# Put-call parity under a regime-switching model

I need some help.
I'm given $$J$$ different regimes, each one characterized by its own parameters $$(r_i, \delta_i,\sigma_i,...)$$ with $$i\in \mathcal{J}= \{1,2,...,J\}$$ ($$r$$ = risk-free interest rate, $$\delta$$ = continuous dividend yield).
For instance, we can have regime $$1$$ under Black-Scholes model with parameters $$(r_1,\delta_1,\sigma_1)$$ and then regime $$2$$ under Variance-Gamma model with parameters $$(r_2,\delta_2,\sigma_2,\nu_2,\theta_2)$$.
Let $$\alpha_{t\{t\in[0,T]\}}$$ be the Markov chain taking values $$i\in \mathcal{J}$$ and $$Q := \{q_{ij},1 \leq i,j \leq J\} \in R^{J×J}$$ its associated intensity matrix (also known as generator), such that the matrix of transition probabilities is defined as $$P(\Delta t)=exp(Q \Delta t)$$.
In this framework, I need to price call options in Matlab exploiting the put-call parity, however if $$r$$ and $$\delta$$ were constant for all the regimes then, given that $$Prices$$ is the vector containing the put prices for each regime, I could write \begin{align} & for \ \ j=1:J \\ & \ \ \ \ \ \ Prices(j)=Prices(j)+S0*exp(-\delta*T)-K*exp(-r*T); \\ & end \end{align} where $$S0$$ is the spot price, $$K$$ is the strike price and $$T$$ the maturity.
How can I generalize the put-call parity with $$r_i$$ and $$\delta_i$$ depending on state $$i$$? Any help will be highly appreciated.