changes parse_data() to remove trailing double underscores

[SteveBronder]

Summary:

Fixed for stan-dev/stanc3#865 so that rstan can correctly check the user data when the model uses Eigen::Map types.

Intended Effect:

Allow rstan to correctly parse data that's represented as an Eigen::Map

How to Verify:

I complied this rstan branch then ran the following R code

file.remove(system.file("stanc.js", package = "rstan"))
library(rstan)

model_code = "
data {
  int<lower=0> N;
  vector[N] weight;
  vector[N] diam1;
  vector[N] diam2;
  vector[N] canopy_height;
  vector[N] total_height;
  vector[N] density;
  vector[N] group;
}
transformed data {           // log transformations
  vector[N] log_weight = log(weight);
  vector[N] log_diam1 = log(diam1);
  vector[N] log_diam2 = log(diam2);
  vector[N] log_canopy_height = log(canopy_height);
  vector[N] log_total_height = log(total_height);
  vector[N] log_density = log(density);
  matrix[N,6] x = [log_diam1', log_diam2', log_canopy_height',
log_total_height', log_density', group']';
}
parameters {
  real alpha;
  vector[6] beta;
  real<lower=0> sigma;
}
model {
  log_weight ~ normal_id_glm(x, alpha, beta, sigma);
}
"

model_data = list(
  N = 46,
  weight = 
    c(401.3, 513.7, 1179.2, 308, 855.2, 268.7, 155.5, 1253.2, 328, 614.6,
      60.2, 269.6, 448.4, 120.4, 378.7, 266.4, 138.9, 1020.8, 635.7, 621.8, 579.8,
      326.8, 66.7, 68, 153.1, 256.4, 723, 4052, 345, 330.9, 163.5, 1160, 386.6, 693.5,
      674.4, 217.5, 771.3, 341.7, 125.7, 462.5, 64.5, 850.6, 226, 1745.1, 908, 213.5),
  diam1 = 
    c(1.8, 1.7, 2.8, 1.3, 3.3, 1.4, 1.5, 3.9, 1.8, 2.1, 0.8, 1.3, 1.2,
      1.5, 2.8, 1.4, 1.5, 2.4, 1.9, 2.3, 2.1, 2.4, 1, 1.3, 1.1, 1.3, 2.5, 5.2, 2, 1.6,
      1.4, 3.2, 1.9, 2.4, 2.5, 2.1, 2.4, 2.4, 1.9, 2.7, 1.3, 2.9, 2.1, 4.1, 2.8, 1.27),
  diam2 = 
    c(1.15, 1.35, 2.55, 0.85, 1.9, 1.4, 0.5, 2.3, 1.35, 1.6, 0.63, 0.95,
      0.9, 0.7, 1.7, 0.85, 0.6, 2.4, 1.55, 1.6, 1.7, 1.3, 0.4, 0.6, 0.7, 1.2, 2.3, 4,
      1.6, 1.6, 1, 1.9, 1.8, 2.4, 1.8, 1.5, 2.2, 1.7, 1.2, 2.5, 1.1, 2.7, 1, 3.8, 2.5,
      1),
  canopy_height = 
    c(1, 1.33, 0.6, 1.2, 1.05, 1, 0.9, 1.3, 0.6, 0.8, 0.6, 0.95,
      1.2, 0.7, 1.2, 1.1, 0.64, 1.2, 1.2, 1.3, 1, 0.9, 1, 0.5, 0.9, 0.6, 1.4, 2.5,
      1.4, 1.3, 1.1, 1.5, 0.8, 1.1, 1.3, 0.85, 1.5, 1.2, 1.15, 1.5, 0.7, 1.9, 1.5,
      1.5, 1.5, 0.62),
  total_height = 
    c(1.3, 1.35, 2.16, 1.8, 1.55, 1.2, 1, 1.7, 0.8, 1.2, 0.9, 1.35,
      1.4, 1, 1.7, 1.5, 0.65, 1.5, 1.7, 1.7, 1.5, 1.5, 1.2, 0.7, 1.2, 0.8, 1.7, 3,
      1.7, 1.6, 1.5, 1.9, 1.1, 1.6, 2, 1.25, 2, 1.3, 1.45, 2.2, 0.7, 1.9, 1.8, 2, 2.2,
      0.92),
  density = 
    c(1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2,
      1, 1, 1, 1, 5, 9, 1, 1, 1, 3, 1, 3, 7, 1, 2, 2, 2, 3, 1, 1, 2, 2, 1, 1),
  group = 
    c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1)
)

model_fit = stan(
  model_code = model_code,
  data = model_data,
  chains = 1,
  verbose = TRUE,
  model_name = "tester"
)

Side Effects:

Hopefully none

Documentation:

No new docs

Reviewer Suggestions:

Can you run this with your epi package? That should be a good check that this works

Copyright and Licensing

Please list the copyright holder for the work you are submitting (this will be you or your assignee, such as a university or company): Steve Bronder

By submitting this pull request, the copyright holder is agreeing to license the submitted work under the following licenses:

• Code: GPLv3 (https://opensource.org/licenses/GPL-3.0)
• Documentation: CC-BY 4.0 (https://creativecommons.org/licenses/by/4.0/)

[SteveBronder]

@hsbadr feel free to use this patch or yours with the $, I think the regex is right either way

[hsbadr]

Thanks @SteveBronder! I've suggested a few edits and will merge after testing.

[SteveBronder]

Sounds good!

[hsbadr]

@hsbadr feel free to use this patch or yours with the $, I think the regex is right either way

FYI. I've tested this and it fixes the issue, but cc02cf4 is required to avoid removing double underscores within a variable name (e.g., x__1 should not be parsed as x1).

[SteveBronder]

Gotcha makes sense!