extract_gor_from_step.Rmd

---
title: "GOR extraction from step block"
draft: true
output:
  html_document:
    df_print: paged
---


```{r setup, include = FALSE, error=TRUE, message=FALSE, warning=FALSE}
knitr::opts_chunk$set(echo = TRUE, 
                      comment = "#>",
                      collapse = TRUE,
                      error = TRUE,
                      warning = FALSE,
                      message = FALSE,
                      fig.align = 'center',
                      autodep=FALSE
                      )
```

```{r read-prt-file}
library(dplyr)
library(ggplot2)

# read the Eclipse PRT output report
proj_root <- rprojroot::find_rstudio_root_file()
# had to zip the PRT file because it's 225 MB and too big for Github
volve_2016_zip <- file.path(proj_root, "inst/rawdata", "VOLVE_2016.zip")
temp <- tempdir()

volve_2016_txt <- readLines(unzip(volve_2016_zip, exdir = temp))
```

Once with the contents of `VOLVE_2016.PRT` loaded in the object __volve_2016_txt__, we proceed to perform the extraction.

We start by extracting few rows after the __STEP__ keyword.

```{r}
# get a list of rows from " STEP" 

# find the rows where we find the word " STEP"
step_rows <- grep("^ STEP", volve_2016_txt)

# add rows ahead to where the keyword was found
step_info_range <- lapply(seq_along(step_rows), function(x) 
    c(step_rows[x], step_rows[x]+1:2))               # add two extra row indices

step_info_range[[1]]   # sample for report page 1 only
```
```{r}
# get the text from all pages and put them in a list
steps_info_txt_pages <- lapply(seq_along(step_info_range), function(x) 
    volve_2016_txt[step_info_range[[x]]])
```

For example, this is an example of the first page for step #1.
```{r}
steps_info_txt_pages[1]
```

## Extract the GOR value
### Explanation

* `.*GOR=.*?`: any characters up to "GOR="
* `(\\d+.\\d+)`: capture group of digits, dot. and digits
* `.*`: any characters to the end

```{r}
# iterate through the list of pages: dates
gor_dfs <- lapply(seq_along(steps_info_txt_pages), function(x) {
    page <- steps_info_txt_pages[[x]]  # put all pages text in a list
    gor_row_txt <- grep(" GOR", page)
    gor_value <- sub(".*GOR=.*?(\\d+.\\d+).*", "\\1", page[gor_row_txt])
    
    # dataframe
    data.frame(gor = gor_value, stringsAsFactors = FALSE) 
})

gor_df <- do.call("rbind", gor_dfs)
gor_df
```  

## Water Gas Ratio (WGR)
###

```{r}
# iterate through the list of pages: WATERCUT
wgr_dfs <- lapply(seq_along(steps_info_txt_pages), function(x) {
    page <- steps_info_txt_pages[[x]]  # put all pages text in a list
    row_txt <- grep(" WGR=", page)
    wgr_value <- sub(".*WGR=.*?(\\d+.\\d+).*", "\\1", page[row_txt])
    
    # dataframe
    data.frame(wgr = wgr_value, stringsAsFactors = FALSE) 
})

wgr_df <- do.call("rbind", wgr_dfs)
wgr_df
```  


```{r rows.print=50}
# iterate through the list of pages
step_info_dfs <- lapply(seq_along(steps_info_txt_pages), function(x) {
    page <- steps_info_txt_pages[[x]]             # put all pages text in a list
    row_txt <- grep(" STEP", page)
    date_value <- sub(".*?(\\d{1,2}-[A-Z]{3}-\\d{4}).", "\\1", page[row_txt])
    # step number
    step_value <- sub(".*STEP.*?(\\d+)+.*", "\\1", page[row_txt]) # extract the days
    # get step in days
    row_txt <- grep(" TIME", page)
    days_value <- sub(".*TIME=.*?(\\d+.\\d.)+.*", "\\1", page[row_txt]) # extract the days
    # Get the average pressure PAV
    row_txt <- grep(" PAV", page)
    pav_value <- sub(".*PAV=.*?(\\d+.\\d+).*", "\\1", page[row_txt])
    # get the GOR
    row_txt <- grep(" GOR", page)
    gor_value <- sub(".*GOR=.*?(\\d+.\\d+).*", "\\1", page[row_txt])
    # get the WGR
    row_txt <- grep(" WGR=", page)
    wgr_value <- sub(".*WGR=.*?(\\d+.\\d+).*", "\\1", page[row_txt])

    
    # dataframe
    data.frame(step=step_value, date = date_value, days = days_value, 
               pav_bar = pav_value, gor_m3m3 = gor_value, wgr_m3m3 = wgr_value,
               stringsAsFactors = FALSE) 
})

step_info <- do.call("rbind", step_info_dfs)
step_info
```