embulk-parser-poi_excel 0.1.5 → 0.1.12

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: b6b3e37c337cca59783620b59c08c46eb873188d
-  data.tar.gz: 6648241ed3545ccf5e134412274a5dfa25a41903
+  metadata.gz: 0ce575511ee0ce2d0a9496f1d63013909d751504
+  data.tar.gz: f79c800fd6454ba5e2511e6e675a3465c3c2c7d3
 SHA512:
-  metadata.gz: 3c115aa89954e182773b200e09dd90d091d8f3a8ba274bd1fd2b30b6cfdceb857b65ff2b2cdf375e4cbfc4a73bb46d8ecb3e73ea6ed823d53e1306ce3c4809b8
-  data.tar.gz: 60461a21d343976e53ef726dc6a893572fe041849c7f7ed1cdb3f8511584f54c7956ce3716cbec7a40db7a89717358092c891d487c4e1c1431107180f3ade95a
+  metadata.gz: a4ddf5f3506c56924bbe9147c6b23e40f31853356b53e91215120ef9f930e322db50a60a9c2277bc51c863cf588610e44d18cf5ec401a0d8b31d0c7ff30f4b50
+  data.tar.gz: bf22e1256df159efd21cdf40d98ae014c8ddf1dc053848dce95a2635fa68dbac2d174833731f5ff700d814fc98a680ab666774b15cce49c24d7a94eea87d6ac1

data/README.md

@@ -20,7 +20,7 @@ in:
     skip_header_lines: 1	# first row is header.
     columns:
     - {name: row, type: long, value: row_number}
-    - {name: get_date, type: timestamp, value: cell_value, column_number: A}
+    - {name: get_date, type: timestamp, cell_column: A, value: cell_value}
     - {name: orb_type, type: string}
     - {name: orb_name, type: string}
     - {name: orb_shape, type: long}
@@ -28,13 +28,14 @@ in:
 ```
 
 if omit **value**, specified `cell_value`.  
-if omit **column_number** when **valus** is `cell_value`, specified next column.  
+if omit **cell_column** when **value** is `cell_value`, specified next column.  
 
 
 ## Configuration
 
-* **sheets**: sheet name. (list of string, required)
-* **skip_header_lines**: skip rows. (integer, default: `0`)
+* **sheets**: sheet name. can use wildcards `*`, `?`. (list of string, required)
+* **record_type**: record type.  (`row`, `column` or `sheet`. default: `row`)
+* **skip_header_lines**: skip rows when **record_type**=`row` (skip columns when **record_type**=`column`). ignored when **record_type**=`sheet`. (integer, default: `0`)
 * **columns**: column definition. see below. (hash, required)
 * **sheet_options**: sheet option. see below. (hash, default: null)
 
@@ -42,11 +43,19 @@ if omit **column_number** when **valus** is `cell_value`, specified next column.
 
 * **name**: Embulk column name. (string, required)
 * **type**: Embulk column type. (string, required)
-* **value**: value type. see below. (string, defualt: `cell_value`)
-* **column_number**: Excel column number. see below. (string, default: next column)
+* **value**: value type. see below. (string, default: `cell_value`)
+* **column_number**: same as **cell_column**.
+* **cell_column**: Excel column number. see below. (string, default: next column when **record_type**=`row`)
+* **cell_row**: Excel row number. see below. (integer, default: next row when **record_type**=`column`)
+* **cell_address**: Excel cell address such as `A1`, `Sheet1!B3`. (string, not required)
+* **numeric_format**: format of numeric(double) to string such as `%4.2f`. (default: Java's Double.toString())
 * **attribute_name**: use with value `cell_style`, `cell_font`, etc. see below. (list of string)
 * **on_cell_error**: processing method of Cell error. see below. (string, default: `constant`)
+* **formula_handling**: processing method of formula. see below. (`evaluate` or `cashed_value`. default: `evaluate`)
+* **on_evaluate_error**: processing method of evaluate formula error. see below. (string, default: `exception`)
+* **formula_replace**: replace formula before evaluate. see below.
 * **on_convert_error**: processing method of convert error. see below. (string, default: `exception`)
+* **search_merged_cell**: search merged cell when cell is BLANK. (`none`, `linear_search`, `tree_search` or `hash_search`, default: `hash_search`)
 
 ### value
 
@@ -55,6 +64,8 @@ if omit **column_number** when **valus** is `cell_value`, specified next column.
 * `cell_style`: all cell style attributes. returned json string. see **attribute_name**. (**type** required `string`)
 * `cell_font`: all cell font attributes. returned json string. see **attribute_name**. (**type** required `string`)
 * `cell_comment`: all cell comment attributes. returned json string. see **attribute_name**. (**type** required `string`)
+* `cell_type`: cell type. returned Cell.getCellType() of POI.
+* `cell_cached_type`: cell cached formula result type. returned Cell.getCachedFormulaResultType() of POI when CellType==FORMULA, otherwise same as `cell_type` (returned Cell.getCellType()).
 * `sheet_name`: sheet name.
 * `row_number`: row number(1 origin).
 * `column_number`: column number(1 origin).
@@ -63,7 +74,9 @@ if omit **column_number** when **valus** is `cell_value`, specified next column.
   * `constant.`*value*: specified value.
   * `constant`: null.
 
-### column_number
+### cell_column
+
+Basically used for **record_type**=`row`.
 
 * `A`,`B`,`C`,...: column number of "A1 format".
 * *number*: column number (1 origin).
@@ -76,6 +89,12 @@ if omit **column_number** when **valus** is `cell_value`, specified next column.
 * `=`: same column.
 * `=`*name*: same column of name.
 
+### cell_row
+
+Basically used for **record_type**=`column`.
+
+* *number*: row number (1 origin).
+
 ### attribute_name
 
 **value**が`cell_style`, `cell_font`, `cell_comment`のとき、デフォルトでは、全属性を取得してJSON文字列に変換します。  
@@ -83,7 +102,7 @@ if omit **column_number** when **valus** is `cell_value`, specified next column.
 
 ```yaml
     columns:
-    - {name: foo, type: string, column_number: A, value: cell_style}
+    - {name: foo, type: string, cell_column: A, value: cell_style}
 ```
 
 
@@ -93,7 +112,7 @@ attribute_nameを指定することで、指定された属性だけを取得し
 
 ```yaml
     columns:
-    - {name: foo, type: string, column_number: A, value: cell_style, attribute_name: [border_top, border_bottom, border_left, border_right]}
+    - {name: foo, type: string, cell_column: A, value: cell_style, attribute_name: [border_top, border_bottom, border_left, border_right]}
 ```
 
 
@@ -106,8 +125,8 @@ attribute_nameを指定することで、指定された属性だけを取得し
     - {name: bar, type: long, value: cell_font.color}
 ```
 
-なお、`cell_style`や`cell_font`では、**column_number**を省略した場合は直前と同じ列を対象とします。  
-（`cell_value`では、**column_number**を省略すると次の列に移る）
+なお、`cell_style`や`cell_font`では、**cell_column**を省略した場合は直前と同じ列を対象とします。  
+（`cell_value`では、**cell_column**を省略すると次の列に移る）
 
 
 ### on_cell_error
@@ -116,56 +135,96 @@ Processing method of Cell error (`#DIV/0!`, `#REF!`, etc).
 
 ```yaml
     columns:
-    - {name: foo, type: string, column_number: A, value: cell_value, on_cell_error: error_code}
+    - {name: foo, type: string, cell_column: A, value: cell_value, on_cell_error: error_code}
 ```
 
 * `constant`: set null. (default)
-* `constant.`*value*: set value.
+* `constant.`*value*: set specified value.
 * `error_code`: set error code.
 * `exception`: throw exception.
 
 
+### formula_handling
+
+Processing method of formula.
+
+```yaml
+    columns:
+    - {name: foo, type: string, cell_column: A, value: cell_value, formula_handling: cashed_value}
+```
+
+* `evaluate`: evaluate formula. (default)
+* `cashed_value`: cashed value in cell.
+
+
+### on_evaluate_error
+
+Processing method of evaluate formula error.
+
+```yaml
+    columns:
+    - {name: foo, type: string, cell_column: A, value: cell_value, on_evaluate_error: constant}
+```
+
+* `constant`: set null.
+* `constant.`*value*: set specified value.
+* `exception`: throw exception. (default)
+
+
+### formula_replace
+
+Replace formula before evaluate.
+
+```yaml
+    columns:
+    - {name: foo, type: string, cell_column: A, value: cell_value, formula_replace: [{regex: aaa, to: "A${row}"}, {regex: bbb, to: "B${row}"}]}
+```
+
+`${row}` is replaced with the current row number.
+`${column}` is replaced with the current column string.
+
+
 ### on_convert_error
 
 Processing method of convert error. ex) Excel boolean to Embulk timestamp
 
 ```yaml
     columns:
-    - {name: foo, type: timestamp, format: "%Y/%m/%d", column_number: A, value: cell_value, on_convert_error: constant.9999/12/31}
+    - {name: foo, type: timestamp, format: "%Y/%m/%d", cell_column: A, value: cell_value, on_convert_error: constant.9999/12/31}
 ```
 
 * `constant`: set null.
-* `constant.`*value*: set value.
+* `constant.`*value*: set specified value.
 * `exception`: throw exception. (default)
 
 
 ### sheet_options
 
-Options of indivisual sheet.
+Options of individual sheet.
 
 ```yaml
   parser:
     type: poi_excel
     sheets: [Sheet1, Sheet2]
     columns:
-    - {name: date, type: timestamp, column_number: A}
+    - {name: date, type: timestamp, cell_column: A}
     - {name: foo, type: string}
     - {name: bar, type: long}
     sheet_options:
       Sheet1:
         skip_header_lines: 1
         columns:
-          foo: {column_number: B}
-          bar: {column_number: C}
+          foo: {cell_column: B}
+          bar: {cell_column: C}
       Sheet2:
         skip_header_lines: 0
         columns:
-          foo: {column_number: D}
+          foo: {cell_column: D}
           bar: {value: constant.0}
 ```
 
 **sheet_options** is map of sheet name.  
-Map values are **skip_header_lines**, **colums**.
+Map values are **skip_header_lines**, **columns**.
 
 **columns** is map of column name.  
 Map values are same **columns** in **parser** (excluding `name`, `type`).

data/build.gradle

@@ -1,6 +1,6 @@
 plugins {
     id "com.jfrog.bintray" version "1.1"
-    id "com.github.jruby-gradle.base" version "0.1.5"
+    id "com.github.jruby-gradle.base" version "1.5.0"
     id "java"
     id "eclipse"
 }
@@ -13,7 +13,7 @@ configurations {
     provided
 }
 
-version = "0.1.5"
+version = "0.1.12"
 
 sourceCompatibility = 1.7
 targetCompatibility = 1.7
@@ -23,8 +23,8 @@ dependencies {
     provided "org.embulk:embulk-core:0.7.5"
     compile  "org.embulk:embulk-standards:0.7.5"
     // compile "YOUR_JAR_DEPENDENCY_GROUP:YOUR_JAR_DEPENDENCY_MODULE:YOUR_JAR_DEPENDENCY_VERSION"
-    compile group: 'org.apache.poi', name : 'poi', version: '3.13'
-    compile(group: 'org.apache.poi', name : 'poi-ooxml', version: '3.13') {
+    compile group: 'org.apache.poi', name : 'poi', version: '3.17'
+    compile(group: 'org.apache.poi', name : 'poi-ooxml', version: '3.17') {
         exclude group: 'stax', module: 'stax-api'
     }
     testCompile "junit:junit:4.+"
@@ -38,19 +38,29 @@ task classpath(type: Copy, dependsOn: ["jar"]) {
 clean { delete "classpath" }
 
 task gem(type: JRubyExec, dependsOn: ["gemspec", "classpath"]) {
-    jrubyArgs "-rrubygems/gem_runner", "-eGem::GemRunner.new.run(ARGV)", "build"
-    script "${project.name}.gemspec"
+    jrubyArgs "-S"
+    script "gem"
+    scriptArgs "build", "${project.name}.gemspec"
     doLast { ant.move(file: "${project.name}-${project.version}.gem", todir: "pkg") }
 }
 
 task gemPush(type: JRubyExec, dependsOn: ["gem"]) {
-    jrubyArgs "-rrubygems/gem_runner", "-eGem::GemRunner.new.run(ARGV)", "push"
-    script "pkg/${project.name}-${project.version}.gem"
+    jrubyArgs "-S"
+    script "gem"
+    scriptArgs "push", "pkg/${project.name}-${project.version}.gem"
 }
 
-task "package"(dependsOn: ["gemspec", "classpath"]) << {
-    println "> Build succeeded."
-    println "> You can run embulk with '-L ${file(".").absolutePath}' argument."
+task "package"(dependsOn: ["gemspec", "classpath"]) {
+    doLast {
+        println "> Build succeeded."
+        println "> You can run embulk with '-L ${file(".").absolutePath}' argument."
+    }
+}
+
+task gemUnpack(type: JRubyExec) {
+    jrubyArgs "-S"
+    script "gem"
+    scriptArgs "unpack", "pkg/${project.name}-${project.version}.gem"
 }
 
 task gemspec {

data/gradle/wrapper/gradle-wrapper.properties

@@ -1,6 +1,5 @@
-#Wed Feb 04 13:46:12 PST 2015
-distributionBase=GRADLE_USER_HOME
-distributionPath=wrapper/dists
-zipStoreBase=GRADLE_USER_HOME
-zipStorePath=wrapper/dists
-distributionUrl=https\://services.gradle.org/distributions/gradle-2.2.1-bin.zip
+distributionBase=GRADLE_USER_HOME
+distributionPath=wrapper/dists
+zipStoreBase=GRADLE_USER_HOME
+zipStorePath=wrapper/dists
+distributionUrl=https\://services.gradle.org/distributions/gradle-4.1-bin.zip

data/gradlew

@@ -1,4 +1,4 @@
-#!/usr/bin/env bash
+#!/usr/bin/env sh
 
 ##############################################################################
 ##
@@ -6,20 +6,38 @@
 ##
 ##############################################################################
 
-# Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
-DEFAULT_JVM_OPTS=""
+# Attempt to set APP_HOME
+# Resolve links: $0 may be a link
+PRG="$0"
+# Need this for relative symlinks.
+while [ -h "$PRG" ] ; do
+    ls=`ls -ld "$PRG"`
+    link=`expr "$ls" : '.*-> \(.*\)$'`
+    if expr "$link" : '/.*' > /dev/null; then
+        PRG="$link"
+    else
+        PRG=`dirname "$PRG"`"/$link"
+    fi
+done
+SAVED="`pwd`"
+cd "`dirname \"$PRG\"`/" >/dev/null
+APP_HOME="`pwd -P`"
+cd "$SAVED" >/dev/null
 
 APP_NAME="Gradle"
 APP_BASE_NAME=`basename "$0"`
 
+# Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
+DEFAULT_JVM_OPTS=""
+
 # Use the maximum available, or set MAX_FD != -1 to use that value.
 MAX_FD="maximum"
 
-warn ( ) {
+warn () {
     echo "$*"
 }
 
-die ( ) {
+die () {
     echo
     echo "$*"
     echo
@@ -30,6 +48,7 @@ die ( ) {
 cygwin=false
 msys=false
 darwin=false
+nonstop=false
 case "`uname`" in
   CYGWIN* )
     cygwin=true
@@ -40,31 +59,11 @@ case "`uname`" in
   MINGW* )
     msys=true
     ;;
+  NONSTOP* )
+    nonstop=true
+    ;;
 esac
 
-# For Cygwin, ensure paths are in UNIX format before anything is touched.
-if $cygwin ; then
-    [ -n "$JAVA_HOME" ] && JAVA_HOME=`cygpath --unix "$JAVA_HOME"`
-fi
-
-# Attempt to set APP_HOME
-# Resolve links: $0 may be a link
-PRG="$0"
-# Need this for relative symlinks.
-while [ -h "$PRG" ] ; do
-    ls=`ls -ld "$PRG"`
-    link=`expr "$ls" : '.*-> \(.*\)$'`
-    if expr "$link" : '/.*' > /dev/null; then
-        PRG="$link"
-    else
-        PRG=`dirname "$PRG"`"/$link"
-    fi
-done
-SAVED="`pwd`"
-cd "`dirname \"$PRG\"`/" >&-
-APP_HOME="`pwd -P`"
-cd "$SAVED" >&-
-
 CLASSPATH=$APP_HOME/gradle/wrapper/gradle-wrapper.jar
 
 # Determine the Java command to use to start the JVM.
@@ -90,7 +89,7 @@ location of your Java installation."
 fi
 
 # Increase the maximum file descriptors if we can.
-if [ "$cygwin" = "false" -a "$darwin" = "false" ] ; then
+if [ "$cygwin" = "false" -a "$darwin" = "false" -a "$nonstop" = "false" ] ; then
     MAX_FD_LIMIT=`ulimit -H -n`
     if [ $? -eq 0 ] ; then
         if [ "$MAX_FD" = "maximum" -o "$MAX_FD" = "max" ] ; then
@@ -114,6 +113,7 @@ fi
 if $cygwin ; then
     APP_HOME=`cygpath --path --mixed "$APP_HOME"`
     CLASSPATH=`cygpath --path --mixed "$CLASSPATH"`
+    JAVACMD=`cygpath --unix "$JAVACMD"`
 
     # We build the pattern for arguments to be converted via cygpath
     ROOTDIRSRAW=`find -L / -maxdepth 1 -mindepth 1 -type d 2>/dev/null`
@@ -154,11 +154,19 @@ if $cygwin ; then
     esac
 fi
 
-# Split up the JVM_OPTS And GRADLE_OPTS values into an array, following the shell quoting and substitution rules
-function splitJvmOpts() {
-    JVM_OPTS=("$@")
+# Escape application args
+save () {
+    for i do printf %s\\n "$i" | sed "s/'/'\\\\''/g;1s/^/'/;\$s/\$/' \\\\/" ; done
+    echo " "
 }
-eval splitJvmOpts $DEFAULT_JVM_OPTS $JAVA_OPTS $GRADLE_OPTS
-JVM_OPTS[${#JVM_OPTS[*]}]="-Dorg.gradle.appname=$APP_BASE_NAME"
+APP_ARGS=$(save "$@")
+
+# Collect all arguments for the java command, following the shell quoting and substitution rules
+eval set -- $DEFAULT_JVM_OPTS $JAVA_OPTS $GRADLE_OPTS "\"-Dorg.gradle.appname=$APP_BASE_NAME\"" -classpath "\"$CLASSPATH\"" org.gradle.wrapper.GradleWrapperMain "$APP_ARGS"
+
+# by default we should be in the correct project dir, but when run from Finder on Mac, the cwd is wrong
+if [ "$(uname)" = "Darwin" ] && [ "$HOME" = "$PWD" ]; then
+  cd "$(dirname "$0")"
+fi
 
-exec "$JAVACMD" "${JVM_OPTS[@]}" -classpath "$CLASSPATH" org.gradle.wrapper.GradleWrapperMain "$@"
+exec "$JAVACMD" "$@"

data/gradlew.bat

@@ -8,14 +8,14 @@
 @rem Set local scope for the variables with windows NT shell
 if "%OS%"=="Windows_NT" setlocal
 
-@rem Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
-set DEFAULT_JVM_OPTS=
-
 set DIRNAME=%~dp0
 if "%DIRNAME%" == "" set DIRNAME=.
 set APP_BASE_NAME=%~n0
 set APP_HOME=%DIRNAME%
 
+@rem Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
+set DEFAULT_JVM_OPTS=
+
 @rem Find java.exe
 if defined JAVA_HOME goto findJavaFromJavaHome
 
@@ -46,10 +46,9 @@ echo location of your Java installation.
 goto fail
 
 :init
-@rem Get command-line arguments, handling Windowz variants
+@rem Get command-line arguments, handling Windows variants
 
 if not "%OS%" == "Windows_NT" goto win9xME_args
-if "%@eval[2+2]" == "4" goto 4NT_args
 
 :win9xME_args
 @rem Slurp the command line arguments.
@@ -60,11 +59,6 @@ set _SKIP=2
 if "x%~1" == "x" goto execute
 
 set CMD_LINE_ARGS=%*
-goto execute
-
-:4NT_args
-@rem Get arguments from the 4NT Shell from JP Software
-set CMD_LINE_ARGS=%$
 
 :execute
 @rem Setup the command line