Formatting Java Time with Spring Boot using JSON

The aim of this post is to summarize and review ways of formatting Java Time objects using Spring Boot and Jackson library.

This post is organized in five steps. Each step represents one aspect of the issue and it is also related to one commit in example project repository.

Step 0 - Prerequirements

Versions and dependencies

This tutorial is based on Spring Boot version 1.3.1.RELEASE with spring-boot-starter-web. It uses jackson-datatype-jsr310 from com.fasterxml.jackson.datatype in version 2.6.4, which is a default version of Spring Boot. All of these is based on Java 8.

The Code

In the example code repository, you can find one HTTP service made with Spring Boot. This service is a GET operation, which returns a class with Java Time objects.
You can also find the integration test that deserializes the response.

Step 1 - The goal

I would like to return class Clock, containing LocalDate,LocalTime and LocalDateTime, preinitialized in constructor.

Clock - Service response class
1
2
3
4
5
6
public final class Clock {
private final LocalDate localDate;
private final LocalTime localTime;
private final LocalDateTime localDateTime;
...
}

Response class is serialized to JSON Map, which is a default behaviour. To some extent it is correct, but ISO formatted Strings in response are preferable.

LocalDate - response as JSON Map
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
{  
"localDate":{
"year":2016,
"month":"JANUARY",
"era":"CE",
"dayOfYear":1,
"dayOfWeek":"FRIDAY",
"leapYear":true,
"dayOfMonth":1,
"monthValue":1,
"chronology":{
"id":"ISO",
"calendarType":"iso8601"
}

}

}

Integration testing is an appropriate way to test our functionality.

Example of integration test
1
2
3
4
5
6
ResponseEntity<Clock> resp = sut.getForEntity("http://localhost:8080/clock", Clock.class);

assertEquals(OK, resp.getStatusCode());
assertEquals(c.getLocalDate(), resp.getBody().getLocalDate());
assertEquals(c.getLocalTime(), resp.getBody().getLocalTime());
assertEquals(c.getLocalDateTime(), resp.getBody().getLocalDateTime());

Unfortunately, tests are not passing, because of deserialization problems. The exception with message is thrown can not instantiate from JSON object.

Step 2 - Adds serialization

First things first. We have to add JSR-310 module. It is a datatype module to make Jackson recognize Java 8 Date & Time API data types.

Note that in this example jackson-datatype-jsr310 version is inherited from spring-boot-dependencies dependency management.

Dependency in pom.xml
1
2
3
4
<dependency>
<groupId>com.fasterxml.jackson.datatype</groupId>
<artifactId>jackson-datatype-jsr310</artifactId>
</dependency>

Response is now consistent but still, not perfect. Dates are serialized as numbers:
Dates serialized to numbers and integers
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
{  
"version":2,
"localDate":[
2016,
1,
1
],

"localTime":[
10,
24
],

"localDateTime":[
2016,
1,
1,
10,
24
],

"zonedDateTime":1451640240.000000000
}


We are one step closer to our goal. Tests are passing now because this format can deserialized without any additional deserializers.
How do I know?
Start an application server on commit Step 2 - Adds Object Mapper, then checkout to Step 1 - Introduce types and problems, and run integration tests without @WebIntegrationTest annotation.

Step 3 - Enables ISO formatting

ISO 8601 formatting is a standard. I’ve found it in many projects. We are going to enable and use it.
Edit spring boot properties file application.properties and add the following line:

application.properties file - disabling timestamps write
1
spring.jackson.serialization.WRITE_DATES_AS_TIMESTAMPS = false

Now, the response is something that I’ve expected:

Dates serialized as Strings
1
2
3
4
5
6
7
{  
"version":2,
"localDate":"2016-01-01",
"localTime":"10:24",
"localDateTime":"2016-01-01T10:24",
"zonedDateTime":"2016-01-01T10:24:00+01:00"
}

Step 4 - Adds on demand formatting pattern

Imagine one of your client systems does not have a capability of formatting time. It may be a primitive device, or microservice that treats this date as a collection of characters. That is why special formatting is required.

We can change formatting in response class by adding JsonFormat annotation with pattern parameter. Standard SimpleDateFormat rules apply.

Using @JsonFormat annotation
1
2
3
4
5
@JsonFormat(pattern = "dd::MM::yyyy")
private final LocalDate localDate;

@JsonFormat(pattern = "KK:mm a")
private final LocalTime localTime;

Below there is a service response using custom @JsonFormat pattern:

Custom response style
1
2
3
4
5
6
7
{  
"version":2,
"localDate":"01::01::2016",
"localTime":"10:24 AM",
"localDateTime":"2016-01-01T10:24",
"zonedDateTime":"2016-01-01T10:24:00+01:00"
}


Our tests are still passing. It means that this pattern is used for serialization in service and deserialization in tests.

Step 5 - Globally changes formatting

There are situations where you have to resign from ISO 8601 formatting in your whole application, and apply custom made standards.

In this part, we will redefine format pattern for LocalDate. This will change formatting of LocalDate in every endpoint of your API.

We have to define:

  • DateTimeFormatter with our pattern.
  • Serializer using defined pattern.
  • Deserializer using defined pattern.
  • ObjectMapper bean with custom serializer and deserializer.
  • RestTemplate that uses our ObjectMapper.

Bean ObjectMapper is defined with annotation @Primary, to override default configuration.
My custom pattern for LocalDate is dd::MM::yyyy

Object mapper bean with custom pattern
1
2
3
4
5
6
7
8
9
10
11
12
public static final DateTimeFormatter FORMATTER = ofPattern("dd::MM::yyyy");

@Bean
@Primary
public ObjectMapper serializingObjectMapper() {
ObjectMapper objectMapper = new ObjectMapper();
JavaTimeModule javaTimeModule = new JavaTimeModule();
javaTimeModule.addSerializer(LocalDate.class, new LocalDateSerializer());
javaTimeModule.addDeserializer(LocalDate.class, new LocalDateDeserializer());
objectMapper.registerModule(javaTimeModule);
return objectMapper;
}

Definitions of serializer and deserializer for all LocalDate classes:

Custom serializer and deserializer
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
public class LocalDateSerializer extends JsonSerializer<LocalDate> {

@Override
public void serialize(LocalDate value, JsonGenerator gen, SerializerProvider serializers) throws IOException {
gen.writeString(value.format(FORMATTER));
}
}

public class LocalDateDeserializer extends JsonDeserializer<LocalDate> {

@Override
public LocalDate deserialize(JsonParser p, DeserializationContext ctxt) throws IOException {
return LocalDate.parse(p.getValueAsString(), FORMATTER);
}
}

Now, the response is formatted with our custom pattern:
Formatted response
1
2
3
{  
"localDate":"01::01::2016"
}

Tests

When we define custom serializer, our tests start to fail. It is because RestTemplate knows nothing about our deserializer. We have to create custom RestTemplateFactory that creates RestTemplate with object mapper containing our deserializer.

Custom RestTemplateFactory
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
@Configuration
public class RestTemplateFactory {

@Autowired
private ObjectMapper objectMapper;

@Bean
public RestTemplate createRestTemplate() {
RestTemplate restTemplate = new RestTemplate();
List<HttpMessageConverter<?>> converters = new ArrayList<>();
MappingJackson2HttpMessageConverter jsonConverter = new MappingJackson2HttpMessageConverter();
jsonConverter.setObjectMapper(objectMapper);
converters.add(jsonConverter);
restTemplate.setMessageConverters(converters);
return restTemplate;
}
}

Conclusion

Custom formatting Dates is relatively simple, but you have to know how to set up it. Luckily, Jackson works smoothly with Spring. If you know other ways of solving this problem or you have other observations, please comment or let me know.

Photo credits: Banner, Thumbnail

Clojure - Fascinated, Disappointed, Astonished

I’ve had a pleasure to work with Piotrek Jagielski for about two weeks on Clojure project. I’ve learned a lot, but there is still a lot to know about Clojure for me. In this post I’ll write what fascinated, disappointed and astonished me about this programming language.

Clojure & InteliJ IDEA tips

Before you start your journey with Clojure:

  • Use Cursive plugin for InteliJ IDEA. In ‘14 Edition it was not in the standard plug-in repository (remove La Clojure plug-in and Cursive repository manually). For IDEA ‘15 it is in repository.
  • Colored brackets help me a lot. You can find configuration for colored brackets on Misophistful Github.

Fascinated

Syntax

For many people Clojure brackets are reasons to laugh. Jokes like that were funny at first: “How many brackets did you write today?”
I have to admit, that at the beginning using brackets was not easy for me. Once I’ve realized that the brackets are just on the other side of the function name, everything was simple and I could code very fast.
After few days I’ve realized that this brackets structure forces me to think more about the structure of the code. As a result the code is refactored and divided into small functions.
Clojure forces you to use good programming habits.

Data structure is your code

Clojure is homoiconic, which means that the Clojure programs are represented by Clojure data structures. This means that when you are reading a Clojure code you see lists, maps, vectors. How cool is that! You only have to know few things and you can code.

Do not restart your JVM

Because Clojure code is represented as data structures, you can pass data structure (program) to running JVM. Furthermore, compiling your code to bytecode (classes, jars) may be eliminated.

For example, when you want to test something you are not obligated to start new JVM with tests. Instead you can just synchronize your working file with running REPL and run the function.

Traditional way of working with JVM is obsolete.

REPL code synchronization

In the picture above, on the left you can see an editor, on the right there is running REPL.

The same way you can run tests, which is extremely fast. In our project we had ~80 tests. Executing them all took about one second.

Easy to read

Simplicity is the ultimate sophistication.

Leonardo da Vinci

After getting familiar with this language, it was really easy to read code. Of course, I was not aware of everything what was happening under the hood, but consistency of the written program evoked sense of control.

Disapointed

Data structure is your code

When data structure is your code, you need to have some additional operators to write effective programs. You should get to know operators like ‘->>’, ‘->’, ‘let’, ‘letfn’, ‘do’, ‘if’, ‘recur’ …

Even if there is a good documentation (e.g. Let), you have to spend some time on analyzing it, and trying out examples.

As the time goes on, new operators will be developed. But it may lead to multiple Clojure dialects. I can imagine teams (in the same company) using different sets of operators, dealing with the same problems in different ways. It is not good to have too many tools. Nevertheless, this is just my suspicion.

Know what you do

I’ve written a function that rounds numbers. Despite the fact that this function was simple, I wanted to write test, because I was not sure if I had used the API in correct way. There is the test function below:

Rounding test
1
2
(let [result (fixture/round 8.211M)]
(is (= 8.21M result))))

Unfortunately, tests were not passing. This is the only message that I received:
Rounding test
1
2
3
4
:error-while-loading pl.package.calc-test
NullPointerException [trace missing]
(pst)
NullPointerException

Great. There is nothing better than a good exception error. I’ve spent a lot of time trying to solve this, and solution was extremely simple.
My function was defined with defn-, instead of defn. defn- means private scope and test code, could not access testing function.

Do not trust assertions

Assertions can be misleading. When tested code does not work properly and returns wrong results, error messages are like this:

Assertions problems
1
2
3
4
ERROR in math-test/math-operation-test (RT.java:528)
should round using half up
expected: (= 8.31M result)
actual: java.lang.IllegalArgumentException: Don't know how to create ISeq from: java.math.BigDecimal

I hadn’t got time to investigate it, but in my opinion it should work out of the box.

Summary

It is a matter of time, when tools will be better. Those problems will slow you down, and they are not nice to work with.

Astonished

The Clojure concurrency impressed me. Until then, I knew only standard Java synchronization model and Scala actors model. I’ve never though that concurrency problems can be solved in a different way. I will explain Clojure approach to concurrency, in details.

Normal variables

The closest Clojure’s analogy to the variables are vars, which can be created by def.

Vars
1
2
3
4
(defn a01 []
(def amount 10)
(def amount 100)
(println amount))

We also have local variables which are only in let scope. If we re-define scope value of amount, the change will take place only in local context.

Lets
1
2
3
4
5
(defn a02 []
(let [amount 10]
(let [amount 100]
(println amount))

(println amount)))

The following will print:

Lets output
1
2
100
10

Nothing unusual. We might expect this behavior.

Concurrent access variables

The whole idea of concurrent access variables can be written in one sentence. Refs ensures safe shared access to variables via STM, where mutation can only occur via transaction.
Let me explain it step by step.

What is Refs?

Refs (reference) is a special type to hold references to your objects. As you can expect, basic things you can do with it is storing and reading values.

What is STM?

STM stands for Software Transactional Memory. STM is an alternative to lock-based synchronization system. If you like theory, please continue with Wikipedia, otherwise continue reading to see examples.

Using Refs

Refs reads
1
2
3
(defn a03 []
(def amount (ref 10))
(println @amount))

In the second line, we are creating reference. Name of this reference is amount. Current value is 10.
In the third line, we are reading value of the reference called amount. Printed result is 10.

Modifying Refs without transaction

Refs writes without transaction
1
2
3
4
(defn a04 []
(def amount (ref 10))
(ref-set amount 100)
(println @amount))

Using ref-set command, we modify the value of the reference amount to the value 100. But it won’t work. Instead of that we caught exception:

Exception
1
IllegalStateException No transaction running  clojure.lang.LockingTransaction.getEx (LockingTransaction.java:208)

Using transaction

Refs writes with transaction
1
2
3
4
(defn a05 []
(def amount (ref 10))
(dosync (ref-set amount 100))
(println @amount))

To modify the code we have to use dosync operation. By using it, we create transaction and only then the referenced value will be changed.

Complete example

The aim of the previous examples was to get familiar with the new operators and basic behavior.
Below, I’ve prepared an example to illustrate bolts and nuts of STM, transactions and rollbacks.

The problem

Imagine we have two references for holding data:

  • source-vector containing three elements: “A”, “B” and “C”.
  • empty destination-vector.

Our goal is to copy the whole source vector to destination vector. Unfortunately, we can only use function which can copy elements one by one - copy-vector.

Moreover, we have three threads that will do the copy. Threads are started by the future function.

Keep in mind that this is probably not the best way to copy vectors, but it illustrates how STM works.

Refs writes with transaction
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
(defn copy-vector [source destination]
(dosync
(let [head (take 1 @source)
tail (drop 1 @source)
conj (concat head @destination)]

(do
(println "Trying to write destination ... ")
(ref-set destination conj)
(println "Trying to write source ... ")
(ref-set source tail)
(println "Sucessful write " @destination)))
)
)


(defn a06 []
(let [source-vector (ref ["A" "B" "C"]) destination-vector (ref [])]
(do
(future (copy-vector source-vector destination-vector))
(future (copy-vector source-vector destination-vector))
(future (copy-vector source-vector destination-vector))
(Thread/sleep 500)
@destination-vector
))
)

Execution

Below is the output of this function. We can clearly see that the result is correct. Destination vector has three elements. Between Sucessful write messages we can see that there are a lot of messages starting with Trying to write.
What does it mean? The rollback and retry occurred.

Printed messages
1
2
3
4
5
6
7
8
9
10
11
12
13
14
(l/a06)
Trying to write destination ...
Trying to write source ...
Trying to write destination ...
Trying to write destination ...
Sucessful write (A)
Trying to write destination ...
Trying to write destination ...
Trying to write source ...
Sucessful write (B A)
Trying to write destination ...
Trying to write source ...
Sucessful write (C B A)
=> ("C" "B" "A")
Rollback

Each thread started to copy this vector, but only one succeed. The remaining two threads had to rollback work and try again one more time.

Software Transaction Memory

When Thread A (red one) wants to write variable, it notices that the value has been changed by someone else - conflict occurs. As a result, it stops the current work and tries again whole section of dosync. It will try until every write operation succeed.

Pros and cons of STM

Cons:

  • Everything that happens in dosync section has to be pure, without side effects. For example you can not send email to someone, because you might send 10 emails instead of one.
  • From performance perspective, it makes sense when you are reading a lot from Refs, but rarely writing it.

Pros:

  • Written code is easy to read, understand, modify.
  • Refs and transactions are part of standard library, so you can use it in Vanilla Java. Take a look at this blog post for more examples.

Summary

There is a lot that Java developers can gain from Clojure. They can learn how to approach the code and how to express the problem in the code. Also they can discover tools like STM.

If you like to develop your skills, you should definitely experiment with Clojure.

Software Developer on leadership conference? StretchCon Summary

Motivation to this post

Software developer skill set should not be limited to hard programming skills. Also, important for of our work is communication, problem understanding, self-sufficiency and other soft skills.

It this blog post, I would like to show the positive aspects for software developer who participated in agile like conference. You will also find here a lot of information about Stretch Con, on which my experience is based on.

As a conclusion, I’ll present what outcome Stretch conference had on me.

Stretch main scene. Open ceremony

Why I wanted to go?

There are many Agile/Lean/Leadership conferences, Thus, you do not have to choose Stretch. Look around for upcoming events, meetups or trainings. There is always something going on. But in this post I’ll only focus on Stretch.

On my regular basis I am software developer and at least half time of my job I spend on programming. My contribution in a company is not strictly related to any management role.

I wanted to go to Stretch Con, because I’ve belived that:

  • Everyone is a leader of himself/herself. You have to manage your time efficiently.
  • I can learn technical things by myself (by studying them). I did not know how to develop my soft skills – or at least I did not know how to start.
  • Understanding processes, dependencies and co-workers’ and clients’ motivations are crucial and they improve the quality of produced software.
  • You can not connect dots looking forward.

Stretch Con 15

Stretch was different, unlike any other conference I have attended, mainly because of the topic but also because of the fact that it forced me to think and interact.

Open spaces

Open spaces were great. Topics were shaped dynamically (voted via sli.do). It was a place where you could directly see, that other people, from different companies (different countries) have the same problems!

Regular form of open space

Those open spaces had a form of brainstorming ideas, where everyone throws an possible solution to the problem. It gave us the possibility to share and discuss ideas.

Joseph discussion panel

During the open space time, something unexpected happened. Discussion panel with Joseph Pelrine emerged. It started naturally, and eventually a lot of people accumulated around him. Gathered people asked Joseph questions, a he responded with deep explanation. Discussion was about:

  • transparency,
  • environment,
  • estimation,
  • product owner,
  • estimation,
  • ErlangC model.

And many more topics, but I was not able note everything. For such moments, it was worth going there.

Discussion panel with Joseph

Conference talks

In my opinion, on average every second talk was worth watching. I think it is a good score for single track conference. Listening to ‘leaders’, was priceless. Wide variety of subjects, helped me realize that this topic is huge.

Best talks

As usual, I would like to recommend 3 presentations which are worth seeing, but there are many more that migh interest you. Visit Ustream channel to watch them.

James Clear: The Surprising Power of Small Habits

After a great introduction, James presented detailed knowledge about the habits. He showed us techniques for shaping habits. Explained habit triggers. Finally, he also presented tricks how to sustain our habits.
You can read more about habits at James Clear Page.

Conference video available here.

Tim Steigert: Don’t blame the goats. Get a Goatherd

It was a presentation, that forced reflection about me. It helped me tu understand who is a leader, what is the team or company and how all of it this fits into our world. Most importantly, how to ‘get’ a goatherd.

Conference video available here.

Joseph Pelrine: Coaching “self-organizing” teams

Joseph started his lecture with explanation of Complex systems. Then, he started discussing social self organization. Among many concepts that he presented, one particularly stuck in my mind. You have to setup for good thing to hapen naturally. Then you have to monitor them, and decide what to do more and what to stop. If you want to know how to setup things, you have to watch video.

Conference video available here.

Conference itself

Conference scene
I’ll remember this event as something positive. Here is my summary.

Pros:

  • Scene decoration was consistent with name of the conference. You may think it is not important, but it really helped me a lot, to put my brain into good mood.
  • Content of the gift bag, including book “This is Lean“. For the first time I’ve received the book, instead of a useless gadget in gift bag. Great idea!
  • Great venue localization. Venue itself was impressive, too.
  • Selected conference speakers including book authors, people who change things (e.g. John Bunch - Holacracy guru at Zappos). I’ve had a feeling that Conference program committee, made many hard decisions.

This is Lean - main gift in attendee bag

Cons:

  • Everyone were talking in Hungarian. It was so hard to start a discussion during breaks.

Outcome

I’ve made 464 lines of notes, from the whole conference. There were also official notes, in case I missed something. Great concept and great drawing.

Official conference notes

I was there with Wojtek. After the conference we have spend 3 hours talking and discussing Stretch content. We’ve managed only to discuss only about few talks - there were a lot of material presented there.

On my way home, I’ve written down few action points, reflections about myself, that I will try to develop in the upcoming weeks.

  • Consciously shape your habits.
  • Turn the camera to yourself, see your actions and behaviors.
  • Stretch your horizon, to look for new opportunities and possibilities. It will help you with problem solving.
  • Do I really watch carefully?
  • Try not only to hear, but carefully listen to your peers.
  • Find better ways to communicate.
  • Discover Holocracy and decentralized way of running organizations.
  • Engagement, purpose, trust are more important than you think.
  • People mindset exists.
  • Research more about Complex systems.

In conclusion, I highly recommend attending this kind of event from time to time to every software developer. Surely, you will come out as a different person.

Clojure - How to replace current time creation?

Recently, I’ve begun my adventure with Clojure programming language. As a result I’ve decided to share gathered knowledge and my opinion, in this and in few upcoming posts.

The problem

I had to implement algorithm that depends on the current date. Core information for this algorithm is number of days between current date and some date in the future, expressed in days.
Therefore, there is a call somewhere in the code:

Current time with Java 8
1
(. java.time.LocalDate now)

For the tests to be stable, I had to make sure that this call always return the same day.

Approach 1

I’ve decided to extract creation of the current date functionality to the function:

now-date function will return current time
1
(defn now-date [] (. java.time.LocalDate now))

During tests I’ve declared different function:
fixed-date function will return current time
1
(defn fixed-date [] (. java.time.LocalDate of 2015 01 28))

Passing function that creates a current date, solved the problem. It worked great, but it had the following disadvantages:

  • Passing to algorithm function that creates current time.
  • Using java notation (with dot) in Clojure.

Approach 2

Having a function, that returns a current time, I’ve decided to find a way of overwriting its definition in tests. I’ve found out that there is operation called with-redefs-fn, which allows re-defining the function temporarily in the local context. Having defined fixed-date function, block of code looks like this:

Replacing now-date with fixed-date
1
2
3
4
5
6
7
8
9
(deftest my-test
(with-redefs-fn { #'fixture/now-date fixed-date }
#(do
(testing "should work method a"
(let [result (fixture/do-stuff 30)]
(is (.equals 8.22M result))
))

;More tests
))
)

fixture/now-date is a reference to function that I wanted to replace. This time I was amazed by language possibilities. But there was one more problem to solve. I did not want to use java notation.

Approach 3

There is a library called Clj-time. It wraps Joda Time library and makes Clojure code more friendly. I wanted to hold on Java 8 library, but I did not see any alternatives.

So I replaced (. java.time.LocalDate now) to (t/now) and also creation of fixed dates, and then I came up with an idea.

Approach 4

Maybe should I replace the Clj-time itself? My production code will be simpler and the test code will be simpler too!

Replacing t/now with fixed date
1
2
3
4
5
6
7
8
9
(deftest my-test
(with-redefs-fn { #'t/now #(t/date-time 2015 11 19) }
#(do
(testing "should work method a"
(let [result (fixture/do-stuff 30 1000)]
(is (.equals 8.22M result))
))

;More tests
))
)

This is my final solution. I am still impressed how easily it that can be done.

I use Clojure for a week. If you have any other ideas how to solve this problem comment, let me know.

Photo credit: Vintage alarm clock, Thumbnail

The Dream Team Nightmare

My previous experience with agile books was not so good. For one thing, they’re overfilled with advices and strategies, often without context. Put differently, it always required some additional effort, to imagine newly discovered strategies in my past experience. As a consequence, it may not be properly understand and implemented by my.

5 days of team live

Main character of this book is a agile coach. He is hired by a company to help with underperforming team - the Dream Team. The book is sliced into about 250 chapters. In addition, every chapter is a new chance to learn something new. I’ve marked 26 notes, related to:

  • decision making
  • dealing with situations under pressure
  • meeting techniques (e.g. retrospectives)
  • asking correct questions
  • self-organization
  • effective communication
  • good practices (or habits)
  • metering possibilities
  • planning
  • preserving an argument

Most significantly, you can clearly see all these techniques in the same order as in Spock testing framework

  1. Given - full specification of the problem
  2. When - action taken to resolve
  3. Then - result of those actions

Reading this book

You can find it, in a book category on amazon. But it is really a book? I do not think so. It was more like game for me.

Every ~10 chapters you have to made a decision. Every decision you made may lead to fail or success of your mission. At first I was confused about this approach and I came back to previous choices and where they lead me.

After reading a about 50 pages I decided to draw a chapter graph (with decision), and come back to paths that I not chosen when I read the whole book.

Outcome

As a result, I’ve managed to made only good choices and lead the team to happy ending. The graph was on five A4 pages. It looks like this:

Me with chapters diagram

In contrast to good choices, failure paths showed me where the team my end. Learning from someone else mistakes is most important lesson for me.
I would like to thank Wojtek Erbetowski for recommending me this book.

Photo credit: Nightmare