Skip to content

Latest commit

 

History

History
609 lines (486 loc) · 21.4 KB

README.md

File metadata and controls

609 lines (486 loc) · 21.4 KB

ganttrify

ganttrify status badge

ganttrify facilitates the creation of nice-looking Gantt charts, commonly used in project proposals and project management.

If you just want to check this out in an interactive web interface, click here and enjoy. Some more context in this blog post.

Read on for more details and examples.

Motivation

It is possible to find online documented attempts at facilitating the creation of Gantt charts from R. Some of them (e.g. this and this) use ‘ggplot2’, but I feel they do not look very nice. The same goes for the answers I found in the relevant Stackoverflow question.

Even Plotly enables the creation of Gantt charts in R, but again, I don’t like the end result.

I did find a solution that was rather visually satisfying, but it was in base R, and all the cool kids nowadays know that base plotting in R exists only for compatibility with S: not an option! (Hey, I’m joking, don’t @ me!)

Given what is evidently my posh taste for Gantt charts, I had no other option than making this package with a pretentious, gentrified name, instead of the obvious “ganttr”.

Please welcome ganttrify.

Disclaimer

More seriously, this has been a quick attempt at making decent-looking Gantt charts.

And yes, I will enable all the customisations you like, but first I actually need to submit this project.

[Thanks to all who contributed suggestions via issues and pull requests!]

[This package has been developed as a personal endeavour, originally conceived for grant writing at my employer (OBCT/CCI)]. The Shiny interface, now hosted in a separate repository, has been developed for EDJNet, the European Data Journalism Network.

Features

Take an adequately formatted spreadsheet and turn it into a Gantt chart made with ggplot2.

Installation

You can install the development version from GitHub with:

# install.packages("remotes")
remotes::install_github("giocomai/ganttrify")

Or from R-universe:

install.packages("ganttrify",
  repos = c(
    "https://giocomai.r-universe.dev",
    "https://cloud.r-project.org"
  )
)

Example

Here is an example project:

wp activity start_date end_date
WP1 - Whatever admin 1.1. That admin activity 1 6
WP1 - Whatever admin 1.2. Another admin activity 3 6
WP1 - Whatever admin 1.3. Fancy admin activity 4 7
WP2 - Whatever actual work 2.1. Actual stuff 5 10
WP2 - Whatever actual work 2.2. Actual R&D stuff 6 12
WP2 - Whatever actual work 2.3. Really real research 9 12
WP2 - Whatever actual work 2.4. Ethics! 3 5
WP2 - Whatever actual work 2.4. Ethics! 8 9
WP3 - Dissemination 3.1. Disseminate near 6 9
WP3 - Dissemination 3.1. Disseminate near 12 12
WP3 - Dissemination 3.2. Disseminate far 8 12

Month since the beginning of the project are used as reference in order to make it easier to change the date when the project starts without needing to change the timing of all activities.

If you prefer to include dates instead of month numbers, please see additional examples below.

library("ganttrify")

ganttrify(
  project = ganttrify::test_project,
  project_start_date = "2021-03",
  font_family = "Roboto Condensed"
)

[all examples in this page will use the Roboto condensed font; if it is not installed, you can use the default sans. See the Troubleshooting section at the bottom of this readme.]

“But what if I wanted to add spot labels for events, deliverables, outputs, milestones, things like that?”, you asked.

Just put them in a table with these column names, and you will be served.

activity spot_type spot_date
1.1. That admin activity D 5
1.3. Fancy admin activity E 7
2.2. Actual R&D stuff O 7
2.2. Actual R&D stuff O 9
2.2. Actual R&D stuff O 11
WP2 - Whatever actual work M 6
ganttrify(
  project = ganttrify::test_project,
  spots = ganttrify::test_spots,
  project_start_date = "2021-03",
  font_family = "Roboto Condensed"
)

“I can’t read the text, can I change the text size?”, I heard. “Also, is it possible to outline quarters?”

You’re welcome.

ganttrify(
  project = ganttrify::test_project,
  spots = ganttrify::test_spots,
  project_start_date = "2021-03",
  size_text_relative = 1.2,
  mark_quarters = TRUE,
  font_family = "Roboto Condensed"
)

It appears that some of you don’t like having a line for the working package and are fine with just lines for activities. Did you mean it like this? (also, consider setting hide_activities to TRUE if on the contrary you want to hide activities and keep only working packages)

ganttrify(
  project = ganttrify::test_project,
  hide_wp = TRUE,
  font_family = "Roboto Condensed"
)

Or perhaps, you did want to keep the name of working package on the left, but just felt that the horizontal line for the WP is redundant? Let’s make the WP lines invisible, and nobody will ever know they’re there!

ganttrify(
  project = ganttrify::test_project,
  alpha_wp = 0,
  font_family = "Roboto Condensed"
)

I felt that rounded line endings for the working packages, and “butt” ending for activities is the best combination of elegance and clarity.

Also, I like full opacity for the lines, but it’s ok if you don’t:

ganttrify(
  project = ganttrify::test_project,
  project_start_date = "2021-04",
  alpha_wp = 0.9,
  alpha_activity = 0.6,
  line_end_wp = "round", # alternative values: "butt" or "square"
  line_end_activity = "round", # alternative values: "butt" or "square"
  font_family = "Roboto Condensed"
)

If you use spot events, then there’s all sorts of opinions you can have about the color and transparency of spot events, as well as the size and padding around the text. Say, you want a larger box around the text, but a semi-transparent background? There you go!
ganttrify(
  project = ganttrify::test_project,
  spots = ganttrify::test_spots,
  project_start_date = "2021-04",
  font_family = "Roboto Condensed",
  spot_size_text_relative = 1.5,
  spot_fill = ggplot2::alpha(c("white"), 0.7),
  spot_padding = ggplot2::unit(0.4, "lines")
)

Or perhaps you actually just want the label text, without any background, or perhaps change the text colour. Be mindful that the text may become hard to read on darker backgrounds, especially if reviewers then print your proposal, but… you have been warned:
ganttrify(
  project = ganttrify::test_project,
  spots = ganttrify::test_spots,
  project_start_date = "2021-04",
  font_family = "Roboto Condensed",
  spot_text_colour = "grey10",
  spot_fontface = "bold",
  spot_fill = NA,
  spot_border = NA
)

Some of us work on very long projects, and may need to declutter the chart to increase readability. So let’s show the month number only once every three months, and hide the thin vertical lines included by default.

test_36 <- ganttrify::test_project
test_36[11, 4] <- 36

ganttrify(
  project = test_36,
  project_start_date = "2021-04",
  month_breaks = 3,
  show_vertical_lines = FALSE,
  font_family = "Roboto Condensed"
)

If you have many working packages, you may want to adjust the size of the output, and choose palettes with more colours. You can always pass a custom palette (e.g. by setting colour_palette = c("#6ACCEA", "#00FFB8", "#B90000", "#6C919C")), but if you’re looking for some inspiration, the package MetBrewer with palettes inspired by works at the Metropolitan Museum of Art in New York has some options that may make your gantt chart even fancier.

test_project_df <- dplyr::bind_rows(
  ganttrify::test_project,
  tibble::tibble(
    wp = ganttrify::test_project$wp %>% stringr::str_replace(pattern = "1", replacement = "4") %>% stringr::str_replace(pattern = "2", replacement = "5") %>% stringr::str_replace(pattern = "3", replacement = "6"),
    activity = ganttrify::test_project$activity %>% stringr::str_replace(pattern = "^1", replacement = "4") %>% stringr::str_replace(pattern = "^2", replacement = "5") %>% stringr::str_replace(pattern = "^3", replacement = "6"),
    start_date = test_project$start_date + 12,
    end_date = test_project$end_date + 12
  )
)


ganttrify(
  project = test_project_df,
  size_text_relative = 1.2,
  month_breaks = 2,
  project_start_date = "2023-01",
  font_family = "Roboto Condensed",
  colour_palette = MetBrewer::met.brewer("Lakota")
) # or e.g. colour_palette = c("#6ACCEA", "#00FFB8", "#B90000", "#6C919C")
#> Registered S3 method overwritten by 'MetBrewer':
#>   method        from       
#>   print.palette wesanderson

Does right-aligned text bother you?

ganttrify(
  project = ganttrify::test_project,
  spots = ganttrify::test_spots,
  project_start_date = "2021-04",
  axis_text_align = "left",
  font_family = "Roboto Condensed"
)

ganttrify(
  project = ganttrify::test_project,
  spots = ganttrify::test_spots,
  project_start_date = "2021-04",
  axis_text_align = "centre",
  font_family = "Roboto Condensed"
)

Do you have very long names for your activities? The parameter label_wrap is there to help you.

tibble::tribble(
  ~wp, ~activity, ~start_date, ~end_date,
  "WP 1", "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Aliquet eget sit amet tellus", 1, 6,
  "WP 2", "Proin sed libero enim sed faucibus turpis in eu mi. Massa ultricies mi quis hendrerit dolor magna eget est.", 6, 7,
  "WP 2", "Proin sed libero enim sed faucibus turpis in eu mi. Massa ultricies mi quis hendrerit dolor magna eget est.", 3, 4,
  "WP 2", "Neque laoreet suspendisse interdum consectetur libero id faucibus nisl tincidunt.", 4, 8
) %>%
  ganttrify(
    label_wrap = 32,
    project_start_date = "2023-01",
    font_family = "Roboto Condensed"
  )

Finally, keep in mind that ganttrify outputs ggplot objects. Some theming options may not behave exactly as you expect, but for example adding title, subtitle, and captions can be done as you would normally do with any ggplot2 graph.

ganttrify(
  project = ganttrify::test_project,
  spots = ganttrify::test_spots,
  project_start_date = "2020-01",
  font_family = "Roboto Condensed"
) +
  ggplot2::labs(
    title = "My beautiful plans for 2020",
    subtitle = "I will definitely comply with the exact timing of each and all activities*",
    caption = "* I mean, I'll do my best, but if there's a pandemic or something, it's not my fault really"
  )

And since it’s a ggplot2 object, you can export it as you would any other ggplot graph, by using ggplot2::ggsave. This allows also to customise the size of the final input, running e.g. ggplot2::ggsave(filename = "my_gantt.png", width = 12, height = 8, bg = "white") after you run the ganttrify function (if you set the file extension to pdf or svg, you will get the chart in vector format for more clarity at different zoom levels).

If you are using this in an rmarkdown document, keep in mind that you can set the size at the chunk level, e.g. with something like {r fig.width=12, fig.height=8} in the chunk header.

Markdown, html, images… it’s all there

Now, to be honest, nobody asked for this. And I’m not even sure it’s a good idea. But if you are interested in more customisations of how the wp and activity labels appear, anything supported by ggtext should work here: markdown, basic html tags, and even inline images.

funky_project <- tibble::tribble(
  ~wp, ~activity, ~start_date, ~end_date,
  "<span style = 'color:red;'>Red</span> flavour", "Considering <sup>upper</sup> styles", 1, 6,
  "<span style = 'color:red;'>Red</span> flavour", "Or **bold**, or *italic*", 3, 6,
  "Don't forget <span style = 'font-size:6pt'>the small things</span>", "Contribute to Wikidata <img src='https://upload.wikimedia.org/wikipedia/commons/thumb/f/ff/Wikidata-logo.svg/320px-Wikidata-logo.svg.png' width=20>", 5, 10,
  "Don't forget <span style = 'font-size:6pt'>the small things</span>", "And to OpenStreetMap <img src='https://upload.wikimedia.org/wikipedia/commons/thumb/b/b0/Openstreetmap_logo.svg/256px-Openstreetmap_logo.svg.png' width=20>", 7, 12
)

ganttrify(
  project = funky_project,
  project_start_date = "2024-01",
  font_family = "Roboto Condensed"
) +
  ggplot2::ggtitle("Custom activity labels, nothing else to see here")

Shiny app

If you prefer interactive web interfaces to coding, you can still have a fancy ganttrified chart. In order to reduce dependencies and facilitate checks, this is now available in a separate package, shinyganttrify.

Step 1: Make sure you have installed both ganttrify and shinyganttrify

remotes::install_github("giocomai/ganttrify")
remotes::install_github("giocomai/shinyganttrify")

Then run:

shinyganttrify::shiny_ganttrify()

You can check it online with no further ado at the following link:

https://ganttrify.europeandatajournalism.eu/

(N.B.: not all features are exposed in the Shiny app)

Shiny app on Docker

Alright, you don’t know like R, but you know how Docker works?

This is all you need to find yourself a nice web app on localhost

docker run -p 80:80 giocomai/ganttrify

You can of course build yourself the docker image using the Dockerfile included in this repo.

Additional input formats

Alright, you prefer to use dates rather than month numbers from the beginning of the project. You’re welcome: just format the date as follows, and remember to include the month_number_label = FALSE parameter. You can also use exact dates (e.g. 2021-01-01), but by default they would still be converted to include the entire month were that given day falls.

knitr::kable(ganttrify::test_project_date_month)
wp activity start_date end_date
WP1 - Whatever admin 1.1. That admin activity 2021-01 2021-06
WP1 - Whatever admin 1.2. Another admin activity 2021-03 2021-06
WP1 - Whatever admin 1.3. Fancy admin activity 2021-04 2021-07
WP2 - Whatever actual work 2.1. Actual stuff 2021-05 2021-10
WP2 - Whatever actual work 2.2. Actual R&D stuff 2021-06 2021-12
WP2 - Whatever actual work 2.3. Really real research 2021-09 2021-12
WP2 - Whatever actual work 2.4. Ethics! 2021-03 2021-05
WP2 - Whatever actual work 2.4. Ethics! 2021-08 2021-09
WP3 - Dissemination 3.1. Disseminate near 2021-06 2021-09
WP3 - Dissemination 3.1. Disseminate near 2021-12 2021-12
WP3 - Dissemination 3.2. Disseminate far 2021-08 2021-12
ganttrify(
  project = ganttrify::test_project_date_month,
  spots = ganttrify::test_spots_date_month,
  by_date = TRUE,
  size_text_relative = 1.2,
  mark_quarters = TRUE,
  font_family = "Roboto Condensed"
)

As it turns out, someone wants more detail: they’d like to be able to input activities with an exact start and end date. I start to suspect that ganttrify at this stage may not be exactly what you’re looking for, but perhaps this works for you?

knitr::kable(ganttrify::test_project_date_day)
wp activity start_date end_date
Data team Data collection 2020-09-01 2020-09-10
Data team Data processing 2020-09-08 2020-09-14
Data team Reporting 2020-09-14 2020-09-16
Data team Data visualisation 2020-10-23 2020-10-30
Investigative team Fieldwork 2020-09-05 2020-09-15
Investigative team Fieldwork 2020-10-10 2020-10-20
Investigative team Writing 2020-10-21 2020-10-31
Social media team Draft outputs 2020-10-25 2020-10-28
Social media team Active promo 2020-10-31 2020-12-15
ganttrify(
  project = ganttrify::test_project_date_day,
  spots = ganttrify::test_spots_date_day,
  by_date = TRUE,
  exact_date = TRUE,
  size_text_relative = 1.2,
  month_number_label = FALSE,
  font_family = "Roboto Condensed"
)

Troubleshooting

Structure of the input table

At this stage, the package has strong expectations about the input format, and does not provide meaningful error messages. If you see unexpected results, please consider that ideally:

  • no cell in the activity column must be empty
  • an activity cannot be called the same as a wp
  • activities in different wp should have different names (or at least add a space at the end or something so that they look different to the computer).

Some of these limitations have been mitigated in recent versions. For example, it is now possible to have activities with the same name in different working packages, e.g.:

activities_with_same_name_in_different_wp_project <- tibble::tribble(
  ~wp, ~activity, ~start_date, ~end_date,
  "WP1", "Admin", 1, 6,
  "WP1", "Research", 3, 6,
  "WP1", "Dissemination", 4, 7,
  "WP2", "Admin", 5, 10,
  "WP2", "Research", 6, 12,
  "WP2", "Research", 9, 12,
  "WP2", "Dissemination", 3, 5,
  "WP2", "Dissemination", 8, 9,
  "WP3", "Admin", 6, 9,
  "WP3", "Admin", 12, 12
)

ganttrify(
  project = activities_with_same_name_in_different_wp_project,
  project_start_date = "2024-01",
  font_family = "Roboto Condensed"
)

But be mindful that this may lead to unexpected results if you’re using spots; since their placement is currently based exclusively on the activity name, possible long term fixes would inevitably involve a change in the input format. An easy (if inelegant) workaround is simply to add one or more spaces at the end of activities that need to be differentiated: e.g.to differentiate an activity named admin and present in both WP1 and WP2 you could name it “Admin” in WP1, but “Admin” in WP2 (notice the additional space in the second case).

Fonts

By default, this package uses a generic sans font but it is recommended to use a narrow (or condensed font such as Roboto Condensed font - a free font that can be downloaded and installed on any desktop) as they make more efficient use of text space.

On Fedora, you can install it with sudo dnf install google-roboto-condensed-fonts

On Debian, you can install it with sudo apt-get install fonts-roboto-fontface

After installation, you should make sure the font is available to R by installing the extrafont package, and running extrafont::font_import().

You can check available fonts also with the package systemfonts and the command systemfonts::system_fonts().