FM Data Analytics: a data visualization tool based on R & Shiny - V3 RELEASED!

Jay86 · January 16, 2021

Hi, V1 of this worked for me but since deleting and installing V2 the FMapp won't launch. Followed all the steps from tutorial but can't seem to get it working, any idea where i'm going wrong???? Thanks!

gam945 · January 16, 2021

22 minutes ago, Jay86 said:

Hi, V1 of this worked for me but since deleting and installing V2 the FMapp won't launch. Followed all the steps from tutorial but can't seem to get it working, any idea where i'm going wrong???? Thanks!

PM me with the content of FMapp, and a printscreen of Rscript.exe's location

gam945 · January 16, 2021

@Jay86's issue resolved. FMapp.bat was updated on Github.

For the more curious basically the path for his Rscript.exe file had a space in it. Using the following syntax

"variable=value"

to set the variable to a value doesn't allow spaces in the value (path).

Instead, the correct syntax to allow spaces in the path is

variable="value"

. Which is now used in the FMapp.bat

Also, as this is a very recurrent error from the user's side: when exporting from the FM shortlist, be sure to scroll from top to bottom of the shortlist. Otherwise your data will be incomplete and this will cause errors

Edited January 16, 2021 by gam945

Mikal · January 16, 2021

10 hours ago, gam945 said:
@Jay86's issue resolved. FMapp.bat was updated on Github.

For the more curious basically the path for his Rscript.exe file had a space in it. Using the following syntax
"variable=value"
to set the variable to a value doesn't allow spaces in the value (path).

Instead, the correct syntax to allow spaces in the path is
variable="value"
. Which is now used in the FMapp.bat

Also, as this is a very recurrent error from the user's side: when exporting from the FM shortlist, be sure to scroll from top to bottom of the shortlist. Otherwise your data will be incomplete and this will cause errors

No luck from me with the new update. I was having the same issue as above, but since downloading the new update the .bat loads, gives the below message and then just closes.

jtaylor1409 · January 16, 2021

Same issue as @Mikal this end.

Edited January 16, 2021 by jtaylor1409

gam945 · January 16, 2021

@Mikal @jtaylor1409 add a "PAUSE" at the bottom of your FMapp.bat like this and save the file.

<my code blablabla...>
PAUSE

As you can see the batch file launches R, launches also the app, but you're stuck at the package install step. PM me with a print screen of the command prompt (black window) when you re-run FMapp.bat

r70lla · January 16, 2021

Try running the FMapp.bat as administrator

Mikal · January 16, 2021

1 hour ago, gam945 said:
@Mikal @jtaylor1409 add a "PAUSE" at the bottom of your FMapp.bat like this and save the file.
<my code blablabla...>
PAUSE
As you can see the batch file launches R, launches also the app, but you're stuck at the package install step. PM me with a print screen of the command prompt (black window) when you re-run FMapp.bat

Thanks. Got a bit further but get following error now.

gam945 · January 16, 2021

3 minutes ago, Mikal said:

Thanks. Got a bit further but get following error now.

Well that is exactly what I suspected

Refer to my first post:

So either run as an admin or uninstall R and install it in a folder you'll create in C:\

Do not forget to adjust again the path in FMapp.bat (no need to reinstall my app)

Mikal · January 16, 2021

Just now, gam945 said:

Well that is exactly what I suspected

Refer to my first post:

So either run as an admin or uninstall R and install it in a folder you'll create in C:\

Do not forget to adjust again the path in FMapp.bat (no need to reinstall my app)

It is on the C drive though? Do you mean don't install it under program files? I didn't have any issues with the first version.

gam945 · January 16, 2021

5 minutes ago, Mikal said:

It is on the C drive though? Do you mean don't install it under program files? I didn't have any issues with the first version.

Yes I mean do not install R under program files. This version is different as the package installation process occurs direct from the main R script and not from a batch file.

Create a folder (for example MyPrograms in C:\) and re-install R in it.

EDIT: added a more explicit direction to follow in my first post regarding R's installation

Edited January 16, 2021 by gam945

Mikal · January 16, 2021

Uninstalled R, re-installed under C:/MyPrograms, updated the fmapp.bat with the correct link (I kept the "PAUSE" at the bottom), ran the batch and it worked and installed the packages. Thanks

Please don't laugh at my poor strike force

Edited January 16, 2021 by Mikal

gam945 · January 16, 2021

Thanks @Mikal

For all users - please follow this direction from my first post to avoid errors:

** It is preferable to install R in a folder you have created in the C:\ directory (instead of installing it in "C:\Program Files") to avoid errors when running the batch file that launches the app

For example:
(1) Create the MyPrograms folder in C:\
(2) Install R in the following directory: C:\MyPrograms

Hovis Dexter · January 16, 2021

I installed both R and FM Data in their own folders in C:\ and following the issue of V2 everything seems to working correctly. Many thanks to @gam945 for a really useful utility.

gam945 · January 16, 2021

43 minutes ago, Hovis Dexter said:

I installed both R and FM Data in their own folders in C:\ and following the issue of V2 everything seems to working correctly. Many thanks to @gam945 for a really useful utility.

Great! Feel free to share your feedback and/or experience with the app!

Gdawg69 · January 18, 2021

@gam945

You sir, are a breath of fresh air and an absolute genius.

Really appreciate your work mate. Well done, genuinely. Top top work.

gam945 · January 18, 2021

37 minutes ago, Gdawg69 said:

@gam945

You sir, are a breath of fresh air and an absolute genius.

Really appreciate your work mate. Well done, genuinely. Top top work.

Appreciate this mate

gam945 · January 28, 2021

I am planning to release a final update to include the performance tracking side in the app. It will be the final update done by me and that adds features, but there may be some other updates on the user interface side (in which I'm a lot less proficient).

My idea is the following:

Nothing changes for the shortlist/scouting analysis. However, for the squad analysis, instead of naming your FM html output "squad.html", you'd need to name it "squad_yyyymmdd.html" where yyyymmdd is the in-game date. In the squad analysis, the code would take the most recent squad data, so no change on that. However, I will add a feature when you can compare your squad's attributes and/or DNA rating and statistics between different in-game dates. The user would see this information on tables (and plots maybe?) basically. And there would be a sub-section where you could decide which in-game dates you want to compare. Of course if you have the following files in your data folder:
"squad_20210101.html", "squad_20210201.html", "squad_202010301.html" well you could only compare in-games dates Jan 1st, Feb 1st and March 1st....you get my point.

It seems basic, but I'm seeding this idea here so you -the users of my app- can maybe make it a better idea.

Hovis Dexter · January 28, 2021

That would be a brilliant addition

Mcfc1894 · January 29, 2021

Have I done something wrong here it's seems to be taking a lot longer than it did on your guide video

Mcfc1894 · January 29, 2021

17 minutes ago, Mcfc1894 said:

Have I done something wrong here it's seems to be taking a lot longer than it did on your guide video

Ignore this I'm on now just needed to be patience thanks it's a great app would help me a lot in game

gam945 · January 29, 2021

1 hour ago, Mcfc1894 said:

Ignore this I'm on now just needed to be patience thanks it's a great app would help me a lot in game

Yes it can take some time to install all the R packages if it's your first time ever. I use R a lot so I had the packages installed already in my video. When the packages are installed they are just loaded after, so it will take a lot less time Usually in 10sec the app is launched.

Edited January 29, 2021 by gam945

Oufti215 · January 31, 2021

Hello,

thanks for this programm ! I tested with success the version 1.

Unfortunately impossible to run version 2.

I followed the recommendations on installing R in a different folder than the original. And change url in FMapp.

Here is the screen I get during the installation (I'm French).

After the window disappears. When I restart, it's marked "the loading required the ggplot2 package" and stop (screen)...

Edited January 31, 2021 by Oufti215

gam945 · February 1, 2021

17 hours ago, Oufti215 said:

Hello,

thanks for this programm ! I tested with success the version 1.

Unfortunately impossible to run version 2.

I followed the recommendations on installing R in a different folder than the original. And change url in FMapp.

Here is the screen I get during the installation (I'm French).

After the window disappears. When I restart, it's marked "the loading required the ggplot2 package" and stop (screen)...

providing the solution to this issue just in case: clean reinstall of R solved the problem

Oufti215 · February 1, 2021

Just be carreful if you had already R before using this programm.

As you know you need to install R in specific folder. So if you have already R in your PC under program files, you need uninstall R.

When you have uninstall R, unfortunatly it remains a library folder : you need to uninstall this folder manually too.

gam945 · February 2, 2021

On 28/01/2021 at 05:10, Hovis Dexter said:

That would be a brilliant addition

So far I tried to keep it as realistic as possible and hide attributes as much as possible. The only part of the app that uses attributes is the DNA Rating, and it could be interpreted as a rating your coaches give to each players according to a specified (via attributes) playing style. So you can't really see the attributes.

For the "performance tracking", I will for sure add a table to monitor each per 90min stats, but I'm not sure if and how I can add attributes monitoring. I know that in the game, it's pretty difficult to see the progress in the attributes (loan a player and you'll loose this info, other multiple bugs) but still I want to keep my app realistic.

How do you think I should build that? Enable to monitor the attributes by category (mental/technical/physical/gk, attacking/defending/movement/... or another categorization)? Just show the attributes as in the game? No attributes at all?

I'm open to suggestions for this part

pow4h · February 7, 2021

How to translate this so I can use it with other game language ?

gam945 · February 7, 2021

4 hours ago, pow4h said:

How to translate this so I can use it with other game language ?

You'd need to modify every occurrences of the data point/attributes/statistics in the code. There's is a data dictionary in the help folder, where the long name, short name and categories of each data point is described. You'd basically need to use the fm views I have, export the view and using the data dictionary, match each data point's short name in British English with the new language. Example below, every data point between `` should be translated to the correct short name in the new language:

And if you want to translate the UI also, you'd need to translate it in the code also. Example below:

It is something I don't personally need and don't have the time to do so if anyone wants to do it, be my guest.

Edited February 7, 2021 by gam945

Hovis Dexter · February 7, 2021

On 02/02/2021 at 22:05, gam945 said:

So far I tried to keep it as realistic as possible and hide attributes as much as possible. The only part of the app that uses attributes is the DNA Rating, and it could be interpreted as a rating your coaches give to each players according to a specified (via attributes) playing style. So you can't really see the attributes.

For the "performance tracking", I will for sure add a table to monitor each per 90min stats, but I'm not sure if and how I can add attributes monitoring. I know that in the game, it's pretty difficult to see the progress in the attributes (loan a player and you'll loose this info, other multiple bugs) but still I want to keep my app realistic.

How do you think I should build that? Enable to monitor the attributes by category (mental/technical/physical/gk, attacking/defending/movement/... or another categorization)? Just show the attributes as in the game? No attributes at all?

I'm open to suggestions for this part

I was just looking forward to the highlighted additions that you mentioned here:

On 28/01/2021 at 00:24, gam945 said:

I am planning to release a final update to include the performance tracking side in the app. It will be the final update done by me and that adds features, but there may be some other updates on the user interface side (in which I'm a lot less proficient).

My idea is the following:

Nothing changes for the shortlist/scouting analysis. However, for the squad analysis, instead of naming your FM html output "squad.html", you'd need to name it "squad_yyyymmdd.html" where yyyymmdd is the in-game date. In the squad analysis, the code would take the most recent squad data, so no change on that. However, I will add a feature when you can compare your squad's attributes and/or DNA rating and statistics between different in-game dates. The user would see this information on tables (and plots maybe?) basically. And there would be a sub-section where you could decide which in-game dates you want to compare.H Of course if you have the following files in your data folder:
"squad_20210101.html", "squad_20210201.html", "squad_202010301.html" well you could only compare in-games dates Jan 1st, Feb 1st and March 1st....you get my point.

It seems basic, but I'm seeding this idea here so you -the users of my app- can maybe make it a better idea.

pow4h · February 8, 2021

Hi I have problem that most stats in app are multiplied by 100.

Like value 14 000 000 instead od 1 400 000,

307 key passes per 90 minutes instead of 3.07 etc

Do you know what cause a problem?

EDIT:

Ok I solved it.

Problem was that I had Decimal Symbol as comma not Dot

EDIT2:

Its fixed but there is problem in other way with other stats:

With wages - where wage is above 1000 it shows 0 when beyond 1000 it shows correct value.

Games played: for many players it shows 0 games played while they are mine key players. For less importans players that played less games it shows correct value.

Crossing is still wrong - its multiplied by 1000 like 7050 crosses per 90 minutes.

EDIT3:

I fixed this by setting formats like that:

Quote

Preferences - Advanced - Formats

Positive: €1,00 (The Symbol of yout currency has to be before the value)
Decimal Symbol: Dot
Digit Grouping Symbol: Comma

Edited February 8, 2021 by pow4h

gam945 · February 9, 2021

On 08/02/2021 at 10:12, pow4h said:

Hi I have problem that most stats in app are multiplied by 100.

Like value 14 000 000 instead od 1 400 000,

307 key passes per 90 minutes instead of 3.07 etc

Do you know what cause a problem?

EDIT:

Ok I solved it.

Problem was that I had Decimal Symbol as comma not Dot

EDIT2:

Its fixed but there is problem in other way with other stats:

With wages - where wage is above 1000 it shows 0 when beyond 1000 it shows correct value.

Games played: for many players it shows 0 games played while they are mine key players. For less importans players that played less games it shows correct value.

Crossing is still wrong - its multiplied by 1000 like 7050 crosses per 90 minutes.

EDIT3:

I fixed this by setting formats like that:

Is it all resolved now?

gam945 · February 9, 2021

Preview of additional visualizations: new Home page

So basically a summary of goal contributions and playing time for the squad... is there another visualization you'd want to see in this section?

Edited February 9, 2021 by gam945

pow4h · February 10, 2021

21 hours ago, gam945 said:

Is it all resolved now?

Yes.

I think it will be good if you post correct settings (format) in first post.

QP · February 15, 2021

On 09/02/2021 at 22:05, gam945 said:

Preview of additional visualizations: new Home page

So basically a summary of goal contributions and playing time for the squad... is there another visualization you'd want to see in this section?

I don't know if it is possible but I prefer to have more stats and compare many of them within each other If it's possible that would be amazing. And dark mode would be amazing.

For passes;

Key passes per 90 / chances created

For tackles;

interceptions per 90 / Tackles per 90

Tackles per 90 / % of tackles

Key Tackles per 90 / Tackles attempted

For header and aerial challenges;

Key headers per 90 / % of headers

And If there is a tutorial to expand the comparison graphs with more stats that would be really helpful because we can also create our own comparison graphics too without too much hassle.

Edited February 15, 2021 by QP

DavyDepuydt1 · February 15, 2021

On 16/01/2021 at 03:03, gam945 said:
@Jay86's issue resolved. FMapp.bat was updated on Github.

For the more curious basically the path for his Rscript.exe file had a space in it. Using the following syntax
"variable=value"
to set the variable to a value doesn't allow spaces in the value (path).

Instead, the correct syntax to allow spaces in the path is
variable="value"
. Which is now used in the FMapp.bat

Also, as this is a very recurrent error from the user's side: when exporting from the FM shortlist, be sure to scroll from top to bottom of the shortlist. Otherwise your data will be incomplete and this will cause errors

I don't know if this has been mentioned, but selecting a random row to focus the grid folllowed by ctrl+a (select all) and export, also does the trick to have the entire grid, is a bit easier than scrolling large grids... (The 'select all' command can take a second of 3 in large grids before selection is complete)

Edited February 16, 2021 by DavyDepuydt1

gam945 · February 21, 2021

UPDATE LOG: V3 (2021/02/21)

* "Home" menu added:

Info boxes: number of players at the club & players in the shortlist
Summary charts: bar chart of the goal contribution distribution, boxplots of the playing time and wages distribution

* "Progress" submenu added in the "Squad Analysis" menu: dates selection, performance and attributes progress between selected dates

* "Shortlist" submenu in the "Scouting" menu updated: now shows min and max asking prices and wage demands, presence of release clause and conditional release clauses instead of replicate table of the "Filters" submenus

* DT tables now used instead of Reactables: styling and formatting improved

* To use the V3 version you need to name your squad output files "squad_yyyymmdd", where "yyyymmdd" is the in-game date. This allows to compute the variations (progress) between given in-game dates. I updated the data folder in my Github to give you an example.

* It is possible thanks to the above feature to store multiple squad data files. Keep in mind that the more there is squad files in your data folder, the longer the app will take to load. You can find out how much time the different tasks (load data & format data for example) take time by paying attention to the command prompt that is launched when you double-click on FMapp.bat.

Following is a video that shows the new features:

https://streamable.com/0spqup

To use the V3 version of the app, you only need the new "app.R" from my Github. Download the entire folder, but only replace the old "app.R" by the new one. You do not need to install R again or set up the batch file again!

For those who installed manually the required packages (Windows 7 users), you will need to install the "DT" package:

install.packages("DT")

https://github.com/ybenadjal/FM-Data-Analytics

Edited February 21, 2021 by gam945

Kung · February 27, 2021

Its so ridicilous, i can´t get it to work on my win 10 laptop, i followed several of times all the instructions, but it does not launch chrome

# Setup -------------------------------------------------------------------

# status message
cat("\nCheck if packages need to be installed and install if needed...")
Sys.sleep(2)

# required packages
requiredpack <- c(
  "tidyverse",
  "shiny",
  "shinydashboard",
  "shinyWidgets",
  "plotly",
  "XML",
  "tictoc",
  "DT"
)
# already installed packages
installedpack <- as.data.frame(installed.packages())$Package
# extract missing packages i.e. required packages that are not installed
missingpack <- requiredpack[!is.element(requiredpack, installedpack)]
# install missing packages if there is more than 0 missing packages
if (length(missingpack) > 0) install.packages(missingpack, repos = "https://cran.rstudio.com/")

# status message
cat("\n\nLoad required packages...\n")
Sys.sleep(2)

# load required packages
library(shiny) # shiny package
library(shinydashboard) # shinydashboard package
library(shinyWidgets) # shinyWidgets package
library(tictoc)
library(magrittr) # tidyverse package
library(plotly) # plotly package
library(DT) # DT package
library(XML) # XML package
library(tidyverse) # tidyverse package

# working directory
# function to get current file location
# from 'https://stackoverflow.com/questions/47044068/get-the-path-of-current-script'
# see Juan Bernabe's answer
getCurrentFileLocation <-  function()
{
  this_file <- commandArgs() %>% 
    tibble::enframe(name = NULL) %>%
    tidyr::separate(col=value, into=c("key", "value"), sep="=", fill='right') %>%
    dplyr::filter(key == "--file") %>%
    dplyr::pull(value)
  if (length(this_file)==0)
  {
    this_file <- rstudioapi::getSourceEditorContext()$path
  }
  return(dirname(this_file))
}
# get this script's directory
sdir <- getCurrentFileLocation()
# set working directory as script's directory
setwd(sdir)


# App inputs 1 ------------------------------------------------------------
# independant of the data

# unique FM positions
positions <- c("GK", # goalkeepers
               "D(L)", "D(C)", "D(R)", # defenders
               "WB(L)", "WB(R)", # wingbacks
               "DM", # defensive midfielders
               "M(L)", "M(C)", "M(R)", # midfielders
               "AM(L)", "AM(C)", "AM(R)", # attacking midfielders
               "ST(C)" # strikers
)

# FM data points: attributes
fm_attributes <- c("Aer","Cmd","Com","Ecc","Han","Kic","1v1","Pun","Ref","TRO",
                "Thr","Agg","Ant","Bra","Cmp","Cnt","Dec","Det","Fla","Ldr",
                "OtB","Pos","Tea","Vis","Wor","Acc","Agi","Bal","Jum","Nat",
                "Pac","Sta","Str","Cor","Cro","Dri","Fin","Fir","Fre","Hea",
                "Lon","L Th","Mar","Pas","Pen","Tck","Tec")

# FM data points: statistics (chalkboard)
fm_statschalkboard <- c("Hdrs A","Aer A/90","Hdrs","Hdrs W/90","Hdr %",
                     "K Hdrs","Cr A","Cr C/A","Cr C","Gl Mst","Distance",
                     "Dist/90","Drb","DrbPG","Off","Asts/90","CCC","Ch C/90",
                     "K Pas","K Ps/90","Pas A","Ps A/90","Pas %","Ps C",
                     "Ps C/90","Svh","Svp","Svt","Shots","ShT","ShT/90",
                     "Shot %","Shot/90","Itc","Int/90","K Tck","Tck A",
                     "Tck R","Tck W","Tck1")

# FM data points: statistics (general)
fm_statsgeneral <- c("AT Apps","AT Gls","AT Lge Apps","AT Lge Gls","Apps",
                  "Ast","Mins/Gl","Av Rat","Clean sheets","Con/90","FA",
                  "Fls","Gwin","D","Lost","G. Mis","Won","Gls","Conc",
                  "Gls/90","Last 5 FT Games","Last 5 Games","Mins",
                  "Mins/Gm","Last C","Last Gl","Pens","Pens Faced",
                  "Pens Saved","Pens Saved Ratio","Pens S","Pen/R","PoM",
                  "Pts/Gm","Red","Starts","Tcon/90","Tgls","Tcon","Tgls/90",
                  "xG", "Yel","Int Apps","Int Ast","Int Av Rat","Int Conc")

# FM data points: custom category - financials
fm_financials <- c("Appearance Fee","Assist Bonus","Cln Sheet Bonus","Goal Bonus",
                "Int Cap Bonus","SLAB","SLGAB","SLGB","Team Year Bonus",
                "Top Score Bonus","Unused Sub Fee","WaCLG","Injury Rls",
                "Min Fee Rls","Min Fee Rls Clubs In Cont Comp",
                "Min Fee Rls Clubs Mjr Cont Comp","Min Fee Rls to Higher Div",
                "Min Fee Rls to Domestic Clubs","Min Fee Rls to Foreign Clubs",
                "Non Prom Rls Cls","Non-Playing Rel","Relegation Release",
                "Wage","Wage After Tax","Wage Contrib.","New Wage","Max AP",
                "Max WD","Min AP","Min WD","Asking Price","Fee",
                "Last Trans. Fee","Ovr","Transfer Fees Received","Value")

# FM data points: custom category - stats for progress tables
fm_stats <- c("Aer A/90", "Hdrs W/90", "Hdr %", "K Hdrs", "Gl Mst", "Dist/90",
            "DrbPG", "Off", "Cr A", "Cr C", "Cr C/A", "Ps A/90", "Ps C/90",
            "Pas %", "K Ps/90", "Ch C/90", "Asts/90", "Shot/90", "ShT/90",
            "Shot %", "Gls/90", "Int/90", "K Tck", "Tck A", "Tck W", "Tck R",
            "Svh", "Svp", "Svt", "Con/90", "Apps", "Starts", "Mins", "Av Rat",
            "Clean sheets", "Gwin", "PoM", "Pts/Gm", "Tcon/90", "Tgls/90",
            "Pens Faced", "Pens Saved", "Pens Saved Ratio", "Pens", "Pens S",
            "Pen/R", "FA", "Fls", "Yel", "Red")

# User defined functions --------------------------------------------------

# reads and parse HTML outputs from FM
# import all data as characters
# uses UTF-8 encoding to allow for special characters such as currency symbols
parse_FMoutput <- function(htmlpath) {
  # read HTML table using UTF-8 encoding
  data <- readHTMLTable(htmlpath, header = TRUE, encoding = "UTF-8") %>% .[[1]]
  # make columns unique in case of duplicate names
  colnames(data) %<>% make.unique(sep = "")
  # return data
  data
}

# position parsing
# example1: "D/M(C)" to "D(C), M(C)"
# example2: "D/WB(L), M/AM(L), ST(C)" to "D(L), WB(L), M(L), AM(L), ST(C)"
pos_parse <- function(pos) {
  # remove all spaces, split using commas as separators
  # transform the list output into a vector
  # split using opening parenthesis as separator
  # remove all closing parenthesis
  pos %<>% str_remove_all(" ") %>% str_split(",") %>% unlist() %>% #trimws() %>%
    str_split("\\(") %>% sapply(function(x) str_remove_all(x, "\\)"))
  # transform into list if necessary
  # special case if position is one string ("GK" or "DM") do nothing
  if((typeof(pos) != "list") & (length(pos) != 1)) pos %<>% split(col(.))
  # transform list into dataframe
  # first column is the vertical position on the pitch (GK/D/WB/DM/M/AM/ST)
  # second column is the horizontal position on the pitch (L/C/R)
  n <- length(pos)
  pos_df <- data.frame(vpos = rep(NA,n), hpos = rep(NA,n))
  for (i in 1:n) {
    pos_df[i, 1] <- pos[[i]][1]
    pos_df[i, 2] <- pos[[i]][2]
  }
  # replace NAs by ""
  pos_df %<>% replace(is.na(.), "")
  # break the vertical position by using "/" as separators
  # remove original row and add a row for each vertical position
  n <- dim(pos_df)[1]
  temp_pos_df <- data.frame()
  cols_to_remove <- vector()
  for (i in 1:n) {
    k <- str_count(pos_df[i, 1], "/") + 1
    if (k-1 > 0) {
      vpos <- rep(pos_df[i, 1], k)
      hpos <- pos_df[i, 2]
      cols_to_remove <- c(cols_to_remove,i)
      for (j in 1:k) vpos[j] %<>% str_split("/") %>% sapply("[", j)
      df <- data.frame(vpos = vpos, hpos = hpos)
      temp_pos_df %<>% rbind(df)
    }
  }
  if (length(cols_to_remove) > 0) pos_df %<>% slice(-cols_to_remove)
  pos_df %<>% rbind(temp_pos_df)
  # break the horizontal position by using "" as separators
  # remove original row and add a row for each horizontal position
  n <- dim(pos_df)[1]
  temp_pos_df <- data.frame()
  cols_to_remove <- vector()
  for (i in 1:n) {
    k <- nchar(pos_df[i, 2])
    if (k > 1) {
      vpos <- pos_df[i, 1]
      hpos <- rep(pos_df[i, 2], k)
      cols_to_remove <- c(cols_to_remove,i)
      for (j in 1:k) hpos[j] %<>% str_split("") %>% sapply("[", j)
      df <- data.frame(vpos = vpos, hpos = hpos)
      temp_pos_df %<>% rbind(df)
    }
  }
  if (length(cols_to_remove) > 0) pos_df %<>% slice(-cols_to_remove)
  pos_df %<>% rbind(temp_pos_df)
  # custom order for vertical and horizontal positions
  vpos_order <- c("GK", "D", "WB", "DM", "M", "AM", "ST")
  hpos_order <- c("L", "C", "R", "")
  # order the dataframe using the custom order
  # add a column for the position = vertical position(horizontal position)
  pos_df %<>%
    mutate(vpos = factor(vpos, levels = vpos_order)) %>%
    mutate(hpos = factor(hpos, levels = hpos_order)) %>%
    arrange(vpos, hpos) %>%
    mutate(pos = paste0(vpos, "(", hpos, ")"))
  # concatenate all positions into a single strings
  # positions are separated by commas
  # replace special cases DM() & GK() by the FM format DM & GK
  pos_df$pos %>%
    paste0(collapse = ", ") %>%
    str_replace_all("DM\\(\\)", "DM") %>%
    str_replace_all("GK\\(\\)", "GK")
}

# converts to percent format; if NA shows NA
formatPct <- function(x) {
  if_else(
    condition = is.na(x),
    true = NA_character_,
    false = paste0(round(x*100, digits = 1), " %")
  )
}

# cleans the raw data extracted from FM
cleandata <- function(data) {
  
  format_money <- function(x) {
    x %>%
      # remove currency symbols and separators
      str_remove_all("\u20AC|\u00A3|\u0024") %>% str_remove_all(",") %>%
      # remove wage frequency symbols
      str_remove_all("p/w") %>% str_remove_all("p/a") %>% str_remove_all("p/m") %>%
      # replace decimal prefix units by their values
      str_replace_all(" ", " 1") %>% str_replace_all("K", " 1000") %>%
      str_replace_all("M", " 1000000") %>%
      # split string and multiply
      str_split(" ") %>% lapply(as.numeric) %>% lapply(prod) %>% unlist()
  }
  
  data %>%
    # Replace "not available" data by actual NAs
    replace(. == "-", NA) %>%
    replace(. == "Scouting Required", NA) %>%
    replace(. == "N/A", NA) %>%
    replace(. == "Unknown", NA) %>%
    replace(. == "-", NA) %>%
    # Name: Removes the nationality in 'shortlist', doesn't affect 'squad'
    mutate(`Name` = `Name`  %>% str_split("-") %>% sapply("[[", 1) %>% trimws()) %>%
    # Rec; scouting recommandation
    mutate_at(vars(one_of("Rec")), function(x) {
      str_remove_all(x, "-") %>% trimws() %>% as.numeric() %>% replace_na(-1)}) %>%
    # position: add the parsed position
    mutate(
      `PositionParsed` = sapply(`Position`, pos_parse)%>% unname(),
      .after = `Position`
    ) %>%
    # Attributes: mean of upper and lower bound if no exact values known
    mutate_at(fm_attributes, function(x) {
      str_split(x, "-") %>% lapply(as.numeric) %>% lapply(mean) %>% unlist()}) %>%
    # Remove the league from the based nation; league is available in separate datapoint
    mutate(`Based` = str_split(`Based`, "\\(") %>% sapply("[[", 1) %>% trimws()) %>%
    # Stats: remove units and divide ratio type values by 100
    mutate(`Hdr %` = as.numeric(str_remove_all(`Hdr %`, "%"))/100) %>%
    mutate(`Cr C/A` = as.numeric(str_remove_all(`Cr C/A`, "%"))/100) %>%
    mutate(`Distance` = str_remove_all(`Distance`, "km")) %>%
    mutate(`Dist/90` = str_remove_all(`Dist/90`, "km")) %>%
    mutate(`Pas %` = as.numeric(str_remove_all(`Pas %`, "%"))/100) %>%
    mutate(`Shot %` = as.numeric(str_remove_all(`Shot %`, "%"))/100) %>%
    mutate(`Tck R` = as.numeric(str_remove_all(`Tck R`, "%"))/100) %>%
    mutate(`Apps` = str_remove_all(`Apps`,"[//(//)]") %>% str_split(" ") %>%
             lapply(as.numeric) %>% lapply(sum) %>% unlist()) %>% # total appearances
    mutate(`Gwin` = as.numeric(str_remove_all(`Gwin`, "%"))/100) %>%
    mutate(`Pens Saved Ratio` = as.numeric(str_remove_all(`Pens Saved Ratio`, "%"))/100) %>%
    mutate(`Pen/R` = as.numeric(str_remove_all(`Pen/R`, "%"))/100) %>%
    mutate(`Int Starts` = str_remove_all(`Int Apps`,"[//(//)]") %>% str_split(" ") %>%
             lapply(as.numeric) %>% sapply("[[",1), .after = `Int Apps`) %>% # create international starts
    mutate(`Int Apps` = str_remove_all(`Int Apps`,"[//(//)]") %>% str_split(" ") %>%
             lapply(as.numeric) %>% lapply(sum) %>% unlist()) %>% # total int appearances
    # Stats as numeric values
    mutate_at(c(fm_statschalkboard,fm_statsgeneral),
              function(x) {x %>% str_remove_all(",") %>% as.numeric}) %>%
    # Format financial values as numeric
    mutate_at(fm_financials, format_money) %>%
    # Value and Wage: 0 if NA
    mutate(`Value` = replace_na(`Value`, 0)) %>%
    mutate(`Wage` = replace_na(`Wage`, 0)) %>%
    # Club: free player/amateur if NA
    mutate(`Club` = replace_na(`Club`, "Free player")) %>%
    # Division: free player/unknown if NA
    mutate(`Division` = replace_na(`Division`, "Free player/unknown"))
}

# assigns a level to the scout recommendation
# used to determine the marker color in the visualizations
# and the less saturated color in the tables, depending on the variable 'plot'
colorcatRec <- function(rec, plot = TRUE) {
  case_when(
    rec == -1 ~ if_else(plot == TRUE, "-1", "#bf3eff"), # darkorchid1
    between(rec, 0, 49) ~ if_else(plot == TRUE, "1", "#ff0000"), # red
    between(rec, 50, 59) ~ if_else(plot == TRUE, "2", "#ffa500"), # orange
    between(rec, 60, 69) ~ if_else(plot == TRUE, "3", "#ffff00"), # yellow
    between(rec, 70, 84) ~ if_else(plot == TRUE, "4", "#00ee00"), # green2
    between(rec, 85, 100) ~ if_else(plot == TRUE, "5", "#008b00") # green4
  )
}

# assigns a numeric value to Minutes/90
# used to determine the marker size in the visualizations
sizemapMinsp90 <- function(minsp90) {
  case_when(
    minsp90 <= 5 ~ 1.5,
    minsp90 > 5 & minsp90 <= 10 ~ 2,
    minsp90 > 10 & minsp90 <= 20 ~ 2.5,
    minsp90 > 20 & minsp90 <= 30 ~ 3,
    minsp90 > 30 ~ 3.5
  )
}

# generates a plotly scatter plot
plot_it <- function(data, x, y, xlabel, ylabel, targetratio) {
  plot_ly(
    # add a column of "1"s to avoid the duplicate legends for the ratio trace
    data = data %>% mutate(`ratiotracecolor` = "1"), x = ~x, y = ~y,
    type = "scatter", mode = "markers",
    # adds text when marker is hovered on
    hoverinfo = "text",
    # define text logic: name, age, position, mins, ratio and marker value
    text = ~paste0(
      "<b>Name: ", Name,
      "<br>Age: </b>", Age,
      "<br><b>Club: </b>", Club,
      "<br><b>Nation: </b>", Nat1,
      "<br><b>Position: </b>", Position,
      "<br><b>Games played: </b>", Mins,
      "<br><b>", xlabel, ": </b>", round(x, digits = 1),
      "<br><b>", ylabel, ": </b>", round(y, digits = 1),
      # add the targetratio*0 to transform the entire expression to NA is targetration = 0
      "<br><b>Ratio: </b>", formatPct(targetratio*0 + y/x)
    ),
    # size determined by game played/mins per 90
    # mapped with sizemapMinsp90()
    size = ~sizemapMinsp90(Mins), fill = "",
    # color determined by scout recommendation value
    color = ~colorcatRec(Rec),
    # color mapping
    colors = c(
      "-1" = "darkviolet",
      "1" = "red",
      "2" = "orange",
      "3" = "yellow",
      "4" = "green2",
      "5" = "green4"
    ),
    # outline color
    strokes = "black",
    # hide legend for the markers
    showlegend = FALSE
  ) %>%
    # add name on top of the marker
    add_text(
      text = ~Name, textposition = "top right", textfont = list(
        color = "black", size = 11
      )
    ) %>%
    # add a trend line. if targetratio = NA it won't show on the plot
    add_trace(
      y = ~targetratio*x,
      mode = "lines", line = list(width = 1, dash = "dash", color = "grey"),
      name = paste0(round(targetratio*100,digits = 0), " % ratio"),
      hoverinfo = "none", showlegend = TRUE, color = ~ratiotracecolor
    ) %>%
    # add axis labels, no title
    layout(
      xaxis = list(title = xlabel), yaxis = list(title = ylabel),
      showlegend = TRUE
    )
}

# checks if any substring from a vector appears in a string
grepl_op <- function(pattern_array, x) {
  check <- 0
  for (i in 1:length(pattern_array)) {
    check <- check + grepl(pattern_array[i], x, fixed = TRUE)
  }
  as.logical(check)
}

# DT tables for the progress submenus
# tabulates attributes/performance data difference
tabulate_progress <- function(dataList, prevdate, currdate, type) {
  
  # transform prevdate and currdate from the inputs format into the original format
  prevdate %<>% as.Date("%d %b %Y") %>% format("%Y%m%d")
  currdate %<>% as.Date("%d %b %Y") %>% format("%Y%m%d")
  
  # determine if working with attributes or performance stats
  if (type == "attr") {
    datapoints <- fm_attributes
  } else {
    datapoints <- fm_stats
  }
  
  # previous date data
  previous <- dataList[[prevdate]]() %>% as.data.frame() %>%
    # select required columns: base 
    select(Name, Age, Position, Rec, all_of(datapoints))
  
  # current date data
  current <- dataList[[currdate]]() %>% as.data.frame() %>%
    select(Name, Age, Position, Rec, all_of(datapoints))
  
  # difference dataframe
  # name column: union of all names in the prev & curr data sets
  difference <- data.frame(Name = base::union(previous$Name, current$Name))
  # add a comment column: check if player is new or out
  # stack() to transform named vector to dataframe
  comment <- stack(sapply(difference$Name, function(x) {
    case_when(
      # if name in prev but not in curr: player is out
      (x %in% previous$Name) & !(x %in% current$Name) ~ "Player no longer at club",
      # if name not in prev but in curr: player is new
      !(x %in% previous$Name) & (x %in% current$Name) ~ "New player",
      # if name in prev and curr: player has stayed; no comment
      (x %in% previous$Name) & (x %in% current$Name) ~ ""
    )
  })) %>%
    # rename values column to facilitate the inner_join later
    mutate(Comment = values, .keep = "unused")
  
  # add corresponding comment to the player
  difference %<>% inner_join(comment, by = c("Name" = "ind"))
  
  # difference: prev version contains all names from difference
  difference_prev <- difference %>% left_join(previous, by = c("Name" = "Name"))
  # difference: curr version contains all names from difference
  difference_curr <- difference %>% left_join(current, by = c("Name" = "Name"))
  
  # set difference as current version
  difference <- difference_curr
  # compute per 90 statistics for appropriate data if tabulating performance progress
  if (type != "attr") {
    for (i in list(difference_curr, difference_prev)) {
      i %<>%
        mutate(`K Hdrs/90` = `K Hdrs`/Mins, .after = `K Hdrs`) %>%
        mutate(`Gl Mst/90` = `Gl Mst`/Mins, .after = `Gl Mst`) %>%
        mutate(`Off/90` = `Off`/Mins, .after = `Off`) %>%
        mutate(`Cr A/90` = `Cr A`/Mins, .after = `Cr A`) %>%
        mutate(`Cr C/90` = `Cr C`/Mins, .after = `Cr C`) %>%
        mutate(`K Tck/90` = `K Tck`/Mins, .after = `K Tck`) %>%
        mutate(`Tck A/90` = `Tck A`/Mins, .after = `Tck A`) %>%
        mutate(`Tck W/90` = `Tck W`/Mins, .after = `Tck W`) %>%
        mutate(`Svh/90` = `Svh`/Mins, .after = `Svh`) %>%
        mutate(`Svp/90` = `Svp`/Mins, .after = `Svp`) %>%
        mutate(`Svt/90` = `Svt`/Mins, .after = `Svt`) %>%
        mutate(`FA/90` = `FA`/Mins, .after = `FA`) %>%
        mutate(`Fls/90` = `Fls`/Mins, .after = `Fls`) %>%
        mutate(`Yel/90` = `Yel`/Mins, .after = `Yel`) %>%
        mutate(`Red/90` = `Red`/Mins, .after = `Red`) %>%
        select(-c(
          `K Hdrs`, `Gl Mst`, `Off`, `Cr A`, `Cr C`, `K Tck`, `Tck A`,
          `Tck W`, `Svh`, `Svp`, `Svt`, `FA`, `Fls`, `Yel`, `Red`
        ))
    }
  }
  
  # compute difference between curr and prev
  for (i in c("Rec", datapoints)) {
    difference[i] <- difference_curr[[i]] - difference_prev[[i]]
  }
  
  # add overall column if tabulating the attributes progress
  if (type == "attr") {
    difference %<>% mutate(Overall = rowSums(select(., all_of(datapoints))), .after = "Rec")
    # round values to 2 digits
    for (i in c("Rec", datapoints)) {
      difference[i] <- round(difference[[i]], digits = 2)
    }
  }
  
  # initialise DT column options
  DTcols_options <- list()
  # options depending on dataset type: attr or stats
  if (type == "attr") {
    # grouped columns
    DTcols_options$cols <- htmltools::withTags(table(
      class = "display",
      thead(
        tr(
          # group columns and hover info
          th(rowspan = 2, "Name", title = "Player Name"),
          th(rowspan = 2, "Comment", title = "Comment"),
          th(rowspan = 2, "Age", title = "Age"),
          th(rowspan = 2, "Position", title = "Position"),
          th(rowspan = 2, "DNA Rating", title = "DNA Rating"),
          th(rowspan = 2, "Overall", title = "Total variation of attributes"),
          th(colspan = 11, "Goalkeeping", aligne = "center"),
          th(colspan = 14, "Mental", aligne = "center"),
          th(colspan = 8, "Physical", aligne = "center"),
          th(colspan = 14, "Technical", aligne = "center")
        ),
        tr(
          # individual columns and hover info
          th("Aer", title = "Aerial Reach"),
          th("Cmd", title = "Command Of Area"),
          th("Com", title = "Communication"),
          th("Ecc", title = "Eccentricity"),
          th("Han", title = "Handling"),
          th("Kic", title = "Kicking"),
          th("1v1", title = "One On Ones"),
          th("Pun", title = "Punching (Tendency)"),
          th("Ref", title = "Reflexes"),
          th("TRO", title = "Rushing Out (Tendency)"),
          th("Thr", title = "Throwing"),
          th("Agg", title = "Aggression"),
          th("Ant", title = "Anticipation"),
          th("Bra", title = "Bravery"),
          th("Cmp", title = "Composure"),
          th("Cnt", title = "Concentration"),
          th("Dec", title = "Decisions"),
          th("Det", title = "Determination"),
          th("Fla", title = "Flair"),
          th("Ldr", title = "Leadership"),
          th("OtB", title = "Off The Ball"),
          th("Pos", title = "Positioning"),
          th("Tea", title = "Teamwork"),
          th("Vis", title = "Vision"),
          th("Wor", title = "Work Rate"),
          th("Acc", title = "Acceleration"),
          th("Agi", title = "Agility"),
          th("Bal", title = "Balance"),
          th("Jum", title = "Jumping Reach"),
          th("Nat", title = "Natural Fitnes"),
          th("Pac", title = "Pace"),
          th("Sta", title = "Stamina"),
          th("Str", title = "Strength"),
          th("Cor", title = "Corners"),
          th("Cro", title = "Crossing"),
          th("Dri", title = "Dribbling"),
          th("Fin", title = "Finishing"),
          th("Fir", title = "First Touch"),
          th("Fre", title = "Free Kicks"),
          th("Hea", title = "Heading"),
          th("Lon", title = "Long Shots"),
          th("L Th", title = "Long Throws"),
          th("Mar", title = "Marking"),
          th("Pas", title = "Passing"),
          th("Pen", title = "Penalty Taking"),
          th("Tck", title = "Tackling"),
          th("Tec", title = "Technique")
        )
      )
    ))
    # background color for grouped columns names
    DTcols_options$group_JSrender <- "
    function(thead) {
    $(thead).closest('thead').find('th').eq(6).css('background-color', '#D9E1F2');
    $(thead).closest('thead').find('th').eq(7).css('background-color', '#8EA9DB');
    $(thead).closest('thead').find('th').eq(8).css('background-color', '#D9E1F2');
    $(thead).closest('thead').find('th').eq(9).css('background-color', '#8EA9DB')
    }
    "
    # sort order
    DTcols_options$order <- list(
      list(1, "asc"), # comment
      list(5, "desc"), # overall
      list(4, "desc"), # DNA rating
      list(2, "desc") # Age
    )
  } else {
    # grouped columns
    DTcols_options$cols <- htmltools::withTags(table(
      class = "display",
      thead(
        tr(
          # group columns and hover info
          th(rowspan = 2, "Name", title = "Player Name"),
          th(rowspan = 2, "Comment", title = "Comment"),
          th(rowspan = 2, "Age", title = "Age"),
          th(rowspan = 2, "Position", title = "Position"),
          th(rowspan = 2, "DNA Rating", title = "DNA Rating"),
          th(colspan = 4, "Aerial Challenges"),
          th(colspan = 1, "Mistakes", aligne = "center"),
          th(colspan = 3, "Movement", aligne = "center"),
          th(colspan = 3, "Crosses", aligne = "center"),
          th(colspan = 6, "Passes", aligne = "center"),
          th(colspan = 4, "Shots", aligne = "center"),
          th(colspan = 5, "Tackles & Interceptions", aligne = "center"),
          th(colspan = 4, "Saves", aligne = "center"),
          th(colspan = 20, "General", aligne = "center")
        ),
        tr(
          # individual columns and hover info
          th("Aer A/90", title = "Aerial Challenges attempts per 90 minutes"),
          th("Hdrs W/90", title = "Headers won per 90 minutes"),
          th("Hdr %", title = "Headers Won Ratio"),
          th("K Hdrs/90", title = "Key Headers per 90 minutes"),
          th("Gl Mst/90", title = "Mistakes Leading To Goal per 90 minutes"),
          th("Dist/90", title = "Distance Covered Per 90 Minutes"),
          th("DrbPG", title = "Dribbles Made Per Game"),
          th("Off/90", title = "Offsides per 90 minutes"),
          th("Cr A/90", title = "Cross Attempts per 90 minutes"),
          th("Cr C/90", title = "Crosses Completed per 90 minutes"),
          th("Cr C/A", title = "Cross Completion Ratio"),
          th("Ps A/90", title = "Pass attempts per 90 minutes"),
          th("Ps C/90", title = "Passes completed per 90 minutes"),
          th("Pas %", title = "Pass Completion Ratio"),
          th("K Ps/90", title = "Key Passes per 90 minutes"),
          th("Ch C/90", title = "Chances Created per 90 minutes"),
          th("Asts/90", title = "Assists per 90 minutes"),
          th("Shot/90", title = "Shots per 90 minutes"),
          th("ShT/90", title = "Shorts on target per 90 minutes"),
          th("Shot %", title = "Shots On Target Ratio"),
          th("Gls/90", title = "Goals per 90 minutes"),
          th("Int/90", title = "Interceptions per 90 minutes"),
          th("K Tck/90", title = "Key Tackles per 90 minutes"),
          th("Tck A/90", title = "Tackle attempts per 90 minutes"),
          th("Tck W/90", title = "Tackles Completed per 90 minutes"),
          th("Tck R", title = "Tackle completion ratio"),
          th("Svh/90", title = "Saves Held per 90 minutes"),
          th("Svp/90", title = "Saves Parried per 90 minutes"),
          th("Svt/90", title = "Saves Tipped per 90 minutes"),
          th("Con/90", title = "Conceded per 90 minutes"),
          th("Apps", title = "Appearances"),
          th("Starts", title = "Starting Appearances"),
          th("Mins", title = "Minutes"),
          th("Av Rat", title = "Average rating"),
          th("Clean sheets", title = "Clean sheets"),
          th("Gwin", title = "Game win ratio"),
          th("PoM", title = "Player Of The Match"),
          th("Pts/Gm", title = "Points won per game"),
          th("Tcon/90", title = "Team conceded per 90 minutes"),
          th("Tgls/90", title = "Team goals per 90 minutes"),
          th("Pens Faced", title = "Penalties Faced"),
          th("Pens Saved", title = "Penalties Saved"),
          th("Pens Saved Ratio", title = "Penalties Saved Ratio"),
          th("Pens", title = "Penalties"),
          th("Pens S", title = "Penalties scored"),
          th("Pen/R", title = "Penalties scored ratio"),
          th("FA/90", title = "Fouls Against per 90 minutes"),
          th("Fls/90", title = "Fouls Made per 90 minutes"),
          th("Yel/90", title = "Yellow cards per 90 minutes"),
          th("Red/90", title = "Red cards per 90 minutes")
        )
      )
    ))
    # background color for grouped columns names
    DTcols_options$group_JSrender <- "
    function(thead) {
    $(thead).closest('thead').find('th').eq(5).css('background-color', '#D9E1F2');
    $(thead).closest('thead').find('th').eq(6).css('background-color', '#8EA9DB');
    $(thead).closest('thead').find('th').eq(7).css('background-color', '#D9E1F2');
    $(thead).closest('thead').find('th').eq(8).css('background-color', '#8EA9DB');
    $(thead).closest('thead').find('th').eq(9).css('background-color', '#D9E1F2');
    $(thead).closest('thead').find('th').eq(10).css('background-color', '#8EA9DB');
    $(thead).closest('thead').find('th').eq(11).css('background-color', '#D9E1F2');
    $(thead).closest('thead').find('th').eq(12).css('background-color', '#8EA9DB');
    $(thead).closest('thead').find('th').eq(13).css('background-color', '#D9E1F2')
    }
    "
    # sort order
    DTcols_options$order <- list(
      list(1, "asc"), # comment
      list(4, "desc"), # DNA rating
      list(2, "desc") # Age
    )
  }
  
  # number of freezed columns
  nb_freezed_cols <- if_else(type == "attr", 6, 5)
  
  # DT table
  DT::datatable(
    data = difference,
    # no row names displayed
    rownames = FALSE,
    # replace Rec column name: DNA rating
    colnames = c("DNA Rating" = "Rec"),
    # required extensions
    extensions = "FixedColumns",
    # options
    options = list(
      # freeze first 6 columns for attributes and 5 for stats
      fixedColumns = list(leftColumns = nb_freezed_cols),
      # no paging, scroll instead
      paging = FALSE,
      # grouped columns styling
      headerCallback = JS(DTcols_options$group_JSrender),
      # order
      order = DTcols_options$order,
      # horizontal scrolling enabled
      scrollX = TRUE,
      # vertical scrolling enabled
      scrollY = "450px"
    ),
    # format grouped columns
    container = DTcols_options$cols
  ) %>%
    # format font color and cell background depending on data
    formatStyle(
      5:dim(difference)[1],
      backgroundColor = styleInterval(c(-.0001,0), c("#FFC7CE","#FFEB9C", "#C6EFCE")),
      color = styleInterval(c(-.0001,0), c("#9C0006","#9C6500", "#006100"))
    )
}

# DT tables for the filtered data
tabulate_filtdata <- function(data, dataset) {
  # name of the Rec column in the table
  Rec_name <- if_else(dataset == "squad", "DNA Rating", "Scouting Rec")
  # replace name in data
  names(data)[2] <- Rec_name
  
  # DT table
  DT::datatable(
    data = data,
    # no row names displayed
    rownames = FALSE,
    # replace Rec column name by appropriate name: dataset
    # add "p/w" to wage column name
    colnames = c("Wage (p/w)" = "Wage"),
    # options
    options = list(
      # no paging, scroll instead
      paging = FALSE,
      # order
      order = list(
        list(7, "desc"), # games played
        list(5, "desc"), # value
        list(6, "desc"), # wage
        list(1, "desc"), # rec
        list(4, "desc") # age
      ),
      # horizontal scrolling enabled
      scrollX = TRUE,
      # vertical scrolling enabled
      scrollY = "500px"
    )
  ) %>%
    # format font color and cell background depending on data
    formatStyle(
      # format all row based on Rec's value
      2, target = "row",
      # original colors
      # c("#bf3eff", "#ff0000", "#ffa500", "#ffff00", "#00ee00", "#008b00")
      backgroundColor = styleInterval(
        c(-1, 49, 59, 69, 84),
        c("#e5b1ff", "#ff9999", "#ffdb99", "#ffff99", "#92ff92", "#20ff20")
      ),
      color = styleInterval(
        c(-1, 49, 59, 69, 84),
        c("#6c00a2", "#990000", "#996300", "#666600", "#008100", "#004000")
      )
    ) %>%
    # format Value and Wage as currency
    formatCurrency(
      columns = c("Value", "Wage (p/w)"),
      currency = "\u20ac ",
      interval = 3, mark = ",",
      digits = 2, dec.mark = ".",
      before = TRUE
    )
}

# creates a subset of the data using and index
# Name, Rec, Age, Position, Mins always part of the data: used in the plot
# {...} represents columns that are selecting depending of the submenu
subsetdata <- function(data, reactindex, ...) {
  as.data.frame(data) %>%
    dplyr::filter(Name %in% reactindex) %>%
    dplyr::select(
      Name,
      Rec,
      Age,
      Position,
      Mins,
      Nat1,
      Club,
      ...
    ) %>%
    dplyr::mutate(
      Mins = (Mins/90) %>% round(digits = 2)
    )
}

# Data Import -------------------------------------------------------------

# import squad data and measure execution time
cat("\nImporting squad data...")
tic("\nImported squad data")
# extract all squad files in the data folder
squads <- list.files("data/") %>% .[grepl("^squad", list.files("data"))]
# extract the in-game dates of each squad snapshot
squads_dates <- squads %>% str_remove_all(".html|squad_")
# extract the most recent in-game date from the squad snapshots
squads_latest_date <- squads_dates %>% as.numeric() %>% max() %>% as.character()
# empty list to contain all the squad snapshots
squads <- list()
# import all squad snapshots in a list
for (i in squads_dates) {
  # import squad snapshot
  squads[[i]] <- parse_FMoutput(paste0("data/squad_", i, ".html"))
  # # import squad snapshot
  # squads$tempname <- parse_FMoutput(paste0("data/squad_", i, ".html"))
  # # replace the name by the date
  # names(squads) %<>% replace(. == "tempname", i)
}
toc()

# import shortlist data and measure execution time
cat("\nImporting shortlist data...")
tic("\nImported shortlist data")
shortlist <- parse_FMoutput("data/shortlist.html")
toc()

# Data cleaning & formatting ----------------------------------------------

# clean & format squad data and measure execution time
cat("\nCleaning & formatting squad data...")
tic("\nCleaned & formatted squad data")
# clean & format all squad snapshot data
for (i in squads_dates) {
  squads[[i]] %<>% mutate(Rec = NA, .after = Name) %>% cleandata()
}
toc()

# clean & format shortlist data and measure execution time
cat("\nCleaning & formatting shortlist data...")
tic("\nCleaned & formatted shortlist data")
shortlist %<>% cleandata()
toc()

# App inputs 2 ------------------------------------------------------------
# dependant on the data

# status message
cat("\nLoading app data...")
tic("\nLoaded app data")

# unique clubs appearing in the squad
# i.e. the managed clubs + clubs with players on loan
# managed club first on the list
sqclubs <- squads[[squads_latest_date]]$Club %>%
  # get frequencies of Club in the list
  cbind(freq = ave(squads[[squads_latest_date]]$Club, squads[[squads_latest_date]]$Club, FUN = length)) %>%
  unique() %>%
  as.data.frame() %>%
  # sort by desc order using the frequencies: managed club in first position
  dplyr::arrange(desc(freq)) %>%
  # pull list of clubs as vector
  dplyr::pull(1)

# unique divisions appearing in the shortlist (put free player at the end)
divisions <- shortlist$Division %>% unique() %>% sort() %>%
  .[. != "Free player/unknown"] %>% c("Free player/unknown")


# App ui: header ----------------------------------------------------------

Header <- dashboardHeader(
  title = "FM Data Analytics"
)

# App ui: sidebar ---------------------------------------------------------

Sidebar <- dashboardSidebar(
  sidebarMenu(
    menuItem(
      "Home", tabName = "home", icon = icon("home"),
      selected = TRUE
    ),
    
    menuItem(
      "Squad Analysis", tabName = "squad", icon = icon("users"),
      startExpanded = FALSE,
      menuSubItem(
        "DNA Model", tabName = "sqmodel", icon = icon("sliders-h")
      ),
      menuSubItem(
        "Filters", tabName = "sqfilter", icon = icon("filter")
      ),
      menuSubItem(
        "General", tabName = "sqgen"
      ),
      menuSubItem(
        "Passing", tabName = "sqpass"
      ),
      menuSubItem(
        "Shooting", tabName = "sqshoot"
      ),
      menuSubItem(
        "Movement", tabName = "sqmove"
      ),
      menuSubItem(
        "Aerial Challenges", tabName = "sqaerial"
      ),
      menuSubItem(
        "Tackling & Intercepting", tabName = "sqtackle"
      ),
      menuSubItem(
        "Progress", tabName = "sqprog", icon = icon("chart-line")
      )
    ),
    
    menuItem(
      "Scouting", tabName = "scouting", icon = icon("search"),
      startExpanded = FALSE,
      menuSubItem(
        "Filters", tabName = "scfilter", icon = icon("filter")
      ),
      menuSubItem(
        "General", tabName = "scgen"
      ),
      menuSubItem(
        "Passing", tabName = "scpass"
      ),
      menuSubItem(
        "Shooting", tabName = "scshoot"
      ),
      menuSubItem(
        "Movement", tabName = "scmove"
      ),
      menuSubItem(
        "Aerial Challenges", tabName = "scaerial"
      ),
      menuSubItem(
        "Tackling & Intercepting", tabName = "sctackle"
      ),
      menuSubItem(
        "Shortlist", tabName = "sclist", icon = icon("list")
      )
    ),
    
    menuItem(
      "About/more info", tabName = "aboutapp", icon = icon("info-circle")
    )
  )
)

# App ui: body ------------------------------------------------------------

Body <- dashboardBody(
  tabItems(
    #### Home ####
    tabItem(
      tabName = "home",
      fluidRow(
        # valuebox for the number of players at the club
        valueBox(
          value = squads[[squads_latest_date]] %>% nrow(),
          subtitle = "players at the club",
          icon = icon("users"), color = "aqua",
          width = 4
        ),
        # valuebox for the number of players in the shortlist
        valueBox(
          value = shortlist %>% nrow(),
          subtitle = "players in the shortlist",
          icon = icon("search"), color = "aqua",
          width = 4
        )
      ),
      tabBox(
        title = "Summary",
        width = 12, height = "70vh",
        tabPanel(
          # goal contributions distribution bar plot
          title = "Goal Contributions",
          plotlyOutput(outputId = "home_plot_gcd")
        ),
        tabPanel(
          # playing time boxplot
          title = "Playing Time",
          plotlyOutput(outputId = "home_plot_pt"),
          HTML(
            "
            <p><br></p>
            <p><em>* Use the boxplot diagrams to understand the data distribution. Actual points are available on the left of each diagram and information is available when hovered on.</em></p>
            "
          )
        ),
        tabPanel(
          # wage boxplot
          title = "Wage",
          plotlyOutput(outputId = "home_plot_wage"),
          HTML(
            "
            <p><br></p>
            <p><em>* Use the boxplot diagrams to understand the data distribution. Actual points are available on the left of each diagram and information is available when hovered on.</em></p>
            <p><em>** <strong>Loaned players</strong> and <strong>loaned out players</strong> are <u>not included</u>.</em></p>
            "
          )
        )
      )
    ),
    
    #### Squad: DNA Model submenu ####
    tabItem(
      tabName = "sqmodel",
      box(
        title = "Attributes Selection",
        width = 12, height = "85vh",
        background = "purple",
        HTML(
          "
          <p>Constitute your DNA model by selecting key attributes for your playing style. The selected attributes will be used to calculate your players DNA rating according to the model.</p>
          "
        ),
        splitLayout(
          cellWidths = c("25%", "25%", "25%", "25%"),
          # technical attributes
          checkboxGroupInput(
            inputId = "sqmodel_technical",
            label = "Technical",
            choices = c(
              "Corners" = "Cor",
              "Crossing" = "Cro",
              "Dribbling" = "Dri",
              "Finishing" = "Fin",
              "First Touch" = "Fir",
              "Free Kicks" = "Fre",
              "Heading" = "Hea",
              "Long Shots" = "Lon",
              "Long Throws" = "L Th",
              "Marking" = "Mar",
              "Passing" = "Pas",
              "Penalty Taking" = "Pen",
              "Tackling" = "Tck",
              "Technique" = "Tec"
            ),
            selected = NULL
          ),
          # mental attributes
          checkboxGroupInput(
            inputId = "sqmodel_mental",
            label = "Mental",
            choices = c(
              "Aggression" = "Agg",
              "Anticipation" = "Ant",
              "Bravery" = "Bra",
              "Composure" = "Cmp",
              "Concentration" = "Cnt",
              "Decisions" = "Dec",
              "Determination" = "Det",
              "Flair" = "Fla",
              "Leadership" = "Ldr",
              "Off The Ball" = "OtB",
              "Positioning" = "Pos",
              "Teamwork" = "Tea",
              "Vision" = "Vis",
              "Work Rate" = "Wor"
            ),
            selected = NULL
          ),
          # physical attributes
          checkboxGroupInput(
            inputId = "sqmodel_physical",
            label = "Physical",
            choices = c(
              "Acceleration" = "Acc",
              "Agility" = "Agi",
              "Balance" = "Bal",
              "Jumping Reach" = "Jum",
              "Natural Fitnes" = "Nat",
              "Pace" = "Pac",
              "Stamina" = "Sta",
              "Strength" = "Str"
            ),
            selected = NULL
          ),
          # goalkeeping attributes
          checkboxGroupInput(
            inputId = "sqmodel_gk",
            label = "Goalkeeping",
            choices = c(
              "Aerial Reach" = "Aer",
              "Command Of Area" = "Cmd",
              "Communication" = "Com",
              "Eccentricity" = "Ecc",
              "Handling" = "Han",
              "Kicking" = "Kic",
              "One On Ones" = "1v1",
              "Punching (Tendency)" = "Pun",
              "Reflexes" = "Ref",
              "Rushing Out (Tendency)" = "TRO",
              "Throwing" = "Thr"
            ),
            selected = NULL
          )
        ),
        # deselect buttons for each of the 4 attributes categories
        splitLayout(
          cellWidths = c("25%", "25%", "25%", "25%"),
          actionButton(
            inputId = "sqmodel_deselect_tec", label = " Deselect all technical",
            width = '100%', style = 'margin-bottom:0.5px'
          ),
          actionButton(
            inputId = "sqmodel_deselect_men", label = " Deselect all mental",
            width = '100%', style = 'margin-bottom:0.5px'
          ),
          actionButton(
            inputId = "sqmodel_deselect_phy", label = " Deselect all physical",
            width = '100%', style = 'margin-bottom:0.5px'
          ),
          actionButton(
            inputId = "sqmodel_deselect_gk", label = " Deselect all goalkeeping",
            width = '100%', style = 'margin-bottom:0.5px'
          )
        ),
        # Remarks about the DNA rating
        HTML(
          "
          <br><p><em>* The rating is calculated as follows: Average(key attributes)/20 rounded to the nearest integer. For instance 15(or 12 or 8) on all key attributes yield a DNA rating of 75(or 60 or 40).</em></p>
          <p><em>** Outfield players rating do not use Goalkeeping attributes. Goalkeepers ratings use the Goalkeeping attributes instead of the Technical attributes. No attributes selected yield a DNA rating of -1 (same as scouting recommandation for non-scouted players).</em></p>
          "
        )
      )
    ),
    
    #### Squad: Filters submenu ####
    tabItem(
      tabName = "sqfilter",
      fluidRow(
        # filters box
        box(
          title = "Filters",
          width = 3, height = "85vh",
          background = "purple",
          # division filter
          pickerInput(
            inputId = "squad_club", label = "Club(s):",
            choices = sqclubs, selected = sqclubs, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 7,
              `selected-text-format` = "count > 0"
            )
          ),
          # position filter
          pickerInput(
            inputId = "squad_pos", label = "Position(s):",
            choices = positions, selected = positions, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 7,
              `selected-text-format` = "count > 0"
            )
          ),
          # age filter
          sliderInput(
            inputId = "squad_age", label = "Age:",
            min = 16, max = 45, value = c(16,45), step = 1
          ),
          # minimal total appearances filter
          sliderInput(
            inputId = "squad_minapps", label = "Min total appearances:",
            min = 0, max = 30, value = 0, step = 1
          )
        ),
        # filtered subset overview
        box(
          title = "Filtered subset overview",
          width = 9, height = '85vh',
          status = "primary", solidHeader = FALSE,
          DT::DTOutput(
            outputId = "squad_filtered"
          )
        )
      )
    ),
    
    #### Squad: General submenu ####
    tabItem(
      tabName = "sqgen",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "sqgen_rating",
            label = "1 - Rating Consistency",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqgen_pointsw",
            label = "2 - Team Results Consistency",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqgen_teamggoalass",
            label = "3 - Goals + Assists Contribution",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqgen_goalass",
            label = "4 - Goals and Assists",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqgen_cards",
            label = "5 - Cards",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "sqgen_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "sqgen_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "sqgen_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "sqgen_plot3")
          ),
          tabPanel(
            title = "4", value = "panel4",
            plotlyOutput(outputId = "sqgen_plot4")
          ),
          tabPanel(
            title = "5", value = "panel5",
            plotlyOutput(outputId = "sqgen_plot5")
          )
        )
      )
    ),
    
    #### Squad: Passing submenu ####
    tabItem(
      tabName = "sqpass",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "sqpass_attcomp",
            label = "1 - Attempted vs Completed Passes",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqpass_compkey",
            label = "2 - Completed vs Key Passes",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqpass_compchan",
            label = "3 - Completed Passes vs Chances Created",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqpass_compass",
            label = "4 - Completed Passes vs Assists",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqpass_teamgass",
            label = "5 - Assists Contribution",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqpass_crossattcomp",
            label = "6 - Attempted vs Completed Crosses",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "sqpass_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "sqpass_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "sqpass_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "sqpass_plot3")
          ),
          tabPanel(
            title = "4", value = "panel4",
            plotlyOutput(outputId = "sqpass_plot4")
          ),
          tabPanel(
            title = "5", value = "panel5",
            plotlyOutput(outputId = "sqpass_plot5")
          ),
          tabPanel(
            title = "6", value = "panel6",
            plotlyOutput(outputId = "sqpass_plot6")
          )
        )
      )
    ),
    
    #### Squad: Shooting submenu ####
    tabItem(
      tabName = "sqshoot",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "sqshoot_atttarg",
            label = "1 - Attempted vs On Target Shots",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqshoot_targgoal",
            label = "2 - On Target Shots vs Goals",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqshoot_teamggoal",
            label = "3 - Goals Contribution",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqshoot_attxg",
            label = "4 - Attempted Shots vs xG",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqshoot_xggoal",
            label = "5 - xG vs Goals",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "sqshoot_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "sqshoot_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "sqshoot_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "sqshoot_plot3")
          ),
          tabPanel(
            title = "4", value = "panel4",
            plotlyOutput(outputId = "sqshoot_plot4")
          ),
          tabPanel(
            title = "5", value = "panel5",
            plotlyOutput(outputId = "sqshoot_plot5")
          )
        )
      )
    ),
    
    #### Squad: Movement submenu ####
    tabItem(
      tabName = "sqmove",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "sqmove_drib",
            label = "1 - Dribbling",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqmove_offs",
            label = "2 - Offsides",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqmove_dist",
            label = "3 - Distance",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "sqmove_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "sqmove_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "sqmove_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "sqmove_plot3")
          )
        )
      )
    ),
    
    #### Squad: Aerial submenu ####
    tabItem(
      tabName = "sqaerial",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "sqaerial_attwon",
            label = "1 - Attempted vs Won Headers",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqaerial_wonkey",
            label = "2 - Won vs Key Headers",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "sqaerial_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "sqaerial_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "sqaerial_plot2")
          )
        )
      )
    ),
    
    #### Squad: Tackling submenu ####
    tabItem(
      tabName = "sqtackle",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "sqtackle_attcomp",
            label = "1 - Attempted vs Completed Tackles",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqtackle_compkey",
            label = "2 - Completed vs Key Tackles",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqtackle_attfoul",
            label = "3 - Attempted Tackles vs Fouls",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqtackle_attcard",
            label = "4 - Attempted Tackles vs Yellow/Red Cards",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqtackle_int",
            label = "5 - Interceptions",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sqtackle_mist",
            label = "6 - Mistakes Contribution",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "sqtackle_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "sqtackle_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "sqtackle_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "sqtackle_plot3")
          ),
          tabPanel(
            title = "4", value = "panel4",
            plotlyOutput(outputId = "sqtackle_plot4")
          ),
          tabPanel(
            title = "5", value = "panel5",
            plotlyOutput(outputId = "sqtackle_plot5")
          ),
          tabPanel(
            title = "6", value = "panel6",
            plotlyOutput(outputId = "sqtackle_plot6")
          )
        )
      )
    ),
    
    #### Squad: Progress submenu ####
    tabItem(
      tabName = "sqprog",
      # progress analysis tabbox
      tabBox(
        width = 12, height = "85vh",
        tabPanel(
          title = "Dates selection",
          HTML(
            "
            <p>Select a <em>&quot;start date&quot;</em> and an <em>&quot;end date&quot;</em> to compare your squad&apos;s data between the selected dates. The available dates to select from are inferred from the name of your squad data files extracted from Football Manager and stored in your <strong>data</strong> folder. To avoid errors, name all your squad data as follows: <strong><em>&quot;</em></strong><strong><em>squad_yyyymmdd&quot;</em></strong>, where <em>&quot;yyyymmdd&quot;</em> represents the in-game date.</p>
            <p><br></p>
            "
          ),
          # start date
          pickerInput(
            inputId = "squad_startdate", label = "Start date: ",
            choices = squads_dates %>% as.Date(format = "%Y%m%d") %>% sort() %>% format("%d %b %Y"),
            selected = squads_dates %>% min() %>% as.Date(format = "%Y%m%d") %>% format("%d %b %Y"),
            multiple = FALSE,
            options = list(
              `size` = 5
            )
          ),
          # end date
          pickerInput(
            inputId = "squad_enddate", label = "End date: ",
            choices = squads_dates %>% as.Date(format = "%Y%m%d") %>% sort() %>% format("%d %b %Y"),
            selected = squads_latest_date %>% as.Date(format = "%Y%m%d") %>% format("%d %b %Y"),
            multiple = FALSE,
            options = list(
              `size` = 5
            )
          )
        ),
        tabPanel(
          title = "Performance",
          DT::DTOutput(
            outputId = "squad_progperf"
          )
        ),
        tabPanel(
          title = "Attributes",
          DT::DTOutput(
            outputId = "squad_progattr"
          )
        )
      )
    ),
    
    #### Scouting: Filters submenu ####
    tabItem(
      tabName = "scfilter",
      fluidRow(
        # filters box
        box(
          title = "Filters",
          width = 3, height = "85vh",
          background = "purple",
          # division filter
          pickerInput(
            inputId = "scout_div", label = "Division(s):",
            choices = divisions, selected = divisions, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 7,
              `selected-text-format` = "count > 0"
            )
          ),
          # position filter
          pickerInput(
            inputId = "scout_pos", label = "Position(s):",
            choices = positions, selected = positions, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 7,
              `selected-text-format` = "count > 0"
            )
          ),
          # age filter
          sliderInput(
            inputId = "scout_age", label = "Age:",
            min = 16, max = 45, value = c(16,45), step = 1
          ),
          # maximal value and wage filters
          splitLayout(
            cellWidths = c('50%','50%'),
            autonumericInput(
              inputId = "scout_maxvalue", label = "Max value:",
              value = 50000000, maximumValue = 500000000, minimumValue = 0,
              align = "right",
              currencySymbol = "\u20ac ", currencySymbolPlacement = "p",
              decimalCharacter = ".", digitGroupSeparator = ",",
              decimalPlaces = 0
            ),
            autonumericInput(
              inputId = "scout_maxwage", label = "Max wage (p/w):",
              value = 500000, maximumValue = 5000000, minimumValue = 0,
              align = "right",
              currencySymbol = "\u20ac ", currencySymbolPlacement = "p",
              decimalCharacter = ".", digitGroupSeparator = ",",
              decimalPlaces = 0
            )
          ),
          # minimal total appearances filter
          sliderInput(
            inputId = "scout_minapps", label = "Min total appearances:",
            min = 0, max = 30, value = 0, step = 1
          )
        ),
        # filtered subset overview
        box(
          title = "Filtered subset overview",
          width = 9, height = '85vh',
          status = "primary", solidHeader = FALSE,
          DT::DTOutput(
            outputId = "scout_filtered"
          )
        )
      )
    ),
    
    #### Scouting: General submenu ####
    tabItem(
      tabName = "scgen",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "scgen_rating",
            label = "1 - Rating Consistency",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scgen_pointsw",
            label = "2 - Team Results Consistency",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scgen_teamggoalass",
            label = "3 - Goals + Assists Contribution",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scgen_goalass",
            label = "4 - Goals and Assists",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scgen_cards",
            label = "5 - Cards",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "scgen_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "scgen_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "scgen_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "scgen_plot3")
          ),
          tabPanel(
            title = "4", value = "panel4",
            plotlyOutput(outputId = "scgen_plot4")
          ),
          tabPanel(
            title = "5", value = "panel5",
            plotlyOutput(outputId = "scgen_plot5")
          )
        )
      ),
      fluidRow(
        # shorlist builder
        box(
          title = "Shortlist builder",
          width = 12, height = '15vh',
          background = "purple",
          pickerInput(
            inputId = "scgen_shortl", label = NULL,
            choices = shortlist$Name, selected = shortlist$Name, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 10,
              `selected-text-format` = "count > 0"
            )
          )
        )
      )
    ),
    
    #### Scouting: Passing submenu ####
    tabItem(
      tabName = "scpass",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "scpass_attcomp",
            label = "1 - Attempted vs Completed Passes",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scpass_compkey",
            label = "2 - Completed vs Key Passes",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scpass_compchan",
            label = "3 - Completed Passes vs Chances Created",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scpass_compass",
            label = "4 - Completed Passes vs Assists",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scpass_teamgass",
            label = "5 - Assists Contribution",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scpass_crossattcomp",
            label = "6 - Attempted vs Completed Crosses",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "scpass_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "scpass_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "scpass_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "scpass_plot3")
          ),
          tabPanel(
            title = "4", value = "panel4",
            plotlyOutput(outputId = "scpass_plot4")
          ),
          tabPanel(
            title = "5", value = "panel5",
            plotlyOutput(outputId = "scpass_plot5")
          ),
          tabPanel(
            title = "6", value = "panel6",
            plotlyOutput(outputId = "scpass_plot6")
          )
        )
      ),
      fluidRow(
        # shorlist builder
        box(
          title = "Shortlist builder",
          width = 12, height = '15vh',
          background = "purple",
          pickerInput(
            inputId = "scpass_shortl", label = NULL,
            choices = shortlist$Name, selected = shortlist$Name, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 10,
              `selected-text-format` = "count > 0"
            )
          )
        )
      )
    ),
    
    #### Scouting: Shooting submenu ####
    tabItem(
      tabName = "scshoot",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "scshoot_atttarg",
            label = "1 - Attempted vs On Target Shots",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scshoot_targgoal",
            label = "2 - On Target Shots vs Goals",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scshoot_teamggoal",
            label = "3 - Goals Contribution",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scshoot_attxg",
            label = "4 - Attempted Shots vs xG",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scshoot_xggoal",
            label = "5 - xG vs Goals",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "scshoot_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "scshoot_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "scshoot_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "scshoot_plot3")
          ),
          tabPanel(
            title = "4", value = "panel4",
            plotlyOutput(outputId = "scshoot_plot4")
          ),
          tabPanel(
            title = "5", value = "panel5",
            plotlyOutput(outputId = "scshoot_plot5")
          )
        )
      ),
      fluidRow(
        # shorlist builder
        box(
          title = "Shortlist builder",
          width = 12, height = '15vh',
          background = "purple",
          pickerInput(
            inputId = "scshoot_shortl", label = NULL,
            choices = shortlist$Name, selected = shortlist$Name, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 10,
              `selected-text-format` = "count > 0"
            )
          )
        )
      )
    ),
    
    #### Scouting: Movement submenu ####
    tabItem(
      tabName = "scmove",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "scmove_drib",
            label = "1 - Dribbling",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scmove_offs",
            label = "2 - Offsides",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scmove_dist",
            label = "3 - Distance",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "scmove_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "scmove_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "scmove_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "scmove_plot3")
          )
        )
      ),
      fluidRow(
        # shorlist builder
        box(
          title = "Shortlist builder",
          width = 12, height = '15vh',
          background = "purple",
          pickerInput(
            inputId = "scmove_shortl", label = NULL,
            choices = shortlist$Name, selected = shortlist$Name, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 10,
              `selected-text-format` = "count > 0"
            )
          )
        )
      )
    ),
    
    #### Scouting: Aerial submenu ####
    tabItem(
      tabName = "scaerial",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "scaerial_attwon",
            label = "1 - Attempted vs Won Headers",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "scaerial_wonkey",
            label = "2 - Won vs Key Headers",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "scaerial_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "scaerial_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "scaerial_plot2")
          )
        )
      ),
      fluidRow(
        # shorlist builder
        box(
          title = "Shortlist builder",
          width = 12, height = '15vh',
          background = "purple",
          pickerInput(
            inputId = "scaerial_shortl", label = NULL,
            choices = shortlist$Name, selected = shortlist$Name, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 10,
              `selected-text-format` = "count > 0"
            )
          )
        )
      )
    ),
    
    #### Scouting: Tackling submenu ####
    tabItem(
      tabName = "sctackle",
      fluidRow(
        # visualization selection box
        box(
          title = "Visualization Selection",
          width = 3, height = '70vh',
          background = "purple",
          actionButton(
            inputId = "sctackle_attcomp",
            label = "1 - Attempted vs Completed Tackles",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sctackle_compkey",
            label = "2 - Completed vs Key Tackles",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sctackle_attfoul",
            label = "3 - Attempted Tackles vs Fouls",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sctackle_attcard",
            label = "4 - Attempted Tackles vs Yellow/Red Cards",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sctackle_int",
            label = "5 - Interceptions",
            width = '100%', style = 'margin-bottom:10px'
          ),
          actionButton(
            inputId = "sctackle_mist",
            label = "6 - Mistakes Contribution",
            width = '100%', style = 'margin-bottom:10px'
          )
        ),
        # visualizations box
        tabBox(
          id = "sctackle_tabbox", title = "Visualizations",
          width = 9, height = '70vh',
          tabPanel(
            title = "1", value = "panel1",
            plotlyOutput(outputId = "sctackle_plot1")
          ),
          tabPanel(
            title = "2", value = "panel2",
            plotlyOutput(outputId = "sctackle_plot2")
          ),
          tabPanel(
            title = "3", value = "panel3",
            plotlyOutput(outputId = "sctackle_plot3")
          ),
          tabPanel(
            title = "4", value = "panel4",
            plotlyOutput(outputId = "sctackle_plot4")
          ),
          tabPanel(
            title = "5", value = "panel5",
            plotlyOutput(outputId = "sctackle_plot5")
          ),
          tabPanel(
            title = "6", value = "panel6",
            plotlyOutput(outputId = "sctackle_plot6")
          )
        )
      ),
      fluidRow(
        # shorlist builder
        box(
          title = "Shortlist builder",
          width = 12, height = '15vh',
          background = "purple",
          pickerInput(
            inputId = "sctackle_shortl", label = NULL,
            choices = shortlist$Name, selected = shortlist$Name, multiple = TRUE,
            width = '100%',
            options = list(
              `actions-box` = TRUE,
              `live-search` = TRUE,
              `size` = 10,
              `selected-text-format` = "count > 0"
            )
          )
        )
      )
    ),
    
    #### Scouting: Shortlist ####
    tabItem(
      tabName = "sclist",
      box(
        title = "Shortlist output",
        width = 12, height = '85vh',
        status = "primary",
        DT::DTOutput(
          outputId = "scout_listfinal"
        )
      )
    ),
    
    #### About/more info tab ####
    tabItem(
      tabName = "aboutapp",
      fluidRow(
        box(
          title = "About",
          width = 6, height = '85vh',
          status = "primary",
          HTML(
            "
            <p>Designed to mirror professional football club's performance analysis tools, this app enables FM managers to use in-game statistics at their full potential. Whether it be to decide which players you&apos;ll target from the hundreds on your shortlist, or to identify sub-performing or over-performing players within your own squad, you will have the possibility to use the app&apos;s interactive visualizations to support your decisions.</p>
            <p>However, don&apos;t forget &mdash; statistics only tell one part of the story!</p>
            <p><br></p>
            <p><strong>Developed by:</strong>
            <br>gam945 (<em>SI forums</em>)/ybenadjal (<em>GitHub</em>)</p>
            <p><br></p>
            <p>PM me on the SI forums to contribute ;)</p>
            "
          )
        ),
        box(
          title = "More info",
          width = 6, height = '85vh',
          status = "primary",
          HTML(
            '
            <p>Github link:</p>
            <p><a href="https://github.com/ybenadjal/FM-Data-Analytics"><em>Github</em></a></p>
            <p><br></p>
            <p>For updates, guides, known issues or to report a bug:</p>
            <p><a href="https://community.sigames.com/forums/topic/545140-a-tool-based-on-r-shiny-for-the-fm-data-driven-manager/"><em>SI Forums thread</em></a></p>
            '
          )
        )
      )
    )
    
  )
)

# App ui ------------------------------------------------------------------

ui <- dashboardPage(Header, Sidebar, Body, skin = "purple")

# App server --------------------------------------------------------------

server <- function(input, output, session) {
  
  #### Home ####
  
  # goal contributions distribution plot
  output$home_plot_gcd <- renderPlotly({
    plot_ly(
      # goal contribution distribution data
      # use latest snapshot of the squad
      data = squads[[squads_latest_date]] %>%
        # required data points: name, club, goals, assists
        select(Name, Club, Gls, Ast, Position, Wage, Age, Mins) %>%
        # remove NA values
        drop_na() %>%
        # keep only players that are currently at the club, i.e remove loaned players
        dplyr::filter(Club == sort(table(Club), decreasing = TRUE) %>% names() %>% .[1]) %>%
        # remove players that have no goals 
        dplyr::filter(Gls + Ast != 0),
      # order goals and assists by decreasing order: left to right
      # add goals data
      x = ~reorder(reorder(Name, -Ast), -Gls), y = ~Gls, name = "Goals",
      type = "bar",
      # adds text when marker is hovered on
      hoverinfo = "text",
      # define text logic: name, age, position, mins, ratio and marker value
      hovertext = ~paste0(
        "<b>Name: ", Name,
        "<br>Age: </b>", Age,
        "<br><b>Position: </b>", Position,
        "<br><b>Wage: </b> \u20AC ", formatC(Wage, format="f", big.mark = ",", digits=0), " p/w",
        "<br><b>Games played: </b>", round(Mins/90, digits = 1)
      ),
      # text on the bar
      text = ~Gls, textposition = "auto",
      marker = list(color = "purple")
    ) %>%
      add_trace(
        # add assists data
        y = ~Ast, name = "Assists",
        # adds text when marker is hovered on
        hoverinfo = "text",
        # define text logic
        hovertext = ~paste0(
          "<b>Name: ", Name,
          "<br>Age: </b>", Age,
          "<br><b>Position: </b>", Position,
          "<br><b>Wage: </b> \u20AC ", formatC(Wage, format="f", big.mark = ",", digits=0), " p/w",
          "<br><b>Games played: </b>", round(Mins/90, digits = 1)
        ),
        # text on the bar
        text = ~Ast, textposition = "auto",
        marker = list(color = "cyan")
      ) %>%
      # no axis labels, add title
      layout(
        # title: X goals and Y assists in Z competitive matches
        title = ~paste0(
          sum(replace_na(Gls, 0)),
          " goals and ",
          sum(replace_na(Ast, 0)),
          " assists in ",
          # total number of minutes divided by 11 players divided by 90 minutes
          # gives total of competitive matches played
          sum(replace_na(Mins, 0)) %>% prod(1/11) %>% prod(1/90),
          " competitive matches"
        ),
        xaxis = list(title = ""), yaxis = list(title = ""),
        # data used for the dynamic plot title
        data = squads[[squads_latest_date]] %>%
          select(Club, Gls, Ast, Mins) %>%
          dplyr::filter(Club == sort(table(Club), decreasing = TRUE) %>% names() %>% .[1])
      )
  })
  
  # playing time distribution plot
  output$home_plot_pt <- renderPlotly({
    # dataframe for player status (agreed playing time) and status' category
    apt_category <- data.frame(apt = c(
      "Star Player", "Important Player", "Regular Starter",
      "Squad Player", "Impact Sub", "Fringe Player",
      "Emergency Backup", "B Team Regular", "Surplus to Requirements",
      "Breakthrough Prospect", "Future Prospect", "Youngster",
      "First-Choice Goalkeeper", "Cup Goalkeeper", "Backup",
      "Loaned out, Age\u226523", "Loaned out, 19\u2264Age\u226422", "Loaned out, Age\u226418"
    ), cat = c(rep("Senior Squad", 9), rep("Youth Players", 3), rep("Goalkeepers", 3), rep("Loaned Players", 3))) %>%
      # add number of occurrences for each player status
      mutate(
        apt = paste0(
          apt, " (", table(squads[[squads_latest_date]]$`Agreed Playing Time`)[apt], ")"
        ) %>%
          # replace NA by 0 in the occurrences
          str_replace_all("NA", "0")
      )
    
    plot_ly(
      # playing time data
      data = squads[[squads_latest_date]] %>%
        # required data points: playing time, age, minutes
        select(Name, Age, Position, Wage, `Agreed Playing Time`, Mins, Gls, Ast) %>%
        # games played = minutes/90
        mutate(`Games Played` = round(`Mins`/90, digits = 1) %>% replace_na(0), .keep = "unused") %>%
        # subset loaned out categories by age
        mutate(`Agreed Playing Time` = if_else(
          condition = is.na(`Agreed Playing Time`),
          true = case_when(
            `Age` <= 18 ~ "Loaned out, Age\u226418",
            between(`Age`,19,22) ~ "Loaned out, 19\u2264Age\u226422",
            `Age`>= 23 ~ "Loaned out, Age\u226523"
          ),
          false = `Agreed Playing Time`
        )) %>%
        # number of occurrences of each status
        mutate(freq = table(`Agreed Playing Time`)[`Agreed Playing Time`]) %>%
        # add frequency to status
        mutate(`Agreed Playing Time` = paste0(`Agreed Playing Time`, " (", freq, ")")) %>%
        # left join to add the categories
        left_join(y = apt_category, by = c("Agreed Playing Time" = "apt")) %>%
        # add loaned out category
        mutate(cat = replace_na(cat, "Loaned Players")),
      x = ~`Agreed Playing Time`, y = ~`Games Played`,
      type = "box", boxpoints = "all",
      color = ~cat,
      colors = c(
        "Senior Squad" = "purple", # squad
        "Youth Players" = "deepskyblue2", # youth
        "Goalkeepers" = "green3", # gk
        "Loaned Players" = "orange" # loaned
      ),
      # adds text when marker is hovered on
      hoverinfo = "text",
      # define text logic
      hovertext = ~paste0(
        "<b>Name: ", Name,
        "<br>Age: </b>", Age,
        "<br><b>Position: </b>", Position,
        "<br><b>Wage: </b> \u20AC ", formatC(Wage, format="f", big.mark = ",", digits=0), " p/w",
        "<br><b>Games played: </b>", replace_na(`Games Played`, 0),
        "<br><b>Goals: </b>", replace_na(Gls, 0),
        "<br><b>Assists: </b>", replace_na(Ast, 0)
      )
    ) %>%
      layout(
        xaxis = list(
          # no title axis
          title = "",
          # order x axis
          categoryorder = "array", categoryarray = apt_category$apt
        ),
        # y axis title
        yaxis = list(title = "Games Played")
      )
  })
  
  # playing time distribution plot
  output$home_plot_wage <- renderPlotly({
    # dataframe for player status (agreed playing time) and status' category
    apt_category <- data.frame(apt = c(
      "Star Player", "Important Player", "Regular Starter",
      "Squad Player", "Impact Sub", "Fringe Player",
      "Emergency Backup", "B Team Regular", "Surplus to Requirements",
      "Breakthrough Prospect", "Future Prospect", "Youngster",
      "First-Choice Goalkeeper", "Cup Goalkeeper", "Backup"
    ), cat = c(rep("Senior Squad", 9), rep("Youth Players", 3), rep("Goalkeepers", 3))) %>%
      # add number of occurrences for each player status
      mutate(
        apt = paste0(
          apt, " (", table((squads[[squads_latest_date]] %>% dplyr::filter(is.na(`Loan Expires`)))$`Agreed Playing Time`)[apt], ")"
        ) %>%
          # replace NA by 0 in the occurrences
          str_replace_all("NA", "0")
      )
    
    plot_ly(
      # wage data
      data = squads[[squads_latest_date]] %>%
        # required data points
        select(Name, Age, Position, `Agreed Playing Time`, Wage, `Loan Expires`, Mins, Gls, Ast) %>%
        # remove loaned in and loaned out players
        dplyr::filter(is.na(`Loan Expires`)) %>%
        # remove loan expiration column
        select(-`Loan Expires`) %>%
        # number of occurrences of each status
        mutate(freq = table(`Agreed Playing Time`)[`Agreed Playing Time`]) %>%
        # add frequency to status
        mutate(`Agreed Playing Time` = paste0(`Agreed Playing Time`, " (", freq, ")")) %>%
        # left join to add the categories
        left_join(y = apt_category, by = c("Agreed Playing Time" = "apt")),
      x = ~`Agreed Playing Time`, y = ~Wage,
      type = "box", boxpoints = "all",
      color = ~cat,
      colors = c(
        "Senior Squad" = "purple", # squad
        "Youth Players" = "deepskyblue2", # youth
        "Goalkeepers" = "green3" # gk
      ),
      # adds text when marker is hovered on
      hoverinfo = "text",
      # define text logic
      hovertext = ~paste0(
        "<b>Name: ", Name,
        "<br>Age: </b>", Age,
        "<br><b>Position: </b>", Position,
        "<br><b>Wage: </b> \u20AC ", formatC(Wage, format="f", big.mark = ",", digits=0), " p/w",
        "<br><b>Games played: </b>", round(replace_na(Mins, 0)/90, digits = 1),
        "<br><b>Goals: </b>", replace_na(Gls, 0),
        "<br><b>Assists: </b>", replace_na(Ast, 0)
      )
    ) %>%
      layout(
        xaxis = list(
          # no title axis
          title = "",
          # order x axis
          categoryorder = "array", categoryarray = apt_category$apt
        ),
        # y axis title
        yaxis = list(title = "Wage (p/w)")
      )
  })
  
  #### Squad: DNA Model ####
  
  # deselect all attributes from a category when the "Deselect all <category>" button is clicked on
  observeEvent(
    input$sqmodel_deselect_tec, {
      updateCheckboxGroupInput(
        session = session, inputId = "sqmodel_technical", selected = character(0)
      )
    }
  )
  observeEvent(
    input$sqmodel_deselect_men, {
      updateCheckboxGroupInput(
        session = session, inputId = "sqmodel_mental", selected = character(0)
      )
    }
  )
  observeEvent(
    input$sqmodel_deselect_phy, {
      updateCheckboxGroupInput(
        session = session, inputId = "sqmodel_physical", selected = character(0)
      )
    }
  )
  observeEvent(
    input$sqmodel_deselect_gk, {
      updateCheckboxGroupInput(
        session = session, inputId = "sqmodel_gk", selected = character(0)
      )
    }
  )
  
  # calculate the DNA ratings
  # initial empty reactive list
  squads_DNArated <- reactiveValues()
  # calculate each date's DNA ratings
  for (i in squads_dates) {
    squads_DNArated[[i]] <- reactive({
      if (length(c(input$sqmodel_technical, input$sqmodel_mental, input$sqmodel_physical, input$sqmodel_gk)) == 0) {
        squads[[i]] %>% mutate(`Rec` = -1)
      } else {
        # key outfield attributes
        outf_keyattr <- c(input$sqmodel_technical, input$sqmodel_mental, input$sqmodel_physical)
        outf_n <- length(outf_keyattr)
        # key gk attributes
        gk_keyattr <- c(input$sqmodel_mental, input$sqmodel_physical, input$sqmodel_gk)
        gk_n <- length(gk_keyattr)
        # subset data: outfield players
        outf_players <- squads[[i]] %>%
          dplyr::select(Name, Position, all_of(outf_keyattr)) %>%
          dplyr::filter(Position != "GK")
        # subset data: goalkeepers
        gk_players <- squads[[i]] %>%
          dplyr::select(Name, Position, all_of(gk_keyattr)) %>%
          dplyr::filter(Position == "GK")
        # get ratings for outfield players
        outf_players %<>% cbind(Rec = dplyr::select(outf_players, -c(Name, Position)) %>% rowSums()) %>%
          dplyr::select(Name, Position,Rec) %>%
          mutate(Rec = round(Rec*100/20/outf_n, digits = 0))
        # get ratings for outfield players
        gk_players %<>% cbind(Rec = dplyr::select(gk_players, -c(Name, Position)) %>% rowSums()) %>%
          dplyr::select(Name, Position,Rec) %>%
          mutate(Rec = round(Rec*100/20/gk_n, digits = 0))
        # ratings
        ratings <- rbind(outf_players, gk_players)
        # inner join by name and position
        squads[[i]] %>% dplyr::select(-Rec) %>%
          inner_join(y = ratings, by = c("Name" = "Name", "Position" = "Position"))
      }
    })
  }
  
  #### Squad: Filters ####
  
  # filtered subset data for DT table presentation
  squad_filtsubset <- reactive({
    squads_DNArated[[squads_latest_date]]() %>% as.data.frame() %>%
      select(Name, Rec, Position, PositionParsed, Division, Club, Age, Value, Wage, Mins) %>%
      # filter using the filters inputs
      dplyr::filter(Club %in% input$squad_club) %>%
      dplyr::filter(
        grepl_op(pattern_array = input$squad_pos, x = PositionParsed)
      ) %>%
      dplyr::filter(between(Age, input$squad_age[1], input$squad_age[2])) %>%
      # dplyr::filter(Value <= input$scout_maxvalue) %>%
      # dplyr::filter(Wage <= input$scout_maxwage) %>%
      dplyr::filter(Mins >= input$squad_minapps*90) %>%
      dplyr::select(-c(Division, PositionParsed)) %>%
      mutate(
        `Games Played` = round(Mins/90, digits = 1), .keep = "unused", .after = Mins
      ) %>%
      drop_na()
  })
  
  # filtered subset DT table
  output$squad_filtered <- DT::renderDT({
    tabulate_filtdata(data = squad_filtsubset(), dataset = "squad")
  })
  
  # filtered subset names: used as an index
  squad_index <- reactive({
    squad_filtsubset() %>% as.data.frame() %>% dplyr::select(Name)# %>% as.vector()
  })
  
  # filtered subset DT table for plots
  squad_data <- reactive({
    subsetdata(
      data = squads_DNArated[[squads_latest_date]]() %>% as.data.frame(),
      reactindex = as.vector(as.data.frame(squad_index())$Name),
      `Av Rat`, `Pts/Gm`, `Clean sheets`, # general
      `Ps A/90`, `Ps C/90`, `K Ps/90`, `Ch C/90`, `Asts/90`, `Cr A`, `Cr C`, # passing
      `Shot/90`, `ShT/90`, `Gls/90`, `Tgls/90`, `xG`, # shooting
      `DrbPG`, `Dist/90`, `Off`, # movement
      `Aer A/90`, `Hdrs W/90`, `K Hdrs`, # aerial
      `Tck R`, `Tck W`, `K Tck`, `Int/90`, `Fls`, `Red`, `Yel`, # tackling 1
      `Tcon/90`, `Gl Mst`#, # tackling 2
    ) %>%
      mutate(`Gls plus Asts` = `Gls/90` + `Asts/90`) %>%
      mutate(`Cr A` = `Cr A`/Mins) %>%
      mutate(`Cr C` = `Cr C`/Mins) %>%
      mutate(`xG` = `xG`/Mins) %>%
      mutate(`Off` = `Off`/Mins) %>%
      mutate(`K Hdrs` = `K Hdrs`/Mins) %>%
      # for some reason no attempted tackles valid data so using ratio and won to calculate it
      mutate(`Tck A` = (`Tck W`/`Tck R`)/`Mins`) %>%
      mutate(`Tck W` = `Tck W`/Mins) %>%
      mutate(`K Tck` = `K Tck`/Mins) %>%
      mutate(`Fls` = `Fls`/Mins) %>%
      mutate(`Red` = `Red`/Mins) %>%
      mutate(`Yel` = `Yel`/Mins) %>%
      mutate(`Cards` = `Red` + `Yel`) %>%
      mutate(`Gl Mst` = `Gl Mst`/Mins) #%>%
    # mutate(`` = ``/Mins)
  })
  
  #### Squad: General ####
  
  # link buttons with tabs
  observeEvent(input$sqgen_rating, {updateTabsetPanel(
    session, inputId = "sqgen_tabbox", selected = "panel1"
  )})
  observeEvent(input$sqgen_pointsw, {updateTabsetPanel(
    session, inputId = "sqgen_tabbox", selected = "panel2"
  )})
  observeEvent(input$sqgen_teamggoalass, {updateTabsetPanel(
    session, inputId = "sqgen_tabbox", selected = "panel3"
  )})
  observeEvent(input$sqgen_goalass, {updateTabsetPanel(
    session, inputId = "sqgen_tabbox", selected = "panel4"
  )})
  observeEvent(input$sqgen_cards, {updateTabsetPanel(
    session, inputId = "sqgen_tabbox", selected = "panel5"
  )})
  
  # generate plots
  output$sqgen_plot1 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Av Rat`,
      xlabel = "Games played",
      ylabel = "Average rating",
      targetratio = NA
    )
  })
  output$sqgen_plot2 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Pts/Gm`,
      xlabel = "Games played",
      ylabel = "Points won per game",
      targetratio = NA
    )
  })
  output$sqgen_plot3 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tgls/90`, y = data$`Gls plus Asts`,
      xlabel = "Team goals per 90min",
      ylabel = "Total goals + assists per 90min",
      targetratio = NA
    )
  })
  output$sqgen_plot4 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Gls/90`, y = data$`Asts/90`,
      xlabel = "Goals per 90min",
      ylabel = "Assists per 90min",
      targetratio = NA
    )
  })
  output$sqgen_plot5 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Yel`, y = data$`Red`,
      xlabel = "Yellow cards per 90min",
      ylabel = "Red cards per 90min",
      targetratio = NA
    )
  })
  
  #### Squad: Passing ####
  
  # link buttons with tabs
  observeEvent(input$sqpass_attcomp, {updateTabsetPanel(
    session, inputId = "sqpass_tabbox", selected = "panel1"
  )})
  observeEvent(input$sqpass_compkey, {updateTabsetPanel(
    session, inputId = "sqpass_tabbox", selected = "panel2"
  )})
  observeEvent(input$sqpass_compchan, {updateTabsetPanel(
    session, inputId = "sqpass_tabbox", selected = "panel3"
  )})
  observeEvent(input$sqpass_compass, {updateTabsetPanel(
    session, inputId = "sqpass_tabbox", selected = "panel4"
  )})
  observeEvent(input$sqpass_teamgass, {updateTabsetPanel(
    session, inputId = "sqpass_tabbox", selected = "panel5"
  )})
  observeEvent(input$sqpass_crossattcomp, {updateTabsetPanel(
    session, inputId = "sqpass_tabbox", selected = "panel6"
  )})
  
  # generate plots
  output$sqpass_plot1 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Ps A/90`, y = data$`Ps C/90`,
      xlabel = "Attempted passes per 90min",
      ylabel = "Completed passes per 90min",
      targetratio = 0.85
    )
  })
  output$sqpass_plot2 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Ps C/90`, y = data$`K Ps/90`,
      xlabel = "Completed passes per 90min",
      ylabel = "Key passes per 90min",
      targetratio = NA
    )
  })
  output$sqpass_plot3 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Ps C/90`, y = data$`Ch C/90`,
      xlabel = "Completed passes per 90min",
      ylabel = "Chances created per 90min",
      targetratio = NA
    )
  })
  output$sqpass_plot4 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Ps C/90`, y = data$`Asts/90`,
      xlabel = "Completed passes per 90min",
      ylabel = "Assists per 90min",
      targetratio = NA
    )
  })
  output$sqpass_plot5 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tgls/90`, y = data$`Asts/90`,
      xlabel = "Team goals per 90min",
      ylabel = "Assists per 90min",
      targetratio = NA
    )
  })
  output$sqpass_plot6 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Cr A`, y = data$`Cr C`,
      xlabel = "Attempted crosses per 90min",
      ylabel = "Completed crosses per 90min",
      targetratio = 0.2
    )
  })
  
  #### Squad: Shooting ####
  
  # link buttons with tabs
  observeEvent(input$sqshoot_atttarg, {updateTabsetPanel(
    session, inputId = "sqshoot_tabbox", selected = "panel1"
  )})
  observeEvent(input$sqshoot_targgoal, {updateTabsetPanel(
    session, inputId = "sqshoot_tabbox", selected = "panel2"
  )})
  observeEvent(input$sqshoot_teamggoal, {updateTabsetPanel(
    session, inputId = "sqshoot_tabbox", selected = "panel3"
  )})
  observeEvent(input$sqshoot_attxg, {updateTabsetPanel(
    session, inputId = "sqshoot_tabbox", selected = "panel4"
  )})
  observeEvent(input$sqshoot_xggoal, {updateTabsetPanel(
    session, inputId = "sqshoot_tabbox", selected = "panel5"
  )})
  
  # generate plots
  output$sqshoot_plot1 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Shot/90`, y = data$`ShT/90`,
      xlabel = "Attempted shots per 90min",
      ylabel = "On target shots per 90min",
      targetratio = 0.5
    )
  })
  output$sqshoot_plot2 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`ShT/90`, y = data$`Gls/90`,
      xlabel = "On target shots per 90min",
      ylabel = "Goals per 90min",
      targetratio = 0.3
    )
  })
  output$sqshoot_plot3 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tgls/90`, y = data$`Gls/90`,
      xlabel = "Team goals per 90min",
      ylabel = "Goals per 90min",
      targetratio = NA
    )
  })
  output$sqshoot_plot4 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Shot/90`, y = data$`xG`,
      xlabel = "Attempted shots per 90min",
      ylabel = "Average xG per 90min",
      targetratio = NA
    )
  })
  output$sqshoot_plot5 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`xG`, y = data$`Gls/90`,
      xlabel = "Average xG per 90min",
      ylabel = "Goals per 90min",
      targetratio = NA
    )
  })
  
  #### Squad: Movement ####
  
  # link buttons with tabs
  observeEvent(input$sqmove_drib, {updateTabsetPanel(
    session, inputId = "sqmove_tabbox", selected = "panel1"
  )})
  observeEvent(input$sqmove_offs, {updateTabsetPanel(
    session, inputId = "sqmove_tabbox", selected = "panel2"
  )})
  observeEvent(input$sqmove_dist, {updateTabsetPanel(
    session, inputId = "sqmove_tabbox", selected = "panel3"
  )})
  
  # generate plots
  output$sqmove_plot1 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`DrbPG`,
      xlabel = "Games played",
      ylabel = "Dribbles per 90min",
      targetratio = NA
    )
  })
  output$sqmove_plot2 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Off`,
      xlabel = "Games played",
      ylabel = "Offsides per 90min",
      targetratio = NA
    )
  })
  output$sqmove_plot3 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Dist/90`,
      xlabel = "Games played",
      ylabel = "Distance covered per 90min",
      targetratio = NA
    )
  })
  
  #### Squad: Aerial ####
  
  # link buttons with tabs
  observeEvent(input$sqaerial_attwon, {updateTabsetPanel(
    session, inputId = "sqaerial_tabbox", selected = "panel1"
  )})
  observeEvent(input$sqaerial_wonkey, {updateTabsetPanel(
    session, inputId = "sqaerial_tabbox", selected = "panel2"
  )})
  
  # generate plots
  output$sqaerial_plot1 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Aer A/90`, y = data$`Hdrs W/90`,
      xlabel = "Attempted aerial challenges per 90min",
      ylabel = "Won aerial challenges per 90min",
      targetratio = NA
    )
  })
  output$sqaerial_plot2 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Hdrs W/90`, y = data$`K Hdrs`,
      xlabel = "Won aerial challenges per 90min",
      ylabel = "Key aerial challenges per 90min",
      targetratio = NA
    )
  })
  
  #### Squad: Tackling ####
  
  # link buttons with tabs
  observeEvent(input$sqtackle_attcomp, {updateTabsetPanel(
    session, inputId = "sqtackle_tabbox", selected = "panel1"
  )})
  observeEvent(input$sqtackle_compkey, {updateTabsetPanel(
    session, inputId = "sqtackle_tabbox", selected = "panel2"
  )})
  observeEvent(input$sqtackle_attfoul, {updateTabsetPanel(
    session, inputId = "sqtackle_tabbox", selected = "panel3"
  )})
  observeEvent(input$sqtackle_attcard, {updateTabsetPanel(
    session, inputId = "sqtackle_tabbox", selected = "panel4"
  )})
  observeEvent(input$sqtackle_int, {updateTabsetPanel(
    session, inputId = "sqtackle_tabbox", selected = "panel5"
  )})
  observeEvent(input$sqtackle_mist, {updateTabsetPanel(
    session, inputId = "sqtackle_tabbox", selected = "panel6"
  )})
  
  # generate plots
  output$sqtackle_plot1 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tck A`, y = data$`Tck W`,
      xlabel = "Attempted tackles per 90min",
      ylabel = "Won tackles per 90min",
      targetratio = 0.6
    )
  })
  output$sqtackle_plot2 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tck W`, y = data$`K Tck`,
      xlabel = "Won tackles per 90min",
      ylabel = "Key tackles per 90min",
      targetratio = NA
    )
  })
  output$sqtackle_plot3 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tck A`, y = data$`Fls`,
      xlabel = "Attempted tackles per 90min",
      ylabel = "Fouls made per 90min",
      targetratio = NA
    )
  })
  output$sqtackle_plot4 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tck A`, y = data$`Cards`,
      xlabel = "Attempted tackles per 90min",
      ylabel = "Cards per 90min",
      targetratio = NA
    )
  })
  output$sqtackle_plot5 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Int/90`,
      xlabel = "Games played",
      ylabel = "Interceptions per 90min",
      targetratio = NA
    )
  })
  output$sqtackle_plot6 <- renderPlotly({
    data <- squad_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tcon/90`, y = data$`Gl Mst`,
      xlabel = "Team goals conceded per 90min",
      ylabel = "Mistakes leading to goal per 90min",
      targetratio = NA
    )
  })
  
  #### Squad: Progress ####
  
  # stats progress
  output$squad_progperf <- DT::renderDT({
    tabulate_progress(squads_DNArated, input$squad_startdate, input$squad_enddate, "stats")
  })
  # attributes progress
  output$squad_progattr <- DT::renderDT({
    tabulate_progress(squads_DNArated, input$squad_startdate, input$squad_enddate, "attr")
  })
  
  #### Scouting: Filters ####
  
  # filtered subset data for DT table presentation
  scout_filtsubset <- reactive({
    shortlist %>%
      select(Name, Rec, Position, PositionParsed, Division, Club, Age, Value, Wage, Mins) %>%
      # filter using the filters inputs
      dplyr::filter(Division %in% input$scout_div) %>%
      dplyr::filter(
        grepl_op(pattern_array = input$scout_pos, x = PositionParsed)
      ) %>%
      dplyr::filter(between(Age, input$scout_age[1], input$scout_age[2])) %>%
      dplyr::filter(Value <= input$scout_maxvalue) %>%
      dplyr::filter(Wage <= input$scout_maxwage) %>%
      dplyr::filter(Mins >= input$scout_minapps*90) %>%
      dplyr::select(-c(Division, PositionParsed)) %>%
      mutate(
        `Games Played` = round(Mins/90, digits = 1), .keep = "unused", .after = Mins
      ) %>%
      drop_na()
  })
  
  # filtered subset DT table
  output$scout_filtered <- DT::renderDT({
    tabulate_filtdata(data = scout_filtsubset(), dataset = "scouting")
  })
  
  # filtered subset names: used as an index
  scout_index <- reactive({
    scout_filtsubset() %>% as.data.frame() %>% dplyr::select(Name)# %>% as.vector()
  })
  
  # filtered subset DT table for plots
  scout_data <- reactive({
    subsetdata(
      data = shortlist, reactindex = as.vector(as.data.frame(scout_index())$Name),
      `Av Rat`, `Pts/Gm`, `Clean sheets`, # general
      `Ps A/90`, `Ps C/90`, `K Ps/90`, `Ch C/90`, `Asts/90`, `Cr A`, `Cr C`, # passing
      `Shot/90`, `ShT/90`, `Gls/90`, `Tgls/90`, `xG`, # shooting
      `DrbPG`, `Dist/90`, `Off`, # movement
      `Aer A/90`, `Hdrs W/90`, `K Hdrs`, # aerial
      `Tck R`, `Tck W`, `K Tck`, `Int/90`, `Fls`, `Red`, `Yel`, # tackling 1
      `Tcon/90`, `Gl Mst`#, # tackling 2
    ) %>%
      mutate(`Gls plus Asts` = `Gls/90` + `Asts/90`) %>%
      mutate(`Cr A` = `Cr A`/Mins) %>%
      mutate(`Cr C` = `Cr C`/Mins) %>%
      mutate(`xG` = `xG`/Mins) %>%
      mutate(`Off` = `Off`/Mins) %>%
      mutate(`K Hdrs` = `K Hdrs`/Mins) %>%
      # for some reason no attempted tackles valid data so using ratio and won to calculate it
      mutate(`Tck A` = (`Tck W`/`Tck R`)/`Mins`) %>%
      mutate(`Tck W` = `Tck W`/Mins) %>%
      mutate(`K Tck` = `K Tck`/Mins) %>%
      mutate(`Fls` = `Fls`/Mins) %>%
      mutate(`Red` = `Red`/Mins) %>%
      mutate(`Yel` = `Yel`/Mins) %>%
      mutate(`Cards` = `Red` + `Yel`) %>%
      mutate(`Gl Mst` = `Gl Mst`/Mins) #%>%
      # mutate(`` = ``/Mins)
  })
  
  #### Scouting: General ####
  
  # link buttons with tabs
  observeEvent(input$scgen_rating, {updateTabsetPanel(
    session, inputId = "scgen_tabbox", selected = "panel1"
  )})
  observeEvent(input$scgen_pointsw, {updateTabsetPanel(
    session, inputId = "scgen_tabbox", selected = "panel2"
  )})
  observeEvent(input$scgen_teamggoalass, {updateTabsetPanel(
    session, inputId = "scgen_tabbox", selected = "panel3"
  )})
  observeEvent(input$scgen_goalass, {updateTabsetPanel(
    session, inputId = "scgen_tabbox", selected = "panel4"
  )})
  observeEvent(input$scgen_cards, {updateTabsetPanel(
    session, inputId = "scgen_tabbox", selected = "panel5"
  )})
  
  # generate plots
  output$scgen_plot1 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Av Rat`,
      xlabel = "Games played",
      ylabel = "Average rating",
      targetratio = NA
    )
  })
  output$scgen_plot2 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Pts/Gm`,
      xlabel = "Games played",
      ylabel = "Points won per game",
      targetratio = NA
    )
  })
  output$scgen_plot3 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tgls/90`, y = data$`Gls plus Asts`,
      xlabel = "Team goals per 90min",
      ylabel = "Total goals + assists per 90min",
      targetratio = NA
    )
  })
  output$scgen_plot4 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Gls/90`, y = data$`Asts/90`,
      xlabel = "Goals per 90min",
      ylabel = "Assists per 90min",
      targetratio = NA
    )
  })
  output$scgen_plot5 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Yel`, y = data$`Red`,
      xlabel = "Yellow cards per 90min",
      ylabel = "Red cards per 90min",
      targetratio = NA
    )
  })
  
  #### Scouting: Passing ####
  
  # link buttons with tabs
  observeEvent(input$scpass_attcomp, {updateTabsetPanel(
    session, inputId = "scpass_tabbox", selected = "panel1"
  )})
  observeEvent(input$scpass_compkey, {updateTabsetPanel(
    session, inputId = "scpass_tabbox", selected = "panel2"
  )})
  observeEvent(input$scpass_compchan, {updateTabsetPanel(
    session, inputId = "scpass_tabbox", selected = "panel3"
  )})
  observeEvent(input$scpass_compass, {updateTabsetPanel(
    session, inputId = "scpass_tabbox", selected = "panel4"
  )})
  observeEvent(input$scpass_teamgass, {updateTabsetPanel(
    session, inputId = "scpass_tabbox", selected = "panel5"
  )})
  observeEvent(input$scpass_crossattcomp, {updateTabsetPanel(
    session, inputId = "scpass_tabbox", selected = "panel6"
  )})
  
  # generate plots
  output$scpass_plot1 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Ps A/90`, y = data$`Ps C/90`,
      xlabel = "Attempted passes per 90min",
      ylabel = "Completed passes per 90min",
      targetratio = 0.85
    )
  })
  output$scpass_plot2 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Ps C/90`, y = data$`K Ps/90`,
      xlabel = "Completed passes per 90min",
      ylabel = "Key passes per 90min",
      targetratio = NA
    )
  })
  output$scpass_plot3 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Ps C/90`, y = data$`Ch C/90`,
      xlabel = "Completed passes per 90min",
      ylabel = "Chances created per 90min",
      targetratio = NA
    )
  })
  output$scpass_plot4 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Ps C/90`, y = data$`Asts/90`,
      xlabel = "Completed passes per 90min",
      ylabel = "Assists per 90min",
      targetratio = NA
    )
  })
  output$scpass_plot5 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tgls/90`, y = data$`Asts/90`,
      xlabel = "Team goals per 90min",
      ylabel = "Assists per 90min",
      targetratio = NA
    )
  })
  output$scpass_plot6 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Cr A`, y = data$`Cr C`,
      xlabel = "Attempted crosses per 90min",
      ylabel = "Completed crosses per 90min",
      targetratio = 0.2
    )
  })
  
  #### Scouting: Shooting ####
  
  # link buttons with tabs
  observeEvent(input$scshoot_atttarg, {updateTabsetPanel(
    session, inputId = "scshoot_tabbox", selected = "panel1"
  )})
  observeEvent(input$scshoot_targgoal, {updateTabsetPanel(
    session, inputId = "scshoot_tabbox", selected = "panel2"
  )})
  observeEvent(input$scshoot_teamggoal, {updateTabsetPanel(
    session, inputId = "scshoot_tabbox", selected = "panel3"
  )})
  observeEvent(input$scshoot_attxg, {updateTabsetPanel(
    session, inputId = "scshoot_tabbox", selected = "panel4"
  )})
  observeEvent(input$scshoot_xggoal, {updateTabsetPanel(
    session, inputId = "scshoot_tabbox", selected = "panel5"
  )})
  
  # generate plots
  output$scshoot_plot1 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Shot/90`, y = data$`ShT/90`,
      xlabel = "Attempted shots per 90min",
      ylabel = "On target shots per 90min",
      targetratio = 0.5
    )
  })
  output$scshoot_plot2 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`ShT/90`, y = data$`Gls/90`,
      xlabel = "On target shots per 90min",
      ylabel = "Goals per 90min",
      targetratio = 0.3
    )
  })
  output$scshoot_plot3 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tgls/90`, y = data$`Gls/90`,
      xlabel = "Team goals per 90min",
      ylabel = "Goals per 90min",
      targetratio = NA
    )
  })
  output$scshoot_plot4 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Shot/90`, y = data$`xG`,
      xlabel = "Attempted shots per 90min",
      ylabel = "Average xG per 90min",
      targetratio = NA
    )
  })
  output$scshoot_plot5 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`xG`, y = data$`Gls/90`,
      xlabel = "Average xG per 90min",
      ylabel = "Goals per 90min",
      targetratio = NA
    )
  })
  
  #### Scouting: Movement ####
  
  # link buttons with tabs
  observeEvent(input$scmove_drib, {updateTabsetPanel(
    session, inputId = "scmove_tabbox", selected = "panel1"
  )})
  observeEvent(input$scmove_offs, {updateTabsetPanel(
    session, inputId = "scmove_tabbox", selected = "panel2"
  )})
  observeEvent(input$scmove_dist, {updateTabsetPanel(
    session, inputId = "scmove_tabbox", selected = "panel3"
  )})
  
  # generate plots
  output$scmove_plot1 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`DrbPG`,
      xlabel = "Games played",
      ylabel = "Dribbles per 90min",
      targetratio = NA
    )
  })
  output$scmove_plot2 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Off`,
      xlabel = "Games played",
      ylabel = "Offsides per 90min",
      targetratio = NA
    )
  })
  output$scmove_plot3 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Dist/90`,
      xlabel = "Games played",
      ylabel = "Distance covered per 90min",
      targetratio = NA
    )
  })
  
  #### Scouting: Aerial ####
  
  # link buttons with tabs
  observeEvent(input$scaerial_attwon, {updateTabsetPanel(
    session, inputId = "scaerial_tabbox", selected = "panel1"
  )})
  observeEvent(input$scaerial_wonkey, {updateTabsetPanel(
    session, inputId = "scaerial_tabbox", selected = "panel2"
  )})
  
  # generate plots
  output$scaerial_plot1 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Aer A/90`, y = data$`Hdrs W/90`,
      xlabel = "Attempted aerial challenges per 90min",
      ylabel = "Won aerial challenges per 90min",
      targetratio = NA
    )
  })
  output$scaerial_plot2 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Hdrs W/90`, y = data$`K Hdrs`,
      xlabel = "Won aerial challenges per 90min",
      ylabel = "Key aerial challenges per 90min",
      targetratio = NA
    )
  })
  
  #### Scouting: Tackling ####
  
  # link buttons with tabs
  observeEvent(input$sctackle_attcomp, {updateTabsetPanel(
    session, inputId = "sctackle_tabbox", selected = "panel1"
  )})
  observeEvent(input$sctackle_compkey, {updateTabsetPanel(
    session, inputId = "sctackle_tabbox", selected = "panel2"
  )})
  observeEvent(input$sctackle_attfoul, {updateTabsetPanel(
    session, inputId = "sctackle_tabbox", selected = "panel3"
  )})
  observeEvent(input$sctackle_attcard, {updateTabsetPanel(
    session, inputId = "sctackle_tabbox", selected = "panel4"
  )})
  observeEvent(input$sctackle_int, {updateTabsetPanel(
    session, inputId = "sctackle_tabbox", selected = "panel5"
  )})
  observeEvent(input$sctackle_mist, {updateTabsetPanel(
    session, inputId = "sctackle_tabbox", selected = "panel6"
  )})
  
  # generate plots
  output$sctackle_plot1 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tck A`, y = data$`Tck W`,
      xlabel = "Attempted tackles per 90min",
      ylabel = "Won tackles per 90min",
      targetratio = 0.6
    )
  })
  output$sctackle_plot2 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tck W`, y = data$`K Tck`,
      xlabel = "Won tackles per 90min",
      ylabel = "Key tackles per 90min",
      targetratio = NA
    )
  })
  output$sctackle_plot3 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tck A`, y = data$`Fls`,
      xlabel = "Attempted tackles per 90min",
      ylabel = "Fouls made per 90min",
      targetratio = NA
    )
  })
  output$sctackle_plot4 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tck A`, y = data$`Cards`,
      xlabel = "Attempted tackles per 90min",
      ylabel = "Cards per 90min",
      targetratio = NA
    )
  })
  output$sctackle_plot5 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Mins`, y = data$`Int/90`,
      xlabel = "Games played",
      ylabel = "Interceptions per 90min",
      targetratio = NA
    )
  })
  output$sctackle_plot6 <- renderPlotly({
    data <- scout_data() %>% as.data.frame()
    plot_it(
      data = data,
      x = data$`Tcon/90`, y = data$`Gl Mst`,
      xlabel = "Team goals conceded per 90min",
      ylabel = "Mistakes leading to goal per 90min",
      targetratio = NA
    )
  })
  
  #### Scouting: Shortlist ####
  
  # render final table
  output$scout_listfinal <- DT::renderDataTable({
    # intersection of all players chosen in the different scouting submenus
    listfinal <- Reduce(intersect, list(
      input$scgen_shortl,
      input$scpass_shortl,
      input$scshoot_shortl,
      input$scmove_shortl,
      input$scaerial_shortl,
      input$sctackle_shortl
    ))
    # filter filtered subset scouting data with only the final names
    listfinal_data <-  shortlist %>%
      select(
        Name, Rec, Position, Age,
        `Min AP`, `Max AP`, `Min WD`, `Max WD`,
        `Min Fee Rls`,
        `Injury Rls`, `Min Fee Rls Clubs In Cont Comp`, `Min Fee Rls Clubs Mjr Cont Comp`,
        `Min Fee Rls to Higher Div`, `Min Fee Rls to Domestic Clubs`, `Min Fee Rls to Foreign Clubs`,
        `Non Prom Rls Cls`, `Non-Playing Rel`, `Relegation Release`
      ) %>%
      mutate(Rls = if_else(is.na(`Min Fee Rls`), "No", "Yes"), .keep = "unused") %>%
      mutate(CondRls = if_else(
        is.na(`Injury Rls`) &&
          is.na(`Min Fee Rls Clubs In Cont Comp`) &&
          is.na(`Min Fee Rls Clubs Mjr Cont Comp`) &&
          is.na(`Min Fee Rls to Higher Div`) &&
          is.na(`Min Fee Rls to Domestic Clubs`) &&
          is.na(`Min Fee Rls to Foreign Clubs`) &&
          is.na(`Non Prom Rls Cls`) && 
          is.na(`Non-Playing Rel`) &&
          is.na(`Relegation Release`),
        "No", "Yes"), .keep = "unused"
      ) %>%
      dplyr::filter(Name %in% listfinal)
    
    # DT table
    DT::datatable(
      data = listfinal_data,
      # no row names displayed
      rownames = FALSE,
      # replace Rec column name by appropriate name: dataset
      # add "p/w" to wage column name
      colnames = c(
        "Scouting Rec" = "Rec",
        "Min WD (p/w)" = "Min WD", "Max WD (p/w)" = "Max WD",
        "Rls Cls?" = "Rls", "Conditional Rls Cls?" = "CondRls"
      ),
      # options
      options = list(
        # no paging, scroll instead
        paging = FALSE,
        # order
        order = list(
          list(1, "desc"), # rec
          list(5, "desc"), # max value
          list(4, "desc"), # min value
          list(7, "desc"), # max wage
          list(6, "desc"), # min wage
          list(3, "desc") # age
        ),
        # horizontal scrolling enabled
        scrollX = TRUE,
        # vertical scrolling enabled
        scrollY = "500px"
      )
    ) %>%
    # format font color and cell background depending on data
    formatStyle(
      # format all row based on Rec's value
      2, target = "row",
      # original colors
      # c("#bf3eff", "#ff0000", "#ffa500", "#ffff00", "#00ee00", "#008b00")
      backgroundColor = styleInterval(
        c(-1, 49, 59, 69, 84),
        c("#e5b1ff", "#ff9999", "#ffdb99", "#ffff99", "#92ff92", "#20ff20")
      ),
      color = styleInterval(
        c(-1, 49, 59, 69, 84),
        c("#6c00a2", "#990000", "#996300", "#666600", "#008100", "#004000")
      )
    ) %>%
    # format Value and Wage as currency
    formatCurrency(
      columns = 5:8,
      currency = "\u20ac ",
      interval = 3, mark = ",",
      digits = 2, dec.mark = ".",
      before = TRUE
    )
  })
  
}

# App data loaded...check line 458
toc()

# Run app -----------------------------------------------------------------

# status message
cat("\nApp will launch in 3...")
Sys.sleep(1)
cat("2...")
Sys.sleep(1)
cat("1...\n")
Sys.sleep(1)
# run without warnings
shinyApp(ui = ui, server = server, options = list(launch.browser = TRUE))

Edited February 27, 2021 by Kung
more info

FMFutbol · March 2, 2021

Looks great! I have done something similar but in python and in spanish for FM20 and now FM21.

I have some recommendations for you since it seems you use the plotly graphing library. Try styling the graphs with using both the size and color of the markers to provide more information.

These are two charts I like. Minutes played vs age. And Shots at goal vs xG vs Goals scored.

A little more information about what I have done is here:

http://medium.com/@footballmanager19

gam945 · March 2, 2021

5 hours ago, FMFutbol said:

Looks great! I have done something similar but in python and in spanish for FM20 and now FM21.

I have some recommendations for you since it seems you use the plotly graphing library. Try styling the graphs with using both the size and color of the markers to provide more information.

These are two charts I like. Minutes played vs age. And Shots at goal vs xG vs Goals scored.

A little more information about what I have done is here:

http://medium.com/@footballmanager19

It's been done since first release mate

Size is relative to the number of games played and the color is relative to the Scouting Recommandation (shortlist) & DNA Rating (squad). Also additional information is available when the point is hovered on.

I also considered Python, but for now Dash is far behind Shiny and I wanted to build an interactive tool, not something like a Jupyter notebook. And I didn't want to mess with HTML and CSS :lol: Anyway, great tool also mate

Edited March 2, 2021 by gam945

pow4h · March 8, 2021

Any way to filter out players that are in second team (U-23, U-18 etc) ?

gam945 · March 8, 2021

34 minutes ago, pow4h said:

Any way to filter out players that are in second team (U-23, U-18 etc) ?

I don't have much time to do updates, but it should be fairly simple.

The Yth/Full time data point is available, so it should be added in the filter box, then in the filters section (#### Squad (or scouting): Filters ####), add the previously added filter.

QP · April 5, 2021

Hi @gam945

I have to ask you something; do you think it is possible or easier to do add more comparisons to scouting tabs?

If It is easier can you share a how to so I can add my own comparisons to suit my own needs.

As far as I understand;

Something like this creates the scatter plots so;

EDIT:

Well I've added my own comparisons and somehow It worked;

Edited April 5, 2021 by QP

gam945 · April 5, 2021

10 hours ago, QP said:

Hi @gam945

I have to ask you something; do you think it is possible or easier to do add more comparisons to scouting tabs?

If It is easier can you share a how to so I can add my own comparisons to suit my own needs.

As far as I understand;

Something like this creates the scatter plots so;

EDIT:

Well I've added my own comparisons and somehow It worked;

Hi mate, just saw your post. Great that you figured it out yourself. By curiosity, do you have programming (R?) experience/knowledge or you figured it out from nothing?

As you surely understood it, you have to add the button in Visualization Selection, add a tab in Visualizations and link it to the button, add an output (PlotlyOutput) in the tab, then in the server side, add a render (renderPlotly iirc). So IMO pretty easy to do, and the code to add is exactly the same as the one included, the only thing to change is the statistics and the various plot labels.

QP · April 6, 2021

10 hours ago, gam945 said:

Hi mate, just saw your post. Great that you figured it out yourself. By curiosity, do you have programming (R?) experience/knowledge or you figured it out from nothing?

As you surely understood it, you have to add the button in Visualization Selection, add a tab in Visualizations and link it to the button, add an output (PlotlyOutput) in the tab, then in the server side, add a render (renderPlotly iirc). So IMO pretty easy to do, and the code to add is exactly the same as the one included, the only thing to change is the statistics and the various plot labels.

I've known nothing about programming but I got the concept when I look at codes in graphics sections then repeated the same code with changing numbers and adding my own elements to comparison.

It is actually very easy to customise because of your very detailed coding, It really helps to make trai and error.

Edited April 6, 2021 by QP

NFLFM · April 12, 2021

Hello

Firstly thanks for your huge work, I really appreciate it !

However I have a concern when I tried to put many squad files it never works do you know the reasons and what to put inside the folder ?

I mean what I gotta write, I know it gotta be years month day but I don't know what to write for the other teams I tried to rename with a 2 or (2), I tried to replace the same file, I tried to create sub folder by team and put in each the squad files, I tried to advance ingame and create 2 files for each date but nothing happened it either doesn't work or always keep the last squad I saved

And we agree that for save multiple squads, we gotta go in each page by team and select players - senior squad, obviously keep the custom view then save ?

Would be nice to find a solution because I don't understand what I should do, sorry for my english btw.

Edited April 12, 2021 by NFLFM

NFLFM · April 13, 2021

Anyone who can help please ? Would be nice

gam945 · April 13, 2021

On 11/04/2021 at 21:42, NFLFM said:

Hello

Firstly thanks for your huge work, I really appreciate it !

However I have a concern when I tried to put many squad files it never works do you know the reasons and what to put inside the folder ?

I mean what I gotta write, I know it gotta be years month day but I don't know what to write for the other teams I tried to rename with a 2 or (2), I tried to replace the same file, I tried to create sub folder by team and put in each the squad files, I tried to advance ingame and create 2 files for each date but nothing happened it either doesn't work or always keep the last squad I saved

And we agree that for save multiple squads, we gotta go in each page by team and select players - senior squad, obviously keep the custom view then save ?

Would be nice to find a solution because I don't understand what I should do, sorry for my english btw.

You only have two type of data that the app can read: squad_<date> (you can have multiple snapshots of the squad at unique & different dates) which is for YOUR squad, and shortlist which covers external players.

If you want to analyse other teams, add their squad to your shortlist. The squads files are only for YOUR squad.

Edited April 13, 2021 by gam945

NFLFM · April 13, 2021

4 hours ago, gam945 said:

You only have two type of data that the app can read: squad_<date> (you can have multiple snapshots of the squad at unique & different dates) which is for YOUR squad, and shortlist which covers external players.

If you want to analyse other teams, add their squad to your shortlist. The squads files are only for YOUR squad.

Ok I understand thanks for your answer but why on your screenshot and video I saw 23 items/many squads on squad analysis ?

And if im correct you said that with the V3 it’s now possible to add many squad

Edited April 13, 2021 by NFLFM

NFLFM · April 13, 2021

On 01/01/2021 at 00:38, gam945 said:

* It is possible thanks to the above feature to store multiple squad data files. Keep in mind that the more there is squad files in your data folder, the longer the app will take to load. You can find out how much time the different tasks (load data & format data for example) take time by paying attention to the command prompt that is launched when you double-click on FMapp.bat.

Im taking about this and the screenshot above

NFLFM · April 14, 2021

Sorry to bother but I also realize that we can't put our own players in the shortlist so it means that it's impossible to compare for example with all the squads of the championship

Edited April 14, 2021 by NFLFM

Sign In

FM Data Analytics: a data visualization tool based on R & Shiny - V3 RELEASED!

Recommended Posts

Link to post

Share on other sites

Top Posters In This Topic

Top Posters In This Topic

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites

Link to post

Share on other sites