don't want to kill parallelism when using forceToDiskExecution. | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
index bb0518629..084dae04f 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
+++ b/scalding-beam/src/main/sc... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala"
] | [
{
"comment": "can you add a comment to the code here? It looks to me like everything in the `Iterator` is going to be read in a single thread.\r\n\r\nI can't really see how that's not the case? That seems like it is going to really hard performance since you don't want to kill parallelism when using forceToDisk... | true | ||
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | I think you need to iterate through the forcedPipes here and delete them all. If we don't do this, you will probably have problems leaking space.
We had many issues like that in the past. | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
index 9061c801d..b57811d42 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala"
] | [
{
"comment": "I think you need to iterate through the forcedPipes here and delete them all. If we don't do this, you will probably have problems leaking space.\r\n\r\nWe had many issues like that in the past.",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala",
"h... | true | ||
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | can we change this to:
```scala
case OptimizedWrite(pipe, toWrite @ (ToIterable(_) | Force(_)) if !tempSources.contains(pipe) =>
// now we know matching below is total
``` | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
index 9061c801d..b57811d42 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala"
] | [
{
"comment": "can we change this to:\r\n\r\n```scala\r\ncase OptimizedWrite(pipe, toWrite @ (ToIterable(_) | Force(_)) if !tempSources.contains(pipe) =>\r\n // now we know matching below is total\r\n```",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala",
"hunk":... | true | ||
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | can we add a total match and comment:
```scala
case OptimizedWrite(_, ToIterable(_) | Force(_)) =>
// we know that tempSources.contains(pipe) on this branch, which means it was already computed.
``` | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
index 9061c801d..b57811d42 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala"
] | [
{
"comment": "can we add a total match and comment:\r\n\r\n```scala\r\ncase OptimizedWrite(_, ToIterable(_) | Force(_)) =>\r\n // we know that tempSources.contains(pipe) on this branch, which means it was already computed.\r\n```",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/Bea... | true | ||
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | would be good to test somehow that the forceToDisk data has been cleaned by this point. | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 85fe1dc01..c7df08104 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | [
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "would be good to test somehow that the forceToDisk data has been cleaned by this point.",
"path": "scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala",
"hunk": "@@ -389,6 +389,31 @@ class BeamBackendTests extends FunSuite with BeforeAndAfter {\n )\n ... | true | ||
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | After adding logic of deletion, this started to fail with files not found. This is because iterator is reading the output files as it is. For what I know is that we don't use `toIterableExecution` as a final step, we always map the iterator to do another computation. | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 85fe1dc01..c7df08104 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | [
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "After adding logic of deletion, this started to fail with files not found. This is because iterator is reading the output files as it is. For what I know is that we don't use `toIterableExecution` as a final step, we always map the iterator to do another computation. ",
"path": "scalding-beam/... | true | ||
twitter/scalding | 1,967 | comment_to_fix | Implement ForceToDisk for Beam runner | can we add a total match:
```scala
case _ => Future.failed(new IllegalStateException(s"TypedPipe = $initial has no existing Iterable"))
``` | 262285fbd16f143e9b6541f4167cd91633dfae84 | aa24f06856802c3504ec3f289d1a50cf042ff697 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
index 9061c801d..b57811d42 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala"
] | [
{
"comment": "can we add a total match:\r\n```scala\r\ncase _ => Future.failed(new IllegalStateException(s\"TypedPipe = $initial has no existing Iterable\"))\r\n```",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala",
"hunk": "@@ -1,62 +1,70 @@\n package com.twitt... | true | ||
twitter/scalding | 1,976 | issue_to_patch | [minor fix] codegen build version for logging | based on this note in https://github.com/twitter/scalding/pull/1975: https://github.com/twitter/scalding/pull/1975#issuecomment-1067147645
the scalding build version is sometimes logged for info. But is is not kept up to date. This uses the https://github.com/sbt/sbt-buildinfo plugin to codegen the build version ins... | 262285fbd16f143e9b6541f4167cd91633dfae84 | 6e9d0f755ef47ce42c91c243cc667e856de1ea77 | diff --git a/build.sbt b/build.sbt
index 5ebfd42ac..9fbdef8e3 100644
--- a/build.sbt
+++ b/build.sbt
@@ -300,8 +300,13 @@ lazy val scaldingCore = module("core")
"org.slf4j" % "slf4j-api" % slf4jVersion,
"org.slf4j" % "slf4j-log4j12" % slf4jVersion % "provided"
),
+ // builInfo here refers to https... | [
"build.sbt",
"project/plugins.sbt",
"scalding-core/src/main/scala/com/twitter/package.scala"
] | [] | true | ||
twitter/scalding | 1,925 | issue_to_patch | [build] bump slf4j to latest version | This solves several issues (see http://www.slf4j.org/news.html for
details) including specific memory leaks.
(noticed when I was looking at other versions. Not for a specific
feature) | 1de7f10731de70e05018d273380547b7dc8b4a91 | 8fecc0caafb568308b10cd58c682b37b2c977f83 | diff --git a/build.sbt b/build.sbt
index 8e2fd3e84..e8f0fb328 100644
--- a/build.sbt
+++ b/build.sbt
@@ -33,7 +33,7 @@ val scalaTestVersion = "3.0.1"
val scroogeVersion = "19.8.0"
val sparkVersion = "2.4.0"
val beamVersion = "2.29.0"
-val slf4jVersion = "1.6.6"
+val slf4jVersion = "1.7.30"
val thriftVersion = "0.9.... | [
"build.sbt"
] | [] | true | ||
twitter/scalding | 1,969 | issue_to_patch | remove scalding-jbdc | based on https://github.com/twitter/scalding/issues/1968. We are not aware of any popular use of this module. Even if someone is using it it will likely need to be re-written in future scalding releases because:
1. it is hard-wired to cascading
2. it depends on cascading ecosystem packages that have not been main... | 1312b22b389b6ae226be6e7ab9a3af44bac2fde5 | 2bee7e0566fa13242336dff7459f2aa8233f9f8f | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index d2426e2e96..0d3b80c6e1 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -1,13 +1,14 @@
name: CI
-# run on all branches
on:
pull_request:
branches:
- '**'
+ # it's annoying to have CI double run on PR/push
+... | [
".github/workflows/CI.yml",
"build.sbt",
"scalding-jdbc/src/main/scala/com/twitter/scalding/jdbc/ColumnDefiner.scala",
"scalding-jdbc/src/main/scala/com/twitter/scalding/jdbc/ConnectionSpec.scala",
"scalding-jdbc/src/main/scala/com/twitter/scalding/jdbc/DriverColumnDefiner.scala",
"scalding-jdbc/src/main/... | [
{
"comment": "Agree",
"path": ".github/workflows/CI.yml",
"hunk": "@@ -1,13 +1,14 @@\n name: CI\n \n-# run on all branches\n on:\n pull_request:\n branches:\n - '**'\n+ # it's annoying to have CI double run on PR/push\n+ # limit push to only final integration branches like develop",
"r... | diff --git a/scalding-jdbc/src/test/scala/com/twitter/scalding/jdbc/JDBCSourceCompileTest.scala b/scalding-jdbc/src/test/scala/com/twitter/scalding/jdbc/JDBCSourceCompileTest.scala
deleted file mode 100644
index dcb1cd152c..0000000000
--- a/scalding-jdbc/src/test/scala/com/twitter/scalding/jdbc/JDBCSourceCompileTest.sc... | true | |
twitter/scalding | 1,970 | issue_to_patch | Use a null flow process when one cannot be found | The Beam runner for Scalding does not work with Hadoop counters(Stat). When Scalding jobs that use the Stat API are run using the Beam runner, they result in the following error:
`Error in job deployment, the FlowProcess for unique id %s isn't available".format(uniqueId)`
It looks like currently it is not possibl... | 1312b22b389b6ae226be6e7ab9a3af44bac2fde5 | ece327bffb839913b590205233b2e10aae9e9b69 | diff --git a/scalding-core/src/main/scala/com/twitter/scalding/Stats.scala b/scalding-core/src/main/scala/com/twitter/scalding/Stats.scala
index 9ac3087d0..b533449b5 100644
--- a/scalding-core/src/main/scala/com/twitter/scalding/Stats.scala
+++ b/scalding-core/src/main/scala/com/twitter/scalding/Stats.scala
@@ -169,7 +... | [
"scalding-core/src/main/scala/com/twitter/scalding/Stats.scala",
"scalding-core/src/test/scala/com/twitter/scalding/WrappedJoinerTest.scala"
] | [] | diff --git a/scalding-core/src/test/scala/com/twitter/scalding/WrappedJoinerTest.scala b/scalding-core/src/test/scala/com/twitter/scalding/WrappedJoinerTest.scala
index 42b8561c2..d951a7451 100644
--- a/scalding-core/src/test/scala/com/twitter/scalding/WrappedJoinerTest.scala
+++ b/scalding-core/src/test/scala/com/twit... | true | |
twitter/scalding | 1,962 | issue_to_patch | ToIterable execution in Beam runner | We use the coder from the PCollection and store data as bytes in temp location using FileIO.
We later read this file using the same coder and get the data on the submitter node.
TESTS: Added unit tests + Tested with Dataflow Runner | 1de7f10731de70e05018d273380547b7dc8b4a91 | db471730292f9e733bc9ec8bfbbe05d5811897ed | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
index b567d0f72..bb0518629 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
+++ b/scalding-beam/src/main/sc... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala",
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "Maybe a cleaner way to do this?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala",
"hunk": "@@ -74,3 +78,54 @@ object BeamSink extends Serializable {\n pc.asInstanceOf[PCollection[String]].apply(TextIO.write().to(path))\n }\n }\n+\n+class... | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index ef83f4622..85fe1dc01 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | true | |
twitter/scalding | 1,962 | comment_to_fix | ToIterable execution in Beam runner | Maybe a cleaner way to do this? | 1de7f10731de70e05018d273380547b7dc8b4a91 | db471730292f9e733bc9ec8bfbbe05d5811897ed | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
index b567d0f72..bb0518629 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
+++ b/scalding-beam/src/main/sc... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala"
] | [
{
"comment": "Maybe a cleaner way to do this?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala",
"hunk": "@@ -74,3 +78,54 @@ object BeamSink extends Serializable {\n pc.asInstanceOf[PCollection[String]].apply(TextIO.write().to(path))\n }\n }\n+\n+class... | true | ||
twitter/scalding | 1,962 | comment_to_fix | ToIterable execution in Beam runner | this worries me. What if you see an error that is not EOF?
Also, boxing with Try is an extra cost you don't need.
```scala
try {
nextRecord = coder.decode(stream)
hasNextRecord = true
}
catch {
case _: EOFException =>
hasNextRecord = false
// don't catch other errors.
}
``` | 1de7f10731de70e05018d273380547b7dc8b4a91 | db471730292f9e733bc9ec8bfbbe05d5811897ed | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
index b567d0f72..bb0518629 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
+++ b/scalding-beam/src/main/sc... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala"
] | [
{
"comment": "this worries me. What if you see an error that is not EOF?\r\n\r\nAlso, boxing with Try is an extra cost you don't need.\r\n```scala\r\ntry {\r\n nextRecord = coder.decode(stream)\r\n hasNextRecord = true\r\n}\r\ncatch {\r\n case _: EOFException =>\r\n hasNextRecord = false\r\n // don't cat... | true | ||
twitter/scalding | 1,962 | comment_to_fix | ToIterable execution in Beam runner | I think this needs to be thread-safe, so it may be best to use `ConcurrentHashMap` from java or use a mutex somehow.
| 1de7f10731de70e05018d273380547b7dc8b4a91 | db471730292f9e733bc9ec8bfbbe05d5811897ed | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
index 6130a71a2..9061c801d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala"
] | [
{
"comment": "I think this needs to be thread-safe, so it may be best to use `ConcurrentHashMap` from java or use a mutex somehow.\r\n",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala",
"hunk": "@@ -4,14 +4,21 @@ import com.stripe.dagon.Rule\n import com.twitter... | true | ||
twitter/scalding | 1,962 | comment_to_fix | ToIterable execution in Beam runner | this should probably be a `require`. An assertion is something that *must* be true by construction, not a requirement. | 1de7f10731de70e05018d273380547b7dc8b4a91 | db471730292f9e733bc9ec8bfbbe05d5811897ed | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
index 6130a71a2..9061c801d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala"
] | [
{
"comment": "this should probably be a `require`. An assertion is something that *must* be true by construction, not a requirement.",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala",
"hunk": "@@ -48,7 +75,16 @@ class BeamWriter(val beamMode: BeamMode) extends W... | true | ||
twitter/scalding | 1,962 | comment_to_fix | ToIterable execution in Beam runner | nit: can we use our safer `widen` function here (general cast is more dangerous). | 1de7f10731de70e05018d273380547b7dc8b4a91 | db471730292f9e733bc9ec8bfbbe05d5811897ed | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
index b567d0f72..bb0518629 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala
+++ b/scalding-beam/src/main/sc... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala"
] | [
{
"comment": "nit: can we use our safer `widen` function here (general cast is more dangerous).",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamMode.scala",
"hunk": "@@ -74,16 +73,15 @@ object BeamSink extends Serializable {\n \n def textLine(path: String): BeamSink[Strin... | true | ||
twitter/scalding | 1,962 | comment_to_fix | ToIterable execution in Beam runner | nit: I think `matchedResources.iterator.flatMap { ` clearer than manually reimplementing via fold.
Now that I think about this though, we are never closing the input stream.
I think we may need something like this:
```scala
def iterator = {
// an empty Iterator that closes an InputStream when it is iterate... | 1de7f10731de70e05018d273380547b7dc8b4a91 | db471730292f9e733bc9ec8bfbbe05d5811897ed | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
index 6130a71a2..9061c801d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamWriter.scala"
] | [
{
"comment": "nit: I think `matchedResources.iterator.flatMap { ` clearer than manually reimplementing via fold.\r\n\r\nNow that I think about this though, we are never closing the input stream.\r\n\r\nI think we may need something like this:\r\n\r\n```scala\r\ndef iterator = {\r\n // an empty Iterator that cl... | true | ||
twitter/scalding | 1,965 | issue_to_patch | Improve CI | follow on to #1963
1. there was a duplicated version string (2.11 twice vs 2.12)
2. we were never testing scalding-beam 😬
3. use https://github.com/coursier/cache-action and remove the old caching approach.
cc @nownikhil | 6ad1d72c547baac59e336ae3b4f11015e2fe1c27 | 4f2dc94f4227876607203bcf57d9f5b211c066d7 | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index 40ebe594c..d2426e2e9 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -33,7 +33,7 @@ jobs:
BUILD: "base"
TEST_TARGET: "scalding-avro scalding-hraven scalding-commons scalding-parquet scalding-parquet-... | [
".github/workflows/CI.yml",
"scripts/packDeps.sh",
"scripts/run_test.sh",
"scripts/unpackDeps.sh"
] | [] | diff --git a/scripts/run_test.sh b/scripts/run_test.sh
index 4b78f22d8..bd85b4cf9 100755
--- a/scripts/run_test.sh
+++ b/scripts/run_test.sh
@@ -5,7 +5,6 @@ BASE_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )"/.. && pwd )"
cd $BASE_DIR
export JVM_OPTS="-XX:+CMSClassUnloadingEnabled -XX:+UseConcMarkSweepGC -XX:Reserve... | true | |
twitter/scalding | 1,964 | issue_to_patch | Update README.md | Fixing CI badge link complementing fixes in https://github.com/twitter/scalding/pull/1963 | 66f4a50e086bbbb443abd8c3e7f2c0d1f8993f81 | a598c2c844e834acc613d104f5e0147299c3fedb | diff --git a/README.md b/README.md
index c1610cff7..9e6242c4a 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
# Scalding
-[](http://travis-ci.org/twitter/scalding)
+[![Build status](https://github.com/twitter/scalding/actions/workflows... | [
"README.md"
] | [] | true | ||
twitter/scalding | 1,963 | issue_to_patch | Add GitHub CI (migrate from Travis CI) | 
Missing features from travis (follow-up PR work):
1. build cache -- I think this is fair for a followup PR since it's not essential.
2. Scala lib resolution on scald.rb i... | 1de7f10731de70e05018d273380547b7dc8b4a91 | c9c2f359e3188a87eb7dead7ff3677af861d8bda | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
new file mode 100644
index 000000000..40ebe594c
--- /dev/null
+++ b/.github/workflows/CI.yml
@@ -0,0 +1,133 @@
+name: CI
+
+# run on all branches
+on:
+ pull_request:
+ branches:
+ - '**'
+ push:
+ branches:
+ - '**'
+
+jobs:
+ run-base... | [
".github/workflows/CI.yml",
".travis.yml",
"scripts/build_assembly_no_test.sh",
"scripts/common.sh",
"scripts/packDeps.sh",
"scripts/run_test.sh",
"scripts/testValidator.sh",
"scripts/test_execution_tutorial.sh",
"scripts/test_matrix_tutorials.sh",
"scripts/test_repl_tutorial.sh",
"scripts/test_... | [
{
"comment": "this will be removed, just for testing on my fork",
"path": ".github/workflows/CI.yml",
"hunk": "@@ -0,0 +1,129 @@\n+name: CI\n+\n+on:\n+ pull_request:\n+ branches:\n+ - develop\n+ - master\n+ - main\n+ push:\n+ branches:\n+ - add_github_ci",
"resolving_sha"... | diff --git a/scripts/build_assembly_no_test.sh b/scripts/build_assembly_no_test.sh
index e5dc766ee..3bf70c71e 100755
--- a/scripts/build_assembly_no_test.sh
+++ b/scripts/build_assembly_no_test.sh
@@ -1,5 +1,5 @@
-#!/bin/bash
-set -e # first error should stop execution of this script
+#!/bin/bash -exv
+
# Identify the... | true | |
twitter/scalding | 1,963 | comment_to_fix | Add GitHub CI (migrate from Travis CI) | this will be removed, just for testing on my fork | 1de7f10731de70e05018d273380547b7dc8b4a91 | c9c2f359e3188a87eb7dead7ff3677af861d8bda | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
new file mode 100644
index 000000000..40ebe594c
--- /dev/null
+++ b/.github/workflows/CI.yml
@@ -0,0 +1,133 @@
+name: CI
+
+# run on all branches
+on:
+ pull_request:
+ branches:
+ - '**'
+ push:
+ branches:
+ - '**'
+
+jobs:
+ run-base... | [
".github/workflows/CI.yml"
] | [
{
"comment": "this will be removed, just for testing on my fork",
"path": ".github/workflows/CI.yml",
"hunk": "@@ -0,0 +1,129 @@\n+name: CI\n+\n+on:\n+ pull_request:\n+ branches:\n+ - develop\n+ - master\n+ - main\n+ push:\n+ branches:\n+ - add_github_ci",
"resolving_sha"... | true | ||
twitter/scalding | 1,963 | comment_to_fix | Add GitHub CI (migrate from Travis CI) | something about bash isn't escaping properly with the `&&` command chaining need to fix it. | 1de7f10731de70e05018d273380547b7dc8b4a91 | c9c2f359e3188a87eb7dead7ff3677af861d8bda | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
new file mode 100644
index 000000000..40ebe594c
--- /dev/null
+++ b/.github/workflows/CI.yml
@@ -0,0 +1,133 @@
+name: CI
+
+# run on all branches
+on:
+ pull_request:
+ branches:
+ - '**'
+ push:
+ branches:
+ - '**'
+
+jobs:
+ run-base... | [
".github/workflows/CI.yml"
] | [
{
"comment": "something about bash isn't escaping properly with the `&&` command chaining need to fix it.",
"path": ".github/workflows/CI.yml",
"hunk": "@@ -0,0 +1,129 @@\n+name: CI\n+\n+on:\n+ pull_request:\n+ branches:\n+ - develop\n+ - master\n+ - main\n+ push:\n+ branches:\n+ ... | true | ||
twitter/scalding | 1,960 | issue_to_patch | Use BeamJoiner function in `planMapGroup` | User job was OOMing at groupBy step and the job finished after making this change. | e9aaa99cfd845fa4a103c1baf8db85368279ace8 | 43ce1048810a9e58eb5bf495089ae8e8f438421a | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.scala
index 1cb2b31ae..903c2ad4a 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.scala
+++ b/scalding-beam/src/m... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [] | true | ||
twitter/scalding | 1,959 | issue_to_patch | scalding-beam custom join functions | Following the discussion #1958, attempting a custom join implementation for Beam backend. | de2c7c20d7a36a209b81fefa0408a74ef947833c | 2007dc25fadc0fc988df67edcec44c60e49c68fc | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.scala
new file mode 100644
index 000000000..1cb2b31ae
--- /dev/null
+++ b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.sca... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "Questionable, please review.",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamJoiner.scala",
"hunk": "@@ -0,0 +1,283 @@\n+package com.twitter.scalding.beam_backend\n+\n+import com.twitter.scalding.serialization.Externalizer\n+import com.twitter.scalding.typed.J... | true | ||
twitter/scalding | 1,957 | issue_to_patch | scalding-beam improvements on resulting DAG | - Implements missing Merge and Fork
- Improvement to have names and better grouping of some operations so these show as fewer boxes on the Dataflow UI. These boxes can be expanded if needed to show the inner steps.
- Caching of runs of BeamOp now means PCollection used by multiple steps don't recompute. | b0ba993ac817e6b1e52126e8b1cfb1054cc00dad | f44eb68cbaeec488f2c08fac5bd6ab1e2a9c1952 | diff --git a/build.sbt b/build.sbt
index 142a5d631..8e2fd3e84 100644
--- a/build.sbt
+++ b/build.sbt
@@ -339,7 +339,9 @@ lazy val scaldingBeam = module("beam")
"org.apache.beam" % "beam-runners-direct-java" % beamVersion % "test",
// Including this dependency since scalding configuration depends on hadoop... | [
"build.sbt",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "we don't need the explicit `apply` do we ? `tuple(HashJoinTransform(keyCoder, joiner))` would work, no?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -176,6 +180,54 @@ object BeamOp extends Serializable {\n }\n }\n \n+ final cas... | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index c3d2b2533..ef83f4622 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | true | |
twitter/scalding | 1,957 | comment_to_fix | scalding-beam improvements on resulting DAG | do we want to add `require(operations.size > 1, s"expected at least two items in Merged: $operations")` just to guard against a refactoring creating suboptimal plans here? | b0ba993ac817e6b1e52126e8b1cfb1054cc00dad | f44eb68cbaeec488f2c08fac5bd6ab1e2a9c1952 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 04ed23e62..af9e9aa1e 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "do we want to add `require(operations.size > 1, s\"expected at least two items in Merged: $operations\")` just to guard against a refactoring creating suboptimal plans here?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -176,6 +180,54 ... | true | ||
twitter/scalding | 1,953 | issue_to_patch | Migrate scalding from scalariform to scalafmt | Scala community migrated of scalariform towards scalafmt during last couple of years.
In this PR I've replaced scalariform with scalafmt in Scalding, used scalafmt config from algebird, and reformatted scalding repo.
I'm planning to follow up with another PR to enable scalafmt check on CI so it would be impossible to... | a9500fd154fa3a318e2fad063e6d3d15e70e8b17 | 284c476af41733f0cfad532ed2c2ab9091548103 | diff --git a/.scalafmt.conf b/.scalafmt.conf
new file mode 100644
index 0000000000..9f2ddaac28
--- /dev/null
+++ b/.scalafmt.conf
@@ -0,0 +1,8 @@
+version = 3.0.4
+maxColumn = 110
+docstrings.style = Asterisk
+newlines.alwaysBeforeMultilineDef = false
+newlines.penalizeSingleSelectMultiArgList = false
+align.openParenC... | [
".scalafmt.conf",
".travis.yml",
"build.sbt",
"project/plugins.sbt",
"scalding-hadoop-test/src/test/scala/com/twitter/scalding/platform/PlatformTest.scala",
"scalding-hadoop-test/src/test/scala/com/twitter/scalding/platform/TestJobsWithDescriptions.scala",
"scalding-quotation/src/test/scala/com/twitter/... | [] | diff --git a/scalding-hadoop-test/src/test/scala/com/twitter/scalding/platform/PlatformTest.scala b/scalding-hadoop-test/src/test/scala/com/twitter/scalding/platform/PlatformTest.scala
index 002dc6eaae..c26a1f9a56 100644
--- a/scalding-hadoop-test/src/test/scala/com/twitter/scalding/platform/PlatformTest.scala
+++ b/sc... | true | |
twitter/scalding | 1,949 | issue_to_patch | Support `sortedTake` in beam runner | We flatten all the input PriorityQueues and construct a new PQ using the monoid provided.
TESTS: Updated failing unit test | a9500fd154fa3a318e2fad063e6d3d15e70e8b17 | 49af384bd38ec02156cbabd7c65174041d52e37f | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala
index cb6d359367..0ff91be696 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala
+++ b/scalding-beam/... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala",
"scalding-core/src/main/scala/com/twitter/scalding/ReduceO... | [
{
"comment": "I don't *think* this should either compile correctly. I probably wrote some bad code in my example. The `SumAll` type is `TraversableOnce[A] => Iterator[A]` where in this case `A` = `V`. So when you pattern match on `PriorityQueueMonoid[T]` that means `V = PriorityQueue[T]`, so, it shouldn't be `P... | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 5095c0bfde..515aafcc68 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b... | true | |
twitter/scalding | 1,949 | comment_to_fix | Support `sortedTake` in beam runner | note, you may want to use
https://github.com/twitter/scalding/blob/a9500fd154fa3a318e2fad063e6d3d15e70e8b17/scalding-core/src/main/scala/com/twitter/scalding/serialization/Externalizer.scala#L29
(and in other locations), e.g.
```scala
private class SerializableComparator[A](comp: Comparator[A]) extends Comparato... | a9500fd154fa3a318e2fad063e6d3d15e70e8b17 | 49af384bd38ec02156cbabd7c65174041d52e37f | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 207cd560d4..bc27d524ae 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "note, you may want to use \r\nhttps://github.com/twitter/scalding/blob/a9500fd154fa3a318e2fad063e6d3d15e70e8b17/scalding-core/src/main/scala/com/twitter/scalding/serialization/Externalizer.scala#L29\r\n\r\n(and in other locations), e.g.\r\n```scala\r\nprivate class SerializableComparator[A](comp: ... | true | ||
twitter/scalding | 1,949 | comment_to_fix | Support `sortedTake` in beam runner | why do we use both `TopCombineFn` and `pqm`? I'm confused why one isn't enough? Seems like we should be able to return `output` to me.
| a9500fd154fa3a318e2fad063e6d3d15e70e8b17 | 49af384bd38ec02156cbabd7c65174041d52e37f | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 207cd560d4..bc27d524ae 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "why do we use both `TopCombineFn` and `pqm`? I'm confused why one isn't enough? Seems like we should be able to return `output` to me.\r\n\r\n",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -61,19 +70,24 @@ object BeamOp extends Seriali... | true | ||
twitter/scalding | 1,949 | comment_to_fix | Support `sortedTake` in beam runner | I think you can lift this out of the `apply` method to reduce allocations. We shouldn't allocate a new one on every input should we? | a9500fd154fa3a318e2fad063e6d3d15e70e8b17 | 49af384bd38ec02156cbabd7c65174041d52e37f | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 207cd560d4..bc27d524ae 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "I think you can lift this out of the `apply` method to reduce allocations. We shouldn't allocate a new one on every input should we?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -59,6 +70,30 @@ object BeamOp extends Serializable {\n ... | true | ||
twitter/scalding | 1,948 | issue_to_patch | Adds support for blob, tinyint and timestamp to TypedJDBCSource | 767a9457cf763f84b02f50a687bf6f75d67cb9db | 1bfc5c6fa2ae7f649a304ae16eeef781ce705491 | diff --git a/scalding-db/src/main/scala/com/twitter/scalding/db/ColumnDefiner.scala b/scalding-db/src/main/scala/com/twitter/scalding/db/ColumnDefiner.scala
index a6cc0187cc..8063cd0002 100644
--- a/scalding-db/src/main/scala/com/twitter/scalding/db/ColumnDefiner.scala
+++ b/scalding-db/src/main/scala/com/twitter/scald... | [
"scalding-db/src/main/scala/com/twitter/scalding/db/ColumnDefiner.scala",
"scalding-db/src/main/scala/com/twitter/scalding/db/macros/impl/ColumnDefinitionProviderImpl.scala",
"scalding-db/src/main/scala/com/twitter/scalding/db/macros/impl/handler/BlobTypeHandler.scala",
"scalding-db/src/test/scala/com/twitter... | [
{
"comment": "why use `assert` here but return `Success`?\r\n\r\nWhy not:\r\n```scala\r\nif (defaultValue.nonEmpty || annotationInfo.nonEmpty) Failure(new Exception(s\"invalid args: defaultValue = $defaultValue annotationInfo = $annotationInfo\"))\r\nelse Success(List(ColumnFormat(c)(accessorTree, \"BLOB\", Non... | diff --git a/scalding-db/src/test/scala/com/twitter/scalding/db/macros/MacrosUnitTests.scala b/scalding-db/src/test/scala/com/twitter/scalding/db/macros/MacrosUnitTests.scala
index c27ace132d..0919af1a1d 100644
--- a/scalding-db/src/test/scala/com/twitter/scalding/db/macros/MacrosUnitTests.scala
+++ b/scalding-db/src/t... | true | ||
twitter/scalding | 1,951 | issue_to_patch | Fixes Execution#recoverWith when partial function doesn't match | When a recoverWith was used and the function didn't match a MatchError would be thrown. This is due to calling apply on the partial function instead of passing as a partial to the recoverWith of Future.
The provided unit test fails without this fix. | e1988110f61a2849a077b92250e1e1d58511cd90 | 28537cb86b1a45d6ce0d5df01994e4ba367c0264 | diff --git a/scalding-core/src/main/scala/com/twitter/scalding/Execution.scala b/scalding-core/src/main/scala/com/twitter/scalding/Execution.scala
index ef58ed34fa..a049bb1a7f 100644
--- a/scalding-core/src/main/scala/com/twitter/scalding/Execution.scala
+++ b/scalding-core/src/main/scala/com/twitter/scalding/Execution... | [
"scalding-core/src/main/scala/com/twitter/scalding/Execution.scala",
"scalding-core/src/test/scala/com/twitter/scalding/ExecutionTest.scala"
] | [
{
"comment": "This here was causing the MatchError to be thrown.",
"path": "scalding-core/src/main/scala/com/twitter/scalding/Execution.scala",
"hunk": "@@ -521,16 +521,17 @@ object Execution {\n fut\n .map {v => (v, CancellationHandler.empty) } // map this to the right shape\n... | diff --git a/scalding-core/src/test/scala/com/twitter/scalding/ExecutionTest.scala b/scalding-core/src/test/scala/com/twitter/scalding/ExecutionTest.scala
index bddda3f9c9..cd32e127f2 100644
--- a/scalding-core/src/test/scala/com/twitter/scalding/ExecutionTest.scala
+++ b/scalding-core/src/test/scala/com/twitter/scaldi... | true | |
twitter/scalding | 1,946 | issue_to_patch | Adds Args.empty | Adds a convenient `Args.empty` instance. | 767a9457cf763f84b02f50a687bf6f75d67cb9db | 3ac16e22c8b4ac32e286d1bc7f4d1bf6ab820e4f | diff --git a/scalding-args/src/main/scala/com/twitter/scalding/Args.scala b/scalding-args/src/main/scala/com/twitter/scalding/Args.scala
index 2c487322ae..a069a1ce56 100644
--- a/scalding-args/src/main/scala/com/twitter/scalding/Args.scala
+++ b/scalding-args/src/main/scala/com/twitter/scalding/Args.scala
@@ -25,6 +25,... | [
"scalding-args/src/main/scala/com/twitter/scalding/Args.scala"
] | [] | true | ||
twitter/scalding | 1,945 | issue_to_patch | Add support for cogroups in beam-backend | In this change we are adding support for `HashCoGroup` and `CoGroupedPipe`.
For evaluating HashCoGroup we are creating a ParDo transformation on the larger pipe with smaller pipe as side input.
For evaluating CoGroupedPipe we are using the `MultiJoinFunction` to evaluate the final output iterator.
Also as part of ... | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala
index 62a2a0249..cb6d35936 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala
+++ b/scalding-beam/sr... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamBackend.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backe... | [
{
"comment": "maybe allocate `private[this] val emptyUs: java.lang.Iterable[U] = Seq.empty[U].asJava` so you don't need to allocate multiple times.\r\n\r\nkeep in mind this gets called for each element on the left hand side. Each allocation avoided could be a win. I would do:\r\n\r\n```scala\r\nval it = sideInp... | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 451eb7ee3..5095c0bfd 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | true | |
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | maybe allocate `private[this] val emptyUs: java.lang.Iterable[U] = Seq.empty[U].asJava` so you don't need to allocate multiple times.
keep in mind this gets called for each element on the left hand side. Each allocation avoided could be a win. I would do:
```scala
val it = sideInputMap.get(key) match {
case n... | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala
index b9da0ddd7..ce651929e 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala
+++ b/scalding-b... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala"
] | [
{
"comment": "maybe allocate `private[this] val emptyUs: java.lang.Iterable[U] = Seq.empty[U].asJava` so you don't need to allocate multiple times.\r\n\r\nkeep in mind this gets called for each element on the left hand side. Each allocation avoided could be a win. I would do:\r\n\r\n```scala\r\nval it = sideInp... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | I think `HashJoin` has access to `Ordering[K]` which in some cases will be a serialization of `K` if it is an `OrderedSerialization[K]`. I think you should accept the ordering here to potentially use that serialization.
https://github.com/twitter/scalding/blob/e445ae8d931845ddccec0795f2a0cb61c57d31d7/scalding-beam/s... | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "I think `HashJoin` has access to `Ordering[K]` which in some cases will be a serialization of `K` if it is an `OrderedSerialization[K]`. I think you should accept the ordering here to potentially use that serialization.\r\n\r\nhttps://github.com/twitter/scalding/blob/e445ae8d931845ddccec0795f2a0cb... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | rather than use `asInstanceOf`, which is dangerous: if scomeone refactors we won't get a compilation error at any cast site. Instead have one safe cast function:
```scala
def widenPCollection[A, B >: A](p: PCollection[A]): PCollection[B] = p.asInstanceOf[PCollection[B]]
```
if we have that one function, then we see... | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "rather than use `asInstanceOf`, which is dangerous: if scomeone refactors we won't get a compilation error at any cast site. Instead have one safe cast function:\r\n```scala\r\ndef widenPCollection[A, B >: A](p: PCollection[A]): PCollection[B] = p.asInstanceOf[PCollection[B]]\r\n```\r\nif we have ... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | I think we should have a `TupleCoder` which uses the coder for the key and the value. The key coder may be found from the Ordering. | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "I think we should have a `TupleCoder` which uses the coder for the key and the value. The key coder may be found from the Ordering.",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -85,21 +101,110 @@ object BeamOp extends Serializable {\n... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | we should use the `widenPCollection` function I mentioned. | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "we should use the `widenPCollection` function I mentioned.",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -85,21 +101,110 @@ object BeamOp extends Serializable {\n }\n }\n \n- final case class TransformBeamOp[A, B](source: BeamOp... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | cg has `Ordering[K]`. We should use that to potentially get an `Coder[K]`
https://github.com/twitter/scalding/blob/e445ae8d931845ddccec0795f2a0cb61c57d31d7/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/KryoCoder.scala#L37 | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "cg has `Ordering[K]`. We should use that to potentially get an `Coder[K]` \r\n\r\nhttps://github.com/twitter/scalding/blob/e445ae8d931845ddccec0795f2a0cb61c57d31d7/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/KryoCoder.scala#L37",
"path": "scalding-beam/src/main/scala/com/twi... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | can we add tests with multiple values per keys on left, right, both... and also filtering: where some keys only appear on the right or left.
Also, can we test join, leftJoin, rightJoin, outerJoin since those are the most common join functions? | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 451eb7ee3..5095c0bfd 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | [
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "can we add tests with multiple values per keys on left, right, both... and also filtering: where some keys only appear on the right or left.\r\n\r\nAlso, can we test join, leftJoin, rightJoin, outerJoin since those are the most common join functions?",
"path": "scalding-beam/src/test/scala/com... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | can we test with multiple values per key? | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 451eb7ee3..5095c0bfd 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | [
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "can we test with multiple values per key?",
"path": "scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala",
"hunk": "@@ -136,6 +136,23 @@ class BeamBackendTests extends FunSuite with BeforeAndAfter {\n )\n }\n \n+ test(\"HashJoin\"){\n+ beamMatche... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | are we not materializing this more than one? Should we not do something like:
```scala
private[this] var mapRight: java.util.Map[K, java.lang.Iterable[U]] = null
@ProcessElement
def processElement(c: DoFn[(K, V), (K, W)]#ProcessContext): Unit = {
if (mapRight == null) {
mapRight = c.sideIn... | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala
index b9da0ddd7..ce651929e 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala
+++ b/scalding-b... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamFunctions.scala"
] | [
{
"comment": "are we not materializing this more than one? Should we not do something like:\r\n```scala\r\nprivate[this] var mapRight: java.util.Map[K, java.lang.Iterable[U]] = null\r\n\r\n @ProcessElement\r\n def processElement(c: DoFn[(K, V), (K, W)]#ProcessContext): Unit = {\r\n if (mapRight == nu... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | how about a test with hashLeftJoin also? | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala b/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
index 451eb7ee3..5095c0bfd 100644
--- a/scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala
+++ b/s... | [
"scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala"
] | [
{
"comment": "how about a test with hashLeftJoin also?",
"path": "scalding-beam/src/test/scala/com/twitter/scalding/beam_backend/BeamBackendTests.scala",
"hunk": "@@ -136,6 +136,23 @@ class BeamBackendTests extends FunSuite with BeforeAndAfter {\n )\n }\n \n+ test(\"HashJoin\"){\n+ beamMatches... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | could we use `widenPCollection` here? | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "could we use `widenPCollection` here?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -85,21 +101,113 @@ object BeamOp extends Serializable {\n }\n }\n \n- final case class TransformBeamOp[A, B](source: BeamOp[A], f: PTransform[PC... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | can't we use `.setCoder(KvCoder.of(keyCoder, kryoCoder))` here? if not, can we comment why? | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "can't we use `.setCoder(KvCoder.of(keyCoder, kryoCoder))` here? if not, can we comment why?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -85,21 +101,113 @@ object BeamOp extends Serializable {\n }\n }\n \n- final case class Tra... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | can't we use `keyCoder` here at all? | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "can't we use `keyCoder` here at all?",
"path": "scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala",
"hunk": "@@ -85,21 +101,113 @@ object BeamOp extends Serializable {\n }\n }\n \n- final case class TransformBeamOp[A, B](source: BeamOp[A], f: PTransform[PCo... | true | ||
twitter/scalding | 1,945 | comment_to_fix | Add support for cogroups in beam-backend | seems like this should take a `Coder[K]` and `Coder[V]` and build a `TupleCoder[K, V]` from those rather than use kryo for the whole thing. At least we can save the serialization tag that the value is a tuple, since we know that statically. | e445ae8d931845ddccec0795f2a0cb61c57d31d7 | 09b594fd7ae6d9cc4a0baf5f60d83e26194d5322 | diff --git a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala b/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
index 6d6898ce8..207cd560d 100644
--- a/scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala
+++ b/scalding-beam/src/main/scala/co... | [
"scalding-beam/src/main/scala/com/twitter/scalding/beam_backend/BeamOp.scala"
] | [
{
"comment": "seems like this should take a `Coder[K]` and `Coder[V]` and build a `TupleCoder[K, V]` from those rather than use kryo for the whole thing. At least we can save the serialization tag that the value is a tuple, since we know that statically.",
"path": "scalding-beam/src/main/scala/com/twitter/s... | true |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.