Last active
September 7, 2021 08:17
-
-
Save KiruyaMomochi/b3f99ef1e0f7ad795509217fd8bb19fa to your computer and use it in GitHub Desktop.
Checking Priconne (Taiwan) SRT data
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
param ( | |
[Parameter(ParameterSetName = "json", Mandatory)] | |
[string] | |
$SrtJson, | |
[Parameter(ParameterSetName = "objects", Mandatory, ValueFromPipeline)] | |
[pscustomobject[]] | |
$SrtPanels | |
) | |
class SrtPanel { | |
[int]$reading_id | |
[string]$reading | |
[int]$read_type | |
[int]$panel_id | |
[string]$head_symbol | |
[string]$tail_symbol | |
[string]$detail_text | |
[int]$version | |
} | |
function Get-Bopomofo { | |
[OutputType([string[]])] | |
param ( | |
[Parameter(Mandatory, ValueFromPipeline)] | |
[string] | |
$word | |
) | |
begin { | |
$bopomofo = "ㄅㄆㄇㄈㄉㄊㄋㄌㄍㄎㄏㄐㄑㄒㄓㄔㄕㄖㄗㄘㄙㄧㄨㄩㄚㄛㄜㄝㄞㄟㄠㄡㄢㄣㄤㄥㄦ".ToCharArray(); | |
} | |
process { | |
if ($word -in $bopomofo) { | |
return $word; | |
} | |
$result = Invoke-RestMethod ("https://www.moedict.tw/uni/" + $word) -Proxy "http://127.0.0.1:10809"; | |
$other = $result.heteronyms.definitions.def | |
| ForEach-Object { return [regex]::Matches($_, '「(.+?)」的異體字。') } | |
| ForEach-Object { return $_.Groups[1].Value } | |
| Foreach-Object { return Get-Bopomofo $_}; | |
Write-Host $result.heteronyms.definitions.def; | |
[string[]] $result = $result.heteronyms.bopomofo | Where-Object { $_ } | |
$result += $other; | |
return $result | |
} | |
} | |
function Get-Rhyme { | |
param ( | |
[Parameter(Mandatory, ValueFromPipeline)] | |
[string] | |
$bopomofo | |
) | |
process { | |
$keep = "[^ㄓㄔㄕㄖㄗㄘㄙㄧㄨㄩㄚㄛㄜㄝㄞㄟㄠㄡㄢㄣㄤㄥㄦ]"; | |
$rhyme = [regex]::Replace($bopomofo, $keep, ''); | |
$remove = "^[ㄓㄔㄕㄖㄗㄘㄙ]?(.+)$"; | |
$rhyme = [regex]::Replace($rhyme, $remove, '$1'); | |
return $rhyme; | |
} | |
} | |
function Check { | |
param ( | |
[Parameter(Mandatory, ValueFromPipeline)] | |
$srt | |
) | |
begin { | |
$nonRhyme = "[ㄅㄆㄇㄈㄉㄊㄋㄌㄍㄎㄏㄐㄑㄒ]"; | |
} | |
process { | |
$id = $srt.reading_id; | |
$reading = [regex]::Replace($srt.reading, $nonRhyme, ''); | |
$sonet_head = $srt.head_symbol; | |
$sonet_tail = $srt.tail_symbol; | |
try { | |
$moe_head = Get-Bopomofo ($reading[0]) | Get-Rhyme | Get-Unique; | |
$moe_tail = Get-Bopomofo ($reading[-1]) | Get-Rhyme | Get-Unique; | |
} | |
catch { | |
Write-Host "? [$id] ${reading}: $($reading[0]) $sonet_head($moe_head), $($reading[-1]) $sonet_tail($moe_tail)" -ForegroundColor Yellow; | |
continue; | |
} | |
if (($sonet_head -notin $moe_head) -or ($sonet_tail -notin $moe_tail)) { | |
Write-Host "- [$id] ${reading}:" -NoNewline -ForegroundColor Red; | |
Write-Host " $($reading[0]) $sonet_head" -NoNewline; | |
if ($sonet_head -notin $moe_head) { | |
Write-Host "($moe_head)" -NoNewline -ForegroundColor Red; | |
} | |
Write-Host ", $($reading[-1]) $sonet_tail" -NoNewline; | |
if ($sonet_tail -notin $moe_tail) { | |
Write-Host "($moe_tail)" -NoNewline -ForegroundColor Red; | |
} | |
Write-Host | |
} | |
else { | |
Write-Host "o [$id] ${reading}: $($reading[0]) $sonet_head($moe_head), $($reading[-1]) $sonet_tail($moe_tail)" -ForegroundColor DarkGray; | |
} | |
} | |
} | |
if ($SrtJson.Length -ne 0) { | |
$script:SrtPanels = [SrtPanel[]](Get-Content $SrtJson | ConvertFrom-Json) | |
} | |
$script:SrtPanels | Check |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment