[SQL] Filtriranje

A kdo ve kako bi lahko odfiltriral izpis tako, da mi izpiše zadnjo vrstico po datumu tako, da je [Jezik] po defaultu 'SLO' in če ni da izpiše 'ANG'.
Zaželeno pa da se ne segmentira v dodaten select.
Je možno da pred prvim "WHERE" preuredi vrstice tako, da je SLO pred ANG ali obratno?

WITH [FilteredSource] AS (
SELECT TOP 1000	[MaticniPodatki].[MpSifra] AS [Šifra],
	[MaticniPodatki].[MpNaziv] AS [Naziv],
	[MaticniPodatki].[MpDoNaziv] AS [Dolgi naziv],
	[MaticniPodatki].[MpRisbaFName] AS [Image],
	REPLACE([MaticniPodatki].[MpSifEnoteMere1], CAST('KOS' AS CHAR(3)), CAST('PCS' AS CHAR(3))) AS [Unit],
	[NFaktura].[Datum] AS [Datum],
	[NNalP].[NNaPStNar] AS [Nabavni nalog],
	[NNalP].[NNaPStPos] AS [Posx],
	[NNalP].[NNaPSifra] AS [KosId],
	[NNalP].[NNaPKolNar] AS [QTY],
	[KomOpisMat].[KOMJezik] AS [Jezik],
	[KomOpisMat].[KOMSifMp] AS [KOMSifMp],
	ISNULL([KomOpisMat].[KOMOpis], [MaticniPodatki].[MpNaziv]) AS [Kom. opis],
	ISNULL([KomOpisMat].[KOMDaljsiOpis], [MaticniPodatki].[MpDoNaziv]) AS [Kom. daljši opis],
	[NFakP].[NFaPKol] AS [NFaPKol],
	[NFakP].[NFaPZnesekDav] AS [NFaPZnesekDav],
	[NFakP].[NFaPZnesekSkupaj] AS [NFaPZnesekSkupaj],
	ROW_NUMBER() OVER (PARTITION BY [MaticniPodatki].[MpSifra] ORDER BY [NFaktura].[Datum] DESC) AS [Partition_MaticniPodatki_MpSifra]
FROM [NNalP]
	FULL JOIN [MaticniPodatki] ON [NNalP].[NNaPSifra] = [MaticniPodatki].[MpSifra]
	FULL JOIN [KomOpisMat] ON [MaticniPodatki].[MpSifra] = [KomOpisMat].[KOMSifMp]
	FULL JOIN [NFakP] ON [MaticniPodatki].[MpSifra] = [NFakP].[NFaPSifra]
	FULL JOIN [NFaktura] ON [NFakP].[NFaPStNFak] = [NFaktura].[NFaZapSt]
WHERE	 ([NNalP].[NNaPStNar] = PARSE('200044' AS INT USING 'Sl-SI'))
	AND ([KomOpisMat].[KOMJezik] LIKE CAST('%SLO%' AS CHAR(5)) 
	OR [KomOpisMat].[KOMJezik] LIKE CAST('%ANG%' AS CHAR(5)) 
	OR [KomOpisMat].[KOMJezik] IS NULL
	)
	ORDER BY [KomOpisMat].[KOMJezik] DESC
)
SELECT TOP 1000
	[FilteredSource].[Partition_MaticniPodatki_MpSifra] AS [Part],
	[FilteredSource].[Šifra] AS [Šifra],
	[FilteredSource].[Kom. opis] AS [Kom. opis],
	[FilteredSource].[Kom. daljši opis] AS [Kom. daljši opis],
	[FilteredSource].[Image] AS [Image],
	[FilteredSource].[QTY] AS [QTY],
	[FilteredSource].[Unit] AS [Unit],
	( SUM([FilteredSource].[NFaPZnesekSkupaj]) - SUM([FilteredSource].[NFaPZnesekDav]) ) / NULLIF( SUM([FilteredSource].[NFaPKol]) , 0) AS [Price],
	[FilteredSource].[Jezik] AS [Jezik]
FROM [FilteredSource]
--WHERE	[FilteredSource].[Partition_MaticniPodatki_MpSifra] <= 1
GROUP BY
	[FilteredSource].[Partition_MaticniPodatki_MpSifra],
	[FilteredSource].[Šifra],
	[FilteredSource].[Naziv],
	[FilteredSource].[Dolgi naziv],
	[FilteredSource].[Image],
	[FilteredSource].[Unit],
	[FilteredSource].[Datum],
	[FilteredSource].[Nabavni nalog],
	[FilteredSource].[Posx],
	[FilteredSource].[KosId],
	[FilteredSource].[QTY],
	[FilteredSource].[Jezik],
	[FilteredSource].[KOMSifMp],
	[FilteredSource].[Kom. opis],
	[FilteredSource].[Kom. daljši opis]
ORDER BY
	[Posx] ASC
;

Mogoče je ena opcija, da imaš dve [FilteredSource] tabeli:

- prva [FilteredSourceSLO], v kateri imaš tiste s slovenskim jezikom
- potem druga pa [FilteredSourceENG], kjer v where pogoju napišeš, da se ID ne sme nahaja v tabeli "SLO".

Se pravi, prvo zlistaj vse IDje z SLO (FilteredSourceSLO), ter potem zlistaj vse IDje z ENG, ki niso v FilteredSourceSLO.

Potem pa te dve tabele združiš:

SELECT * FROM [FilteredSourceSLO] AS `slo`
UNION
SELECT * FROM [FilteredSourceENG] AS `eng`

Torej naredim novo kolono, kjer v select oštevilčim po jeziku ter nato to kolono sortiram?
Bo potem obrnlo 138 in 139 vrstico ter izpisalo SLO pred ANG?

Morda bi šlo tudi tako:
ROW_NUMBER() OVER (PARTITION BY [MaticniPodatki].[MpSifra] ORDER BY [NFaktura].[Datum] as ST : s tem dobiš številšenje po datumu
dodaš tole vrstico:
count( ?ime polja?) over (partition by [MaticniPodatki].[MpSifra]) as TOTAL :s tem dobiš total glede na mp sifra

pol pa uporabiš CASE: to ti zamenja jezik le na zadnjem zapisu

case
when st=total then 'SLO'
else 'ANG'
end as jezik