embulk-parser-jsonl 0.0.1 → 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/.gitignore +3 -0
- data/CHANGELOG.md +7 -0
- data/README.md +3 -3
- data/build.gradle +79 -0
- data/embulk-parser-jsonl.gemspec +1 -1
- data/gradle/wrapper/gradle-wrapper.jar +0 -0
- data/gradle/wrapper/gradle-wrapper.properties +6 -0
- data/gradlew +164 -0
- data/gradlew.bat +90 -0
- data/lib/embulk/guess/jsonl.rb +16 -67
- data/lib/embulk/parser/jsonl.rb +3 -67
- data/settings.gradle +1 -0
- data/src/main/java/org/embulk/parser/jsonl/JsonlParserPlugin.java +282 -0
- data/src/test/java/org/embulk/parser/jsonl/TestJsonlParserPlugin.java +228 -0
- metadata +28 -18
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA1:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 973fe894b7704f01da0d8d0ba2b93eaa3804ea7d
|
|
4
|
+
data.tar.gz: fe6ca73d3100595bd64e95e0e08269490671314e
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 74ad34c4f29980e75f36ac88b9160fd784a975a83b906d6d5c30738edab617211c0653f22d0a9c01ad533bbcc92394b3ceccad20dcc8805b1627298695b5d6ac
|
|
7
|
+
data.tar.gz: dd611c8bff080d4ca1c61a8c071a5baeb2d982b89b2812457af67f9ecaacd5c3e17b42c230356087c26ed53ecb51a7a0728a1fde14c67a441974483a53fcd58b
|
data/.gitignore
CHANGED
data/CHANGELOG.md
ADDED
data/README.md
CHANGED
|
@@ -10,7 +10,7 @@ TODO: Write short description here and embulk-parser-jsonl.gemspec file.
|
|
|
10
10
|
## Configuration
|
|
11
11
|
|
|
12
12
|
- **type**: specify this parser as jsonl
|
|
13
|
-
- **
|
|
13
|
+
- **columns**: specify column name and type (array, required)
|
|
14
14
|
|
|
15
15
|
## Example
|
|
16
16
|
|
|
@@ -19,7 +19,7 @@ in:
|
|
|
19
19
|
type: any file input plugin type
|
|
20
20
|
parser:
|
|
21
21
|
type: jsonl
|
|
22
|
-
|
|
22
|
+
columns:
|
|
23
23
|
- {name: first_name, type: string}
|
|
24
24
|
- {name: last_name, type: string}
|
|
25
25
|
- {name: age, type: long}
|
|
@@ -35,5 +35,5 @@ $ embulk guess -g jsonl config.yml -o guessed.yml
|
|
|
35
35
|
## Build
|
|
36
36
|
|
|
37
37
|
```
|
|
38
|
-
$
|
|
38
|
+
$ ./gradlew gem classpath
|
|
39
39
|
```
|
data/build.gradle
ADDED
|
@@ -0,0 +1,79 @@
|
|
|
1
|
+
plugins {
|
|
2
|
+
id "com.jfrog.bintray" version "1.1"
|
|
3
|
+
id "com.github.jruby-gradle.base" version "0.1.5"
|
|
4
|
+
id "java"
|
|
5
|
+
id "jacoco"
|
|
6
|
+
}
|
|
7
|
+
import com.github.jrubygradle.JRubyExec
|
|
8
|
+
repositories {
|
|
9
|
+
mavenCentral()
|
|
10
|
+
jcenter()
|
|
11
|
+
}
|
|
12
|
+
configurations {
|
|
13
|
+
provided
|
|
14
|
+
}
|
|
15
|
+
|
|
16
|
+
version = "0.1.0"
|
|
17
|
+
|
|
18
|
+
compileJava.options.encoding = 'UTF-8' // source encoding
|
|
19
|
+
sourceCompatibility = 1.7
|
|
20
|
+
targetCompatibility = 1.7
|
|
21
|
+
|
|
22
|
+
dependencies {
|
|
23
|
+
compile "org.embulk:embulk-core:0.8.2"
|
|
24
|
+
provided "org.embulk:embulk-core:0.8.2"
|
|
25
|
+
|
|
26
|
+
testCompile "junit:junit:4.+"
|
|
27
|
+
testCompile "org.embulk:embulk-core:0.8.2:tests"
|
|
28
|
+
testCompile "org.embulk:embulk-standards:0.8.2"
|
|
29
|
+
}
|
|
30
|
+
|
|
31
|
+
task classpath(type: Copy, dependsOn: ["jar"]) {
|
|
32
|
+
doFirst { file("classpath").deleteDir() }
|
|
33
|
+
from (configurations.runtime - configurations.provided + files(jar.archivePath))
|
|
34
|
+
into "classpath"
|
|
35
|
+
}
|
|
36
|
+
clean { delete "classpath" }
|
|
37
|
+
|
|
38
|
+
task gem(type: JRubyExec, dependsOn: ["gemspec", "classpath"]) {
|
|
39
|
+
jrubyArgs "-rrubygems/gem_runner", "-eGem::GemRunner.new.run(ARGV)", "build"
|
|
40
|
+
script "${project.name}.gemspec"
|
|
41
|
+
doLast { ant.move(file: "${project.name}-${project.version}.gem", todir: "pkg") }
|
|
42
|
+
}
|
|
43
|
+
|
|
44
|
+
task gemPush(type: JRubyExec, dependsOn: ["gem"]) {
|
|
45
|
+
jrubyArgs "-rrubygems/gem_runner", "-eGem::GemRunner.new.run(ARGV)", "push"
|
|
46
|
+
script "pkg/${project.name}-${project.version}.gem"
|
|
47
|
+
}
|
|
48
|
+
|
|
49
|
+
task "package"(dependsOn: ["gemspec", "classpath"]) << {
|
|
50
|
+
println "> Build succeeded."
|
|
51
|
+
println "> You can run embulk with '-L ${file(".").absolutePath}' argument."
|
|
52
|
+
}
|
|
53
|
+
|
|
54
|
+
task gemspec {
|
|
55
|
+
ext.gemspecFile = file("${project.name}.gemspec")
|
|
56
|
+
inputs.file "build.gradle"
|
|
57
|
+
outputs.file gemspecFile
|
|
58
|
+
doLast { gemspecFile.write($/
|
|
59
|
+
Gem::Specification.new do |spec|
|
|
60
|
+
spec.name = "${project.name}"
|
|
61
|
+
spec.version = "${project.version}"
|
|
62
|
+
spec.authors = ["Shunsuke Mikami"]
|
|
63
|
+
spec.summary = "Jsonl parser plugin for Embulk"
|
|
64
|
+
spec.description = "Parses Jsonl files read by other file input plugins."
|
|
65
|
+
spec.email = ["shun0102@gmail.com"]
|
|
66
|
+
spec.licenses = ["MIT"]
|
|
67
|
+
spec.homepage = "https://github.com/shun0102/embulk-parser-jsonl"
|
|
68
|
+
|
|
69
|
+
spec.files = `git ls-files`.split("\n") + Dir["classpath/*.jar"]
|
|
70
|
+
spec.test_files = spec.files.grep(%r{^(test|spec)/})
|
|
71
|
+
spec.require_paths = ["lib"]
|
|
72
|
+
|
|
73
|
+
spec.add_development_dependency 'bundler', ['~> 1.0']
|
|
74
|
+
spec.add_development_dependency 'rake', ['~> 10.0']
|
|
75
|
+
end
|
|
76
|
+
/$)
|
|
77
|
+
}
|
|
78
|
+
}
|
|
79
|
+
clean { delete "${project.name}.gemspec" }
|
data/embulk-parser-jsonl.gemspec
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
|
|
2
2
|
Gem::Specification.new do |spec|
|
|
3
3
|
spec.name = "embulk-parser-jsonl"
|
|
4
|
-
spec.version = "0.0
|
|
4
|
+
spec.version = "0.1.0"
|
|
5
5
|
spec.authors = ["Shunsuke Mikami"]
|
|
6
6
|
spec.summary = "Jsonl parser plugin for Embulk"
|
|
7
7
|
spec.description = "Parses Jsonl files read by other file input plugins."
|
|
Binary file
|
data/gradlew
ADDED
|
@@ -0,0 +1,164 @@
|
|
|
1
|
+
#!/usr/bin/env bash
|
|
2
|
+
|
|
3
|
+
##############################################################################
|
|
4
|
+
##
|
|
5
|
+
## Gradle start up script for UN*X
|
|
6
|
+
##
|
|
7
|
+
##############################################################################
|
|
8
|
+
|
|
9
|
+
# Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
|
|
10
|
+
DEFAULT_JVM_OPTS=""
|
|
11
|
+
|
|
12
|
+
APP_NAME="Gradle"
|
|
13
|
+
APP_BASE_NAME=`basename "$0"`
|
|
14
|
+
|
|
15
|
+
# Use the maximum available, or set MAX_FD != -1 to use that value.
|
|
16
|
+
MAX_FD="maximum"
|
|
17
|
+
|
|
18
|
+
warn ( ) {
|
|
19
|
+
echo "$*"
|
|
20
|
+
}
|
|
21
|
+
|
|
22
|
+
die ( ) {
|
|
23
|
+
echo
|
|
24
|
+
echo "$*"
|
|
25
|
+
echo
|
|
26
|
+
exit 1
|
|
27
|
+
}
|
|
28
|
+
|
|
29
|
+
# OS specific support (must be 'true' or 'false').
|
|
30
|
+
cygwin=false
|
|
31
|
+
msys=false
|
|
32
|
+
darwin=false
|
|
33
|
+
case "`uname`" in
|
|
34
|
+
CYGWIN* )
|
|
35
|
+
cygwin=true
|
|
36
|
+
;;
|
|
37
|
+
Darwin* )
|
|
38
|
+
darwin=true
|
|
39
|
+
;;
|
|
40
|
+
MINGW* )
|
|
41
|
+
msys=true
|
|
42
|
+
;;
|
|
43
|
+
esac
|
|
44
|
+
|
|
45
|
+
# For Cygwin, ensure paths are in UNIX format before anything is touched.
|
|
46
|
+
if $cygwin ; then
|
|
47
|
+
[ -n "$JAVA_HOME" ] && JAVA_HOME=`cygpath --unix "$JAVA_HOME"`
|
|
48
|
+
fi
|
|
49
|
+
|
|
50
|
+
# Attempt to set APP_HOME
|
|
51
|
+
# Resolve links: $0 may be a link
|
|
52
|
+
PRG="$0"
|
|
53
|
+
# Need this for relative symlinks.
|
|
54
|
+
while [ -h "$PRG" ] ; do
|
|
55
|
+
ls=`ls -ld "$PRG"`
|
|
56
|
+
link=`expr "$ls" : '.*-> \(.*\)$'`
|
|
57
|
+
if expr "$link" : '/.*' > /dev/null; then
|
|
58
|
+
PRG="$link"
|
|
59
|
+
else
|
|
60
|
+
PRG=`dirname "$PRG"`"/$link"
|
|
61
|
+
fi
|
|
62
|
+
done
|
|
63
|
+
SAVED="`pwd`"
|
|
64
|
+
cd "`dirname \"$PRG\"`/" >&-
|
|
65
|
+
APP_HOME="`pwd -P`"
|
|
66
|
+
cd "$SAVED" >&-
|
|
67
|
+
|
|
68
|
+
CLASSPATH=$APP_HOME/gradle/wrapper/gradle-wrapper.jar
|
|
69
|
+
|
|
70
|
+
# Determine the Java command to use to start the JVM.
|
|
71
|
+
if [ -n "$JAVA_HOME" ] ; then
|
|
72
|
+
if [ -x "$JAVA_HOME/jre/sh/java" ] ; then
|
|
73
|
+
# IBM's JDK on AIX uses strange locations for the executables
|
|
74
|
+
JAVACMD="$JAVA_HOME/jre/sh/java"
|
|
75
|
+
else
|
|
76
|
+
JAVACMD="$JAVA_HOME/bin/java"
|
|
77
|
+
fi
|
|
78
|
+
if [ ! -x "$JAVACMD" ] ; then
|
|
79
|
+
die "ERROR: JAVA_HOME is set to an invalid directory: $JAVA_HOME
|
|
80
|
+
|
|
81
|
+
Please set the JAVA_HOME variable in your environment to match the
|
|
82
|
+
location of your Java installation."
|
|
83
|
+
fi
|
|
84
|
+
else
|
|
85
|
+
JAVACMD="java"
|
|
86
|
+
which java >/dev/null 2>&1 || die "ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
|
|
87
|
+
|
|
88
|
+
Please set the JAVA_HOME variable in your environment to match the
|
|
89
|
+
location of your Java installation."
|
|
90
|
+
fi
|
|
91
|
+
|
|
92
|
+
# Increase the maximum file descriptors if we can.
|
|
93
|
+
if [ "$cygwin" = "false" -a "$darwin" = "false" ] ; then
|
|
94
|
+
MAX_FD_LIMIT=`ulimit -H -n`
|
|
95
|
+
if [ $? -eq 0 ] ; then
|
|
96
|
+
if [ "$MAX_FD" = "maximum" -o "$MAX_FD" = "max" ] ; then
|
|
97
|
+
MAX_FD="$MAX_FD_LIMIT"
|
|
98
|
+
fi
|
|
99
|
+
ulimit -n $MAX_FD
|
|
100
|
+
if [ $? -ne 0 ] ; then
|
|
101
|
+
warn "Could not set maximum file descriptor limit: $MAX_FD"
|
|
102
|
+
fi
|
|
103
|
+
else
|
|
104
|
+
warn "Could not query maximum file descriptor limit: $MAX_FD_LIMIT"
|
|
105
|
+
fi
|
|
106
|
+
fi
|
|
107
|
+
|
|
108
|
+
# For Darwin, add options to specify how the application appears in the dock
|
|
109
|
+
if $darwin; then
|
|
110
|
+
GRADLE_OPTS="$GRADLE_OPTS \"-Xdock:name=$APP_NAME\" \"-Xdock:icon=$APP_HOME/media/gradle.icns\""
|
|
111
|
+
fi
|
|
112
|
+
|
|
113
|
+
# For Cygwin, switch paths to Windows format before running java
|
|
114
|
+
if $cygwin ; then
|
|
115
|
+
APP_HOME=`cygpath --path --mixed "$APP_HOME"`
|
|
116
|
+
CLASSPATH=`cygpath --path --mixed "$CLASSPATH"`
|
|
117
|
+
|
|
118
|
+
# We build the pattern for arguments to be converted via cygpath
|
|
119
|
+
ROOTDIRSRAW=`find -L / -maxdepth 1 -mindepth 1 -type d 2>/dev/null`
|
|
120
|
+
SEP=""
|
|
121
|
+
for dir in $ROOTDIRSRAW ; do
|
|
122
|
+
ROOTDIRS="$ROOTDIRS$SEP$dir"
|
|
123
|
+
SEP="|"
|
|
124
|
+
done
|
|
125
|
+
OURCYGPATTERN="(^($ROOTDIRS))"
|
|
126
|
+
# Add a user-defined pattern to the cygpath arguments
|
|
127
|
+
if [ "$GRADLE_CYGPATTERN" != "" ] ; then
|
|
128
|
+
OURCYGPATTERN="$OURCYGPATTERN|($GRADLE_CYGPATTERN)"
|
|
129
|
+
fi
|
|
130
|
+
# Now convert the arguments - kludge to limit ourselves to /bin/sh
|
|
131
|
+
i=0
|
|
132
|
+
for arg in "$@" ; do
|
|
133
|
+
CHECK=`echo "$arg"|egrep -c "$OURCYGPATTERN" -`
|
|
134
|
+
CHECK2=`echo "$arg"|egrep -c "^-"` ### Determine if an option
|
|
135
|
+
|
|
136
|
+
if [ $CHECK -ne 0 ] && [ $CHECK2 -eq 0 ] ; then ### Added a condition
|
|
137
|
+
eval `echo args$i`=`cygpath --path --ignore --mixed "$arg"`
|
|
138
|
+
else
|
|
139
|
+
eval `echo args$i`="\"$arg\""
|
|
140
|
+
fi
|
|
141
|
+
i=$((i+1))
|
|
142
|
+
done
|
|
143
|
+
case $i in
|
|
144
|
+
(0) set -- ;;
|
|
145
|
+
(1) set -- "$args0" ;;
|
|
146
|
+
(2) set -- "$args0" "$args1" ;;
|
|
147
|
+
(3) set -- "$args0" "$args1" "$args2" ;;
|
|
148
|
+
(4) set -- "$args0" "$args1" "$args2" "$args3" ;;
|
|
149
|
+
(5) set -- "$args0" "$args1" "$args2" "$args3" "$args4" ;;
|
|
150
|
+
(6) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" ;;
|
|
151
|
+
(7) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" ;;
|
|
152
|
+
(8) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" "$args7" ;;
|
|
153
|
+
(9) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" "$args7" "$args8" ;;
|
|
154
|
+
esac
|
|
155
|
+
fi
|
|
156
|
+
|
|
157
|
+
# Split up the JVM_OPTS And GRADLE_OPTS values into an array, following the shell quoting and substitution rules
|
|
158
|
+
function splitJvmOpts() {
|
|
159
|
+
JVM_OPTS=("$@")
|
|
160
|
+
}
|
|
161
|
+
eval splitJvmOpts $DEFAULT_JVM_OPTS $JAVA_OPTS $GRADLE_OPTS
|
|
162
|
+
JVM_OPTS[${#JVM_OPTS[*]}]="-Dorg.gradle.appname=$APP_BASE_NAME"
|
|
163
|
+
|
|
164
|
+
exec "$JAVACMD" "${JVM_OPTS[@]}" -classpath "$CLASSPATH" org.gradle.wrapper.GradleWrapperMain "$@"
|
data/gradlew.bat
ADDED
|
@@ -0,0 +1,90 @@
|
|
|
1
|
+
@if "%DEBUG%" == "" @echo off
|
|
2
|
+
@rem ##########################################################################
|
|
3
|
+
@rem
|
|
4
|
+
@rem Gradle startup script for Windows
|
|
5
|
+
@rem
|
|
6
|
+
@rem ##########################################################################
|
|
7
|
+
|
|
8
|
+
@rem Set local scope for the variables with windows NT shell
|
|
9
|
+
if "%OS%"=="Windows_NT" setlocal
|
|
10
|
+
|
|
11
|
+
@rem Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
|
|
12
|
+
set DEFAULT_JVM_OPTS=
|
|
13
|
+
|
|
14
|
+
set DIRNAME=%~dp0
|
|
15
|
+
if "%DIRNAME%" == "" set DIRNAME=.
|
|
16
|
+
set APP_BASE_NAME=%~n0
|
|
17
|
+
set APP_HOME=%DIRNAME%
|
|
18
|
+
|
|
19
|
+
@rem Find java.exe
|
|
20
|
+
if defined JAVA_HOME goto findJavaFromJavaHome
|
|
21
|
+
|
|
22
|
+
set JAVA_EXE=java.exe
|
|
23
|
+
%JAVA_EXE% -version >NUL 2>&1
|
|
24
|
+
if "%ERRORLEVEL%" == "0" goto init
|
|
25
|
+
|
|
26
|
+
echo.
|
|
27
|
+
echo ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
|
|
28
|
+
echo.
|
|
29
|
+
echo Please set the JAVA_HOME variable in your environment to match the
|
|
30
|
+
echo location of your Java installation.
|
|
31
|
+
|
|
32
|
+
goto fail
|
|
33
|
+
|
|
34
|
+
:findJavaFromJavaHome
|
|
35
|
+
set JAVA_HOME=%JAVA_HOME:"=%
|
|
36
|
+
set JAVA_EXE=%JAVA_HOME%/bin/java.exe
|
|
37
|
+
|
|
38
|
+
if exist "%JAVA_EXE%" goto init
|
|
39
|
+
|
|
40
|
+
echo.
|
|
41
|
+
echo ERROR: JAVA_HOME is set to an invalid directory: %JAVA_HOME%
|
|
42
|
+
echo.
|
|
43
|
+
echo Please set the JAVA_HOME variable in your environment to match the
|
|
44
|
+
echo location of your Java installation.
|
|
45
|
+
|
|
46
|
+
goto fail
|
|
47
|
+
|
|
48
|
+
:init
|
|
49
|
+
@rem Get command-line arguments, handling Windowz variants
|
|
50
|
+
|
|
51
|
+
if not "%OS%" == "Windows_NT" goto win9xME_args
|
|
52
|
+
if "%@eval[2+2]" == "4" goto 4NT_args
|
|
53
|
+
|
|
54
|
+
:win9xME_args
|
|
55
|
+
@rem Slurp the command line arguments.
|
|
56
|
+
set CMD_LINE_ARGS=
|
|
57
|
+
set _SKIP=2
|
|
58
|
+
|
|
59
|
+
:win9xME_args_slurp
|
|
60
|
+
if "x%~1" == "x" goto execute
|
|
61
|
+
|
|
62
|
+
set CMD_LINE_ARGS=%*
|
|
63
|
+
goto execute
|
|
64
|
+
|
|
65
|
+
:4NT_args
|
|
66
|
+
@rem Get arguments from the 4NT Shell from JP Software
|
|
67
|
+
set CMD_LINE_ARGS=%$
|
|
68
|
+
|
|
69
|
+
:execute
|
|
70
|
+
@rem Setup the command line
|
|
71
|
+
|
|
72
|
+
set CLASSPATH=%APP_HOME%\gradle\wrapper\gradle-wrapper.jar
|
|
73
|
+
|
|
74
|
+
@rem Execute Gradle
|
|
75
|
+
"%JAVA_EXE%" %DEFAULT_JVM_OPTS% %JAVA_OPTS% %GRADLE_OPTS% "-Dorg.gradle.appname=%APP_BASE_NAME%" -classpath "%CLASSPATH%" org.gradle.wrapper.GradleWrapperMain %CMD_LINE_ARGS%
|
|
76
|
+
|
|
77
|
+
:end
|
|
78
|
+
@rem End local scope for the variables with windows NT shell
|
|
79
|
+
if "%ERRORLEVEL%"=="0" goto mainEnd
|
|
80
|
+
|
|
81
|
+
:fail
|
|
82
|
+
rem Set variable GRADLE_EXIT_CONSOLE if you need the _script_ return code instead of
|
|
83
|
+
rem the _cmd.exe /c_ return code!
|
|
84
|
+
if not "" == "%GRADLE_EXIT_CONSOLE%" exit 1
|
|
85
|
+
exit /b 1
|
|
86
|
+
|
|
87
|
+
:mainEnd
|
|
88
|
+
if "%OS%"=="Windows_NT" endlocal
|
|
89
|
+
|
|
90
|
+
:omega
|
data/lib/embulk/guess/jsonl.rb
CHANGED
|
@@ -1,84 +1,33 @@
|
|
|
1
1
|
require 'json'
|
|
2
|
+
require "embulk/parser/jsonl.rb"
|
|
2
3
|
|
|
3
4
|
module Embulk
|
|
4
5
|
module Guess
|
|
6
|
+
# $ embulk guess -g "jsonl" partial-config.yml
|
|
5
7
|
|
|
6
|
-
# TODO
|
|
7
|
-
# $ embulk guess -g "jsonl" partial-config.yml
|
|
8
|
-
#
|
|
9
|
-
# Depending on the file format the plugin uses, you can use choose
|
|
10
|
-
# one of binary guess (GuessPlugin), text guess (TextGuessPlugin),
|
|
11
|
-
# or line guess (LineGuessPlugin).
|
|
12
|
-
|
|
13
|
-
require "embulk/parser/jsonl.rb"
|
|
14
|
-
|
|
15
|
-
#class JsonlParserGuessPlugin < GuessPlugin
|
|
16
|
-
# Plugin.register_guess("jsonl", self)
|
|
17
|
-
#
|
|
18
|
-
# def guess(config, sample_buffer)
|
|
19
|
-
# if sample_buffer[0,2] == GZIP_HEADER
|
|
20
|
-
# guessed = {}
|
|
21
|
-
# guessed["type"] = "jsonl"
|
|
22
|
-
# guessed["property1"] = "guessed-value"
|
|
23
|
-
# return {"parser" => guessed}
|
|
24
|
-
# else
|
|
25
|
-
# return {}
|
|
26
|
-
# end
|
|
27
|
-
# end
|
|
28
|
-
#end
|
|
29
|
-
|
|
30
|
-
#class JsonlParserGuessPlugin < TextGuessPlugin
|
|
31
|
-
# Plugin.register_guess("jsonl", self)
|
|
32
|
-
#
|
|
33
|
-
# def guess_text(config, sample_text)
|
|
34
|
-
# js = JSON.parse(sample_text) rescue nil
|
|
35
|
-
# if js && js["mykeyword"] == "keyword"
|
|
36
|
-
# guessed = {}
|
|
37
|
-
# guessed["type"] = "jsonl"
|
|
38
|
-
# guessed["property1"] = "guessed-value"
|
|
39
|
-
# return {"parser" => guessed}
|
|
40
|
-
# else
|
|
41
|
-
# return {}
|
|
42
|
-
# end
|
|
43
|
-
# end
|
|
44
|
-
#end
|
|
45
|
-
|
|
46
|
-
class JsonlParserGuessPlugin < LineGuessPlugin
|
|
8
|
+
class Jsonl < LineGuessPlugin # TODO should use GuessPlugin instead of LineGuessPlugin
|
|
47
9
|
Plugin.register_guess("jsonl", self)
|
|
48
10
|
|
|
49
11
|
def guess_lines(config, sample_lines)
|
|
12
|
+
#return {} unless config.fetch("parser", {}).fetch("type", "jsonl") == "jsonl"
|
|
13
|
+
|
|
14
|
+
rows = []
|
|
15
|
+
|
|
50
16
|
columns = {}
|
|
51
17
|
sample_lines.each do |line|
|
|
52
|
-
|
|
53
|
-
hash.each do |k, v|
|
|
54
|
-
columns[k] = get_embulk_type(v)
|
|
55
|
-
end
|
|
18
|
+
rows << JSON.parse(line)
|
|
56
19
|
end
|
|
57
|
-
schema = []
|
|
58
|
-
columns.each do |k,v|
|
|
59
|
-
schema << {'name' => k, 'type' => v}
|
|
60
|
-
end
|
|
61
|
-
guessed = {}
|
|
62
|
-
guessed["type"] = "jsonl"
|
|
63
|
-
guessed["schema"] = schema
|
|
64
|
-
return {"parser" => guessed}
|
|
65
|
-
end
|
|
66
20
|
|
|
67
|
-
|
|
21
|
+
return {} if rows.size <= 3
|
|
68
22
|
|
|
69
|
-
|
|
70
|
-
|
|
71
|
-
|
|
72
|
-
|
|
73
|
-
when FalseClass
|
|
74
|
-
return "boolean"
|
|
75
|
-
when Integer
|
|
76
|
-
return "long"
|
|
77
|
-
when Float
|
|
78
|
-
return "double"
|
|
79
|
-
else
|
|
80
|
-
return "string"
|
|
23
|
+
columns = Embulk::Guess::SchemaGuess.from_hash_records(rows).map do |c|
|
|
24
|
+
column = {name: c.name, type: c.type}
|
|
25
|
+
column[:format] = c.format if c.format
|
|
26
|
+
column
|
|
81
27
|
end
|
|
28
|
+
parser_guessed = {"type" => "jsonl"}
|
|
29
|
+
parser_guessed["columns"] = columns
|
|
30
|
+
return {"parser" => parser_guessed}
|
|
82
31
|
end
|
|
83
32
|
end
|
|
84
33
|
end
|
data/lib/embulk/parser/jsonl.rb
CHANGED
|
@@ -1,67 +1,3 @@
|
|
|
1
|
-
|
|
2
|
-
|
|
3
|
-
|
|
4
|
-
module Parser
|
|
5
|
-
|
|
6
|
-
class JsonlParserPlugin < ParserPlugin
|
|
7
|
-
Plugin.register_parser("jsonl", self)
|
|
8
|
-
|
|
9
|
-
def self.transaction(config, &control)
|
|
10
|
-
parser_task = config.load_config(Java::LineDecoder::DecoderTask)
|
|
11
|
-
task = {
|
|
12
|
-
"decoder_task" => DataSource.from_java(parser_task.dump),
|
|
13
|
-
"schema" => config.param("schema", :array)
|
|
14
|
-
}
|
|
15
|
-
columns = task["schema"].each_with_index.map do |c, i|
|
|
16
|
-
Column.new(i, c["name"], c["type"].to_sym)
|
|
17
|
-
end
|
|
18
|
-
yield(task, columns)
|
|
19
|
-
end
|
|
20
|
-
|
|
21
|
-
def init
|
|
22
|
-
@decoder_task = task.param("decoder_task", :hash).load_task(Java::LineDecoder::DecoderTask)
|
|
23
|
-
end
|
|
24
|
-
|
|
25
|
-
def run(file_input)
|
|
26
|
-
decoder = Java::LineDecoder.new(file_input.instance_eval { @java_file_input }, @decoder_task)
|
|
27
|
-
schema = @task["schema"]
|
|
28
|
-
|
|
29
|
-
while decoder.nextFile
|
|
30
|
-
while line = decoder.poll
|
|
31
|
-
begin
|
|
32
|
-
hash = JSON.parse(line)
|
|
33
|
-
@page_builder.add(make_record(schema, hash))
|
|
34
|
-
rescue
|
|
35
|
-
# TODO: logging
|
|
36
|
-
end
|
|
37
|
-
end
|
|
38
|
-
end
|
|
39
|
-
page_builder.finish
|
|
40
|
-
end
|
|
41
|
-
|
|
42
|
-
private
|
|
43
|
-
|
|
44
|
-
def make_record(schema, e)
|
|
45
|
-
schema.map do |c|
|
|
46
|
-
val = e[c["name"]]
|
|
47
|
-
v = val.nil? ? "" : val
|
|
48
|
-
case c["type"]
|
|
49
|
-
when "string"
|
|
50
|
-
v
|
|
51
|
-
when "long"
|
|
52
|
-
v.to_i
|
|
53
|
-
when "double"
|
|
54
|
-
v.to_f
|
|
55
|
-
when "boolean"
|
|
56
|
-
["yes", "true", "1"].include?(v.downcase)
|
|
57
|
-
when "timestamp"
|
|
58
|
-
v.empty? ? nil : Time.strptime(v, c["time_format"])
|
|
59
|
-
else
|
|
60
|
-
raise "Unsupported type #{c['type']}"
|
|
61
|
-
end
|
|
62
|
-
end
|
|
63
|
-
end
|
|
64
|
-
end
|
|
65
|
-
|
|
66
|
-
end
|
|
67
|
-
end
|
|
1
|
+
Embulk::JavaPlugin.register_parser(
|
|
2
|
+
"jsonl", "org.embulk.parser.jsonl.JsonlParserPlugin",
|
|
3
|
+
File.expand_path('../../../../classpath', __FILE__))
|
data/settings.gradle
ADDED
|
@@ -0,0 +1 @@
|
|
|
1
|
+
rootProject.name = 'embulk-parser-jsonl'
|
|
@@ -0,0 +1,282 @@
|
|
|
1
|
+
package org.embulk.parser.jsonl;
|
|
2
|
+
|
|
3
|
+
import com.google.common.base.Optional;
|
|
4
|
+
import com.google.common.collect.ImmutableMap;
|
|
5
|
+
import org.embulk.config.Config;
|
|
6
|
+
import org.embulk.config.ConfigDefault;
|
|
7
|
+
import org.embulk.config.ConfigException;
|
|
8
|
+
import org.embulk.config.ConfigSource;
|
|
9
|
+
import org.embulk.config.Task;
|
|
10
|
+
import org.embulk.config.TaskSource;
|
|
11
|
+
import org.embulk.spi.Column;
|
|
12
|
+
import org.embulk.spi.ColumnVisitor;
|
|
13
|
+
import org.embulk.spi.DataException;
|
|
14
|
+
import org.embulk.spi.Exec;
|
|
15
|
+
import org.embulk.spi.FileInput;
|
|
16
|
+
import org.embulk.spi.PageBuilder;
|
|
17
|
+
import org.embulk.spi.PageOutput;
|
|
18
|
+
import org.embulk.spi.ParserPlugin;
|
|
19
|
+
import org.embulk.spi.Schema;
|
|
20
|
+
import org.embulk.spi.SchemaConfig;
|
|
21
|
+
import org.embulk.spi.json.JsonParser;
|
|
22
|
+
import org.embulk.spi.time.TimestampParser;
|
|
23
|
+
import org.embulk.spi.util.LineDecoder;
|
|
24
|
+
import org.embulk.spi.util.Timestamps;
|
|
25
|
+
import org.msgpack.core.MessageTypeException;
|
|
26
|
+
import org.msgpack.value.BooleanValue;
|
|
27
|
+
import org.msgpack.value.FloatValue;
|
|
28
|
+
import org.msgpack.value.IntegerValue;
|
|
29
|
+
import org.msgpack.value.Value;
|
|
30
|
+
import org.slf4j.Logger;
|
|
31
|
+
|
|
32
|
+
import java.util.Map;
|
|
33
|
+
|
|
34
|
+
import static org.msgpack.value.ValueFactory.newString;
|
|
35
|
+
|
|
36
|
+
public class JsonlParserPlugin
|
|
37
|
+
implements ParserPlugin
|
|
38
|
+
{
|
|
39
|
+
public interface PluginTask
|
|
40
|
+
extends Task, LineDecoder.DecoderTask, TimestampParser.Task
|
|
41
|
+
{
|
|
42
|
+
@Config("columns")
|
|
43
|
+
@ConfigDefault("null")
|
|
44
|
+
Optional<SchemaConfig> getSchemaConfig();
|
|
45
|
+
|
|
46
|
+
@Config("schema")
|
|
47
|
+
@ConfigDefault("null")
|
|
48
|
+
@Deprecated
|
|
49
|
+
Optional<SchemaConfig> getOldSchemaConfig();
|
|
50
|
+
|
|
51
|
+
@Config("stop_on_invalid_record")
|
|
52
|
+
@ConfigDefault("false")
|
|
53
|
+
boolean getStopOnInvalidRecord();
|
|
54
|
+
}
|
|
55
|
+
|
|
56
|
+
private final Logger log;
|
|
57
|
+
|
|
58
|
+
private String line = null;
|
|
59
|
+
private long lineNumber = 0;
|
|
60
|
+
private Map<String, Value> columnNameValues;
|
|
61
|
+
|
|
62
|
+
public JsonlParserPlugin()
|
|
63
|
+
{
|
|
64
|
+
this.log = Exec.getLogger(JsonlParserPlugin.class);
|
|
65
|
+
}
|
|
66
|
+
|
|
67
|
+
@Override
|
|
68
|
+
public void transaction(ConfigSource configSource, Control control)
|
|
69
|
+
{
|
|
70
|
+
PluginTask task = configSource.loadConfig(PluginTask.class);
|
|
71
|
+
control.run(task.dump(), getSchemaConfig(task).toSchema());
|
|
72
|
+
}
|
|
73
|
+
|
|
74
|
+
// this method is to keep the backward compatibility of 'schema' option.
|
|
75
|
+
private SchemaConfig getSchemaConfig(PluginTask task)
|
|
76
|
+
{
|
|
77
|
+
if (task.getOldSchemaConfig().isPresent()) {
|
|
78
|
+
log.warn("Please use 'columns' option instead of 'schema' because the 'schema' option is deprecated. The next version will stop 'schema' option support.");
|
|
79
|
+
}
|
|
80
|
+
|
|
81
|
+
if (task.getSchemaConfig().isPresent()) {
|
|
82
|
+
return task.getSchemaConfig().get();
|
|
83
|
+
}
|
|
84
|
+
else if (task.getOldSchemaConfig().isPresent()) {
|
|
85
|
+
return task.getOldSchemaConfig().get();
|
|
86
|
+
}
|
|
87
|
+
else {
|
|
88
|
+
throw new ConfigException("Attribute 'columns' is required but not set");
|
|
89
|
+
}
|
|
90
|
+
}
|
|
91
|
+
|
|
92
|
+
@Override
|
|
93
|
+
public void run(TaskSource taskSource, Schema schema, FileInput input, PageOutput output)
|
|
94
|
+
{
|
|
95
|
+
PluginTask task = taskSource.loadTask(PluginTask.class);
|
|
96
|
+
|
|
97
|
+
setColumnNameValues(schema);
|
|
98
|
+
|
|
99
|
+
final TimestampParser[] timestampParsers = Timestamps.newTimestampColumnParsers(task, getSchemaConfig(task));
|
|
100
|
+
final LineDecoder decoder = newLineDecoder(input, task);
|
|
101
|
+
final JsonParser jsonParser = newJsonParser();
|
|
102
|
+
final boolean stopOnInvalidRecord = task.getStopOnInvalidRecord();
|
|
103
|
+
|
|
104
|
+
try (final PageBuilder pageBuilder = new PageBuilder(Exec.getBufferAllocator(), schema, output)) {
|
|
105
|
+
while (decoder.nextFile()) { // TODO this implementation should be improved with new JsonParser API on Embulk v0.8.3
|
|
106
|
+
lineNumber = 0;
|
|
107
|
+
|
|
108
|
+
while ((line = decoder.poll()) != null) {
|
|
109
|
+
lineNumber++;
|
|
110
|
+
|
|
111
|
+
try {
|
|
112
|
+
Value value = jsonParser.parse(line);
|
|
113
|
+
|
|
114
|
+
if (!value.isMapValue()) {
|
|
115
|
+
throw new JsonRecordValidateException("Json string is not representing map value.");
|
|
116
|
+
}
|
|
117
|
+
|
|
118
|
+
final Map<Value, Value> record = value.asMapValue().map();
|
|
119
|
+
|
|
120
|
+
schema.visitColumns(new ColumnVisitor() {
|
|
121
|
+
@Override
|
|
122
|
+
public void booleanColumn(Column column)
|
|
123
|
+
{
|
|
124
|
+
Value v = record.get(getColumnNameValue(column));
|
|
125
|
+
if (isNil(v)) {
|
|
126
|
+
pageBuilder.setNull(column);
|
|
127
|
+
}
|
|
128
|
+
else {
|
|
129
|
+
try {
|
|
130
|
+
pageBuilder.setBoolean(column, ((BooleanValue) v).getBoolean());
|
|
131
|
+
}
|
|
132
|
+
catch (MessageTypeException e) {
|
|
133
|
+
throw new JsonRecordValidateException(e);
|
|
134
|
+
}
|
|
135
|
+
}
|
|
136
|
+
}
|
|
137
|
+
|
|
138
|
+
@Override
|
|
139
|
+
public void longColumn(Column column)
|
|
140
|
+
{
|
|
141
|
+
Value v = record.get(getColumnNameValue(column));
|
|
142
|
+
if (isNil(v)) {
|
|
143
|
+
pageBuilder.setNull(column);
|
|
144
|
+
}
|
|
145
|
+
else {
|
|
146
|
+
try {
|
|
147
|
+
pageBuilder.setLong(column, ((IntegerValue) v).asLong());
|
|
148
|
+
}
|
|
149
|
+
catch (MessageTypeException e) {
|
|
150
|
+
throw new JsonRecordValidateException(e);
|
|
151
|
+
}
|
|
152
|
+
}
|
|
153
|
+
}
|
|
154
|
+
|
|
155
|
+
@Override
|
|
156
|
+
public void doubleColumn(Column column)
|
|
157
|
+
{
|
|
158
|
+
Value v = record.get(getColumnNameValue(column));
|
|
159
|
+
if (isNil(v)) {
|
|
160
|
+
pageBuilder.setNull(column);
|
|
161
|
+
}
|
|
162
|
+
else {
|
|
163
|
+
try {
|
|
164
|
+
pageBuilder.setDouble(column, ((FloatValue) v).toDouble());
|
|
165
|
+
}
|
|
166
|
+
catch (MessageTypeException e) {
|
|
167
|
+
throw new JsonRecordValidateException(e);
|
|
168
|
+
}
|
|
169
|
+
}
|
|
170
|
+
}
|
|
171
|
+
|
|
172
|
+
@Override
|
|
173
|
+
public void stringColumn(Column column)
|
|
174
|
+
{
|
|
175
|
+
Value v = record.get(getColumnNameValue(column));
|
|
176
|
+
if (isNil(v)) {
|
|
177
|
+
pageBuilder.setNull(column);
|
|
178
|
+
}
|
|
179
|
+
else {
|
|
180
|
+
try {
|
|
181
|
+
pageBuilder.setString(column, v.toString());
|
|
182
|
+
}
|
|
183
|
+
catch (MessageTypeException e) {
|
|
184
|
+
throw new JsonRecordValidateException(e);
|
|
185
|
+
}
|
|
186
|
+
}
|
|
187
|
+
}
|
|
188
|
+
|
|
189
|
+
@Override
|
|
190
|
+
public void timestampColumn(Column column)
|
|
191
|
+
{
|
|
192
|
+
Value v = record.get(getColumnNameValue(column));
|
|
193
|
+
if (isNil(v)) {
|
|
194
|
+
pageBuilder.setNull(column);
|
|
195
|
+
}
|
|
196
|
+
else {
|
|
197
|
+
try {
|
|
198
|
+
pageBuilder.setTimestamp(column, timestampParsers[column.getIndex()].parse(v.toString()));
|
|
199
|
+
}
|
|
200
|
+
catch (MessageTypeException e) {
|
|
201
|
+
throw new JsonRecordValidateException(e);
|
|
202
|
+
}
|
|
203
|
+
}
|
|
204
|
+
}
|
|
205
|
+
|
|
206
|
+
@Override
|
|
207
|
+
public void jsonColumn(Column column)
|
|
208
|
+
{
|
|
209
|
+
Value v = record.get(getColumnNameValue(column));
|
|
210
|
+
if (isNil(v)) {
|
|
211
|
+
pageBuilder.setNull(column);
|
|
212
|
+
}
|
|
213
|
+
else {
|
|
214
|
+
try {
|
|
215
|
+
pageBuilder.setJson(column, v);
|
|
216
|
+
}
|
|
217
|
+
catch (MessageTypeException e) {
|
|
218
|
+
throw new JsonRecordValidateException(e);
|
|
219
|
+
}
|
|
220
|
+
}
|
|
221
|
+
}
|
|
222
|
+
|
|
223
|
+
private boolean isNil(Value v)
|
|
224
|
+
{
|
|
225
|
+
return v == null || v.isNilValue();
|
|
226
|
+
}
|
|
227
|
+
});
|
|
228
|
+
|
|
229
|
+
pageBuilder.addRecord();
|
|
230
|
+
}
|
|
231
|
+
catch (JsonRecordValidateException e) {
|
|
232
|
+
if (stopOnInvalidRecord) {
|
|
233
|
+
throw new DataException(String.format("Invalid record at line %d: %s", lineNumber, line), e);
|
|
234
|
+
}
|
|
235
|
+
log.warn(String.format("Skipped line %d (%s): %s", lineNumber, e.getMessage(), line));
|
|
236
|
+
}
|
|
237
|
+
}
|
|
238
|
+
}
|
|
239
|
+
|
|
240
|
+
pageBuilder.finish();
|
|
241
|
+
}
|
|
242
|
+
}
|
|
243
|
+
|
|
244
|
+
private void setColumnNameValues(Schema schema)
|
|
245
|
+
{
|
|
246
|
+
ImmutableMap.Builder<String, Value> builder = ImmutableMap.builder();
|
|
247
|
+
for (Column column : schema.getColumns()) {
|
|
248
|
+
String name = column.getName();
|
|
249
|
+
builder.put(name, newString(name));
|
|
250
|
+
}
|
|
251
|
+
columnNameValues = builder.build();
|
|
252
|
+
}
|
|
253
|
+
|
|
254
|
+
private Value getColumnNameValue(Column column)
|
|
255
|
+
{
|
|
256
|
+
return columnNameValues.get(column.getName());
|
|
257
|
+
}
|
|
258
|
+
|
|
259
|
+
public LineDecoder newLineDecoder(FileInput input, PluginTask task)
|
|
260
|
+
{
|
|
261
|
+
return new LineDecoder(input, task);
|
|
262
|
+
}
|
|
263
|
+
|
|
264
|
+
public JsonParser newJsonParser()
|
|
265
|
+
{
|
|
266
|
+
return new JsonParser();
|
|
267
|
+
}
|
|
268
|
+
|
|
269
|
+
static class JsonRecordValidateException
|
|
270
|
+
extends DataException
|
|
271
|
+
{
|
|
272
|
+
JsonRecordValidateException(String message)
|
|
273
|
+
{
|
|
274
|
+
super(message);
|
|
275
|
+
}
|
|
276
|
+
|
|
277
|
+
JsonRecordValidateException(Throwable cause)
|
|
278
|
+
{
|
|
279
|
+
super(cause);
|
|
280
|
+
}
|
|
281
|
+
}
|
|
282
|
+
}
|
|
@@ -0,0 +1,228 @@
|
|
|
1
|
+
package org.embulk.parser.jsonl;
|
|
2
|
+
|
|
3
|
+
import com.google.common.collect.ImmutableList;
|
|
4
|
+
import com.google.common.collect.Lists;
|
|
5
|
+
import org.embulk.EmbulkTestRuntime;
|
|
6
|
+
import org.embulk.config.ConfigSource;
|
|
7
|
+
import org.embulk.config.TaskSource;
|
|
8
|
+
import org.embulk.spi.ColumnConfig;
|
|
9
|
+
import org.embulk.spi.DataException;
|
|
10
|
+
import org.embulk.spi.FileInput;
|
|
11
|
+
import org.embulk.spi.ParserPlugin;
|
|
12
|
+
import org.embulk.spi.Schema;
|
|
13
|
+
import org.embulk.spi.SchemaConfig;
|
|
14
|
+
import org.embulk.spi.TestPageBuilderReader.MockPageOutput;
|
|
15
|
+
import org.embulk.spi.time.Timestamp;
|
|
16
|
+
import org.embulk.spi.type.Type;
|
|
17
|
+
import org.embulk.spi.util.InputStreamFileInput;
|
|
18
|
+
import org.embulk.spi.util.Pages;
|
|
19
|
+
import org.junit.Before;
|
|
20
|
+
import org.junit.Rule;
|
|
21
|
+
import org.junit.Test;
|
|
22
|
+
|
|
23
|
+
import java.io.ByteArrayInputStream;
|
|
24
|
+
import java.io.IOException;
|
|
25
|
+
import java.io.InputStream;
|
|
26
|
+
import java.util.List;
|
|
27
|
+
|
|
28
|
+
import static org.embulk.spi.type.Types.BOOLEAN;
|
|
29
|
+
import static org.embulk.spi.type.Types.DOUBLE;
|
|
30
|
+
import static org.embulk.spi.type.Types.JSON;
|
|
31
|
+
import static org.embulk.spi.type.Types.LONG;
|
|
32
|
+
import static org.embulk.spi.type.Types.STRING;
|
|
33
|
+
import static org.embulk.spi.type.Types.TIMESTAMP;
|
|
34
|
+
import static org.junit.Assert.assertEquals;
|
|
35
|
+
import static org.junit.Assert.assertNull;
|
|
36
|
+
import static org.junit.Assert.assertTrue;
|
|
37
|
+
import static org.junit.Assert.fail;
|
|
38
|
+
import static org.msgpack.value.ValueFactory.newArray;
|
|
39
|
+
import static org.msgpack.value.ValueFactory.newMap;
|
|
40
|
+
import static org.msgpack.value.ValueFactory.newString;
|
|
41
|
+
|
|
42
|
+
public class TestJsonlParserPlugin
|
|
43
|
+
{
|
|
44
|
+
@Rule
|
|
45
|
+
public EmbulkTestRuntime runtime = new EmbulkTestRuntime();
|
|
46
|
+
|
|
47
|
+
private ConfigSource config;
|
|
48
|
+
private JsonlParserPlugin plugin;
|
|
49
|
+
private MockPageOutput output;
|
|
50
|
+
|
|
51
|
+
@Before
|
|
52
|
+
public void createResource()
|
|
53
|
+
{
|
|
54
|
+
config = config().set("type", "jsonl");
|
|
55
|
+
plugin = new JsonlParserPlugin();
|
|
56
|
+
recreatePageOutput();
|
|
57
|
+
}
|
|
58
|
+
|
|
59
|
+
private void recreatePageOutput()
|
|
60
|
+
{
|
|
61
|
+
output = new MockPageOutput();
|
|
62
|
+
}
|
|
63
|
+
|
|
64
|
+
@Test
|
|
65
|
+
public void skipRecords()
|
|
66
|
+
throws Exception
|
|
67
|
+
{
|
|
68
|
+
SchemaConfig schema = schema(
|
|
69
|
+
column("_c0", BOOLEAN), column("_c1", LONG), column("_c2", DOUBLE),
|
|
70
|
+
column("_c3", STRING), column("_c4", TIMESTAMP), column("_c5", JSON));
|
|
71
|
+
ConfigSource config = this.config.deepCopy().set("columns", schema);
|
|
72
|
+
|
|
73
|
+
transaction(config, fileInput(
|
|
74
|
+
"[]",
|
|
75
|
+
"\"embulk\"",
|
|
76
|
+
"10",
|
|
77
|
+
"true",
|
|
78
|
+
"false",
|
|
79
|
+
"null"
|
|
80
|
+
));
|
|
81
|
+
|
|
82
|
+
List<Object[]> records = Pages.toObjects(schema.toSchema(), output.pages);
|
|
83
|
+
assertEquals(0, records.size());
|
|
84
|
+
}
|
|
85
|
+
|
|
86
|
+
@Test
|
|
87
|
+
public void throwDataException()
|
|
88
|
+
throws Exception
|
|
89
|
+
{
|
|
90
|
+
SchemaConfig schema = schema(
|
|
91
|
+
column("_c0", BOOLEAN), column("_c1", LONG), column("_c2", DOUBLE),
|
|
92
|
+
column("_c3", STRING), column("_c4", TIMESTAMP), column("_c5", JSON));
|
|
93
|
+
ConfigSource config = this.config.deepCopy().set("columns", schema).set("stop_on_invalid_record", true);
|
|
94
|
+
|
|
95
|
+
try {
|
|
96
|
+
transaction(config, fileInput(
|
|
97
|
+
"\"not_map_value\""
|
|
98
|
+
));
|
|
99
|
+
fail();
|
|
100
|
+
}
|
|
101
|
+
catch (Throwable t) {
|
|
102
|
+
assertTrue(t instanceof DataException);
|
|
103
|
+
}
|
|
104
|
+
}
|
|
105
|
+
|
|
106
|
+
@Test
|
|
107
|
+
public void writeNils()
|
|
108
|
+
throws Exception
|
|
109
|
+
{
|
|
110
|
+
SchemaConfig schema = schema(
|
|
111
|
+
column("_c0", BOOLEAN), column("_c1", LONG), column("_c2", DOUBLE),
|
|
112
|
+
column("_c3", STRING), column("_c4", TIMESTAMP), column("_c5", JSON));
|
|
113
|
+
ConfigSource config = this.config.deepCopy().set("columns", schema);
|
|
114
|
+
|
|
115
|
+
transaction(config, fileInput(
|
|
116
|
+
"{}",
|
|
117
|
+
"{\"_c0\":null,\"_c1\":null,\"_c2\":null}",
|
|
118
|
+
"{\"_c3\":null,\"_c4\":null,\"_c5\":null}",
|
|
119
|
+
"{}"
|
|
120
|
+
));
|
|
121
|
+
|
|
122
|
+
List<Object[]> records = Pages.toObjects(schema.toSchema(), output.pages);
|
|
123
|
+
assertEquals(4, records.size());
|
|
124
|
+
|
|
125
|
+
for (Object[] record : records) {
|
|
126
|
+
for (int i = 0; i < 6; i++) {
|
|
127
|
+
assertNull(record[i]);
|
|
128
|
+
}
|
|
129
|
+
}
|
|
130
|
+
}
|
|
131
|
+
|
|
132
|
+
@Test
|
|
133
|
+
public void useNormal()
|
|
134
|
+
throws Exception
|
|
135
|
+
{
|
|
136
|
+
SchemaConfig schema = schema(
|
|
137
|
+
column("_c0", BOOLEAN), column("_c1", LONG), column("_c2", DOUBLE),
|
|
138
|
+
column("_c3", STRING), column("_c4", TIMESTAMP, config().set("format", "%Y-%m-%d %H:%M:%S %Z")), column("_c5", JSON));
|
|
139
|
+
List<ConfigSource> configs = Lists.newArrayList(
|
|
140
|
+
this.config.deepCopy().set("columns", schema),
|
|
141
|
+
this.config.deepCopy().set("schema", schema)
|
|
142
|
+
);
|
|
143
|
+
|
|
144
|
+
for (ConfigSource config : configs) {
|
|
145
|
+
transaction(config, fileInput(
|
|
146
|
+
"{\"_c0\":true,\"_c1\":10,\"_c2\":0.1,\"_c3\":\"embulk\",\"_c4\":\"2016-01-01 00:00:00 UTC\",\"_c5\":{\"k\":\"v\"}}",
|
|
147
|
+
"[1, 2, 3]",
|
|
148
|
+
"{\"_c0\":false,\"_c1\":-10,\"_c2\":1.0,\"_c3\":\"エンバルク\",\"_c4\":\"2016-01-01 00:00:00 +0000\",\"_c5\":[\"e0\",\"e1\"]}"
|
|
149
|
+
));
|
|
150
|
+
|
|
151
|
+
List<Object[]> records = Pages.toObjects(schema.toSchema(), output.pages);
|
|
152
|
+
assertEquals(2, records.size());
|
|
153
|
+
|
|
154
|
+
Object[] record;
|
|
155
|
+
{
|
|
156
|
+
record = records.get(0);
|
|
157
|
+
assertEquals(true, record[0]);
|
|
158
|
+
assertEquals(10L, record[1]);
|
|
159
|
+
assertEquals(0.1, (Double) record[2], 0.0001);
|
|
160
|
+
assertEquals("embulk", record[3]);
|
|
161
|
+
assertEquals(Timestamp.ofEpochSecond(1451606400L), record[4]);
|
|
162
|
+
assertEquals(newMap(newString("k"), newString("v")), record[5]);
|
|
163
|
+
}
|
|
164
|
+
{
|
|
165
|
+
record = records.get(1);
|
|
166
|
+
assertEquals(false, record[0]);
|
|
167
|
+
assertEquals(-10L, record[1]);
|
|
168
|
+
assertEquals(1.0, (Double) record[2], 0.0001);
|
|
169
|
+
assertEquals("エンバルク", record[3]);
|
|
170
|
+
assertEquals(Timestamp.ofEpochSecond(1451606400L), record[4]);
|
|
171
|
+
assertEquals(newArray(newString("e0"), newString("e1")), record[5]);
|
|
172
|
+
}
|
|
173
|
+
|
|
174
|
+
recreatePageOutput();
|
|
175
|
+
}
|
|
176
|
+
}
|
|
177
|
+
|
|
178
|
+
private ConfigSource config()
|
|
179
|
+
{
|
|
180
|
+
return runtime.getExec().newConfigSource();
|
|
181
|
+
}
|
|
182
|
+
|
|
183
|
+
private void transaction(ConfigSource config, final FileInput input)
|
|
184
|
+
{
|
|
185
|
+
plugin.transaction(config, new ParserPlugin.Control()
|
|
186
|
+
{
|
|
187
|
+
@Override
|
|
188
|
+
public void run(TaskSource taskSource, Schema schema)
|
|
189
|
+
{
|
|
190
|
+
plugin.run(taskSource, schema, input, output);
|
|
191
|
+
}
|
|
192
|
+
});
|
|
193
|
+
}
|
|
194
|
+
|
|
195
|
+
private FileInput fileInput(String... lines)
|
|
196
|
+
throws Exception
|
|
197
|
+
{
|
|
198
|
+
StringBuilder sb = new StringBuilder();
|
|
199
|
+
for (String line : lines) {
|
|
200
|
+
sb.append(line).append("\n");
|
|
201
|
+
}
|
|
202
|
+
|
|
203
|
+
ByteArrayInputStream in = new ByteArrayInputStream(sb.toString().getBytes());
|
|
204
|
+
return new InputStreamFileInput(runtime.getBufferAllocator(), provider(in));
|
|
205
|
+
}
|
|
206
|
+
|
|
207
|
+
private InputStreamFileInput.IteratorProvider provider(InputStream... inputStreams)
|
|
208
|
+
throws IOException
|
|
209
|
+
{
|
|
210
|
+
return new InputStreamFileInput.IteratorProvider(
|
|
211
|
+
ImmutableList.copyOf(inputStreams));
|
|
212
|
+
}
|
|
213
|
+
|
|
214
|
+
private SchemaConfig schema(ColumnConfig... columns)
|
|
215
|
+
{
|
|
216
|
+
return new SchemaConfig(Lists.newArrayList(columns));
|
|
217
|
+
}
|
|
218
|
+
|
|
219
|
+
private ColumnConfig column(String name, Type type)
|
|
220
|
+
{
|
|
221
|
+
return column(name, type, config());
|
|
222
|
+
}
|
|
223
|
+
|
|
224
|
+
private ColumnConfig column(String name, Type type, ConfigSource option)
|
|
225
|
+
{
|
|
226
|
+
return new ColumnConfig(name, type, option);
|
|
227
|
+
}
|
|
228
|
+
}
|
metadata
CHANGED
|
@@ -1,41 +1,41 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: embulk-parser-jsonl
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.0
|
|
4
|
+
version: 0.1.0
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Shunsuke Mikami
|
|
8
|
-
autorequire:
|
|
8
|
+
autorequire:
|
|
9
9
|
bindir: bin
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date:
|
|
11
|
+
date: 2016-02-22 00:00:00.000000000 Z
|
|
12
12
|
dependencies:
|
|
13
13
|
- !ruby/object:Gem::Dependency
|
|
14
|
-
name: bundler
|
|
15
14
|
requirement: !ruby/object:Gem::Requirement
|
|
16
15
|
requirements:
|
|
17
|
-
- -
|
|
16
|
+
- - ~>
|
|
18
17
|
- !ruby/object:Gem::Version
|
|
19
18
|
version: '1.0'
|
|
20
|
-
|
|
19
|
+
name: bundler
|
|
21
20
|
prerelease: false
|
|
21
|
+
type: :development
|
|
22
22
|
version_requirements: !ruby/object:Gem::Requirement
|
|
23
23
|
requirements:
|
|
24
|
-
- -
|
|
24
|
+
- - ~>
|
|
25
25
|
- !ruby/object:Gem::Version
|
|
26
26
|
version: '1.0'
|
|
27
27
|
- !ruby/object:Gem::Dependency
|
|
28
|
-
name: rake
|
|
29
28
|
requirement: !ruby/object:Gem::Requirement
|
|
30
29
|
requirements:
|
|
31
|
-
- -
|
|
30
|
+
- - ~>
|
|
32
31
|
- !ruby/object:Gem::Version
|
|
33
32
|
version: '10.0'
|
|
34
|
-
|
|
33
|
+
name: rake
|
|
35
34
|
prerelease: false
|
|
35
|
+
type: :development
|
|
36
36
|
version_requirements: !ruby/object:Gem::Requirement
|
|
37
37
|
requirements:
|
|
38
|
-
- -
|
|
38
|
+
- - ~>
|
|
39
39
|
- !ruby/object:Gem::Version
|
|
40
40
|
version: '10.0'
|
|
41
41
|
description: Parses Jsonl files read by other file input plugins.
|
|
@@ -45,36 +45,46 @@ executables: []
|
|
|
45
45
|
extensions: []
|
|
46
46
|
extra_rdoc_files: []
|
|
47
47
|
files:
|
|
48
|
-
-
|
|
48
|
+
- .gitignore
|
|
49
|
+
- CHANGELOG.md
|
|
49
50
|
- Gemfile
|
|
50
51
|
- LICENSE.txt
|
|
51
52
|
- README.md
|
|
52
53
|
- Rakefile
|
|
54
|
+
- build.gradle
|
|
53
55
|
- embulk-parser-jsonl.gemspec
|
|
56
|
+
- gradle/wrapper/gradle-wrapper.jar
|
|
57
|
+
- gradle/wrapper/gradle-wrapper.properties
|
|
58
|
+
- gradlew
|
|
59
|
+
- gradlew.bat
|
|
54
60
|
- lib/embulk/guess/jsonl.rb
|
|
55
61
|
- lib/embulk/parser/jsonl.rb
|
|
62
|
+
- settings.gradle
|
|
63
|
+
- src/main/java/org/embulk/parser/jsonl/JsonlParserPlugin.java
|
|
64
|
+
- src/test/java/org/embulk/parser/jsonl/TestJsonlParserPlugin.java
|
|
65
|
+
- classpath/embulk-parser-jsonl-0.1.0.jar
|
|
56
66
|
homepage: https://github.com/shun0102/embulk-parser-jsonl
|
|
57
67
|
licenses:
|
|
58
68
|
- MIT
|
|
59
69
|
metadata: {}
|
|
60
|
-
post_install_message:
|
|
70
|
+
post_install_message:
|
|
61
71
|
rdoc_options: []
|
|
62
72
|
require_paths:
|
|
63
73
|
- lib
|
|
64
74
|
required_ruby_version: !ruby/object:Gem::Requirement
|
|
65
75
|
requirements:
|
|
66
|
-
- -
|
|
76
|
+
- - '>='
|
|
67
77
|
- !ruby/object:Gem::Version
|
|
68
78
|
version: '0'
|
|
69
79
|
required_rubygems_version: !ruby/object:Gem::Requirement
|
|
70
80
|
requirements:
|
|
71
|
-
- -
|
|
81
|
+
- - '>='
|
|
72
82
|
- !ruby/object:Gem::Version
|
|
73
83
|
version: '0'
|
|
74
84
|
requirements: []
|
|
75
|
-
rubyforge_project:
|
|
76
|
-
rubygems_version: 2.
|
|
77
|
-
signing_key:
|
|
85
|
+
rubyforge_project:
|
|
86
|
+
rubygems_version: 2.1.9
|
|
87
|
+
signing_key:
|
|
78
88
|
specification_version: 4
|
|
79
89
|
summary: Jsonl parser plugin for Embulk
|
|
80
90
|
test_files: []
|