Merge branch 'master' of https://github.com/dfinke/ImportExcel into add-helper-functions

This commit is contained in:
dfinke
2022-04-30 08:21:18 -04:00
13 changed files with 263 additions and 89 deletions

View File

@@ -0,0 +1,7 @@
Import-Module $PSScriptRoot\..\..\ImportExcel.psd1 -Force
$xlfile = "$PSScriptRoot\yearlySales.xlsx"
$result = Import-Excel $xlfile * -NotAsDictionary
$result | Measure-Object

View File

@@ -0,0 +1,9 @@
Import-Module $PSScriptRoot\..\..\ImportExcel.psd1 -Force
$xlfile = "$PSScriptRoot\yearlySales.xlsx"
$result = Import-Excel $xlfile *
foreach ($sheet in $result.Values) {
$sheet
}

Binary file not shown.

View File

@@ -6,7 +6,7 @@
RootModule = 'ImportExcel.psm1'
# Version number of this module.
ModuleVersion = '7.4.3'
ModuleVersion = '7.5.0'
# ID used to uniquely identify this module
GUID = '60dd4136-feff-401a-ba27-a84458c57ede'
@@ -49,6 +49,7 @@ Check out the How To Videos https://www.youtube.com/watch?v=U3Ne_yX4tYo&list=PL5
'Copy-ExcelWorksheet',
'DoChart',
'Enable-ExcelAutoFilter',
'Enable-ExcelAutofit',
'Expand-NumberFormat',
'Export-Excel',
'Export-ExcelSheet',
@@ -65,7 +66,6 @@ Check out the How To Videos https://www.youtube.com/watch?v=U3Ne_yX4tYo&list=PL5
'Import-UPS',
'Import-USPS',
'Invoke-AllTests',
'Enable-ExcelAutofit',
'Invoke-ExcelQuery',
'Invoke-Sum',
'Join-Worksheet',
@@ -83,8 +83,8 @@ Check out the How To Videos https://www.youtube.com/watch?v=U3Ne_yX4tYo&list=PL5
'PieChart',
'Pivot',
'Read-Clipboard',
'ReadClipboardImpl',
'Read-OleDbData',
'ReadClipboardImpl',
'Remove-Worksheet',
'Select-Worksheet',
'Send-SQLDataToExcel',

View File

@@ -15,7 +15,7 @@
[Alias('Sheet')]
[Parameter(Position = 1)]
[ValidateNotNullOrEmpty()]
[String]$WorksheetName,
[String[]]$WorksheetName,
[Parameter(ParameterSetName = 'PathB' , Mandatory)]
[Parameter(ParameterSetName = 'PackageB', Mandatory)]
[String[]]$HeaderName ,
@@ -36,7 +36,8 @@
[string[]]$AsDate,
[ValidateNotNullOrEmpty()]
[String]$Password,
[Int[]]$ImportColumns
[Int[]]$ImportColumns,
[Switch]$NotAsDictionary
)
end {
$sw = [System.Diagnostics.Stopwatch]::StartNew()
@@ -64,7 +65,7 @@
try {
if ($ImportColumns) {
$end = $Worksheet.Dimension.End.Column
$end = $sheet.Dimension.End.Column
# Check $ImportColumns
if ($ImportColumns[0] -le 0) { throw "The first entry in ImportColumns must be equal or greater 1" ; return }
# Check $StartColumn and $EndColumn
@@ -95,7 +96,7 @@
foreach ($C in $Columns) {
#allow "False" or "0" to be column headings
$Worksheet.Cells[$StartRow, $C] | Where-Object { -not [string]::IsNullOrEmpty($_.Value) } | Select-Object @{N = 'Column'; E = { $C } }, Value
$sheet.Cells[$StartRow, $C] | Where-Object { -not [string]::IsNullOrEmpty($_.Value) } | Select-Object @{N = 'Column'; E = { $C } }, Value
}
}
}
@@ -125,108 +126,128 @@
}
try {
#Select worksheet
if (-not $WorksheetName) { $Worksheet = $ExcelPackage.Workbook.Worksheets[1] }
if ($WorksheetName -eq '*') { $Worksheet = $ExcelPackage.Workbook.Worksheets }
elseif (-not $WorksheetName) { $Worksheet = $ExcelPackage.Workbook.Worksheets[1] }
elseif (-not ($Worksheet = $ExcelPackage.Workbook.Worksheets[$WorksheetName])) {
throw "Worksheet '$WorksheetName' not found, the workbook only contains the worksheets '$($ExcelPackage.Workbook.Worksheets)'. If you only wish to select the first worksheet, please remove the '-WorksheetName' parameter." ; return
}
#region Get rows and columns
#If we are doing dataonly it is quicker to work out which rows to ignore before processing the cells.
if (-not $EndRow ) { $EndRow = $Worksheet.Dimension.End.Row }
if (-not $EndColumn) { $EndColumn = $Worksheet.Dimension.End.Column }
$endAddress = [OfficeOpenXml.ExcelAddress]::TranslateFromR1C1("R[$EndRow]C[$EndColumn]", 0, 0)
if ($DataOnly) {
#If we are using headers startrow will be the header-row so examine data from startRow + 1,
if ($NoHeader) { $range = "A" + ($StartRow ) + ":" + $endAddress }
else { $range = "A" + ($StartRow + 1 ) + ":" + $endAddress }
#We're going to look at every cell and build 2 hash tables holding rows & columns which contain data.
#Want to Avoid 'select unique' operations & large Sorts, becuse time time taken increases with square
#of number of items (PS uses heapsort at large size). Instead keep a list of what we have seen,
#using Hash tables: "we've seen it" is all we need, no need to worry about "seen it before" / "Seen it many times".
$colHash = @{ }
$rowHash = @{ }
foreach ($cell in $Worksheet.Cells[$range]) {
if ($null -ne $cell.Value ) { $colHash[$cell.Start.Column] = 1; $rowHash[$cell.Start.row] = 1 }
}
$rows = ( $StartRow..$EndRow ).Where( { $rowHash[$_] })
$columns = ($StartColumn..$EndColumn).Where( { $colHash[$_] })
}
else {
$Columns = $StartColumn .. $EndColumn ; if ($StartColumn -gt $EndColumn) { Write-Warning -Message "Selecting columns $StartColumn to $EndColumn might give odd results." }
if ($NoHeader) { $rows = $StartRow..$EndRow ; if ($StartRow -gt $EndRow) { Write-Warning -Message "Selecting rows $StartRow to $EndRow might give odd results." } }
elseif ($HeaderName) { $rows = $StartRow..$EndRow }
else {
$rows = (1 + $StartRow)..$EndRow
if ($StartRow -eq 1 -and $EndRow -eq 1) {
$rows = 0
$xlBook = [Ordered]@{}
foreach ($sheet in $Worksheet) {
$targetSheetname = $sheet.Name
$xlBook["$targetSheetname"] = @()
#region Get rows and columns
#If we are doing dataonly it is quicker to work out which rows to ignore before processing the cells.
if (-not $EndRow ) { $EndRow = $sheet.Dimension.End.Row }
if (-not $EndColumn) { $EndColumn = $sheet.Dimension.End.Column }
$endAddress = [OfficeOpenXml.ExcelAddress]::TranslateFromR1C1("R[$EndRow]C[$EndColumn]", 0, 0)
if ($DataOnly) {
#If we are using headers startrow will be the header-row so examine data from startRow + 1,
if ($NoHeader) { $range = "A" + ($StartRow ) + ":" + $endAddress }
else { $range = "A" + ($StartRow + 1 ) + ":" + $endAddress }
#We're going to look at every cell and build 2 hash tables holding rows & columns which contain data.
#Want to Avoid 'select unique' operations & large Sorts, becuse time time taken increases with square
#of number of items (PS uses heapsort at large size). Instead keep a list of what we have seen,
#using Hash tables: "we've seen it" is all we need, no need to worry about "seen it before" / "Seen it many times".
$colHash = @{ }
$rowHash = @{ }
foreach ($cell in $sheet.Cells[$range]) {
if ($null -ne $cell.Value ) { $colHash[$cell.Start.Column] = 1; $rowHash[$cell.Start.row] = 1 }
}
$rows = ( $StartRow..$EndRow ).Where( { $rowHash[$_] })
$columns = ($StartColumn..$EndColumn).Where( { $colHash[$_] })
}
else {
$Columns = $StartColumn .. $EndColumn ; if ($StartColumn -gt $EndColumn) { Write-Warning -Message "Selecting columns $StartColumn to $EndColumn might give odd results." }
if ($NoHeader) { $rows = $StartRow..$EndRow ; if ($StartRow -gt $EndRow) { Write-Warning -Message "Selecting rows $StartRow to $EndRow might give odd results." } }
elseif ($HeaderName) { $rows = $StartRow..$EndRow }
else {
$rows = (1 + $StartRow)..$EndRow
if ($StartRow -eq 1 -and $EndRow -eq 1) {
$rows = 0
}
}
# ; if ($StartRow -ge $EndRow) { Write-Warning -Message "Selecting $StartRow as the header with data in $(1+$StartRow) to $EndRow might give odd results." } }
}
#endregion
#region Create property names
if ((-not $Columns) -or (-not ($PropertyNames = Get-PropertyNames -Columns $Columns -StartRow $StartRow))) {
throw "No column headers found on top row '$StartRow'. If column headers in the worksheet are not a requirement then please use the '-NoHeader' or '-HeaderName' parameter."; return
}
if ($Duplicates = $PropertyNames | Group-Object Value | Where-Object Count -GE 2) {
throw "Duplicate column headers found on row '$StartRow' in columns '$($Duplicates.Group.Column)'. Column headers must be unique, if this is not a requirement please use the '-NoHeader' or '-HeaderName' parameter."; return
}
#endregion
if (-not $rows) {
Write-Warning "Worksheet '$WorksheetName' in workbook '$Path' contains no data in the rows after top row '$StartRow'"
}
else {
#region Create one object per row
if ($AsText -or $AsDate) {
<#join items in AsText together with ~~~ . Escape any regex special characters...
# ; if ($StartRow -ge $EndRow) { Write-Warning -Message "Selecting $StartRow as the header with data in $(1+$StartRow) to $EndRow might give odd results." } }
}
#endregion
#region Create property names
if ((-not $Columns) -or (-not ($PropertyNames = Get-PropertyNames -Columns $Columns -StartRow $StartRow))) {
throw "No column headers found on top row '$StartRow'. If column headers in the worksheet are not a requirement then please use the '-NoHeader' or '-HeaderName' parameter."; return
}
if ($Duplicates = $PropertyNames | Group-Object Value | Where-Object Count -GE 2) {
throw "Duplicate column headers found on row '$StartRow' in columns '$($Duplicates.Group.Column)'. Column headers must be unique, if this is not a requirement please use the '-NoHeader' or '-HeaderName' parameter."; return
}
#endregion
if (-not $rows) {
Write-Warning "Worksheet '$WorksheetName' in workbook '$Path' contains no data in the rows after top row '$StartRow'"
}
else {
#region Create one object per row
if ($AsText -or $AsDate) {
<#join items in AsText together with ~~~ . Escape any regex special characters...
# which turns "*" into "\*" make it ".*". Convert ~~~ to $|^ and top and tail with ^%;
So if we get "Week", "[Time]" and "*date*" ; make the expression ^week$|^\[Time\]$|^.*Date.*$
$make a regex for this which is case insensitive (option 1) and compiled (option 8)
#>
$TextColExpression = ''
if ($AsText) {
$TextColExpression += '(?<astext>^' + [regex]::Escape($AsText -join '~~~').replace('\*', '.*').replace('~~~', '$|^') + '$)'
}
if ($AsText -and $AsDate) {
$TextColExpression += "|"
}
if ($AsDate) {
$TextColExpression += '(?<asDate>^' + [regex]::Escape($AsDate -join '~~~').replace('\*', '.*').replace('~~~', '$|^') + '$)'
}
$TextColRegEx = New-Object -TypeName regex -ArgumentList $TextColExpression , 9
}
else { $TextColRegEx = $null }
foreach ($R in $rows) {
#Disabled write-verbose for speed
# Write-Verbose "Import row '$R'"
$NewRow = [Ordered]@{ }
if ($TextColRegEx) {
foreach ($P in $PropertyNames) {
$MatchTest = $TextColRegEx.Match($P.value)
if ($MatchTest.groups.name -eq "astext") {
$NewRow[$P.Value] = $Worksheet.Cells[$R, $P.Column].Text
}
elseif ($MatchTest.groups.name -eq "asdate" -and $Worksheet.Cells[$R, $P.Column].Value -is [System.ValueType]) {
$NewRow[$P.Value] = [datetime]::FromOADate(($Worksheet.Cells[$R, $P.Column].Value))
}
else { $NewRow[$P.Value] = $Worksheet.Cells[$R, $P.Column].Value }
$TextColExpression = ''
if ($AsText) {
$TextColExpression += '(?<astext>^' + [regex]::Escape($AsText -join '~~~').replace('\*', '.*').replace('~~~', '$|^') + '$)'
}
}
else {
foreach ($P in $PropertyNames) {
$NewRow[$P.Value] = $Worksheet.Cells[$R, $P.Column].Value
# Write-Verbose "Import cell '$($Worksheet.Cells[$R, $P.Column].Address)' with property name '$($p.Value)' and value '$($Worksheet.Cells[$R, $P.Column].Value)'."
if ($AsText -and $AsDate) {
$TextColExpression += "|"
}
if ($AsDate) {
$TextColExpression += '(?<asDate>^' + [regex]::Escape($AsDate -join '~~~').replace('\*', '.*').replace('~~~', '$|^') + '$)'
}
$TextColRegEx = New-Object -TypeName regex -ArgumentList $TextColExpression , 9
}
[PSCustomObject]$NewRow
else { $TextColRegEx = $null }
foreach ($R in $rows) {
#Disabled write-verbose for speed
# Write-Verbose "Import row '$R'"
$NewRow = [Ordered]@{ }
if ($TextColRegEx) {
foreach ($P in $PropertyNames) {
$MatchTest = $TextColRegEx.Match($P.value)
if ($MatchTest.groups.name -eq "astext") {
$NewRow[$P.Value] = $sheet.Cells[$R, $P.Column].Text
}
elseif ($MatchTest.groups.name -eq "asdate" -and $sheet.Cells[$R, $P.Column].Value -is [System.ValueType]) {
$NewRow[$P.Value] = [datetime]::FromOADate(($sheet.Cells[$R, $P.Column].Value))
}
else { $NewRow[$P.Value] = $sheet.Cells[$R, $P.Column].Value }
}
}
else {
foreach ($P in $PropertyNames) {
$NewRow[$P.Value] = $sheet.Cells[$R, $P.Column].Value
# Write-Verbose "Import cell '$($Worksheet.Cells[$R, $P.Column].Address)' with property name '$($p.Value)' and value '$($Worksheet.Cells[$R, $P.Column].Value)'."
}
}
$xlBook["$targetSheetname"] += [PSCustomObject]$NewRow
}
#endregion
}
#endregion
}
}
catch { throw "Failed importing the Excel workbook '$Path' with worksheet '$WorksheetName': $_"; return }
finally {
$EndRow = 0
$EndColumn = 0
if ($Path) { $stream.close(); $ExcelPackage.Dispose() }
if ($NotAsDictionary) {
foreach ($entry in $xlbook.GetEnumerator()) {
$entry.Value
}
}
elseif ($Worksheet.Count -eq 1) {
$xlBook["$targetSheetname"]
}
else {
$xlBook
}
}
}
}

View File

@@ -63,6 +63,10 @@ Describe "Import-Excel on a sheet with no headings" {
}
}
AfterAll {
Remove-Item $PSScriptRoot\testImportExcelSparse.xlsx -ErrorAction SilentlyContinue
}
It "Import-Excel should have this shape" {
$actual = @(Import-Excel $xlfile)
@@ -230,6 +234,7 @@ Describe "Import-Excel on a sheet with no headings" {
$actual.Count | Should -Be 1
Remove-Item $xlfile
# Looks like -DataOnly does not handle empty columns
# $actual[0].FirstName | Should -BeExactly 'Jean-Claude'
# $actual[0].SecondName | Should -BeExactly 'Vandamme'

View File

@@ -0,0 +1,63 @@
#Requires -Modules Pester
[Diagnostics.CodeAnalysis.SuppressMessageAttribute('PSUseDeclaredVarsMoreThanAssignments', '', Justification = 'False Positives')]
param()
Import-Module $PSScriptRoot\..\..\ImportExcel.psd1 -Force
Describe 'Different ways to import sheets' -Tag ImportExcelReadSheets {
BeforeAll {
$xlFilename = "$PSScriptRoot\yearlySales.xlsx"
}
Context 'Test reading sheets' {
It 'Should read one sheet' {
$actual = Import-Excel $xlFilename
$actual.Count | Should -Be 100
$actual[0].Month | Should -BeExactly "April"
$actual[99].Month | Should -BeExactly "April"
}
It 'Should read two sheets' {
$actual = Import-Excel $xlFilename march, june
$actual.keys.Count | Should -Be 2
$actual["March"].Count | Should -Be 100
$actual["June"].Count | Should -Be 100
}
It 'Should read all the sheets' {
$actual = Import-Excel $xlFilename *
$actual.keys.Count | Should -Be 12
$actual["January"].Count | Should -Be 100
$actual["February"].Count | Should -Be 100
$actual["March"].Count | Should -Be 100
$actual["April"].Count | Should -Be 100
$actual["May"].Count | Should -Be 100
$actual["June"].Count | Should -Be 100
$actual["July"].Count | Should -Be 100
$actual["August"].Count | Should -Be 100
$actual["September"].Count | Should -Be 100
$actual["October"].Count | Should -Be 100
$actual["November"].Count | Should -Be 100
$actual["December"].Count | Should -Be 100
}
It 'Should throw if it cannot find the sheet' {
{ Import-Excel $xlFilename april, june, notthere } | Should -Throw
}
It 'Should return an array not a dictionary' {
$actual = Import-Excel $xlFilename april, june -NotAsDictionary
$actual.Count | Should -Be 200
$group = $actual | Group-Object month -NoElement
$group.Count | Should -Be 2
$group[0].Name | Should -BeExactly 'April'
$group[1].Name | Should -BeExactly 'June'
}
}
}

View File

@@ -0,0 +1,57 @@
[Diagnostics.CodeAnalysis.SuppressMessageAttribute('PSUseDeclaredVarsMoreThanAssignments', '', Justification = 'False Positives')]
Param()
Import-Module $PSScriptRoot\..\..\ImportExcel.psd1 -Force
Describe "Test reading multiple XLSX files of differernt row count" -Tag ReadMultipleXLSX {
It "Should find these xlsx files" {
Test-Path -Path $PSScriptRoot\rows05.xlsx | Should -BeTrue
Test-Path -Path $PSScriptRoot\rows10.xlsx | Should -BeTrue
}
It "Should find two xlsx files" {
(Get-ChildItem $PSScriptRoot\row*xlsx).Count | Should -Be 2
}
It "Should get 5 rows" {
(Import-Excel $PSScriptRoot\rows05.xlsx).Count | Should -Be 5
}
It "Should get 10 rows" {
(Import-Excel $PSScriptRoot\rows10.xlsx).Count | Should -Be 10
}
It "Should get 15 rows" {
$actual = Get-ChildItem $PSScriptRoot\row*xlsx | Import-Excel
$actual.Count | Should -Be 15
}
It "Should get 4 property names" {
$actual = Get-ChildItem $PSScriptRoot\row*xlsx | Import-Excel
$names = $actual[0].psobject.properties.name
$names.Count | Should -Be 4
$names[0] | Should -BeExactly "Region"
$names[1] | Should -BeExactly "State"
$names[2] | Should -BeExactly "Units"
$names[3] | Should -BeExactly "Price"
}
It "Should have the correct data" {
$actual = Get-ChildItem $PSScriptRoot\row*xlsx | Import-Excel
# rows05.xlsx
$actual[0].Region | Should -BeExactly "South"
$actual[0].Price | Should -Be 181.52
$actual[4].Region | Should -BeExactly "West"
$actual[4].Price | Should -Be 216.56
# rows10.xlsx
$actual[5].Region | Should -BeExactly "South"
$actual[5].Price | Should -Be 199.85
$actual[14].Region | Should -BeExactly "East"
$actual[14].Price | Should -Be 965.25
}
}

Binary file not shown.

Binary file not shown.

Binary file not shown.

Binary file not shown.

View File

@@ -1,3 +1,15 @@
# v7.5.0
## Fixes
- Importing multiple files with Import-Excel by pipeline uses only the first file for the row count https://github.com/dfinke/ImportExcel/issues/1172
## New Features
- Import-Excel now supports importing multiple sheets. It can either return a dictionary of all sheets, or as a single array of all sheets combined.
- `Import-Excel $xlfile *` # reads all sheets, returns all data in a dictionary
- `Import-Excel $xlfile * -NotAsDictionary` # reads all sheets, returns all data in a single array
# v7.4.2
- Thank you [James Mueller](https://github.com/jamesmmueller) Updated `ConvertFrom-ExcelToSQLInsert` to handle single quotes in the SQL statement.