stevedlawrence commented on a change in pull request #446:
URL: https://github.com/apache/incubator-daffodil/pull/446#discussion_r511001995
##########
File path: build.sbt
##########
@@ -216,7 +218,7 @@ lazy val libManagedSettings = Seq(
schemas.map { schema =>
val out = outdir / "org" / "apache" / "daffodil" / "xsd" /
schema.getName
IO.copyFile(schema, out)
- stream.log.info("Generated %s".format(out))
+ stream.log.info("generated %s".format(out))
Review comment:
This was done just to make output more consistent with sbt output.
They've starting using lowercase in lots of places.
##########
File path: daffodil-cli/bin.NOTICE
##########
@@ -106,18 +113,11 @@ Scala Parser Combinators
(lib/org.scala-lang.modules.scala-parser-combinators_2.
LAMP/EPFL (https://lamp.epfl.ch/) and
Lightbend, Inc. (https://www.lightbend.com/).
-Scala XML (lib/org.scala-lang.modules.scala-xml_2.12-1.1.0.jar)
+Scala XML (lib/org.scala-lang.modules.scala-xml_2.12-1.3.0.jar)
scala-xml
Copyright (c) 2002-2020 EPFL
Copyright (c) 2011-2020 Lightbend, Inc.
scala-xml includes software developed at
LAMP/EPFL (https://lamp.epfl.ch/) and
Lightbend, Inc. (https://www.lightbend.com/).
-
-Woodstox (lib/com.fasterxml.woodstox.woodstox-core-5.1.0.jar)
- This product currently only contains code developed by authors
- of specific components, as identified by the source code files.
-
- Since product implements StAX API, it has dependencies to StAX API
- classes.
Review comment:
This appears to be a notice that was only applied to old version of
woodstox. Newer versions do not include a notice that I can find.
##########
File path: daffodil-cli/src/main/scala/org/apache/daffodil/Main.scala
##########
@@ -102,24 +104,6 @@ import org.rogach.scallop.ArgType
import org.rogach.scallop.ScallopOption
import org.rogach.scallop.ValueConverter
-class NullOutputStream extends OutputStream {
- override def close(): Unit = {
- //do nothing
- }
- override def flush(): Unit = {
- //do nothing
- }
- override def write(b: Array[Byte]): Unit = {
- //do nothing
- }
- override def write(b: Array[Byte], off: Int, len: Int): Unit = {
- //do nothing
- }
- override def write(b: Int): Unit = {
- //do nothing
- }
-}
-
Review comment:
Use Apache IO NullOutputStream instead of creating our own.
##########
File path:
daffodil-test/src/test/resources/org/apache/daffodil/section13/text_number_props/TextNumberProps.tdml
##########
@@ -4288,4 +4288,26 @@
</tdml:parserTestCase>
+ <!--
+ Test Name: textStandardFloatPatternNoSeparators2
+ Schema: textNumberPattern
+ Root: tnp96
+ Purpose: This test shows that when the textNumberPattern contains
+ no grouping/decimal separators, when unparsing, ICU will round the
+ result to match the pattern.
+ -->
+
+ <tdml:unparserTestCase name="textStandardFloatPatternNoSeparators2"
root="tnp96" model="textNumberPattern" roundTrip="none">
+
+ <tdml:document>
+ <tdml:documentPart type="text">11</tdml:documentPart>
+ </tdml:document>
+ <tdml:infoset>
+ <tdml:dfdlInfoset>
+ <tnp96>10.9</tnp96>
+ </tdml:dfdlInfoset>
+ </tdml:infoset>
+
+ </tdml:unparserTestCase>
Review comment:
ICU changed the behavior for how floating point numbers are
formatted/unparsed when the textNumberPattern does not contain a decimal
separator. The previous behavior was to unparse the number with the separator.
New behavior rounds the floating point number. Note that when parsing it will
still accept floating point numbers if the pattern does not contain a decimal
separator. This means patterns without decimal separators and the data contains
a deciaml point may not round trip correct. See
textStandardFloatPatternNoSeparators1 above for the parse case.
##########
File path:
daffodil-test/src/test/resources/org/apache/daffodil/section17/calc_value_properties/outputValueCalc2.tdml
##########
@@ -494,7 +494,7 @@
</tdml:dfdlInfoset>
</tdml:infoset>
<tdml:document>
- <tdml:documentPart type="text">XYZ5.0X038.0Y41.23Z008</tdml:documentPart>
+ <tdml:documentPart type="text">XYZ5X038.0Y41.23Z008</tdml:documentPart>
Review comment:
ICU change behabior here, the text pattern here has option decimal point
for the first number (5.0e03). The pattern says the .0 is optional, and so is
now correctly left off when the value is zero.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
